• Demonstration: Professional Monograph
Image for Melatonin (N-acetyl-5-methoxytryptamine)
Melatonin (N-acetyl-5-methoxytryptamine)

Synonyms/Common Names/Related Substances:

  • 5-Methoxy-N-acetyltryptamine, 6-sulfatoxymelatonin, acetamide, agomelatine, aMT6s, beta-methyl-6-chloromelatonin, BMS-214778, CAS 73-31-4, hypnotic, indole, luzindol, luzindole, mel, MEL, melatonia, melatonine, Melaxen®, melaxene, MLT, MT, N-2-(5-methoxyindol-3-ethyl)-acetamide, N-acetil-5-metoxitriptamina, N-acetyl-5-methoxytryptamine, neurohormone, ramelteon (CAS 196597-26-9, TAK-375), tryptophan.
  • Brand names: Accurate Release®; Appleheart Melatonin®; Circadin®; Inspired by Nature®; Mel®; Melatonin Controlled Release®; Melatonin Olympian Labs®; Melatonin Tablets®; Melatonin Time Release®; Melatonin-BioDynamax®; Melatonin-Metabolic Response Modifier®; Melatonin-New Hope Health Products®; Melatonin-Optimum Nutrition®; Melaxen®; Nature's Bounty®; Puritan's Pride®; Rozerem®; Twinlab® Melatonin; Valdoxan®.
  • Combination product examples: Melatonex® (vitamin B6); Melatonin Forte® (Piper methysticum, kavalactones, valeriana); Melatonin PM Complex® (vitamin B6, vitamin B2, vitamin B3); Melatonin spray® (gamma-aminobutyric acid, pyridoxal-5-phosphate); Super Snooze with Melatonin® (valerian root, hop, skullcap, chamomile, passion flower).

Clinical Bottom Line/Effectiveness

Brief Background

  • Endogenous melatonin is an indole neurohormone produced in the brain by the pineal gland, from the amino acid tryptophan (1), with regulation of day-night changes in synthesis regulated by changes in the activity of serotonin N-acetyltransferase (2). The synthesis and release of melatonin are stimulated by darkness and suppressed by light, suggesting the involvement of melatonin in circadian rhythm and regulation of diverse body functions. Levels of melatonin in the blood are highest prior to bedtime. Melatonin acts on MT(1) and MT(2) melatonin receptors located in the hypothalamic suprachiasmatic nuclei, the site of the body's master circadian clock.
  • Due to increases in melatonin levels at night, the most common use of melatonin is to aid in sleep. The uses of melatonin which have the strongest supporting evidence are delayed sleep phase syndrome, insomnia in children and the elderly, jet lag, and sleep disorders in individuals with behavioral, developmental, or intellectual disorders. The weakest evidence in support of melatonin is in relation to work shift sleep disorder. While this is promising and the subject of future clinical research, good evidence in support of melatonin for other uses is lacking.
  • New drugs that block the effects of melatonin are in development, such as BMS-214778 or luzindole, and they may have uses in various disorders (3;4;5). Ramelteon (Rozerem®) and agomelatine (Valdoxan®) (the latter not yet U.S. Food and Drug Administration (FDA) approved) are drugs that have a high affinity for the melatonin receptors MT1 and MT2, present in the suprachiasmatic nucleus, the circadian pacemaker (6). Melatonin agonists, such as beta-methyl-6-chloromelatonin, have also been studied (7). Details and clinical trials of these drugs are not included in this monograph.

Scientific Evidence for Common/Studied Uses:

    Indication

    Evidence Grade

    GRADING SYSTEM LINK

    Delayed sleep phase syndrome (DSPS)

    B

    Insomnia (children)

    B

    Insomnia (elderly)

    B

    Jet lag

    B

    Sleep disorders (individuals with behavioral, developmental, or intellectual disorders)

    B

    Sleep enhancement in healthy people

    B

    Age-related macular degeneration

    C

    Aging (thermoregulation)

    C

    Alzheimer's disease/ cognitive decline

    C

    Anti-inflammatory

    C

    Benzodiazepine tapering

    C

    Cancer treatment

    C

    Cardiovascular disease

    C

    Chronic fatigue syndrome

    C

    Chronic obstructive pulmonary disease

    C

    Circadian rhythm sleep disorders (visually impaired and nonimpaired individuals)

    C

    Delirium

    C

    Depression

    C

    Diabetes (adjunct therapy)

    C

    Exercise performance

    C

    Fertility

    C

    Fibromyalgia

    C

    Gastrointestinal disorders

    C

    Glaucoma

    C

    Headache

    C

    Hepatitis

    C

    High blood pressure (hypertension)

    C

    High cholesterol

    C

    HIV/AIDS

    C

    Memory

    C

    Menopause

    C

    Pain

    C

    Parkinson's disease

    C

    Periodic limb movement disorder

    C

    REM sleep behavior disorder

    C

    Restless leg syndrome

    C

    Rett's syndrome

    C

    Sarcoidosis

    C

    Schizophrenia

    C

    Seasonal affective disorder (SAD)

    C

    Seizure disorder

    C

    Sleep disturbance

    C

    Smoking cessation

    C

    Stroke

    C

    Surgical uses

    C

    Tardive dyskinesia

    C

    Thrombocytopenia (low platelets)

    C

    Tinnitus

    C

    Ulcers

    C

    Urination (nocturia)

    C

    UV-induced erythema prevention/sunburn

    C

    Work shift sleep disorder

    D

    Historical or Theoretical Indications That Lack Sufficient Evidence:

    • Acetaminophen toxicity (8;9), acute respiratory distress syndrome (ARDS) (10;11), adaptogen (12), addiction (13), adrenal insufficiency, aging (general) (14;15;16;17;18;19;20;21;22;23;24), alopecia (25), amenorrhea (26;27), amyotrophic lateral sclerosis (ALS) (28), antioxidant (29;30;31;32;33;34;35;36;37;38;39;40;41;42;43;44;45;46;47;48;49;50;51;52;53;54;55;56;57), anxiety (58), arthritis (59;60), ataxia (Machado-Joseph disease) (61), atopic dermatitis (62), attention deficit hyperactivity disorder (ADHD) (63), autoimmune diseases (demyelination) (5), beta-blocker sleep disturbance (64;65), bipolar disorder (66;67;68;69;70), bladder disorders (71), bone diseases (72;73), bone healing (74;75;76), brain injuries (77;78;79;80), cachexia (81;82), cataracts, chemotherapy toxicity (83;84;85;86;87;88;89), colic (90), contraception, dental conditions (44), dry skin (91), Duchenne muscular dystrophy (92;93), eating disorders (low-level, enhanced circadian rhythm) (94;95), eczema (96), edema, endometriosis (97;98), erectile dysfunction (99), esophagitis (100;101), exercise recovery (102), food preservation (103), fragile X syndrome (104), growth (growing pains) (105), helminthic infections (Schistosoma mansoni) (106), hepatoprotection (107;108;109;110;111;112), hormonal/endocrine disorders (McCune-Albright syndrome) (72), hyperpigmentation, immunomodulation (113;114;115;116), infant development / neonatal care (117;118), interstitial cystitis, ischemia-reperfusion injury protection (80;119;120;121;122;123;124;125;126;127;128;129;130;131;132;133;134), itching, jaundice (135), jellyfish stings (136), kidney protection (137;138;139;140;141), lung inflammation (142;143), malaria (144), melatonin deficiency, metabolic disorders (Sanfilippo syndrome) (145), migraine (impaired pineal function) (146;147;148;149;150;151;152), movement disorders (153;154), multiple sclerosis (155;156;157), neurological disorders (72;79;158;159;160;161;162;163;164;165;166;167;168;169;170;171;172;173;174;175;176;177;178;179;180;181;182;183;184;185;186;187;188;189), nitrate tolerance (190), noise-induced hearing loss (191), obesity (192;193), obstructive sleep apnea (symptoms) (194), osteoporosis, ovarian disorders (195;196), pancreatitis (197;198;199;200;201;202), parasite infection (203), phenylketonuria (PKU) (204), photoprotection (205;206;207), polycystic ovarian syndrome (PCOS) (208), pregnancy nutritional supplement (209;210), premenstrual dysphoric disorder (chronobiological abnormalities of melatonin secretion) (211), psychiatric disorders (212), pulmonary fibrosis (213), radiation protection (214;215;216;217;218;219), retinal protection (220;221;222), scalds (223), sepsis (224;225;226), shock (227;228), spinal cord injury (229;230;231), spine problems (idiopathic scoliosis) (232;233), stomatitis (100), stress (234), sudden infant death syndrome (prevention) (235;236;237), testicular damage (238), toxicity (cadmium, formaldehyde, mercury, aflatoxin, aluminum, cyclosporine, gentamycin, lead, organophosphate, alcohol, toluene) (239;240;241;242;243;244;245;246;247;248;249;250;251;252;253;254;255;256;257;258;259;260;261;262), tuberculosis (263), uterine disorders (hormone-dependent, myometrial functioning) (264), Wilson's disease (265), withdrawal from narcotics (266;267), wound healing (268).

    Expert Opinion and Historic/Folkloric Precedent:

    • Melatonin is widely recommended for various sleep disorders and for prevention of jet lag. In addition, it is used in conditions believed to be associated with low levels of endogenous melatonin, such as aging, sleep disorders in children, and affective disorders. It has also garnered attention as possibly playing a role in, or serving as a treatment for, chronic inflammatory diseases (269), cancer (270), and hypertension (271), as well as an antioxidant therapy to counter aging and a variety of metabolic diseases (272).
    • Many authors have published studies, including reviews, suggesting that melatonin may be of interest for ulcers (100;273;274), ulcerative colitis (275), surgical recovery (postpinealectomy syndrome) (276), ischemic stroke (impaired nocturnal excretion) (277;278), stroke (279), sexual activity enhancement (280), schizophrenia (66;281), myocardial injury (282;283), major depressive disorder (low nocturnal melatonin) (284;285;286), jet lag (287), hemodynamic parameters (288), delirium (289;290;291), Alzheimer's disease (180;292;293;294;295), analgesia (296;297;298), cardiovascular conditions (acute coronary syndromes) (299), cardiovascular conditions (nicotine-induced vasculopathy) (300), cardiovascular conditions (ventricular fibrosis) (301;302), colitis (303;304;305), coronary artery disease (306), gastric ulcers (307;308;309;310), gastritis (100), gastroesophageal reflux disease (GERD) (311), intestinal motility disorders (312), postoperative adjunct (313;314;315;316;317;318;319;320;321), primary insomnia (322;323), pain (324), critical illness (325), hemicrania continua (326), postoperative delirium (290), and various other uses (327). Melatonin was also found to decrease the susceptibility of erythrocytes to oxidation during blood storage (328). It has been hypothesized that mining is protective against prostate cancer due to the supposed increases in melatonin levels in individuals in this profession (329).
    • A review by Bjorvatn and Pallesen outlines how to estimate circadian rhythm based on a careful patient history and use this estimate to administer melatonin or light to treat delayed sleep phase disorder, advanced sleep phase disorder, free-running, irregular sleep-wake rhythm, jet lag disorder, and shift work disorder (330).
    • Acute sleep deprivation lacked an effect on urinary melatonin levels (331). In human research, individual differences in the amplitude reduction of the melatonin rhythm were correlated with the amplitude of cortisol and body temperature (332).
    • The impact of light from computer monitors on melatonin levels has been determined in human research (333). Melatonin concentrations were found to be reduced after exposure to the blue-light goggle experimental condition compared with a dark control condition or a computer monitor only. Diminished melatonin secretion in elderly persons with insomnia was suggested as being caused by insufficient environmental illumination (334).
    • Urinary levels of melatonin and 6-hydroxymelatonin sulfate changed during the life cycle of infants and prepubertal children, with declining levels in preterm babies and increased levels in those aged 4-7 years; night-day differences were lacking before six months of age (335). Nocturnal melatonin patterns in children were also investigated and were found to be affected by pubertal age (336). In children with rest-activity disturbances with septo-optic dysplasia, melatonin levels were absent, normal, or increased with daylight, suggesting that effects of melatonin were not consistent within a subgroup like this (337). Low melatonin production, measured by urinary 6-sulfatoxymelatonin, was found in infants with a life-threatening event (338).
    • Melatonin was suggested as being of interest for idiopathic rapid eye movement sleep behavior disorder (339) and insomnia following an improvised explosive device-induced traumatic brain injury (340). In human research, melatonin levels decreased in patients with epilepsy; however, levels increased following seizures (341) or in untreated patients with active epilepsy (342). Endogenous melatonin profiles were examined in asymptomatic inflammatory bowel disease; however, further details are lacking (343).
    • Peres et al. wrote a review on the possible therapeutic effects of melatonin for cluster headache; this is based on lower levels in these patients, as well as lower levels during the cluster period rather than remission (344). In a review, Dhillon et al. hypothesized that low levels of copper in patients with migraines might exacerbate the deficiency of melatonin, as well as zinc and coenzyme Q10, commonly found in patients with migraines (345).
    • Duman reported on a case of spontaneous hypermelatoninemia possibly related to irregular control of pinealocytes and treatable with propanolol (346).
    • Sanders wrote a review on aeromedical, toxicopharmacological, and analytical aspects of melatonin (347). The authors suggested that use of melatonin by aviation professionals may pose a safety risk for passengers. Sigurdardottir published a systematic review of epidemiological studies examining circadian disruption, sleep loss, and prostate cancer risk (348).
    • A physician survey indicated that 25% of doctors polled recommended melatonin as a complementary alternative medicine for children with autism (349). Melatonin prescribing among customers of compounding pharmacies in Australia was explored; however, details are lacking (350).
    • Melatonin is not listed in the United States Food and Drug Administration (FDA) Generally Recognized as Safe (GRAS) list.

    Brief Safety Summary:

    • Likely safe: When used in doses commonly studied in clinical trials. Although doses of up to 40mg have been studied over 23 days (351), commonly studied doses range from 1 to 20mg, with mild or no adverse effects in many studies. In children, when used long term (mean use: 3.1 years and mean dose: 2.69mg) due to a lack of negative effect on sleep, puberty, and mental health (352).
    • Possibly unsafe: When used in individuals with genitourinary disorders, such as children with a history of enuresis, due to reports of enuresis in clinical trials (353;354;355;356;357;358). When used in aviation professionals, based on expert opinion suggesting that use of melatonin by aviation professionals may pose a safety risk for passengers (347). When used in individuals with hypotension or hypertension or those using agents that affect blood pressure, based on reports in humans of both blood pressure reductions and increases (359;360;361;362;363;364;365;366;367;368;369). When used in patients with heart disease, due to reports of abnormal heart rhythms, palpitations, fast heart rate, or chest pain, in patients usually also taking other drugs that may account for these symptoms (370;371;372;373;374), as well as reports of increased heart rate in some (368;375), but not all (376;377;378), studies. When used in patients with skin disorders, due to reports of skin reactions (356;358;379;380;381). When used in individuals with, or at risk of, diabetes, or those using hypoglycemic or hyperglycemic medications, based on reports of elevated blood sugar levels in patients with type 1 diabetes (382;383) and reduced glucose tolerance and insulin sensitivity (384;385), and due to research indicating that melatonin in combination with zinc improved postprandial glycemic control in patients with type 2 diabetes (386;387). When used in individuals with gastrointestinal disorders, due to reports of gastrointestinal adverse events in clinical trials and case reports (354;355;357;358;380;381;382;388;389;390;391;392;393;394;395;396;397;398). When used in individuals with blood-clotting disorders or those using medications that affect blood clotting, based on reports of a dose-response relationship between the plasma concentration of melatonin and coagulation activity (399), case reports of alterations in prothrombin time (a measurement of blood clotting ability) in patients taking both melatonin and the blood-thinning medication warfarin (Coumadin®) (372), reports of nosebleeds (358), and experts suggesting that melatonin should be avoided in patients using warfarin and possibly in patients taking other blood-thinning medications or in those with clotting disorders (400); also, in humans, melatonin was associated with lower plasma levels of procoagulant factors, and dose-response relationships between the plasma concentration of melatonin and coagulation activity have been hypothesized (399). In animal research, melatonin enhanced platelet responsiveness (401). In patients with decreased platelets due to cancer therapies, increased platelet counts after melatonin use have been observed (402;403;404;405;406;407;408), and cases of idiopathic thrombocytopenic purpura (ITP) in patients treated with melatonin have been reported (409;410). When used in individuals with musculoskeletal disorders, due to case reports of ataxia following overdose (372). When used in individuals with neurological disorders, as well as individuals driving or operating heavy machinery or using sedatives, depressants, or stimulants, and in individuals with hyperactitivity, due to reports of neurological adverse effects, such as fatigue, dizziness, headache (including migraine), irritability, drowsiness, weakness, fogginess, yawning, nighttime awakening, poor sleep quality, sleepiness, insomnia, hyperactivity, worsening behavior, and other effects (no year (372), 1970-1979 ((381), 1980-1989 (411;412;413;414;415;416), 1990-1999 (391;393;398;417;418;419;420;421;422;423;424;425), 2000-2009 (353;354;355;356;357;358;366;367;371;380;388;390;392;396;426;427;428;429;430;431;432;433;434;435;436;437;438;439;440;441;442;443;444;445)). When used in individuals who often feel cold, due to reports of feeling cold and hypothermia in clinical trials (356;390;392;445;446;447;448). When used in individuals with or at risk of seizures, as melatonin may act as a proconvulsant (449;450) and may lower seizure threshold and increase the risk of seizure (451), particularly in children with severe neurologic disorders, and in adults with recurring symptoms following repeated melatonin administration (372;452), as well as due to reports of developing epilepsy (431) and an increased seizure risk (354;380); however, due to reports of decreased seizures (453;454;455;456;457;458), this remains an area of controversy (449). When used in individuals with glaucoma or other ocular disorders, due to human and laboratory reports of increased intraocular pressure (382) or retinal damage (400;459); however, there is preliminary human evidence that melatonin may actually decrease intraocular pressure in the eye, and it has been suggested as a possible therapy for glaucoma (366;460;461). When used in individuals with psychiatric disorders, due to clinical reports of psychotic symptoms possibly due to overdose (371;462), as well as effects on mood changes (390;391;392;419;440;463), delusions and hallucinations (441), and aggressiveness (433). When used in patients with inflammatory disorders or those using anti-inflammatory agents, due to reports of an increase in proinflammatory cytokines and erythrocyte sedimentation rate (464). When used in patients that are breastfeeding, based on a lack of adequate safety evidence. When used in individuals using agents metabolized by cytochrome P450, as melatonin appears to inhibit CYP1A2 (465;466;467) and induce CYP3A. In animal research, melatonin inhibited the activity of cytochrome P450 2E1, but to a lesser extent than taurine (468). When used in individuals using hormonal agents or those with alterations in hormone levels, based on hormonal effects that have been shown in both human and animal studies (411;469;470;471;472;473;474;475;476;477;478;479;480;481;482;483;484;485;486;487;488;489;490;491;492;493;494;495;496;497;498;499;500;501;502;503;504;505;506;507;508). When used in individuals with immune function disorders or those using immunomodulating agents, due to reports of interactions in human research with immune therapies, such as interferon (509), interleukin-2 (407;408;510;511;512;513;514;515;516;517;518;519;520;521;522;523;524;525;526;527;528), or tumor necrosis factor (525;529;530), as well as other effects on immune response and immune mediators in human and laboratory research (101;114;133;283;402;403;404;406;407;408;531;532;532;533;534;535;536;537;538;539;540;541;542;543;544;545).
    • Likely unsafe: When used in women who are pregnant or are attempting to become pregnant, based on hormonal effects that have been shown in both human and animal studies (411;469;470;471;472;473;474;475;476;477;478;479;480;481;481;482;483;484;485;486;487;488;489;490;491;492;493;494;495;496;497;498;499;500;501;502;503;504;505;506;507;508), including alterations of pituitary contractions (546), a possible increased risk of developmental disorders due to high levels of melatonin during pregnancy (547), decreased sperm count and motility (in human and animal research) (400;548;549), inhibited ovarian function (391), and decreased libido (438). When used in patients with known allergies to melatonin or related products, based on reports of allergic skin reactions after taking melatonin by mouth (382;550).

    Dosing/Toxicology

    General:

    • Doses may be based on those most commonly used in available trials, or on historical practice. However, with natural products, the optimal doses to balance efficacy and safety are often unclear. Preparation of products may vary from manufacturer to manufacturer, and from batch to batch within one manufacturer. As it is often not clear what the active component(s) of a product may be, standardization is often not possible, and the clinical effects of different brands may not be comparable.

    Standardization:

    • There is no well-known standardization for melatonin. Experts note that many brands contain impurities that cannot be characterized, as well as dissimilar amounts of actual hormone. In 2002, Consumer Lab evaluated 18 melatonin-containing supplements (15 quick-release and three time-release products), of which 12 were melatonin-only products. It was reported that 16 of the 18 products contained 100-135% of the claimed amount of melatonin, one rapid-release product contained only 83% of the claimed amount of melatonin, and another rapid-release product contained a small amount of lead (slightly more than 0.5mcg daily based on the recommended serving size of melatonin). Among the 12 melatonin-only products that "passed" these standards are: Nature's Bounty® Melatonin 1mg and 3mg tablets, Puritan's Pride® Inspired by Nature® Melatonin 3mg tablets, Twinlab® Melatonin Caps, Highest Quality, Quick Acting 3mg tablets.
    • U.S. Pharmacopeia melatonin has been obtained from Sigma Company (St. Louis, MO) (551).
    • In a clinical trial, chromatographically pure melatonin was dissolved in a 1:90 mixture of ethanol and 0.9% saline (552).

    Adult (age ≥18):

      Oral:

      • General: Time of melatonin administration is important. When administered orally in the morning, melatonin delays circadian rhythms, but advances circadian rhythms when administered in the afternoon or early evening (553).
      • Age-related macular degeneration: Melatonin 3mg each night at bedtime for six months has been shown to protect the retina and delay macular degeneration (459).
      • Aging (thermoregulation in the elderly): Melatonin 1.5mg nightly for two weeks has been shown to stabilize body temperature rhythm (554).
      • Alzheimer's disease/cognitive decline: Melatonin 1-6mg, administered before bedtime for 1-2 months, has been shown to improve cognition and memory in aging patients (555;556;557;558;559;560). It has been used for as little as 10 days (555) and as long as three years (556). Doses of 2.5-10mg daily were reported in systematic reviews (561;562) for 10 days to 35 months (562).
      • Anti-inflammatory: Melatonin 10mg daily at night over six months has been investigated in patients with rheumatoid arthritis (RA); however, controversial effects on cytokine production were noted, and the anti-inflammatory action and benefit in patients with RA remains unclear (464). Melatonin 5mg the night before and one hour prior to surgery produced anti-inflammatory effects in patients undergoing surgery (563).
      • Asthma: Melatonin 3mg for four weeks improved sleep quality in patients with asthma (primary outcome) but lacked a significant benefit on asthma symptoms (other outcome) (564).
      • Benzodiazepine tapering: Doses ranging from 1-5mg daily have been studied (565;566;567;568;569). Treatment duration most commonly lasted several weeks (up to six weeks); however, one study (dose unclear) (533) continued for a year.
      • Cancer treatment: Various doses of melatonin have been studied in patients with cancer, usually given in addition to other standard treatments such as chemotherapy, radiation therapy, or immune therapy (81;351;512;570;571;572;573;574;575;576). Safety and effectiveness are not proven, and melatonin should not be used instead of more proven therapies. Oral doses have ranged between 1-40mg daily, with the most common dose being 20mg (81;351;512;570;571;572;573;574;575;576;577). Dosing regimens generally mirrored conventional therapeutic time courses (several weeks to months). Patients are advised to discuss cancer treatment plans with an oncologist before considering use of melatonin either alone or with other therapies.
      • Chronic fatigue syndrome: Melatonin 5mg five hours before dim light melatonin onset during three months improved fatigue, concentration, and activity in patients with chronic fatigue syndrome (578;579).
      • Chronic obstructive pulmonary disease (COPD): Melatonin 3mg has been taken orally each evening two hours before bed for three months (580).
      • Circadian rhythm sleep disorders (visually impaired and nonimpaired individuals): A single dose of 5mg given at 11 p.m. has been found to increase sleep time and sleep efficiency in visually impaired patients (478). A dose of 0.5mg daily at 9 p.m. for 26-81 days was also shown to improve circadian rhythm in visually impaired patients (480). Melatonin has been taken daily for eight weeks (581). Melatinin 3mg has been taken once as a sustained-release supplement (582).
      • Delayed sleep phase syndrome (DSPS): Dosages of melatonin for DSPS range from 0.3mg up to 6mg; however, the most frequent dose was 5mg. Melatonin was most commonly administered daily in the hours preceding the desired sleep period for durations of two weeks to three months (390;583;584;585;586;587;588;589;590;591).
      • Delirium: 0.5mg of melatonin has been taken every evening for up to 14 days (442).
      • Depression: Melatonin 6mg slow-release at bedtime for four weeks has been shown to improve mood (592).
      • Exercise performance: Melatonin 6mg one hour before a heavy-resistance exercise session has been used (508). Melatonin 5mg at bedtime had little benefit on physical performance the morning after (593).
      • Fertility: 3mg of melatonin has been taken between 10 and 11 p.m. from the third to the fifth day of the menstrual cycle until the day when human chorionic gonadotropin (hCG) was injected according to a standard gonadotropin-releasing hormone (GnRH) agonist regimen (594). Melatonin 3mg has been started on the day of GnRH administration (duration unknown) (595).
      • Fibromyalgia: 3mg of melatonin has been taken at bedtime for four weeks (596). 5mg of melatonin has been taken each night for 60 days (597).
      • Gastrointestinal disorders: Melatonin 5mg in the evening for 12 weeks reduced dyspeptic symptoms (598). For IBS, 3mg of melatonin administered daily at bedtime for 2-8 weeks has been used (438;599;600;601;601;602). 3-10mg of melatonin daily (duration unclear) was reported in a systematic review (603).
      • Headache: Studies have evaluated regular use of 5-10mg of melatonin taken nightly by mouth for up to 14 days (604;605;606). 6mg of melatonin has been taken daily (duration unclear), according to a case report (607). In a systematic review, use of 2mg of prolonged-release melatonin one hour before bedtime daily for eight weeks was reported (440).
      • Hepatitis: 5mg of melatonin has been taken twice daily for 12 weeks (445).
      • High blood pressure (hypertension): Melatonin dosages studied were 1-3mg, with most studies investigating treatment using one daily dose prior to bedtime (377;378;507;608), though one study reported daytime administration of a single dose of 1mg (506). The treatment period was up to four weeks in most of the included studies. Melatonin 5mg has been taken daily for 90 days (375) or two months (609). According to a systematic review, 5mg of fast-release melatonin and 2-3mg of controlled-release melatonin have been used for 7-90 days (443).
      • High cholesterol: Melatonin 5mg has been taken daily for two months (609).
      • Insomnia (elderly): Melatonin 0.3-5mg, regular or controlled-release, at or up to 120 minutes prior to bedtime for up to eight weeks has been used in most studies (373;374;379;389;610;611;612;613;614;615;616). Although treatment in most clinical trials were often several weeks, some studies continued for several months (617;618). Melatonin-rich night milk was used for eight weeks in one study (618). One study found that low doses (0.1-0.3mg nightly) appear to be as equally effective as higher doses (3-5mg nightly) (446). The most common product used in these studies was the slow-release melatonin, Circadin® (Neurim Pharmaceuticals Ltd., Tel Aviv, Israel).
      • Jet lag: Melatonin 0.1-0.5mg, usually started on the day of travel (close to the target bedtime at the destination), then taken every 24 hours for several days (424;619), or a more common dose of 5mg (416;419;424;426;432;619;620;621;622;623) has been used. Higher doses of 6-8mg have also been studied (370). Overall, 0.5mg appears to be slightly less effective than 5mg for improvement of sleep quality and latency (371;429), although this area remains controversial, and other research suggests no statistically significant differences between these doses (370;624). Slow-release melatonin may not be as effective as standard (quick-release) formulations (424) for sleep onset, but may be more effective for sleep maintenance. If the dose is taken too early in the day, it may result in excessive daytime sleepiness and greater difficulty adapting to the destination time zone. Formulations were rarely specified; however, one study reported using 2mg of Circadin®, a pharmaceutical-grade time-released formulation of melatonin (Neurim Pharmaceuticals, Tel Aviv, Israel) on circadian sleep and performance in an air crew (623).
      • Memory: Melatonin 3mg by mouth prior to exposure to a laboratory stressor and subsequent recall evaluation was used (625).
      • Menopause: The use of 3mg at bedtime for six months has been studied for the ability to normalize pituitary and thyroid function in menopausal women (488). Another study administered 3mg of pure melatonin for three months for the relief of menopausal symptoms (626). 3mg of melatonin daily (duration unclear) was also reported in a systematic review (627), and 3mg daily for six months has been used in a clinical trial (628).
      • Parkinson's disease: Melatonin 3.0-6.6g daily has been used in patients, some of whom were also treated with levodopa (381). According to a systematic review, participants in melatonin studies were administered 3-50mg of melatonin daily before bedtime for 2-10 weeks (444).
      • Periodic limb movement disorder: One study delivered 3mg of melatonin nightly for a six-week period (629).
      • REM sleep behavior disorder: Melatonin 3-12mg daily has been used for REM sleep behavior disorder (441;630;631;632). The duration was four weeks in one study (632).
      • Restless leg syndrome: A single dose of 3mg of melatonin improved leg discomfort in patients with restless leg syndrome (633).
      • Sarcoidosis: One study reported a treatment regimen lasting two years of 20mg daily in the first year, and 10mg in the second (531).
      • Schizophrenia: In a systematic review, schizophrenic participants were orally administered 2-10mg of melatonin daily to treat tardive dyskinesia (TD) (634).
      • Seasonal affective disorder (SAD): One study examined 2mg of sustained-release melatonin 1-2 hours before bed for three weeks (635). One study has evaluated 0.5mg of melatonin, sublingually, for six days (636).
      • Seizure disorders: Doses used in children and adults in a systematic review were 3-10mg daily for 2-4 weeks, with a conditional extension of two months in one study (637).
      • Sleep (general): Doses of melatonin were 1-10mg in healthy participants or 0.3-10mg in patients with insomnia, or physiologic doses of melatonin (0.1-0.3mg) have been used for an unknown duration (391).
      • Sleep disorders (individuals with behavioral, developmental, or intellectual disorders): In studies that assessed the effects of melatonin in adults, participants were administered 0.1-10mg of melatonin daily for up to one year (354;398;638).
      • Sleep disturbance (Alzheimer's disease): Melatonin 1.5-10mg taken nightly for up to 35 months has been studied in patients with Alzheimer's disease (555;557;559;560;639;640). In one study, 5mg in combination with light exposure for 10 weeks increased daytime wake time and activity levels and strengthened the rest-activity rhythm (641). Melatonin capsules (8.5mg of immediate-release and 1.5mg of sustained-release) has been taken daily for 10 days (642).
      • Sleep disturbance (asthma): Melatonin 3mg for four weeks improved sleep quality in patients with asthma (the primary outcome), but lacked a significant benefit on asthma symptoms (another outcome) (564).
      • Sleep disturbance (autism): In adult patients with autism, a dose of 3mg (increased up to 9mg if ineffective) before bedtime for six months was studied (643). A systematic review of patients with autism noted regimens with doses ranging from 0.75mg to 10mg, administered over two weeks to two months (353).
      • Sleep disturbance (COPD): 3mg of melatonin has been been used daily at 10 p.m. (duration unknown) (644).
      • Sleep disturbance (cystic fibrosis): Melatonin 3mg at bedtime for 21 days improved sleep measures in patients with cystic fibrosis (645).
      • Sleep disturbance (depression): Melatonin 0.5-10mg for three to four weeks has been used in patients with depressive symptoms (592;635;636;639;646). Another study administered melatonin 5-10mg for four weeks in conjunction with fluoxetine (647).
      • Sleep disturbance (healthy people): Doses studied ranged from 0.1 to 80mg (373;374;393;417;427;448;593;648;649;650;651;652;653;654;655;656;657;658;659;660). Melatonin was generally administered daily in the evening hours immediately preceding the desired rest period, though some studies investigated the effect of earlier dosing on nighttime sleep (661;662), as well as daytime naps (663). Most of the previously mentioned studies occurred over the course of one or several days; however, some (373;374;417;652;660) lasted up to 26 weeks.
      • Sleep disturbance (hemodialysis): In hemodialysis patients, 3mg of melatonin for six weeks has been used (664).
      • Sleep disturbance (hospitalized and medically ill): 3-5.4mg of an evening dose of melatonin has been used in hospitalized and medically ill patients (duration unclear) (640;665).
      • Sleep disturbance (intellectual disabilities): A meta-analysis of melatonin therapy in patients with intellectual disabilities noted regimens from included studies with doses ranging from 0.5mg to 9mg and lasting from 32 to 73 days (inclusive of washout periods) (666).
      • Sleep disturbance (Parkinson's disease): Melatonin 3, 5, and 50mg at bedtime for 2-4 weeks have improved sleep disturbances in patients with Parkinson's disease (435;667).
      • Sleep disturbance (postoperatively): In laparoscopic cholecystectomy patients, a dose of 5mg for three nights postoperatively was used (668).
      • Sleep disturbance (psychiatric disorders): Melatonin 2-12mg daily before the desired rest period for up to 12 weeks has been used in patients with psychiatric disorders (669;670;671;672;673).
      • Sleep disturbance (traumatic brain injury): A 5mg dose of melatonin for one month has been used (674).
      • Sleep disturbance (tuberous sclerosis complex): Melatonin 5mg has been taken 20 minutes prior to the usual bedtime for two weeks (675).
      • Smoking cessation: One study used a 0.3mg dose of melatonin given 3.5 hours after nicotine withdrawal (676).
      • Surgical uses: Melatonin 3-15mg orally or sublingually or 0.05-0.2mg/kg sublingually, either as a monotherapy or in combination with other sedatives prior to surgery, has been studied (most often 90 minutes prior to surgery or the night before and 90 minutes prior to surgery) (366;563;677;678;679;680;681;682;683).
      • Tardive dyskinesia (TD): One study delivered 2mg daily for four weeks (671). Another, later study expanded treatment to 10mg daily for six weeks (684). In a systematic review, schizophrenic participants were orally administered 2-10mg of melatonin daily to treat TD (634). 20mg daily for 12 weeks has been used (685).
      • Thrombocytopenia (low platelets): One study administered 20mg daily in the evening for two months (686).
      • Tinnitus: Melatonin 3mg daily for up to 80 days has been used (687;688;689;690;691).
      • Ulcers: Melatonin 5mg has been taken twice daily for 21 days as an adjunct to other medications (692;693).
      • Urination (nocturia): Studies report using 2mg daily for four weeks (694;695).
      • Work shift sleep disorder: Regimens studied included doses which ranged from 1.8mg to 10mg and were generally administered up to three times daily (for up to six days) prior to daytime sleep following a night shift schedule (434;437;696;697;698;699;700;701;702;703;704).
      • Other: There are other uses with limited study and unclear effectiveness or safety. Use of melatonin for these conditions should be discussed with a primary healthcare provider and should not be substituted for more proven therapies.
      • Note: Multiple studies performed to date on the functions of endogenous melatonin have utilized exogenous melatonin, often at high doses. However, the most reliable data are obtained at low doses of exogenous melatonin, at which plasma levels are within a physiological range (661;705). It is not entirely clear what relationships exist between melatonin secretion and pharmacological effects observed at higher concentrations(413).

      Intravenous:

      • Preoperative sedation / anxiolysis: Melatonin 3-10mg and/or 0.05-0.5mg/kg, either as a monotherapy or in combination with other sedatives prior to surgery, has been studied (366;563;677;678;679;680;681;682;706;707).
      • Sleep disturbance (healthy individuals): One study administered melatonin 50mg intravenously (duration unclear) (659).

      Transbuccal:

      • Insomnia (elderly): Transbuccal melatonin 0.5mg for four nights has been used (447).

      Topical:

      • Cancer treatment: Various doses of melatonin have been studied in patients with cancer, usually given in addition to other standard treatments, such as chemotherapy, radiation therapy, or immune therapy. In studies of patients with melanoma, melatonin preparations have been applied to the skin (708). Patients are advised to discuss cancer treatment plans with an oncologist before considering use of melatonin either alone or with other therapies. Safety and effectiveness are not proven, and melatonin should not be used instead of more proven therapies.
      • UV-induced erythema prevention/sunburn: The use of melatonin in topical applications has been studied in preparations of 20 or 100mg dissolved in 70% ethanol (709), in concentrations of 0.05, 0.1, and 0.5% in 0.12mL gel immediately after UV irradiation (710), and as 0.6mg/m2 from 15 minutes before to 240 minutes after UV irradiation (711), alone or in combination with ascorbic acid and vitamin E (712). 5% melatonin in ethanol, propylene glycol, and water vehicle has been studied (713). Participants received 5.85mcL of solutions containing 1-2.5% melatonin on a 1.5cm2 area of the lower back at predetermined time intervals following UV exposure (information regarding these time intervals was lacking), alone, or in combination with vitamins C and E (714).

      Children (age <18):

        Oral:

        • General: There is limited study of melatonin supplements in children, and safety is not established.
        • Anti-inflammatory: Ten doses of melatonin, at doses of 10mg/kg (the first four doses separated by 2-hour intervals, the fifth and sixth doses separated by four-hour intervals, the seventh and eighth doses separated by eight-hour intervals, and the ninth and tenth doses separated by 12-hour intervals) were administered to newborns (11).
        • Circadian rhythm sleep disorders (visually impaired and nonimpaired individuals): Children and adolescents included in clinical trials were administered 3-12mg of melatonin daily for eight weeks (581).
        • Delayed sleep phase syndrome (DSPS): In a systematic review, children in the included studies were administered 3-6mg of melatonin daily for 10-28 days (390).
        • Insomnia (children): Melatonin 1-5mg once daily at bedtime for up to two months has been used (392;431;715;716). 0.05-0.15mg/kg melatonin has been used nightly (between 5:30 p.m. and 7:30 p.m.) for one week (356).
        • Respiratory distress syndrome: Ten doses of melatonin, at doses of 10 mg/kg each (the first four doses separated by 2-hour intervals, the fifth and sixth doses separated by four-hour intervals, the seventh and eighth doses separated by eight-hour intervals, and the ninth and tenth doses separated by 12-hour intervals) were administered to newborns (11).
        • Rett's syndrome: Melatonin 2.5-7.5mg once daily at bedtime for up to two years has been used (717;718).
        • Sedation (children): One study used 3 and 6mg doses of melatonin administered 10 minutes prior to standard oral sedation (719).
        • Seizure disorder (children): Case reports have evaluated 1.5-9mg of melatonin taken daily over treatment durations ranging from two weeks to three months (380;454;455;456;720;721;722). Two (if under nine years old or weighing <30kg) or three (if over nine years or weighing >30kg) tablets containing 3mg of rapid-release melatonin have been taken for 28-32 days (723). Research is limited in this area, and there are other reports that melatonin may actually increase risk of seizure or lower seizure threshold (372;451;452). In another study exploring the effect of melatonin on insomnia in children, a reported case of mild generalized epilepsy developing four months after the start of the trial was noted (431). Therefore, caution is advised, and use of melatonin should be discussed with the child's primary healthcare provider. Doses used in children and adults in a systematic review were 3-10mg daily for 2-4 weeks, with a conditional extension of two months in one study (637).
        • Sepsis: Two 10mg doses, separated by one hour, administered 12 hours after diagnosis has been studied (225).
        • Sleep disorders (children with behavioral, developmental, or intellectual disorders): Dosages usually varied from 3 to 6mg and were administered daily prior to bedtime (354;355;357;358;398;433;436;445;638;643;724;725;726;727;728;729;730;730;731;732;733;734;735;736). Some studies employed lower (0.1-1mg) (398;445;551;737) or higher (up to 10mg) (353;354;355;445;638;643;724;726;734;735;738;739;740) dosages. Treatment duration ranged from one week to 72 months, with the majority lasting several weeks. In one study, participants received pharmaceutical-grade controlled-release melatonin 5mg followed by a three-month open-label study, during which the dose was gradually increased until the therapy showed optimal beneficial effects (587).
        • Sleep disturbance: A case report of one patient reported a dose of 2mg given nightly for four weeks (741) after pinealectomy. Melatonin 6-9mg fast-release melatonin (dose varied depending on the age or size of child), one hour before bedtime, as an add-on to sodium valproate therapy, for four weeks, has been shown to improve sleep behavior in children with epilepsy (397). Melatonin 5mg has been taken 20 minutes prior to the usual bedtime for two weeks (675). Doses of 5mg and 10mg have been used for the treatment of sleep disturbance related to tuberous sclerosis complex (742;743). 3mg of melatonin has been taken daily for three months in children with epilepsy (744).
        • Surgical uses: 0.1mg/kg of melatonin has been used as an adjunct to 2-2.5mg/kg of paracetamol (acetaminophen), orally as a solution 40-45 minutes prior to anesthesia (745). 0.1, 0.25, or 0.5mg/kg of melatonin has been taken orally mixed with acetaminophen 15mg/kg (706). 0.05, 0.2, and 0.4mg/kg of melatonin have been taken (maximum dose: 20mg) in one or three doses about 45 minutes prior to anesthesia (707).
        • Urination (nocturia): 5mg of melatonin has been taken at 8 p.m. each evening for three months, and then the patients were observed for three months after treatment termination (746).

        Intravenous:

        • Pain: 10mg/kg of melatonin has been given intravenously as a single dose (552).

        Toxicology:

        • In human research, melatonin was found to be well tolerated and without grade 3/4 toxicity (747).
        • In a case report, bladder rupture resulting in death was reported after an intentional medication overdose, including melatonin, carisoprodol, ativan, and clonazepam, in a woman with a medical history of diabetes, depression with past suicide attempts, and suicidal ideation (748). The role of melatonin is not clear.
        • Psychotic symptoms have been reported in at least two cases, including hallucinations and paranoia, possibly due to overdose (371;462).
        • The LD50 in mice has been reported to be greater than 800mg/kg; in clinical trials, toxicity appeared to be minimal (749) and included mild effects such as diarrhea, headache, and abdominal cramps. A case of melatonin overdose was reported with ingestion of over 24mg for relaxation and sleep prior to surgery inducing symptoms of lethargy and disorientation (422). According to case reports, ataxia (difficulties with walking and balance) or disorientation may occur following melatonin overdose (83). Psychotic symptoms have been reported in at least two cases, possibly due to overdose, and included hallucinations and paranoia (371;462).
        • According to a letter to the editor, contaminants were found in over-the-counter meatonin products purchased in the Rochester, MN, area (750). There were seven contaminants, all detected at the 0.1-0.5% level of parent melatonin, and five were melatonin structural analog contaminants previously obtained from L-tryptophan made by Showa Denko K.K. Toxic effects from these contaminants had not been reported at time of writing. Contaminants in commercial preparations of melatonin have also been discussed separately (751).

        Precautions/Contraindications

        Allergy:

        • Avoid with known allergies to melatonin or related products. There are rare reports of allergic skin reactions after taking melatonin by mouth (382;550). Melatonin has been linked to a case of autoimmune hepatitis (394;430).

        Adverse Effects/Post-Market Surveillance:

        • General: Based on information from available studies and clinical use, melatonin is generally regarded as safe in recommended doses for short-term use (three months or less) (752). Similarly, the literature appears free of reported complications associated with long-term use of melatonin in children and adults, and it is even well-tolerated in neonates (117). Overall adverse effects or serious adverse effects are not significantly more common with melatonin than placebo, and are lacking in many clinical trials (371;374;377;388;389;424;425;444;445;567;568;587;590;592;605;626;627;632;648;666;667;671;675;685;691;694;723;724;735;736;745;753;754). The most commonly reported adverse events were headaches, dizziness, nausea, and drowsiness (388;393;419;431;436). However, case reports raise concerns about increased risk of seizure (449;453;454;455;456;457;458) and disorientation with overdose (372). Also, the timing of the melatonin dose appears important: if it is taken early in the day, it may cause sleepiness and delay adaptation to local time (if taken to mediate jet lag) and may diminish neurobehavioral performance (423). Case reports suggest that people with epilepsy and patients taking warfarin may experience harm from melatonin (371;372;400;429). In human research, melatonin doses of 0.4 and 2mg resulted in stable renal and liver function parameters after six weeks of use (755).
        • Cardiovascular: Hypotension or blood pressure lowering has occurred in both animal (756) and human research (359;360;361;362;363;364;365;366;369). However, hypertension was also reported in a clinical trial (367). There is some evidence of increases in cholesterol levels and atherosclerotic plaque buildup in human research (757) and both increases and decreases in cholesterol levels in animal research (758;759;760). Also, there are reports of decreased triglyceride and LDL cholesterol levels in human research (609;761;762). There are several rare or poorly described reports of abnormal heart rhythms, palpitations, fast heart rate, or chest pain, although in most cases, patients were taking other drugs that could account for these symptoms (370;371;372;373;374).
        • Dermatologic: Pruritus was reported in one patient receiving 2mg of melatonin for three weeks (379). Papular skin rash was also reported in a clinical trial (380). Side effects of melatonin reported in a clinical trial included flushing, red earlobes, and pallor (356;381), and itching or painful lumps on the skin (358).
        • Endocrine (hormonal effects): In clinical and laboratory studies, melatonin has been reported as producing varying hormonal effects. Such reports include changes in levels of luteinizing hormone (469;470;471;472;473;474;475;476), cortisol (477;478;479;480;481), progesterone (481;482;483;484;485), estradiol (482), thyroid hormone (T4 and T3) (486;487;488), testosterone (487;489), growth hormone (411;476;490;491;492;493;494), prolactin (411;495;496;497;498;499;500), oxytocin and vasopressin (490;501;502;503), adrenocorticotrophic hormone (478), and gonadotropin-inhibitory hormone (504). Melatonin has further been shown to alter pituitary hormone (LH and FSH) profiles in menopausal women to more "juvenile" profiles (488). In human research, in combination with estradiol treatment, melatonin reduced peak values of norepinephrine and increased epinephrine levels in some, but not all, stimulus situations (505;506;507). Effects on cortisol, norepinephrine, and epinephrine were lacking in some human research (625). In human research, during a heavy-resistance exercise session, melatonin increased the area under the curve of growth hormone (508). Variations may occur based on underlying patient characteristics. Gynecomastia (increased breast size) has been reported in men, as well as decreased sperm count (both which resolved with cessation of melatonin) (400). Decreased sperm motility has also been reported in rats (548) and humans (549). Inhibited ovarian function was reported in a clinical trial (391).
        • Endocrine (hyperglycemia): Elevated blood sugar levels (hyperglycemia) have been reported in patients with type 1 diabetes (insulin-dependent diabetes) (382;383), and low doses of melatonin have reduced glucose tolerance and insulin sensitivity (384;385). In patients with type 2 diabetes who had a suboptimal response to the oral hypoglycemic agent metformin, melatonin and zinc acetate administration improved impaired fasting and postprandial glycemic control and decreased the level of glycated hemoglobin (386;387). However, in other research, melatonin supplementation was found to lack an effect on measures of glucose homeostasis (763).
        • Gastrointestinal: Mild gastrointestinal distress has been reported in clinical trials and case reports, including nausea, vomiting, cramping, stomach pain, decreased appetite, an odd taste in the mouth, or diarrhea (354;355;357;358;380;381;382;388;389;390;391;392;393;764). Altered taste has also been noted (393). Melatonin has been linked to a case of autoimmune hepatitis (394) and with triggering of Crohn's disease symptoms (395). In human research, melatonin in combination with somatostatin, retinoids, vitamin D, bromocriptine, and cyclophosphamide resulted in mild diarrhea, nausea and vomiting, and drowsiness of grade 1-2 were reported, but the role of melatonin is unclear (396). A study in children with ADHD suffering from insomnia noted abnormal feces at a long-term follow-up (357;358). Increased appetite was noted in one study (397). In one study, a child administered melatonin showed severe reflux esophagitis (398). In animal research, high doses of melatonin have been shown to inhibit motility by interacting with serotonin and cholecystokinin-2 (CCK2) (765). It has been suggested in a review that melatonin may have a direct effect on bowel function, reducing gut contractions induced by serotonin and inhibiting proliferation of epithelium (766). In one participant, treatment with melatonin resulted in increased alkaline phosphatase levels (390).
        • Genitourinary: A systematic review of novel and emerging treatments for autism noted that an adverse effect associated with melatonin was increased enuresis; this has also been shown in clinical trials (353;354;355;356;357). Similarly, a study in children with ADHD suffering from insomnia noted bedwetting at a long-term follow-up (358). Decreased libido was noted in a clinical trial (438;438).
        • Hematologic (blood clotting abnormalities): It has been suggested that there might be a dose-response relationship between the plasma concentration of melatonin and coagulation activity (399). There are at least six reported cases of alterations in prothrombin time (a measurement of blood clotting ability) in patients taking both melatonin and the blood-thinning medication warfarin (Coumadin®) (372). These cases have noted decreases in prothrombin time (PT), which would tend to decrease the effects of warfarin and increase the risk of blood clots. However, blood clots have not been noted in these patients. Minor bleeding was noted in two of these cases (nosebleed and internal eye bleed). Nosebleeds have been reported in other human research (358). This may have been due to the blood-thinning effects of warfarin alone, without a correlation with melatonin use, or may possibly have been due to an interaction between melatonin and warfarin. Evidence pertaining to whether melatonin has effects on blood clotting in people who are not taking warfarin is currently lacking. Based on these reports, melatonin should be avoided in patients using warfarin and possibly in patients taking other blood-thinning medications or those with clotting disorders (400).
        • Musculoskeletal: According to case reports, ataxia (difficulties with walking and balance) may occur following melatonin overdose (372). In human research, melatonin lacked negative effects on postural stability (367). Compared to baseline, participants with chronic fatigue syndrome treated with melatonin showed a significant worsening of bodily pain (579). Weakened muscle power was reported in a clinical trial (598).
        • Neurologic (general): Commonly reported adverse effects include fatigue, dizziness, headache (including migraine), irritability, drowsiness, weakness, fogginess, yawning, nighttime awakening, poor sleep quality, vertigo, insomnia, and sleepiness (353;354;355;356;357;358;366;367;371;372;380;381;381;388;390;391;392;393;393;396;416;417;419;421;424;426;429;430;431;435;438;440;441;443;445;598;764). These symptoms are also indications of jet lag, and in some cases, causality may be unclear. Fatigue may occur with morning use or high doses (greater than 50mg) (415), and irregular sleep-wake cycles may occur (420). In a child, melatonin resulted in increased insomnia (398) and worsened behavior (355). One study reported that exogenous melatonin also may suppress the secretion of endogenous melatonin (670). A case report of severe migraine has also been noted (436). Disorientation, confusion, sleepwalking, vivid dreams, and nightmares have also been noted, with effects often resolving after cessation of melatonin (371;372;414;418;422;426;433;434;437;440;442;443). Nausea, vomiting, amnesia and somnambulia to the point of incapacitation, confusion, and morning sleepiness were also reported in patients treated with a combination of melatonin and zolpidem (432). Due to risk of daytime sleepiness, caution should be taken by those driving or operating heavy machinery (411;415;418;421;422). Exogenous melatonin may also cause decrements in mental performance, including a slowing of choice-reaction time (412;428), neurobehavioral performance (423), or learning (413). Other studies have failed to confirm a decrement in performance (767). Rarely, sleeping difficulties have also been reported (419). A systematic review of novel and emerging treatments for autism noted adverse effects associated with melatonin have included morning drowsiness, nighttime awakening, and excitement before going to sleep (353). Other reported adverse effects include feeling cold and hypothermia (356;390;392;445;446;447;448), although this did not occur in all studies (614), and hyperactivity (357;427). Although adverse effects were lacking in included studies, authors of a systematic review suggested that general side effects associated with melatonin may include grogginess the next day, irritability, and headache (444). Although reports of significant adverse events were lacking in a systematic review, the results collected indicated that sedation and impaired performance on psychomotor tests occurred (683). In a rodent model of Parkinson's disease, melatonin was found to potentiate neurodegeneration (768).
        • Neurologic (seizure risk): Melatonin may act as a proconvulsant (449;450) and may lower seizure threshold and increase the risk of seizure (451), particularly in children with severe neurologic disorders, and in adults with recurring symptoms following repeated melatonin administration (372;452). However, patients in these studies had no statistically significant exacerbations of seizure disorders requiring a discontinuation of melatonin therapy (452). Another study exploring the effect of melatonin on insomnia in children reported one case of mild generalized epilepsy developing four months after the start of the trial (431), and increased seizure risk has been reported in other clinical trials and systematic reviews (354). In contrast, multiple case reports indicated reduced incidence of seizure with regular melatonin use (453;454;455;456;457;458). This remains an area of controversy (449). Patients with seizure disorder taking melatonin should be monitored closely by a healthcare professional. In one study, the initial dose of melatonin used in the study (3.0mg) was lessened (1.5mg) following the observation of increased seizure frequency in two patients (380).
        • Ocular: It has been theorized that due to effects on photoreceptor renewal in the eye, high doses of melatonin may increase intraocular pressure and the risk of glaucoma, age-related maculopathy, and myopia (382), or retinal damage (400). However, there is preliminary human evidence that melatonin may actually decrease intraocular pressure in the eye, and it has been suggested as a possible therapy for glaucoma (366;460;461). In a case-control study of 100 patients with age-related macular degeneration, eight eyes showed more retinal bleeding and six eyes more retinal exudates during treatment with melatonin (459); however, the authors concluded that these outcomes were likely unrelated to melatonin administration. Human research has suggested that oral melatonin may reduce the function of retinal cones (769). In a clinical trial, a side effect of melatonin was red eyes (356).
        • Psychiatric: In human research, mood changes have been reported, including giddiness, dysphoria (sadness), mood dip, nervousness, hyperactivity, irritability, and transient depression (358;390;391;392;419;440;463). Psychotic symptoms have been reported in at least two cases, including hallucinations and paranoia, possibly due to overdose (371;462). Delusions and hallucinations have been reported in other clinical trials (441). An isolated incident of aggressiveness was also noted in a child diagnosed with ADHD and taking prescribed methylphenidate (433). Patients with underlying major depression or psychotic disorders taking melatonin should be monitored closely by a healthcare professional. In human research, melatonin attenuated the improvement of atypical symptoms and physical parameters of quality of life compared to placebo in subjects with weather-associated syndrome, positive type (635).
        • Respiratory: Melatonin lacked negative effects on sleep-disordered breathing in cardiac risk patients (770).
        • Other: In patients with rheumatoid arthritis, melatonin led to an increase in proinflammatory cytokines and erythrocyte sedimentation rate (464). In a clinical trial, melatonin was associated with increased side effects when compared with placebo (771) and adverse effects occurred in two children (431) or were few and minor (596;730); however, details are lacking.

        Precautions/Warnings/Contraindications:

        • Use cautiously in individuals with genitourinary disorders, such as children with a history of enuresis, due to reports of enuresis in clinical trials (353;354;355;356;357;358).
        • Use cautiously in aviation professionals, based on expert opinion suggesting that use of melatonin by aviation professionals may pose a safety risk for passengers (347).
        • Use cautiously in individuals with hypotension or hypertension or those using agents that affect blood pressure, based on reports in humans of both blood pressure reductions and increases (359;360;361;362;363;364;365;366;367;368;369).
        • Use cautiously in patients with heart disease, due to reports of abnormal heart rhythms, palpitations, fast heart rate, or chest pain, in patients usually also taking other drugs that may account for these symptoms (370;371;372;373;374), as well as reports of increased heart rate in some (368;375), but not all (376;377;378), studies.
        • Use cautiously in patients with skin disorders, due to reports of skin reactions (356;358;379;380;381).
        • Use cautiously in individuals with, or at risk of, diabetes, or those using hypoglycemic or hyperglycemic medications, based on reports of elevated blood sugar levels in patients with type 1 diabetes (382;383) and reduced glucose tolerance and insulin sensitivity (384;385), and due to research indicating that melatonin in combination with zinc improved postprandial glycemic control in patients with type 2 diabetes (386;387).
        • Use cautiously in individuals with gastrointestinal disorders, due to reports of gastrointestinal adverse events in clinical trials and case reports (354;355;357;358;380;381;382;388;389;390;391;392;393;394;395;396;397;398).
        • Use cautiously in individuals with blood clotting disorders or those using medications that affect blood clotting, based on reports of a dose-response relationship between the plasma concentration of melatonin and coagulation activity (399), case reports of alterations in prothrombin time (a measurement of blood clotting ability) in patients taking both melatonin and the blood-thinning medication warfarin (Coumadin®) (372), reports of nosebleeds (358), and experts suggesting that melatonin should be avoided in patients using warfarin and possibly in patients taking other blood-thinning medications or those with clotting disorders (400); also, in humans, melatonin was associated with lower plasma levels of procoagulant factors, and dose-response relationships between the plasma concentration of melatonin and coagulation activity have been hypothesized (399). In animal research, melatonin enhanced platelet responsiveness (401). In patients with decreased platelets due to cancer therapies, increased platelet counts after melatonin use have been observed (402;403;404;405;406;407;408), and cases of idiopathic thrombocytopenic purpura (ITP) treated with melatonin have been reported (409;410).
        • Use cautiously in individuals with musculoskeletal disorders, due to case reports of ataxia following overdose (372).
        • Use cautiously in individuals with neurological disorders, as well as individuals driving or operating heavy machinery or using sedatives, depressants or stimulants, and in individuals with hyperactitivity, due to reports of neurological adverse effects, such as fatigue, dizziness, headache (including migraine), irritability, drowsiness, weakness, fogginess, yawning, nighttime awakening, poor sleep quality, sleepiness, insomnia, hyperactivity, worsening behavior, and other effects (No year (372), 1970-1979 (381), 1980-1989 (411;412;413;414;415;416), 1990-1999 (391;393;398;417;418;419;420;421;422;423;424;425), 2000-2009 (353;354;355;356;357;358;366;367;371;380;388;390;392;396;426;427;428;429;430;431;432;433;434;435;436;437;438;439;440;441;442;443;444;445)).
        • Use cautiously in individuals who often feel cold, due to reports of feeling cold and hypothermia in clinical trials (356;390;392;445;446;447;448).
        • Use cautiously in individuals with or at risk of seizures, as melatonin may act as a proconvulsant (449;450) and may lower seizure threshold and increase the risk of seizure (451), particularly in children with severe neurologic disorders, and in adults with recurring symptoms following repeated melatonin administration (372;452), and due to reports of developing epilepsy (431) and increased seizure risk (354;380); however, due to reports of decreased seizures (453;454;455;456;457;458), this remains an area of controversy (449).
        • Use cautiously in individuals with glaucoma or other ocular disorders, due to human and laboratory reports of increased intraocular pressure (382) or retinal damage (400;459); however, there is preliminary human evidence that melatonin may actually decrease intraocular pressure in the eye, and it has been suggested as a possible therapy for glaucoma (366;460;461).
        • Use cautiously in individuals with psychiatric disorders, due to clinical reports of psychotic symptoms possibly due to overdose (371;462), as well as effects on mood changes (390;391;392;419;440;463), delusions and hallucinations (441), and aggressiveness (433).
        • Use cautiously in patients with inflammatory disorders or those using anti-inflammatory agents, due to reports of an increase in proinflammatory cytokines and erythrocyte sedimentation rate (464).
        • Use cautiously in patients that are breastfeeding, based on a lack of adequate safety evidence.
        • Use cautiously in individuals using agents metabolized by cytochrome P450, as melatonin appears to inhibit CYP1A2 (465;466;467) and induce CYP3A, and in animal research, melatonin inhibited the activity of cytochrome P450 2E1, but to a lesser extent than taurine (468).
        • Use cautiously in individuals using hormonal agents or with alterations in hormone levels, based on hormonal effects that have been shown in both human and animal studies (411;469;470;471;472;473;474;475;476;477;478;479;480;481;482;483;484;485;486;487;488;489;490;491;492;493;494;495;496;497;498;499;500;501;502;503;504;505;506;507;508).
        • Use cautiously in individuals with immune function disorders or those using immunomodulating agents, due to reports of interactions in human research with immune therapies, such as interferon (509), interleukin-2 (407;408;510;511;512;513;514;515;516;517;518;519;520;521;522;523;524;525;526;527;528), or tumor necrosis factor (525;529;530), as well as other effects on immune response and immune mediators in human and laboratory research (101;114;133;283;402;403;404;406;407;408;531;532;533;534;535;536;537;538;539;540;541;542;543;544;545).
        • Avoid in women who are pregnant or are attempting to become pregnant, based on hormonal effects that have been shown in both human and animal studies (411;469;470;471;472;473;474;475;476;477;478;479;480;481;482;483;484;485;486;487;488;489;490;491;492;493;494;495;496;497;498;499;500;501;502;503;504;505;506;507;508), including alterations of pituitary contractions (546), a possible increased risk of developmental disorders due to high levels of melatonin during pregnancy (547), decreased sperm count and motility (in human and animal research) (400;548;549), inhibited ovarian function (391), and decreased libido (438;438).
        • Avoid with known allergies to melatonin or related products.

        Pregnancy & Lactation:

        • The pineal hormone melatonin is synthesized from the amino acid tryptophan and is known to regulate sleep. Breast milk contains tryptophan, and a temporal relationship has been observed between the circadian rhythm of 6-sulfatoxymelatonin (a metabolite of melatonin excreted in the urine) of breastfed babies and that of tryptophan in the mother's milk (772).
        • Melatonin has been determined in the breast milk of nursing mothers; average levels at night were 23ng/L, with undetectable (<10ng/L) levels during the day (773). In a separate study, watching humorous videos increased breast milk levels of melatonin in both healthy mothers and mothers with eczema (20.8ng/L and 19.9ng/L, respectively) (774).
        • Avoid in women who are pregnant or are attempting to become pregnant, based on hormonal effects that have been shown in both human and animal studies (411;469;470;471;472;473;474;475;476;477;478;479;480;481;482;483;484;485;486;487;488;489;490;491;492;493;494;495;496;497;498;499;500;501;502;503;504;505;506;507;508) including alterations of pituitary contractions (546), and a possible increased risk of developmental disorders due to high levels of melatonin during pregnancy (547). Also, in human and animal research, effects of melatonin included decreased sperm count and motility (400;548;549), inhibited ovarian function (391), and decreased libido (438;438).
        • In animal studies, exogenous melatonin has been shown to improve placental efficiency and birthweight in undernourished pregnancies (209).
        • Beneficial effects of melatonin in preeclampsia were lacking (775).
        • According to the National Institute of Health's Lactation and Toxicology Database (LactMed), based on studies investigating the effect of melatonin on blood levels of melatonin in women, and an average peak breast milk concentration of 0.02mcg/L, the concentration of melatonin received by the infant would be much less than that used in clinical studies of neonates (10mg/kg).

        Interactions

        Most herbs and supplements have not been thoroughly tested for interactions with other herbs, supplements, drugs, or foods. The interactions listed below are based on reports in scientific publications, laboratory experiments, or traditional use. You should always read product labels. If you have a medical condition, or are taking other drugs, herbs, or supplements, you should speak with a qualified healthcare provider before starting a new therapy.

        Melatonin/Drug Interactions:

        • Note: This section discusses both endogenous and exogenous melatonin and the effects of other agents on melatonin and when taken concomitantly with melatonin. As a powerful antioxidant and immunomodulator, melatonin has been widely studied as a pharmacological means of mitigating oxidative damage caused by a number of substances. Specific mention of such interactions are generally omitted due to their positive effect.
        • Multiple drugs are reported to lower natural levels of melatonin in the body. It is not clear that there are any health hazards of lowered melatonin levels, or if replacing melatonin with supplements is beneficial. Examples of drugs that may reduce production or secretion of melatonin include nonsteroidal anti-inflammatory drugs (NSAIDs), such as ibuprofen (Motrin®, Advil®) or naproxen (Naprosyn®, Aleve®) (776;777); beta-blocker blood pressure medications, such as propranolol (Inderal®) (778), atenolol (Tenormin®) and metoprolol (Lopressor®, Toprol®) (779;780); and medications that reduce levels of vitamin B6 in the body, such as oral contraceptives, hormone replacement therapy, loop diuretics, hydralazine, and theophylline (781;782;783;784). Anesthesia using 7% sevoflurane decreased melatonin blood concentrations (785). However, using 5% isoflurane, blood levels of melatonin increased (785).
        • Melatonin is metabolized in the liver via the hepatic microsome cytochrome P450 system, primarily (but not exclusively) by the CYP2C19 and CYP1A family (particularly CYP1A2) and possibly CYP2C9. It appears to inhibit CYP1A2 and induce CYP3A. Thus, there are potential for interactions and altered levels of drugs and melatonin if used with agents that are substrates, inducers, or inhibitors of these isoenzymes.
        • Other agents that may alter synthesis or release of melatonin include caffeine (786;787), with a more pronounced effect in nonsmokers (788), diazepam (782;783), estradiol (789), vitamin B12 (790), verapamil (791), temazepam (792), and somatostatin (793).
        • Alcohol: In human research, alcohol consumption lacked effects on urinary levels of 6-sulfatoxymelatonin, a marker of melatonin (794).
        • Alzheimer's agents: Melatonin levels are often lower in patients with Alzheimer's disease (795;796;797;798;799;800). Some randomized controlled trials suggest a possible benefit of melatonin in patients with dementia (558;561). In vitro studies suggest a synergy between tacrine, a cholinesterase inhibitor, and melatonin (801).
        • Analgesics: In humans, melatonin use decreased the need for analgesics (366;605;681;682;683) and reduced levels of pain (552;596;682;683). However, compared to baseline, participants with chronic fatigue syndrome treated with melatonin showed a significant worsening of bodily pain (579).
        • Anesthetics: In human research melatonin augmented standard general anesthetics (677;679;680;802;803;804;805;806). However, not all trials have been positive (807). In vitro studies indicate that some anesthetics have also been found to alter blood melatonin concentrations in humans (isoflurane increasing and sevoflurane decreasing) (785). Plasma levels of melatonin increased during administration of propofol in humans (802) as well as in rats (808). In humans, melatonin premedication significantly decreased the doses of both propofol and thiopental required to induce anesthesia (679;680).
        • Angiotensin-converting enzyme (ACE) inhibitors: In human research, melatonin normalized ACE in six patients with high levels at baseline (531).
        • Antiaging agents: Melatonin has been identified as countering some of the deleterious effects of aging in human, animal, and in vitro research (14;15;16;17;18;19;20;21;22;23).
        • Antiarthritics: Based on mechanisms of action in vitro, melatonin has been suggested as possibly playing a beneficial role in osteoarthritis (60) and other rheumatic diseases (59).
        • Antiasthmatics: Asthmatics were found to have lower levels of endogenous melatonin (809;810); however, elevated levels at night were associated with worsening of symptoms (811;812).
        • Anticoagulants and antiplatelets: According to experts, melatonin may decrease prothrombin time (a measurement of blood clotting ability) (372;400). In humans, melatonin was associated with lower plasma levels of procoagulant factors, and dose-response relationships between the plasma concentration of melatonin and coagulation activity have been hypothesized (399). In animal research, melatonin enhanced platelet responsiveness (401). Increased platelet counts after melatonin use have been observed in patients with decreased platelets due to cancer therapies (402;403;404;405;406;407;408), and cases of idiopathic thrombocytopenic purpura (ITP) treated with melatonin have been reported (409;410).
        • Anticonvulsants: It has been suggested that melatonin may act as a proconvulsant (449) and may lower seizure threshold and increase the risk of seizure, particularly in children with severe neurologic disorders (372;451;452). In a study exploring the effect of melatonin on insomnia in children, a reported case of mild generalized epilepsy developing four months after the start of the trial was noted; the child was initiated on valproate, and although melatonin was not discontinued, further seizures were lacking (431). In contrast, several case reports indicated reduced incidence of seizure with regular melatonin use (453;454;455;456;457;458). Both anticonvulsant (458;813;814;815) and proconvulsant (449) properties have been associated with melatonin in preclinical studies. This remains an area of controversy (449). Increases in the anticonvulsant effects of valproate have been observed in mice (458;813). In human research, add-on melatonin administration in epileptic children did not alter valproate serum concentrations (816); however, treatment with valproate and carbamazepine increased urinary 6-sulfatoxymelatonin, a marker of melatonin, which had decreased during the period of epileptic seizures (817). In one human study, valproate decreased the sensitivity of melatonin to light in patients with bipolar disorder (818).
        • Antidepressants: In human research, antidepressants (fluoxetine, fluvoxamine, duloxetine, and Hypericum perforatum) increased melatonin and 6-hydroxymelatonin (metabolite) levels, and fluvoxamine both increased melatonin bioavailability and decreased melatonin metabolism (819;820;821;822;823). In human research, concurrent use of fluvoxamine and melatonin resulted in increased levels of melatonin, likely due to reduced metabolism of melatonin by inhibiting CYP1A2 and/or CYP2C9 (465;466;467). Venlafaxine lacked effects on nocturnal melatonin concentrations in a human study (824). Commonly reported adverse effects of melatonin in clinical trials include fatigue, dizziness, headache (including migraine), irritability, drowsiness, weakness, fogginess, yawning, nighttime awakening, poor sleep quality, vertigo, insomnia, and sleepiness (353;354;355;356;357;358;366;367;371;372;380;381;388;390;391;392;393;396;416;417;419;421;424;426;429;430;431;435;438;440;441;443;445;598;764). These symptoms are also indications of jet lag, and in some cases, causality may be unclear. In human research, mood changes have been reported, including giddiness, dysphoria (sadness), mood dip, nervousness, hyperactivity, irritability, and transient depression (357;358;390;391;392;419;427;440;463). Psychotic symptoms have also been reported in human research, including hallucinations, delusions, and paranoia, possibly due to overdose (371;441;462).
        • Antidiabetics: Elevated blood sugar levels (hyperglycemia) have been reported in patients with type 1 diabetes (insulin-dependent diabetes) (382;383), and low doses of melatonin have reduced glucose tolerance and insulin sensitivity (384;385). In patients with type 2 diabetes mellitus who had a suboptimal response to the oral hypoglycemic agent metformin, melatonin and zinc acetate administration improved impaired fasting and postprandial glycemic control and decreased the level of glycated hemoglobin (386;387). However, in other research, melatonin supplementation was found to lack an effect on measures of glucose homeostasis (763).
        • Antihypertensives: In animal and human research, hypotension, blood-pressure lowering effects, and hypertension have been reported (359;360;361;362;363;364;365;366;367;369;443;506;609;756;825;826;827;828;829), although melatonin did not alter blood pressure in some animal or human research (376;830) or had mixed effects on night and day blood pressure, with decreases at night (375;377;378;608). In human research, suppression of nocturnal melatonin secretion with atenolol (a beta1-adrenoreceptor antagonist) increased total wake time and decreased REM and slow-wave sleep; these effects were reversed if melatonin was given after the antagonist (64). Serum melatonin levels decreased noticeably with propranolol treatment (778). In animals, melatonin reduced the effects of the alpha-adrenergic agonist clonidine (756). In contrast, in humans, blood pressure increases have been observed when 5mg of melatonin was taken at the same time as the calcium-channel blocker nifedipine (368;465). Verapamil increased urinary melatonin excretion significantly (by 67%), but left excretion of 6-sulphatoxy-melatonin unaffected in healthy adults infused with calcium as a model for hyperkalemia (791).
        • Anti-inflammatories: In human research, melatonin had anti-inflammatory effects in infants with respiratory distress (11), decreased the upregulation of proinflammatory cytokines in laboratory and human research (47;62;101;109;115;761;831;832), and inhibited nitric oxide (NO) and malondialdehyde (MDA) production and increased glutathione levels (833;834). However, there is conflicting evidence from human trials, where melatonin induced a proinflammatory response, increasing levels of certain inflammatory cytokines (p>0.05), as well as plasma kynurenine concentrations (p<0.05) in individuals with rheumatoid arthritis (464). Also, in human research, melatonin lacked effects on C-reactive protein (CRP) levels (835).
        • Antilipemics: There is some evidence of increases in cholesterol levels and atherosclerotic plaque buildup in human research (757) and animal research (759;760). In contrast, there are also reports of decreases in cholesterol levels in animal research (758) and decreased triglyceride and LDL cholesterol levels in human research (609;761;762).
        • Antineoplastics: According to the "melatonin hypothesis" of cancer, the exposure to light at night and anthropogenic electric and magnetic fields may be related to the increased incidence of cancer and childhood leukemia via melatonin disruption (836). Based on theoretical antioxidant mechanisms and in human research, melatonin has anticarcinogenic effects (351;402;403;521;571;573;575;837;838;839;840;841;841;842;843;844;845;846;847;848;849;850;851;852;853;854;855;856;857;858;859;860;861). Melatonin has been combined with other types of treatment, including chemotherapies (such as cisplatin, etoposide, or irinotecan) (351;403;404;405;406;528;850;854;862;863;864;865;866), COX-2 inhibitors (867), or immune therapies, such as interferon (509), interleukin (IL)-2 or IL-12 (407;408;510;511;512;513;514;515;516;517;518;519;520;521;522;523;524;525;526;527;868), or tumor necrosis factor (525;529;530). A number of studies have established melatonin's ability to prevent or mitigate damage from a number of chemical sources including (but not limited to) the following: methamphetamines (57;869), organophosphorus compounds (258;259;260), alcohol (261;870), nicotine (300), beta-cyfluthrin (871), and benzo(a)pyrene (872). Results of a meta-analysis of clinical trials suggested that melatonin had a significant effect on tumor remission and the one-year survival rate, as well as an ability to decrease side effects related to radiochemotherapy, including thrombocytopenia, fatigue, and neurotoxicity (577).
        • Antiobesity agents: In laboratory research, melatonin inhibited adipocyte differentiation (193) and reduced gut motility (873). Other animal research has indicated that exogenous melatonin, however, lacks effect on leptin secretion (192). In patients with type 2 diabetes, nocturnal plasma melatonin levels were higher in obese subjects vs. nonobese subjects and lean nondiabetic controls (874).
        • Antiparasitics: In animal research, melatonin therapy controlled Trypanosoma cruzi proliferation by stimulating the host's immune response (203;875).
        • Antiparkinson agents: In human research, melatonin lacked an effect on signs of parkinsonism or levodopa effects, although it was well tolerated, but with side effects such as skin flushing, diarrhea, abdominal cramps, somnolence during the day, scotoma lucidum, and headaches (381).
        • Antipsychotics: Chronic treatment with antipsychotic drugs significantly improved psychotic symptomatology in schizophrenics, but did not change the secretory pattern of melatonin (876). The increase in melatonin secretion, which occurs with the initiation of neuroleptic therapy, may be responsible for the delay in the antipsychotic effects of neuroleptics and may also account for the lag in the development of drug-induced parkinsonism, as well as its disappearance (877). Preliminary human and laboratory reports suggest that melatonin had mixed effects on mood, sleep, and tardive dyskinesia in patients with schizophrenia, often treated with haloperidol (634;671;673;684;878;879;880;881;882). In human research, quetiapine did not appear to alter melatonin levels (883).
        • Anti-ulcer agents: In human research, melatonin improved the healing of ulcers (692;693).
        • Antivirals: In animal research, the protective effect of melatonin against Venezuelan equine encephalomyelitis virus was likely mediated by melatonin receptor activation (884).
        • Anxiolytics: In humans, melatonin has been widely reported as having general and synergistic anxiolytic effects (563;677;679;680;803;804;805;806); however, evidence is mixed from a systematic review and well-designed clinical trials with respect to melatonin for anxiety prevention during surgery (678;681;683).
        • Benzodiazepines: A small amount of research has examined the use of melatonin to assist with tapering or cessation of benzodiazepines such as diazepam (Valium®) or lorazepam (Ativan®), and in general, results are promising (565;567;568;569). Although melatonin has demonstrated effectiveness in reducing benzodiazepine consumption in older patients with established insomnia (565), low doses of immediate release melatonin (3mg) lacked usefulness for benzodiazepine tapering in older patients with minor sleep disturbances (565). In human research, melatonin was found to improve the quality of sleep in combination with benzodiazepines (885).
        • Caffeine: Caffeine is reported to raise natural melatonin levels in the body (787) with a more pronounced effect in nonsmokers (788), possibly due to effects on the liver enzyme cytochrome P450 1A2 (886). It has been proposed that caffeine may increase the bioavailability of endogenous melatonin (887). Caffeine may also alter circadian rhythms in humans, with effects on melatonin secretion (788). It has been reported that caffeine reduced the onset of nighttime melatonin levels for women in the luteal phase, but had little effect on melatonin levels for oral contraceptive users (888). Another human study has shown that a single dose of 200mg of caffeine reduced natural melatonin levels (786), though a more recent human study using a twice-daily dose of 200mg of caffeine over seven days found a lack of effect on nighttime salivary melatonin (889).
        • Calcium channel blockers: In human research, melatonin increased blood pressure in patients treated with the calcium channel blocker nifedipine (368). Verapamil increased urinary melatonin excretion significantly (by 67%), but left excretion of 6-sulphatoxy-melatonin unaffected in healthy adults infused with calcium as a model for hyperkalemia (791). A review of the role of melatonin in the pathology of the cardiovascular system noted that further evaluation of the clinical safety and efficacy of melatonin as an antihypertensive therapy is necessary and that such study must take into account melatonin's antagonism of calcium channel inhibitors (890).
        • Cardiovascular agents: In human research, low levels of platelet melatonin was found to be associated with angiographic no-reflow after primary percutaneous coronary intervention in patients with ST-segment elevation myocardial infarction (891). It has been proposed that melatonin acts directly on the cardiovascular system rather than modulating cardiac autonomic activity (892). In a poor-quality study, the inclusion of melatonin in the combined treatment of cardiovascular disease resulted in anti-ischemic, antianginal, antioxidant, and hypotensive effects (369). There is some evidence of increases in cholesterol levels and atherosclerotic plaque buildup in human research (757) and animal research (759;760). In contrast, there are also reports of decreases in cholesterol levels in animal research (758) and decreased triglyceride and LDL cholesterol levels in human research (609;761;762). In animal and human research, hypotension, blood pressure-lowering effects, and hypertension have been reported (359;360;361;362;363;364;365;366;367;369;443;506;609;756;825;826;827;828;829), although melatonin did not alter blood pressure in some animal or human research (376;830) or had mixed effects on night and day blood pressure, with decreases at night (375;377;378;608).
        • CNS depressants: In theory, based on possible risk of daytime sleepiness (67;354;411;415;418;421;422;435;667) and reported negative effects on certain cognitive tasks in humans in some, but not all, studies (412;413;428;767;893;894), melatonin may exacerbate the amount of drowsiness and reduced mental acuity caused by CNS depressants. Increased daytime drowsiness was reported when melatonin was used at the same time as the prescription sleep aid zolpidem (Ambien®), although it is not clear that effects were greater than with the use of zolpidem alone (114). In human research, an effect of remifentanil on melatonin concentration and an effect of melatonin on remifentanil-induced sleep disturbance were lacking (439).
        • CNS stimulants: In human research, there was an isolated case of aggression in a child diagnosed with ADHD and taking prescribed methylphenidate (433). In animal research, melatonin increased the adverse effects of methamphetamine on the nervous system (895). Melatonin has been implicated as having dosing time-dependent effects on the action of psychostimulant drugs such as cocaine and amphetamines (896).
        • Cognitive agents: In human research, exogenous melatonin caused decrements in performance, including a slowing of choice-reaction time (412;428) or learning (413); however, some studies have failed to confirm a decrement in performance (767;893;894), including a study of high-dose melatonin (50mg) in elderly persons (mean age: 84.5 years) (897).
        • Contraceptives: In patients undergoing in vitro fertilization embryo transfer (IVF-ET), although melatonin benefited oocyte maturation, effects on fertilization and pregnancy were lacking (594;898). Similarly, melatonin has been shown to improve viability of sperm (899) and embryos (118;900;901;902;903) produced with in vitro fertilization. In animal research, reproductive effects of melatonin have also been found (280;487;904;905;906;907;908;909;910).
        • Cytochrome P450-metabolized agents: Melatonin is metabolized in the liver via the hepatic microsome cytochrome P450 system, primarily (but not exclusively) by the CYP2C19 and CYP1A family (particularly CYP1A2) (911;912) and possibly CYP2C9. It appears to inhibit CYP1A2 (465;466;467) and induce CYP3A. In human research, concurrent use of fluvoxamine and melatonin resulted in increased levels of melatonin, likely due to reduced metabolism of melatonin by inhibiting CYP1A2 and/or CYP2C9 (465;466;467). Caffeine is reported to raise natural melatonin levels in the body (787) with a more pronounced effect in nonsmokers (788), possibly due to effects on the liver enzyme cytochrome P450 1A2 (788). This effect was more pronounced in nonsmokers (788). Other human studies suggest that interactions between exogenous melatonin and substrates metabolized by CYP1A2 may differ in individuals before and after smoking abstinence (913). In animal research, melatonin inhibited the activity of cytochrome P450 2E1, but to a lesser extent than taurine (468).
        • Dental agents: In human research, salivary and gingival crevicular fluid melatonin levels were lower in individuals with periodontal disease (914).
        • Dermatologic agents: Dermatologic use of melatonin has been proposed because of its immunomodulatory and antioxidant abilities. Research findings indicate that melatonin accumulates in the stratum corneum (709). In human research, free radical scavenging was suggested as a possible mechanism of action in the protection against UV-induced erythema (711).
        • Dextromethorphan: In animal research, dextromethorphan interacted synergistically with melatonin in relieving neuropathic pain (188).
        • Diuretics: In clinical trials, an adverse effect associated with melatonin was increased enuresis (353;354;355;356;357). A study in children with ADHD suffering from insomnia noted bedwetting at a long-term follow-up (358).
        • Drugs that affect GABA: In animal research, results suggested a possible role of the GABAergic system in melatonin's effects (915). In human research, melatonin was found to potentiate the effects of gamma-amino butyric acid (GABA) (885).
        • Drugs that may lower seizure threshold: It has been suggested that melatonin may act as a proconvulsant (449) and may lower seizure threshold and increase the risk of seizure, particularly in children with severe neurologic disorders (372;451;452). In a study exploring the effect of melatonin on insomnia in children, a reported case of mild generalized epilepsy developing four months after the start of the trial was noted; the child was initiated on valproate, and although melatonin was not discontinued, further seizures were lacking (431). In contrast, several case reports indicated reduced incidence of seizure with regular melatonin use (453;454;455;456;457;458). Both anticonvulsant (458;813;814;815) and proconvulsant (449) properties have been associated with melatonin in preclinical studies. This remains an area of controversy (449).
        • Drugs used for osteoporosis: In laboratory research, melatonin impaired osteoclast activity and bone resorption (916;917;918). In human research, melatonin lacked effects on bone density, N-terminal telopeptide (NTX), or osteocalcin (OC), although the NTX:OC ratio in the melatonin group was reduced (628).
        • Epithalamin: In human research, epithalamin normalized the circadian rhythm of melatonin (919).
        • Estrogens: Human and laboratory studies have suggested that melatonin mimics the effect of drugs that act through the estrogen receptor, interfering with the effects of endogenous estrogens, as well as those that interfere with the synthesis of estrogens by inhibiting the enzymes controlling the interconversion from their androgenic precursors (920). Mechanisms of melatonin's oncostatic action may include regulation of estrogen receptor expression and transactivation (921) and antiestrogenic effects (922;923;924). MCF-7 human breast cancer cultured cells have been reported as melatonin sensitive, as well as estrogen receptor positive and estrogen responsive (925), although this finding was not confirmed in a subsequent study (926). Melatonin has been reported to elicit an increase in estrogen receptor activity in breast tumors (927). Low plasma melatonin concentrations were associated with greater amounts of estrogen or progesterone receptors on primary tumors (928). In a review on the anticarcinogenic role of melatonin, potential mechanisms included the inhibition of initiation and growth of hormone-dependent tumors by decreasing the expression of estrogen receptors, as well as aromatase activity, resulting in the inhibition of cancer cell proliferation, a decrease in oxidative stress, and an increase in the activity of the immune system (929).
        • Exercise performance agents: In human research, during a heavy-resistance exercise session, melatonin increased the area under the curve of growth hormone (508) and protected against the overexpression of inflammatory mediators and inhibited the expression of proinflammatory cytokines in exercising individuals (930).
        • Fertility agents: In patients undergoing in vitro fertilization embryo transfer (IVF-ET), although melatonin benefited oocyte maturation, effects on fertilization and pregnancy were lacking (594;898). Similarly, melatonin has been shown to improve viability of sperm (899) and embryos (118;900;901;902;903) produced with in vitro fertilization. In animal research, reproductive effects of melatonin have also been found (280;487;904;905;906;907;908;909;910).
        • Flumazenil: In hamsters, the administration of the benzodiazepine antagonist flumazenil blunted the activity of melatonin in these behaviors (931).
        • Gastrointestinal agents: Preliminary research has indicated that melatonin aids symptoms of functional dyspepsia (598), gastroesophageal reflux disease (GERD) (602), Crohn's disease and ulcerative colitis (932), and irritable bowel syndrome (mixed evidence) (438;599;600;601;603;933).
        • Genitourinary tract agents: In clinical trials, an adverse effect associated with melatonin was increased enuresis (353;354;355;356;357). A study in children with ADHD suffering from insomnia noted bedwetting at a long-term follow-up (358).
        • Glaucoma agents: Preliminary human evidence suggests that melatonin may decrease intraocular pressure in the eye (366;460;461); however, according to reviews, high doses of melatonin may increase intraocular pressure and the risk of glaucoma, age-related maculopathy, and myopia (382), as well as retinal damage (400).
        • Headache agents: Evidence is mixed from human research with respect to preventive effects of melatonin on headaches, including migraines (146;440;604;605;934;935;936;937;938;939;940).
        • Heart rate regulating agents: Melatonin has been shown to increase heart rate when administered in patients taking nifedipine (a calcium channel blocker antihypertensive) (368) and in other studies (375); however, effects were lacking in other human research (376;377;378). When measured in the morning, the relationship between salivary melatonin and exercise-induced heart rate changes was steeper than when measured in the evening (941). Clinical significance is unclear. There are several rare or poorly described reports of abnormal heart rhythms, palpitations, fast heart rate, or chest pain, although in most cases, patients were taking other drugs that may account for these symptoms (370;371;372;373;374).
        • Hepatotoxins: In patients with nonalcoholic steatohepatitis (NASH), use of melatonin resulted in improvements in liver function (445). In patients with steatohepatitis, melatonin decreased levels of proinflammatory cytokines, triglycerides, and GGTP (761). In human research, melatonin resulted in stable renal and liver function parameters after six weeks of use (755). Decreased transaminases have been shown in other human research (754). However, in one participant, treatment with melatonin resulted in increased alkaline phosphatase levels (390).
        • Hormonal agents: In humans, hormone replacement therapy (HRT) is reported to cause a decrease in daily melatonin secretion without disturbing circadian rhythm (942;943). In clinical and laboratory studies, melatonin has also been reported as producing varying hormonal effects. Such reports include changes in levels of luteinizing hormone (469;470;471;472;473;474;475;476), cortisol (477;478;479;480;481), progesterone (481;482;483;484;485), estradiol (482), thyroid hormone (T4 and T3) (486;487;488), testosterone (487;489), growth hormone (411;476;490;491;492;493;494), prolactin (411;495;496;497;498;499;500), oxytocin and vasopressin (490;501;502;503), adrenocorticotrophic hormone (478), and gonadotropin-inhibitory hormone (504). Melatonin has further been shown to alter pituitary hormone (LH and FSH) profiles in menopausal women to more "juvenile" profiles (488). In clinical trials, melatonin affected hormone levels in patients with hormonal-related cancers and had synergistic effects with tamoxifen (850;862;863). Other human studies report a lack of significant hormonal effects (496;617;944;945). Gynecomastia (increased breast size) has been reported in men, as well as decreased sperm count (both which resolved with cessation of melatonin) (400). Decreased sperm motility has also been reported in rats (548) and humans (549). Other human and laboratory studies have suggested that melatonin mimics the effect of drugs that act through the estrogen receptor interfering with the effects of endogenous estrogens, as well as those that interfere with the synthesis of estrogens by inhibiting the enzymes controlling the interconversion from their androgenic precursors (920). In females, blood pressure decreased only in hormone replacement therapy or birth control users and not nonusers (506;507). In human research, progesterone modulated melatonin secretion in postmenopausal women (946). In human research, in combination with estradiol treatment, melatonin reduced peak values of norepinephrine and increased epinephrine levels in some, but not all, stimulus situations (505;506;507). In human research, during a heavy-resistance exercise session, melatonin increased the area under the curve of growth hormone (508).
        • Immunosuppressants: In human research, melatonin was found to interact positively with immune therapies, such as interferon (509), interleukin-2 (407;408;510;511;512;513;514;515;516;517;518;519;520;521;522;523;524;525;526;527;528), or tumor necrosis factor (525;529;530). Based on limited human research, researchers concluded that melatonin may be an effective treatment for sarcoidosis (531). Exogenous melatonin has been shown to enhance immune response following veterinary vaccination (532). Researchers noted increased platelet counts after melatonin use in patients with decreased platelets due to cancer chemotherapy (402;403;404;406;407;408;533). According to a review, activation of melatonin receptors was associated with the release of cytokines by type 1 T-helper cells (Th1), including gamma-interferon (gamma-IFN) and IL-2, as well as novel opioid cytokines (534). Melatonin has been reported to promote neutrophil apoptosis in patients receiving hepatectomy involving ischemia and reperfusion of the liver (283;535;536;537). A combination hormone therapy including melatonin was found to improve leucocyte function in ovariectomized aged rats (538). In laboratory research, melatonin suppressed TNF-alpha, IL-1 beta, and IL-6 (101); inhibited Th1 cells (114); stimulated humoral activity and antibody production (532;539;540); inhibited NF-kappaB (541), as well as IKK, and JNK pathways (133); prevented T cell apoptosis (542); and stimulated mononuclear cell production (543). In human research, combined therapy with low-dose subcutaneous IL-2 and melatonin improved the mean number of lymphocytes, eosinophils, T lymphocytes, natural killer (NK) cells, and CD25- and DR-positive lymphocytes, and increased the mean CD4:CD8 ratio (544). In cancer patients who achieved disease control, melatonin induced a decrease in the number of regulatory T lymphocytes; this change was lacking in individuals with progressed disease (545).
        • Isoniazid: In vitro, melatonin increased the effects of isoniazid against Mycobacterium tuberculosis (263).
        • Lithium: In human research, lithium had a significant effect on sensitivity to light but not on overall melatonin synthesis (947).
        • Magnetic fields: It has been theorized that chronic exposure to magnetic fields or recurrent cellular telephone use may alter melatonin levels and circadian rhythms. However, several studies suggest that this is not the case (948;949;950;951). Melatonin was shown to reduce the effects of lipid peroxidation, less effectively than vitamin E, in rats exposed to static magnetic fields under laboratory conditions (952).
        • Methamphetamines: In human research, there was an isolated case of aggression in a child diagnosed with ADHD and taking prescribed methylphenidate (433). In animal research, melatonin increased the adverse effects of methamphetamine on the nervous system (895).
        • Methoxamine: In animals, melatonin reduced the effects of the alpha-adrenergic agonist methoxamine (756).
        • Musculoskeletal agents: According to case reports, ataxia (difficulties with walking and balance) may occur following melatonin overdose (372). In human research, melatonin lacked negative effects on postural stability (367). Compared to baseline, participants with chronic fatigue syndrome treated with melatonin showed a significant worsening of bodily pain (579). Weakened muscle power was reported in a clinical trial (598).
        • Neurologic agents: It has been proposed that melatonin may reduce the amount of neurologic damage patients experience after stroke, based on antioxidant properties (953;954;955;956;957;958;959;960). A significant body of basic research has indicated that melatonin may possess neuroprotective properties (79;167;168;169;170;171;172;173;174;175;176;177;179;180;181;182;183;184;185;186;187), meriting reviews in the contexts of neurodegenerative diseases (961), the peripheral nervous system (962), and traumatic nervous system injury (80). However, commonly reported adverse effects of melatonin in clinical trials include fatigue, dizziness, headache (including migraine), irritability, drowsiness, weakness, fogginess, yawning, nighttime awakening, poor sleep quality, vertigo, insomnia, and sleepiness (353;354;355;356;357;358;366;367;371;372;380;381;388;390;391;392;393;396;416;417;419;421;424;426;429;430;431;435;438;440;441;443;445;598;764). These symptoms are also indications of jet lag, and in some cases, causality may be unclear. In human research, mood changes have been reported, including giddiness, dysphoria (sadness), mood dip, nervousness, hyperactivity, irritability, and transient depression (357;358;390;391;392;419;427;440;463). Psychotic symptoms have also been reported in human research, including hallucinations, delusions, and paranoia, possibly due to overdose (371;441;462).
        • Neuromuscular blockers: In laboratory research, melatonin increased the neuromuscular blocking effect of the muscle relaxant succinylcholine, but not vecuronium (963).
        • Ophthalmic agents: In limited human research, melatonin stabilized vision in patients suffering from age-related macular degeneration (459). Preliminary human evidence also suggests that melatonin may decrease intraocular pressure in the eye (366;460;461); however, according to reviews, high doses of melatonin may increase intraocular pressure and the risk of glaucoma, age-related maculopathy, and myopia (382), as well as retinal damage (400). Use of transitions lenses as part of chromotherapy for macular degeneration were found to maintain the physiological balance of melatonin (964). In human research, use of eye masks increased melatonin levels (965).
        • Opioids: In animals, researchers have concluded that melatonin acutely reversed and prevented tolerance to and dependence on morphine (966;967) and reduced the incidence of naloxone-induced withdrawal (967). In human research, melatonin reduced the need for morphine (563;681).
        • Otic agents: In human research, melatonin attenuated the muscle sympathetic nerve activity (vestibulosympathetic reflex) response to baroreceptor unloading while lacking effects on the vestibulocollic reflexes (968). In human research, use of ear plugs increased melatonin levels (965).
        • Radioprotective drugs: Due to its well-known antioxidant properties, it has been suggested that melatonin possesses a protective effect against damage caused by ionizing radiation, a hypothesis borne out of preliminary animal and in vitro research (206;207;215;216;217;219;969). Melatonin has been shown to ameliorate oxidative injury due to ionizing radiation in vitro (214;970;971). The specific mechanisms may involve downregulation of apoptotic pathways via control of oxidative load (972).
        • Remifentanil: In human research, remifentanil did not decrease melatonin concentration (439). Melatonin administration also did not prevent remifentanil-induced sleep disturbance.
        • Renally eliminated agents: In human research, melatonin resulted in decreased renal blood flow velocity and conductance (376). In human research, melatonin resulted in stable renal function parameters after six weeks of use (755).
        • Respiratory agents: In a clinical trial, melatonin reduced dyspnea; however, changes in lung function were lacking (580).
        • Sedatives: In human research, melatonin has been shown to decrease sleep latency (390;752) and benefit sleep quality and duration in children, older and younger adults, individuals with disabilities, and visually impaired individuals (322;354;355;356;357;373;374;379;389;391;392;431;446;581;613;615;638;659;734). In human research, exogenous melatonin exerted hypnotic effects primarily when circulating levels of endogenous melatonin were low (653), and even very low doses caused sleep in some studies when ingested before endogenous melatonin onset (418;649;655;661;662). Also, in human research, melatonin has been shown to decrease the amount of anesthesia required during surgery (679;680;719;973).
        • Sevoflurane: In human research, sevoflurane resulted in a reduction of postoperative plasma melatonin levels (974).
        • Tacrine: In vitro studies suggest the possibility of a synergy between tacrine, a cholinesterase inhibitor, and melatonin, based on mechanism of action (801).
        • Thermoregulating agents: In human research, hypothermic effects of melatonin have been reported with doses from 15mg (427;657;975;976), and ingestion of 1.6mg of melatonin was reported to result in approximately 0.4ºC decrease of body temperature in humans (975;977;978;979). Reports of feeling cold or hypothermia exist in other clinical literature (356;390;392;445), but not in all studies (614).
        • Thyroid hormones: In human research, thyroid-stimulating hormone (TSH) serum levels were lower and those of free thyroxine (FT4) were increased at night when endogenous melatonin levels were higher (980). In clinical and laboratory studies, melatonin has also been reported to change levels of thyroid hormone (T4 and T3) (486;487;488).
        • Valproic acid: It has been suggested that melatonin may act as a proconvulsant (449) and may lower seizure threshold and increase the risk of seizure, particularly in children with severe neurologic disorders (372;451;452). In a study exploring the effect of melatonin on insomnia in children, a reported case of mild generalized epilepsy developing four months after the start of the trial was noted; the child was initiated on valproate, and although melatonin was not discontinued, further seizures were lacking (431). In contrast, several case reports indicated reduced incidence of seizure with regular melatonin use (453;454;455;456;457;458). Both anticonvulsant (458;813;814;815) and proconvulsant (449) properties have been associated with melatonin in preclinical studies. This remains an area of controversy (449). Increases in the anticonvulsant effects of valproate have been observed in mice (458;813). In human research, add-on melatonin administration in epileptic children did not alter valproate serum concentrations (816); however, treatment with valproate and carbamazepine increased urinary 6-sulfatoxymelatonin, a marker of melatonin, which had been decreased during the period of epileptic seizures (817). In one human study, valproate decreased the sensitivity of melatonin to light in patients with bipolar disorder (818).
        • Vaccines: Exogenous melatonin has been shown to enhance immune response following veterinary vaccination (532).
        • Vasodilators: In healthy male volunteers, melatonin significantly increased peripheral blood flow, as measured by distal to proximal skin temperature gradient and finger pulse volume (981). In human research, melatonin resulted in decreased renal blood flow velocity and conductance , increased forearm blood flow and vascular conductance, and lacked an effect on cerebral blood flow (376).

        Melatonin/Herb/Supplement Interactions:

        • Note: This section discusses both endogenous and exogenous melatonin and the effects of other agents on melatonin and when taken concomitantly with melatonin. As a powerful antioxidant and immunomodulator, melatonin has been widely studied as a pharmacological means of mitigating oxidative damage caused by a number of substances. Specific mention of such interactions are generally omitted due to their positive effect.
        • Multiple drugs are reported to lower natural levels of melatonin in the body. It is not clear that there are any health hazards of lowered melatonin levels, or if replacing melatonin with supplements is beneficial. Examples of drugs that may reduce production or secretion of melatonin include nonsteroidal anti-inflammatory drugs (NSAIDs), such as ibuprofen (Motrin®, Advil®) or naproxen (Naprosyn®, Aleve®) (776;777); beta-blocker blood pressure medications, such as propranolol (Inderal®) (778), atenolol (Tenormin®) and metoprolol (Lopressor®, Toprol®) (779;780); and medications that reduce levels of vitamin B6 in the body, such as oral contraceptives, hormone replacement therapy, loop diuretics, hydralazine, and theophylline (781;782;783;784). Anesthesia using 7% sevoflurane decreased melatonin blood concentrations (785). However, using 5% isoflurane, blood levels of melatonin increased (785).
        • Melatonin is metabolized in the liver via the hepatic microsome cytochrome P450 system, primarily (but not exclusively) by the CYP2C19 and CYP1A family (particularly CYP1A2) and possibly CYP2C9. It appears to inhibit CYP1A2 and induce CYP3A. Thus, there are potential for interactions and altered levels of drugs and melatonin if used with agents that are substrates, inducers, or inhibitors of these isoenzymes.
        • Other agents that may alter synthesis or release of melatonin include caffeine (786;787), with a more pronounced effect in nonsmokers (788), diazepam (782;783), estradiol (789), vitamin B12 (790), verapamil (791), temazepam (792), and somatostatin (793).
        • 5-Hydroxytryptophan(5-HTP): In a case report, treatment with 5-HTP was found to normalize the melatonin profile (982).
        • Alzheimer's agents: Melatonin levels are often lower in patients with Alzheimer's disease (795;796;797;798;799;800). Some randomized controlled trials suggest a possible benefit of melatonin in patients with dementia (558;561). In vitro studies suggest a synergy between tacrine, a cholinesterase inhibitor, and melatonin (801).
        • Analgesics: In humans, melatonin use decreased the need for analgesics (366;605;681;682;683) and reduced levels of pain (552;596;682;683). However, compared to baseline, participants with chronic fatigue syndrome treated with melatonin showed a significant worsening of bodily pain (579).
        • Anesthetics: In human research, melatonin augmented standard general anesthetics (677;679;680;802;803;804;805;806). However, not all trials have been positive (807).
        • Angiotensin-converting enzyme (ACE) inhibitors: In human research, melatonin normalized ACE in six patients with high levels at baseline (531).
        • Antiaging agents: Melatonin has been identified as countering some of the deleterious effects of aging in human, animal, and in vitro research (14;15;16;17;18;19;20;21;22;23).
        • Antianxiety herbs and supplements: In humans, melatonin has been widely reported as having general and synergistic anxiolytic effects (563;677;679;680;803;804;805;806); however, evidence is mixed from a systematic review and well-designed clinical trials with respect to melatonin for anxiety prevention during surgery (678;681;683).
        • Antiarthritics: Based on mechanisms of action in vitro, melatonin has been suggested as possibly playing a beneficial role in osteoarthritis (60) and other rheumatic diseases (59).
        • Antiasthmatics: Asthmatics were found to have lower levels of endogenous melatonin (809;810); however, elevated levels at night were associated with worsening of symptoms (811;812).
        • Anticoagulants and antiplatelets: According to experts, melatonin may decrease prothrombin time (a measurement of blood clotting ability) (372;400). In humans, melatonin was associated with lower plasma levels of procoagulant factors, and dose-response relationships between the plasma concentration of melatonin and coagulation activity have been hypothesized (399). In animal research, melatonin enhanced platelet responsiveness (401). Increased platelet counts after melatonin use have been observed in patients, with decreased platelets due to cancer therapies (402;403;404;405;406;407;408), and cases of idiopathic thrombocytopenic purpura (ITP) treated with melatonin have been reported (409;410).
        • Anticonvulsants: It has been suggested that melatonin may act as a proconvulsant (449) and may lower seizure threshold and increase the risk of seizure, particularly in children with severe neurologic disorders (372;451;452). In a study exploring the effect of melatonin on insomnia in children, a reported case of mild generalized epilepsy developing four months after the start of the trial was noted; the child was initiated on valproate, and although melatonin was not discontinued, further seizures were lacking (431). In contrast, several case reports indicated reduced incidence of seizure with regular melatonin use (453;454;455;456;457;458). Both anticonvulsant (458;813;814;815) and proconvulsant (449) properties have been associated with melatonin in preclinical studies. This remains an area of controversy (449).
        • Antidepressants: In human research, antidepressants increased melatonin and 6-hydroxymelatonin (metabolite) levels, and increased melatonin bioavailability and decreased melatonin metabolism (819;820;821;822;823). Commonly reported adverse effects of melatonin in clinical trials include fatigue, dizziness, headache (including migraine), irritability, drowsiness, weakness, fogginess, yawning, nighttime awakening, poor sleep quality, vertigo, insomnia, and sleepiness (353;354;355;356;357;358;366;367;371;372;380;381;388;390;391;392;393;396;416;417;419;421;424;426;429;430;431;435;438;440;441;443;445;598;764). These symptoms are also indications of jet lag, and in some cases, causality may be unclear. In human research, mood changes have been reported, including giddiness, dysphoria (sadness), mood dip, nervousness, hyperactivity, irritability, and transient depression (357;358;390;391;392;419;427;440;463). Psychotic symptoms have also been reported in human research, including hallucinations, delusions, and paranoia, possibly due to overdose (371;441;462).
        • Anti-inflammatories: In human research, melatonin had anti-inflammatory effects in infants with respiratory distress (11), decreased the upregulation of proinflammatory cytokines in laboratory and human research (47;62;101;109;115;761;831;832), and inhibited NO and MDA production and increased glutathione levels (833;834). However, there is conflicting evidence from human trials, where melatonin induced a proinflammatory response, increasing levels of certain inflammatory cytokines (p>0.05), as well as plasma kynurenine concentrations (p<0.05) in individuals with rheumatoid arthritis (464). Also, in human research, melatonin lacked effects on CRP levels (835).
        • Antilipemics: There is some evidence of increases in cholesterol levels and atherosclerotic plaque buildup in human research (757) and animal research (759;760). In contrast, there are also reports of decreases in cholesterol levels in animal research (758) and decreased triglyceride and LDL cholesterol levels in human research (609;761;762).
        • Antineoplastics: According to the "melatonin hypothesis" of cancer, the exposure to light at night and anthropogenic electric and magnetic fields may be related to the increased incidence of cancer and childhood leukemia via melatonin disruption (836). Based on theoretical antioxidant mechanisms and in human research, melatonin has anticarcinogenic effects (351;402;403;521;571;573;575;837;838;839;840;841;841;842;843;844;845;846;847;848;849;850;851;852;853;854;855;856;857;858;859;860;861). Results of a meta-analysis of clinical trials suggested that melatonin had a significant effect on tumor remission and the one year survival rate, as well as an ability to decrease side effects related to radiochemotherapy, including thrombocytopenia, fatigue, and neurotoxicity (577).
        • Antiobesity agents: In laboratory research, melatonin inhibited adipocyte differentiation (193) and reduced gut motility (873). Other animal research has indicated that exogenous melatonin, however, lacks effect on leptin secretion (192). In patients with type 2 diabetes, nocturnal plasma melatonin levels were higher in obese subjects vs. nonobese subjects and lean nondiabetic controls (874).
        • Antioxidants: A variety of in vitro and in vivo studies have reported on the antioxidative effects of melatonin in a range of tissues and oxidative injury contexts (158;283;535;536;537;953;954;983;984;985;986;987;988;989;990;991;992;993;994;995;996;997;998;999;1000;1001;1002;1003;1004); (8;31;32;33;34;35;36;37;38;39;40;41;42;43;44;45;46;47;48;49;50;51;52;53;225;609;711;757;762;955;1005;1006;1007;1008;1009;1010;1011;1012;1013;1014;1015;1016;1017;1018;1019;1020;1021;1022;1023;1024). In vivo studies have generally used rats as their model system. Melatonin has been reported as being a more efficient antioxidant than glutathione (1025), vitamin C (1026;1027), or vitamin E (1028;1029;1030;1031;1032;1033). Synergy has also been observed with other antioxidants (1034;1035). Reports are by and large positive; however, select failures to observe ameliorative antioxidant function also appear in the literature (1036;1037).
        • Antiparasitics: In animal research, melatonin therapy controlled Trypanosoma cruzi proliferation by stimulating the host's immune response (203;875).
        • Antiparkinsonians: In human research, melatonin lacked an effect on signs of parkinsonism or levodopa effects, although it was well tolerated, but with side effects such as skin flushing, diarrhea, abdominal cramps, somnolence during the day, scotoma lucidum, and headaches (381).
        • Antipsychotics: Chronic treatment with antipsychotic drugs significantly improved psychotic symptomatology in schizophrenics, but did not change the secretory pattern of melatonin (876). The increase in melatonin secretion, which occurs with the initiation of neuroleptic therapy, may be responsible for the delay in the antipsychotic effects of neuroleptics and may also account for the lag in the development of drug-induced parkinsonism, as well as its disappearance (877). Preliminary human and laboratory reports suggest that melatonin had mixed effects on mood, sleep, and tardive dyskinesia in patients with schizophrenia, often treated with haloperidol (634;671;673;684;878;879;880;881;882).
        • Anti-ulcer agents: In human research, melatonin improved the healing of ulcers (692;693).
        • Antivirals: In animal research, the protective effect of melatonin against Venezuelan equine encephalomyelitis virus was likely mediated by melatonin receptor activation (884).
        • Caffeine-containing agents: Caffeine is reported to raise natural melatonin levels in the body (787) with a more pronounced effect in nonsmokers (788), possibly due to effects on the liver enzyme cytochrome P450 1A2 (886). It has been proposed that caffeine may increase the bioavailability of endogenous melatonin (887). Caffeine may also alter circadian rhythms in humans, with effects on melatonin secretion (788). It has been reported that caffeine reduced the onset of nighttime melatonin levels for women in the luteal phase, but had little effect on melatonin levels for oral contraceptive users (888). Another human study has shown that a single dose of 200mg of caffeine reduced natural melatonin levels (786), though a more recent human study using a twice-daily dose of 200mg of caffeine over seven days found a lack of effect on nighttime salivary melatonin (889).
        • Cardiovascular agents: In human research, low levels of platelet melatonin were found to be associated with angiographic no-reflow after primary percutaneous coronary intervention in patients with ST-segment elevation myocardial infarction (891). It has been proposed that melatonin acts directly on the cardiovascular system rather than modulating cardiac autonomic activity (892). In a poor-quality study, the inclusion of melatonin in the combined treatment of cardiovascular disease resulted in anti-ischemic, antianginal, antioxidant, and hypotensive effects (369). There is some evidence of increases in cholesterol levels and atherosclerotic plaque buildup in human research (757) and animal research (759;760). In contrast, there are also reports of decreases in cholesterol levels in animal research (758) and decreased triglyceride and LDL cholesterol levels in human research (609;761;762). In animal and human research, hypotension, blood pressure-lowering effects, and hypertension have been reported (359;360;361;362;363;364;365;366;367;369;443;506;609;756;825;826;827;828;829), although melatonin did not alter blood pressure in some animal or human research (376;830) or had mixed effects on night and day blood pressure, with decreases at night (375;377;378;608).
        • Chasteberry: In human research, chasteberry increased the natural secretion of melatonin (1038).
        • CNS depressants: In theory, based on possible risk of daytime sleepiness (67;354;411;415;418;421;422;435;667) and reported negative effects on certain cognitive tasks in humans in some, but not all, studies (412;413;428;767;893;894), melatonin may exacerbate the amount of drowsiness and reduced mental acuity caused by CNS depressants. Increased daytime drowsiness was reported when melatonin was used at the same time as the prescription sleep aid zolpidem (Ambien®), although it is not clear that effects were greater than with the use of zolpidem alone (114). In human research, an effect of remifentanil on melatonin concentration and an effect of melatonin on remifentanil-induced sleep disturbance were lacking (439).
        • CNS stimulants: In human research, there was an isolated case of aggression in a child diagnosed with ADHD and taking prescribed methylphenidate (433). In animal research, melatonin increased the adverse effects of methamphetamine on the nervous system (895). Melatonin has been implicated as having dosing time-dependent effects on the action of psychostimulant drugs such as cocaine and amphetamines (896).
        • Cognitive agents: In human research, exogenous melatonin caused decrements in performance, including a slowing of choice-reaction time (412;428) or learning (413); however, some studies have failed to confirm a decrement in performance (767;893;894), including a study of high-dose melatonin (50mg) in elderly persons (mean age: 84.5 years) (897).
        • Contraceptives: In patients undergoing in vitro fertilization embryo transfer (IVF-ET), although melatonin benefited oocyte maturation, effects on fertilization and pregnancy were lacking (594;898). Similarly, melatonin has been shown to improve viability of sperm (899) and embryos (118;900;901;902;903) produced with in vitro fertilization. In animal research, reproductive effects of melatonin have also been found (280;487;904;905;906;907;908;909;910).
        • Cytochrome P450-metabolized herbs and supplements: Melatonin is metabolized in the liver via the hepatic microsome cytochrome P450 system, primarily (but not exclusively) by the CYP2C19 and CYP1A family (particularly CYP1A2) (911;912) and possibly CYP2C9. It appears to inhibit CYP1A2 (465;466;467) and induce CYP3A. In human research, concurrent use of fluvoxamine and melatonin resulted in increased levels of melatonin, likely due to reduced metabolism of melatonin by inhibiting CYP1A2 and/or CYP2C9 (465;466;467). Caffeine is reported to raise natural melatonin levels in the body (787), with a more pronounced effect in nonsmokers (788), possibly due to effects on the liver enzyme cytochrome P450 1A2 (788). This effect was more pronounced in nonsmokers (788). Other human studies suggest that interactions between exogenous melatonin and substrates metabolized by CYP1A2 may differ in individuals before and after smoking abstinence (913). In animal research, melatonin inhibited the activity of cytochrome P450 2E1, but to a lesser extent than taurine (468).
        • Dental agents: In human research, salivary and gingival crevicular fluid melatonin levels were lower in individuals with periodontal disease (914).
        • Dermatologic agents: Dermatologic use of melatonin has been proposed because of its immunomodulatory and antioxidant abilities. Study findings indicate that melatonin accumulates in the stratum corneum (709). In human research, free radical scavenging was suggested as a possible mechanism of action in the protection against UV-induced erythema (711).
        • DHEA: In mice, DHEA and melatonin have been noted to stimulate immune function, with slight additive effects when used together (1039).
        • Diuretics: In clinical trials, an adverse effect associated with melatonin was increased enuresis (353;354;355;356;357). A study in children with ADHD suffering from insomnia noted bedwetting at a long-term follow-up (358).
        • Echinacea: In mice, a combination of echinacea and melatonin has been noted to slow the maturation of some types of immune cells, which may reduce immune function (1040).
        • Exercise performance agents: In human research, during a heavy-resistance exercise session, melatonin increased the area under the curve of growth hormone (508), protected against the overexpression of inflammatory mediators, and inhibited the expression of proinflammatory cytokines in exercising individuals (930).
        • Fertility agents: In patients undergoing in vitro fertilization embryo transfer (IVF-ET), although melatonin benefited oocyte maturation, effects on fertilization and pregnancy were lacking (594;898). Similarly, melatonin has been shown to improve viability of sperm (899) and embryos (118;900;901;902;903) produced with in vitro fertilization. In animal research, reproductive effects of melatonin have also been found (280;487;904;905;906;907;908;909;910).
        • Folate: In animal research, severe folate deficiency reduced the body's natural levels of melatonin (1041).
        • Gamma-aminobutyric acid (GABA): In animal research, results suggested a possible role of the GABAergic system in melatonin's effects (915). In human research, melatonin was found to potentiate the effects of gamma-amino butyric acid (GABA) (885).
        • Gastrointestinal agents: Preliminary research has indicated that melatonin aids symptoms of functional dyspepsia (598), gastroesophageal reflux disease (GERD) (602), Crohn's disease and ulcerative colitis (932), and irritable bowel syndrome (mixed evidence) (438;599;600;601;603;933).
        • Genitourinary tract agents: In clinical trials, an adverse effect associated with melatonin was increased enuresis (353;354;355;356;357). A study in children with ADHD suffering from insomnia noted bedwetting at a long-term follow-up (358).
        • Glaucoma agents: Preliminary human evidence suggests that melatonin may decrease intraocular pressure in the eye (366;460;461); however, according to reviews, high doses of melatonin may increase intraocular pressure and the risk of glaucoma, age-related maculopathy, and myopia (382), as well as retinal damage (400).
        • Headache agents: Evidence is mixed from human research with respect to preventive effects of melatonin on headaches, including migraines (146;440;604;605;934;935;936;937;938;939;940)
        • Heart rate regulating agents: Melatonin has been shown to increase heart rate when administered in patients taking nifedipine (a calcium channel blocker antihypertensive) (368) and in other studies (375); however, effects were lacking in other human research (376;377;378). When measured in the morning, the relationship between salivary melatonin and exercise-induced heart rate changes was steeper than when measured in the evening (941). Clinical significance is unclear. There are several rare or poorly described reports of abnormal heart rhythms, palpitations, fast heart rate, or chest pain, although in most cases, patients were taking other drugs that may account for these symptoms (370;371;372;373;374).
        • Hepatics: In patients with nonalcoholic steatohepatitis (NASH), use of melatonin resulted in improvements in liver function (445). In patients with steatohepatitis, melatonin decreased levels of proinflammatory cytokines, triglycerides, and GGTP (761). In human research, melatonin resulted in stable renal and liver function parameters after six weeks of use (755). Decreased transaminases have been shown in other human research (754). However, in one participant, treatment with melatonin resulted in increased alkaline phosphatase levels (390).
        • Herbs/supplements that affect GABA: In animal research, results suggested a possible role of the GABAergic system in melatonin's effects (915). In human research, melatonin was found to potentiate the effects of gamma-amino butyric acid (GABA) (885).
        • Hormonal agents: In humans, hormone replacement therapy (HRT) is reported to cause a decrease in daily melatonin secretion without disturbing circadian rhythm (942;943). In clinical and laboratory studies, melatonin has also been reported as producing varying hormonal effects. Such reports include changes in levels of luteinizing hormone (469;470;471;472;473;474;475;476), cortisol (477;478;479;480;481), progesterone (481;482;483;484;485), estradiol (482), thyroid hormone (T4 and T3) (486;487;488), testosterone (487;489), growth hormone (411;476;490;491;492;493;494), prolactin (411;495;496;497;498;499;500), oxytocin and vasopressin (490;501;502;503), adrenocorticotrophic hormone (478), and gonadotropin-inhibitory hormone (504). Melatonin has further been shown to alter pituitary hormone (LH and FSH) profiles in menopausal women to more "juvenile" profiles (488). In clinical trials, melatonin affected hormone levels in patients with hormonal-related cancers and had synergistic effects with tamoxifen (850;862;863). Other human studies reported a lack of significant hormonal effects (496;617;944;945). Gynecomastia (increased breast size) has been reported in men, as has decreased sperm count (both which resolved with cessation of melatonin) (400). Decreased sperm motility has also been reported in rats (548) and humans (549). Other human and laboratory studies have suggested that melatonin mimics the effect of drugs that act through the estrogen receptor interfering with the effects of endogenous estrogens, as well as those that interfere with the synthesis of estrogens by inhibiting the enzymes controlling the interconversion from their androgenic precursors (920). In females, blood pressure decreased only in hormone replacement therapy or birth control users and not nonusers (506;507). In human research, progesterone modulated melatonin secretion in postmenopausal women (946). In human research, in combination with estradiol treatment, melatonin reduced peak values of norepinephrine and increased epinephrine levels in some, but not all, stimulus situations (505;506;507). In human research, during a heavy-resistance exercise session, melatonin increased the area under the curve of growth hormone (508).
        • Hypoglycemics: Elevated blood sugar levels (hyperglycemia) have been reported in patients with type 1 diabetes (insulin-dependent diabetes) (382;383), and low doses of melatonin have reduced glucose tolerance and insulin sensitivity (384;385). In patients with type 2 diabetes mellitus who had a suboptimal response to the oral hypoglycemic agent metformin, melatonin and zinc acetate administration improved impaired fasting and postprandial glycemic control and decreased the level of glycated hemoglobin (386;387). However, in other research, melatonin supplementation was found to lack an effect on measures of glucose homeostasis (763).
        • Hypotensives: In animal and human research, hypotension, blood pressure-lowering effects, and hypertension have been reported (359;360;361;362;363;364;365;366;367;369;443;506;609;756;825;826;827;828;829), although melatonin did not alter blood pressure in some animal or human research (376;830) or had mixed effects on night and day blood pressure, with decreases at night (375;377;378;608). In human research, suppression of nocturnal melatonin secretion with atenolol (a beta1-adrenoreceptor antagonist) increased total wake time and decreased REM and slow-wave sleep; these effects were reversed if melatonin was given after the antagonist (64). Serum melatonin levels decreased noticeably with propranolol treatment (778). In animals, melatonin reduced the effects of the alpha-adrenergic agonist clonidine (756). In contrast, in humans, blood pressure increases have been observed when 5mg of melatonin was taken at the same time as the calcium-channel blocker nifedipine (368;465). Verapamil increased urinary melatonin excretion significantly (by 67%), but left excretion of 6-sulphatoxy-melatonin unaffected in healthy adults infused with calcium as a model for hyperkalemia (791).
        • Immunomodulators: In human research, melatonin was found to interact positively with immune therapies, such as interferon (509), interleukin-2 (407;408;510;511;512;513;514;515;516;517;518;519;520;521;522;523;524;525;526;527;528), or tumor necrosis factor (525;529;530). Based on limited human research, researchers concluded that melatonin may be an effective treatment for sarcoidosis (531). Exogenous melatonin has been shown to enhance immune response following veterinary vaccination (532). Researchers noted increased platelet counts after melatonin use in patients with decreased platelets due to cancer chemotherapy (402;403;404;406;407;408;533). According to a review, activation of melatonin receptors was associated with the release of cytokines by type 1 T-helper cells (Th1), including gamma-interferon (gamma-IFN) and IL-2, as well as novel opioid cytokines (534). Melatonin has been reported to promote neutrophil apoptosis in patients receiving hepatectomy involving ischemia and reperfusion of the liver (283;535;536;537). A combination hormone therapy including melatonin was found to improve leucocyte function in ovariectomized aged rats (538). In laboratory research, melatonin suppressed TNF-alpha, IL-1 beta, and IL-6 (101); inhibited Th1 cells (114); stimulated humoral activity and antibody production (532;539;540); inhibited NF-kappaB (541), as well as IKK, and JNK pathways (133); prevented T cell apoptosis (542); and stimulated mononuclear cell production (543). In human research, combined therapy with low-dose subcutaneous IL-2 and melatonin improved the mean number of lymphocytes, eosinophils, T lymphocytes, natural killer (NK) cells, and CD25- and DR-positive lymphocytes, and increased the mean CD4:CD8 ratio (544). In cancer patients who achieved disease control, melatonin induced a decrease in the number of regulatory T lymphocytes; this change was lacking in individuals with progressed disease (545)
        • Light therapy: Melatonin and light treatment have been used in combination in various human studies for illnesses such as depression (1042). In human research, bright light therapy increased the steepness of night melatonin levels (1043). In human research, use of an eyeglass LED delivery system suppressed melatonin secretion (1044). Nonvisual light for mood and cognitive enhancement decreased melatonin (1045). The effect of evening phototherapy for insomnia on melatonin was investigated; further details are lacking (1046). In a review, the effect of bright light therapy for mood disorders and melatonin levels was discussed; futher details are lacking (1047).
        • Lithium: In human research, lithium had a significant effect on sensitivity to light but not on overall melatonin synthesis (947).
        • Magnetic fields: It has been theorized that chronic exposure to magnetic fields or recurrent cellular telephone use may alter melatonin levels and circadian rhythms. However, several studies suggest that this is not the case (948;949;950;951). Melatonin was shown to reduce the effects of lipid peroxidation, less effectively than vitamin E, in rats exposed to static magnetic fields under laboratory conditions (952).
        • Meditation: In human research, meditation increased nighttime plasma melatonin levels (1048).
        • Musculoskeletal agents: According to case reports, ataxia (difficulties with walking and balance) may occur following melatonin overdose (372). In human research, melatonin lacked negative effects on postural stability (367). Compared to baseline, participants with chronic fatigue syndrome treated with melatonin showed a significant worsening of bodily pain (579). Weakened muscle power was reported in a clinical trial (598).
        • Music therapy: In patients with Alzheimer's disease, music therapy increased serum melatonin levels (1049).
        • Neurologic agents: It has been proposed that melatonin may reduce the amount of neurologic damage patients experience after stroke, based on antioxidant properties (953;954;955;956;957;958;959;960). A significant body of basic research has indicated that melatonin may possess neuroprotective properties (79;167;168;169;170;171;172;173;174;175;176;177;179;180;181;182;183;184;184;185;186;187;187), meriting reviews in the contexts of neurodegenerative diseases (961), the peripheral nervous system (962), and traumatic nervous system injury (80). However, commonly reported adverse effects of melatonin in clinical trials include fatigue, dizziness, headache (including migraine), irritability, drowsiness, weakness, fogginess, yawning, nighttime awakening, poor sleep quality, vertigo, insomnia, and sleepiness (353;354;355;356;357;358;366;367;371;372;380;381;388;390;391;392;393;396;416;417;419;421;424;426;429;430;431;435;438;440;441;443;445;598;764). These symptoms are also indications of jet lag, and in some cases, causality may be unclear. In human research, mood changes have been reported, including giddiness, dysphoria (sadness), mood dip, nervousness, hyperactivity, irritability, and transient depression (357;358;390;391;392;419;427;440;463). Psychotic symptoms have also been reported in human research, including hallucinations, delusions, and paranoia, possibly due to overdose (371;441;462).
        • Neuromuscular blockers: In laboratory research, melatonin increased the neuromuscular blocking effect of the muscle relaxant succinylcholine, but not vecuronium (963).
        • Ocular agents: In limited human research, melatonin stabilized vision in patients suffering from age-related macular degeneration (459). Preliminary human evidence also suggests that melatonin may decrease intraocular pressure in the eye (366;460;461); however, according to reviews, high doses of melatonin may increase intraocular pressure and the risk of glaucoma, age-related maculopathy, and myopia (382), as well as retinal damage (400). Use of transition lenses as part of chromotherapy for macular degeneration was found to maintain the physiological balance of melatonin (964). In human research, use of eye masks increased melatonin levels (965).
        • Osteoporosis agents: In laboratory research, melatonin impaired osteoclast activity and bone resorption (916;917;918). In human research, melatonin lacked effects on bone density, NTX, or OC, although the NTX:OC ratio in the melatonin group was reduced (628).
        • Otic agents: In human research, melatonin attenuated the muscle sympathetic nerve activity (vestibulosympathetic reflex) response to baroreceptor unloading while lacking effects on the vestibulocollic reflexes (968). In human research, use of ear plugs increased melatonin levels (965).
        • Phytoestrogens: Human and laboratory studies have suggested that melatonin mimics the effect of drugs that act through the estrogen receptor interfering with the effects of endogenous estrogens, as well as those that interfere with the synthesis of estrogens by inhibiting the enzymes controlling the interconversion from their androgenic precursors (920). Mechanisms of melatonin's oncostatic action may include regulation of estrogen receptor expression and transactivation (921) and antiestrogenic effects (922;923;924). MCF-7 human breast cancer cultured cells have been reported as melatonin sensitive, as well as estrogen receptor positive and estrogen responsive (925), although this finding was not confirmed in a subsequent study (926). Melatonin has been reported to elicit an increase in estrogen receptor activity in breast tumors (927). Low plasma melatonin concentrations were associated with greater amounts of estrogen or progesterone receptors on primary tumors (928). In a review on the anticarcinogenic role of melatonin, potential mechanisms included the inhibition of initiation and growth of hormone-dependent tumors by decreasing the expression of estrogen receptors, as well as aromatase activity, resulting in the inhibition of cancer cell proliferation, a decrease in oxidative stress, and an increase in the activity of the immune system (929).
        • Radioprotective agents: Due to its well-known antioxidant properties, it has been suggested that melatonin possesses a protective effect against damage caused by ionizing radiation, a hypothesis borne out of preliminary animal and in vitro research (206;207;215;216;217;219;969). Melatonin has been shown to ameliorate oxidative injury due to ionizing radiation in vitro (214;970;971). The specific mechanisms may involve downregulation of apoptotic pathways via control of oxidative load (972).
        • Renally eliminated agents: In human research, melatonin resulted in decreased renal blood flow velocity and conductance (376). In human research, melatonin resulted in stable renal function parameters after six weeks of use (755).
        • Respiratory agents: In a clinical trial, melatonin reduced dyspnea; however, changes in lung function were lacking (580).
        • Sedatives: In theory, based on possible risk of daytime sleepiness (67;354;411;415;418;421;422;435;667) and reported negative effects on certain cognitive tasks in humans in some, but not all, studies (412;413;428;767;893;894), melatonin may exacerbate the amount of drowsiness and reduced mental acuity caused by CNS depressants. In human research, melatonin has been shown to decrease sleep latency (390;752) and benefit sleep quality and duration in children, older and younger adults, individuals with disabilities, and visually impaired individuals (322;354;355;356;357;373;374;379;389;391;392;431;446;581;613;615;638;659;734). In human research, exogenous melatonin exerted hypnotic effects, primarily when circulating levels of endogenous melatonin were low (653), and even very low doses caused sleep in some studies when ingested before endogenous melatonin onset (418;649;655;661;662). Also, in human research, melatonin has been shown to decrease the amount of anesthesia required during surgery (679;680;719;973).
        • Seizure threshold-lowering agents: It has been suggested that melatonin may act as a proconvulsant (449) and may lower seizure threshold and increase the risk of seizure, particularly in children with severe neurologic disorders (372;451;452). In a study exploring the effect of melatonin on insomnia in children, a reported case of mild generalized epilepsy developing four months after the start of the trial was noted; the child was initiated on valproate, and although melatonin was not discontinued, further seizures were lacking (431). In contrast, several case reports indicated reduced incidence of seizure with regular melatonin use (453;454;455;456;457;458). Both anticonvulsant (458;813;814;815) and proconvulsant (449) properties have been associated with melatonin in preclinical studies. This remains an area of controversy (449).
        • Thermoregulating agents: In human research, hypothermic effects of melatonin have been reported with doses from 15mg (427;657;975;976), and ingestion of 1.6mg of melatonin was reported to result in approximately 0.4ºC decrease of body temperature in humans (975;977;978;979). Reports of feeling cold or hypothermia exist in other clinical literature (356;390;392;445;448), but not in all studies (614).
        • Thyroid agents: In human research, thyroid-stimulating hormone (TSH) serum levels were lower and those of free thyroxine (FT4) were increased at night when endogenous melatonin levels were higher (980). In clinical and laboratory studies, melatonin has also been reported to change levels of thyroid hormone (T4 and T3) (486;487;488).
        • Vasodilators: In healthy male volunteers, melatonin significantly increased peripheral blood flow, as measured by distal to proximal skin temperature gradient and finger pulse volume (981). In human research, melatonin resulted in decreased renal blood flow velocity and conductance, increased forearm blood flow and vascular conductance, and lacked an effect on cerebral blood flow (376).

        Melatonin/Food Interactions:

        • General: The gastrointestinal effects of melatonin are likely dependent on food intake (1050;1051). Food deprivation was found to impair daily rhythms of melatonin content by altering the activity of melatonin-synthesizing enzymes (1051). According to a review, in animal research, fasting lacked effects on melatonin-induced intestinal bicarbonate secretion (1052).
        • Melatonin-containing foods: Some foods, such as oats, sweet corn, rice, ginger, tomatoes, bananas, and barley, contain small amounts of melatonin and may increase melatonin levels (1053;1054). In human research, consumption of diets enriched with Jerte Valley cherry (a food source of melatonin (1055)) increased urinary levels of 6-sulfatoxymelatonin (1056).
        • Vegetables: Increased consumption of vegetables raised circulatory melatonin concentrations (1057).

        Melatonin/Lab Interactions:

        • 8-isoprostanes: In human research, melatonin resulted in decreased 8-isoprostane levels (580).
        • Blood glucose: Elevated blood sugar levels (hyperglycemia) have been reported in patients with type 1 diabetes (insulin-dependent diabetes) (382;383), and low doses of melatonin have reduced glucose tolerance and insulin sensitivity (384;385). In patients with type 2 diabetes mellitus who had a suboptimal response to the oral hypoglycemic agent metformin, melatonin and zinc acetate administration improved impaired fasting and postprandial glycemic control and decreased the level of glycated hemoglobin (386;387). However, in other research, melatonin supplementation was found to lack an effect on measures of glucose homeostasis (763).
        • Blood pressure: In animal and human research, hypotension, blood pressure-lowering effects, and hypertension have been reported (359;360;361;362;363;364;365;366;367;369;443;506;609;756;825;826;827;828;829), although melatonin did not alter blood pressure in some animal or human research (376;830) or had mixed effects on night and day blood pressure, with decreases at night (375;377;378;608). In human research, suppression of nocturnal melatonin secretion with atenolol (a beta1-adrenoreceptor antagonist) increased total wake time and decreased REM and slow-wave sleep; these effects were reversed if melatonin was given after the antagonist (64). Serum melatonin levels decreased noticeably with propranolol treatment (778). In animals, melatonin reduced the effects of the alpha-adrenergic agonist clonidine (756). In contrast, in humans, blood pressure increases have been observed when 5mg of melatonin was taken at the same time as the calcium-channel blocker nifedipine (368;465). Verapamil increased urinary melatonin excretion significantly (by 67%), but left excretion of 6-sulphatoxy-melatonin unaffected in healthy adults infused with calcium as a model for hyperkalemia (791).
        • Body temperature: In human research, hypothermic effects of melatonin have been reported with doses from 15mg (427;657;975;976), and ingestion of 1.6mg of melatonin was reported to result in approximately 0.4ºC decrease of body temperature in humans (975;977;978;979). Reports of feeling cold or hypothermia exist in other clinical literature (356;390;392;445;448).
        • Bone markers: In laboratory research, melatonin impaired osteoclast activity and bone resorption (916;917;918). In human research, melatonin lacked effects on bone density, NTX, or OC, although the NTX:OC ratio in the melatonin group was reduced (628).
        • Catecholamines: In human research, effects of melatonin levels on catecholamines were lacking (506).
        • Coagulation panel: According to experts, melatonin may decrease prothrombin time (a measurement of blood clotting ability) (372;400). In humans, melatonin was associated with lower plasma levels of procoagulant factors, and dose-response relationships between the plasma concentration of melatonin and coagulation activity have been hypothesized (399). In animal research, melatonin enhanced platelet responsiveness (401). Increased platelet counts after melatonin use have been observed in patients with decreased platelets due to cancer therapies (402;403;404;405;406;407;408;533), and cases of idiopathic thrombocytopenic purpura (ITP) treated with melatonin have been reported (409;410).
        • Electroencephalogram (EEG): In human research, effects of melatonin on EEG characteristics were lacking (1058).
        • Glycated hemoglobin (HA1c): In patients with type 2 diabetes mellitus who had a suboptimal response to the oral hypoglycemic agent metformin, melatonin and zinc acetate administration improved impaired fasting and postprandial glycemic control and decreased the level of glycated hemoglobin (386;387).
        • Heart rate: Melatonin has been shown to increase heart rate when administered in patients taking nifedipine (a calcium channel blocker antihypertensive) (368) and in other studies (375); however, effects were lacking in other human research (376;377;378). When measured in the morning, the relationship between salivary melatonin and exercise-induced heart rate changes was steeper than when measured in the evening (941). Clinical significance is unclear. There are several rare or poorly described reports of abnormal heart rhythms, palpitations, fast heart rate, or chest pain, although in most cases, patients were taking other drugs that may account for these symptoms (370;371;372;373;374).
        • Hormone panel: In clinical and laboratory studies, melatonin has been reported as producing varying hormonal effects. Such reports include changes in levels of luteinizing hormone (469;470;471;472;473;474;475;476), cortisol (477;478;479;480;481), progesterone (481;482;483;484;485), estradiol (482), thyroid hormone (T4 and T3) (486;487;488), testosterone (487;489), growth hormone (411;476;490;491;492;493;494), prolactin (411;495;496;497;498;499;500), oxytocin and vasopressin (490;501;502;503), adrenocorticotrophic hormone (478), and gonadotropin-inhibitory hormone (504). Melatonin has further been shown to alter pituitary hormone (LH and FSH) profiles in menopausal women to more "juvenile" profiles (488). In human research, in combination with estradiol treatment, melatonin reduced peak values of norepinephrine and increased epinephrine levels in some, but not all, stimulus situations (505;506;507). Effects on cortisol, norepinephrine, and epinephrine were lacking in some human research (625). In human research, during a heavy-resistance exercise session, melatonin increased the area under the curve of growth hormone (508).
        • Immune panel: In human research, combined therapy with low-dose subcutaneous IL-2 and melatonin improved the mean number of lymphocytes, eosinophils, T lymphocytes, natural killer (NK) cells, and CD25- and DR-positive lymphocytes, and increased the mean CD4:CD8 ratio (544). In cancer patients who achieved disease control, melatonin induced a decrease in the number of regulatory T lymphocytes; this change was lacking in individuals with progressed disease (545). In human research, melatonin enhanced IL-2-, with or without IL-12, induced lymphocytosis, and reversed lymphcytopenia induced by IL-12 (868). In laboratory research, melatonin suppressed TNF-alpha, IL-1 beta, and IL-6 (101). In human research, melatonin reduced levels of IL-6, IL-8, IL-10, and IL-12 (552).
        • Inflammatory markers: In human research, melatonin had anti-inflammatory effects in infants with respiratory distress (11), decreased the upregulation of proinflammatory cytokines in laboratory and human research (47;62;101;109;115;761;831;832), and inhibited NO and MDA production and increased glutathione levels (833;834). In exercising individuals, melatonin protected against the overexpression of inflammatory mediators and inhibited the expression of proinflammatory cytokines in exercising individuals (930). However, there is conflicting evidence from human trials, where melatonin induced a proinflammatory response, increasing levels of certain inflammatory cytokines (p>0.05), as well as plasma kynurenine concentrations (p<0.05) in individuals with rheumatoid arthritis (464). Also, in human research, melatonin lacked effects on CRP levels (835).
        • Insulin-like growth factor: In human research, melatonin lacked effects on insulin-like growth factor I (IGF-1) or insulin-like growth factor-binding protein 3 (IGFBP-3), or their ratio (747).
        • Intraocular pressure: Preliminary human evidence also suggests that melatonin may decrease intraocular pressure in the eye (366;460;461); however, according to reviews, high doses of melatonin may increase intraocular pressure and the risk of glaucoma, age-related maculopathy, and myopia (382), as well as retinal damage (400).
        • Lipid profile: There is some evidence of increases in cholesterol levels and atherosclerotic plaque buildup in human research (757) and animal research (759;760). In contrast, there are also reports of decreases in cholesterol levels in animal research (758) and decreased triglyceride and LDL cholesterol levels in human research (609;761;762).
        • Liver function: In patients with nonalcoholic steatohepatitis (NASH), use of melatonin resulted in improvements in liver function (445). In patients with steatohepatitis, melatonin decreased levels of proinflammatory cytokines, triglycerides, and GGTP (761). In human research, melatonin resulted in stable renal and liver function parameters after six weeks of use (755). Decreased transaminases have been shown in other human research (754). However, in one participant, treatment with melatonin resulted in increased alkaline phosphatase levels (390).
        • Melatonin levels: In human research, melatonin supplementation was found to increase plasma levels of melatonin, occasionally into the daylight hours (446;508;665;692;754;761;1053;1054;1059).
        • Nitric oxide: In human research, the production of NO was suggested to be regulated by melatonin; however, further details are lacking (1060).
        • Nitrate/nitrite: In human research, melatonin decreased nitrite and nitrate levels (11;645).
        • Pigments: In human research, melatonin induced and augmented lentigines and nevi (1061). This report was commented on (1062).
        • Plasma kynurenine concentrations: In human research, melatonin induced an increase in plasma kynurenine concentrations in individuals with rheumatoid arthritis (464).
        • Renal function: In human research, melatonin resulted in decreased renal blood flow velocity and conductance (376). In human research, melatonin resulted in stable renal function parameters after six weeks of use (755).
        • Seizure threshold: It has been suggested that melatonin may act as a proconvulsant (449) and may lower seizure threshold and increase the risk of seizure, particularly in children with severe neurologic disorders (372;451;452). In a study exploring the effect of melatonin on insomnia in children, a reported case of mild generalized epilepsy developing four months after the start of the trial was noted; the child was initiated on valproate, and although melatonin was not discontinued, further seizures were lacking (431). In contrast, several case reports indicated reduced incidence of seizure with regular melatonin use (453;454;455;456;457;458). Both anticonvulsant (458;813;814;815) and proconvulsant (449) properties have been associated with melatonin in preclinical studies. This remains an area of controversy (449).
        • Sperm function: In animal research, exposure to exogenous melatonin increased sperm motility, ejaculate volume, sperm concentration, total sperm output, and total function sperm fraction, and decreased means of reaction time, dead sperm, and abnormal sperm (487). Melatonin implants have also been shown to improve semen characteristics in sheep (905).

        Melatonin/Nutrient Depletion:

        • Caffeine: Caffeine is reported to raise natural melatonin levels in the body (787), with a more pronounced effect in nonsmokers (788), possibly due to effects on the liver enzyme cytochrome P450 1A2 (886). It has been proposed that caffeine may increase the bioavailability of endogenous melatonin (887). Caffeine may also alter circadian rhythms in humans, with effects on melatonin secretion (788). It has been reported that caffeine reduced the onset of nighttime melatonin levels for women in the luteal phase, but had little effect on melatonin levels for oral contraceptive users (888). Another human study has shown that a single dose of 200mg of caffeine reduced natural melatonin levels (786), though a more recent human study using a twice-daily dose of 200mg of caffeine over seven days found a lack of effect on nighttime salivary melatonin (889).
        • Cholesterol: There is some evidence of increases in cholesterol levels and atherosclerotic plaque buildup in human research (757) and animal research (759;760). In contrast, there are also reports of decreases in cholesterol levels in animal research (758) and decreased triglyceride and LDL cholesterol levels in human research (609;761;762).
        • Glucose: Elevated blood sugar levels (hyperglycemia) have been reported in patients with type 1 diabetes (insulin-dependent diabetes) (382;383), and low doses of melatonin have reduced glucose tolerance and insulin sensitivity (384;385). In patients with type 2 diabetes mellitus who had a suboptimal response to the oral hypoglycemic agent metformin, melatonin and zinc acetate administration improved impaired fasting and postprandial glycemic control and decreased the level of glycated hemoglobin (386;387). However, in other research, melatonin supplementation was found to lack an effect on measures of glucose homeostasis (763).
        • Natural melatonin levels (decreases): Multiple drugs are reported to lower natural levels of melatonin in the body. It is not clear that there are any health hazards of lowered melatonin levels, or if replacing melatonin with supplements is beneficial. Examples of drugs that may reduce production or secretion of melatonin include nonsteroidal anti-inflammatory drugs (NSAIDs), such as ibuprofen (Motrin®, Advil®) or naproxen (Naprosyn®, Aleve®) (776;777); beta-blocker blood pressure medications, such as atenolol (Tenormin®) or metoprolol (Lopressor®, Toprol®) (779;780); and medications that reduce levels of vitamin B6 in the body, such as oral contraceptives, hormone replacement therapy, loop diuretics, hydralazine, and theophylline (781;782;783;784). Anesthesia using 7% sevoflurane decreased melatonin blood concentrations (785). Asthmatics may have lower levels of endogenous melatonin (809;810). Melatonin levels in serum decreased noticeably with propranolol treatment (778). Other drugs that may reduce melatonin levels (by inducing P450 1A2) include the following: carbamazepine, insulin, 3-methyl cholanthrene, modafinil, nafcillin, nicotine, omeprazole, phenobarbital, phenytoin, primidone, rifampin, and ritonavir, according to anecdotal information.
        • Sevoflurane: In human research, sevoflurane resulted in a reduction of postoperative plasma melatonin levels (974).
        • Verapamil: Verapamil increased urinary melatonin excretion significantly (by 67%), but left excretion of 6-sulphatoxy-melatonin unaffected in healthy adults infused with calcium as a model for hyperkalemia (791).

        Mechanism of Action

        Pharmacology:

        • Constituents: Endogenous melatonin is an indole neurohormone synthesized from the amino acid tryptophan (1;312).
        • General: Melatonin acts on the MT(1) and MT(2) melatonin receptors located in the hypothalamic suprachiasmatic nuclei, the site of the body's master circadian clock (1063). Based on animal experimental models, melatonin has antidopaminergic effects; repeated administration of melatonin may modify the plasticity of behaviors mediated by central dopaminergic systems (1064).
        • Endogenous melatonin: Melatonin is an indole neurohormone synthesized from tryptophan in the pineal gland (1;312), with regulation of day-night changes in synthesis by serotonin N-acetyltransferase (2). Some synthesis takes place additionally in the retina, gastrointestinal tract, bone marrow, bile, and cells of the immune system (1065;1066;1067;1068). The formation and release of melatonin are stimulated by darkness and inhibited by light (1069;1070;1071), without significant differences between polarized and nonpolarized light (1072;1073). The primary melatonin-controlled events take place in the rods rather than in the cones of retina (1073). This response to light may remain in blind subjects, despite apparently complete loss of visual function (1074;1075).
        • Since beta1-adrenoreceptor antagonists almost completely inhibit the normal nighttime rise in melatonin, it is thought that human pineal adrenoreceptors are of beta1-subtype (779;780;780;1076;1077;1078;1079). Similar arguments indicate the involvement of alpha1-adrenoreceptors in the stimulation of melatonin synthesis (1080), while alpha2-adrenoreceptors presumably operate as downregulators (1081;1082). As such, the production of melatonin is considered to be regulated by the sympathetic nervous system and its neurotransmitter norepinephrine. With the onset of darkness, norepinephrine release is followed by stimulation of postsynaptic pineal adrenoreceptors and enhancement of melatonin synthesis (1;1083). The involvement of norepinephrine is supported by an increase in nighttime plasma melatonin in humans treated with an inhibitor of norepinephrine uptake (1084;1085). Activation of the sympathetic nervous system appears to accelerate melatonin synthesis (1086;1087;1088). Animal studies suggest that endogenous melatonin is involved in the suppression of sympathetic activity, with possible negative feedback inhibition (1089). Melatonin may reduce circulating norepinephrine in young individuals and in postmenopausal women receiving estradiol replacement, but not in menopausal women (507).
        • Melatonin secretion occurs at a constant rate in both young and older men and women (1090). Circadian melatonin rhythm appears at the end of the neonatal period and persists thereafter (1091;1092), despite functioning vision.
        • In humans, the daytime melatonin plasma level is less than 40pM/L. In the middle of the night, the concentration increases to approximately 260pM/L (1). In some species, these values may reach a 50-fold difference (918). Both high- and low-affinity binding sites have been identified in several brain regions (superchiasmatic nuclei of the hypothalamus, thalamus, and pituitary gland), retina, gut, ovaries, blood vessels, lymphocytes, and platelets (1;705;1093;1094). A recently discovered class of melatonin-binding sites, called orphan receptors, presumably mediate the ability of melatonin to regulate gene expression (918;1095).
        • In humans, a pacemaker located in the suprachiasmatic nucleus of the hypothalamus controls the circadian rhythm of melatonin production. By binding to the receptors in the nucleus, melatonin may alter a phase of this pacemaker (e.g., reset the biological clock) (1096;1097;1098;1099;1100;1101;1102). The suprachiasmatic nuclei are the target sites for the effect of exogenous melatonin on the amplitude of the endogenous melatonin rhythm (1103).
        • The following factors may modulate the synthesis, release, or bioavailability of melatonin: nonsteroidal anti-inflammatory drugs (777), diazepam (782;783), vitamin B12 (790), GABA (1104;1105), ethanol (1106), micronutrient accumulation and depletion (1107), gonadotropin-releasing hormone, gonadotropins/gonadal steroids (1108;1109), estrogen plus progesterone (784), testosterone (1108), duration of gestation (1110), prenatal growth restriction (1111), interleukin-2, cancer (1112), posture (1113), balance (1114), phenelzine (1115), Thorazine® (927), sleep deprivation (1087), hypercalcemia/verapamil (791), temazepam, zopiclone (792), levodopa-related motor complications (1116), agnus castus (1038), desipramine (1117), prazosin (1081), intravenous L-tryptophan (1067), caffeine (887), and exercise (1118). In terms of modulating the synthesis and release of melatonin, current research reports a lack of an effect using somatostatin (793), oral administration of 5-hydoxytryptophan (1119), exposure to pulsating magnetic fields (949), nifedipine (1120), midazolam, sodium thiopental (1121), electroconvulsive therapy, TRH-injection, L-dopa, or bromergocryptine (150;1122).
        • Disturbances in the circadian rhythm of melatonin (and declines in nighttime melatonin) have been associated with aging (1;1123;1124;1125;1126;1127;1128;1129).
        • The lack of light signal in blind persons leads to various unusual free-running melatonin secretory patterns (1130;1131;1132). Low-dose melatonin has been noted to be more effective entraining blind people with a free-running melatonin secretory pattern due to the potential of excess hormone discharging into the wrong zone of the phase-response curve (1133). Free-running patterns are also observed after pineal gland damage (1134) or under special working regimens (1135;1136). The nocturnal onset of melatonin secretion strongly correlates with a steep rise in sleep propensity and precedes it by approximately two hours (1137). The specific events taking place during this interval currently remain obscure. It is possible that melatonin does not actively induce sleep but switches off a wakefulness-generated mechanism that opposes a sleep-inducing mechanism (1137), the alerting process being dependent on the suprachiasmatic nucleus (1097).
        • While measuring endogenous melatonin, some authors have not found a link between melatonin secretion and the sleep-waking cycle in humans (1138). It has been suggested by some that natural sleep is largely determined by a functioning circadian system without melatonin involvement (1102;1124;1139;1140).
        • Analgesic effects: According to a review, research has indicated that melatonin may modulate pain perception via activation of MT(1) and MT(2) receptors, opening of potassium channels, inhibition of 5-lipoxygenase and cyclooxygenase-2 expression, and indirect activation of opioid receptors (298). Ebadi et al. wrote a review on the potential analgesic effects of melatonin and the potential for its involvement in decreased overnight pain (1141).
        • Antiaging effects: Melatonin has been identified as countering some of the deleterious effects of aging (14;15;16;17;18;19;20;21;22;23). Animal experimentation has indicated that melatonin may improve longevity (cellular and otherwise) by preventing age-related mitochondrial impairment (19), maintaining youthful rhythmic activity (20), improving monoaminergic neurotransmission (1142), and reversing immunosenescence (538;1143). An experimental model of age-induced neuronal apoptosis indicated that melatonin may exert a protective effect via prosurvival Akt and prevention of DNA damage (1144).
        • Antiarthritic effects: As shown in vitro, melatonin has been suggested as possibly playing a beneficial role in osteoarthritis (60) and other rheumatic diseases (59). Possible mechanisms include the enhancement of cartilage matrix synthesis (60) and the inhibition of fibroblast-like synoviocyte proliferation (59).
        • Anticancer effects: Researchers have investigated the outcome of melatonin use as an adjunct to chemotherapy, interleukin-2, radiotherapy, support therapy, and tumor necrosis factor (TNF) therapy (866). According to the "melatonin hypothesis" of cancer, the exposure to light at night and anthropogenic electric and magnetic fields may be related to the increased incidence of cancer and childhood leukemia via melatonin disruption (836).
        • In vitro, at pharmacological concentrations, melatonin exhibited cytotoxic activity in cancer cells (921;1145). At both physiological and pharmacological concentrations, melatonin acted as a differentiating agent in some cancer cells and lowered their invasive and metastatic capabilities through alterations in adhesion molecules and maintenance of gap junctional intercellular communication (921). In other cancer cell types, melatonin, either alone or in combination with other agents, induced apoptotic cell death. Biochemical and molecular mechanisms of melatonin's oncostatic action may include regulation of estrogen receptor expression and transactivation, calcium/calmodulin activity, protein kinase C activity, cytoskeletal architecture and function, intracellular pH, melatonin receptor-mediated signal transduction cascades, and fatty acid transport and metabolism (921). These and other possible anticancer mechanisms may include or operate by modulation of apoptosis (1146;1147;1148;1149;1150;1151), downregulation of HIF-1 alpha expression (antiangiogenic) (1152), antiestrogenic effects (922;923;924), antiproliferative effects (1153;1154), suppression of linoleic acid uptake and metabolism (1155), SIRT1 inhibition (21), cytoskeletal dynamics (inhibition of cancer cell migration) (1156;1157), epigenetic regulation (1158), reduction of oxidative stress (1159), modulation of melatonin receptor expression (1160), and stimulation of melatonin receptors and associated biomolecular cascades (1161).
        • Nuclear signaling appears to play a central role in the function of melatonin (1162). At physiological circulating concentrations, melatonin may inhibit cancer cell proliferation via cell cycle-specific effects identified in vitro (921;1145;1163;1164). Women with endometrial cancer have been found to have melatonin plasma levels six times lower than tumor-free controls. It is not clear if this is causative or adaptive, although melatonin antiproliferative activity is hypothesized (1165;1166;1167).
        • As a chemoprotective agent, daily administration of melatonin for six months was reported to induce a protective effect against the formation of mammary tumors in rat models (1168). MT1 (high-affinity melatonin receptor subtype) receptors have been detected in normal and malignant breast epithelium, with high receptor levels occurring more frequently in tumor cells (p<0.001); tumors with moderate or strong reactivity are more likely to be high nuclear grade (p<0.045). It has been proposed that these findings have implications for the use of melatonin in breast cancer therapy (1169).
        • An oncostatic effect of melatonin (cessation of cell cycle progression) has been reported in human prostate cancer cells (1170;1171); nuclear rather than membrane melatonin receptors have been implicated, as have MT1 receptor subtypes. MCF-7 human breast cancer cultured cells have been reported as melatonin sensitive, as well as estrogen receptor positive and estrogen responsive (925), although this finding was not confirmed in a subsequent study (926). Inhibition of cell proliferation may be mediated through G-protein coupled membrane receptors, or via retinoid orphan receptors with involvement of the calcium signaling pathway (1172). Downregulation of the receptor level indicates the possibility of an estrogen-responsive mechanism. This notion is supported by melatonin-induced stimulation of expression of the growth-inhibitory factor, TGF-beta (925). However, other investigators have failed to reproduce the oncostatic effect of melatonin on MCF-7 cells or ZR-75-1 and T-47D cell lines (1173). Melanoma M 2R cells also responded to melatonin (1174). It has also been suggested that melatonin-related growth inhibition of breast cancer cells may be related to enhanced TGF-beta(1) secretion (1175).
        • Melatonin has been reported to elicit an increase in estrogen receptor activity in breast tumors (927). Low plasma melatonin concentrations were associated with greater amounts of estrogen or progesterone receptors on primary tumors (928).
        • Pinealectomy is reported to enhance tumor growth and metastasis in animals (1176). In addition, increased neoplastic growth has been noted in some animal models (Morris hepatoma 7288 CTC in rats), when animals are maintained in a constant weak light (0.2 lux), thereby inducing inhibition of melatonin synthesis (1177).
        • A review has been written on the anticarcinogenic role of melatonin (929). Potential mechanisms of action mentioned in the review included the inhibition of initiation and growth of hormone-dependent tumors by decreasing the expression of estrogen receptors, as well as aromatase activity, resulting in the inhibition of cancer cell proliferation, a decrease in oxidative stress, and an increase in the activity of the immune system.
        • Anticonvulsant effects: Both anticonvulsant (458;813;814;815) and proconvulsant (449) properties have been associated with melatonin in preclinical studies. In animals, intraventricular injection of antimelatonin antibody has elicited transitory epileptiform abnormalities in the electroencephalogram (1178). Proposed mechanisms of action include altered brain GABAergic neurotransmission, interactions with benzodiazepine brain receptors, tryptophan metabolite activity, free radical scavenger activity, or modulation of brain amino acids and nitric oxide (NO) production (449;798;814).
        • Antidiabetic effects: Elevated blood sugar levels (hyperglycemia) have been reported in patients with type 1 diabetes (insulin-dependent diabetes) (382;383), and low doses of melatonin have reduced glucose tolerance and insulin sensitivity (384;385). In patients with type 2 diabetes mellitus who had a suboptimal response to the oral hypoglycemic agent metformin, melatonin and zinc acetate administration improved impaired fasting and postprandial glycemic control and decreased the level of glycated hemoglobin (386;387). However, in other research, melatonin supplementation was found to lack an effect on measures of glucose homeostasis (763).
        • Melatonin may stimulate glycogen synthesis via the PKCzeta-Akt-GSK3beta pathway (1179) as well as inhibit insulin secretion via stimulation of melatonin receptors in pancreatic islet cells (1180). Exogenous melatonin has also been shown to improve glucose tolerance, increase muscle insulin receptor and GLUT 4 expression, and enhance glucose clearance from the blood in bats (1181). Based on findings in rats, melatonin has been suggested as influencing gene expression in insulinoma beta-cells (1182).
        • In patients with metabolic syndrome, overnight levels of melatonin production (measured using urinary 6-oxymelatonin sulfate (6-SOMT) levels) were found to be depressed with a negative correlation between levels of this metabolite in urine and plasma levels of insulin, glucose, and leptin (1183). In patients with type 2 diabetes, however, nocturnal plasma melatonin levels were higher in obese subjects vs. nonobese subjects and lean nondiabetic controls (874).
        • Anti-inflammatory effects: In human research, melatonin had anti-inflammatory effects in infants with respiratory distress (11). Melatonin has been reported to decrease upregulation of proinflammatory cytokines in laboratory and human research (47;101;109;761;831;832). In animal research, melatonin has also been reported to reduce cardiac inflammatory injury induced by acute exercise (1184). Other anti-inflammatory effects may be related to inhibition of NO and malondialdehyde (MDA) production or an increase of glutathione levels (833;834); the suppression of TNF-alpha and IL-6 (115); the inhibition of phospholipase A2 (1185), mitogen-activated protein kinases (231), NF-kappaB (40;142;832;1184), or IL-4 and interferon (IFN)-gamma (62); or the regulation of mast cells (113). Use in inflammatory conditions has been proposed (1186). However, there is conflicting evidence from human trials, where melatonin induced a proinflammatory response, increasing levels of certain inflammatory cytokines, as well as plasma kynurenine concentrations in individuals with rheumatoid arthritis (464).
        • Antiobesity effects: Melatonin has been suggested as possibly playing a role in body weight control, possibly via inhibition of adipocyte differentiation (193) or reducing gut motility (873). Other animal research has indicated that exogenous melatonin lacks an effect on leptin secretion (192). In patients with type 2 diabetes, nocturnal plasma melatonin levels were higher in obese subjects vs. nonobese subjects and lean nondiabetic controls (874).
        • Antioxidant effects: There was a reduction (p<0.05) of malondialdehyde and 4-hydroxylalkenals to normal levels (p<0.05) in septic newborns treated with melatonin (225). In human research, melatonin improved antioxidant status (762). In human research, free radical scavenging was suggested as a possible mechanism of action in the protection against UV-induced erythema (711). In patients with metabolic syndrome, melatonin improved antioxidant status (609).
        • Many of melatonin's proposed therapeutic or preventive uses are based on its antioxidant activity (8;103;283;464;535;536;537;563;953;954;955;987;988;995;996;1001;1003;1004;1006;1007;1008;1009;1010;1011;1012;1013;1017;1018;1029;1034;1187;1188;1189;1190;1191;1192). There are many laboratory and animal studies of the antioxidant (free radical-scavenging) properties of melatonin (103;1020;1021;1022;1023;1024;1029;1034;1187;1188;1189;1190;1191;1192;1193;1194). As a result, melatonin has been proposed as a supplement to prevent or treat many conditions that are associated with oxidative damage. In vitro, melatonin has been observed to act as a direct free radical scavenger with the ability to detoxify both reactive oxygen and reactive nitrogen species (158;983;984;1005). Melatonin may reduce oxidative damage under a variety of conditions in which excessive free radical generation is believed to be involved (31;32;33;34;35;36;37;38;39;40;41;42;43;44;45;46;47;48;49;50;51;52;53;54;56;57;1003;1004). This reduction of oxidative damage has been observed in various animal models of ischemia and reperfusion injury (8;80;120;121;122;123;124;125;126;127;128;129;130;131;132;133;134;134;283;535;536;537;953;954;955;987;988;995;996;1001;1006;1007;1008;1009;1010;1011;1012;1013;1017;1018;1019), as well as in nerve tissues, including brain, spinal cord, optic nerve, and spinal cord white matter (991). However, in a rat model, melatonin was not effective in attenuating alcohol-induced loss of Purkinje cells (1036). Nonetheless, due to its high lipophilicity, melatonin is likely able to reach most tissues (1032).
        • In preclinical studies, melatonin protected against toxicity related to oxidative damage, such as alloxan-induced pancreatic toxicity (1002), 6-hydroxydopamine damage to neuronal PC12 cells, kainic acid injury (997;998;999), homocysteine-mediated oxidative stress (757;1000;1016), iron-induced oxidative injury (283;535;536;537), radiation-induced damage of various cell lines (992;993;994;1014;1015), ultraviolet light-induced damage (989;990), and copper-induced LDL oxidation (986). Melatonin has been reported as being a more efficient antioxidant than glutathione (1025), vitamin C (1026;1027), or vitamin E (1028;1029;1030;1031;1032;1033). Synergy has been observed with other antioxidants (1034;1035).
        • Melatonin was found to decrease the susceptibility of erythrocytes to oxidation during blood storage (328).
        • Some data suggest that melatonin as a free radical scavenger may inhibit the microsomal production of hydrogen peroxide in rats treated with aflatoxin (985), although more recent research reports that melatonin does not directly scavenge hydrogen peroxide (1195). Other conflicting evidence also exists: exogenous melatonin administered to rats via intraperitoneal injection increased photoreceptor susceptibility to light-induced damage; pinealectomy has been shown to protect photoreceptors, while subsequent melatonin injection has increased destruction (1037). It has been proposed that melatonin may reduce the amount of neurologic damage patients experience after stroke, based on antioxidant properties (953;954;955;956;957;958;959;960) . In addition, melatonin levels may be altered in people immediately after stroke (277;1196).
        • Melatonin produced an increase in the activity of antioxidant enzymes, glutathione peroxidase, and glutathione reductase in epileptic children receiving valproate (720;816), improved the hyperoxidative status of patients with muscular dystrophy (93), and reduced oxidative stress associated with exercise (930) and hypokinesis (1197). It has been suggested that such activity may help protect neurons from oxidative stress and damage.
        • In animal research, the antioxidant effects of melatonin were found to be less than those of taurine (468). Other studies have suggested a lack of antioxidant effects of melatonin (771;835). Further laboratory studies have been conducted, but details are lacking (1198;1199).
        • Antiparasitic effects: In animal research, melatonin therapy controlled Trypanosoma cruzi proliferation by stimulating the host's immune response (203;875).
        • Antiviral effects: In animal research, the protective effect of melatonin against Venezuelan equine encephalomyelitis virus was likely mediated by melatonin receptor activation (884).
        • Bone effects: Through free radical-scavenging and antioxidant properties, melatonin may impair osteoclast activity and bone resorption (916;917;918). In human research, melatonin lacked effects on bone density, NTX, or OC, although the NTX:OC ratio in the melatonin group was reduced (628).
        • Cardiovascular effects: In human research, low levels of platelet melatonin was found to be associated with angiographic no-reflow after primary percutaneous coronary intervention in patients with ST-segment elevation myocardial infarction (891). In patients with metabolic syndrome, melatonin reduced blood pressure and LDL cholesterol levels (609). In a poor-quality study, the inclusion of melatonin in the combined treatment of cardiovascular disease resulted in anti-ischemic, antianginal, antioxidant, and hypotensive effects (369).
        • In animal and human research, hypotension, blood pressure-lowering effects, and hypertension have been reported (359;360;361;362;363;364;365;366;367;369;443;506;609;756;825;826;827;828;829), although melatonin did not alter blood pressure in some animal or human research (376;830) or had mixed effects on night and day blood pressure, with decreases at night (375;377;378;608).
        • Melatonin may act directly on the cardiovascular system rather than modulating cardiac autonomic activity (892). In humans, melatonin, at doses as low as 1mg, has been observed to reduce blood pressure and decrease catecholamine levels after 90 minutes, possibly via direct effects on the hypothalamus, through antioxidant activity, by decreasing catecholamine levels, or by relaxing smooth muscle in the aorta wall (1200). In animal research, melatonin lowered blood pressure via GABA(A) receptors (825) and the reduction of oxidative load and restoration of the NO pathway (1201). However, a review of the role of melatonin in the pathology of the cardiovascular system noted that further evaluation of the clinical safety and efficacy of melatonin as an antihypertensive therapy is necessary and that such study must take into account melatonin's antagonism of calcium channel inhibitors (890). Dose-dependent relaxation of precontracted rat aorta and reduction of contractile response to alpha-adrenergic but not beta-adrenergic agonists have been observed (756). In healthy humans, melatonin is reported to decrease the pulsatility index and systolic and diastolic blood pressures, blunt noradrenergic activation (359;361;362), and increase cardiac vagal tone (363). In healthy postmenopausal women, hypotensive action has been observed only during hormone replacement therapy (364). High concentrations of melatonin may attenuate the reflex sympathetic increases that occur in response to orthostatic stress (1202).
        • Coagulation effects: According to experts, melatonin may decrease prothrombin time (a measurement of blood-clotting ability) (372;400). In humans, melatonin was associated with lower plasma levels of procoagulant factors, and dose-response relationships between the plasma concentration of melatonin and coagulation activity have been hypothesized (399). In animal research, melatonin enhanced platelet responsiveness (401).
        • Melatonin may play a role in mediating the influence of the psychoendocrine system and of lighting conditions on hematopoiesis. Melatonin seems to regulate hematopoietic cell growth by influencing apoptosis-related mechanisms (686). In humans, melatonin was associated with lower plasma levels of procoagulant factors, and dose-response relationships between the plasma concentration of melatonin and coagulation activity have been hypothesized (399).
        • Increased platelet counts after melatonin use have been observed in patients with decreased platelets due to cancer therapies (402;403;404;405;406;407;408). Stimulation of platelet production (thrombopoiesis) has been suggested but not clearly demonstrated. Cases of idiopathic thrombocytopenic purpura (ITP) treated with melatonin have been reported (409;409;410).
        • Cognitive effects: Although vigilance, reaction time, and tasks in humans undergo circadian variations, they do not seem to correlate with endogenous melatonin levels (418;423;1203). Exogenous melatonin may cause decrements in performance, including a slowing of choice-reaction time (412;428) or learning (413). Some studies have failed to confirm a decrement in performance (767;893;894), including a study of high-dose melatonin (50mg) in elderly persons (mean age: 84.5 years) (897). Animal research suggests a possible role of the GABAergic system (915).
        • Cytochrome P450 effects: Melatonin is metabolized in the liver via the hepatic microsome cytochrome P450 system, primarily (but not exclusively) by the CYP2C19 and CYP1A family (particularly CYP1A2) (911;912) and possibly CYP2C9. It appears to inhibit CYP1A2 (465;466;467) and induce CYP3A. In human research, concurrent use of fluvoxamine and melatonin resulted in increased levels of melatonin, likely due to reduced metabolism of melatonin by inhibiting CYP1A2 and/or CYP2C9 (465;466;467). Caffeine is reported to raise natural melatonin levels in the body (787) with a more pronounced effect in nonsmokers (788), possibly due to effects on the liver enzyme cytochrome P450 1A2 (788). This effect was more pronounced in nonsmokers (788). Other human studies suggest that interactions between exogenous melatonin and substrates metabolized by CYP1A2 may differ in individuals before and after smoking abstinence (913). In animal research, melatonin inhibited the activity of cytochrome P450 2E1, but to a lesser extent than taurine (468).
        • Dental effects: In human research, salivary and gingival crevicular fluid melatonin levels were lower in individuals with periodontal disease (914).
        • Dermatologic effects: Dermatologic use of melatonin has been proposed because of its immunomodulatory and antioxidant abilities. Study findings indicate that melatonin may accumulate in the stratum corneum (709). In human research, free radical scavenging was suggested as a possible mechanism of action in the protection against UV-induced erythema (711).
        • Exercise performance: In human research, during a heavy-resistance exercise session, melatonin increased the area under the curve of growth hormone (508). In human research, melatonin protected against the overexpression of inflammatory mediators and inhibited the expression of proinflammatory cytokines in exercising individuals (930). In human research, melatonin decreased exercise-induced increases in triglyceride levels and improved antioxidant status (762).
        • Endocrine effects: In clinical and laboratory studies, melatonin has been reported as producing varying hormonal effects. Such reports include changes in levels of luteinizing hormone (469;470;471;472;473;474;475;476), cortisol (477;478;479;480;481), progesterone (481;482;483;484;485), estradiol (482), thyroid hormone (T4 and T3) (486;487;488), testosterone (487;489), growth hormone (411;476;490;491;492;493;494), prolactin (411;495;496;497;498;499;500), oxytocin and vasopressin (490;501;502;503), adrenocorticotrophic hormone (478), and gonadotropin-inhibitory hormone (504). Melatonin has further been shown to alter pituitary hormone (LH and FSH) profiles in menopausal women to more "juvenile" profiles (488). In human research, in combination with estradiol treatment, melatonin reduced peak values of norepinephrine and increased epinephrine levels in some, but not all, stimulus situations (505;506;507). Effects on cortisol, norepinephrine, and epinephrine were lacking in some human research (625). In human research, during a heavy-resistance exercise session, melatonin increased the area under the curve of growth hormone (508).
        • In healthy young men, melatonin lacked an effect on suppressing hypothalamic-pituitary-adrenal system activity (495). However, melatonin did increase plasma prolactin concentrations (p<0.01) and reduce systolic blood pressure in the time interval following induced hypoglycemia (p<0.05). Researchers have studied the effects of the hypothalamic-pituitary-thyroid axis and melatonin in thermogenesis (body temperature regulation). Using a healthy male population, one study indicated that the pineal hormone may modulate hypothalamic-pituitary-thyroid axis function and affect body temperature (980). At night, when melatonin levels increase, thyroid-stimulating hormone (TSH) serum levels were lower and those of free thyroxine (FT4) increased. It has been proposed that administering melatonin to aging women may lead to a recovery of pituitary and thyroid function (488;626).
        • In animal research, melatonin was found to stimulate the release of gonadotropin-inhibitory hormone in quail (504) and inhibit gonadotropin-releasing-hormone-receptor-mediated oxytocin release in rats (501). In human research, exogenous melatonin enhanced the stimulatory effect of hypothalamic gonadotropin-releasing hormone on pituitary luteinizing hormone in women (472;473;474;475). This response to melatonin was found to be distorted in patients with menstrual abnormalities (1088;1204), was absent in postmenopausal women (1205;1206), and was not observed in men (944;1207;1208), in whom only a decrease in basal luteinizing hormone level was noted (469;470;476).
        • Melatonin decreased progesterone and estradiol plasma levels in healthy women (482) and enhanced the stimulatory effect of chorionic gonadotropin on progesterone production in cell culture (483). Research has suggested that melatonin may mimic the effect of drugs that act through the estrogen receptor interfering with the effects of endogenous estrogens, as well as those that interfere with the synthesis of estrogens by inhibiting the enzymes controlling the interconversion from their androgenic precursors (920). In contrast, melatonin has also been shown to significantly increase progesterone and androstenedione synthesis in bats (484). Melatonin is also involved in the control of testosterone secretion (489). In women with a luteal phase defect, melatonin improved serum progesterone concentrations; in these women, the melatonin concentration in the follicular fluid was correlated with progesterone concentrations (485). The authors suggested that the improved progesterone levels were due to antioxidant protection of the luteal cells.
        • Melatonin enhanced basal levels of growth hormone and its stimulation by hypothalamic-releasing hormone or exercise (490;491;492;493;494). This result may be mediated via the serotonin pathway (1209;1210;1211) or through naloxone-sensitive opioid-mediation (1212;1213). However, this effect has not been confirmed in other studies (945;1214). Exogenous melatonin has been reported to generate dopamine circadian rhythms in mice (1215).
        • Both endogenous (1216) and exogenous melatonin appeared to elevate plasma concentrations of prolactin without affecting the temporal pattern of its pulsatile secretion (496;497;498;499;500).
        • In human research, melatonin normalized the temporal pattern of plasma adrenocorticotropic hormone (ACTH) and cortisol concentrations and decreased cortisol nadir values in visually impared individuals (478).
        • Melatonin did not affect basal levels of cortisol in young men (500;1217). Cortisol increases are reported in older women, but not young women (481). However, melatonin was found to decrease circulating cortisol in goldfish (477).
        • In rats, melatonin significantly affected vasopressin secretion (502). It suppressed plasma arginine vasopressin in a rat model of hyperthyroidism (503). In humans, melatonin did not directly alter basal and angiotensin-2-stimulated arginine-vasopressin levels (1218).
        • Gastrointestinal effects: Research has been conducted concerning the effects of melatonin in patients with duodenal ulcers, dyspepsia, gastroesophageal reflux disease (GERD), and irritable bowel syndrome (IBS). In patients with GERD, a melatonin-containing supplement inhibited gastric acid secretion and synthesis of nitric oxide (311); nitric oxide affects lower esophageal sphincter relaxation, a major mechanism. Melatonin also influenced gastrointestinal motility in an animal model (765), regulated pancreatic secretion and maintained the integrity of the pancreas according to a review (1219), and affected bowel functions, reduced gut contractions induced by serotonin, and inhibited proliferation of epithelium, according to a review (766). In animal research, high doses of melatonin have been shown to inhibit motility by interacting with serotonin and CCK2 (765). Protective effects of melatonin in the gastrointestinal tract may be due to its effects on prostaglandins and cytoprotection from its free radical-scavenging activity (100;158;1005;1220;1221;1222).
        • Melatonin has also been shown to be beneficial in animal models of gastric ulcer by downregulating matrix metalloproteinases-9 and -3 (308). Melatonin showed similar benefit in a model of NSAID-induced gastropathy by preventing activation of the mitochondrially mediated apoptosis (by mitigating oxidative stress) (1223).
        • According to a review, in animal research, fasting lacked effects on melatonin-induced intestinal bicarbonate secretion (1052).
        • Graft effects: In animal research, melatonin improved the survival of human-derived ovarian grafts (1224).
        • Hepatoprotective effects: In patients with nonalcoholic steatohepatitis (NASH), use of melatonin resulted in improvements in liver function (445). In patients with steatohepatitis, melatonin decreased levels of proinflammatory cytokines, triglycerides, and GGTP (761). In human research, melatonin resulted in stable renal and liver function parameters after six weeks of use (755). Decreased transaminases have been shown in other human research (754). In one participant, treatment with melatonin resulted in increased alkaline phosphatase levels (390).
        • Hypolipidemic effects: There is some evidence of increases in cholesterol levels and atherosclerotic plaque buildup in human research (757) and animal research (759;760). In contrast, there are also reports of decreases in cholesterol levels in animal research (758) and decreased triglyceride and LDL cholesterol levels in human research (609;761;762).
        • Immune effects: Researchers have conducted studies concerning the effects of melatonin as they relate to cachexia, chemotherapy side effects, HIV/AIDS, ischemia-reperfusion injury, and sepsis. Researchers noted increased platelet counts after melatonin use in patients with decreased platelets due to cancer chemotherapy (402;403;404;406;407;408;533). Although not clearly demonstrated, studies have indicated possible stimulation of platelet production (thrombopoiesis). According to a review, activation of melatonin receptors was associated with the release of cytokines by type 1 T-helper cells (Th1), including gamma-interferon (gamma-IFN) and IL-2, as well as novel opioid cytokines (534). There is indirect evidence that melatonin may amplify the immunostimulatory effect of IL-2, as measured by an increase in the number of T-lymphocytes, natural killer cells, and eosinophils in cancer patients (862;1225;1226;1227;1228). In animal research, inhibition of the circadian synthesis of melatonin has been associated with reversible immunosuppression (1229) and elicited T cell autoimmunity in mice (5;1230). Melatonin has been reported to promote neutrophil apoptosis in patients receiving hepatectomy involving ischemia and reperfusion of the liver (283;535;536;537). A combination hormone therapy including melatonin was found to improve leukocyte function in ovariectomized aged rats (538).
        • Preliminary clinical studies suggest that combined therapy with low-dose subcutaneous IL-2 and melatonin improved the immune status in AIDS patients with CD4 cell counts below 200/mm3, who generally do not respond to IL-2 alone. The mean number of lymphocytes, eosinophils, T lymphocytes, natural killer (NK) cells, and CD25- and DR-positive lymphocytes increased with the treatment. An increase in the subjects' CD4:CD8 mean ratio was noted (544). In cancer patients who achieved disease control, melatonin induced a decrease in the number of regulatory T lymphocytes; this change was lacking in individuals with progressed disease (545).
        • According to a review, administration of melatonin agonists has reportedly reduced neophobia, and treatment with a melatonin antagonist during the dark period has reportedly reversed the anxiolytic-like effect of endogenous melatonin (312). A study in rats showed that melatonin attenuated kainic acid (KA)-induced neuronal death, lipid peroxidation, and microglial activation, and the number of DNA breaks (1231).
        • Other mechanisms through which melatonin may modulate the immune system, according to laboratory research, include the following: suppression of TNF-alpha, IL-1 beta, and IL-6 (101); inhibition of Th1 cells (114); stimulation of humoral activity/antibody production (532;539;540); inhibition of NF-kappaB (541), IKK, and JNK pathways (133); prevention of T cell apoptosis (542); and stimulation of mononuclear cell production (543).
        • Neurologic effects: It has been proposed that melatonin may reduce the amount of neurologic damage patients experience after stroke, based on its antioxidant properties (953;954;955;956;957;958;959;960).
        • According to a review, melatonin mechanisms are related to headache pathophysiology in many ways, including its anti-inflammatory effect, toxic free radical scavenging, reduction of proinflammatory cytokine upregulation, nitric oxide synthase activity and dopamine release inhibition, membrane stabilization, GABA and opioid analgesia potentiation, glutamate neurotoxicity protection, neurovascular regulation, 5-HT modulation, and its similarity in chemical structure to indomethacin (1232).
        • A significant body of basic research has indicated that melatonin may possess neuroprotective properties (79;167;168;169;170;171;172;173;174;175;176;177;179;180;181;182;183;184;184;185;186;187;187), meriting reviews in the contexts of neurodegenerative diseases (961), the peripheral nervous system (962), and traumatic nervous system injury (80). One possible mechanism may be the prevention of FGF9 downregulation (1233). Other possibilities, as revealed in experimental models of Parkinson's disease in mice, may involve the prevention of the induction of mitochondrial NO synthase (1234) or the inhibition of 6-hydroxydopamine production (1235). However, in an alternate rodent model of Parkinson's disease, melatonin was found to potentiate neurodegeneration (768). A number of preliminary animal studies have also suggested that melatonin may aid in recovery from or mitigate spinal cord injury (229;230;231); however, not all findings have been positive (1236).
        • Melatonin has also been suggested as a possible therapeutic strategy for Alzheimer's disease, having been shown in laboratory study to attenuate amyloid-beta-induced phosphorylation of tau-protein and prevent GSK-3beta activation and neuroinflammation (295), and mitigate amyloid-beta-mediated mitochondrial dysfunction (293) and amyloid-beta load (294).
        • Other animal research has indicated that melatonin preserved hippocampal cytochrome oxidase and sirtuin-1 expression following sleep deprivation (1237).
        • Opioid tolerance: In animals, researchers have concluded that melatonin acutely reversed and prevented tolerance to and dependence on morphine (966;967), and reduced the incidence of naloxone-induced withdrawal (967); however, the exact mechanism is not well understood.
        • Optic/ocular effects: In limited human research, melatonin stabilized vision in patients suffering from age-related macular degeneration (459). Preliminary human evidence also suggests that melatonin may decrease intraocular pressure in the eye (366;460;461); however, according to reviews, high doses of melatonin may increase intraocular pressure and the risk of glaucoma, age-related maculopathy, and myopia (382), as well as retinal damage (400).
        • Psychiatric effects: In patients with schizophrenia, melatonin lacked effects on P50 suppression (suggesting a lack of effect on sensory gating in these patients); however, increases in the P50 ratio occurred in some individuals with high baseline P50 suppression (1238).
        • Radioprotective effects: Due to its well-known antioxidant properties, it has been suggested that melatonin possesses a protective effect against damage caused by ionizing radiation, a hypothesis borne out of preliminary animal and in vitro research (206;207;215;216;217;219;969). The specific mechanisms may involve downregulation of apoptotic pathways via control of oxidative load (972).
        • Renal effects: In human research, melatonin resulted in stable renal function parameters after six weeks of use (755).
        • Reproductive effects: Due to its antioxidant potential and role in determining sexual status in certain mammals, melatonin has been suggested as a means of improving reproductive success in a variety of conditions (904). Exposure to exogenous melatonin increased sperm motility, ejaculate volume, sperm concentration, total sperm output, total function sperm fraction, and blood testosterone concentration, and decreased means of reaction time, dead sperm, abnormal sperm, and blood triiodothyronine concentration in Damascus goats (487). Melatonin implants have also been shown to improve semen characteristics in sheep (905). Melatonin treatment has been shown to regulate follicular development and oocyte competence in sheep (906) and goats (907). Additional veterinary research has indicated that melatonin may increase reproductive success (908;909;910) or activity (280). Similarly, melatonin has been shown to improve viability of sperm (899) and embryos (118;900;901;902;903) produced with in vitro fertilization. In vitro, melatonin has been shown to decrease kisspeptin gene expression, while increasing that of RFRP-3 in cell lines from rat hypothalamus (1239). In other research, melatonin lacked an effect on luteal blood flow or function in humans (1240). In a rat model of endometriosis, melatonin was shown to induce regression of endometriotic foci (97). Other reproductive applications of melatonin include the suppression of estrus (in cats) (1241), induction of estrus (in goats) (1242), and inhibition of gonad function (in fish) (1243).
        • Sleep effects: Melatonin, administered in the day or night in doses beyond the physiological range, appears to elicit a hypnotic effect. In human research, exogenous melatonin exerted hypnotic effects primarily when circulating levels of endogenous melatonin were low (653) and even very low doses may cause sleep when ingested before endogenous melatonin onset (418;649;655;661;662), although some studies have failed to confirm this finding (650). Also, in human research, melatonin has been shown to decrease the amount of anesthesia required during surgery (679;680;719;973). In human research, melatonin was found to potentiate the effects of gamma-amino butyric acid (GABA) and benzodiazepines and improve quality of sleep in combination with benzodiazepines (885). Melatonin may interact directly with the GABA-benzodiazepine-chloride ion channel as suggested in both human and animal research (792;1244), but not with the benzodiazepine receptor as suggested in human research (1245). In human research, as seen in functional magnetic resonance imaging, melatonin may play a role in priming sleep-associated brain activation patterns in anticipation of sleep (1246).
        • Randomized clinical trials have demonstrated some effect of melatonin on circadian rhythm entraining (1247;1248;1249). Endogenous circadian rhythmicity influences autonomic control of heart rate, and the timing of these endogenous rhythms may be altered by extended sleep/rest episodes and associated changes in the photoperiod, as well as by melatonin treatment (1250), with no evidence of changes in the duration of endogenous melatonin secretion or pituitary or gonadal hormones (1251). In visually impared individuals, disturbances of sleep and sleep-related neuroendocrine patterns may be caused by the absence of light cues. In individuals who are completely blind, a single administration of a pharmacological dose of melatonin improved sleep function by synchronizing the inhibition of pituitary-adrenal activity with central nervous sleep processes (478).
        • In human research, exogenous melatonin is able to shift circadian rhythms, as well as endogenous melatonin secretion and core body temperature (975;977;978;979); however, light appears to be a stronger regulator of circadian rhythm than melatonin itself (553;1071;1252;1253;1254;1255;1256;1257). The time of administration of melatonin is of critical importance, since it may cause both phase delay and phase advance. In human research it was determined that for phase delay, melatonin should be administered in the early morning; however, for phase advance, melatonin should be administered 1-2 hours before 9 p.m. (553).
        • Thermoregulating effects: The central thermoregulatory centers, including the preoptic area of the anterior hypothalamus, are likely to be involved with the regulation of core temperature following melatonin administration (1258). In human research, hypothermic effects of melatonin have been reported with doses from 15mg (657;975;976) and ingestion of 1.6mg of melatonin was reported to result in approximately 0.4ºC decrease of body temperature in humans (975;977;978;979). The hypothermic effects may be mediated by GABA receptor activity (1259). Melatonin may also influence chloride flux or other intracellular actions via a different mechanism that is not well understood (1245). Melatonin did not appear to exert hypothermic effects by central benzodiazepine receptors (1245). Melatonin appeared to increase heat loss and decrease heat production when taken during the day (1259). A parallel relationship was found between rectal core body temperature and the decline in sleep onset latency following melatonin administration (1259).
        • Toxicity protective effects: A number of studies have established melatonin's ability to prevent or mitigate damage from a number of chemical sources, including (but not limited to) methamphetamines (57;869), organophosphorus compounds (258;259;260), alcohol (261;870), nicotine (300), beta-cyfluthrin (871), and benzo(a)pyrene (872).
        • Vasodilating effects: In healthy male volunteers, melatonin significantly increased peripheral blood flow, as measured by distal to proximal skin temperature gradient and finger pulse volume (981). In human research, melatonin resulted in decreased renal blood flow velocity and conductance, increased forearm blood flow and vascular conductance, and lacked an effect on cerebral blood flow (376).
        • Vestibular effects: In human research, melatonin attenuated the muscle sympathetic nerve activity (vestibulosympathetic reflex) response to baroreceptor unloading, while lacking effects on the vestibulocollic reflexes (968).
        • Other: In models of experimental ischemia-reperfusion, melatonin reduced damage to tissues and limited cardiac pathophysiology (119).

        Pharmacodynamics/Kinetics:

        • Antioxidant activity: Melatonin was reported as having an IC50 of 1.46mcM and an inhibition constant of 0.82 ± 0.28mcM vs. lactoperoxidase in vitro (1260).
        • Cytochrome P450 effects: Melatonin appears to inhibit CYP1A2 (465;466;467) and induce CYP3A. In animal research, melatonin inhibited the activity of cytochrome P450 2E1, but to a lesser extent than taurine (468).
        • Absorption: The calculated oral bioavailability of melatonin was 3-76% (1053;1054;1059;1261). Some foods, such as oats, sweet corn, rice, ginger, tomatoes, bananas, and barley, contain small amounts of melatonin and may increase melatonin levels (1053;1054). Melatonin may be delivered transdermally in humans (1262;1263), or transmucosally to mimic physiological activity (100) and readily passes through the blood-brain barrier (158;1264). In human research, the peak value following a 2.5mg dose was 2,630pg/mL (returning to baseline within seven hours) and the peak value following a 75mg dose was 64,730pg/mL (not at baseline at last testing period) (417).
        • Distribution: Melatonin has been determined in both salivary and gingival crevicular fluid; however, a correlation between the two was lacking (914). Intraplatelet measurements of melatonin have been conducted (891).
        • Drug concentration levels: In human research, when administered in gelatin capsules, melatonin reached peak levels after 60-150 minutes (350-10,000 times higher than nighttime concentration) (1054;1265). In human research, melatonin supplementation was found to increase plasma levels of melatonin, occasionally into the daylight hours (446;508;665;692;754;761;1053;1054;1059).
        • In human research, it has been suggested that melatonin may be monitored by its serum metabolite, 6-sulphatoxymelatonin by radioimmunoassay (335;1122;1266), and in saliva (1267;1268), although melatonin concentrations measured in saliva did not consistently reflect absolute concentrations in blood (1267). Urine melatonin levels were found to correlate well with plasma levels (1269).
        • Salivary melatonin levels using an in-home sampling method were found to be in agreement overall with laboratory data and were useful for assessing circadian phase using dim light melatonin onset (DLMO) (1270).
        • According to a review, variations (daytime: 10pg/mL, nighttime: 30-120pg/mL) in the blood concentration of melatonin can be determined by radioimmunoassay, high-performance liquid chromatography, and gas chromatography-mass spectroscopy analytical techniques (347).
        • De Almeida et al. published a review on the measurement of melatonin in body fluids (1271).
        • In human research, fluvoxamine coadministration with melatonin increased the area under concentration-time curve (AUC) by 17-fold and the serum peak concentration (Cmax) of melatonin by 12-fold (822). It has been proposed that caffeine may increase the bioavailability of endogenous melatonin (887).
        • Metabolism: Melatonin was primarily inactivated by 6-hydroxylation in the liver, followed by conjugation and excretion as the sulfate or glucuronide. First-pass hepatic metabolism was extensive (up to 60% of the oral dose) (1053;1054;1059). 6-Sulphatoxymelatonin is an inactive metabolite of melatonin (1122;1266). In patients with liver cirrhosis, melatonin levels were elevated compared to controls (912;1272). Melatonin may be monitored by its serum metabolite, 6-sulphatoxymelatonin, by radioimmunoassay (1122;1266), and in saliva (335;1267;1268).
        • In human research, consumption of diets enriched with Jerte Valley cherries (a food source of melatonin) increased urinary levels of 6-sulfatoxymelatonin (1056).
        • In human case studies in patients with insomnia, a loss of response to melatonin treatment was found to be associated with a slow melatonin metabolism (1273). The authors suggested that in some patients, there may be decreased activity of CYP1A2, resulting in decreased melatonin metabolism and therefore persistent melatonin in plasma at levels of >50pg/mL up to four hours after supplementation.
        • Melatonin is metabolized in the liver via the hepatic microsome cytochrome P450 system, primarily (but not exclusively) by the CYP2C19 and CYP1A family (particularly CYP1A2) (911;912) and possibly CYP2C9. In human research, concurrent use of fluvoxamine and melatonin resulted in increased levels of melatonin, likely due to reduced metabolism of melatonin by inhibiting CYP1A2 and/or CYP2C9 (465;466;467). Caffeine is reported to raise natural melatonin levels in the body (787) with a more pronounced effect in nonsmokers (788), possibly due to effects on the liver enzyme cytochrome P450 1A2 (788). This effect was more pronounced in nonsmokers (788). Other human studies suggest that interactions between exogenous melatonin and substrates metabolized by CYP1A2 may differ in individuals before and after smoking abstinence (913). According to secondary sources, CYP1B1 is involved in the metabolism of melatonin.
        • Time to peak: In humans, melatonin secretion increased after the onset of darkness, peaked in the middle of the night (between 2 and 4 a.m.), and then gradually decreased (1). In human research, gender lacked an effect on the suppression of melatonin by light (1274). Pharmacological effects appeared to depend on the time of administration. Research shows that the time delay between administration and maximal effect varied linearly from 220 minutes at noon, to 60 minutes at 9 p.m. (566).
        • In older adults with insomnia complaints, the time to maximum level lacked differences when a low (0.4mg) and higher (2mg) dose were compared (755). The time to maximum level was 1.3-1.5 hours. The maximum concentrations were 405 ± 93pg/mL and 3,999 ± 700pg/mL (low and high dose, respectively). Use of 2mg of melatonin resulted in a maintenance of melatonin levels >50pg/mL for an average of 10 hours.
        • According to a review, melatonin has a rapid onset of action (Tmax of 30-60 minutes) (638).
        • Half-life: The physiologic half-life of melatonin was approximately 30-60 minutes (379;1054;1059;1275;1276). Nutritional supplements did not appear to mimic the physiologic release of melatonin, as dissolution testing has ranged from four to 12 hours (1277), with controlled-release formulations available (1278). In older adults with insomnia complaints, the apparent total clearance and elimination half-life lacked differences when a low (0.4mg) and higher (2mg) dose were compared (755). The elimination half-life was 1.8-2.1 hours, and the apparent total clearance was 379-478L per hour. According to a review, melatonin has a short elimination half-life (T1/2: 30-50 minutes) (638).
        • Excretion: Up to 85% of 6-hydroxymelatonin sulfate was excreted in urine; 6-sulphatoxymelatonin levels during melatonin treatment correlated with a normal circadian rhythm of excretion (200). In normal children, total 6-sulfatoxymelatonin excretion ranged from 11.1 to 40.2mcg (19.0 ± 7.4mcg) (742). Urine melatonin levels were found to correlate well with plasma levels (1269).

        History

        • According to secondary sources, the pineal gland was originally believed to be the seat of the soul by Descartes (1279). Physicians identified the pineal gland as belonging to the endocrine system in the early 1900s. Aaron Lerner, a dermatologist from Yale, and his team of researchers discovered melatonin in 1958 when investigating a potential treatment for vitiligo (1280).
        • According to secondary sources, melatonin's sedative properties emerged from research in the 1970s and 1980s, leading to its widespread use in the treatment of sleep disorders.
        • According to secondary sources, melatonin's popularity grew dramatically in the United States following reports in 1995 of its ability to promote sleep and alleviate jet lag, perhaps its best known application.
        • According to secondary sources, synthetic melatonin is sold in the United States as a food supplement due to small amounts found in some foods such as bananas and rice. Some authors suggest that this categorization interferes with the standardization and quality improvement of melatonin.

        Evidence Table

        ConditionStudy DesignAuthor, YearNStatistically Significant?Quality of Study
        0-2=poor
        3-4=good
        5=excellent
        Magnitude of BenefitARRNNTComments
        Delayed sleep phase syndrome (DSPS)Meta-analysisvan Geijlswijk, 2010Nine trialsYesNAMediumNANASmall studies and most studies were in children.
        Delayed sleep phase syndrome (DSPS)Meta-analysisBuscemi, 200514 trialsVariedNASmallNANAMelatonin was not effective in treating most primary sleep disorders with short-term use.
        Delayed sleep phase syndrome (DSPS)Systematic reviewBuscemi, 200449 trialsNANANANANAInvestigated the effect of melatonin on various sleep disorders.
        Delayed sleep phase syndrome (DSPS)Systematic reviewWagner, 199818 trialsNANANANANAMelatonin had mixed effects for insomnia associated with benzodiazepine tapering.
        Delayed sleep phase syndrome (DSPS)Randomized controlled trial, crossoverKayumov, 200122Yes4MediumNANAMelatonin 5mg or placebo.
        Delayed sleep phase syndrome (DSPS)Randomized controlled trial, crossoverRahman, 201020Yes3LargeNANAMelatonin 5mg or placebo.
        Delayed sleep phase syndrome (DSPS)Randomized controlled trialMundey, 200522Yes3SmallNANANo between-group comparison.
        Insomnia (children)Systematic reviewBuscemi, 200449 trialsNANANANANAInvestigated the effect of melatonin on various sleep disorders.
        Insomnia (children)Randomized controlled trialsvan Geijlswijk, 201072Yes5LargeNANAWell-designed study showing benefits of melatonin for insomnia in children.
        Insomnia (children)Randomized controlled trialsvan der Heijden, 2005110Yes4MediumNANAMelatonin 5mg or placebo. Study combined two previously published studies (Smits, 2001 and Smits, 2003).
        Insomnia (children)Randomized controlled trialSmits, 200370Yes4MediumNANA5mg of melatonin or placebo for four weeks.
        Insomnia (children)Randomized controlled trialSmits, 200140Yes4MediumNANAMelatonin 5mg.
        Insomnia (elderly)Systematic reviewBuscemi, 200449 trialsNANANANANAInvestigated the effect of melatonin on various sleep disorders.
        Insomnia (elderly)Systematic reviewWagner, 199818 trialsNANANANANAMelatonin had mixed effects for insomnia associated with benzodiazepine tapering.
        Insomnia (elderly)Randomized controlled trialWade, 2010791Yes5MediumNANAImproved outcomes in elderly subjects only.
        Insomnia (elderly)Randomized controlled trial, crossoverBaskett, 200340No5NANANAMelatonin 5mg, fast-release.
        Insomnia (elderly)Randomized controlled trial, crossoverGarfinkel, 199515Yes4MediumNANACrossover study.
        Insomnia (elderly)Randomized controlled trialWade, 2011791Yes3MediumNANAImproved outcomes in elderly subjects only.
        Insomnia (elderly)Randomized controlled trialLuthringer, 200940Yes3SmallNANA2mg of prolonged-release melatonin or placebo.
        Insomnia (elderly)Randomized controlled trialLemoine, 2007170Yes3LargeNANAMelatonin PR 2mg or placebo for three weeks.
        Insomnia (elderly)Randomized controlled trialWade, 2007354Yes3SmallNANAMelatonin PR 2mg vs. placebo for three weeks.
        Insomnia (elderly)Randomized controlled trialZhdanova, 200130 completersYes2MediumNANAThree melatonin doses (0.1, 0.3, and 3.0 mg) orally 30 minutes before bedtime for a week.
        Jet lagSystematic reviewBuscemi, 2006Nine trials (efficacy)NANANANANAOutcomes were similar for melatonin and placebo, but jet lag was not a specific endpoint.
        Jet lagSystematic reviewHerxheimer, 2001; Herxheimer, 200210 trialsNANANANA2Nine out of 10 trials suggested a benefit of melatonin.
        Jet lagSystematic reviewWagner, 199818 trialsNANANANANAMelatonin had mixed effects for insomnia associated with benzodiazepine tapering.
        Jet lagRandomized controlled trialSpitzer, 1999257No2NANANAMelatonin 5mg or 0.5mg, or placebo.
        Sleep disorders (individuals with behavioral, developmental, or intellectual disorders)Systematic review and meta-analysisRossignol, 201135 trials (systematic review); five trials (meta-analysis)YesNAMediumNANAOverall evidence of benefit in children with autistic spectrum disorders.
        Sleep disorders (individuals with behavioral, developmental, or intellectual disorders)Meta-analysisBraam, 2009Nine trialsYesNAMediumNANAPatients with intellectual disabilities.
        Sleep disorders (individuals with behavioral, developmental, or intellectual disorders)Systematic reviewMalow, 2012Eight trialsNANANANANAOverall evidence of benefit in children with autistic spectrum disorders.
        Sleep disorders (individuals with behavioral, developmental, or intellectual disorders)Systematic reviewGuénolé, 2011Eight trialsNANANANANAOverall evidence of benefit in children and adults with autistic spectrum disorders.
        Sleep disorders (individuals with behavioral, developmental, or intellectual disorders)Systematic reviewHollway, 201114 trialsNANANANANAOverall evidence of benefit in individuals (children and adults) with developmental disabilities.
        Sleep disorders (individuals with behavioral, developmental, or intellectual disorders)Systematic reviewBendz, 2010Four trialsYesNAMediumNANATwenty pieces of literature evaluated. Melatonin suggested to be safe and effective in children with ADHD.
        Sleep disorders (individuals with behavioral, developmental, or intellectual disorders)Systematic reviewRossignol, 200913 trialsYesNAMediumNANAMelatonin suggested as a novel treatment for symptoms associated with autism.
        Sleep disorders (individuals with behavioral, developmental, or intellectual disorders)Systematic reviewPhillips, 2004Three trialsNANANANANAPossibly effective in reducing sleep latency.
        Sleep disorders (individuals with behavioral, developmental, or intellectual disorders)Systematic reviewLancioni, 1999Eight trialsNANANANANAMixed effects depending on patient group.
        Sleep disorders (individuals with behavioral, developmental, or intellectual disorders)Systematic reviewWagner, 199818 trialsNANANANANAMelatonin had mixed effects for insomnia associated with benzodiazepine tapering.
        Sleep disorders (individuals with behavioral, developmental, or intellectual disorders)Randomized controlled trialGringras, 2012146Yes5SmallNANADisagreement between diaries and actigraphy with respect to total sleep time improvements.
        Sleep disorders (individuals with behavioral, developmental, or intellectual disorders)Randomized controlled trial, crossover, followed by open-label studyWasdell, 200851Yes5MediumNANAMelatonin 5mg vs. placebo.
        Sleep disorders (individuals with behavioral, developmental, or intellectual disorders)Randomized controlled trial, crossoverWeiss, 200627Yes5MediumNANATwo-phase treatment study; melatonin 5mg or placebo.
        Sleep disorders (individuals with behavioral, developmental, or intellectual disorders)Randomized controlled trialWright, 201120Yes4LargeNANAImproved sleep latency and duration in children with autism.
        Sleep disorders (individuals with behavioral, developmental, or intellectual disorders)Randomized controlled trialBraam, 200858Yes4MediumNANAMelatonin or placebo in intellectually disabled subjects.
        Sleep disorders (individuals with behavioral, developmental, or intellectual disorders)Randomized controlled trial, crossoverGarstang, 200611Yes4LargeNANAMelatonin 5mg vs. placebo; children with autism.
        Sleep disorders (individuals with behavioral, developmental, or intellectual disorders)Randomized controlled trial, crossoverWirojanan, 200918Yes3SmallNANA3mg at bedtime. Only sleep onset time significantly changed, though other measures showed tendencies toward improvement.
        Sleep disorders (individuals with behavioral, developmental, or intellectual disorders)Randomized controlled trialVan der Heijden, 2007107Yes3LargeNANAMelatonin 3mg or 6mg or placebo.
        Sleep disorders (individuals with behavioral, developmental, or intellectual disorders)Randomized controlled trial, crossover, followed by open-trialCoppola, 200425No3NANANAMelatonin 3mg vs. placebo in children with mental retardation.
        Sleep disorders (individuals with behavioral, developmental, or intellectual disorders)Randomized controlled trialBraam, 201066Yes2MediumNANAPatients with intellectual disabilities.
        Sleep disorders (individuals with behavioral, developmental, or intellectual disorders)Randomized controlled trial, crossover trialNiederhofer, 200320Yes2MediumNANAMelatonin vs. placebo in mentally retarded children.
        Sleep disorders (individuals with behavioral, developmental, or intellectual disorders)Randomized controlled trial; open label (two trials)Jan, 200016; 42No2NANANACompared fast-release melatonin and controlled-release melatonin.
        Sleep enhancement in healthy peopleMeta-analysisBrzezinski, 200517 studies; 284 participantsYesNASmallNANARigorous criteria for inclusion/exclusion. Limited by meta-analytic nature.
        Sleep enhancement in healthy peopleSystematic reviewBuscemi, 200449 trialsNANANANANAInvestigated the effect of melatonin on various sleep disorders.
        Sleep enhancement in healthy peopleSystematic reviewMorera, 20008 trialsNANANANA NAImprovement in sleep quality and latency.
        Sleep enhancement in healthy peopleSystematic reviewWagner, 199818 trialsNANANANANAMelatonin had mixed effects for insomnia associated with benzodiazepine tapering.
        Sleep enhancement in healthy peopleRandomized controlled trialWade, 2010791No5NANANAImproved outcomes in elderly subjects only. Lack of effect overall.
        Sleep enhancement in healthy peopleRandomized controlled trial, crossoverTzischinsky, 19948Yes5LargeNANALimited by male-only population.
        Sleep enhancement in healthy peopleRandomized controlled trialWyatt, 200626Yes4MediumNANALimited to healthy adults. Well-characterized study.
        Sleep enhancement in healthy peopleRandomized controlled trial (two parallel trials)Kunz, 2004 16Yes4MediumNANACrossover design.
        Sleep enhancement in healthy peopleRandomized controlled trialWade, 2011791Yes3MediumNANAImproved outcomes in elderly subjects only.
        Sleep enhancement in healthy peopleRandomized controlled trialPaul, 200423Yes3SmallNANAMagnitude of benefit for total sleep time and latency in melatonin compared to placebo.
        Sleep enhancement in healthy peopleRandomized controlled trialPinto, 200440Yes3MediumN/AN/AMale-only test population.
        Sleep enhancement in healthy peopleRandomized controlled trial, crossover Almeida Montes, 200310No3NANANAMelatonin 0.3 or 1mg or placebo.
        Sleep enhancement in healthy peopleRandomized controlled trialCajochen, 1996Eight (in each of two experiments)Yes3NANANALimited by homogeneity of subject population (all males of similar age).
        Sleep enhancement in healthy peopleRandomized controlled trial, single-blind, equivalence trialSatomura, 20017Yes2SmallNANAMelatonin vs. triazolam vs. placebo.
        Sleep enhancement in healthy peopleRandomized controlled trialEllis, 199615No2NANANAMelatonin 5mg or placebo.
        Aging (thermoregulation)Randomized controlled trialGubin, 200697Yes2SmallNANA1.5mg melatonin daily or placebo for two weeks.
        Alzheimer's disease/ cognitive declineSystematic reviewDe Jonghe, 2010Nine trialsNANANANANAOverall evidence of benefit from mainly poor-quality studies.
        Alzheimer's disease/ cognitive declineSystematic reviewJansen, 2006Three trialsNANANANANAInsufficient evidence to support the use of melatonin in managing cognitive and noncognitive sequelae of dementia.
        Alzheimer's disease/ cognitive declineRandomized controlled studyPeck, 200430Yes4SmallNANAMelatonin 1mg or placebo each night for four weeks.
        Anti-inflammatoryRandomized controlled trialForrest, 200775Mixed results3NoneNANAMelatonin group suggested proinflammatory action in individuals with rheumatoid arthritis.
        Anti-inflammatoryRandomized controlled trialGitto, 200440Yes1Medium-largeNANAIL-6, IL-8, and TNF-alpha levels lower in melatonin group (in preterm infants with respiratory distress).
        Benzodiazepine taperingRandomized controlled trial, crossoverGarfinkel, 199721Yes4MediumNANAElderly subjects received controlled release melatonin or placebo.
        Benzodiazepine taperingRandomized controlled trialCardinali, 200245No3NANANAPatients received 3mg of fast-release melatonin or placebo for six weeks.
        Benzodiazepine taperingRandomized controlled trialGarfinkel, 199934Yes3Large53%2Patients received 2mg of CR melatonin or placebo for six weeks.
        Benzodiazepine taperingRandomized controlled trial, crossoverPeles, 200780No2NANANAPatients received 5mg daily of melatonin or placebo for six weeks.
        Cancer treatmentMeta-analysisWang, 2012Eight trialsYesNAMediumNANAImprovements in remission when used as adjunct therapy.
        Cancer treatmentMeta-analysisMills, 200510 trials; 643 patients NANANANANAAuthors suggested potential for melatonin in treating cancer.
        Cancer treatmentSystematic reviewBlock, 2007Four trials of melatoninNANANANANAIndividual effects of melatonin lacking.
        Cancer treatmentSystematic reviewErnst, 2006,13 trialsNANANANANAAuthors mentioned encouraging but not conclusive results.
        Chronic fatigue syndromeRandomized controlled trialWilliams, 200242No1NANANALack of effect on symptoms or quality of life.
        Chronic obstructive pulmonary diseaseBefore-and-after studyde Matos, 201236Mixed3Medium (dyspnea); NA (lung function)NANAStatistics vs. placebo lacking. Main endpoint was exhalation of oxidation marker.
        Circadian rhythm sleep disorders (visually impaired and nonimpaired individuals)Systematic reviewKhan, 2011Two trialsNANANANANAImprovement in visually impaired children.
        Circadian rhythm sleep disorders (visually impaired and nonimpaired individuals)Randomized controlled trialPaul, 201111Yes1LargeNANAImprovement in circadian phase shift.
        DeliriumRandomized controlled trialAl-Aama, 2011145Yes3Medium19%5The odds ratio for delirium was 0.19.
        DepressionRandomized controlled trialSerfaty, 201033No5NANANA6mg of slow-release melatonin or placebo for four weeks.
        Exercise performanceRandomized controlled trial, crossoverMero, 200610No2NANANAOral ingestion of melatonin (6mg) during daytime with heavy resistance exercise. Effect on strength capacity lacking.
        FertilityRandomized controlled trialBatioglu, 201285No2NANANALack of effect on the number of mature oocytes.
        FertilityRandomized controlled trialEryilmaz 201163Yes2LargeNANALarge improvement in the primary outcome of oocyte quality; however, fertilization and pregnancy rates did not change.
        FertilityRandomized controlled trialRizzo, 201065Yes2MediumNANAIncrease in the number of mature oocytes; however, changes in fertilization and pregnancy were lacking.
        FibromyalgiaEquivalence trialHussain, 2011101Yes vs. baseline; unclear vs. fluoxetine2MediumNANAImprovement in Fibromyalgia Impact Questionnaire and quality of life.
        Gastrointestinal disordersSystematic reviewMozaffari, 2010Nine trialsNANANANANAImprovements in individuals with various types of gastrointestinal issues.
        Gastrointestinal disordersRandomized controlled trial, crossoverLu, 200932No4NANANA3mg of melatonin or placebo. A significant effect was observed in normal, but not IBS patients.
        Gastrointestinal disordersRandomized controlled trialSaha, 200718Yes3MediumNANA3mg of melatonin or placebo at bedtime for eight weeks.
        Gastrointestinal disordersRandomized controlled trial, crossoverLu, 200524Yes3Medium41%3Female patients received 3mg of melatonin or placebo for eight weeks. Incorrect method of randomization.
        Gastrointestinal disordersRandomized controlled trialSong, 200540Yes (abdominal pain)3LargeNANAFemale patients received 3mg melatonin or placebo at bedtime for two weeks.
        Gastrointestinal disordersRandomized controlled trialKlupińska, 200760Yes2Large79.3%25mg of melatonin or placebo for 12 weeks.
        Headache Systematic reviewFrancis, 2010One trialNANANANANAReduction in frequency in one study.
        Headache Randomized controlled trialAlstadhaug, 201048No4NANANALack of effect on migraine frequency, although there was some evidence of benefit for insomnia in some patients.
        Headache Randomized controlled trialLeone, 199620Yes2MediumNANAMelatonin 10mg.
        HepatitisRandomized controlled trialGonciarz, 201245Yes1SmallNANASmall improvements in AST and ALT.
        High blood pressure (hypertension)Meta-analysisGrossman, 2011Seven trialsNo (overall); Yes (for slow-release form)NASmall (slow release)NANAMeta-analysis suggested that effects of melatonin on nocturnal blood pressure were lacking. Subanalysis of studies using a slow-release form suggested blood pressure reductions.
        High blood pressure (hypertension)Randomized controlled trialGrossman, 200638Yes3SmallNANAPatients with controlled hypertension.
        High blood pressure (hypertension)Randomized controlled trialCagnacci, 200522Yes3MediumNANANo effects on diurnal BP but did reduce systolic, diastolic, and mean BP.
        High blood pressure (hypertension)Randomized controlled trialRechciński, 201062Yes2Small20%55mg of melatonin or placebo for 90 days.
        High blood pressure (hypertension)Randomized controlled trial, crossoverScheer, 200416Yes2SmallNANAMale patients with essential hypertension.
        High blood pressure (hypertension)Randomized controlled trialCagnacci, 200031Yes2SmallNANAEffect was only observed in HRT-treated women (not in those not on HRT).
        High blood pressure (hypertension)Randomized controlled trialCagnacci, 199712Yes2LargeNANAMelatonin 1mg.
        MenopauseSystematic reviewKelley, 2010One trialNANANANANALimited by inclusion of one study on melatonin.
        MenopauseRandomized controlled trialSecreto, 2004262NA5NANANASoy isoflavones (80mg) + melatonin (3mg), soy isoflavones alone, melatonin alone, or placebo for three months. Statistical analysis not done due to high placebo effect.
        Menopause Randomized controlled trialKotlarczyk, 201219Mixed4Small (physical symptoms)NANALack of effect on quality of life.
        Menopause Randomized controlled trialBellipanni, 200179Yes3MixedNANA3mg of melatonin or placebo at bedtime.
        PainRandomized controlled trialGitto, 201260Yes1LargeNANALack of double-blinding.
        Parkinson's diseaseSystematic reviewSeppi, 2011Two trialsNANANANANAPotential for improved sleep in PD patients based on small studies.
        REM sleep behavior disorderSystematic reviewAurora, 2010Six trialsNANANANANAGrade B recommendation from reviewers for melatonin.
        REM sleep behavior disorderRandomized controlled trialKunz, 20108Yes5SmallNANAVery small number of patients.
        SchizophreniaSystematic reviewAnderson, 2012Three trialsNANANANANAOverall mixed results for schizophrenia and tardive dyskinesia.
        Seasonal affective disorder (SAD)Randomized controlled trialLeppämäki, 200358Yes3MediumNANAHealthy adults with SAD.
        Seasonal affective disorder (SAD)Randomized controlled trialDanilenko, 200516 (SAD) + 17 (healthy controls)No2NANANAFemale patients with SAD.
        Seizure disorderSystematic reviewBrigo, 20124 studesNANANANANAConclusions lacking. Meta-analysis not possible.
        Seizure disorder Randomized controlled trialGupta, 200431No5NANANAPrimary outcome was subjective change in quality of life of add-on melatonin in children on valproate monotherapy via questionnaire.
        Sleep disturbanceSystematic reviewBuscemi, 200449 studiesNANANANANAInvestigated the effect of melatonin on various sleep disorders.
        Sleep disturbanceRandomized controlled trialSerfaty, 201033No5NANANA6mg of slow-release melatonin or placebo for four weeks.
        Sleep disturbanceRandomized controlled trialGögenur, 2009121No5NANANAIn postoperative patients.
        Sleep disturbance Randomized controlled trialIbrahim, 200632No5NANANAPatients with tracheostomy.
        Sleep disturbanceRandomized controlled trial Gupta, 200531Yes5MediumNANASleep parameters were affected by administration of melatonin in combination with valproate. Study in children with seizures.
        Sleep disturbance Randomized controlled trial, crossoverGarzón, 200922Yes4MediumNANAIn elderly patients with sleep or behavioral problems.
        Sleep disturbance Randomized controlled trialMedeiros, 200720Yes4SmallNANASleep quality was improved according to a sleep index.
        Sleep disturbance Randomized controlled trial, crossoverDowling, 200543Yes4SmallNANASmall improvements were observed.
        Sleep disturbance Randomized controlled trialCampos, 200422Yes4MediumNANAImproved sleep; no effects on asthma symptoms.
        Sleep disturbance Randomized controlled trialAndrade, 200133Yes (specific days)4UnclearNANAMelatonin average dose: 5.4mg.
        Sleep disturbance Randomized controlled trialO'Callaghan, 19997Yes4SmallNANASmall but clinically significant improvement.
        Sleep disturbanceRandomized controlled trialde Castro-Silva, 201019Yes3SmallNANAPatients with cystic fibrosis.
        Sleep disturbanceRandomized controlled trialGehrman, 200941No3NANANAPatients with Alzheimer's disease.
        Sleep disturbanceRandomized controlled trial, crossoverKoch, 200920Yes3SmallNANA3mg of melatonin or placebo over 18 weeks.
        Sleep disturbance Randomized controlled trialSuresh, 200740Yes3MediumNANAPatients with schizophrenia.
        Sleep disturbance Randomized controlled trial, crossoverShamir, 200019No3NANANAMelatonin 2mg in patients with tardive dyskinesia.
        Sleep disturbance Randomized controlled trial, crossoverShamir, 200014Yes3MediumNANA2mg of melatonin improved alertness test.
        Sleep disturbance Randomized controlled trialDolberg, 199819Yes3MediumNANAMelatonin patients with depression.
        Sleep disturbance Randomized controlled trialLeibenluft, 19975No3NANANAMelatonin 10mg had negative effects in patients with bipolar disorder.
        Sleep disturbance Randomized controlled trialDowling, 200850 completersNo2NANANANo effects of melatonin on nighttime sleep but improvement in daytime somnolence.
        Sleep disturbance Randomized equivalence trial, crossoverKemp, 20047No2NANANANo differences between melatonin and amitriptyline compared to baseline.
        Smoking cessationRandomized, controlled trial, crossoverZhdanova, 200012Yes (self-ratings); No (performance tests)3SmallNANAOnly self-ratings of mood were significantly affected.
        Surgical usesSystematic reviewYousaf, 201010 trialsNANANANANAEight of the studies looked at pain and anxiety, and two studies looked only at pain.
        Surgical usesRandomized controlled trialCaumo, 200963Yes5Medium66%2; 16 (depending on anxiety level)5mg following total abdominal hysterectomy. Melatonin as effective as standard treatment.
        Surgical usesRandomized controlled trialCapuzzo, 2006150No5NANANAStudy conducted in elderly patients.
        Surgical usesRandomized controlled trialBorazan, 201052Yes4MediumNANA6mg of melatonin or placebo the evening before and one hour prior to surgery.
        Surgical usesRandomized controlled trialIsmail, 200940Yes4MediumNANA10mg of melatonin or placebo 90 minutes prior to surgery.
        Surgical usesRandomized controlled trialCaumo, 200735Yes4MediumUnknown3Melatonin or placebo the night before and one hour prior to surgery.
        Surgical usesRandomized controlled trialOzcengiz, 2011100Yes3Small24%4Small reduction in agitation in children.
        Surgical usesRandomized controlled trialSamarkandi, 2005105Yes3SmallNANAStudy conducted in children.
        Surgical usesRandomized controlled trialAcil, 200466Yes2MediumNANAEffects lacking on postoperative psychomotor performance.
        Surgical usesRandomized dosing and equivalence trialKain, 2009148No2NANANAMelatonin vs midazolam. Midazolam was significantly more effective than melatonin.
        Tardive dyskinesiaSystematic reviewAnderson, 2012Three trialsNANANANANAOverall mixed results for schizophrenia and tardive dyskinesia.
        Tardive dyskinesiaRandomized, controlled trialCastro, 201113No4NANANALack of effect on TD symptoms in a small study.
        Tardive dyskinesiaRandomized, controlled crossover trialShamir, 200124Yes4SmallNANAPatients with schizophrenia and TD.
        Tardive dyskinesiaRandomized, controlled crossover trialShamir, 200019No4NANANA2mg of melatonin daily or placebo for four weeks.
        TinnitusRandomized controlled trialHurtuk, 201184Yes3Small32%3High dropout rate.
        TinnitusRandomized controlled trialLopez-Bonzalez, 2007120Unclear3NANANAMelatonin alone better than placebo; melatonin plus sulpiride most effective. Significance of results unclear.
        UlcersRandomized controlled trialCelinski, 201142Yes2Small14-29%4-7Increased gastric and duodenal ulcer healing.
        UlcersRandomized controlled trialCelinski, 201142Yes1Large50%2Increased ulcer healing.
        Urination (nocturia)Randomized, placebo controlled trial, crossoverDrake, 200420Yes3SmallNANAMen received 2mg of melatonin or placebo for four weeks.
        Urination (nocturia)Randomized controlled trialMerks, 201224No1NANANARandomization code was broken due to slow recruitment.
        UV-induced erythema prevention/sunburnRandomized controlled trialDreher, 19996No2NANANAMelatonin 1-2.5% with or without other agents.
        UV-induced erythema prevention/sunburnRandomized controlled trialFischer, 199920Yes2LargeNANAMelatonin 0.6mg/cm2 skin reduced erythema when applied before UV but not after.
        UV-induced erythema prevention/sunburnRandomized controlled trialBangha, 199620Yes2MediumNANAMelatonin 0.05, 0.1, 0.5% in gel. 0.5% reduced erythema eight hours after UV.
        Work shift sleep disorder Systematic reviewBuscemi, 2006Nine trials (427 subjects)NANANANANANo evidence that melatonin is effective in treating secondary sleep disorders.
        Work shift sleep disorderRandomized controlled trial, crossoverBjorvatn, 2007UnclearYes4ModerateNANAModest increases in sleep were reported (15-20 minutes).
        Work shift sleep disorderRandomized controlled trial, crossoverCavallo, 200545No3NANANAMelatonin 3mg before bedtime in the morning after the night shift.
        Work shift sleep disorderRandomized controlled trial (not double-blind)Dawson, 199536No1NANANAMelatonin 2mg may have had small effect on sleep quality but most endpoints were not significantly changed.

        Evidence Discussion

        Note

        • Due to the popularity of melatonin research, the evidence discussion paragraphs are limited to melatonin clinical trials. Conditions lacking high-quality human evidence (i.e., only case reports) are included in the Historical or Theoretical Indications section. Also, the table is limited to meta-analyses, systematic reviews, and randomized controlled trials. Clinical trials included in systematic reviews and meta-analyses are not necessarily described individually in detail.

        Delayed sleep phase syndrome (DSPS)

        • Summary: Delayed sleep phase syndrome is a condition that results in delayed sleep onset, despite normal sleep architecture and sleep duration. Several randomized controlled studies have reported improvements in sleep latency (390;752). Nonrandomized clinical trials have also been conducted (583;588) and, overall, the results agree with those observed in the RCTs. Further well-designed research is required before conclusions can be drawn.
        • Meta-analysis and systematic review: van Geijlswijk et al. conducted a systematic review and meta-analysis of nine studies (358;392;431;436;584;585;586;590;1281) to assess the effect of melatonin on sleep-wake rhythm in patients with DSPD (390). Relevant randomized, double-blind, placebo controlled studies published in English between January 1990 and September 2009 were pooled from PubMed and Embase. All randomized controlled trials included in the meta-analysis had to involve participants who had DSPD and had to include one or more of the following outcome measures: dim light melatonin onset (DLMO), sleep onset (SOT), wake-up time (WUT), sleep-onset latency (SOL), and total sleep time (TST). Studies assessing melatonin agonists or other comorbidities were excluded. Adult participants in the included studies were administered 0.3-5mg melatonin daily for two to four weeks. Children in the included studies were administered 3-6mg melatonin daily for 10-28 days. Information on standardization was lacking. Five of the studies reported headaches (approximately 6-7% of participants overall). Other adverse events seen with melatonin treatment included feeling cold, a mood dip, dizziness, and decreased appetite. One participant experienced generalized epilepsy following melatonin treatment. In one participant, treatment with melatonin resulted in increased alkaline phosphatase levels. Information regarding dropouts and interactions was lacking. Outcome measures included changes in DLMO, SOT, SOL, TST, and WUT. In adults, treatment with melatonin significantly advanced DLMO (-1.69 hours, 95% CI -2.31 to -1.07, Z=5.34), SOT (-0.70 hours, 95% CI -1.04 to -0.36, Z=4.08). Statistically significant changes in WUT, SOL, or TST were lacking. In children, treatment with melatonin significantly advanced DLMO (-1.13 hours, 95% CI -1.47 to -0.80, Z=6.62) and SOT (-0.64 hours, 95% CI -0.93 to -0.36, Z=4.42), as well as significantly reduced SOL (-16.04 minutes, 95% CI -23.77 to -8.32, Z=4.07) and significantly prolonged TST (28.39 minutes, 95% CI 13.06 to 43.72, Z=3.36). A statistically significant effect on WUT was lacking. The reviewers concluded that melatonin may be an effective treatment for DSPD, but that the review was limited by the small number of evaluated studies, which tended to have small participant populations, the inability to exclude the potential of carryover effect in the crossover studies, and the fact that most of the studies were conducted only in children.
        • Buscemi et al. conducted a meta-analysis and systematic review to evaluate the efficacy and safety of exogenous melatonin in the management of primary sleep disorders (752). A number of electronic databases were searched. Bibliographies of included studies and relevant reviews were reviewed, and hand-searching was conducted. Randomized controlled trials were eligible for the efficacy review, and controlled trials were eligible for the safety review. One reviewer extracted data, while the other verified the extracted data. The Random Effects Model was used to analyze data. Fourteen randomized controlled trials were relevant to the efficacy review, encompassing 279 participants (379;392;393;431;446;447;567;584;590;610;611;640;648;651). Melatonin decreased sleep onset latency (weighted mean difference [WMD]: -11.7 minutes; 95% confidence interval [CI], -18.2, -5.2); it was decreased to a greater extent in people with delayed sleep phase syndrome (WMD: -38.8 minutes; 95% CI, -50.3, -27.3; N=2) compared with people with insomnia (WMD: -7.2 minutes; 95% CI, -12.0, -2.4; N=12). Evidence of adverse effects from melatonin was lacking. The researchers concluded that there was a lack of evidence in support of melatonin for most primary sleep disorders with short-term use (four weeks or less); however, additional large-scale randomized controlled trials are needed before firm conclusions can be drawn. There is some evidence to suggest that melatonin is effective in treating delayed sleep phase syndrome with short-term use.
        • Buscemi et al. conducted a systematic review of 49 studies (379;392;393;415;424;432;446;447;448;470;559;560;584;590;610;612;622;640;648;649;650;651;653;655;657;658;672;675;698;699;700;701;704;717;885;976;1282;1283;1284;1285;1286;1287;1288;1289;1290;1291;1292;1293;1294) to assess the effects of melatonin on sleep disorders (764). Relevant studies published primarily in English were pooled from Medline (through June 2003), Cochrane Central Register of Controlled Trials (through the third quarter, 2003), Science Citation Index (through July 4, 2003), Biological Abstracts (through July 4, 2003), International Pharmaceutical Abstracts (through August 2003), NLM Gateway (through August 13, 2003), OCLC Papers First and Proceedings First (through July 11, 2003), and Toxline (through July 4, 2003). Studies published in non-English languages were included if the studies published in English were biased. Clinical trials, quasi-randomized controlled trials, prospective cohorts, case series, and reviews were included in this review. However, information from prospective cohorts and case series has been excluded from this summary. In addition, information related to the pharmacology or mechanism of action of melatonin has been excluded from this summary focusing on the clinical effects of treatment. Information regarding specific doses, frequency and duration of treatments, and standardization of treatment was lacking from the review. According to the reviewers, adverse effects included headaches, dizziness, nausea, and drowsiness. Information regarding toxic effects, dropouts, and interactions was lacking. Outcome measures included sleep onset latency, sleep efficiency, rapid eye movement (REM) latency, sleep quality, wakefulness after sleep onset, total sleep time, and percentage of time in REM sleep. Based on a pooled analysis of results from 20 studies, treatment with melatonin resulted in a significant reduction in sleep onset latency vs. placebo in healthy individuals (weighted mean difference (WMD) -3.92 minutes, 95% CI -5.28 to -2.55 minutes, Z=5.63, p<0.00001). However, funnel plot analysis indicated that possible publication bias existed. In individuals with primary sleep disorders, treatment with melatonin resulted in a significant reduction in sleep onset latency vs. placebo (WMD -10.66, 95% CI -17.61 to -3.72, Z=3.01, p=0.003). In individuals with secondary sleep disorders or in people experiencing sleep restriction, treatment with melatonin lacked a statistically significant effect on sleep onset latency. Based on a pooled analysis of results from 13 studies, treatment with melatonin resulted in a significant increase in sleep efficiency vs. placebo in healthy individuals (WMD 2.3%, 95% CI 0.7 to 3.9%, Z=2.83, p=0.005). In individuals with primary sleep disorders, treatment with melatonin lacked a statistically significant effect on sleep efficiency, sleep quality, wakefulness after sleep onset, total sleep time, or percentage time spent in REM sleep. In individuals with secondary sleep disorders, treatment with melatonin lacked a significant effect on sleep efficiency, wakefulness after sleep onset, and percentage of time in REM sleep, but showed a significant beneficial effect on total sleep time (WMD 15.6 minutes, 95% CI 7.2 to 24.0 minutes). In individuals experiencing sleep restriction, treatment with melatonin lacked a significant effect on sleep efficiency, sleep quality, wakefulness after sleep onset, or percentage of time in REM sleep, but showed a significant beneficial effect on total sleep time (WMD 18.2 minutes, 95% CI 8.1 to 28.3 minutes). Based on a pooled analysis of 11 studies, treatment with melatonin lacked a statistically significant effect on REM latency in healthy individuals. Overall, the reviewers gave melatonin a grade A for efficacy in treating sleep disorders, with a level of evidence of 1b. Limitations of this review included the lack of detailed information regarding dosing, duration of treatment, standardization of treatment, dropouts, and interactions that occurred during the included studies.
        • Wagner et al. conducted a systematic review to assess the clinical efficacy and safety of current therapies available for insomnia treatment (391). The effects of barbiturates, antidepressants, benzodiazepines, cloral hydrate, zaleplon, zolpidem, zopiclone, and valerian were assessed in this review but are excluded from this summary focusing on melatonin. Of the included studies that assessed melatonin, those that focused on only the pharmacokinetics of melatonin, in vivo results, or case reports were excluded from this summary; results for 18 included trials (370;379;393;411;417;463;482;584;612;649;650;651;653;655;657;658;1295;1296) and three reviews (1098;1227;1297) that assessed the clinical effects of melatonin are included in this summary. The authors performed a MEDLINE search to find case reports, reviews, abstracts, and clinical studies from April 1992 to December 1997. The references of the pooled articles were manually reviewed to identify additional relevant studies. Articles selected for inclusion had to include a specific patient population presenting with a particular condition, had to provide specific details regarding the treatment, and had to specifically describe the outcome measures used to assess the treatment. All included studies were published in English. According to the reviewers, doses of melatonin used for treatment generally ranged from 1-10mg in healthy participants or 0.3-10mg in patients with insomnia. Also, participants in some included trials were administered only physiologic doses of melatonin (0.1-0.3mg). Specific details regarding the duration of treatments was lacking. Information on standardization was lacking from the review. According to the reviewers, adverse effects were lacking in most of the included studies. However, in some studies, adverse effects included stomach cramps, dizziness, fatigue, headache, irritability, drowsiness in the day, increased depression, and inhibited ovarian function. Information on toxic effects, dropouts, and interactions was lacking in the review. Primary outcome measures included safety and efficacy outcomes of the different insomnia treatments. For inducing sleep, some studies concluded that high doses of melatonin were effective. The reviewers stated that melatonin has been used to treat jet lag and delayed sleep-phase syndrome, as well as decrease sleep latency and increase total sleep time. The reviewers reported that melatonin has been shown to have mild soporific effects when given in the day or early evening and hypnotic effects when taken before sleeping. Low doses of melatonin administered close to bedtime showed mixed results. According to the reviewers, melatonin may be the most beneficial when given two hours before sleeping rather than immediately prior to sleeping. Also, the reviewers concluded that melatonin may be more effective in treating insomnia in older individuals, as well as multi-disabled or neurologically impaired children. The authors concluded that melatonin has mixed effects depending on the patient population, dosage, time of administration and experimental design. The review was limited by lack of information regarding treatment frequency and duration, as well as specific data regarding treatment outcomes in the included studies.
        • Evidence: Kayumov et al. conducted a randomized, double-blind, placebo controlled crossover trial to investigate the effects of exogenous melatonin on delayed sleep phase syndrome (590). Patients with DSPS were included if they were over age 15 and did not do shift work, and had no presence of other sleep disorders, alcohol or drug abuse, use of psychotropic medications, active behavioral treatment, or severe psychiatric or neurological disorders. Twenty-two patients with delayed sleep phase syndrome received placebo or melatonin (5mg) daily for four weeks; these patients underwent a one-week washout period, and then were given the other treatment for an additional four weeks. Two consecutive overnight polysomnographic recordings were performed on three occasions: at baseline (before treatment), after four weeks of melatonin treatment, and after four weeks of placebo treatment. Twenty patients completed the study, with two subjects withdrawing (one subject moved and the other refused to sleep in the clinic). Sleep onset latency was reduced (20.2 minutes vs. 58.9 minutes for placebo vs. 35.8 minutes at baseline, p<0.05) while subjects were taking melatonin. Melatonin had no effect on total sleep time vs. baseline, but total sleep time was decreased in the placebo group vs. baseline. Melatonin lacked an effect on scores of subjective measures of sleepiness, fatigue, and alertness, which were administered at different times of the day. After an imposed conventional sleep period (from midnight to 8 a.m.), subjects taking melatonin reported being less sleepy and fatigued than they did while taking placebo. Noted adverse effects were lacking. The method of randomization was not indicated in this study. Otherwise, this study was well designed.
        • Rahman et al. conducted a randomized, double-blind, placebo controlled crossover trial to examine the hypothesis that exogenous melatonin (5mg) may attenuate symptoms of depression in delayed sleep phase syndrome patients (589). A total of 20 patients (13 males aged 35.6 ± 14.0 years and seven females aged 30.8 ± 12.4 years) with an established diagnosis of DSPS participated in the study. Subjects were excluded if they had other sleep disorders, did shift work, were under 16 years old, were alcohol or drug abusers, were current users of psychotropic medications or any other form of medication affecting melatonin secretion, were receiving active behavioral treatment, or had severe psychiatric or neurological disorders. Subjects were randomly allocated to the melatonin treatment group or placebo group. Each subject was given 5mg of melatonin or placebo capsule daily between 7 and 9 p.m. The total duration of the study was nine weeks. Before the randomization, a two-night sleep study was conducted to establish the baseline. After the second night of study and randomization, treatment continued for four weeks, followed by a one-week washout period before treatment crossover. Two-night sleep studies were later conducted again on days 27 and 28 (the first four-week treatment) and on days 62 and 63 (the second four-week treatment). Each group received the treatment of melatonin or placebo capsule between 7 and 9 p.m. daily. Subjects were primarily assessed by clinical interviews, psychometric evaluation for depression (Center for Epidemiologic Studies Depression Scale and Hamilton Depression Rating Scale-17), and overnight polysomnographic sleep studies (carried out at baseline and at the end of melatonin and placebo treatments). Melatonin secretion rhythm was measured by urinary 6-sulphatoxymelatonin level to determine circadian phase. Significant decreases (p<0.05) in HDRS-17 and CES-D scores after melatonin treatment were observed in group I (DSPS patients with comorbid depressive symptoms; N=8) while placebo treatment lacked an effect on either score. Changes in group II (DSPS patients without comorbid depressive symptoms; N=12) were not significant. The sleep onset latency was significantly reduced (p=0.03) in the melatonin treatment arm compared to both placebo and baseline in group I. Polysomnographic findings in group II indicated significant advance (p=0.03) in sleep onset latency in the melatonin treatment arm compared to placebo and baseline. The circadian profile of DSPS patients with marked depressive symptoms showed an abnormal pattern of endogenous melatonin secretion on placebo treatment; however, DSPS patients without depression showed normal melatonin production. This was a well-designed study; however, it should be noted that the study population was small.
        • Mundey et al. conducted a randomized double-blind placebo controlled trial to test the effectiveness of melatonin to advance the timing of sleep and circadian phase in individuals with delayed sleep phase syndrome (DSPS) (585). Twenty-two subjects were included. Subjects received 0.3 (N=5) or 3mg (N=4) melatonin or placebo (N=4) capsules taken for a period of four weeks, between 1.5 and 6.5 hours before dim light melatonin onset. The primary outcomes included the circadian phase shift markers dim-light melatonin onset (DLMO) and timing of the fitted nocturnal temperature minimum, measured both objectively and subjectively. Actigraphs (Actiware-Sleep, Mini Mitter Co., Inc., Bend, OR) measured objective sleep outcomes. Subjective outcomes (bed/lights-out times from sleep-diary) were compared to actigraphy data to calculate secondary outcome sleep latency. During the baseline week and on the fourth week of treatment, sleep onset, sleep offset, sleep efficiency, and sleep latency averages were calculated. The circadian phase of endogenous melatonin progressed in the groups after taking both doses of melatonin. Compared to baseline, advances (average time in hours ± SD) in dim-light melatonin onset (1.75 ± 0.89; p<0.001) and timing of the fitted nocturnal temperature minimum (1.63 ± 1.79; p<0.05) were noted in the melatonin groups. There was a strong correlation between the size of the phase advance in dim-light melatonin onset and the time that the subjects took the melatonin. The earlier the time of melatonin administration, the more effective it was (r2=0.94; p<0.0001). There was a correlation, although weak, between the time at which the subjects took the melatonin and change in sleep time. Limitations of this study include a lack of a power calculation, intent-to-treat analysis, and between-groups comparison. This study was supported by the Northwestern Drug Discovery Program and by AG00810. Standardization, adverse effects, toxic effects, and interactions were not discussed.
        • Studies of lesser methodological strength (not included in the Evidence Table): Nagtegaal et al. conducted a double-blind, placebo controlled, crossover trial to determine the importance of dim-light melatonin onset in patients suffering from DSPS (586). Patients were excluded if they were under 12 years of age, or had prior use of melatonin, liver diseases, renal failure, or psychiatric disorders. Thirty patients with DSPS received melatonin 5mg for two weeks in a double-blind setting and two weeks in an open setting, successively or interrupted by two weeks of placebo. Endpoints included measurements of the 24-hour curves of endogenous melatonin production and rectal temperature (N=14), polysomnography (N=22), actigraphy (N=13), sleep log (N=22), and subjective sleep quality (N=25). Mean dim light melatonin onset (± SD), before treatment, occurred at 23.17 hours (± 138 minutes). Melatonin was administered five hours before the individual dim-light melatonin onset. After melatonin treatment, the onset of the nocturnal melatonin profile was significantly advanced by approximately 1.5 hours. Melatonin use resulted in a significant reduction in sleep onset latency (15.3 minutes with melatonin vs. 25.3 minutes with placebo). Otherwise, an influence on sleep architecture was lacking. During melatonin treatment, patients reported feeling more refreshed in the morning. This study was not randomized, and the identical nature of supplements was not discussed.
        • Van Maaned et al. conducted a study to examine the effect of melatonin supplementation and then withdrawal on sleep, health, behavior, and parent stress in children with delayed Dim Light Melatonin Onset (N=41) (1298). The children were treated with melatonin for three weeks and were then discontinued, first by taking a half dose for one week, and then treatment was ceased for another week. Endpoints included sleep based on sleep diaries filled in by parents and actometers worn by children. During the stop week, sleep latency was longer, sleep start was later, and actual sleep time was shorter. Sleep efficiency also deteriorated and the Dim Light Melatonin Onset was no longer earlier (as it was during the treatment phase). Use of the melatonin was found to improve children's health and behavior, as well as parenting stress; treatment discontinuation resulted in deterioration of health only, with effects on behavior problems and parenting stress lacking.
        • In a case report, melatonin was used in a patient with a circadian rhythm sleep disorder (1299). Further details are lacking.
        • Nagtegal et al. conducted an open study to examine the effect of melatonin on quality of life in patients with delayed sleep phase syndrome (N=43) (591). Patients were treated with 5mg melatonin. A quality of life questionnaire was completed by the patients and responses were compared with other patients. Melatonin treatment was found to improve the scales, except for "role due to emotional problems." Further details are lacking.

        Insomnia (children)

        • Summary: Several randomized clinical trials, including at least one well randomized study, have shown benefits of melatonin in children with insomnia (356;392;431). Nonrandomized clinical trials have also been conducted (715;1300), and, overall, the results agree with those of the randomized controlled trials. A review of the effects of melatonin on sleep-wake disorders in children and adolescents was published by Jan et al. in 1999 (1301). More well-designed studies with a focus on safety in this population are needed.
        • Systematic review: Buscemi et al. conducted a systematic review of 49 studies (379;392;393;415;424;432;446;447;448;470;559;560;584;590;610;612;622;640;648;649;650;651;653;655;657;658;672;675;698;699;700;701;704;717;885;976;1282;1283;1284;1285;1286;1287;1288;1289;1290;1291;1292;1293;1294) to assess the effects of melatonin on sleep disorders (764). Relevant studies published primarily in English were pooled from Medline (through June 2003), Cochrane Central Register of Controlled Trials (through the third quarter, 2003), Science Citation Index (through July 4, 2003), Biological Abstracts (through July 4, 2003), International Pharmaceutical Abstracts (through August 2003), NLM Gateway (through August 13, 2003), OCLC Papers First and Proceedings First (through July 11, 2003), and Toxline (through July 4, 2003). Studies published in non-English languages were included if the studies published in English were biased. Clinical trials, quasi-randomized controlled trials, prospective cohorts, case series, and reviews were included in this review. However, information from prospective cohorts and case series has been excluded from this summary. In addition, information related to the pharmacology or mechanism of action of melatonin has been excluded from this summary focusing on the clinical effects of treatment. Information regarding specific doses, frequency and duration of treatments, and standardization of treatment was lacking from the review. According to the reviewers, adverse effects included headaches, dizziness, nausea, and drowsiness. Information regarding toxic effects, dropouts, and interactions was lacking. Outcome measures included sleep onset latency, sleep efficiency, rapid eye movement (REM) latency, sleep quality, wakefulness after sleep onset, total sleep time, and percentage of time in REM sleep. Based on a pooled analysis of results from 20 studies, treatment with melatonin resulted in a significant reduction in sleep onset latency vs. placebo in healthy individuals (weighted mean difference (WMD) -3.92 minutes, 95% CI -5.28 to -2.55 minutes, Z=5.63, p<0.00001). However, funnel plot analysis indicated that possible publication bias existed. In individuals with primary sleep disorders, treatment with melatonin resulted in a significant reduction in sleep onset latency vs. placebo (WMD -10.66, 95% CI -17.61 to -3.72, Z=3.01, p=0.003). In individuals with secondary sleep disorders or in people experiencing sleep restriction, treatment with melatonin lacked a statistically significant effect on sleep onset latency. Based on a pooled analysis of results from 13 studies, treatment with melatonin resulted in a significant increase in sleep efficiency vs. placebo in healthy individuals (WMD 2.3%, 95% CI 0.7 to 3.9%, Z=2.83, p=0.005). In individuals with primary sleep disorders, treatment with melatonin lacked a statistically significant effect on sleep efficiency, sleep quality, wakefulness after sleep onset, total sleep time, or percentage time spent in REM sleep. In individuals with secondary sleep disorders, treatment with melatonin lacked a significant effect on sleep efficiency, wakefulness after sleep onset, and percentage of time in REM sleep, but showed a significant beneficial effect on total sleep time (WMD 15.6 minutes, 95% CI 7.2 to 24.0 minutes). In individuals experiencing sleep restriction, treatment with melatonin lacked a significant effect on sleep efficiency, sleep quality, wakefulness after sleep onset, or percentage of time in REM sleep, but showed a significant beneficial effect on total sleep time (WMD 18.2 minutes, 95% CI 8.1 to 28.3 minutes). Based on a pooled analysis of 11 studies, treatment with melatonin lacked a statistically significant effect on REM latency in healthy individuals. Overall, the reviewers gave melatonin a grade A for efficacy in treating sleep disorders, with a level of evidence of 1b. Limitations of this review included the lack of detailed information regarding dosing, duration of treatment, standardization of treatment, dropouts, and interactions that occurred during the included studies.
        • Evidence: van Geijlswijk et al. conducted a randomized, double-blind, placebo controlled trial to assess the effects of melatonin on chronic sleep onset insomnia (CSOI) in children (N=72) (356). Children aged 6-12 years who experienced CSOI for at least five nights weekly for over one year were included if they lacked improvement with changes to sleep hygiene. In addition, all included participants had an average sleep onset latency (SOL) greater than 30 minutes and a normal hypnogram within the previous two months. Participants were excluded if they had an underlying psychiatric or pedagogic disease state, learning disability, pervasive developmental disorder, chronic pain, liver or kidney dysfunction, epilepsy, or a history of melatonin, stimulant, neuroleptic, benzodiazepine, clonidine, antidepressant, hypnotic, or beta-blocker use within the previous four weeks. Following a one-week run-in period, subjects were randomized to receive 0.05mg/kg melatonin, 0.1mg/kg melatonin, 0.15mg/kg melatonin, or placebo nightly (between 17:30-19:30) for one week. Melatonin was supplied by Pharma Nord, Denmark; however, further details regarding standardization were lacking. Side effects of melatonin included flushing, red earlobes and eyes, and yawning within one hour of taking melatonin (N=15); palor, dizziness, and feeling cold (N=8); higher rates of nighttime urination (N=2), and bedwetting (N=1). Side effects of placebo included headache (N=2); nausea and upset stomach (N=1); and dizziness and nausea (N=1). Information regarding toxic doses was lacking. Ten patients were excluded from the final analysis. Participants were excluded from the 0.05mg group due to parental concern about the trial negatively influencing the child's sleep behavior (N=1), lack of saliva sample (N=1), and dim light melatonin onset (DLMO) data that lacked the ability to be interpreted (N=1). In the 0.1mg group, two participants were excluded due to lack of appropriate actigraph data. In the 0.15mg group, participants were excluded due to development of mononucleosis (N=1) and lack of appropriate saliva samples (N=1). In the placebo group, participants were excluded due to lack of interpretable DLMO data (N=2) and poor adherence to the actigraph test (N=1). Although several participants were taking other medications during the trial, information regarding interactions was lacking. Outcome measures included changes in DLMO, sleep onset (SO), and SOL. Compared to the placebo group, participants treated with 0.05mg melatonin daily showed a statistically significant change in SO shift (42±10 minutes, p<0.001) and SOL shift (31±10 minutes, p=0.007). Compared to placebo, participants administered 0.1mg melatonin daily showed a significant change in DLMO shift (105±26 minutes, p<0.001), SO shift (50±11 minutes, p<0.001), and SOL shift (36±10 minutes, p<0.001). Compared to placebo, participants administered 0.15mg melatonin daily showed statistically significant changes in DLMO (91±24 minutes, p<0.001), SO shift (56±10 minutes, p<0.001), and SOL shift (42±9 minutes, p<0.001). Based on a bivariate correlation analysis, melatonin dose significantly correlated with all outcome measures when the placebo group was included. Also, DLMO shift significantly correlated with SO shift (r=0.38, p=0.003) and SOL shift (r=0.36, p=0.05) when the placebo group was included. However, if the placebo group was excluded, the dose-response correlation and the association of DLMO shift with SOL shift and SO shift was lacking. According to the authors, treatment with melatonin had beneficial effects on CSOI in children when administered 1-2 hours prior to bedtime. Limitations of this trial include a lack of data regarding long term outcomes, toxicities, allergies, and standardization. The trial also lacked power due to a small study population and reported differences in outcomes based on time of melatonin administration.
        • Van der Heijden combined the data of two randomized controlled trials (N=110, ages 6-12 years) (392;431) to investigate the efficacy of melatonin on pretreatment dim-light melatonin onset in children with chronic sleep onset insomnia (716). Inclusion and exclusion criteria were described previously with these references. The methodology was previously described separately for each study. The primary outcome measures for this study were pre- to post-treatment changes in dim-light melatonin onset (DeltaDLMO), sleep onset (DeltaSO), sleep latency (DeltaSL), and total sleep duration (DeltaTSD). Melatonin improved dim-light melatonin onset (+1:12 hours; p<0.001), sleep onset +0:42 hours; p=0.004), and sleep latency (25 minutes; p=0.019). Melatonin lacked a significant effect on total sleep duration vs. placebo. In the melatonin-treated group, but not in the placebo-treated group, the pretreatment dim-light melatonin onset was related to the DeltaDLMO [F(1, 29)=7.28, p=0.012] and the DeltaSO [F(1, 25)=7.72, p=0.010]. The time interval between treatment administration and pretreatment dim-light melatonin onset (INT) was only related to the DeltaSO [F(1,26)=5.40, p=0.028]. This study is limited by the republication of data. Otherwise, the study was well designed.
        • Smits et al. conducted a randomized, placebo controlled, double-blind trial to investigate the effect of melatonin treatment on health status and sleep in 70 children with idiopathic sleep-onset insomnia (392). The children, 6-12 years of age, had suffered for more than one year from idiopathic chronic sleep onset insomnia. They received 5mg of melatonin or placebo at 7 p.m. The study consisted of a one-week baseline period, followed by a four-week treatment. Adverse effects in the melatonin group included a cold feeling, decreased appetite, dizziness, and reduced mood. Eight patients in the melatonin group withdrew consent in the baseline period upon getting information on melatonin. There was also missing information from various patients in both groups (e.g., questionnaires, diaries). Health status was measured with the RAND General Health Rating Index (RAND-GHRI) and Functional Status II (FS-II) questionnaires. Lights-off time, sleep onset, and wake-up time were recorded in a diary, and endogenous dim-light melatonin onset was measured in saliva. The total scores of the RAND-GHRI and FS-II improved significantly more during melatonin treatment compared to placebo. The magnitude of change was much higher in the melatonin group than in the placebo group, with standardized response means for the RAND-GHRI of 0.69 vs. 0.07 and for the FS-II of 1.61 vs. 0.64. Compared with baseline, melatonin treatment significantly advanced sleep onset by 57 minutes, sleep offset by nine minutes, and melatonin onset by 82 minutes, and decreased sleep latency by 17 minutes. Lights-off time and total sleep time did not change. The method of randomization was not clear. Also, some of the children used medication, such as methylphenidates.
        • Smits et al. conducted a randomized, double-blind, placebo controlled trial to establish the efficacy of melatonin treatment in childhood sleep onset insomnia in 40 children (431). Elementary school children, 6-12 years of age, who had suffered for more than one year from chronic sleep onset insomnia were included. Children were excluded if they had severe learning disabilities, prior use of melatonin, liver or kidney disease, pain, use of various medications, and other sleep concerns (e.g., sleep maintenance concerns). Following a one-week baseline period, the children received 5mg of melatonin or placebo for four weeks. Adverse effects occurred in two children during the first two days of the melatonin treatment. Endpoints included lights-off time, sleep onset, and wake-up time, recorded in a diary (N=33 due to missing diaries in some patients), sleep onset (actigraph) (N=25), endogenous dim-light melatonin onset in saliva (N=27), and sustained attention (Bourdon-Vos reaction time test; N=36). Melatonin resulted in significant advancement in lights-off time by 34 (6-63) minutes, diary sleep onset time by 63 (32-94) minutes, actigraphic sleep onset by 75 (36-114) minutes, and melatonin onset by 57 (24-89) minutes. The total sleep time increased by 41 (19-62) minutes. Statistically significant differences between the treatment groups in the change of sleep latency, wake-up time, and sustained attention reaction times were lacking. Of the 38 children who finished the study and continued with open-label melatonin use, 13 stopped melatonin because their sleep problem was solved and one because sleep was not improved. One child developed mild generalized epilepsy four months after the start of the trial. Although he was initiated on valproate, the parents did not want to stop use of melatonin due to benefit in terms of sleep. Thus, he continued to use both valproate and melatonin with no further seizures. Limitations include the fact that the method of randomization was not clear. Also, some of the children used medication, such as methylphenidates.

        Insomnia (elderly)

        • Summary: Mean excretion of 6-sulfatoxymelatonin (a metabolite of melatonin) has been found to be lower in patients aged 55 years or older with insomnia than patients of the same age or younger (1302). Several randomized controlled studies have reported improvements in insomnia in the elderly with melatonin supplementation (373;374;379;389;446;613;615). Some nonrandomized clinical trials and clinical trials using combination treatments including melatonin have reported similar results (614;617;1303). Further well-designed research is required before firm conclusions can be drawn.
        • Systematic review: Buscemi et al. conducted a systematic review of 49 studies (379;392;393;415;424;432;446;447;448;470;559;560;584;590;610;612;622;640;648;649;650;651;653;655;657;658;672;675;698;699;700;701;704;717;885;976;1282;1283;1284;1285;1286;1287;1288;1289;1290;1291;1292;1293;1294) to assess the effects of melatonin on sleep disorders (764). Relevant studies published primarily in English were pooled from Medline (through June 2003), Cochrane Central Register of Controlled Trials (through the third quarter, 2003), Science Citation Index (through July 4, 2003), Biological Abstracts (through July 4, 2003), International Pharmaceutical Abstracts (through August 2003), NLM Gateway (through August 13, 2003), OCLC Papers First and Proceedings First (through July 11, 2003), and Toxline (through July 4, 2003). Studies published in non-English languages were included if the studies published in English were biased. Clinical trials, quasi-randomized controlled trials, prospective cohorts, case series, and reviews were included in this review. However, information from prospective cohorts and case series has been excluded from this summary. In addition, information related to the pharmacology or mechanism of action of melatonin has been excluded from this summary focusing on the clinical effects of treatment. Information regarding specific doses, frequency and duration of treatments, and standardization of treatment was lacking from the review. According to the reviewers, adverse effects included headaches, dizziness, nausea, and drowsiness. Information regarding toxic effects, dropouts, and interactions was lacking. Outcome measures included sleep onset latency, sleep efficiency, rapid eye movement (REM) latency, sleep quality, wakefulness after sleep onset, total sleep time, and percentage of time in REM sleep. Based on a pooled analysis of results from 20 studies, treatment with melatonin resulted in a significant reduction in sleep onset latency vs. placebo in healthy individuals (weighted mean difference (WMD) -3.92 minutes, 95% CI -5.28 to -2.55 minutes, Z=5.63, p<0.00001). However, funnel plot analysis indicated that possible publication bias existed. In individuals with primary sleep disorders, treatment with melatonin resulted in a significant reduction in sleep onset latency vs. placebo (WMD -10.66, 95% CI -17.61 to -3.72, Z=3.01, p=0.003). In individuals with secondary sleep disorders or in people experiencing sleep restriction, treatment with melatonin lacked a statistically significant effect on sleep onset latency. Based on a pooled analysis of results from 13 studies, treatment with melatonin resulted in a significant increase in sleep efficiency vs. placebo in healthy individuals (WMD 2.3%, 95% CI 0.7 to 3.9%, Z=2.83, p=0.005). In individuals with primary sleep disorders, treatment with melatonin lacked a statistically significant effect on sleep efficiency, sleep quality, wakefulness after sleep onset, total sleep time, or percentage time spent in REM sleep. In individuals with secondary sleep disorders, treatment with melatonin lacked a significant effect on sleep efficiency, wakefulness after sleep onset, and percentage of time in REM sleep, but showed a significant beneficial effect on total sleep time (WMD 15.6 minutes, 95% CI 7.2 to 24.0 minutes). In individuals experiencing sleep restriction, treatment with melatonin lacked a significant effect on sleep efficiency, sleep quality, wakefulness after sleep onset, or percentage of time in REM sleep, but showed a significant beneficial effect on total sleep time (WMD 18.2 minutes, 95% CI 8.1 to 28.3 minutes). Based on a pooled analysis of 11 studies, treatment with melatonin lacked a statistically significant effect on REM latency in healthy individuals. Overall, the reviewers gave melatonin a grade A for efficacy in treating sleep disorders, with a level of evidence of 1b. Limitations of this review included the lack of detailed information regarding dosing, duration of treatment, standardization of treatment, dropouts, and interactions that occurred during the included studies.
        • Wagner et al. conducted a systematic review to assess the clinical efficacy and safety of current therapies available for insomnia treatment (391). The effects of barbiturates, antidepressants, benzodiazepines, cloral hydrate, zaleplon, zolpidem, zopiclone, and valerian were assessed in this review but are excluded from this summary focusing on melatonin. Of the included studies that assessed melatonin, those that focused on only the pharmacokinetics of melatonin, in vivo results, or case reports were excluded from this summary; results for 18 included trials (370;379;393;411;417;463;482;584;612;649;650;651;653;655;657;658;1295;1296) and three reviews (1098;1227;1297) that assessed the clinical effects of melatonin are included in this summary. The authors performed a MEDLINE search to find case reports, reviews, abstracts, and clinical studies from April 1992 to December 1997. The references of the pooled articles were manually reviewed to identify additional relevant studies. Articles selected for inclusion had to include a specific patient population presenting with a particular condition, had to provide specific details regarding the treatment, and had to specifically describe the outcome measures used to assess the treatment. All included studies were published in English. According to the reviewers, doses of melatonin used for treatment generally ranged from 1-10mg in healthy participants or 0.3-10mg in patients with insomnia. Also, participants in some included trials were administered only physiologic doses of melatonin (0.1-0.3mg). Specific details regarding the duration of treatments was lacking. Information on standardization was lacking from the review. According to the reviewers, adverse effects were lacking in most of the included studies. However, in some studies, adverse effects included stomach cramps, dizziness, fatigue, headache, irritability, drowsiness in the day, increased depression, and inhibited ovarian function. Information on toxic effects, dropouts, and interactions was lacking in the review. Primary outcome measures included safety and efficacy outcomes of the different insomnia treatments. For inducing sleep, some studies concluded that high doses of melatonin were effective. The reviewers stated that melatonin has been used to treat jet lag and delayed sleep-phase syndrome, as well as decrease sleep latency and increase total sleep time. The reviewers reported that melatonin has been shown to have mild soporific effects when given in the day or early evening and hypnotic effects when taken before sleeping. Low doses of melatonin administered close to bedtime showed mixed results. According to the reviewers, melatonin may be the most beneficial when given two hours before sleeping rather than immediately prior to sleeping. Also, the reviewers concluded that melatonin may be more effective in treating insomnia in older individuals, as well as multi-disabled or neurologically impaired children. The authors concluded that melatonin has mixed effects depending on the patient population, dosage, time of administration and experimental design. The review was limited by lack of information regarding treatment frequency and duration, as well as specific data regarding treatment outcomes in the included studies.
        • Evidence: Wade et al. conducted a randomized, double-blind, placebo controlled, parallel group study to assess the effect of melatonin on primary insomnia in adult outpatients (N=791) (373). Men and women between the ages of 18-80 years who had primary insomnia (based on the Diagnostic and Statistical Manual of Mental Disorders fourth edition (DSM-IV) criteria) with a sleep latency of more than 20 minutes were included. Exclusion criteria were: 1) benzodiazepine or non-benzodiazepine hypnotic use in the preceding two weeks; 2) psychoactive treatment in the preceding three months; 3) sleep disorders due to psychiatric conditions or secondary to other conditions; 4) concomitant use of psychotropic treatments including antidepressants, antiepileptics, anxiolytics, barbiturates, first-generation antihistamines, hypnotics, lithium, neuroleptics, and treatments used as hypnotics such as barbiturates, benzodiazepines, buspirone, hydroxyzine, zaleplon, zolpidem, or zopiclone; 5) consuming alcohol to excess; or 6) leading a lifestyle that may interfere with sleep. In addition, participants who experienced short-term changes in their condition or who lacked compliance with treatment during the run-in period were excluded. A single-blind, placebo run-in period (two weeks) was followed by a double-blind treatment period (weeks 1-3) and a double-blind extension period (26 weeks: weeks 4-29). During the first double-blind period, participants were randomized in a 1:1 ratio to receive one tablet of 2mg prolonged-release melatonin (Circadin® 2mg, Neurim Pharmaceuticals Ltd., Tel Aviv, Israel) or placebo daily two hours before bedtime for three weeks. During the double-blind extension period, all participants initially randomized to the melatonin group continued with that treatment for an additional 26 weeks, while participants initially randomized to the placebo group were randomized in a 1:1 ratio to continue placebo treatment or to begin receiving melatonin treatment for 26 weeks. This was followed by a two-week, single-blind placebo run-out period. A description of standardization was lacking. Adverse events occurred in both placebo and treatment groups. Adverse events included arthralgia, diarrhea, headache, lower and upper respiratory tract infections, and nasopharyngitis. One subject experienced palpitations in the treatment group. A description of toxic effects was lacking. In the treatment group, reasons for dropouts included unwillingness to continue (N=59), occurrence of an adverse event (N=26), ineligibility to continue (N=9), lost to follow-up (N=9), lacking compliance (N=4), withdrawal of consent (N=1), and unknown reasons (N=3). In the placebo group, reasons for dropouts included unwillingness to continue (N=24), occurrence of an adverse event (N=10), ineligibility to continue (N=6), lost to follow-up (N=1), and unknown reasons (N=2). A description of interactions was lacking. The main outcome measure was change in sleep latency after the first three-weeks of treatment. Other outcomes included Clinical Global Impression of Improvement (CGI-I), Pittsburgh Sleep Quality Index (PSQI) score, and Quality of Life (based on the World Health Organzaton-5 Well-being Index). For participants aged 18-80 years who were low excretors (6-sulphatoxymelatonin levels ≤8mcg nightly) a significant benefit of melatonin treatment on sleep latency and most other outcome parameters was lacking. However, a reduction in sleep disturbances (PSQI component 5; -0.10, 95% CI -0.18 to -0.03, p=0.008) and quality of life (1.21, 95% CI 0.22 to 2.20, p=0.016) was observed for this group vs. placebo after three weeks. After six months of treatment, these participants showed a significant improvement in total sleep time (13.1 minutes, 95% CI 1.0 to 25.2, p=0.035), PSQI global scores (-0.66, 95% CI -1.30 to -0.01, p=0.046), WHO-5 Index scores (0.91, 95% CI 0.16 to 1.66, p=0.017), CGI-I scores (-0.25, 95% CI -0.49 to -0.01, p=0.042), and PSQI question two (-11.6, 95% CI -22.0 to -1.1 minutes, p=0.030) compared to the placebo group. For participants aged 65-80 years, a significant improvement in sleep latency (-15.6min, 95% CI -25.3 to -6.0, p=0.002), PSQI question two (-13.7, 95% CI -23.5 to -3.9, p=0.006), PSQI component 2 (-0.23, 95% CI -0.41 to -0.04, p=0.018), sleep maintenance (-0.17, 95% CI -0.33 to 0.00, p=0.046), time going to bed (-0.22, 95% CI -0.39 to -0.05, p=0.012), and PSQI global score (-0.64, 95% CI -1.25 to -0.02, p=0.042) was observed vs. placebo after three weeks. After six months of treatment, participants aged 65-80 years showed significant improvement in sleep latency (-14.5 minutes, 95% CI -21.4 to -7.7, p<0.001), time going to bed (-0.21 hours, 95% CI -0.33 to -0.08, p=0.002), PSQI global score (-0.70, 95% CI -1.17 to -0.23, p=0.003), component one of the PSQI score (-0.15, 95% CI -0.25 to -0.04, p=0.006), component two of the PSQI score (-0.24, 95% CI -0.38 to -0.10, p=0.001), PSQI score question two (-12.1 minutes, 95% CI -19.1 to -5.1, p=0.001), morning alertness (-0.10, 95% CI -0.19 to -0.01, p=0.032), and CGI score (-0.20, 95% CI -0.38 to -0.02, p=0.027) compared to the placebo group. This was a well-designed trial.
        • Baskett et al. conducted a randomized, double-blind, placebo controlled, crossover trial to determine the effect of melatonin on quality of sleep in healthy older people with age-related sleep maintenance problems (610). Patients were excluded if they were aged under 65 years with a score below 26 (out of 30) points on the mini-mental status examination (MMSE) or a score >6 on the geriatric depression score (GDS). Other exclusions included non-age-related sleep problems, advanced or delayed sleep phase syndrome, poor sleep hygiene, medical conditions significantly interfering with sleep (including sleep apnea), changes in medication during the study, use of hypnosedatives, or creatinine clearance <0.41mL per second. Twenty normal and 20 problem sleepers were included in the study from a larger sample of 60 normal and 60 problem sleepers. Twenty-four-hour urine 6-sulphatoxymelatonin was measured to estimate melatonin secretion in each participant. Patients were randomized to 5mg of melatonin, or matching placebo, at bedtime for four weeks, separated by a four-week washout period. Endpoints included sleep quality (sleep diaries), the Leeds Sleep Evaluation Questionnaire, and actigraphy. There was a significant difference between the groups in self-reported sleep quality indicators at entry, but a difference in melatonin secretion was lacking. Melatonin did not significantly improve any sleep parameter measured in either group. This study was well designed and well reported.
        • Garfinkel et al. conducted a randomized, controlled crossover trial of 15 elderly subjects who complained of long-term insomnia and were receiving different medications at baseline to measure 6-sulphatoxymelatonin levels and to assess the effect of controlled-release melatonin on sleep hygiene (379). Participants received tablets of 2mg of controlled-release melatonin (Circadin®, Neurim Pharmaceuticals, Tel Aviv, Israel) or a placebo. Sleep variables were measured for three consecutive nights at the end of the three-week treatment periods, and averages were taken. Wrist actigraphs (Somnitor®, Neurim Pharmaceuticals, Tel Aviv, Israel) assessed sleep-wake patterns. Indicators for sleep quality were assessed in the subjects for three nights in a row. A specific primary outcome was lacking; latency is assumed. Latency (time between bedtime and sleep onset), efficiency (total time asleep as a percentage of total time in bed), total sleep time (time spent asleep after sleep onset), and wake after sleep onset (WASO; accumulated time awake after sleep onset) were assessed. 6-Sulphatoxymelatonin excretion, onset, and peak time were also measured. Compared to the melatonin group, the placebo group exhibited a longer latency period. The mean sleep latency was 19 minutes [SE: 5; range: 3-49] in the melatonin group compared to the placebo period 33 minutes [SE: 7], but a statistically significant decrease was lacking (p=0.088). Sleep efficiency in the melatonin period showed an increase (83% [SE: 4]), when compared to the placebo period (75% [SE 3]) (p<0.001). Wake after sleep onset in the melatonin period was 49 minutes [SE: 14], and 73 minutes [SE: 13] in the placebo group (p<0.001). A significant effect of melatonin was lacking on total sleep time (p=0.49). It was hypothesized that, due to the short half-life of melatonin, the controlled-release melatonin is helpful for sleep maintenance. Limitations of this study include a lack of a power calculation and explanation of randomization. Toxic effects, standardization, and interactions were not discussed.
        • Wade et al. conducted a randomized, double-blind, placebo controlled trial to assess the effects of prolonged release melatonin (PRM) on primary insomnia (N=791) (374). Referred or self-referred individuals aged 18-80 years with primary insomnia were included in the study. Primary insomnia was defined as sleep latency >20 minutes (according to the Diagnostic and Statistical Manual for Mental Disorders, fourth edition). Individuals were excluded if they had taken hypnotics in the past two weeks or psychoactive treatment in the past three months, had a psychiatric disorder associated with their sleep disorder, had a sleep disorder that was secondary to another medical disorder (such as chronic pain) or lifestyle (such as shift work or jet lag), or consumed alcohol excessively. Also, participants were excluded if they used the following medications: antidepressants, antiepileptics, antihistamines (first generation), anxiolytics, barbiturates, hyptnotics, lithium, or neuroleptics. Following a two-week run-in period, during which time all participants received placebo treatment, participants were randomized to receive 2mg PRM (Circadin®) or placebo daily 1-2 hours before bed for three weeks. Following the first three weeks of treatment, participants in the placebo arm were randomized again in a 1:1 ratio to receive 2mg PRM or placebo for a 26-week extension period. All participants in the treatment group continued receiving 2mg PRM daily during the extension period. Information on standardization was lacking. Side effects that the authors reported to be possibly treatment-related included heart palpitations. Adverse effects associated with placebo treatment included burning sensation, labrynthitis, and pharyngolaryngeal pain. Overall, the frequency of adverse events was similar in the treatment and placebo groups. Information on toxic effects was lacking. Of the 791 participants randomized in the three week trial, 69 dropped out before the treatment was complete. The most common reasons for dropouts included consent withdrawal, lost to follow-up, or an adverse event. Of the 722 participants who completed the first three weeks of treatment, 711 were included in the extension phase. Of these participants, 156 withdrew due to consent withdrawal or an adverse event. Additional details regarding specific reasons for dropouts were reported by the authors. Information on interactions was lacking. The primary outcome measure was sleep latency, measured using a sleep diary. The Pittsburgh Sleep Quality Index (PSQI) global score, WHO-5 Well-being Index score (1998 version), and the Clinical Global Impressions Scale score for severity of illness (CGI-S) were also used to assess sleep quality. Compared to the placebo group, three weeks of treatment with melatonin lacked a statistically significant effect on sleep latency for participants aged 18-80 years and the subgroup of participants aged 18-54 years. However, for participants aged 55-80 years, treatment with melatonin for three weeks resulted in a significant difference in sleep latency vs. placebo (-15.4 vs. -5.5 minutes, p=0.014). When the older (55-80 years) and younger (18-54 years) subgroups of participants were compared, the effect of three weeks of melatonin treatment on sleep latency in the older participants population was significantly different vs. the younger participants (p=0.034). When all participants were considered, treatment with melatonin for three weeks resulted in significant difference in PSQI Q2 score (-7.8 minutes, 95% CI -13.4 to -2.2, p=0.006), PSQI C2 score (-0.13, 95% CI -0.24 to -0.02, p=0.023), time going to bed (-0.14 hours from midnight, 95% CI -0.24 to -0.03, p=0.011), and PSQI global score (-0.44, 95% CI -0.84 to -0.05, p=0.027) compared to the placebo group. Significant between-group differences in total sleep time and quality of life were lacking. For the younger participants, treatment with melatonin for three weeks lacked a significant effect on PSQI Q2 score, PSQI C2 score, total sleep time, time going to bed, sleep quality, and quality of life. For participants aged 55-80 years, treatment with melatonin for three weeks resulted in a statistically significant between-group difference in PSQI Q2 score (-9.5 minutes, 95% CI -5.8 to -3.3, p=0.003), PSQI C2 score (-0.17, 95% CI -0.29 to -0.05, p=0.005), time going to bed (-0.15 hours from midnight, -0.26 to -0.03, p=0.014), total sleep time (0.15 hours, 95% CI 0.00 to 0.31, p=0.048), sleep quality (-0.65, 95% CI -1.09 to -0.21, p=0.003), and quality of life (0.65, 95% CI 0.12 to 1.19, p=0.017) compared to the placebo group. Considering all ages after six months of treatment, participants administered melatonin showed significant difference in sleep latency based on diary (-6.0 minutes, 95% CI -10.0 to -2.1, p=0.003), PSQI C2 score (-0.10, 95% CI -0.20 to -0.01, p=0.032), PSQI Q2 score (-6.8 minutes, 95% CI -10.9 to -2.6, p=0.001), time going to bed (-0.13 hours from midnight, 95% CI -0.20 to -0.05, p=0.002), sleep quality (-0.39, 95% CI -0.71 to -0.08, p=0.014), quality of sleep (-0.08, 95% CI -0.15 to 0.00, p=0.046), daytime functioning (-0.07, 95% CI -0.13 to 0.00, p=0.040), morning alertness (-0.07, 95% CI -0.13 to 0.00, p=0.047), quality of life (0.46, 95% CI 0.11 to 0.81, p=0.011), and clinical status (-0.12, 95% CI -0.24 to -0.01, p=0.036) vs. placebo. The participants aged 18-54 years lacked these between-group differences, while participants aged 55-80 years showed all of these differences except for morning alertness after six months of treatment with melatonin vs. placebo. The authors concluded that PRM has both short-term and long-term benefits on insomnia, particularly for individual aged 55-80 year. The study was limited by the high dropout rate, and lack of blinding and randomization description.
        • Luthringer et al. conducted a randomized, double-blind, sleep laboratory phase II study to investigate the effects of prolonged-release melatonin (PRM) on sleep and daytime performance in patients with insomnia (615). Subjects were 40 patients aged 55 years or older with primary insomnia, as defined by the fourth revision of the Diagnostic and Statistical Manual of Mental Disorders of the American Psychiatric Association. Patients with other sleeping disorders (related to breathing disorders or restless leg/periodic leg movements), a history of severe cardiac or neurologic disorders, or those who took sedative hypnotics in the preceding month were excluded from this study. Patients were treated with a nightly single-blinded placebo for two weeks. They were then randomized in a double-blind fashion to prolonged-release melatonin (PRM) 2mg or placebo two hours before bedtime for three weeks. Following the double-blind randomization, a withdrawal period in which no drugs were given was conducted for three weeks. Eleven patients in each treatment group reported adverse events; however, none were considered to be treatment related. Sleep was assessed objectively using polysomnography, all-night electroencephalography spectral analysis, and subjective questionnaires (the Leeds Sleep Evaluation Questionnaire, LSEQ). Daytime psychomotor performance was objectively assessed using the Leeds Psychomotor Tests battery. At the end of the double-blind treatment period, the PRM treatment group had significantly shorter sleep onset latency (nine minutes; p=0.02) compared to placebo group. The PRM group also had significantly better Critical Flicker Fusion scores (p=0.008), without negative effects on sleep structure and architecture. Based on the LSEQ questionnaire, quality of sleep was significantly improved in the PRM group (p=0.004), but changes were lacking in the placebo group. More patients in the PRM group also reported substantial improvement in sleep quality at home compared to placebo (50% vs. 15%; p=0.018). However, a significant difference in objective measures of sleep quality was lacking between the groups. The authors concluded that nightly treatment with PRM effectively induced sleep and improved perceived quality of sleep. It should be noted that this study was funded by the manufacturer of the melatonin supplement used in this study.
        • Lemoine et al. conducted a randomized, double-blind, placebo controlled, parallel study to assess the efficacy and safety of melatonin in improving quality of sleep and morning alertness in 170 patients aged 55 years or older (389). Patients had primary insomnia for at least one month and had consistent complaints of poor sleep quality during the single-blind placebo run-in period. Patients with breathing-related sleep disorder, circadian rhythm sleep disorder, dyssomnia not otherwise specified, sleep disorder because of a general medical condition, significant psychiatric or neurological disorders (anxiety, depression, dementia), or use of any medications that affected the central nervous system or sleep-wake function within two weeks prior to the first day of the placebo run-in period were excluded. The study began with a two-week, single-blind, run-in phase with placebo treatment followed by an evaluation. Eligible patients were randomized to placebo or prolonged-release (PR) melatonin 2mg (Circadin®; Neurim Pharmaceuticals Ltd., Tel Aviv, Israel) for three weeks. Subjects were instructed to take the medication daily after the evening meal, between one and two hours before bedtime, preferably between 9 and 10 p.m. Prolonged-release melatonin significantly improved quality of sleep (p=0.047), quality-of-sleep ratings (p=0.003), and morning alertness (p=0.002) compared with placebo. The authors suggested that the improvements in quality of sleep and morning alertness were linked, indicating a beneficial treatment effect on the restorative value of sleep. Rebound insomnia or withdrawal effects were lacking upon treatment discontinuation. Adverse effects were mild and an increased occurrence was lacking in the melatonin group. Randomization and blinding were not described.
        • Wade et al. conducted a randomized, double-blind, placebo controlled study to evaluate whether or not treatment with prolonged release (PR)-melatonin would improve quality of sleep and next day alertness in 354 older patients with primary insomnia (613). Patients with primary insomnia were included following exclusion of neuropsychiatric disorders and other potential reasons for development of insomnia. Patients were randomized to receive 2mg of melatonin (N=177; Circadin®, Neurim Pharmaceuticals Ltd., Tel Aviv, Israel) or placebo (N=177) for three weeks, one tablet two hours before bedtime. Twenty patients failed to complete visit 3 (eight in the melatonin group; 12 in the placebo group), with 334 patients comprising the full analysis set. This treatment period was preceded by a two-week run-in period. Endpoints included quality of sleep and morning alertness. Improvements were noted in quality of sleep and morning alertness (26% vs. 15%; p=0.014). A statistically significant and clinically relevant shortening of sleep latency to the same extent as most frequently used sleep medications was also found (-24.3 vs.-12.9 minutes; p=0.028). Quality of life also improved (p=0.034). Blinding was not described.
        • Zhdanova et al. conducted a randomized, double-blind, placebo controlled trial to determine if melatonin at various doses could restore sleep in elderly subjects, 30 of which completed the study (446). Inclusion criteria included chronic insomnia and increased daytime sleepiness (or frequent napping), which gradually developed with advancing age (after subjects reached 40 years of age) and were present for at least one year before the recruitment. Subjective symptoms (International Classification of Sleep Disorders) consistent with psychophysiological insomnia as well as objective symptoms (polysomnographic (PSG) criteria showing latency to sleep onset of more than 30 minutes) also had to be present. Exclusion criteria included major psychiatric diagnosis on Axis I of the DSM-IV; regular use (once a week or more) of hypnotics, melatonin, stimulants, or other medications that may affect melatonin levels, and unwillingness or inability to discontinue the occasional use of these medications for at least four weeks before beginning the protocol; and acute or unstable chronic conditions (e.g., diabetes, uncontrolled thyroid disease; kidney, prostate, or bladder conditions; congestive heart failure, angina, or other severe cardiovascular disorders; hepatitis; asthma or severe respiratory allergies; stroke; cancer if less than one year since the end of treatment; conditions associated with chronic pain; neurological disorders; more than "moderate" alcohol use; unwillingness or inability to maintain a regular sleep-wake cycle during the entire period of the study; sleep apnea/hypopnea index >10,; and periodic (more than 10) limb movements in sleep per hour). Thirty subjects completed the study, but the number starting the study is not clear. Subjects received a placebo on alternate (odd-numbered) weeks throughout the study, starting with the first "run-in" week, thus providing washout periods between and after active treatments. During the second week and thereafter on each even-numbered treatment week, subjects received melatonin (0.1, 0.3, or 3.0mg) mixed with microcrystalline cellulose or the placebo (cellulose) daily, administered for four days at home and then for three days as inpatients in a research center. Melatonin or placebo capsules were ingested half an hour before each subject's fixed bedtime. Treatments were separated by one-week washout periods. Sleep data were obtained by polysomnography on the last three nights of each treatment period. The physiologic melatonin dose (0.3mg) restored sleep efficiency (p<0.0001), acting principally in the middle third of the night; it also elevated plasma melatonin levels (p<0.0008) to normal. The highest dose (3.0mg) was also effective in improving sleep; however, it also induced hypothermia and caused plasma melatonin to remain elevated into the daylight hours. Melatonin did not affect the sleep of the control subjects, even though they also had low melatonin levels. Description of randomization, blinding, and withdrawals is lacking.
        • Studies of lesser methodological strength (not included in the Evidence Table): Haimov et al. conducted a double-blind clinical trial to investigate the effects of melatonin replacement therapy on melatonin-deficient elderly insomniacs (611;612). The study comprised a running-in, no-treatment period and four experimental periods. During the second, third, and fourth periods, subjects were administered tablets for seven consecutive days, two hours before desired bedtime. The tablets were 2mg of melatonin administered as sustained-release or fast-release formulations, or an identical-looking placebo. The fifth period, which concluded the study, was a two-month period of daily administration of 1mg of sustained-release melatonin two hours before desired bedtime. During each of these five experimental periods, sleep-wake patterns were monitored by wrist-worn actigraphs. Analysis of the first three one-week periods revealed that a one-week treatment with 2mg of sustained-release melatonin was effective for sleep maintenance (i.e., sleep efficiency and activity level) of elderly insomniacs, while sleep initiation was improved by the fast-release melatonin treatment. The authors stated that sleep maintenance and initiation were further improved following the two-month, 1mg sustained-release melatonin treatment, indicating that tolerance had not developed. After cessation of treatment, sleep quality deteriorated. This study was not randomized.
        • Valtonen et al. conducted two long-term double-blind, placebo controlled studies to investigate the effects of a melatonin-rich nighttime milk on sleep and activity in institutionalized elderly patients (618). Inclusion and exclusion criteria were lacking. In study 1, 70 patients receiving various medications for chronic illnesses participated in a crossover study. The subjects were divided into two groups (I and II). Group I (N=31) used night milk (experimental) for eight weeks and normal commercial milk (placebo) for eight weeks, with a washout period of one week in between. Group II (N=31) started with normal daytime milk for eight weeks and switched to night milk after the washout period. Each subject drank about 0.5L of milk daily. The study was performed between March and July. Based on subjective evaluation, quality of sleep was good during the whole study, with almost all scores between 8 and 10 (10=most restful). A difference in sleep quality was lacking between night milk and normal milk periods for group I, but for group II, which consumed night milk second, sleep quality decreased during the later period (p<0.01). Sleep quality was reduced towards the summer, indicating a seasonal effect that was more evident in subjects with severe dementia. This seasonal effect was statistically significant. In study 2, the design was changed slightly, since the subjective evaluation in study 1 was considered to be too limited. Also, the parameter most strongly related to season (light and melatonin) was nocturnal activity. Eighty-one subjects participated in this study, and all subjects consumed milk regularly. The participants were divided into three groups. Group III (N=23) started by consuming night milk for eight weeks and then, after a washout period of one week, consumed normal daytime milk for another eight weeks. Group IV (N=26) started with normal milk for eight weeks and changed to night milk after the washout period. Group V (N=32), living in another rest home, consumed normal daytime milk during the whole experiment and served as a control group for evaluating the effect of season. The study was performed between October and February. For Group III, sleep quality was rated significantly (p<0.001) higher during the normal milk period. Changes in morning or evening activity, and an effect of night milk, were lacking. In Group IV, sleep quality, morning activity, and evening activity increased (p<0.001) when night milk was consumed. In Group V, sleep quality was slightly better (p</0.05) when night milk was consumed. A difference between the periods in morning or evening activity was lacking, which may indicate a lack of seasonal effect in activity. Randomization, blinding, and dropouts were not described.
        • Dawson et al. conducted a double-blind, placebo controlled, crossover trial of 12 elderly patients to assess the effect of exogenous melatonin on age-related sleep maintenance insomnia (447). Subjects were over 55 years of age (mean=65.67 years, SEM=1.68) and had all been suffering from sleep maintenance insomnia for at least six months. Subjects were treated with 0.5mg of transbuccal (delivered via a patch placed on the gums) melatonin or a placebo at 7 p.m. for two sessions of four consecutive nights, at least three days apart. Nightly urine samples were assayed for the melatonin metabolite 6-sulfatoxy-melatonin. Body temperature was measured continually from 9 p.m. to 7 a.m. Subjects self-selected lights-out times, and sleep was assessed using standard polysomnographic (PSG) measures. Outcome measures included polysomnographic recordings (EEG, EMG, EOG) and associated variables (total sleep time, sleep onset latency, REM onset latency, early morning awake, percentage time awake, sleep efficiency, stage changes in sleep period, time of sleep onset, and wake after sleep onset), and body temperature (rectal probe). Analysis revealed that, compared to the placebo, transbuccal melatonin administration significantly reduced core body temperature (p<0.05), but a positive, statistically significant effect on any PSG measure of sleep quality was lacking. This study found a lack of statistically significant effects of 0.5mg of transbuccal melatonin in treating age-related sleep maintenance insomnia. The authors speculated that both the dose size and timing may have influenced the negative result. Interactions, allergies and adverse effects, as well as dropouts, were not discussed.
        • Wurtman et al. reported the results of a small controlled study, in a letter to the editor, investigating melatonin's ability to improve sleep quality in aged insomniacs (614). Using a Latin square design, nine subjects (ages 51-78) were given melatonin (0.3mg) or placebo for three consecutive days, 30 minutes before bedtime. Core temperature and motor activity were recorded every minute for six days. Subjects also completed questionnaires about their previous night's sleep. Melatonin administration reduced the number of movements per night (480 ± 96 to 240 ± 55), latency to sleep onset (49 ± 8.9 to 27 ± 4.5), and awakenings per night (2.56 ± 0.32 to 0.88 ± 0.26). Subjects also reported that melatonin improved subjective sleep quality and did not increase morning sleepiness. Core body temperature was unaffected. Due to the nature of the article, it is difficult to evaluate the rigor of the study. Statistical findings were not reported.

        Jet lag

        • Summary: Several randomized, placebo controlled human trials suggest that melatonin taken by mouth, started on the day of travel (close to the target bedtime at the destination) and continued for several days, reduces the number of days required to establish a normal sleep pattern, diminishes the time it takes to fall asleep ("sleep latency"), improves alertness, and reduces daytime fatigue (371;429). Melatonin has been used in combination with zaleplon, zopiclone, and temazepam to compare the drugs' hypnotic effects and to observe drowsiness levels (654). Other research has attempted to determine optimal formulations in preparation for travel (1304). Further well-designed trials are necessary to confirm these findings, to determine optimal dosing, and to evaluate use in combination with prescription sleep aids.
        • Systematic reviews: Buscemi et al. conducted a systematic review of the efficacy and safety of exogenous melatonin in managing secondary sleep disorders and sleep disorders accompanying sleep restriction, such as jet lag and shift work disorder (388). Thirteen electronic databases were searched. Reference lists of relevant reviews, as well as a random sample of included studies, were reviewed. Abstracts of meetings of Associated Professional Sleep Society abstracts (1999-2003) were also searched. MEDLINE and Embase were searched again in 2004 to identify more recent published studies. Randomized controlled trials were assessed using the Jadad Scale and criteria by Schulz et al., and nonrandomized controlled trials were assessed by the Downs and Black checklist. One reviewer extracted data and another reviewer verified the data extracted. The inverse variance method was used to weigh studies, and the random effects model was used to analyze data. A range of doses were used in the various studies. The duration of administration varied from days to weeks. Dosage and duration of melatonin administration described a considerable amount of heterogeneity across studies. Efficacy: Nine trials (427 participants) (424;432;622;655;698;699;700;701;704) were included in the efficacy analysis for secondary sleep disorders. The median quality score was 4 out of 5 (Jadad score). Evidence that melatonin had an effect on sleep onset latency in people who had sleep disorders accompanying sleep restriction was lacking (-1.0 minutes (-2.3 to 0.3)). Sleep onset latency: Six randomized controlled trials with 97 participants lacked evidence in support of melatonin for sleep onset latency in people with secondary sleep disorders (weighted mean difference: -13.2 minutes (95% confidence interval, -27.3 to 0.9)). Safety: Ten studies (487 participants) were included in the safety analysis. The most commonly reported adverse events were headaches, dizziness, nausea, and drowsiness. The occurrence of these outcomes was similar for melatonin and placebo. A clear effect of melatonin specifically for jet lag is lacking from this study.
        • Herxheimer et al. conducted a systematic review to assess the effectiveness of oral melatonin taken in different dosage regimens for alleviating jet lag after air travel across several time zones (371;429). Cochrane Controlled Trials Register, MEDLINE, Embase, PsychLit, and Science Citation Index electronically, and the journals Aviation, Space and Environmental Medicine and Sleep were searched by hand. Citation lists of relevant studies for other relevant trials were also searched. Principal authors of relevant studies were asked about unpublished trials. Randomized trials on airline passengers, airline staff, or military personnel given oral melatonin, compared with placebo or other medication, were included. Outcome measures consisted of subjective ratings of jet lag or related components, such as subjective well-being, daytime tiredness, onset and quality of sleep, psychological functioning, duration of return to normal, or indicators of circadian rhythms. Ten trials met the inclusion criteria (370;416;419;424;425;426;621;624;1305;1306). All compared melatonin with placebo; in addition, one compared it with a hypnotic, zolpidem. Nine of the trials were of adequate quality to contribute to the assessment; one had a design fault and could not be used in the assessment. Reports of adverse events outside trials were found through MEDLINE, Reactions Weekly, and in the WHO UMC database. Nine of the 10 trials found that melatonin, taken close to the target bedtime at the destination (10 p.m. to midnight), decreased jet lag from flights crossing five or more time zones. Daily doses of melatonin between 0.5 and 5mg were similarly effective, except that people fell asleep faster and slept better after 5mg than 0.5mg. Additional effects of doses above 5mg were lacking. The relative ineffectiveness of 2mg of slow-release melatonin suggests that a short-lived higher peak concentration of melatonin works better. Based on the review, the number needed to treat (NNT) is 2. The benefit is likely to be greater the more time zones are crossed, and less for westward flights. The timing of the melatonin dose is important: if it is taken at the wrong time, or early in the day, it may cause sleepiness and delay adaptation to local time. The incidence of other side effects was reportedly low. Case reports suggest that people with epilepsy and patients taking warfarin may experience harm from melatonin.
        • Wagner et al. conducted a systematic review to assess the clinical efficacy and safety of current therapies available for insomnia treatment (391). The effects of barbiturates, antidepressants, benzodiazepines, cloral hydrate, zaleplon, zolpidem, zopiclone, and valerian were assessed in this review but are excluded from this summary focusing on melatonin. Of the included studies that assessed melatonin, those that focused on only the pharmacokinetics of melatonin, in vivo results, or case reports were excluded from this summary; results for 18 included trials (370;379;393;411;417;463;482;584;612;649;650;651;653;655;657;658;1295;1296) and three reviews (1098;1227;1297) that assessed the clinical effects of melatonin are included in this summary. The authors performed a MEDLINE search to find case reports, reviews, abstracts, and clinical studies from April 1992 to December 1997. The references of the pooled articles were manually reviewed to identify additional relevant studies. Articles selected for inclusion had to include a specific patient population presenting with a particular condition, had to provide specific details regarding the treatment, and had to specifically describe the outcome measures used to assess the treatment. All included studies were published in English. According to the reviewers, doses of melatonin used for treatment generally ranged from 1-10mg in healthy participants or 0.3-10mg in patients with insomnia. Also, participants in some included trials were administered only physiologic doses of melatonin (0.1-0.3mg). Specific details regarding the duration of treatments was lacking. Information on standardization was lacking from the review. According to the reviewers, adverse effects were lacking in most of the included studies. However, in some studies, adverse effects included stomach cramps, dizziness, fatigue, headache, irritability, drowsiness in the day, increased depression, and inhibited ovarian function. Information on toxic effects, dropouts, and interactions was lacking in the review. Primary outcome measures included safety and efficacy outcomes of the different insomnia treatments. For inducing sleep, some studies concluded that high doses of melatonin were effective. The reviewers stated that melatonin has been used to treat jet lag and delayed sleep-phase syndrome, as well as decrease sleep latency and increase total sleep time. The reviewers reported that melatonin has been shown to have mild soporific effects when given in the day or early evening and hypnotic effects when taken before sleeping. Low doses of melatonin administered close to bedtime showed mixed results. According to the reviewers, melatonin may be the most beneficial when given two hours before sleeping rather than immediately prior to sleeping. Also, the reviewers concluded that melatonin may be more effective in treating insomnia in older individuals, as well as multi-disabled or neurologically impaired children. The authors concluded that melatonin has mixed effects depending on the patient population, dosage, time of administration and experimental design. The review was limited by lack of information regarding treatment frequency and duration, as well as specific data regarding treatment outcomes in the included studies.
        • Evidence: Spitzer et al. conducted a randomized, double-blind, placebo controlled trial to validate a new rating scale for measuring severity of jet lag and to compare the efficacy of contrasting melatonin regimens to alleviate jet lag (619). The subjects were 257 Norwegian physicians who had visited New York for five days. Subjects were randomly assigned to placebo (N=60) or one of three alternative regimens of melatonin administration beginning on the day of travel and continuing daily for the five days after travel. On the sixth day after travel, the subjects made ratings but did not take any study capsules. The first experimental group (N=64) took 5mg of melatonin at bedtime. The second group (N=70) took 0.5mg at bedtime. This small dose produced total blood levels similar to those secreted endogenously over the entire night (although there was a distinctly higher initial peak concentration 30 minutes after administration, followed for at least nine hours by an exponential decline). The third group (N=63) also received 0.5mg of melatonin, but capsules were taken one hour earlier each day beginning in the early evening. The timing of the first dose (on the day of travel) was 11 hours after the subject's usual wake-up time. Jet lag ratings were made on the day of travel from New York back to Oslo (six hours eastward) and for the next six days in Norway. The main outcome measures were scale and item scores from a new, syndrome-specific instrument, the Columbia Jet Lag Scale, which identifies prominent daytime symptoms of jet lag distress. There was a marked increase in total jet lag score in all four treatment groups on the first day at home, followed by progressive improvement over the next five days. However, significant group differences or group-by-time interactions were lacking. In addition, a group effect for sleep onset, time of awakening, hours slept, or hours napping was lacking. Ratings on a summary jet lag item were highly correlated with total jet lag scores (from a low of r=0.54 on the day of travel to a high of r=0.80 on day 3). The internal consistency of the total jet lag score was high on each day of the study. Randomization and dropouts were not described.
        • Studies of lesser methodological strength: Paul et al. conducted a repeated-measures, controlled crossover trial of 30 subjects to explore the effectiveness of lower doses of zopiclone and melatonin in inducing early circadian sleep and how the treatments affect the performance of air crew (623). The subjects were given capsules of a placebo, 2mg of sustained-release melatonin (Circadin®, a pharmaceutical-grade time-released formulation of melatonin made by Neurim Pharmaceuticals, Tel Aviv, Israel), or zopiclone 5mg, all with an identical appearance. Objective measures included: sleep latency, wake after sleep time, total sleep time, time taken to get to sleep, number of awakenings, and time spent awake after having fallen sleep, using wrist actigraphy (Precision Control Design, Fort Walton Beach, FL). Fatigue, sleep quality and mood, drug effects/side effects, and amount of alcohol ingested were measured subjectively using a self-reported questionnaire (using seven-point Likert scales). According to the actigraphic data, subjects slept longer while using melatonin (p<0.02) and zopiclone (p<0.005) compared to placebo. Melatonin (p<0.01) and zopiclone (p<0.003) decreased sleep latency. Compared to placebo, subjects woke up less after having fallen asleep when taking melatonin (p<0.004) and zopiclone (p<0.01). Relative to placebo, subjects spent less time awake after sleep onset while taking melatonin (p<0.01) and zopiclone (p<0.05). A statistically significant difference between melatonin and zopiclone was lacking in total sleep time, time taken to get to sleep (latency), number of awakenings, or time spent awake after having fallen sleep. Subjects found it easier to fall asleep on melatonin (p<0.0001) and zopiclone (p<0.001) when compared to placebo. Relative to the placebo group, subjects woke up less and reported it easier to fall back asleep after awakening when using melatonin (p<0.005 and p<0.0003, respectively) and zopiclone (p<0.001 and p<0.0003, respectively). The quality of sleep was improved in the melatonin (p<0.0003) and zopiclone (p<0.0004) groups compared to placebo. All of these findings were statistically significant. However, a statistically significant difference was lacking between the zopiclone and melatonin groups in difficulty getting to sleep, reported number of awakenings, difficulty returning to sleep after awakening, or quality of sleep. Limitations of this study include a lack of a power calculation, randomization, adequate description of blinding process, and intention-to-treat analysis, as well as the suboptimal study design, length of study (just one dose, one day), and statistical interpretation. Generalizability is questionable, as the sample was from an air crew and predominantly male. Standardization, allergies, adverse effects, toxic effects, and interactions were not discussed.

        Sleep disorders (individuals with behavioral, developmental, or intellectual disorders)

        • Summary: Several randomized controlled studies have reported melatonin use in children with various neuropsychiatric disorders, including autism, psychiatric disorders, visual impairment, or epilepsy (354;355;357;638;734). Nonrandomized clinical trials have also been conducted (433;717;726;727;728;729;738;740;1295;1307;1308;1309;1310;1311) and, overall, the results agree with those observed in the RCTs. Well-designed controlled trials in select patient populations are needed before a conclusion can be made.
        • Meta-analyses and systematic reviews: Rossignol et al. conducted a systematic review and meta-analysis to investigate any information about melatonin and its impact in autism spectrum disorders (ASD) (355). This review included 35 studies which either examined the biochemistry and physiology of melatonin in ASD or examined melatonin as a treatment for ASD (349;353;587;643;717;718;725;728;731;735;1297;1300;1309;1312;1313;1314;1315;1316;1317;1318;1319;1320;1321;1322;1323;1324;1325;1326;1327;1328;1329;1330;1331;1332;1333), five of which were included in the meta-analysis (N=61) (587;717;725;731;735). Articles assessing the effects of melatonin on sleep parameters in participants with ASD published prior to October 2010 were pooled from ERIC, Google Scholar, EMBASE, CINAHL, PubMed and Scopus. For inclusion in the meta-analysis, studies had to be randomized double-blind, and placebo controlled. Any studies that included participants with developmental disorders other than ASD or did not report data on total sleep time, sleep onset latency, or number of nighttime awakenings were excluded from the meta-analysis. The daily dose of melatonin used in all studies ranged from 0.75-10mg daily, and in the meta-analysis doses given included 5mg, 2.5-7.5mg fast release, 1mg fast release and 4mg controlled release, and 2-10mg fast release which was titrated for effect. Trials included in the meta-analysis ranged from 25 days to eight months, and for all included articles durations ranged from 10 days to over four years. Information on standardization was lacking. According to the review, noted minor adverse effects throughout the studies included worse behavior, morning drowsiness, increased enuresis, headache, tiredness, dizziness, and diarrhea that were generally rare, although one outlier study reported side effects in 34% of the participants. This review asserts that overall, no serious adverse events were reported in any of the studies. Information on toxic effects, dropouts, and interactions was lacking. Primary outcome measures for all studies reviewing melatonin as a treatment included prevalence values and any changes in sleep parameters. In the meta-analysis, outcome measures included total sleep time, sleep onset latency, and number of nighttime awakenings. In studies that assessed the biochemistry or physiology of melatonin in ASD, nine of nine studies that reported melatonin and its metabolites concentration found derangements. Seven studies found levels to be lower than in healthy individuals, and two reported that daytime melatonin levels were higher than normal. Five studies examined genetic variations in melatonin homeostasis. Of these five studies, four examined the ASMT gene, involved in melatonin synthesis, and all four found decreased ASMT activity which significantly correlated with lower melatonin levels. Additionally, two of these studies evaluated melatonin receptor levels: one study found a significant decrease in ASD, and one had found no significant differences. Four of these studies also reported a correlation between lower melatonin levels and increases in ASD symptoms such as increased problems with verbal communication, play, and daytime sleepiness. Three surveys completed reported melatonin use among individuals with ASD to range from 2.9-10.8% with a mean prevalence of 7.2% (95% confidence interval [CI] 5.6-8.7%) and a prevalence of physicians recommending melatonin use in ASD to be 32.4% (95% CI 30.6-34.2%). In studies that evaluated the effect of melatonin on sleep in ASD, 12 of 18 studies reported an overall improvement rate of 84.2% (95% CI 81.4-88.9%). Six studies which examined daytime behavior changes with the administration of nighttime melatonin showed improvements in behavioral rigidity, ease of management by care-takers, social interactions, outbursts, irritability, alertness, and academic performance. For the five trials included in the meta-analysis, a significant improvement was reported using Hedge's G effect size (95% CI) in total sleep duration (44-73 minutes longer) and sleep onset latency (39-66 minutes shorter) by all studies with melatonin vs. placebo (p<0.001 for duration and latency) and baseline (p<0.001 for duration and latency). Changes in nighttime awakenings resulted a nonsignificant improvement over placebo (p<0.10) and baseline (p<0.10). The authors noted that the manner in which the data was collected seemed to highly impact the results. Using sleep diaries filled out by parents over actigraphy resulted in higher total sleep times, reduced sleep latency times, and decreased nighttime awakenings (0.6 vs. 13.3). This review found overall improvements in sleep with the use of melatonin in autism spectrum disorders in addition to correlations between melatonin levels and daytime behavior and between ASD and melatonin level derangements. Information on interactions, toxic effects, and dropouts is lacking in this review, and there is great variety in the design and quality of studies reviewed.
        • Braam et al. conducted a meta-analysis to investigate whether melatonin is effective for sleep problems in individuals with intellectual disabilities (666). The databases PubMed, MEDLINE, and Embase were searched for randomized controlled trials, written in English, that used melatonin as a treatment intervention for individuals with a diagnosis of intellectual disabilities. Search criteria under the heading of developmental problems included "intellectual disability," "developmental disability," "learning disorder," "mental retardation," and "neurodevelopmental disability." Furthermore, for inclusion, studies had to contain quantitative data on one of the following measures: sleep latency, total sleep time, or number of wakes per night. Nine studies were found to meet inclusion criteria (551;587;717;724;725;730;732;1292). Doses used ranged from 0.5mg to 9mg, with regimens lasting from 32 to 73 days (inclusive of washout periods). Of the nine included studies, only four specified the formulation of melatonin used. McArthur et al. and both studies by Braam et al. used immediate-release melatonin, whereas the Wasdell et al. study used a combination immediate-release (1mg)/extended-release (4mg) melatonin product. Analysis of the included trials revealed that melatonin treatment was found to decrease median time to sleep by 34 minutes (p<0.001), increase median length of sleep by 50 minutes (p<0.001), and decrease median nighttime wakes (p=0.024). The mean quality of included studies was 25.28 out of 32. The two reviewers had an intraclass correlation of 0.90. A substantial effect on publication bias was lacking. Overall, most adverse effects were small, with rates comparable to placebo. It should be noted that, of the nine included trials, five had less than 10 subjects. Also, none of the studies compared the same developmental disabilities.
        • Malow et al. conducted a systematic review to assess the effects of pharmacologic, behavioral, and complementary and alternative therapies on insomnia in children with autism spectrum disorders (ASDs) to ultimately develop practice guidelines for the management of insomnia in these patients (734). The effects of behavioral interventions, massage therapy, aromatherapy, risperidone, secretin, L-carnitine, niaprazine, mirtazapine, clonidine, iron, and multivitamins were assessed in this review, but are excluded from this summary focusing on melatonin; eight studies included in the review assessed the effects of melatonin in children with ASD suffering from insomnia (587;725;728;731;735;1312;1321;1334). Articles assessing the effects of melatonin on insomnia in children up to 18 years of age with ASD published in English between January 1995 and July 2010 were pooled from OVID, Embase, Cochrane Database of Systematic reviews databases, CINAHL, and the Database of Abstracts and Review Database of Abstracts of Reviews and Effects. Studies were excluded if they were reviews, nonintervention trials, case studies with less than ten individuals, or included children without ASD. In the included trials 1-10mg of melatonin nightly for two weeks to 24 months was studied. Information regarding the doses used in several of the cited trials was unclear. Information regarding standardization, allergies, side effects, toxic effects, dropouts, and interactions was lacking in this review. Outcomes were measured by self-reported sleep problems, sleep diaries, parental sleep charts and the Children's Sleep Habits Questionnaire (CSHQ). Three small trials showed melatonin to be safe and efficacious in treating insomnia for children with ASD. The four other trials showed that melatonin significantly improved sleep latency in children with ASD. Sleep duration increased significantly in three trials. The number of nightly awakenings significantly decreased in two trials and increased in one trial. The authors stated that further trials are needed to determine the safety and efficacy of melatonin. Ultimately, the authors recommend behavioral therapies as first line therapy in children with ASD suffering with insomnia, and as a second line sleep medications, including melatonin. Limitations of this review include a lack of statistical analyses of included trials, unknown significance of results, and a lack of information on standardization, allergies, side effects, dropouts, interactions, specific doses, and toxic effects. The review also excluded trials written in foreign languages.
        • Guénolé et al. conducted a systematic review of 12 studies and reports to assess the safety and efficacy of oral melatonin for treating sleep disorders in individuals who have been diagnosed with autism spectrum disorders (N=205) (354). Of the 12 studies, this review is focused on the results of the three retrospective studies (643;1321;1324), two open-label clinical trials (728;1335), and three placebo controlled trials (725;731;735). To locate these articles a Pubmed search was conducted. The search was performed all citations that were published from 1958 until November 2010. All languages were included. In addition, textbooks, the Cochrane library, and the references of pooled articles were manually reviewed to identify additional relevant studies. In the reviewed studies, children in the treatment groups were administered 0.75-10mg melatonin daily 30-60 minutes before bedtime for one week to two years. Adult participants were administered 3-9mg melatonin daily 45 minutes before bedtime for six months. Information regarding standardization was lacking. Adverse effects included daytime sleepiness or general tiredness (N=4), fogginess (N=1), dizziness (N=1), increased nocturnal enuresis (N=1), headache (N=1), and diarrhea (N=1). In children with ASD, one of the studies reported an increase in frequency of seizures and a higher rate of developing new onset seizure activity. Information regarding toxic effects was lacking. One patient dropped out due to adverse effects, which stopped immediately after discontinuing the melatonin. Information regarding interactions was lacking. Multiple different sleep ailments were assessed including long sleep latency (LSL), night wakenings (NW), settling difficulties (SD), early morning awakenings (EMA), and short sleep duration (SSD). Daytime sleepiness and the behavior of the children were assessed using the Karolinska Sleepiness Scale and the Child Behavior Check List in some of the trials. One retrospective study reported an improvement in total sleep duration in 56% of cases treated with melatonin. Another study reported a full resolution of sleep issues in 25% of cases and partial improvement in 60% of cases. In an open-label trial, a significant decrease in LSL and mean nocturnal activity was reported with melatonin treatment. However, after the discontinuation of melatonin, there was a significant decrease in total sleep duration and increased nocturnal activity. One study reported that melatonin was associated with improvements in LSL, NW, and sleep duration. In one of the placebo controlled studies, the investigators reported that melatonin was associated with a significant decrease in LSL, NW, and increased total sleep time when compared with placebo. Another study reported a small but significant increase in total sleep time (21 minutes) during melatonin treatment, as well as a significant decrease in LSL (28 minutes). One study reported that melatonin produced a significantly shorter sleep latency (52 minutes vs. 10 minutes) and a longer sleep duration (56 minutes vs. 8 minutes) vs. placebo. Limitations of this review included the fact that several studies failed to use standardized tools to assess sleep. In addition, many of the studies were small in size, and most symptoms were graded subjectively. Often the children's symptoms were assessed by their parents.
        • Hollway and Aman conducted a systematic review of 58 studies to assess the effects of treatments for sleep disturbances in individuals with developmental disabilities (638). The effects of antihistamines, alpha-adrenergic agonists, antidepressants, antipsychotics, ramelteon, benzodiazepines, and nonbenzodiazepines were assessed in this review but are excluded from this summary focusing on melatonin. The effects of melatonin were evaluated in 14 of the included studies (358;392;431;436;551;587;675;717;724;725;730;732;737;742). Studies published in English that assessed the effects of various agents on sleep disturbances were pooled from PsycInfo and Medline databases. References of the included articles were manually reviewed to identify additional relevant studies. All included studies involved children, teens, or adults and evaluated the effect of treatment on specific sleep disorders that influenced sleep initiation or maintenance. Studies that reported only on symptoms of other co-morbid illnesses were excluded, as were studies that evaluated obstructive sleep apnea (OSA) and narcolepsy. In studies that assessed the effects of melatonin in children or adolescents, participants were administered 0.1-9mg melatonin daily for 10 days to 10-63 days. In studies that assessed the effects of melatonin in adults, participants were administered 0.1-10mg melatonin daily for 2-9 weeks. Information regarding standardization was lacking. Minor adverse effects were reported in some of the included studies, but additional details regarding these events were lacking. The reviewers noted that, overall, the adverse effects were similar for participants administered melatonin vs. placebo. Information regarding toxic effects was lacking. In some of the studies, a significant number of dropouts occurred. Information regarding interactions was lacking. Outcome measures included changes in sleep disturbances based on parent sleep diaries, actigraphy, seizure diaries, the parent rated Quine Sleep Index, and changes in factors such as sleep onset latency (SOL), number of night awakened (NNW), total sleep time (TST), number of night awakenings (NAW), dim light melatonin onset (DLMO), and sleep efficiency (SE). In addition melatonin plasma levels and salivary melatonin samples were assessed. One study found significant improvement in parent-defined sleep variables, subjective SOL (ES 1.63), subjective NNW (ES 0.64), and subjective TST (ES 0.74) for participants treated with melatonin vs. placebo. Another study noted significant improvements in parent-defined sleep variables, subjective SOL (ES 0.56), subjective NAW (ES 0.28), subjective TST (ES 0.54), and DLMO (ES 0.96) for participants administered melatonin vs. placebo. Another study found significant improvements in SOL (ES 0.91), TST (ES 0.69), SE (ES 0.83), and in DLMO (ES 0.96) for participants treated with melatonin vs. placebo. Also, this study noted improvement in parent-defined difficulty falling asleep (ES 1.27). Significant improvements in actigraphic sleep variables, SOL (ES 0.46), TST (ES 0.25), parent diaries subjective SOL (ES 0.83), and subjective TST (ES 0.43) were noted in another trial. Also, one study showed significant improvements in parent-defined subjective SOL, NAW, and TST for participants treated with melatonin. Another study noted significant improvement in both actigraphic-defined and parent-defined sleep variables for participants treated with melatonin vs. placebo. This trial also assessed sleep hygiene and found that it decreased initial insomnia to <60 min in five children (ES 0.67). The study investigators also identified a significant reduction in insomnia (16 minutes) for participants treated with melatonin vs. placebo (ES 0.60). In another study, significant improvements in parent-defined sleep variable and subjective SOL (ES 0.25) were observed for participants treated with melatonin vs. placebo. Also, one study noted significant improvements for three doses of melatonin, with the 0.3mg dose causing the largest response effect vs. placebo. Significant increases in plasma circulating-melatonin levels were observed in participants who received the 0.3mg dose as well. Another study reported significant differences in parent-defined sleep variables, subjective SOL (ES 0.54), subjective SO (ES 1.1), and DLMO (ES 1.3) for participants administered melatonin vs. placebo. Also, one study reported significant improvement in actigraphic-defined SO (ES 1.9) for participants administered melatonin vs. placebo. Parent-defined subjective TST (ES 1.0) and SO (ES 0.87) were also improved. Another study reported significant improvement in parent-defined subjective TST for participants administered melatonin vs. placebo. One study reported that actigraphic sleep initiation variables showed significantly improved times in SOL for participants administered melatonin vs. placebo. In the final included study, significant improvements in subjective TST, NAW, and NNW were lacking for participants administered melatonin vs. placebo. This review lacked information regarding standardization, toxicity and adverse events.
        • Bendz et al. conducted a systematic review of the safety and efficacy of melatonin for the alleviation of insomnia in children with attention-deficit hyperactivity disorder (ADHD) (357). Criteria for inclusion were English-language articles and human studies conducted in children with ADHD and administering melatonin for the treatment of insomnia. Databases searched included MEDLINE (1948-August 2009), EMBASE (1950-August 2009), and Scopus (1960-August 2009). Search terms were "melatonin," "attention-deficit/hyperactivity disorder (ADHD)," "pediatric," "insomnia," "sleep disorder," and "sleep." A total of four studies were found (358;433;436;1336). Doses ranged from 3mg to 6mg of oral melatonin. Tjon Pian Gi et al. conducted a one year open-label study (N=120). Van der Heijden et al. conducted a four-week randomized controlled trial (N=105). Hoebert et al. conducted a three year follow-up to the van der Heijden trial (N=94). Weiss et al. conducted a six-month crossover study (N=28). Tjon Pian Gi et al. reported one case of restless sleep. Van der Heijden et al. reported mild adverse effects including headache, hyperactivity, dizziness, and abdominal pain. Also, a two year safety follow-up showed bedwetting, abnormal feces, and drowsiness. Hoebert et al. had 20% of patients (N=19) report adverse effects, all of which were not serious and did not involve new cases of epilepsy. After discontinuation, there were no reports of withdrawal symptoms or dependence. Weiss et al. reported mild-to-moderate adverse effects, except for one report of severe migraine. Tjon Pian Gi et al. examined sleep onset. Van der Heijden et al. measured sleep parameters including sleep onset, total sleep time, and sleep latency as well as specific core problems, including anger, sleep, and attention. Hoebert et al. measured relapse rates of sleep onset insomnia after discontinuation of melatonin using subjective questionnaires filled out by parents. Weiss et al. measured sleep latency and also implemented a sleep hygiene regimen using actigraphy and somnolog measures. Tjon Pian Gi et al. concluded that, at short-term evaluation, sleep onset was significantly improved (median increase: 135 minutes, N=24), which was then sustained at the long-term evaluation (N=13). Van der Heijden et al. reported significant improvements in sleep parameters (sleep onset, total sleep time, sleep latency, DLMO) and specific core problems (anger, sleep, attention). Hoebert et al. reported that 92% (N=60) of subjects experienced a delay in sleep onset, although, overall, 90% (N=85) of parents believed melatonin to be an effective treatment for SOI in their children with ADHD, 71% (N=67) thought daytime behavior improved, and 61% (N=57) thought mood improved. Weiss et al. determined that those who responded well to melatonin therapy lacked a significant improvement with continued treatment. In most studies, sleep onset was improved by 0.5-2 hours, sleep duration was improved by 0.33-1 hour, and sleep latency was improved by approximately 20 minutes. The review was well conducted; however, there is an overall lack of well-designed studies related to melatonin use. The available studies were limited by small sample sizes and variable outcome measures.
        • Rossignol et al. conducted a systematic review of the literature for novel and emerging treatments for autism spectrum disorders (ASD), among which melatonin was identified (353) in a number of studies (587;725;728;731;1297;1309;1312;1314;1321;1325;1328;1330;1332). Studies were gathered via a systematic search through PubMed and Google Scholar (1966 to April 2009), using the search terms "autism," "autistic," "pervasive," and "PDD" in all languages, in combination with a list of identified treatments. The list of treatments was developed by drawing on the Autism Research Institute, review articles, and the author's knowledge of the relevant literature. Regimens from the included studies ranged from 0.75mg to 10mg administered over two weeks to two months. Reported adverse effects included morning drowsiness, nighttime awakening, increased enuresis, and excitement before going to sleep. Experimental measures consisted of blood melatonin concentrations, nighttime urinary excretion of 6-sulphatoxymelatonin, total night sleep, length of sleep, sleep-wake rhythm, sleep pattern, sleep latency, nighttime awaking, sleep onset time, and seizure occurrences. The Jan et al. study reported that fast-release melatonin improved sleep in over 80% of the children, without any adverse effects or tolerance observed in the retrospective study that recruited children with chronic sleep disorders and a neurologic or developmental disability (including an unspecified number with autism). Hayashi et al. observed prolonged total night sleep and improved the sleep-wake rhythm in a case study. Ishiyaki et al. reported that melatonin administered at bedtime led to improvements in sleep in 84% of participants. Paavonen et al. stated that melatonin significantly improved sleep patterns and decreased sleep latency (p<0.002) in children with Asperger's syndrome. Giannotti et al. stated that melatonin significantly improved sleep (p<0.001) without any adverse effects in autistic children with a sleep disorder, although sleep problems returned when melatonin therapy was stopped (and improved again when melatonin was reintroduced). These improvements were still present at both 12- and 24-month follow-ups. In a retrospective study, Andersen et al. reported that 85% of participating children experienced improvements in sleep. In the same study, only one child had worsening of sleep with melatonin, and three other children had mild adverse effects. Wasdel et al. stated that melatonin therapy improved the length of sleep (p<0.01) and sleep latency (p<0.01) in children with a neurodevelopmental disorder and sleep problems. Garstang et al. reported significant improvement on sleep latency, by nearly 0.9 hours (p<0.05), decreased nighttime awakenings (p<0.05), and increased sleep duration, by 1.1 hours (p<0.05) in 11 ASD children. Finally, Wirojanan et al. stated that the use of melatonin in children with autism and/or fragile X syndrome increased mean sleep duration (p=0.02), sleep latency (p=0.0001), and sleep onset time (p=0.02). The authors concluded that promising treatments for ASD include melatonin, antioxidants, acetylcholinesterase inhibitors, naltrexone, and music therapy, which all received a grade of A according to their evidence-based guideline (based on the availability of two or more randomized controlled trials or one or more systematic reviews).
        • Phillips et al. conducted a systematic review to assess randomized control trial evidence for the use of melatonin in children with neurological and developmental problems (1337). Randomized clinical trials of children (up to 18 years of age) with any type of neurological disorder or neurodevelopmental disability and associated sleep disturbance, where oral melatonin was used and compared to placebo, were identified. Sustained-release formulations were excluded. Three studies (N=35 children) met inclusion criteria (551;717;1292). Two studies reported time to sleep onset and showed a decrease (p<0.05) in this specific outcome where melatonin was compared with a placebo (717;1292). Significant effects of melatonin compared with a placebo on the other outcome measures of total sleep time, nighttime awakenings, and parental opinions were lacking.
        • Lancioni et al. conducted a systematic review to assess the effects of various treatments or strategies used to reduce sleeping problems in subjects suffering from severe or profound mental retardation or other handicaps (398). The effects of sleep scheduling, maintaining a bedtime routine, extinction, bedtime fading, and chronotherapy were assessed in this review but are excluded from this summary focusing on melatonin; eight included studies assessed the effects of melatonin (551;717;1295;1297;1338;1339;1340;1341). Electronic searches were undertaken of PSYCHLIT, ERIC, and MEDLINE EXPRESS databases for articles assessing the effects of melatonin in subjects with disturbed sleep in association with multiple handicaps or severe or profound mental retardation. All included studies were published between 1984 and 1998. Studies were excluded if treatment results were lacking. Doses of melatonin reported in the review ranged from 0.5mg-5mg orally. Treatments were administered to children and/or adults each evening for 14 days to one year. Information regarding standardization was lacking. In one study, a child administered melatonin showed increased insomnia and severe reflux esophagitis. Information regarding toxic effects, dropouts, or interactions was lacking. Outcome measures included the effect of treatment on sleep. Effects were graded as positive, mixed, or negative. In one included case study, treatment with melatonin had a beneficial effect on sleep. In another study, nine children with mental retardation and visual impairment showed significant beneficial effects with melatonin treatment. In a third study, eight children treated with melatonin all showed improvement in sleep condition, although this beneficial effect was only temporary for one of the participants. Of the remaining five studies, a positive overall effect was reported for one study, a negative overall effect was reported for another study, and mixed results were reported for three studies. The authors conclude that, although there are indications of efficacy of melatonin for treating sleep disorders in the physically and mentally disabled, the results of these studies could be confounded by differences in severity of the underlying condition, different doses used, different periods of treatment, and whether other modalities of treatment were used at the same time. Also, the methodology used in some studies is of questionable quality.
        • Wagner et al. conducted a systematic review to assess the clinical efficacy and safety of current therapies available for insomnia treatment (391). The effects of barbiturates, antidepressants, benzodiazepines, cloral hydrate, zaleplon, zolpidem, zopiclone, and valerian were assessed in this review but are excluded from this summary focusing on melatonin. Of the included studies that assessed melatonin, those that focused on only the pharmacokinetics of melatonin, in vivo results, or case reports were excluded from this summary; results for 18 included trials (370;379;393;411;417;463;482;584;612;649;650;651;653;655;657;658;1295;1296) and three reviews (1098;1227;1297) that assessed the clinical effects of melatonin are included in this summary. The authors performed a MEDLINE search to find case reports, reviews, abstracts, and clinical studies from April 1992 to December 1997. The references of the pooled articles were manually reviewed to identify additional relevant studies. Articles selected for inclusion had to include a specific patient population presenting with a particular condition, had to provide specific details regarding the treatment, and had to specifically describe the outcome measures used to assess the treatment. All included studies were published in English. According to the reviewers, doses of melatonin used for treatment generally ranged from 1-10mg in healthy participants or 0.3-10mg in patients with insomnia. Also, participants in some included trials were administered only physiologic doses of melatonin (0.1-0.3mg). Specific details regarding the duration of treatments was lacking. Information on standardization was lacking from the review. According to the reviewers, adverse effects were lacking in most of the included studies. However, in some studies, adverse effects included stomach cramps, dizziness, fatigue, headache, irritability, drowsiness in the day, increased depression, and inhibited ovarian function. Information on toxic effects, dropouts, and interactions was lacking in the review. Primary outcome measures included safety and efficacy outcomes of the different insomnia treatments. For inducing sleep, some studies concluded that high doses of melatonin were effective. The reviewers stated that melatonin has been used to treat jet lag and delayed sleep-phase syndrome, as well as decrease sleep latency and increase total sleep time. The reviewers reported that melatonin has been shown to have mild soporific effects when given in the day or early evening and hypnotic effects when taken before sleeping. Low doses of melatonin administered close to bedtime showed mixed results. According to the reviewers, melatonin may be the most beneficial when given two hours before sleeping rather than immediately prior to sleeping. Also, the reviewers concluded that melatonin may be more effective in treating insomnia in older individuals, as well as multi-disabled or neurologically impaired children. The authors concluded that melatonin has mixed effects depending on the patient population, dosage, time of administration and experimental design. The review was limited by lack of information regarding treatment frequency and duration, as well as specific data regarding treatment outcomes in the included studies.
        • Evidence: Gringras et al. conducted a randomized, double-blind, placebo controlled, multicenter trial to assess the effects of melatonin on severe sleep disorders in children with neurodevelopmental problems (N=146) (445). Children aged 3-15 years with neurodevelopmental disorders (<1.5SD below the mean based on the adaptive behavior assessment system) were included in the study if they had experienced sleep problems for at least five months before the study and standardized advice to parents on sleep behavior over the past 4-6 weeks had lacked a positive outcome. Sleep problems were defined as being unable to fall asleep within one hour of turning off the lights for three out of five nights or getting less than six hours of sleep for three out of five nights. Potential subjects were excluded if they were taking sedative medications or were given melatonin within the last five months. All children were given rapid acting oral melatonin 0.5mg or placebo 45 minutes prior to bedtime. The dose could be increased to 2mg, 6mg, or 12mg melatonin or placebo if deemed necessary. The treatments were given in identical capsules for 12 weeks. Information on standardization was lacking. Adverse effects were similar in both treatment and placebo groups and were described as mild. In the treatment and placebo groups, types and numbers of adverse events included coughing (N=22/28 in treatment/placebo groups), mood swings (N=16/17), vomiting (N=15/18), increased excitability (N=13/16), rash (N=11/8), somnolence (N=9/10), hypothermia (N=6/4), increased activity (N=6/9), nausea (N=3/11), dizziness (N=1/5), breathlessness (N=1/1), hung-over feeling (N=1/0), and seizures (N=0/1). Information on toxic effects was lacking. Of the original 70 participants in the melatonin group and 76 participants in the placebo group, 51 and 59 were included in analysis of the primary outcome, respectively. In the melatonin group, 19 participants were excluded from analysis of the primary outcome due to discontinuation of treatment (N=4), having less than five nights' sleep recorded at baseline or 12 weeks (N=12), having a lost or forgotten sleep diary (N=1), or falling asleep after the parents (N=2). In the placebo group, 17 participants were excluded from final analysis of the primary outcome due to discontinuation of treatment (N=6), having less than five nights' sleep recorded at baseline or 12 weeks (N=10), or having a lost or forgotten sleep diary (N=1). Information on interactions was lacking. Assessment of total sleep time using diaries was the primary outcome. Other assessments were measurement of sleep duration by actigraphy, time to attain sleep, time in bed asleep, child behavior, and functional status of the family. As assessed by sleep diaries, total sleep duration with melatonin was increased by 22.4 minutes compared with placebo (95% CI 0.5 to 44.3, p<0.05). However, statistically significant between-group differences in total sleep were lacking based on actigraphy. Time taken to sleep was significantly reduced compared with placebo based on sleep diary (-37.5 minutes, 95% CI -55.3 to -19.7 p<0.001) and actigraphy (-45.3 minutes, 95% CI -68.8 to -21.9, p<0.001). Children woke earlier on melatonin by 29.9 minutes compared with placebo (95% CI 13.6 to 46.3 minutes p<0.001). Slight improvement was seen with melatonin in family functioning and child behavior compared to baseline, but overall significant between-group differences were lacking. According to the authors, melatonin had only a limited effect on the total duration of sleep. While children fell asleep significantly faster, they woke earlier. Overall, this trial was well-designed.
        • Wasdell et al. conducted a randomized, double-blind, placebo controlled, crossover trial to determine the efficacy of controlled-release (CR) melatonin in the treatment of delayed sleep phase syndrome and impaired sleep maintenance of 51 children with neurodevelopmental disabilities, including autistic spectrum disorders (587). Children (age range: 2-18 years) who did not respond to sleep hygiene intervention were enrolled. Participants received a pharmaceutical grade controlled release (CR) melatonin 5mg followed by a three-month open-label study, during which the dose was gradually increased until the therapy showed optimal beneficial effects. Endpoints included sleep characteristics measured by a caregiver (somnologs and wrist actigraphs), as well as clinician ratings of severity of sleep disorder and improvement from baseline, and caregiver ratings of global functioning and family stress. Fifty patients completed the crossover trial (one withdrew due to illness) and 47 completed the open-label phase. Recordings of total nighttime sleep and sleep latency showed an improvement (p<0.001) with melatonin. Improvement was observed in both clinician and parent ratings (p<0.05). Adverse effects related to melatonin were lacking. This study was well designed; however, some patients had previous experience with melatonin therapy and were only asked to withdraw two weeks prior to the study.
        • Weiss et al. conducted a randomized, double-blind, placebo controlled, crossover trial to evaluate the efficacy of sleep hygiene and melatonin treatment for initial insomnia in children with attention-deficit hyperactivity disorder (ADHD) (436). Twenty-seven stimulant-treated children (6-14 years of age) with ADHD and initial insomnia (>60 minutes) received sleep hygiene intervention and were administered melatonin 5mg and placebo. Each treatment period was 10 days, followed by a five-day washout period with placebo. Sleep hygiene reduced initial insomnia to less than 60 minutes in five cases, with an overall effect size in the group of 0.67. A reduction in initial insomnia of 16 minutes with melatonin relative to placebo was observed (p<0.01), with an effect size of 0.6. The effect size of the combined sleep hygiene and melatonin intervention from baseline to 90 days' post-trial was 1.7, with a mean decrease in initial insomnia of 60 minutes. Improved sleep lacked a demonstrable effect on ADHD symptoms. This trial was well designed.
        • Wright et al. conducted a randomized controlled trial to assess the effect of melatonin on sleep in children with autism (N=20) (735). For inclusion, participants had to have a history of sleep disturbances (i.e. difficulty with sleep latency, frequent nighttime awakenings, or decreased sleep duration) that failed to improve following lifestyle modifications. Children were excluded if they suffered from other sleep disturbances (i.e. night terrors, sleepwalking, and hallucinations), a history of previous melatonin therapy, treatment with psychotropic agents, or had a history of other developmental or neurological disorders. Participants were randomized to capsules of placebo or 2mg of oral melatonin 30 to 40 minutes prior to bedtime for three months. During the study, therapy was allowed to be titrated to a maximum dose of 10mg. Standard release melatonin and placebo were provided by DHP pharma. Information regarding standardization was lacking. Although statistical significance was lacking between melatonin and placebo, side effects reported during the trial were as follows: daytime drowsiness, sore throat, dizziness, anxiety, tearfulness, headaches, vomiting, gastric upset, reduced appetite, low mood, irritability, rashes, reduced alertness, diarrhea, constipation, earaches, asthma, fit or seizure, confusion, mild tremor and bedwetting. Information regarding allergies was lacking. This study resulted in a lack of serious side effects. The authors reported five dropouts due to the potential for poor adherence, inability to complete a sleep diary, diagnosis of flu, perceived significant benefit while on therapy, and perceived ineffectiveness of therapy. Information regarding interactions was lacking. Primary outcome measures included sleep latency, duration of sleep, and nighttime awakenings. Secondary outcome measures included the Sleep Difficulties Questionnaire (SDQ), Developmental Behavior Checklist (DBC), General Health Questionnaire (GHQ) and Side Effects Questionnaire (SEQ). When compared to placebo, participants on melatonin experienced a statistically significant average 46.7±55 minute improvement in sleep latency (p=0.004), an average 52.3±55.1 minute increase in sleep duration (p=0.002), an average 6±10.8 point improvement in DBC total score (p=0.05), and an average -5.67±8.30 point difference in SDQ dysomnia (p=0.041). Statistical significance between melatonin and placebo was lacking for number of daily naps, changes in nighttime routine, and number of night awakenings. In the SEQ, GHQ, and other parameters of the DBC and SDQ, significant differences were lacking between groups. Lastly, all results lacked a statistically significant crossover effect. Limitations of this study include lack of information regarding allergies, interactions, and randomization methods. Strengths of this study include proper blinding methods and a description of withdrawals.
        • Braam et al. conducted a randomized, double-blind, placebo controlled trial to evaluate the effectiveness of melatonin in 58 patients with intellectual disabilities (730). Patients were included with sleep disturbances not due to somatic and psychiatric causes. Standard behavioral treatment and sleep hygiene measures had been previously unsuccessful. Individuals with sleep latency of more than 30 minutes, two or more periods of night awakenings lasting more than 45 minutes each night, or five or more night awakenings lasting more than 15 minutes for at least five nights each week were included. Sleep disturbance was to have been present for at least one year. Exclusion criteria included prior use of melatonin, restless leg syndrome, sleep apnea syndrome, liver disease, renal failure, chronic pain, age less than 24 months, and Smith-Magenis syndrome. The trial consisted of a baseline and qualification period followed by a four-week treatment period. During the treatment period, subjects received melatonin 5mg or placebo at bedtime. Individuals younger than six years received 2.5mg of melatonin. Compared with placebo treatment, mean sleep latency decreased by 29 minutes (p<0.01), mean sleep onset increased by 34 minutes (p<0.01), and mean total sleep time increased by 48 minutes (p<0.05) during melatonin treatment. The mean number of night awakenings per night decreased (p<0.05). Changes in lights-out time, number of nights with night wakings per week, and sleep offset time compared with baseline were lacking. There was a correlation between baseline dim-light melatonin onset and the change in dim-light melatonin onset between baseline and week 4 in the melatonin group (p<0.001). Adverse effects were reported as few and minor; however, specific information was lacking. This trial was well designed.
        • Garstang et al. conducted a randomized, double-blind, placebo controlled, crossover trial to evaluate the effects of melatonin in 11 autistic children with sleep disorders (725). Children who had used melatonin previously were excluded, and all children were aged 4-16. The children received melatonin 5mg or placebo for four weeks. There was a washout period of one week between each of the treatment periods. Seven children completed the trial. Four children did not complete the trial; reasons were moving, a child protection enquiry, and difficulty with the placebo capsules, necessitating a recall. Melatonin significantly reduced sleep latency and the number of night wakings, and increased the total sleep time. Sleep latency was 2.6 hours (95% CI, 2.28-2.93) at baseline, 1.91 hours (95% CI, 1.78-2.03) with placebo, and 1.06 hours (95% CI, 0.98-1.13) with melatonin. Wakings per night were 0.35 (95% CI, 0.18-0.53) at baseline, 0.26 (95% CI, 0.20-0.34) with placebo, and 0.08 (95% CI, 0.04-0.12) with melatonin. Total sleep duration was 8.05 hours (95% CI, 7.65-8.44) at baseline, 8.75 hours (95% CI, 8.56-8.98) with placebo, and 9.84 hours (95% CI, 9.68-9.99) with melatonin. A discussion of adverse effects was lacking. A description of blinding was lacking, but otherwise the study was well designed.
        • Wirojanan et al. conducted a four-week, randomized, double-blind, placebo controlled, crossover trial to determine the efficacy of melatonin on sleep problems in children with autistic spectrum disorder (ASD) and fragile X syndrome (FXS) (731). The experiment was conducted on eighteen subjects (16 boys, two girls) between two and 15.3 years of age who were diagnosed with ASD (either autism or pervasive developmental disorder [PDD-NOS]), fragile X syndrome (FXS), or a combination of both. Participants were given 3mg of melatonin or placebo to take orally at bedtime. The placebo was provided by the Twinlab Corporation to look similar to the melatonin capsule. The study medication was given 30 minutes prior to bedtime over the two-week treatment arm followed by a one-week washout period. Lastly, the participants crossed over to the alternate treatment for an additional two weeks. There were six dropouts from a total of 18 enrolled. Three participants were excluded, even though they finished the study, because their caregivers did not complete the sleep diary or because the Actiwatch data were unreadable. Two participants were excluded since they failed to wear the Actiwatch during one of the treatment arms. Also, a family was excluded due to study protocol violation and an incomplete sleep diary. Objective sleep variables, including sleep onset time (actual time participant fell asleep), total length of sleep duration, sleep onset latency time (time from bedtime to sleep onset time), as well as the number of times the participant woke, were obtained from actigraphy and the subjective daily sleep diary that was completed by the parents or caregivers. A small device worn around the wrist or ankle, the Actiwatch, measured gross motor activity in one-minute intervals. Total night sleep duration tended to be longer (mean increase: 21 minutes) in the melatonin arm, but a statistical significance was lacking (p=0.057). Sleep latency time tended to be shorter (mean decrease: 28 minutes, 5 seconds) in the melatonin arm, but significance was lacking (p=0.10). Sleep onset time was significantly earlier (mean decrease of 42 minutes) for participants (p=0.0017) during melatonin treatment compared with placebo when using a parametric test. A significant difference in the number of nighttime awakenings was lacking in the melatonin group (p=0.73). The study was well designed, albeit with a small participant size.
        • van der Heijden et al. conducted a randomized, double-blind, placebo controlled trial to assess the effects of melatonin in children with ADHD and chronic sleep onset insomnia (SOI) (N=107) (358). Children aged 6-12 years who were diagnosed with ADHD and SOI were included in the study. The study excluded subjects with IQ less than 80, pervasive developmental disorder, chronic pain, kidney or liver dysfunction, epilepsy, previous melatonin consumption, and use of stimulants, neuroleptics, benzodiazepines, clonidine, antidepressants, hypnotics, or beta-blockers within four weeks of the start of treatment. Following a one-week run-in period, participants were randomized to receive tablets containing 3mg or 6mg melatonin according to body weight (3mg if body weight <40kg; 6mg if body weight >40kg) or identical looking placebo. The doses were given daily for four weeks. Information on standardization was lacking. According to the investigators, a significant between-group difference regarding adverse effects was lacking. However, participants in the treatment group experienced headache, hyperactivity, dizziness, abdominal pain, nose bleeding, itching or painful lumps on the skin, diarrhea, decreased mood, and insomnia. These effects were lacking for the placebo group. Information on toxic effects was lacking. One patient in placebo group and one in treatment group were excluded from the study after randomization because they started another treatment. Information on interaction was lacking. Outcome measures included changes in sleep parameters such as sleep onset, total time asleep, difficulty falling asleep, and dim light melatonin onset (DMLO). Other outcome measures included changed in behavioral problems, cognitive performance, and quality of life. Compared to the placebo group, participants treated with melatonin showed significantly reduced DLMO (+13 ± 59.0 vs. -44.4 ± 67.9 minutes, respectively, p<0.0001), sleep onset (+10 ± 37 vs. -27 ± 48 minutes, respectively, p<0.0001), sleep latency (+3.0 ± 31.7 vs. -12.3 ± 33.0 minutes, respectively, p=0.001), and difficulty falling asleep (-0.1 ± 0.8 vs. -1.2 ± 1.3 points, p<0.0001), as well as significantly increased total time asleep (-13.6 ± 50.6 vs. +19.8 ± 61.9 minutes, respectively, p=0.01) and sleep efficiency (-2.1 ± 7.1 vs. +2.6 ± 8.9%, respectively, p=0.011). Also, compared to the placebo group, participants treated with melatonin showed a statistically significant improvement in core problems based on parent assessment (+0.2 ± 0.8 vs. +0.7 ± 0.9, respectively; p=0.002). However, this between-group difference was lacking when core symptoms related to sleep were removed. There was a lack of significant between-group differences in other outcomes related to behavior, cognition, and quality of life. Limitations of this study include lack of information regarding the method of randomization and blinding, as well as lack of information regarding standardization, toxic effects, or interactions. In addition, it was unclear if a power calculation had been conducted to determine the sample size needed to detect between-group differences.
        • Coppola et al. conducted a randomized placebo controlled trial to examine the effects of melatonin on sleep disturbance in 32 patients with mental retardation with or without epileptic seizures (724). Patients more than 12 months old with mental retardation with or without epileptic seizures, with a diagnosis of a circadian rhythm sleep disorder (including delayed onset of sleep, multiple night awakenings, and short duration of night sleep through a baseline period of six months), and exclusion of medical issues such as gastroesophageal reflux, pain, or epileptic seizures mimicking sleep disorders, with persisting sleep disturbances despite maintaining appropriate sleep hygiene, were included if there was no progressive neurological or systemic diseases. Patients were 3.6-26 years old (mean: 10.5 years). Patients received oral synthetic fast-release melatonin (3mg at bedtime) or placebo. However, in case of inefficacy, the dose could be titrated up to 9mg the following two weeks at increments of 3mg per week, unless the patient was unable to tolerate it. Twenty-five patients completed the study; withdrawals were not due to adverse effects. Endpoints included data available from sleep logs, such as sleep latency and total time of sleep, and night awakenings. The analysis of all the sleep logs disclosed a significant treatment effect of melatonin on sleep latency (p=0.019). Melatonin was well tolerated in all patients, and reported side effects were lacking. The authors concluded that melatonin may be effective in young patients with mental disabilities and epileptic seizures in improving the wake-sleep disorders, such as time to fall asleep. Randomization and blinding were not adequately described.
        • Braam et al. combined data from a pair of randomized controlled trials (730;732) to examine the efficacy of melatonin in improving sleep and decreasing behavioral problems in 66 persons with intellectual disabilities (ID) and chronic insomnia (733). Patients had chronic insomnia with comorbid ID who had been referred to the investigators' sleep center by local general practitioners. Insomnia was defined as difficulty falling asleep (30 minutes or greater), waking up twice or more during the night for more than 45 minutes, or waking up five times for no less than 15 minutes each time, with these difficulties persisting for one year or longer. Patients with somatic or psychiatric causes of insomnia, previous use of melatonin, restless legs syndrome, liver or renal disease, chronic pain, or sleep apnea, and those younger than two years of age were excluded. Subjects received melatonin (5mg for those six years of age and older and 2.5mg for those younger) or placebo (N=27 and 22, respectively) for four weeks (younger patients at 6 p.m., older patients at 7 p.m.). Subjective outcome measures that were used to quantify challenging behavior included the Storend Gedragsschaal voor Zwakzinnigen (SGZ; Maladaptive Behavior Scale for the Mentally Retarded). Subjective measures which were used to quantify sleep problems included times lights went out, time to actual sleep, total sleep time and number, and duration of nighttime wakes as logged by a sleep diary. Objective outcome measures included dim-light melatonin onset (DLMO), which was set when salivary melatonin reached 4pg/mL. Out of the initial subject pool, one child was withdrawn due to unwillingness to submit to saliva sampling, and nine children in the placebo group and seven in the melatonin group were excluded due to incomplete sleep logs or SGZ scales. Melatonin therapy was found to significantly reduce mean SGZ-T (p<0.003) and M (p<0.005) scores, though V subscale scores were not (p=0.0617). Though clinical improvement was noted in individual items of the SGZ-A subscale, the mean score was not significantly changed (p=0.076). Melatonin treatment was also found to significantly improve a number of sleep parameters including sleep latency, number of wakes, duration of wakes, and total sleep time (p<0.001, 0.002, 0.034, and 0.043, respectively), as well as significantly advance DLMO (p<0.001). A significant change in time of lights-out, however, was lacking (p=0.697). Limitations of this study include reliance upon subjective measures of sleep and a small subject population. It should also be noted that although specific mention of an identical placebo was made, it is unclear if the study was double-blinded.
        • Niederhofer et al. conducted a randomized, double-blind, placebo controlled trial to determine whether depressed nocturnal melatonin levels and insomnia are associated, and whether various melatonin doses restore sleep in 20 mentally retarded subjects (737). Patients with insomnia of at least one year (latency of sleep of 30 minutes or more, two or more nighttime awakenings, or total period of nighttime sleep less than six hours) were included if self-reports were confirmed by continuous one-week recordings of the motor activity (<85% sleep efficiency) and reasons for insomnia (disease, medications) were excluded. The patients received a placebo on alternate (odd-numbered) weeks throughout the study, starting with the first "run-in" week, thus providing washout periods between and after active treatments. During the second week and thereafter on each even-numbered treatment week, subjects received melatonin (0.1mg or 0.3mg) or the placebo daily (a half-hour before bedtime) for seven days. Sleep was improved by all three melatonin doses, with the 0.3mg dose causing the greatest effect (p<0.0001). The authors reported that melatonin lacked behavioral, dose-related effects on total sleep time and number of awakenings. Randomization, blinding, and withdrawals were not adequately described.
        • Jan et al. conducted a randomized, double-blind crossover trial to compare the effectiveness of fast-release and controlled-release melatonin and to establish the most effective dose in children with chronic and severe sleep disorders (452). In the initial study, 16 multidisabled children (4-21 years of age) with severe sleep-wake cycle disorders were included. These patients were already being treated with fast-release melatonin for more than three months. Then the 16 subjects were randomized to receive controlled- or fast-release melatonin, each for 11 days, and then the drugs were crossed over. A washout period was lacking. The fast-release dose was the same as before the study. The controlled-release formulation was 50% of the fast-release dose. Improvements in sleep patterns were noted in 11 of the children, but the controlled-release formulation had lacked a clear advantage over the fast-release melatonin in five children. Following this initial study, 42 other patients on fast-release melatonin were permitted to change to controlled-release preparation in an open-study design. The average final controlled-release melatonin dose in the 42 patients was 5.7mg (2-12mg). Fast-release melatonin was found to be most effective for delayed sleep onset; controlled-release formulations were more beneficial for sleep maintenance. Randomization was not described, and a placebo was lacking in this study.
        • Studies of lesser methodological quality (not included in the Evidence Table): Hoebert et al. conducted a long-term follow-up of melatonin therapy in pediatric patients with ADHD and chronic sleep onset insomnia (1336). Subjects were 101 patients (aged 6-12 years with confirmed diagnosis of ADHD, total IQ higher than 80, and chronic sleep onset insomnia) who had previously taken part in a randomized clinical trial of melatonin treatment. The mean time to follow-up was 3.7 years. Assessment was effected via questionnaires consisting of 19 questions addressing use, dose, frequency, discontinuation, and adverse events. Of responding subjects (94/101), long-term melatonin therapy was judged to be effective against sleep onset problems in 88% and to improve behavior and mood in 71 and 61% of respondents, respectively. Sixty five percent still used melatonin daily and 12% occasionally. Serious adverse events or melatonin-related comorbidities were lacking. The authors concluded that melatonin was an effective long-term treatment for chronic sleep onset insomnia in children with ADHD.
        • Galli-Carminati et al. conducted a retrospective study of six patients to determine the use of melatonin in adults with autism and sleep disorders (643). Included patients were all followed in a psychiatric unit for patients with intellectual deficiencies. Patients presented with autism associated with mental retardation diagnosed according to the International Classification of Diseases and exhibiting accompanying behavioral problems, including severe psychomotor agitation, autoaggressive behavior (including minor and severe automutilation), and violent behavior directed towards objects and others. Melatonin was initiated at a daily dose of 3mg before bedtime. After four weeks of treatment, if the dose was ineffective, it was increased by increments of 3mg every two weeks, with a maximum of 9mg. Both before and after the six-month treatment period, patients' global severity and global improvement of sleep disorders were evaluated. Improvement in sleep disorders were evaluated using the Clinical Global Impression scale severity and improvement scales. All of the patients' sleep disorders were significantly improved, with a median score of 5.5 before treatment and 1.0 after treatment (p=0.031). Limitations of this study include a small sample size, and lack of a control group.
        • Various other studies of lesser methodological strength add to the information above and generally show positive results (736;1334;1342;1343).

        Sleep enhancement in healthy people

        • Summary: Patients with insomnia appear to have decreased melatonin secretion, and treatment with exogenous melatonin may offer some benefit (322). Most human trials have been small and brief in duration (often single-dose studies). However, the weight of scientific evidence does suggest that melatonin decreases the time it takes to fall asleep ("sleep latency"), increases the feeling of "sleepiness," and may increase the duration of sleep. Melatonin may also aid in inducing daytime sleep (1344). Further well-designed study is required before firm conclusions can be drawn.
        • Meta-analysis: Brzezinski et al. conducted a meta-analysis of randomized, double-blinded trials of varying design to assess the effects of exogenous melatonin on sleep (659). A total of 17 studies were included (379;446;560;584;612;614;649;650;651;655;657;658;663;1285;1293;1345;1346). Subjects included varied widely, representing insomniacs (naturally occurring and artificially induced), schizophrenics, Alzheimer's disease patients, and healthy volunteers. Age and sex distributions also varied from study to study. Authors reported that dosages ranged from 0.3 to 80mg of melatonin delivered orally. One study administered melatonin intravenously (50mg). The therapy schedule also varied from study to study, ranging from multiple times over one experimental session to one treatment daily for up to two months. Included studies employed only adult subjects. Discussion of allergies, adverse and toxic effects, interactions, and dropouts was lacking. Outcome measures included sleep onset latency, total sleep duration, and sleep efficiency. Analysis revealed that melatonin treatment significantly reduced sleep onset latency by 4.0 minutes (95% CI, 2.5-5.4), increased sleep efficiency by 2.2% (95% CI, 0.2-4.2), and increased total sleep duration by 12.8 minutes (95% CI, 2.9-22.8). Since 15 of the 17 studies included enrolled healthy subjects or people with no relevant medical condition other than insomnia, the analysis was also done including only these 15 studies. The sleep onset results were changed to 3.9 minutes (95% CI, 2.5-5.4), sleep efficiency increased to 3.1% (95% CI, 0.7-5.5), and sleep duration increased by 13.7 minutes (95% CI, 3.1-24.3). This study suggests a small but statistically significant increase in sleep onset latency, sleep efficiency, and total sleep duration in response to melatonin treatment as measured via meta-analysis of a number of studies of varying design; however, the disparate nature of the included research makes pooling of data and objective comparison of results somewhat difficult.
        • Systematic review: Buscemi et al. conducted a systematic review of 49 studies (379;392;393;415;424;432;446;447;448;470;559;560;584;590;610;612;622;640;648;649;650;651;653;655;657;658;672;675;698;699;700;701;704;717;885;976;1282;1283;1284;1285;1286;1287;1288;1289;1290;1291;1292;1293;1294) to assess the effects of melatonin on sleep disorders (764). Relevant studies published primarily in English were pooled from Medline (through June 2003), Cochrane Central Register of Controlled Trials (through the third quarter, 2003), Science Citation Index (through July 4, 2003), Biological Abstracts (through July 4, 2003), International Pharmaceutical Abstracts (through August 2003), NLM Gateway (through August 13, 2003), OCLC Papers First and Proceedings First (through July 11, 2003), and Toxline (through July 4, 2003). Studies published in non-English languages were included if the studies published in English were biased. Clinical trials, quasi-randomized controlled trials, prospective cohorts, case series, and reviews were included in this review. However, information from prospective cohorts and case series has been excluded from this summary. In addition, information related to the pharmacology or mechanism of action of melatonin has been excluded from this summary focusing on the clinical effects of treatment. Information regarding specific doses, frequency and duration of treatments, and standardization of treatment was lacking from the review. According to the reviewers, adverse effects included headaches, dizziness, nausea, and drowsiness. Information regarding toxic effects, dropouts, and interactions was lacking. Outcome measures included sleep onset latency, sleep efficiency, rapid eye movement (REM) latency, sleep quality, wakefulness after sleep onset, total sleep time, and percentage of time in REM sleep. Based on a pooled analysis of results from 20 studies, treatment with melatonin resulted in a significant reduction in sleep onset latency vs. placebo in healthy individuals (weighted mean difference (WMD) -3.92 minutes, 95% CI -5.28 to -2.55 minutes, Z=5.63, p<0.00001). However, funnel plot analysis indicated that possible publication bias existed. In individuals with primary sleep disorders, treatment with melatonin resulted in a significant reduction in sleep onset latency vs. placebo (WMD -10.66, 95% CI -17.61 to -3.72, Z=3.01, p=0.003). In individuals with secondary sleep disorders or in people experiencing sleep restriction, treatment with melatonin lacked a statistically significant effect on sleep onset latency. Based on a pooled analysis of results from 13 studies, treatment with melatonin resulted in a significant increase in sleep efficiency vs. placebo in healthy individuals (WMD 2.3%, 95% CI 0.7 to 3.9%, Z=2.83, p=0.005). In individuals with primary sleep disorders, treatment with melatonin lacked a statistically significant effect on sleep efficiency, sleep quality, wakefulness after sleep onset, total sleep time, or percentage time spent in REM sleep. In individuals with secondary sleep disorders, treatment with melatonin lacked a significant effect on sleep efficiency, wakefulness after sleep onset, and percentage of time in REM sleep, but showed a significant beneficial effect on total sleep time (WMD 15.6 minutes, 95% CI 7.2 to 24.0 minutes). In individuals experiencing sleep restriction, treatment with melatonin lacked a significant effect on sleep efficiency, sleep quality, wakefulness after sleep onset, or percentage of time in REM sleep, but showed a significant beneficial effect on total sleep time (WMD 18.2 minutes, 95% CI 8.1 to 28.3 minutes). Based on a pooled analysis of 11 studies, treatment with melatonin lacked a statistically significant effect on REM latency in healthy individuals. Overall, the reviewers gave melatonin a grade A for efficacy in treating sleep disorders, with a level of evidence of 1b. Limitations of this review included the lack of detailed information regarding dosing, duration of treatment, standardization of treatment, dropouts, and interactions that occurred during the included studies.
        • Morera et al. conducted a review to evaluate whether melatonin may be considered as an alternative for the treatment of insomnia (1347). A computerized search spanning a 33 year period was performed, and 93 articles were collected. "Melatonin," "pineal gland," and "insomnia" were used as key words. Of the 93 articles collected, 85 were excluded because they were reviews or were not directly related to the research topic. A total of 111 insomniac patients were treated with melatonin in eight articles. 60% of the patients reported an improvement in sleep quality. Objective sleep measures also improved; there was a decrease in the sleep latency time and the number of awakenings (62% and 50% of patients, respectively) after melatonin treatment.
        • Wagner et al. conducted a systematic review to assess the clinical efficacy and safety of current therapies available for insomnia treatment (391). The effects of barbiturates, antidepressants, benzodiazepines, cloral hydrate, zaleplon, zolpidem, zopiclone, and valerian were assessed in this review but are excluded from this summary focusing on melatonin. Of the included studies that assessed melatonin, those that focused on only the pharmacokinetics of melatonin, in vivo results, or case reports were excluded from this summary; results for 18 included trials (370;379;393;411;417;463;482;584;612;649;650;651;653;655;657;658;1295;1296) and three reviews (1098;1227;1297) that assessed the clinical effects of melatonin are included in this summary. The authors performed a MEDLINE search to find case reports, reviews, abstracts, and clinical studies from April 1992 to December 1997. The references of the pooled articles were manually reviewed to identify additional relevant studies. Articles selected for inclusion had to include a specific patient population presenting with a particular condition, had to provide specific details regarding the treatment, and had to specifically describe the outcome measures used to assess the treatment. All included studies were published in English. According to the reviewers, doses of melatonin used for treatment generally ranged from 1-10mg in healthy participants or 0.3-10mg in patients with insomnia. Also, participants in some included trials were administered only physiologic doses of melatonin (0.1-0.3mg). Specific details regarding the duration of treatments was lacking. Information on standardization was lacking from the review. According to the reviewers, adverse effects were lacking in most of the included studies. However, in some studies, adverse effects included stomach cramps, dizziness, fatigue, headache, irritability, drowsiness in the day, increased depression, and inhibited ovarian function. Information on toxic effects, dropouts, and interactions was lacking in the review. Primary outcome measures included safety and efficacy outcomes of the different insomnia treatments. For inducing sleep, some studies concluded that high doses of melatonin were effective. The reviewers stated that melatonin has been used to treat jet lag and delayed sleep-phase syndrome, as well as decrease sleep latency and increase total sleep time. The reviewers reported that melatonin has been shown to have mild soporific effects when given in the day or early evening and hypnotic effects when taken before sleeping. Low doses of melatonin administered close to bedtime showed mixed results. According to the reviewers, melatonin may be the most beneficial when given two hours before sleeping rather than immediately prior to sleeping. Also, the reviewers concluded that melatonin may be more effective in treating insomnia in older individuals, as well as multi-disabled or neurologically impaired children. The authors concluded that melatonin has mixed effects depending on the patient population, dosage, time of administration and experimental design. The review was limited by lack of information regarding treatment frequency and duration, as well as specific data regarding treatment outcomes in the included studies.
        • Evidence: Wade et al. conducted a randomized, double-blind, placebo controlled, parallel group study to assess the effect of melatonin on primary insomnia in adult outpatients (N=791) (373). Men and women between the ages of 18-80 years who had primary insomnia (based on the Diagnostic and Statistical Manual of Mental Disorders fourth edition (DSM-IV) criteria) with a sleep latency of more than 20 minutes were included. Exclusion criteria were: 1) benzodiazepine or non-benzodiazepine hypnotic use in the preceding two weeks; 2) psychoactive treatment in the preceding three months; 3) sleep disorders due to psychiatric conditions or secondary to other conditions; 4) concomitant use of psychotropic treatments including antidepressants, antiepileptics, anxiolytics, barbiturates, first-generation antihistamines, hypnotics, lithium, neuroleptics, and treatments used as hypnotics such as barbiturates, benzodiazepines, buspirone, hydroxyzine, zaleplon, zolpidem, or zopiclone; 5) consuming alcohol to excess; or 6) leading a lifestyle that may interfere with sleep. In addition, participants who experienced short-term changes in their condition or who lacked compliance with treatment during the run-in period were excluded. A single-blind, placebo run-in period (two weeks) was followed by a double-blind treatment period (weeks 1-3) and a double-blind extension period (26 weeks: weeks 4-29). During the first double-blind period, participants were randomized in a 1:1 ratio to receive one tablet of 2mg prolonged-release melatonin (Circadin® 2mg, Neurim Pharmaceuticals Ltd., Tel Aviv, Israel) or placebo daily two hours before bedtime for three weeks. During the double-blind extension period, all participants initially randomized to the melatonin group continued with that treatment for an additional 26 weeks, while participants initially randomized to the placebo group were randomized in a 1:1 ratio to continue placebo treatment or to begin receiving melatonin treatment for 26 weeks. This was followed by a two-week, single-blind placebo run-out period. A description of standardization was lacking. Adverse events occurred in both placebo and treatment groups. Adverse events included arthralgia, diarrhea, headache, lower and upper respiratory tract infections, and nasopharyngitis. One subject experienced palpitations in the treatment group. A description of toxic effects was lacking. In the treatment group, reasons for dropouts included unwillingness to continue (N=59), occurrence of an adverse event (N=26), ineligibility to continue (N=9), lost to follow-up (N=9), lacking compliance (N=4), withdrawal of consent (N=1), and unknown reasons (N=3). In the placebo group, reasons for dropouts included unwillingness to continue (N=24), occurrence of an adverse event (N=10), ineligibility to continue (N=6), lost to follow-up (N=1), and unknown reasons (N=2). A description of interactions was lacking. The main outcome measure was change in sleep latency after the first three-weeks of treatment. Other outcomes included Clinical Global Impression of Improvement (CGI-I), Pittsburgh Sleep Quality Index (PSQI) score, and Quality of Life (based on the World Health Organzaton-5 Well-being Index). For participants aged 18-80 years who were low excretors (6-sulphatoxymelatonin levels ≤8mcg nightly) a significant benefit of melatonin treatment on sleep latency and most other outcome parameters was lacking. However, a reduction in sleep disturbances (PSQI component 5; -0.10, 95% CI -0.18 to -0.03, p=0.008) and quality of life (1.21, 95% CI 0.22 to 2.20, p=0.016) was observed for this group vs. placebo after three weeks. After six months of treatment, these participants showed a significant improvement in total sleep time (13.1 minutes, 95% CI 1.0 to 25.2, p=0.035), PSQI global scores (-0.66, 95% CI -1.30 to -0.01, p=0.046), WHO-5 Index scores (0.91, 95% CI 0.16 to 1.66, p=0.017), CGI-I scores (-0.25, 95% CI -0.49 to -0.01, p=0.042), and PSQI question two (-11.6, 95% CI -22.0 to -1.1 minutes, p=0.030) compared to the placebo group. For participants aged 65-80 years, a significant improvement in sleep latency (-15.6min, 95% CI -25.3 to -6.0, p=0.002), PSQI question two (-13.7, 95% CI -23.5 to -3.9, p=0.006), PSQI component 2 (-0.23, 95% CI -0.41 to -0.04, p=0.018), sleep maintenance (-0.17, 95% CI -0.33 to 0.00, p=0.046), time going to bed (-0.22, 95% CI -0.39 to -0.05, p=0.012), and PSQI global score (-0.64, 95% CI -1.25 to -0.02, p=0.042) was observed vs. placebo after three weeks. After six months of treatment, participants aged 65-80 years showed significant improvement in sleep latency (-14.5 minutes, 95% CI -21.4 to -7.7, p<0.001), time going to bed (-0.21 hours, 95% CI -0.33 to -0.08, p=0.002), PSQI global score (-0.70, 95% CI -1.17 to -0.23, p=0.003), component one of the PSQI score (-0.15, 95% CI -0.25 to -0.04, p=0.006), component two of the PSQI score (-0.24, 95% CI -0.38 to -0.10, p=0.001), PSQI score question two (-12.1 minutes, 95% CI -19.1 to -5.1, p=0.001), morning alertness (-0.10, 95% CI -0.19 to -0.01, p=0.032), and CGI score (-0.20, 95% CI -0.38 to -0.02, p=0.027) compared to the placebo group. This was a well-designed trial.
        • Tzischinsky and Lavie conducted a crossover, placebo controlled, double-blind trial of eight men in a 7/13 ultrashort sleep-wake paradigm (following an overnight sleep deprivation) to assess the efficacy of melatonin on inducing sleep (662). All subjects were men, with an average age of 27.06 ± 3.7 years. Subjects were administered the Technion Sleep questionnaire and the Horne and Ostberg questionnaire prior to experimentation to exclude sleep disturbances and/or extreme morning or evening types. During the sessions, subjects were administered a pill at four different times (noon, 5 p.m., 7 p.m., and 9 p.m.). In four of the sessions, three of the pills were placebo and one contained 5mg of melatonin. Melatonin administration was varied according to a Latin square design. During a fifth session, all pills administered were placebo. Four of the subjects received the all-placebo trial during the first experimental session, while the other four received it in the last. Outcome measures included polysomnographic recordings (EEG), oral temperature, and subjective reports of sleepiness. Subjects receiving melatonin at noon, 5 p.m., 7 p.m., and 9 p.m. experienced significant effects of treatment (p<0.1, p<0.003, p<0.0001, p<0.004, respectively), trial (p<0.0001, p<0.008, p<0.0001, p<0.0001, respectively), and a significant interaction trial x treatment (p<0.03, p<0.01, p<0.0001, p<0.001, respectively). Wilcoxon tests revealed significant drops in oral temperature following administration of melatonin at all times except 9 p.m. This study suggests an increase in sleep propensity in subjects ingesting melatonin prior to the desired rest period; however, it is somewhat limited by the homogeneity of the population used (adult males of a narrow age range).
        • Wyatt et al. conducted a randomized, double-blind, placebo controlled, parallel-group design trial to investigate the effects of melatonin on sleep latency and sleep efficiency in sleep episodes initiated across a full range of circadian phases (656). Subjects (N=26) were healthy young adults aged 18-30 (male, N=15; female, N=11). Treatment consisted of pharmaceutical-grade melatonin obtained from Regis Technologies (Morton Grove, IL); dose was either 0.3 or 5mg delivered orally and administered 30 minutes prior to each 6.67-hour sleep episode during forced desynchrony from endogenous circadian rhythm. All subjects received placebo capsules during the first and last three days of the protocol. Outcome measures included polysomnographic recordings (EEG), sleep time and latency, core body temperature, and stage-distribution of sleep. Analysis revealed both doses of melatonin improved polysomnographically determined sleep efficiency from 77% in the placebo group (0.3mg=84%, 5.0mg=83%; all p<0.05) for sleep episodes occurring during circadian phases when endogenous melatonin was absent (EM-). However, this remained below the average sleep efficiency of 88% observed during sleep episodes scheduled during the circadian night, when endogenous melatonin was present (EM+). Comparisons within the three drug groups showed that all had statistically significant, lower sleep efficiency during EM- vs. EM+ (all p<0.05). Melatonin did not affect sleep initiation, core body temperature, or relative stage-distribution of sleep in a statistically significant manner. This study reported an increase in sleep efficiency in response to melatonin treatment when administered outside of the increased endogenous melatonin phase of the body's natural circadian rhythm. Although children and the elderly were lacking in this study, this study was thorough and well characterized.
        • Kunz et al. conducted two randomized, double-blind, placebo controlled trials to determine the effects of exogenous melatonin on disturbed REM sleep in healthy adults (652). The patients included in the study had contacted the interdisciplinary sleep clinic voluntarily and were seeking treatment for neuropsychiatric sleep-related disturbances (International Classification of Sleep Disorders criteria). Polysomnography (PSG) was performed in relevant subjects. Patients were included in the study if they were found to have a quantitative reduction in REM sleep duration 25% or more below their age norm. Exclusion criteria included age less than 18 or more than 80, pregnancy, current or recent shift work (during the last year), poor sleep hygiene (sleep log or actigraphic proof of a more than two-hour variation in bedtime during a 14-day recruitment period), definite morning or evening types (regular bedtime outside 10-12 p.m.), transmeridian travel (during or within one month of the study), psychiatric disorders (except mood disorders in remission, Diagnostic and Statistical Manual of Mental Disorders-IV), pathological findings in brain imaging, recent changes in medication (within one month of the study), and the intake of any medication that might interfere with melatonin production or secretion or REM sleep. Sixteen subjects were included, but two were excluded from analysis due to compliance issues. In study 1, subjects were assigned to be treated with melatonin 3mg daily for four weeks. After a 3-5-day washout period, patients treated with melatonin received placebo in study 2, and patients treated with placebo in study 1 received melatonin in study 2 for four weeks. Melatonin was found to be more effective than placebo in terms of significant increases in REM sleep percentage (baseline/melatonin, 14.7/17.8 vs. baseline/placebo, 14.3/12.0, p<0.010) and improvements in subjective measures of daytime dysfunction, as well as clinical global impression score. Melatonin lacked an effect on the shift of circadian phase, but did increase REM sleep continuity and promote decline in rectal temperature during sleep. Dropouts were not described. Otherwise, the study was well designed.
        • Wade et al. conducted a randomized, double-blind, placebo controlled trial to assess the effects of prolonged release melatonin (PRM) on primary insomnia (N=791) (374). Referred or self-referred individuals aged 18-80 years with primary insomnia were included in the study. Primary insomnia was defined as sleep latency >20 minutes (according to the Diagnostic and Statistical Manual for Mental Disorders, fourth edition). Individuals were excluded if they had taken hypnotics in the past two weeks or psychoactive treatment in the past three months, had a psychiatric disorder associated with their sleep disorder, had a sleep disorder that was secondary to another medical disorder (such as chronic pain) or lifestyle (such as shift work or jet lag), or consumed alcohol excessively. Also, participants were excluded if they used the following medications: antidepressants, antiepileptics, antihistamines (first generation), anxiolytics, barbiturates, hyptnotics, lithium, or neuroleptics. Following a two-week run-in period, during which time all participants received placebo treatment, participants were randomized to receive 2mg PRM (Circadin®) or placebo daily 1-2 hours before bed for three weeks. Following the first three weeks of treatment, participants in the placebo arm were randomized again in a 1:1 ratio to receive 2mg PRM or placebo for a 26-week extension period. All participants in the treatment group continued receiving 2mg PRM daily during the extension period. Information on standardization was lacking. Side effects that the authors reported to be possibly treatment-related included heart palpitations. Adverse effects associated with placebo treatment included burning sensation, labrynthitis, and pharyngolaryngeal pain. Overall, the frequency of adverse events was similar in the treatment and placebo groups. Information on toxic effects was lacking. Of the 791 participants randomized in the three week trial, 69 dropped out before the treatment was complete. The most common reasons for dropouts included consent withdrawal, lost to follow-up, or an adverse event. Of the 722 participants who completed the first three weeks of treatment, 711 were included in the extension phase. Of these participants, 156 withdrew due to consent withdrawal or an adverse event. Additional details regarding specific reasons for dropouts were reported by the authors. Information on interactions was lacking. The primary outcome measure was sleep latency, measured using a sleep diary. The Pittsburgh Sleep Quality Index (PSQI) global score, WHO-5 Well-being Index score (1998 version), and the Clinical Global Impressions Scale score for severity of illness (CGI-S) were also used to assess sleep quality. Compared to the placebo group, three weeks of treatment with melatonin lacked a statistically significant effect on sleep latency for participants aged 18-80 years and the subgroup of participants aged 18-54 years. However, for participants aged 55-80 years, treatment with melatonin for three weeks resulted in a significant difference in sleep latency vs. placebo (-15.4 vs. -5.5 minutes, p=0.014). When the older (55-80 years) and younger (18-54 years) subgroups of participants were compared, the effect of three weeks of melatonin treatment on sleep latency in the older participants population was significantly different vs. the younger participants (p=0.034). When all participants were considered, treatment with melatonin for three weeks resulted in significant difference in PSQI Q2 score (-7.8 minutes, 95% CI -13.4 to -2.2, p=0.006), PSQI C2 score (-0.13, 95% CI -0.24 to -0.02, p=0.023), time going to bed (-0.14 hours from midnight, 95% CI -0.24 to -0.03, p=0.011), and PSQI global score (-0.44, 95% CI -0.84 to -0.05, p=0.027) compared to the placebo group. Significant between-group differences in total sleep time and quality of life were lacking. For the younger participants, treatment with melatonin for three weeks lacked a significant effect on PSQI Q2 score, PSQI C2 score, total sleep time, time going to bed, sleep quality, and quality of life. For participants aged 55-80 years, treatment with melatonin for three weeks resulted in a statistically significant between-group difference in PSQI Q2 score (-9.5 minutes, 95% CI -5.8 to -3.3, p=0.003), PSQI C2 score (-0.17, 95% CI -0.29 to -0.05, p=0.005), time going to bed (-0.15 hours from midnight, -0.26 to -0.03, p=0.014), total sleep time (0.15 hours, 95% CI 0.00 to 0.31, p=0.048), sleep quality (-0.65, 95% CI -1.09 to -0.21, p=0.003), and quality of life (0.65, 95% CI 0.12 to 1.19, p=0.017) compared to the placebo group. Considering all ages after six months of treatment, participants administered melatonin showed significant difference in sleep latency based on diary (-6.0 minutes, 95% CI -10.0 to -2.1, p=0.003), PSQI C2 score (-0.10, 95% CI -0.20 to -0.01, p=0.032), PSQI Q2 score (-6.8 minutes, 95% CI -10.9 to -2.6, p=0.001), time going to bed (-0.13 hours from midnight, 95% CI -0.20 to -0.05, p=0.002), sleep quality (-0.39, 95% CI -0.71 to -0.08, p=0.014), quality of sleep (-0.08, 95% CI -0.15 to 0.00, p=0.046), daytime functioning (-0.07, 95% CI -0.13 to 0.00, p=0.040), morning alertness (-0.07, 95% CI -0.13 to 0.00, p=0.047), quality of life (0.46, 95% CI 0.11 to 0.81, p=0.011), and clinical status (-0.12, 95% CI -0.24 to -0.01, p=0.036) vs. placebo. The participants aged 18-54 years lacked these between-group differences, while participants aged 55-80 years showed all of these differences except for morning alertness after six months of treatment with melatonin vs. placebo. The authors concluded that PRM has both short-term and long-term benefits on insomnia, particularly for individual aged 55-80 year. The study was limited by the high dropout rate, and lack of blinding and randomization description.
        • Paul et al. conducted a double-blind, placebo controlled, crossover study to compare the sleep-inducing power of four medications (zopiclone, zaleplon, melatonin, and temazepam) (654). Nine men and 14 women, aged 21-53 years, were assessed for psychomotor performance before and for seven hours after ingestion of a single dose of placebo, zaleplon 10mg, zopiclone 7.5mg, temazepam 15mg, or time-released melatonin 6mg. Subjects wore polysomnographic electrodes to record total sleep and sleep latency during four-minute periods with eyes closed immediately before and after each psychomotor test sequence. Subjective drowsiness was assessed by questionnaire. There were drug x trial interactions for zaleplon, zopiclone, and temazepam for total sleep, sleep latency, and subjective drowsiness. More sleep, shorter sleep latency, and more drowsiness occurred immediately after psychomotor testing compared to before testing for all medications. Melatonin did not cause any sleep prior to psychomotor testing sessions, but caused sleep and reduced sleep latency after psychomotor test sessions from 1.75 hours to 4.75 hours postingestion. The authors reported the sleep-inducing power of the medications before psychomotor testing as zopiclone > zaleplon > melatonin > temazepam. The corresponding effect after psychomotor testing was reported to be zopiclone > melatonin > zaleplon > temazepam.
        • Pinto et al. conducted a placebo controlled, double-blind trial of 40 young adult males using two different criteria for sleep onset (10 minutes of uninterrupted sleep and first three epochs of uninterrupted stage 1 sleep) to assess the effect of melatonin on inducing the onset of sleep (660). All subjects were men, with an average age of 28 ± 5 years. Subjects were administered 10mg of melatonin or a placebo every day for 28 days one hour prior to sleep (10 p.m.). Outcome measures included polysomnographic recordings (EEG, EOG in both eyes, and EMG in both legs). Analysis revealed a statistically significant decrease in sleep latency (p<0.05) using one criterion (10 minutes of uninterrupted sleep). The study is also somewhat limited by the homogeneity of the population used (adult males of a narrow age range).
        • Almeida Montes et al. conducted a randomized, double-blind, placebo controlled, crossover trial to assess the efficacy and safety of melatonin in 10 patients with primary insomnia (648). Patients with circadian rhythm sleep disorders were excluded, as were various diseases and medications (four-week washout was acceptable). Patients (mean age: 50 years, range: 30-72 years) met the DSM-IV criteria for primary insomnia. Patients were administered 0.3mg of melatonin, 1mg of melatonin, or placebo 60 minutes before bedtime. Each patient received each of the three treatments for a seven-day period (with a five-day washout period in between) by crossover method. After each seven-day treatment, nighttime electroencephalographic (EEG) records were collected. Each morning, subjects completed sleep logs and analog-visual scales to document the amount and subjective quality of sleep. Differences were lacking between placebo and the two doses of melatonin in sleep EEG or subjective measures of sleep (time, quality, or latency, REM sleep). Differences in adverse effects were lacking between groups. Randomization and dropouts were not described.
        • Cajochen et al. conducted a randomized, placebo controlled, double-blind crossover trial of eight healthy young men to assess the effect of melatonin on subjective and objective measures of sleepiness (661). Eight male students (average age in experiment 1: 27 ± 4 years old; in experiment 2: 24.8 ± 3.5 years old) were paid to participate in the study. The study comprised two separate experiments during which melatonin was administered at 6 p.m. or 1 p.m. Each subject participated in two consecutive treatment periods made up of one day with placebo and one day of treatment with melatonin, in randomized order. Volunteers reported to the chronobiology laboratory at 3 p.m. (exp. 1) or 9 a.m. (exp. 2) each day for the eight-hour experimental period. Subjects remained awake throughout experimentation. Discussion of toxic or interaction effects was lacking. Dropouts were also not specifically addressed; however, data were reported for all eight subjects in both experiments. Outcome measures included half-hourly self-ratings of fatigue and mood obtained on 100mm visual analogue scales (VAS), as well as subject-completed Åkerstedt Sleepiness Symptoms Check Lists (ASSC) and Åkerstedt Sleepiness Scales (ASS). Furthermore, EEG, EOG, EMG, and ECG recordings were also made. Saliva was collected in parallel for analysis of melatonin. Analysis revealed a significant increase in subjective sleepiness as rated on the VAS, ASSC, and ASS in experiment 1. The ANOVAs for all three scales separately revealed a significant interaction term "treatment x time" (VAS, p<0.02; ASSC, p<0.04; ASS, p<0.001). Following treatment, self-rated sleepiness on the ASS remained higher than after placebo from 7:30 p.m. for the remainder of the evening (p<0.05). In experiment 2, melatonin administration increased subjective sleepiness as rated on the ASS (p<0.05) but differences in the other two scales (VAS and ASSC) were lacking. Statistical analysis of EEG data showed that the theta/alpha-band (5.25-9Hz) was significantly higher following melatonin treatment compared to placebo (p<0.05). Further analysis of the temporal relationship between melatonin concentrations, subjective sleepiness, and theta/alpha-activity in the waking EEG revealed both subjective sleepiness and EEG power center points significantly different from the melatonin center point (p<0.002). In addition, the subjective sleepiness center point occurred 23.4 minutes later than the EEG power center point (p<0.002). Testing of the relationship between melatonin levels and the amount of increase in subjective and objective fatigue as measure in the waking EEG showed a lack of statistically significant correlations between subjective fatigue and the area under the melatonin curve (AUC), but statistically significant, negative correlations between the center point of subjective sleepiness and melatonin AUC (p<0.05). The authors concluded that this study presented evidence that daytime administration of melatonin had an acute and strong effect on subjective and objective measures of sleepiness, though these findings are somewhat limited by the homogeneity of the subject population.
        • Satomura et al. conducted a randomized, single-blind, placebo controlled, equivalence trial to assess the presence or absence of hypnotic action by administering exogenous melatonin during the day when secreted levels of melatonin are low, compared to triazolam (448). Seven male college students were included in the study. Hypnotic action, effects on rectal temperature, and dose dependency by daytime administration were assessed. Melatonin (1mg, 3mg, 6mg), triazolam 0.125mg, or placebo (lactose) was administered at 1:30 p.m.; however, discussion of the duration and frequency was lacking. Exogenous melatonin (1mg, 3mg, and 6mg) significantly increased total sleep time and sleep efficiency compared to placebo (p<0.05). Melatonin 6mg was observed to demonstrate hypnotic effects that were nearly equal to those of triazolam at 0.125mg. Rectal temperature was decreased at melatonin 1mg and 3mg; however, the hypothermic action of triazolam and melatonin 6mg was lacking in statistical significance. Dropouts, randomization, and blinding were not described.
        • Ellis et al. conducted a randomized, double-blind, placebo controlled trial to evaluate the hypnotic action of melatonin in 15 subjects with psychophysiological insomnia (393). Inclusion and exclusion criteria were lacking. Melatonin 5mg or placebo was taken at 8 p.m. for a one-week period. Endpoints included the effects on sleep and wakefulness (visual analog scale and interview). Melatonin lacked an effect on bedtime, sleep onset time, estimated total sleep and wake time, estimates of next-day function, or self-rated sleep quality. The period of melatonin treatment was retrospectively correctly identified by eight of 15 subjects. Seven of 15 subjects reported that sleep had subjectively improved with active treatment. Possible adverse effects to melatonin included mild headache, poor sleep quality, odd taste in mouth, and "muzziness." Limitations include the lack of description of randomization, blinding, and withdrawals.
        • Studies of lesser methodological strength (not included in the Evidence Table): Nave et al. conducted a placebo controlled, double-blind trial of 12 young adults to assess the combined effect of melatonin of two different dosages and presleep dosage intervals on the onset and length of sleep of an early-evening nap (653). Subjects were all students, with an average age of 24.6 ±2.7 years, following more or less the same daily schedule. None had used drugs or medications for at least three months prior to the start of the study. Patients were given one of the following treatments: placebo and placebo, placebo and 3mg of melatonin (or in the reverse order), placebo and 6mg of melatonin (or in the reverse order). There was to be no daytime sleep during the study. Outcome measures included polysomnographic recordings, sleep time and latency, and questionnaires probing subjective sleep experience. All subjects completed the study. Analysis revealed a decrease in sleep latency (p<0.0054) and an increase in sleep time (p<0.0001) in all drug conditions. Additionally, all conditions exhibited an increase in stage 2 sleep time (p<0.0054). The authors neglected to discuss their particular methodology for polysomnographic recording and how they determined sleep time and onset latency from these recordings. Discussion of adverse effects was lacking. Randomization was not discussed.
        • MacFarlane et al. conducted a conducted a double-blind, placebo controlled, crossover study to evaluate the effects of melatonin on total sleep time and daytime alertness in 13 patients with chronic insomnia (417). Patients were otherwise healthy and free of neuroactive medications for at least four weeks. Patients were also prescreened by polysomnogram and subjective ratings of sleep quality and daytime alertness for one week. Patients were administered melatonin 75mg per night or placebo at 10 p.m. daily for 14 consecutive days. The dose was based on a preliminary dosing study in two patients in which 100mg was associated with headaches. An increase in the subjective assessment of total sleep time and daytime alertness (p<0.05) was demonstrated with melatonin but changes with placebo were lacking. However, seven of the 13 patients reported that the active treatment had no effect on subjective feelings of well-being. This study is limited by the lack of randomization and inadequate description of blinding and withdrawals.
        • Stone et al. conducted a controlled trial to determine the effect of different levels of melatonin upon nocturnal and evening sleep, as well as on core body temperature (427). When given at 11:30pm 5mg melatonin reduced the duration of stage 3 in the first 100 minutes of sleep and 0.1mg melatonin reduced body temperature between 6.5 and 7 hours after ingestion. When given at 6:00pm, all doses of melatonin increased total sleep time, sleep efficiency index, and stage 2, and reduced wakefulness. Changes in body temperature were lacking. The authors concluded that melatonin lacked an effect when given for nocturnal sleep but had a hypnotic effect when given in the early evening.
        • Atkinson et al. conducted a controlled trial of 12 subjects to examine the effects of melatonin on the subjective quality of sleep in athletes and any residual effects on physical performance the morning after treatment (593). Exclusion and inclusion criteria were not specifically discussed; however, the subjects included were described as being healthy, physically active sports science undergraduates, ages 19-30 years. The subjects were given 5mg of melatonin or a placebo; a single dose of one capsule was taken at 11:30 p.m., 30 minutes before retiring. A specific statement of the primary outcome was lacking, although subjective measure of sleep quality (maintenance and latency) is assumed. Secondary measures included objectively measured intra-aural temperature, left and right grip strength, and time taken to complete a 4km stationary bike ride. Secondary, subjective measures include perceived exertion. ANOVA was used to analyze grip strength and cycle time data. Significant differences between the placebo and melatonin groups were lacking for right [t11=-0.15(p=0.88)] or left [t11=0.35(p=0.74)] hand grip strength, cycle times [t11=0.44(p=0.67)], or sleep latency [W=25.5 (p=0.33)] and maintenance [W=12.5 (p=0.87)] (Wilcoxon tests, p>0.5). Limitations of this study include a lack of randomization, inclusion and exclusion criteria, and adequate description of the blinding process, as well as the short study duration and overall suboptimal study design. Standardization, allergies and adverse effects, toxic effects, dropouts, and interactions were not discussed.

        Age-related macular degeneration

        • Summary: Melatonin may exert antioxidant effects, which may contribute to its beneficial effects on the eyes. According to clinical research, melatonin may play a role in protecting the retina to delay macular degeneration (459). Well-designed clinical trials are needed before a conclusion can be made.
        • Studies of lesser methodological strength (not included in the Evidence Table): Yi et al. conducted a case control study to evaluate the effects of melatonin in age-related macular degeneration (459). Subjects with dry or wet forms of age-related macular degeneration were included. One hundred patients with age-related macular degeneration were given melatonin 3mg each night at bedtime for at least three months. Fifty-five patients were able to be followed for more than six months. Visual acuity was stable at six months of treatment. Eight eyes showed more retinal bleeding and six eyes more retinal exudates.

        Aging (thermoregulation)

        • Summary: Melatonin may be helpful in regulating age-dependent changes in body temperature rhythm (554). More well-designed trials are needed before a conclusion can be made.
        • Evidence: Gubin et al. conducted a randomized controlled trial to examine the effects of melatonin in 97 patients involved in a larger study dedicated to the study and treatment of age-dependent changes or disturbances in the circadian system in humans (554). Those with mental disorders and alcohol abuse were excluded. Both genders were included. Patients were aged 63-91 years. Patients lived a self-chosen sleep-wake regimen. After one control week, part of the group (N=63) received 1.5mg of melatonin (Melaxen®) daily at 10:30 p.m. for two weeks. The other 34 subjects were given placebo. Younger subjects (58 young adults, both genders, 17-39 years of age) were used as controls. Endpoints included rhythm characteristics of body temperature. The MESOR (36.38 ± 0.19 degrees C vs. 36.17 ± 0.21 degrees C) and circadian amplitude (0.33 ± 0.01 degrees C vs. 0.26 ± 0.01 degrees C) were slightly decreased in the elderly compared to the young adult subjects (p<0.001) at baseline; however, the mean circadian acrophase was similar in both age groups (17.19 ± 1.66 vs. 16.93 ± 3.08 hours). Interindividual differences were higher in the older group (varying between 10 a.m. and 11 p.m.). With melatonin treatment, the MESOR was lower by 0.1 degrees C, and the amplitude increased to 0.34 ± 0.01 degrees C (similar to that found in young adults). A significant effect on the change in the mean acrophase was lacking. This study is limited by a lack of description of randomization, blinding, and withdrawals.

        Alzheimer's disease/ cognitive decline

        • Summary: Limited research has been conducted to examine the effects of melatonin on cognitive disorders. Some randomized controlled trials suggest a possible benefit (558;561). Nonrandomized clinical trials have also been conducted (555;556) and, overall, results agree with those of the RCTs. In elderly patients suffering from mild cognitive impairment, an oily emulsion of DHA-phospholipids containing melatonin and tryptophan resulted in an improvement in MMSE and the olfactory sensitivity assessment; statistically nonsignificant improvements were found for semantic verbal fluency (1348). More well-designed trials are needed before a conclusion can be made.
        • Systematic reviews: de Jonghe et al. conducted a systematic review of nine studies (556;557;559;560;642;1349;1350;1351;1352) to assess the effect of melatonin on circadian rhythm disturbances in individuals with dementia (562). A search was conducted on literature published in any language from1985 and April 2009 using PubMed, CINAHL, Embase, and the Cochrane Database of Systematic Reviews. The references of pooled studies were reviewed to identify additional relevant studies. All of the included studies assessed the effects of melatonin treatment in people with dementia. All prospective studies were included, while studies that appeared to be a commentary, guideline, reviews, or case reports based on the title or abstract were excluded. Studies were excluded if the effect of melatonin alone was unclear or if separate results for patients who were cognitively and noncognitively impaired were lacking. In cases in which the same study was published more than once, the most recent publication that assessed the largest participant population was included. Participants in the included studies were administered 2.5-10mg melatonin daily for 10 days to 35 months. Information regarding standardization of the melatonin products, allergies, adverse events, toxic effects, dropouts, and interactions was lacking. Sundowning/agitated behavior was assessed by multiple tools including a neuropsychiatric inventory (NPI), the agitated behavior rating scale (ABRS), the Cohen-Mansfield Agitation Inventory (CMAI), the Alzheimer's Disease Assessment scale noncognitive section (ADAS non-cog), the coefficient of variation of bedtime, actigraph, and clinical observation. Sleep quality was assessed with an actigraph, daily sleep diaries, the Standard Sleep Quality Questionnaire (SSPQ), a visual analog scale (VAS), structured interviews, and sleep logs. Daytime functioning was assessed with actigraphs, sleeplogs, structured interviews, and the CMAI. Two out of four randomized controlled studies found significant improvement in the sundowning/agitated behavior in participants who took melatonin compared to those who took placebo. Of the five included case series, all reported improvement regarding this parameter. Of the three studies that assessed sundowning/agitated behavior based on clinical observations, disappearance of the syndrome was observed in 86% and 100% of participants. Sleep quality was assessed in eight of the nine studies. For three of the randomized controlled trials that assessed sleep quality based on the actigraph, positive improvement was lacking. However, one randomized controlled trial showed significant improvement in sleep quality based on sleep diary data. Of the five case-series studies, two found significant improvement using the SSPQ and VAS assessment scales, while one showed improvement based on the actigraph. Daytime functioning was assessed in four out of the nine studies. Based on actigraph measurements, improvement in daytime functioning was shown in one randomized controlled trial, while three case series showed a lack of improvement. Limitations of this systematic review included the fact that five of the included studies were case series. In addition, the assessment scales lacked uniformity throughout the studies. Also, the reviewers included studies in which the participants were living under different care conditions (home vs. hospital). Finally, data was lacking regarding dropouts, adverse events, drug interactions, allergies, and standardization.
        • Jansen et al. conducted a systematic review to assess the clinical efficacy and safety of melatonin in the treatment of manifestations of dementia or cognitive impairment (561). The Cochrane Dementia and Cognitive Improvement Group's Specialized Register was searched on 5 October 2005 for trials involving melatonin. The search terms used were "melatonin" and "n-acetyl-5-methoxytryptamine." All relevant randomized controlled trials in which orally administered melatonin in any dosage was compared with a control group for the effect on managing cognitive, behavioral (excluding sleep), and/or affective disturbances of people with dementia of any degree of severity were included. Two or three reviewers independently assessed the retrieved articles for relevance and methodological quality and extracted data from the selected studies. Statistically significant differences in changes in outcomes from baseline to end of treatment between the melatonin and control groups were examined. Each study was summarized using a measure of effect (e.g., mean difference), and meta-analyses were conducted when appropriate. Three studies met the inclusion criteria (557;559;560). This review revealed statistically nonsignificant effects from the pooled estimates of MMSE cognitive and ADAS cognitive change scores. Individual study estimates for treatment effect demonstrated a statistically significant improvement for melatonin compared with placebo in behavioral and affective symptoms as measured by the ADAS noncognitive scale in a study of 20 patients, and the Neuropsychiatric Inventory (NPI) following treatment with 2.5mg of slow-release melatonin daily, but not with 10mg of immediate-release melatonin daily in a larger study of 157 patients. The remainder of the treatment effects were statistically nonsignificant. Further research is required.
        • Evidence: Peck et al. conducted a randomized, double-blind, placebo controlled trial to determine if exogenous melatonin would improve cognitive function in 30 healthy elderly subjects (558). Patients with psychiatric or cognitive problems were excluded, as were those with unstable medical problems or drug concerns. Subjects received melatonin 1mg or placebo each night for four weeks. Subjects completed a sleep questionnaire and a battery of cognitive tests at baseline and after four weeks. Results were included for 26 subjects, because of benzodiazepine use or findings of dementia. Significant improvements were noted in morning restedness in the melatonin group (p<0.04). Melatonin administration also improved scores on the California Verbal Learning Test (p<0.02). Randomization was not adequately described in this study.

        Anti-inflammatory

        • Summary: Melatonin has been reported to decrease upregulation of proinflammatory cytokines (831). Other anti-inflammatory effects may be related to inhibition of nitric oxide (NO) and malondialdehyde (MDA) production, or increase of glutathione levels (833;834). Use in inflammatory conditions has been proposed (1186). Based on limited human research, melatonin may be an effective anti-inflammatory (11), decreasing concentrations of IL-6, IL-8, and TNF-alpha. However, there is conflicting evidence that melatonin may actually induce a proinflammatory response and may increase plasma kynurenine concentrations in certain populations (464). Further well-designed study is required.
        • Evidence: Forrest et al. conducted a randomized controlled trial to investigate the anti-inflammatory effect of melatonin in 75 patients with rheumatoid arthritis (464). Patients were allocated randomly to receive melatonin 10mg at night, in addition to ongoing medication, or placebo, for six months. Blood samples were drawn monthly for endpoints that included the following: disease severity by analysis of inflammatory indicators [CRP, erythrocyte sedimentation rate (ESR), neopterin], proinflammatory cytokines (IL-1beta, IL-6, TNF-alpha), lipid peroxidation products, and the kynurenine pathway metabolites of tryptophan. The authors reported an increase of the erythrocyte sedimentation rate (ESR) (two-way ANOVA F(1,127)=5.24, p=0.024) and neopterin concentrations (F(1,136)=4.64, p=0.033) in patients treated with melatonin compared with controls. In the placebo group, peroxidation products showed a statistically significant decrease, but a decrease was lacking in melatonin-treated patients. The authors reported that these results suggested a proinflammatory response. Although over time, concentrations of proinflammatory cytokines, IL-1beta, IL-6, and TNF-alpha were elevated in the melatonin group compared to placebo, in a majority of measurements taken at specific intervals, a statistical significance in these findings was lacking. The melatonin group also demonstrated increased plasma kynurenine concentrations (F(1,124)=4.24, p= 0.041), again suggesting proinflammatory activity. This finding was statistically significant. The authors concluded that a daily dose of 10mg of melatonin showed a slowly developing antioxidant profile in patients with arthritis and increased the concentrations of some inflammatory indicators, but these effects were not associated with changes in clinical symptoms. Dropouts and double-blinding procedures were unclear.
        • Gitto et al. conducted a randomized controlled trial to determine if treatment with melatonin would influence proinflammatory cytokines (IL-6, IL-8, TNF-alpha), and nitrite/nitrate levels in 40 newborns with grade III or IV respiratory distress syndrome (11). Patients with radiographically confirmed respiratory distress syndrome, diagnosed within the first six hours of life, were included. Compared with the melatonin-treated respiratory distress syndrome newborns, in the untreated infants, the concentrations of IL-6, IL-8, and TNF-alpha were significantly higher at 24 hours, 72 hours, and seven days after the onset of treatment. In addition, the authors noted that nitrite/nitrate levels at all time points were higher in the untreated respiratory distress syndrome newborns than in the melatonin-treated babies. Following melatonin administration, nitrite/nitrate levels decreased significantly, whereas they remained high and increased further in the respiratory distress syndrome infants not given melatonin. Dropouts, blinding, and randomization procedures were unclear.

        Benzodiazepine tapering

        • Summary: A small amount of research has examined the use of melatonin to assist with tapering or cessation of benzodiazepines such as diazepam (Valium®) or lorazepam (Ativan®) (565;567;568;569). Nonrandomized clinical trials have also been conducted (566) and, overall, the results agree with those observed in the randomized controlled trials. Baandrup et al. reported on the protocol for a study investigating the effect of prolonged-release melatonin for benzodiazepine discontinuation in patients with schizophrenia (1353). Although preliminary results are promising, due to weaknesses in the design and reporting of this research, further research is necessary before a firm conclusion can be reached.
        • Evidence: Garfinkel et al. conducted a randomized, double-blind crossover trial to investigate the efficacy of melatonin replacement therapy in improving sleep in 21 elderly subjects taking benzodiazepines for insomnia (567). Inclusion and exclusion criteria were lacking. Subjects were treated for three weeks with 2mg per night of controlled-release melatonin (Circadin®; Neurim Pharmaceuticals Ltd., Tel Aviv, Israel) and for three weeks with placebo, two hours before desired bedtime, with a one-week washout period between treatment periods. Subjects' sleep was assessed by wrist actigraphy. All subjects completed the study. Reported adverse effects were lacking. Melatonin treatment significantly increased sleep efficiency and total sleep time and decreased wake after sleep onset, sleep latency, number of awakenings, and fragmental index, compared to placebo (p<0.05). This study is limited by the lack of description of randomization.
        • Cardinali et al. conducted a randomized, double-blind, placebo controlled study to assess whether melatonin (3mg in fast-release form) could be useful to reduce benzodiazepine dosage in 45 older patients with minor sleep disturbances (565). Patients with organic or psychiatric disorders, past neurological disorders, past abuse of drugs or alcohol, or heavy smoking habits were excluded. The patients (36 females, 70.5 ± 13.1 years old) regularly taking anxiolytic benzodiazepines in low doses received 3mg of melatonin (Melatol®, Elisium S.A., Buenos Aires, Argentina) or placebo 30 minutes before anticipated sleep time for six weeks. Overall quality of morning freshness, daily alertness, sleep quality, and sleep onset and offset time were assessed from structured clinical interviews and from logs completed by the patients. The possible correlation of urinary excretion of 6-sulphatoxymelatonin (aMT6s) before starting treatment and outcome of treatment was also examined. On day 14 of treatment, the benzodiazepine dose was reduced by half and on day 28, it was halted. Twenty-one patients were excluded from the analysis for incomplete compliance. There was a general lack of changes in quality of wakefulness or sleep in patients taking melatonin or placebo vs. baseline. During the first two weeks, sleep quality of patients taking melatonin was lower than that of placebo. Melatonin advanced sleep onset by 27.9 ± 11.9 minutes and decreased the variability of sleep onset time (p=0.03). The urinary concentration of aMT6s prior to the study did not correlate with any parameter examined. This study is limited by the lack of description of randomization and blinding.
        • Garfinkel et al. conducted a randomized, double-blind, placebo controlled trial to assess whether the administration of melatonin could facilitate the discontinuation of benzodiazepine therapy in 34 patients with insomnia (568). Patients had expressed willingness to discontinue therapy and were considered healthy. Patients received melatonin (2mg in a controlled-release formulation; Circadin®; Neurim Pharmaceuticals Ltd., Tel Aviv, Israel) or a placebo nightly for six weeks (period 1). Patients were encouraged to reduce their benzodiazepine dosage 50% during week 2 and 75% during weeks 3 and 4, and to discontinue benzodiazepine therapy completely during weeks 5 and 6. In period 2, melatonin was administered (single-blinded) for six weeks to all subjects, and attempts to discontinue benzodiazepine therapy were resumed. Benzodiazepine consumption and subjective sleep-quality scores were reported daily by all patients. All subjects were then allowed to continue melatonin therapy, and follow-up reassessments were performed six months later. Adverse effects were similar between groups. By the end of period 1, 14 of 18 subjects who had received melatonin therapy, and four of 16 in the placebo group, discontinued benzodiazepine therapy (p=0.006). Sleep-quality scores were higher in the melatonin therapy group (p=0.04). All patients finished the first phase of the study, and 30 agreed to take part in the second phase. Six additional subjects in the placebo group discontinued benzodiazepine therapy when given melatonin in period 2. The six-month follow-up assessments revealed that of the 24 patients who discontinued benzodiazepine and received melatonin therapy, 19 maintained good sleep quality. This study is limited by lack of description of randomization and withdrawals.
        • Peles et al. conducted a randomized, double-blind crossover trial to evaluate the effectiveness of melatonin in attenuating sleep difficulties during benzodiazepine withdrawal in 80 patients (569). All patients who were admitted to the clinic between July 1993 and July 2004 were eligible for inclusion. Patients received melatonin (5mg daily) or placebo with the following schedule: six weeks in one arm, a one-week washout, six weeks in other arm. Outcome measures included urine benzodiazepines, self-reported Pittsburgh Sleep Quality Index (PSQI), and the Center for Epidemiologic Studies Depression (CES-D) questionnaires administered at baseline, and at six, seven, and 13 weeks. Adverse effects were not discussed. PSQI scores were significantly lower (indicating better sleep quality) in the 22 patients who discontinued benzodiazepines (8.9 ± 0.9) than in 39 with urine benzodiazepines (11.2 ± 0.7, p=0.04). Sleep quality in patients who continued abusing benzodiazepines improved slightly but significantly more in the "melatonin first" group than in the "placebo first" group, with no differences in sleep quality improvement in patients who stopped benzodiazepines. Group differences in phase 2 of the study were lacking. An effect on benzodiazepines withdrawal was lacking. Limitations include lack of description of randomization and blinding. Reasons for withdrawals were lacking.
        • Select studies of lesser methodological strength (not included in the Evidence Table): Vissers et al. conducted a placebo controlled trial to examine whether administration of melatonin facilitates discontinuation of benzodiazepine therapy in patients with insomnia (533). Long-term users of benzodiazepines were asked by their general practitioner (GP) to participate in a discontinuation program in combination with melatonin or placebo. The intervention and follow-up period lasted one year. During this period participants received four questionnaires about their use of sleeping medication and several health instruments. The urine of all participants was tested for the presence of benzodiazepines, as proof of the discontinuation. The main outcome measures were discontinuation of benzodiazepine use, measured by questionnaires and urine samples at three assessment points. A total of 503 long-term users were selected by the GPs, of whom 38 patients (16 males and 22 females) participated. After one year, 40% had stopped their benzodiazepine use, both in the intervention group on melatonin and in the placebo control group. Comparing stoppers and nonstoppers did not reveal differences in benzodiazepine use, or awareness of problematic use. The authors concluded that these results did not conclusively indicate that melatonin was helpful for the discontinuation of the use of benzodiazepines, but the average dose of benzodiazepines in the group was low. This study is limited by the lack of randomization and double-blinding.

        Cancer treatment

        • Summary: There are several early-phase and controlled human trials of melatonin in patients with various advanced stage malignancies, including brain, breast, colorectal, gastric, liver, lung, pancreatic, and testicular cancers, as well as lymphoma, melanoma, renal cell carcinoma, and soft-tissue sarcomas. Melatonin has been used in combination with somatostatin, retinoids, other pineal hormones, and vitamin D and other nutritional agents (396;402;574;840;1354;1355;1356), in addition to other types of treatment, including radiation therapy (570;851), chemotherapies (such as cisplatin, etoposide, or irinotecan) (351;403;404;405;406;528;528;571;573;575;576;854;864), hormonal treatments (such as tamoxifen) (850;862;863), or immune therapies, such as interferon (509), IL-2 (407;408;510;511;512;513;514;515;516;517;518;519;520;521;522;523;524;525;526;527), or TNF (525;529;530). It has also been used in poorly designed studies (479). Many of these trials have been published by the same research group and have involved giving melatonin orally, intravenously, or injected into muscle. Results have been mixed, with some patients stabilizing and others progressing. Animal studies with melatonin note reduced severity of heart damage from anthracycline drugs (87;88;89;1357;1358) or lung damage from bleomycin (213;1359). There are some promising reported results from the above studies, including small significant improvements in the survival of patients with non-small-cell lung cancer given oral melatonin with chemotherapy (cisplatin and etoposide). A review of the use of antioxidant supplements during breast cancer treatment noted a possible enhancement of tumor response associated with melatonin; however, the included trials were described as underpowered (1360). Furthermore, it has been suggested that the use of antioxidants, such as melatonin, may mitigate some of the side effects of chemotherapy, although research remains scant (83). In cancer patients, the nocturnal ratio of melatonin/cortisol was reduced (1361). In persons living in the Arctic, melatonin levels were high in the dark winter months, leading to the melatonin hypothesis for cancer prevention (1362). Despite these findings, as well as those from an array of basic, nonclincal studies, which have indicated a possible role of melatonin in the prevention or suppression of cancer (21;922;924;1149;1150;1151;1152;1153;1158;1160;1363;1364;1365;1366;1367), the evidence remains insufficient to recommend the use of melatonin in cancer patients. High-quality follow-up trials are necessary to confirm these preliminary results.
        • Meta-analysis: Wang et al. conducted a systematic review and meta-analysis to assess the effects of melatonin on cancer (N=761) (577). Eight randomized controlled trials were selected for inclusion (571;573;854;864;1368;1369;1370). The researchers used Cochrane library, CNKI, EMBASE, PubMed and Medline to locate trials using melatonin as an adjunct treatment to chemotherapy or radiotherapy. Also, the references of the pooled articles were manually reviewed to identify additional relevant studies. Articles selected for inclusion had to be randomized controlled trials and include patients with a pathology-confirmed tumor. Included articles had to provide information regarding tumor remission, side effects from chemotherapy, or patient survival at one year. Trials with melatonin in combination with nonstandard chemotherapy regimens, pharmacokinetic trials, and animal studies were excluded. In all included studies, participants were administered 20mg melatonin by mouth daily in combination with chemotherapy or radiation. Information regarding the duration of melatonin treatment was lacking. Information on standardization was lacking from the review. A lack of severe adverse events was reported from the trials. Information on toxic effects, dropouts, and interactions was lacking in the review. The primary outcome measures included the relative risk of cancer remission, radiochemotherapy side effects (including thrombocytopenia, neurotoxicity, and fatigue), and one year survival rate. Based on the meta-analysis, melatonin had a significant effect tumor remission (RR 1.95, 95% CI 1.49 to 2.54; Z=4.90, p<0.00001) and the one year survival rate (RR 1.90; 95% CI 1.28 to 2.83; Z=3.81, p=0.001). Furthermore, melatonin decreased side effects related to radiochemotherapy, including thrombocytopenia (RR 0.13; 95% CI, 0.06 to 0.28; Z=5.26, p<0.00001), fatigue (RR 0.37; 95% CI 0.28 to 0.48; Z=7.18, p<0.00001), and neurotoxicity (RR 0.19; 95% CI 0.09 to 0.40; Z=4.38, p<0.0001). The authors concluded that, as an adjuvant cancer therapy, melatonin significantly improves one year survival rate and cancer remission. Strengths of the review are inclusion of only randomized controlled trials and the detailed treatment outcome data provided for all trials. Limitations of this review included lack of information regarding dropouts, interactions, or toxic effects.
        • Mills et al. conducted a systematic review to examine the effects of melatonin in solid tumor cancer patients and its effect on survival at one year (859). Ten electronic databases from inception to October 2004 (unspecified) were covered. Trials using melatonin as either sole treatment or as adjunct treatment were included. Prespecified criteria were used as a guide for assessment of trial quality. A meta-analysis using a random effects model was conducted. Ten RCTs published between 1992 and 2003 with a total of 643 patients were included. All trials included solid tumor cancers. All trials were conducted at the same hospital network and were unblinded. Melatonin reduced the risk of death at one year (relative risk: 0.66, 95% CI, 0.59-0.73, I2=0%, heterogeneity p≤0.56). Effects were consistent across melatonin dose and type of cancer. Severe adverse events were lacking. The authors concluded that the reduction in risk of death, low adverse events reported, and low costs related to this intervention suggest great potential for melatonin in treating cancer.
        • Systematic reviews: Block et al. conducted a systematic review to examine the literature in order to compile results from randomized trials that evaluate concurrent use of antioxidants with chemotherapy (838). MEDLINE, Cochrane, CINAHL, AMED, AltHealthWatch, and Embase databases were searched. Only randomized, controlled clinical trials that reported survival or tumor response were included in the final tally. The literature searches were performed in duplicate following a standardized protocol. Of 845 articles considered, 19 trials met the inclusion criteria. Antioxidants evaluated were glutathione (7), melatonin (4), vitamin A (2), an antioxidant mixture (2), vitamin C (1), N-acetylcysteine (1), vitamin E (1), and ellagic acid (1). Subjects of most studies had advanced or relapsed disease. The authors concluded that none of the trials reported evidence of significant decreases in efficacy from antioxidant supplementation during chemotherapy. Many of the studies indicated that antioxidant supplementation resulted in either increased survival times, increased tumor responses, or both, as well as fewer toxicities than controls; however, lack of adequate statistical power was a consistent limitation. Large, well-designed studies of antioxidant supplementation concurrent with chemotherapy are warranted.
        • Ernst et al. conducted a systematic review to evaluate and critically analyze all randomized clinical trials (RCTs) of "'alternative cancer cures" (ACCs) for breast cancer (1371). The electronic databases Cochrane Central Register of Controlled Trials, MEDLINE, Embase, Allied and Complementary Medicine, Scirus, BIOSIS, CancerLit, and CINAHL were searched; for ongoing trials, the MetaRegister at http://www.controlled-trials.com/ and the National Research Register at http://www.update-software.com/national/ were searched from their inception. Bibliographies of located studies were scanned. Unpublished or ongoing trials were identified through correspondence with experts in the field. The authors' personal files were hand-searched for further RCTs. Review methods included a systematic review of RCTs involving breast cancer patients treated with ACCs, survival, parameters indicative of tumor burden, disease progression, cancer recurrence, and cancer cure. Thirteen RCTs met the inclusion criteria. In most cases, their methodological quality was low, with only two RCTs scoring 4, and four RCTs scoring 3 out of 5 possible points for methodological quality. The treatments tested included various methods of psychosocial support, such as group support therapy, cognitive behavioral therapy cognitive existential group therapy, a combination of muscle relaxation training and guided imagery, the Chinese herbal remedy Shi Quan Da Bu Tang, thymus extract, transfer factor, melatonin, and factor AF2. For melatonin, results were encouraging but not fully convincing. Further study is required.

        Cardiovascular disease

        • Summary: In a poor quality study, the inclusion of melatonin in the combined treatment of cardiovascular disease resulted in anti-ischemic, antianginal, antioxidant, and hypotensive effects (369). Further research is needed.

        Chronic fatigue syndrome

        • Summary: Nonrandomized clinical trials have been conducted to examine the effects of melatonin on chronic fatigue syndrome with possible benefits (578). A lack of effect of melatonin was found in one randomized study (579). More well-designed trials are needed before a conclusion can be made.
        • Evidence: Williams et al. conducted a randomized, placebo controlled crossover study to evaluate the effects of melatonin or bright-light phototherapy on chronic fatigue syndrome (CFS) symptoms (N=42) (579). Included participants were diagnosed with CFS by the Oxford criteria. Excluded participants were reluctant or unable to meet the experimental protocol demands. Of individuals who completed the trial, the average age was 44.5 years of age and the average length of CFS was 3.6 years. Following a 12-week run-in period, participants were randomly assigned to receive 5mg melatonin or phototherapy daily. Melatonin was taken by mouth daily 2.5 hours before bedtime for 12 weeks. Phototherapy was administered for one hour each morning 30 minutes after the average wakeup time (which had been assessed during the previous four weeks) using a 2500Lux lightbox. Following a 12-week washout (placebo) period, participants were crossed over to the other treatment group. After the participants completed 12 weeks of treatment with the second intervention, they underwent a second 12-week washout (placebo) period. Information on standardization was lacking from the study. A lack of treatment-related adverse effects and toxic effects were observed during the study. Ten participants dropped out due to time or social demands, and two participants dropped out due to an employment change. Information on interactions was lacking from the study. The primary outcomes were quality of life and CFS symptoms after treatment. Researchers used a visual analog scale (VAS) to assess CFS symptoms, a Shortform (SF-36) Health Survey to assess quality of life, a mental fatigue inventory (MFI), and Hospital Anxiety and Depression Scale (HAD). Secondary outcomes included treatment effect on body temperature circadian rhythm and melatonin secretion. Compared to baseline, treatment with melatonin lacked a significant effect on fatigue, depression, anxiety, sleep disturbance, feeling refreshed when waking, energy, concentration, and muscle pain. Compared to baseline, participants treated with phototherapy showed a significant improvement in sleep disturbance (6.6 vs. 5.1, p=0.03), but lacked significant improvement in other outcome measures based on the visual analogue scale. Compared to baseline, participants treated with melatonin showed a significant worsening of bodily pain (31 vs. 41, p=0.044), although increased vitality (15 vs. 20, p=0.016) and improved mental health (56 vs. 56, p=0.046) were observed. Participants treated with phototherapy lacked significant change in any outcome measure associated with the SF-46 Health Survey vs. baseline. Both treatment groups lacked significant change in MFI score or HAD score. Treatment with phototherapy resulted in a significant change in temperature rhythm acrophase (1.63 ± 0.3 hours, p<0.05), but a significant change in amplitude of temperature rhythms was lacking for both treatment groups. A significant correlation between the timing of DLMO and the temperature acrophase was lacking for both groups after treatment. The authors concluded that melatonin lacks a significant benefit on CFS symptoms or quality of life in CFS individuals. Limitations to the study include the lack of double blinding, the lack of description of randomization methods, and the high dropout rate.

        Chronic obstructive pulmonary disease

        • Summary: In a clinical trial and compared with baseline, melatonin decreased dyspnea and reduced the exhalation of 8-isoprostane (580). Changes in lung function tests were lacking. Further study is needed.
        • Evidence: de Matos et al. conducted a randomized, double-blind, placebo controlled study in subjects with chronic obstructive pulmonary disease (COPD) to determine the effectiveness of oral melatonin on oxidative stress in the lungs (N=36) (580). Male and female subjects with moderate to very severe COPD (stages II to IV) were selected consecutively as they attended the clinic. Exclusion criteria included exacerbation of COPD in the last four weeks (with or without administration of systemic steroids), hospitalization within eight weeks of the trial, any serious co-morbidities, or abuse of alcohol or drugs. Prior to randomization, all participants completed a two-week run-in period, during which time participants stopped using inhaled steroids or lung medications except for bronchodilators (long-acting) or beta2-antagonist (short-acting). Participants were randomized to receive capsules containing 3mg melatonin or placebo orally each evening two hours before bed for three months. Information on standardization was lacking. Significant adverse effects were reported to be lacking. However, two participants in the melatonin group and two participants in the placebo group reported numbness, mild headache, and dyspnea worsening. Information regarding toxic effects was lacking. Information regarding dropouts and interactions was lacking. The main outcome measure was the level of 8-isoprostane in exhaled breath condensate. Secondary outcome measures included dyspnea severity using the Medical Research Council (MRC) scale, exercise capacity using the 6-minute walk test (6-MWT), lung spirometry, and exhaled breath condensate interleukin-8 (IL-8) concentration. Compared to baseline, participants administered melatonin showed a statistically significant decrease of 8-isoprostane levels with pair wise comparisons after two months (7.729±2.223, p=0.03) and three months (7.708±1.868, p=0.01), but statistically significant change in the placebo group was lacking. There was a statistically significant improvement in the mean MRC dyspnea score (p=0.01) after melatonin treatment; this effect was lacking for the placebo group. Statistically significant changes in 6-MWT or lung function tests were lacking for both groups. There was a statistically significant increase in exhaled breath condensate of IL-8 (1.98±0.73 to 2.41±0.82, p=0.03) in the placebo group only. According to the authors, melatonin (3mg daily) reduced oxidative stress in the lungs of subjects with COPD and induced improvement in dyspnea. The method was double blind with a placebo control, but description of the method of randomization was lacking.

        Circadian rhythm sleep disorders (visually impaired and nonimpaired individuals)

        • Summary: In visually impaired individuals, light and dark stimuli are not received by the eye to trigger melatonin release and the onset of sleep. In these patients, natural melatonin levels peak at a different hour every night to the point where individuals may sleep during the day and awake at night. This is commonly referred to as "free-running" circadian rhythm. Khan et al. conducted a systematic review to determine the effect of melatonin for nonrespiratory sleep disorders in visually impaired children; however, studies meeting the inclusion criteria were lacking and therefore conclusions cannot be drawn (753). In a separate systematic review, melatonin improved sleep latency and time in individuals with visual impairment in poorly designed studies (581). This topic was also investigated in a review (1372). There are multiple published small case series and case reports in the literature, yet limited controlled trials in this population, as well as in some sighted individuals (414;1338;1340;1341;1373;1374;1375;1376;1377;1378;1379;1380;1381;1382;1383;1384;1385) to date. At present, studies and individual cases suggest that melatonin, administered in the evening, may correct circadian rhythm. Large, well-designed controlled trials are needed before a conclusion can be made.
        • Systematic review: Khan et al. conducted a systematic review of two studies (724;1292) to assess the effects of melatonin for treating sleep disorders in children with visual impairment (581). Relevant studies published in any language through August 2010 were pooled from PubMed, EMBASE, Science Citation Index Expanded, CINHAL, and the Cochrane Central Register of Controlled Trials (CENTRAL). The references of the pooled studies were manually reviewed, and citation-tracking software was used to identify additional relevant studies. All included studies were randomized or semi-randomized clinical trials assessing the effects of treatment in participants aged three months to 26 years. Participants were required to have both a sleep disorder and a visual impairment. A sleep disorder could include difficulties with sleep initiation, maintenance, or scheduling. Visual impairment involved a loss of vision from damage to the eye or ocular conditions. Children and adolescents included in the trials were administered 3-12mg melatonin daily for eight weeks. Adults included in the studies were administered 5mg melatonin daily for eight weeks. Information regarding standardization was lacking. One study reported a lack of adverse effects. Information regarding adverse effects was lacking from the remaining study. Information regarding toxic effects, dropouts, and interactions was lacking. Primary outcome measures included changes in sleep latency and night waking. Secondary outcome measures included mood during the day and adverse effects. According to one study, treatment with melatonin resulted in a significant improvement in sleep latency (p=0.019). Mean sleep latency was decreased to 0.3 hours with melatonin treatment, which was lower that at baseline (1.6 hours) and after placebo treatment (0.7 hours). Response to treatment was observed in 29.2% of participants administered 3mg melatonin daily, 45.8% of individuals administered 6mg, 20.8% of participants administered 9mg, and 4.2% of participants administered 12mg melatonin daily. In the other study, 18 of 20 children that were treated with melatonin experienced a significant decrease in sleep latency when compared to placebo or baseline (p<0.05). At baseline the mean sleep latency was 1.2 hours. After treatment with placebo it remained at 1.2 hours, while after treatment with melatonin it was 0.7 hours. The total sleep time improved from 7.1 hours at baseline to 8.1 hours with melatonin (p=0.007). A difference in mean number of night awakenings was lacking. Limitations of this review included the fact that one included study lacked information regarding what constituted a "response" In addition, both studies lacked clear information regarding whether the p-value corresponded to comparisons with baseline or with the control group. Finally, all of the improvements were subjective in nature, and because children were involved, these improvements were often assessed by the parents rather than the participant.
        • Evidence: Paul et al. conducted a randomized, double-blind, placebo controlled, crossover trial to assess the effect of melatonin, light treatment, or combined melatonin and light treatment, on circadian phase advance (N=11) (582). Healthy individuals who lacked underlying illness were included in this study. Individuals who smoked, who took medications that could influence the results (including beta-blockers), or those individuals for whom treatment with melatonin was contraindicated were excluded from the study. Participants were randomized to receive 3mg sustained-release melatonin, placebo supplements (at 1600 hours), light treatment (from 0700-0800 hours), or 3g melatonin at 1600 hours combined with light treatment from 0600 to 0700 hours the following morning. Subjects in each group were treated once during each four-day phase of the study. Following a three-day washout period, participants were crossed over to other treatment groups until all participants had completed each treatment group. The sustained-release melatonin supplements had an eight-hour release profile. The light treatment device emitted narrow bandwidth green light (wavelength 500nm) from two 45cm towers that were situated 45cm apart. Information regarding allergies, adverse effects, toxic effects, dropouts, and interactions was lacking. The main outcome was change in circadian phase shifts. Compared to baseline, treatment with melatonin alone resulted in a statistically significant phase shift of 0.72±0.11 hours (p<0.01). Participants administered placebo or light treatment groups lacked statistically significant change in phase shift. For the combined melatonin and light treatment group, a statistically significant 1.04±0.17 hour advance was noted (p<0.01). A statistically significant main effect was noted between conditions (F(2.71, 27.13)=9.82, p<0.0002). A post-hoc analysis indicated that the phase shifts from the melatonin-only group (p<0.005), and the combined melatonin and light group (p<0.0002) were significantly larger compared to the placebo group. The phase advance from the combined melatonin and light group was larger (p<0.015) than the advance from light treatment only group. Limitations included a lack of description of randomization, inappropriate blinding, and lack of information regarding adverse events, reasons for dropouts, or interactions.
        • Studies of lesser methodological strength (not included in the Evidence Table): Fischer et al. conducted a double-blind crossover trial to examine whether a single administration of melatonin improves sleep and associated neuroendocrine patterns in 12 blind individuals (478). All subjects were totally blind and were in good health, based on a physical examination. The patients were not on any medications. All reported insomnia and daytime sleepiness. Twelve totally blind subjects received 5mg of melatonin and placebo orally one hour before bedtime starting at 11 p.m. A discussion of adverse effects was lacking. Endpoints included sleep quality, including time and efficiency, as well as levels of plasma hormones involved in sleep. Melatonin increased total sleep time and sleep efficiency (p<0.05) and reduced time awake (p<0.05). The increment in total sleep time was primarily due to an increase in stage 2 sleep (p<0.01) and a slight increase in rapid eye movement sleep (p<0.06). Melatonin normalized in parallel the temporal pattern of adrenocorticotropic hormone (ACTH) and cortisol plasma concentration. Placebo lacked an effect on the difference in ACTH and cortisol levels between early and late sleep. However, melatonin induced the typical suppression of pituitary-adrenal activity during early sleep and induced a rise during late sleep (p<0.01, respectively). Melatonin also decreased cortisol nadir values (p<0.05). It was concluded that melatonin could improve sleep function in blind persons by synchronizing in time the inhibition of pituitary-adrenal activity with central nervous sleep processes. Limitations to this study include the lack of randomization and description of blinding and withdrawals.
        • Hack et al. conducted a placebo controlled, single-blind trial to examine the entraining effects of melatonin on the cortisol rhythm and its acute effects on subjective sleep in 10 blind subjects (480). Subjects (nine adult males) had free-running 6-sulphatoxymelatonin (aMT6s) rhythms (circadian period [tau]: 24.23-24.95 hours). Inclusion and exclusion criteria were lacking. The subjects lacked a conscious light perception (NPL). Subjects received 0.5mg of melatonin or placebo daily at 9 p.m. (treatment duration: 26-81 days, depending on individuals' circadian period). Subjective sleep was assessed from daily sleep and nap diaries. Urinary cortisol and aMT6s were assessed for 24-48 hours weekly and measured by radioimmunoassay. While on melatonin, six of the 10 subjects had an entrained cortisol period. In four of these, the period was indistinguishable from 24 hours and in the other two subjects, the rhythm continued to free run for up to 25 days before entrainment. One subject exhibited a shortened cortisol period throughout melatonin treatment. Rhythms returned to prestudy levels after melatonin was ceased. Melatonin resulted in a significant increase in nighttime sleep duration and a significant reduction in the number and duration of daytime naps (p<0.001). The authors suggested that a daily dose of 0.5mg of melatonin was effective at entraining the free-running circadian systems in most of the blind subjects studied. This study is limited by the lack of randomization and double-blinding.

        Delirium

        • Summary: de Jonghe et al. published a protocol for a randomized controlled trial investigating the effects of melatonin on delirium in hip fracture patients (1386). In elderly individuals, melatonin was found to reduce delirium (442). Further study is needed in order for conclusions to be drawn.
        • Evidence: Al-Aama et al. conducted a randomized controlled trial to assess the effect of melatonin on delirium pathogensis (N=145) (442). Inclusion criteria were patients aged 65 years or older who were admitted from the emergency department to in-patient services of the internal medicine department. Exclusion criteria included need for an extended stay, patients unable to communicate in English, patients that were unable to take oral medicines, patients with a life expectancy of less than two days, or those who experienced seizures, intracranial bleeding, an allergy to the specified treatment, or an international normalized ratio (INR) of less than one or more than four with warfarin treatment. Participants were randomized to receive 0.5mg melatonin or placebo every evening for up to 14 days. A description of standardization was lacking. One participant experienced nightmares, and another participant reported the sensation of ''floating around and talking to his dead wife." Side effects reported may have been directly related to the treatment or secondary to delirium. A description of toxic effects was lacking. Two subjects in each group withdrew consent. In addition, nine patients from the treatment group and ten from the placebo group were excluded from final analysis due to inadequate data. Reasons for inadequate data included absenteeism during the assessment period, change in initial condition resulting in aggressiveness, or inability to communicate due to a change in consciousness, discharge, or death. A description of interactions was lacking. The main outcome was delirium, which was defined according to the Confusion Assessment Method (CAM). Secondary outcomes were delirium severity (based on the Memorial Delirium Assessment Scale (MDAS)), use of sedatives, psychotropic medicines, or restraints, timeframe of the stay, sleep/wake disturbance, and mortality. Compared to the placebo group, a lower percentage of participants who received melatonin experienced delirium vs. placebo (12% vs. 31%, p=0.014). The odds ratio for delirium was 0.19 (95% CI 0.06 to 0.62). All other outcomes lacked a statistically significant difference between treatment and placebo groups. Limitations included an incomplete description of double blinding and lack of intent to treat analysis.

        Depression

        • Summary: Melatonin has been suggested as playing a role in and serving as a possible treatment for depression (66;1387), a hypothesis which has garnered some support in animal research (1388;1389). Also, in human research, higher baseline urinary levels of 6-sulfatoxymelatonin was found to be a predictor of clinical outcome in depressive patients (1390). However, human research remains scant and inconclusive and one well designed study suggests that effects are lacking (592). Further research is required.
        • Evidence: Serfaty et al. conducted a randomized, double-blind, placebo controlled trial to determine if exogenous melatonin may act as a sleep promoter and antidepressant in 33 patients with major depressive disorder (MDD) (592). Patients aged 18-65 years were included in the trial if they were diagnosed with major depressive disorder (MDD) and experienced early-morning wakings. Patients were excluded if they had delusions, hallucinations, suicidal ideation, or a recent change in psychotropic medication within the last four weeks. The subjects were administered melatonin (6mg of slow-release at bedtime) or placebo for four weeks in addition to any pre-existing treatment for depression. Actigraphy, sleep diaries, and the Leeds Sleep Evaluation Questionnaire were used as measures of sleep quality. Two subjective questionnaires, the Beck Depression Inventory (BDI) and the 21-item Hamilton Depression Rating Scale (HDRS), were used to measure depression. Of the 33 participants, 31 completed the trial; two dropped out immediately because they felt better. Improvements were noted with time in terms of sleep diary data, HDRS scores, and LSEQ scores, but changes associated with melatonin use were lacking. Significantly different BDI scores were lacking between melatonin and placebo (2.03, 95% CI, -6.30 to 2.25). Adverse effects were similar between the two groups. The study was conducted over a three year, three-month period, with different patients undergoing the four-week randomization throughout this time. The trial stated that 17 people would be required in each group (for a total of 34) in order to power the trial adequately; however, this standard was not met. It should also be noted that participants were allowed to continue any current antidepressive medication, whose effects on these findings are difficult to quantify. This study was well conducted.

        Diabetes (adjunct therapy)

        • Summary: Melatonin in combination with zinc has been found to improve postprandial glycemic control in patients with type 2 diabetes (386;387). However, in other research, melatonin supplementation was found to lack a significant effect upon measures of glucose homeostasis (763). In a systematic review of nutritional supplements in individuals with type 2 diabetes (1391), one included clinical trial assessed the effects of melatonin on antioxidant status, but not endocrine health or clinical endpoints (30). More well-designed trials, using melatonin as a monotherapy, are needed before a conclusion can be made.

        Exercise performance

        • Summary: Daytime administration of melatonin lacked effects on maximal jumping ability and maximal strength (508). More well-designed trials are needed before a conclusion can be made.
        • Evidence: Mero et al. conducted a randomized, double-blind trial to investigate the effects of a heavy resistance exercise session (RES) with the oral daytime ingestion of melatonin on the physiological responses and acute performance in 10 healthy males (508). The men were resistance trained and not using medication or sport supplement. Subjects undertook an 80-minute intensive hypertrophic RES for major muscles of the lower and upper extremities. The subjects were studied on two occasions, receiving melatonin (6mg) or placebo (6mg) in random order 60 minutes before each RES. Blood samples were taken from an antecubital vein, both in fasting conditions in the morning and before RES (60 minutes and 0 minutes before), during RES (middle) and after RES (0, 15, 30, and 60 minutes after). Maximal jumping ability and maximal strength in bench press and squat were measured before and immediately after RES. Melatonin lacked an effect on maximal jumping ability of strength. Serum melatonin concentration increased significantly (p<0.05-0.001) in the melatonin group at all time points. The concentration reached a peak value of 1,171.3 ± 235.2pg/mL in 60 minutes at pre 0. Serum melatonin also increased slightly but significantly (p<0.05) in the placebo group just before RES, in the middle of RES, and after RES (0 and 15 minutes after). There were large differences (p<0.01-0.001) in the serum melatonin concentration between the groups at all time points. There were no differences in the growth hormone (GH), testosterone, and cortisol peak concentrations at any time points between the groups, but the area under the curve of GH during RES (p<0.01) and 60 minutes after RES (p<0.05) was higher in the placebo condition. Discussion of adverse effects was lacking. The authors concluded that oral ingestion of melatonin (6mg) during daytime with heavy-resistance exercise may slightly decrease GH concentrations; however, melatonin administration during daytime lacked acute (1-2 hours) effects, either on the maximal jumping ability or on the maximal strength. This study is limited by the lack of description of randomization, blinding, and withdrawals.
        • Studies of lesser methodological strength (not included in the Evidence Table): Atkinson et al. conducted a controlled trial of 12 subjects to examine the effects of melatonin on the subjective quality of sleep in athletes and residual effects on physical performance the morning after treatment (593). Exclusion and inclusion criteria were lacking; however, the subjects included were described as being healthy, physically active sports science undergraduates, ages 19-30 years. The subjects were given 5mg of melatonin or a placebo; a single dose of one capsule was taken at 11 p.m., 30 minutes before retiring. A discussion on standardization, allergies and adverse effects, toxic effects, dropouts, and interactions was lacking. A specific statement of primary outcome was lacking, although subjective measure of sleep quality (maintenance and latency) is assumed. Secondary measures included objectively measured intra-aural temperature, left and right grip strength, and time taken to complete a 4km stationary bike ride. Secondary subjective measures include perceived exertion. ANOVA was used to analyze grip strength and cycle time data. There was little difference between the means of the placebo vs. melatonin groups in right [t11=-0.15(p=0.88)] or left [t11=0.35(p=0.74)] hand grip strength, or cycle times [t11=0.44(p=0.67)]. A difference in sleep latency [W=25.5 (p=0.33)] and maintenance [W=12.5 (p=0.87)] was lacking between the melatonin and placebo group median values (Wilcoxon tests, p>0.5). Limitations of this study include a lack of randomization, inclusion and exclusion criteria, and adequate description of the blinding process, as well as the short study duration and overall suboptimal study design.

        Fertility

        • Summary: In patients undergoing in vitro fertilization embryo transfer (IVF-ET), although melatonin benefited oocyte maturation, effects on fertilization and pregnancy were lacking (594;898). A combination of myo-inositol and melatonin improved both oocyte quality and fertilization rates in women who failed to conceive in previous IVF cycles (1392). Further well-designed research is needed.
        • Evidence: Batioglu et al. conducted a randomized controlled trial to evaluate melatonin's effect on oocyte quality in women undergoing in vitro fertilization (IVF) (N=85) (595). Participants were women aged 20-40 years who were undergoing IVF, had regular menstrual cycles, and lacked hormonal or nonhormonal therapy in the previous three months. Individuals with serious endometriosis, follicle-stimulating hormone (FSH) levels >13, hypogonatropic hypogonadism, or azoospermia were excluded from the study. Beginning at mid-luteal phase, all participants were down-regulated with gonadotropin-releasing hormone (GnRH) agonist. Participants randomized to the treatment group also began receiving 3mg melatonin daily starting on the day of GnRH administration. Participants in the control group lacked additional supplementation. When appropriate down-regulation had been achieved, a starting dose of 150 IU recombinant FSH (rec-FSH) was administered daily to stimulate; rec-FSH dosing was adjusted based on the response of the participant. When the average follicle diameter was at least 18mm, recombinant human chorionic gonadotropin (rec-hCG) was administered. Thirty-six hours after rec-hCH was administered, the oocytes were retrieved. Information regarding the duration of treatment with melatonin was unclear. Information on standardization, allergies, adverse effects, toxic effects, dropouts, and interactions was lacking from the study. The primary outcome was the number of morphologically mature (MII) oocytes retrieved. Secondary outcomes included embryo quality, rate of fertilization, and the rate of pregnancy. Between-group differences regarding the number of retrieved oocytes and the number of retrieved mature MII oocytes were lacking. However, compared to the placebo group, participants administered melatonin showed a statistically significant increase in the percentage of mature oocytes retrieved (75.8 vs. 81.9%, respectively, p<0.05). A statistically significant between-group difference in fertilization rate was lacking. However, participants treated with melatonin had a higher number of top-quality embryos (3.28 vs. 2.53, p=0.035) and a higher percentage of top-quality embryos (30.4% vs. 33.7%, p=0.004). A statistically significant between-group difference in the rate of pregnancies was lacking. This study was limited by lack of blinding and lack of information on adverse effects.
        • Eryilmaz et al. conducted a randomized controlled trial to assess the effect of melatonin on the in vitro fertilization embryo transfer (IVF-ET) outcomes in subjects with sleep disturbance (N=63) (594). Women scheduled to undergo IVF-ET were included if a psychologist had determined that they had a disturbed sleep pattern. All participants had been diagnosed with infertility that was unexplained based on the World Health Organization criteria. Subjects were excluded if they had any of the following: chronic drug use or a habit of smoking, history of total fertilization failure cycles that occurred more than once, high blood pressure, diabetes mellitus, uterine myoma, or ovarian cyst. All participants underwent IVF-ET. The test participants were administered oral tablets containing 3mg melatonin (Przedsiebiorstwo Farmaceutyczne, Zakroczym, Polland), which were taken between 10-11pm from the third to the fifth day of the menstrual cycle until the day when human chorionic gonadotropin (hCG) was injected according to a standard gonadotropin-releasing hormone (GnRH) agonist regimen. Information on standardization was lacking. Reference to adverse or toxic effects was lacking. There were three dropouts because of failure to take the melatonin as instructed or cancellation of the IVF procedure. Information regarding interactions was lacking. Oocyte quality was the primary outcome (grades I, II, or III, and metaphase II (MII) oocytes). Other comparisons were sleeping status of the participant, the number of follicles with a diameter equal to 16mm, the number of oocytes retrieved, and the rates of fertilization, implantation, and clinical pregnancy. After treatment, the following measures of oocyte quality were significantly higher in the melatonin group vs. the control group: mean MII oocyte count (9.0±5.6 vs. 4.4±3.3, respectively, p=0.0001), MII oocyte ratio (79.6% vs. 62.3%, respectively, p=0.0001), and grade I embryo ratio (69.3% vs. 44.8%, respectively, p<0.05). The mean number of retrieved oocytes in the melatonin group was also higher vs. placebo (11.5±6.3 vs. 6.9±3.8, respectively, p=0.0001). However, fertilization, implantation, and pregnancy rates were similar for both groups, and a statistically significant between-group difference in sleeping status improvement was lacking. According to the authors, treatment with melatonin resulted in an improvement in oocyte quality and oocyte retrieval after IVF. Limitations of this study included the lack of placebo control.
        • Rizzo et al. conducted a prospective, randomized controlled trial to assess the effect of melatonin on oocyte quality in women receiving fertility treatments (N=65) (898). Women aged 35-42 years who were undergoing IVF due to infertility factors (low oocyte quality) were included in the study. All participants received 2g myo-inositol twice daily plus 200mg folic acid daily. Participants in the control group lacked additional treatment. Participants in the treatment group also received 3mg melatonin daily. Melatonin and control treatments began on the first day of a gonadotropin-releasing hormone (GnRH) agonist administration (Decapeptyl; Ipsen, Milan, Italy), which was started at mid-luteal phase. Once the participants had been optimally down-regulated with GnRH agonist, they were stimulated with recombinant follicle stimulating hormone (FSH; starting dose 150 IU/die). When the follicle diameters reached 18mm, participants were administered human chorionic gonadotropin (hCG). The oocytes were retrieved for analysis 36 hours after hCG injection. Information regarding standardization, allergies, adverse effects, toxic effects, dropouts, and interactions was lacking. Outcome measures included meiotic maturation status of the retrieved oocytes, quality of embryos, and rate of pregnancy. Secondary outcomes included total quantity of oocytes and fertilization rate (oocytes retrieved and cleavage rate). There was a lack of a statistically significant, between-groups difference in quantity of oocytes, clinical pregnancy rate, abortion rate, implantation rate, and fertilization rates. However, more mature oocytes (mean number: 6.56 ± 1.64 vs 5.76 ± 1.56; p=0.047) and less immature oocytes (mean number: 1.31 ± 0.74 vs. 1.91 ± 0.68; p=0.001) were noted in the melatonin group vs. the control group. The mean number of high quality embryos (classes 1 and 2) occurred in the melatonin group vs. the control group (1.69 ± 0.64 vs. 1.24 ± 0.75; p=0.01). There were thirteen pregnancies in the melatonin co-treatment group compared to nine in the other group; this difference, however, lacked statistical significance (p=0.26). Limitations included a lack of double-blinding, withdrawals, and intent to treat analysis.

        Fibromyalgia

        • Summary; Melatonin has been found to improve symptoms of fibromyalgia in preliminary studies (596;597). Further research is needed.
        • Evidence: Hussain et al. conducted a randomized, double-blind, equivalence study to compare the effects of melatonin, fluoxetine, or a combination of melatonin and fluoxetine on fibromyalgia syndrome (N=101) (597). Individuals aged 18-65 years who had been diagnosed with primary fibromyalgia (based on the American College of Rheumatology criteria) were included in the study if they lacked other pathological conditions that could influence treatment outcome. All included participants lacked treatment with analgesics, anti-inflammatory agents, aspirin, or antioxidant therapy. Participants were randomized to receive capsules containing 20mg fluoxetine (Cipla, Mumbai, India) plus placebo formula (Group A), capsules containing 5mg melatonin (Rupal Chemicals Ltd., Tarapur, India) plus placebo formula (Group B), capsules containing 20mg fluoxetine plus capsules containing 3mg melatonin (Group C), or capsules containing 20mg fluoxetine plus capsules containing 5mg melatonin (Group D). Treatments were administered daily for 60 days. Capsules of fluoxetine were administered in the morning, while melatonin was administered at night. Information regarding standardization, allergies, adverse effects, toxic effects, dropouts, and interactions was lacking. Outcome measures included changes in clinical symptoms based on the Fibromyalgia Impact Questionnaire (FIQ), as well as health related quality of life parameters. Compared to baseline, participants in Group A showed a significant reduction in total FIQ score (21.5%, p<0.05), FIQ pain subscore (14.3%, p<0.001), FIQ stiffness subscore (23.5%, p<0.05), and FIQ depression subscore (24.5%, p<0.05), as well as statistically significant decreases in physical impairment score (16.7%, p<0.001), feeling good score (23.7%, p<0.05), work missed (32.9%, p<0.001), and ability to do work (20.4%, p<0.05). Participants in Group B showed statistically significant reductions in FIQ total score (18.9%, p<0.05), FIQ pain subscore (27%, p<0.001), FIQ fatigue subscore (23.7%, p<0.05), FIQ rest/sleep subscore (31.3%, p<0.001), FIQ stiffness subscore (23.0%, p<0.05), and FIQ depression subscore (23.3%, p<0.05), as well as statistically significant decrease in feeling good (9.6%, p<0.001), work missed (10.7%, p<0.05), and ability to do work (15.3%, p<0.05). Compared to baseline, participants in Group C showed statistically significant reduction in total FIQ score (28.8%, p<0.001), FIQ pain subscore (27.3%, p<0.001), FIQ fatigue subscore (20.3%, p<0.05), FIQ rest/sleep subscore (36%, p<0.001), FIQ stiffness subscore (37.2%, p<0.001), FIQ anxiety subscore (47.3%, p<0.001), and FIQ depression subscore (39.6%, p<0.001), as well as statistically significant decrease in physical impairment (23.7%, p<0.001), feeling good (24.6%, p<0.001), work missed (29%, p<0.001), and ability to do work (21%, p<0.001). Compared to baseline, participants in Group D showed a statistically significant reduction in FIQ score (28.9%, p<0.001), FIQ pain subscore (30%, p<0.001), FIQ fatigue subscore (34.7%, p<0.05), FIQ rest/sleep subscore (41.3%, p<0.001), FIQ stiffness subscore (25%, p<0.05), FIQ anxiety subscore (21.6%, p<0.001), and FIQ depression subscore (42.3%, p<0.001), as well as a statistically significant decrease in physical impairment (37%, p<0.001), feeling good (25%, p<0.001), work missed (23.8%, p<0.05), and ability to do work (23%, p<0.001). The authors concluded that melatonin has beneficial effects on fibromyalgia when used alone or in combination with other treatments. Limitations of this study included lack of information regarding the methods of randomization or blinding, as well as lack of information regarding adverse effects, dropouts, or interactions.
        • Studies of lesser methodological strength (not included in the Evidence Table): Citera et al. conducted a pilot open-label study to examine the effect of melatonin in patients with fibromyalgia (N=21) (596). Patients were treated with melatonin 3mg at bedtime for four weeks. Adverse events were indicated as being mild and transient but further details are lacking. The study was completed in 19 patients due to withdrawal in two patients (migraine and loss to follow-up). Evaluation included tender point count by palpation, pain score, pain severity (10cm visual analog scale), sleep disturbances, fatigue, depression, anxiety, and patient and physician global assessments (VAS). Urinary 6-sulphatoxymelatonin levels were also measured. Melatonin resulted in a significant improvement in the tender point count, severity of pain, patient and physician global assessments, and VAS for sleep, after 30 days. Statistically significant changes for other endpoints were lacking.

        Gastrointestinal disorders

        • Summary: Preliminary research has indicated that, in patients with functional dyspepsia, treatment with melatonin mitigates or eliminates symptoms (598), a finding possibly corroborated by observed reductions in oxidative damage to cells of the gastric lining (1393). Other research has shown that melatonin supplementation may also be effective as a treatment or adjuvant in gastroesophageal reflux disease (GERD) (602). In patients with Crohn's disease and ulcerative colitis, melatonin resulted in a reduction of inflammation when used in combination with traditional therapies (932). It has also been suggested as a possible therapy for irritable bowel syndrome (933), although findings have been mixed (438;599;600;601;603). Further clinical trials are required before a conclusion can be made.
        • Systematic review: Mozaffari et al. conducted a systematic review to assess the effects of melatonin on irritable bowel syndrome (IBS) or other gastrointestinal conditions (603). In vitro and animal studies were assessed in this review but are excluded from this summary focusing on the clinical effects of melatonin. Nine studies included in the review assessed the effects of melatonin in clinical trials (438;598;599;600;601;646;1394;1395;1396). The reviewers used the databases Cochrane, Google scholar, Pubmed, Scopus, and Web of Science to search for articles published through August 2010 that assessed the effects of melatonin on irritable bowel syndrome. The references of the pooled articles were manually reviewed to identify additional relevant studies. Articles selected for inclusion had to include a specific adult patient population presenting with a particular condition and had to provide specific details regarding the treatment. According to the reviewers, participants were administered 3-10 mg melatonin daily. Information on frequency and duration of treatment, standardization, allergies, adverse effects, toxic effects, dropouts, and interactions was lacking from the review. The primary outcome measure was the disturbance in melatonin concentration and the clinical efficacy of melatonin on IBS and gastrointestinal disorders. Overall, the reviewers found that individuals with IBS had disturbances in endogenous concentration of melatonin. Melatonin supplementation decreased stomach pain and improved the IBS symptom score. In one study, participants with IBS showed improvement in dyspepsia, pain tolerance, psychological factors, and quality of life following treatment with melatonin vs. psychotropic drugs. In another study, participants with IBS and sleep disturbances showed reduced abdominal pain score and improved rectal pain threshold. However, a significant effect on sleep parameters was lacking. In another study, melatonin lacked a significant effect on sleep, anxiety, and depression in individuals with IBS, but a decrease in abdominal pain, abdominal distention, and sensations of defecation in the abdomen were observed vs. placebo. In another study, treatment with melatonin resulted in improvement in IBS symptom score. In some participants with resistant depression, treatment with melatonin resulted in decreased Hamilton Rating Scale for depression, as well as reduced insomnia and fatigue. According to the reviewers, this effect may be beneficial in individuals with IBS as well. In individuals with ulcer-like dyspepsia, treatment with melatonin resulted in improved health, reduced dyspeptic symptoms, and reduced intensity and frequency of nocturnal pain. A final study indicated that treatment with melatonin improved colonic transit time in a control population, but this effect was lacking in IBS patients. The review was limited by the inclusion of studies of lesser methodological strength, a lack of information regarding treatment frequency and duration, and a lack of detailed information regarding study outcomes.
        • Evidence: Lu et al. conducted a randomized, double-blind, placebo controlled, crossover clinical trial to evaluate the efficacy of exogenous melatonin on gut motility in both healthy subjects and those with irritable bowel syndrome (IBS) (601). Subjects were 17 healthy controls and 17 matched IBS patients (all female). Participants were included if they had active IBS symptoms (a minimum period of more than one month prior to the study and whose IBS symptom score indicated at least a moderate severity), normal hematological and biochemical indices, and no abnormalities on barium enema or colonoscopy. Subjects who were pregnant, breastfeeding, or lactose intolerant, or who had gastrointestinal, anal, hepatic, or other systemic disorders, such as diabetes mellitus or previous gastrointestinal surgery (except appendectomy), were excluded. Subjects were treated with 3mg of melatonin or placebo daily for 16 weeks (two eight-week treatment periods reversed following a four-week washout period). All participants were asked to avoid long-distance travel and any drugs known to change gastrointestinal function, during the study and for one month preceding, as well as to follow a normal diet. Colonic transit time (CTT) measurements including Radio-Opaque, Blue Dye, and Bristol Stool Form Score (BSFS) were used to evaluate outcome. A significant increase on mean CTT measured by the radio-opaque method (p=0.04), but not the Blue Dye (p=0.65) or BSFS (p=0.43), was observed in healthy controls compared to baseline CTT. Significant differences in mean CTT were lacking, measured by either the Blue Dye or BSFS methods, in IBS patients after melatonin and placebo treatment. Although significance was lacking in CTT measured by the Blue Dye method in IBS patients (p=0.08), it appeared to be prolonged after melatonin treatment. Within the IBS subgroup, constipation-predominant patients showed slower CTT than their respective normal controls (p=0.03 for the Blue Dye method and p=0.048 for the BSFS method) after melatonin treatment, but significant differences were lacking in diarrhea-predominant patients. This study was well designed.
        • Saha et al. conducted a randomized, placebo controlled trial to determine if melatonin was effective in improving bowel symptoms, extracolonic symptoms, and quality of life (QOL) in 18 irritable bowel syndrome (IBS) patients (438). Adult patients with IBS symptoms (Rome II criteria) for more than three months were included. Cerebral disease, previous history of gastrointestinal surgery, and other systemic disorders were excluded. Patients (aged 18-65 years; six females) were randomly assigned to receive melatonin 3mg (N=9) or matching placebo (N=9) at bed time for eight weeks. The overall IBS scores, extracolonic IBS scores, and QOL scores were assessed at two, four, six, and eight weeks during treatment and at 16, 24, and 48 weeks during follow-up. Compared with placebo, melatonin taken for eight weeks significantly improved overall IBS score (45% vs. 16.66%, p<0.05). The posttreatment overall extracolonic IBS score was significantly lower (49.16% to 13.88%, p<0.05) when compared with the placebo group. The overall improvement in QOL score was 43.63% in the melatonin group and 14.64% in the placebo group, which was statistically significant. Decreased libido was reported as an adverse event in the melatonin group. Drowsiness was reported in both groups. This study is limited by the lack of description of randomization and withdrawals.
        • Lu et al. conducted a placebo controlled, crossover trial to examine the effects of melatonin on irritable bowel syndrome in 24 patients (600). The patients satisfied the Rome II criteria for IBS, with diagnosis at least one month prior to study. Patients with previously diagnosed organic gastrointestinal diseases such as inflammatory bowel disease, previous gastrointestinal surgery, and concurrent severe systemic diseases, including diabetes mellitus, were excluded. Patients received melatonin 3mg or an identically appearing placebo for eight weeks, followed by a four-week washout period, and placebo or melatonin in the reverse order for another eight weeks. Three validated questionnaires (GI symptoms, sleep questionnaires, and the Hospital Anxiety and Depression Scale) were used to assess symptom severity and to compute the IBS, sleep, and anxiety/depression scores, respectively. Results for 17 patients were included in the analysis; withdrawals due to adverse events were lacking. Improvements in mean IBS scores were significantly greater after treatment with melatonin (3.9 ± 2.6) than with placebo (1.3 ± 4.0, p=0.037). Percent response rate, defined as percentage of subjects achieving mild-to-excellent improvement in IBS symptoms, was also greater in the melatonin-treated arm (88% vs. 47%, p=0.04). The changes in mean sleep, anxiety, and depression scores were similar with either melatonin or placebo treatment. This study is limited by the method of randomization; patients were randomized by tossing a coin.
        • Song et al. conducted a randomized, placebo controlled trial to examine the effects of melatonin on symptoms associated with IBS in 40 patients (599). Patients with a diagnosis of IBS (Rome II criteria) who suffered sleep difficulty (defined as experiencing difficulty getting to sleep, awakening during the night, and/or early-morning awakening), with a Pittsburgh sleep quality index score of greater than 5 and sleep disturbance occurring at least two nights per week in the preceding 12 weeks, were included. Patients were asked not to take any medications known to alter gastrointestinal function or sleep conditions within a month before and during the study. Medication history was recorded. Exclusions included pregnancy; breastfeeding; organic gastrointestinal, anal, hepatic, or other systemic disorders; previous gastrointestinal surgery history except appendectomy; or history of cerebral disease or surgery. The IBS patients (aged 20-64 years; 24 female) received melatonin 3mg (N=20) or placebo (N=20) at bedtime for two weeks. Immediately before and after the treatment, subjects completed bowel, sleep, and psychological questionnaires, and underwent rectal manometry and overnight polysomnography. Compared with placebo, melatonin taken for two weeks significantly decreased mean abdominal pain score (2.35 vs. 0.70; p<0.001) and increased mean rectal pain threshold (8.9 vs. -1.2mmHg; p<0.01). Bloating, stool type, stool frequency, and anxiety and depression scores did not differ in a statistically significant manner after treatment in both groups. Data from sleep questionnaires and polysomnography showed that the two-week course of melatonin lacked an influence on sleep parameters. Randomization and withdrawals were not adequately described.
        • Klupińska et al. conducted a randomized, double-blind, placebo controlled study to assess the effects of melatonin on functional dyspepsia (N=60) (598). Individuals presenting with functional, nonulcer dyspepsia were included in the study. All included participants had epigastric pain that was chronic or recurring based on the Rome Criteria II. Participants were excluded from the study if they had upper abdomen disorders, H. pylori infection, high-grade gastritis, organic diseases, neuropsychiatric disorders, a history of surgery, disorders that affected the function of the alimentary tract, allergies, intolerance to certain foods, or use of nonsteroidal anti-inflammatory drugs (NSAIDs). Participants were randomized to receive tablets containing 5mg melatonin (LEK-AM, Zakroczyn, Poland) or placebo daily, one hour before bed, for 12 weeks. Information regarding standardization was lacking. Adverse events reported by two participants in the melatonin group included fatigue, weakened muscle power, and vertigo. For these individuals, the dose of treatment was reduced to 3mg melatonin daily. Information regarding toxic effects, dropouts, and interactions was lacking. Outcome measures included changes in pain (on a scale of 0-10) and sleep propensity. Compared to baseline after one month of treatment, a significant reduction in complaints was observed for participants in both the melatonin (seven participants (23.3%), p<0.05) and placebo groups (eight participants (26.6%), p<0.05). Compared to the placebo group, participants treated with melatonin used fewer alkaline tablets after the first month (30.0 vs. 17.5 tablets, p<0.01). After two months of treatment, another seven participants from the melatonin group reported a lack of dyspeptic symptoms. In addition, the participants in the melatonin group continued to consume fewer alkaline tablets than the participants in the placebo group (7.5 vs. 30.5 tablets, respectively, p<0.01). After 12 weeks of treatment, 56% of participants in the melatonin group lacked dyspeptic symptoms, and 30% reported partial symptom improvement. In the placebo group, 93.3% of participants reported a lack of symptom improvement. Also, after 12 weeks of treatment, 63.3% of participants in the melatonin group reported improved sleep, with 16 of 19 melatonin group participants reporting reduced nocturnal pain. Based on multivariate analysis, dyspeptic symptom improvement correlated positively with melatonin treatment (OR 95.86, 95% CI 3.72 to 2469.37, p<0.01) and correlated negatively with previous H. pylori infection (beta-coefficient -4.82, p<0.05). The authors concluded that melatonin treatment improves sleeping and clinical symptoms of individuals with functional dyspepsia. Limitations of this study included a lack of information regarding the methods of randomization and blinding, as well as a lack of information regarding dropouts and interactions.
        • Studies of lesser methodological strength (not included in the Evidence Table): Kandil et al. conducted a nonblinded equivalence trial to determine the effects of melatonin treatment on gastroesophageal reflux disease (GERD) in 27 patients (602). Sixty patients were initially selected for the study. After undergoing tests, 27 subjects with GERD were selected and compared with nine healthy volunteers (age and sex matched) who served as controls. Subjects with duodenal or gastric ulcers or functional dyspepsia were excluded. Group I (control) lacked treatment. Group II received melatonin 3mg daily at bedtime. Group III received omeprazole 20mg twice daily. Group IV received melatonin 3mg daily at bedtime and omeprazole 20mg twice daily. All medications were administered for eight weeks, with results examined at weeks 4 and 8. The study assessed heartburn, epigastric pain, and lower esophageal sphincter (LES) pressure, residual pressure, relaxation percentage, pH, serum gastrin and mean melatonin level, relaxation duration, and basal acid output (BAO). Results showed a significant decrease in LES pressure, residual pressure, relaxation percentage, pH, serum gastrin and mean melatonin level, and a significant increase in relaxation duration and basal acid output (BAO) relative to the control group. Treatment with melatonin alone (Group II) for four and eight weeks elicited an improvement in GERD symptoms, such as heartburn and epigastric pain, as well significant changes in all other measured criteria (p<0.05). Treatment with omeprazole alone (group III) for four and eight weeks led to improvement of GERD symptoms. Significant changes in BAO and serum gastrin (p<0.05), as well as nonsignificant changes in other measured criteria, were also observed. Treatment with melatonin and omeprazole for four and eight weeks also showed improvement in GERD symptoms, as well as significant (p<0.05) changes in all other criteria. It should be noted that this study was not placebo controlled or blinded.
        • Select combination study (not included in the Evidence Table): Chojnacki et al. conducted a study to examine the effect of melatonin as adjuvant treatment for ulcerative colitis (N=60) (1397). Patients had been in clinical remission for 12 months. Patients were treated for 12 months with mesalazine 2 x 1g daily plus melatonin 5mg daily at bedtime or placebo. In the combination group all patients remained in remission during the 12 months of observation with improvements in the Mayo Clinic Disease Activity Index (MCDAI) values (1.50±0.51 to 2.75±1.86) whereas increases in these values occurred in the placebo group after 6, 9, and 12 months (1.61±0.68 to 5.10±2.22). CRP levels remained within the normal range in the combination group but increased in the placebo group; at the same time in this group there was a decrease in hemoglobin. A statistically significant difference between groups was lacking for anxiety and depression. Further details are lacking. The effects of melatonin alone are not clear.

        Glaucoma

        • Summary: It has been theorized that due to effects on photoreceptor renewal in the eye, high doses of melatonin may increase intraocular pressure and the risk of glaucoma, age-related maculopathy, myopia (382), or retinal damage (400). However, there is preliminary evidence that melatonin may actually decrease intraocular pressure in the eye, and it has been suggested as a possible therapy for glaucoma (460;461). Additional study is necessary in this area. Patients with glaucoma taking melatonin should be monitored by a healthcare professional.

        Headache

        • Summary: Melatonin has gained some notoriety as an alternative therapy for the prevention of headaches (1398). Several small studies have examined the possible role of melatonin in preventing various forms of headache, including migraine, primary stabbing headache, and cluster and tension-type headache (in people who suffer from regular headaches) (146;604;605;934;935;936;937;938;939;940). In a case report of an individual with a pineal cyst gland, 6mg daily melatonin improved nocturnal melatonin levels, as well as nocturnal headaches (607) and this has been the topic of a review (1399). In a poor quality study, melatonin lacked an effect on prophylaxis of cluster headache (1400), and in a well designed study, melatonin lacked an effect on migraine frequency (440). Although further information is lacking, melatonin was also of interest in cases of hypnic headache in children (1401). Although limited initial research suggests possible benefits in these types of headache, well-designed controlled studies are needed before a firm conclusion can be drawn.
        • Systematic review: Francis et al. conducted a systematic review and meta-analysis to assess the effects of various pharmacologic agents on the prevention and treatment of cluster headaches (606). The effects of civamide, suboccipital steroid injection, sodium valproate, sumatriptan, verapamil, cimetidine/chlorpheniramine, lithium, misoprostol, oxygen, capsaicin, nitrate tolerance, prednisone, zolmitriptan, cocaine/lidocaine, octreotide, dihydroergotamine and somatostatin were assessed in this review but are excluded from this summary focusing on melatonin. One included study assessed the effects of meltatonin in preventing cluster headaches (605). Relevant randomized, double-blind, placebo controlled or equivalence studies published by June 2009 were pooled from MEDLINE and EMBASE. All included studies assessed the effects of treatments in participants aged at least 18 years who presented with episodic or chronic cluster headaches. In the included melatonin study, participants were administered 10mg melatonin daily for a duration of two weeks. Information regarding standardization was lacking. There was a lack of adverse events in the patient population. Information regarding toxic effects, dropouts, and interactions was lacking. The main outcome measure was frequency of headaches. Compared to the placebo group, participants administered melatonin showed a significant reduction in headache frequency (+0.12 vs. -1.79, p<0.03). As a result, the authors give a Level C recommendation for the treatment of cluster headaches with 10mg melatonin, indicating that it may be considered for the prevention of cluster headaches. Limitations of this review included the lack of information regarding toxic effects, interactions, and dropout information.
        • Evidence: Alstadhaug et al. conducted a randomized, double-blind, placebo controlled crossover study to assess the effect of melatonin on migraine prophylaxis (N=48) (440). Subjects between the ages of 18-65 years who experienced migraines (with or lacking an aura) at least 2-7 times monthly for a minimum of one year were included in the study. Participants were excluded if they used drugs regularly, used migraine medication within the preceding month, or failed trials with treatment involving at least two migraine medication classes. Acute migraine treatment use was acceptable; however, participants who used hypnosis, sedatives, excessive headache medication, or overuse of headache medication were excluded. Other exclusions were pregnancy or breastfeeding, as well as presenting with chronic migraine, comorbid conditions, conditions needing medical attention or treatment, or psychiatric conditions. Participants were randomized to receive 2mg prolonged-release melatonin or placebo daily one hour before bedtime for eight weeks. Following a six-week washout period, participants were crossed over to the other treatment group for an additional eight weeks. A description of standardization was lacking. Adverse events during melatonin treatment included dizziness and fatigue (N=2), as well as nervousness and nightmares (N=1). During treatment with placebo, adverse events included fatigue (N=1), dry mouth and irritability (N=1), night sweats (N=1), eczema (N=1), and dream activity that was abnormally high (N=1). A description of toxic effects was lacking. One subject originally randomized to the placebo group was dropped from the trial due to mistaken inclusion. Another subject initially randomized to the placebo group withdrew from the treatment group without providing a reason. A description of interactions was lacking. The primary outcome measure was migraine attack frequency. The secondary outcome was sleep quality prior to migraines using the Pittsburgh Sleep Quality Index (PSQI). Overall, there was a lack of statistically significant, between-groups differences in migraine frequency and PSQI scores. However in participants with insomnia (PSQI greater than six at baseline), treatment with melatonin resulted in a statistically significant reduction in PSQI score vs. placebo (final scores 6.8±4.0 vs. 9.4±4.0, p=0.03). Limitations included an inadequate description of double-blinding methods. Otherwise, this was a well-designed trial.
        • Leone et al. conducted a randomized, double-blind, placebo controlled trial to examine the role of melatonin in cluster headaches in 20 patients (605). Patients with cluster headaches diagnosed using the International Headache Society criteria were included. Patients had to have had at least one cluster period, and all cluster periods had to have lasted at least one month. Exclusions included drug and alcohol abuse, liver or kidney disease, psychiatric disorders, and certain medications. During a cluster period, 20 cluster headache patients (two primary chronic, 18 episodic) received oral melatonin 10mg (N=10) or placebo (N=10) for 14 days, taken in a single evening dose. Adverse effects were lacking. Endpoints included headache frequency and analgesic use. Headache frequency was reduced (ANOVA, p<0.03), and there were strong trends towards reduced analgesic consumption (ANOVA, p<0.06) in the treatment group. Five of the 10 treated patients were responders whose attack frequency declined 3-5 days after treatment, and they experienced no further attacks until melatonin was discontinued. The chronic cluster patients did not respond. A response was lacking from patients in the placebo group. Limitations include lack of description of randomization, blinding, and withdrawals.

        Hepatitis

        • Summary: In a poorly conducted study in patients with nonalcoholic steatohepatitis (NASH), use of melatonin resulted in improvements in liver function (445). Further study is needed.
        • Evidence: Gonciarz et al. conducted a randomized, placebo controlled trial to assess the effects of melatonin on hepatic enzymes in subjects with NASH (N=45) (445). Subjects with histologically confirmed NASH were included if they showed elevated levels of plasma aminotransferases but lacked other liver disease. Individuals who had been taking other antioxidants were excluded. All participants had previously participated in a 12-week study assessing the effects of melatonin (1402). During the initial 12-week study, participants were randomized in a 2:1 ratio to receive 5mg melatonin (LEK-AM, Zakroczyn, Poland) or placebo twice daily at 9am and 9pm for 12 weeks. In this follow-up study, participants continued treatment for an additional 12 weeks (24 weeks total) and were observed for 12 weeks following treatment discontinuation. Information on standardization was lacking. Signi?cant adverse reactions to melatonin were reported as lacking, and information on toxic effects was lacking. There were three dropouts from the control group, but reasons were lacking. Information on interactions was lacking. Outcome measures included levels of plasma alanine aminotransferase (ALT), aspartate aminotransferase (AST), alkaline phosphatase (ALP), and gamma-glutamyl transpeptidase (GGT), as well as glucose, total cholesterol, and triglyceride levels. Compared to baseline, significant changes in levels of ALT were observed after 24 weeks for participants in the melatonin (118.0 to 69.5 IU/L, 43% reduction, p<0.05) and control group (120.0 to 76.0 IU/L, 33% reduction, p<0.05). Compared to the control group, the reduction in ALT observed for the melatonin group was statistically significant after 24 weeks. After the 12-week observation period, statistically significant between-group difference was lacking. Compared to baseline, significant changes in levels of AST were observed after 24 weeks for participants in the melatonin group (76.5 to 51.5 IU/L, 33% reduction, p<0.05). Significant changes in levels of AST were lacking for the control group. Compared to baseline, significant changes in levels of GGT were observed after 24 weeks for participants in the melatonin group (113.0 to 54.5 IU/L, 52% reduction, p<0.05). Significant changes in levels of GGT were lacking for the control group. Levels of cholesterol decreased in both groups compared to baseline, but the statistical significance of this change was unclear. Statistically significant changes in ALP, triglycerides, and glucose were lacking. The authors stated that the improvement in ALT, AST, and GGT found in a previous study of 12 weeks' melatonin treatment was maintained with a further 12 weeks of therapy. Limitations of this study included lack of information regarding the method of randomization, lack of blinding, and lack of information regarding dropouts.

        High blood pressure (hypertension)

        • Summary: Several controlled studies in patients with high blood pressure (BP) report small reductions in diastolic and systolic blood pressure when taking melatonin by mouth (orally) or inhaled through the nose (intranasally) (443;826;827;828;829). In a poorly designed study in patients with metabolic syndrome, melatonin treatment improved blood pressure, lipid profile, and parameters of oxidative stress (609). Similarly, in a poor quality study, the inclusion of melatonin in the combined treatment of cardiovascular disease resulted in anti-ischemic, antianginal, antioxidant, and hypotensive effects (369). Melatonin in addition to lisinopril or amlopidine was suggested to normalize metabolic parameters in elderly patients with hypertension (1403). Further details are lacking. A meta-analysis suggests the potential for benefit with slow-release melatonin for nocturnal blood pressure reduction, as opposed to regular melatonin (443). Most trials have been small and not well designed or well reported. Better-designed research is necessary before a firm conclusion can be reached.
        • Meta-analysis: Grossman et al. conducted a meta-analysis (443) of seven studies (368;375;377;378;608;1404;1405) to determine the efficacy and safety of melatonin for nocturnal blood pressure. Randomized controlled trials accessing the effects of melatonin for nocturnal hypertension that were published in English through December 2010 were pooled from PubMed. Studies using melatonin agonists and that had a Jadad score of less than three were excluded. Four studies utilized 5mg fast-release melatonin, and three studies utilized 2-3mg controlled-release melatonin. Duration of treatment varied from seven to 90 days. Information on standardization was lacking. Three trials reported adverse events of nightmares, headache, drowsiness, and weakness. Four studies reported a lack of adverse events. Reports of toxic effects were lacking. Information on dropouts and interactions was lacking. The main outcome measures considered were nocturnal systolic and diastolic blood pressure. Meta-analysis of seven studies showed a lack of significant effect of melatonin on nocturnal systolic and diastolic blood pressure when compared to placebo. In a subanalysis, controlled release melatonin resulted in a statistically significant decrease of systolic blood pressure (-6.1mmHg; 95% confidence interval [CI] -10.7 to -1.5; p=0.009) and in diastolic blood pressure (-3.5mmHg; 95% CI -6.1 to -0.9; p=0.009) while fast-release melatonin lacked a significant effect on both systolic and diastolic blood pressure. The authors concluded that controlled-release melatonin is effective and safe as an adjunct for nocturnal hypertension treatment, and fast-release melatonin is ineffective. A strength of the meta-analysis was that the included studies had a Jadad score of at least three. Limitations of the analysis included lacking information regarding dropouts, and interactions. Authors of this review were funded and some were employed by manufacturers of melatonin.
        • Evidence: Grossman et al. conducted a randomized, double-blind trial to examine the efficacy of melatonin in reduction of nighttime BP in 38 treated hypertensive patients with nocturnal hypertension (608). Treated hypertensive patients (22 males, mean age: 64 ± 11 years) with confirmed nocturnal hypertension (mean nighttime systolic BP >125mmHg), according to repeated 24-hour ambulatory blood pressure monitoring (ABPM), were included. There was a two-week placebo run-in period before randomization. Patients were randomized to receive controlled release (CR)-melatonin 2mg (Circadin®; Neurim Pharmaceuticals Ltd., Tel Aviv, Israel) or placebo two hours before bedtime for four weeks. A 24-hour ABPM was then performed. Adverse effects were lacking. The main endpoint was nocturnal blood pressure. Melatonin treatment reduced nocturnal systolic BP from 136 ± 9 to 130 ± 10mmHg (p=0.011) and diastolic BP from 72 ± 11 to 69 ± 9mmHg (p=0.002), whereas placebo lacked an effect on nocturnal BP. Effects of melatonin on daytime blood pressure measurements were lacking. Following treatment or placebo, there was a two-week washout period. All subjects completed the study. The reduction in nocturnal systolic BP was greater with melatonin than with placebo (p=0.01) and was most prominent between 2 a.m. and 5 a.m. (p=0.002). Limitations include the lack of description of randomization and blinding.
        • Cagnacci et al. conducted a randomized, double-blind trial to determine whether prolonged nocturnal administration of melatonin could influence the daily rhythm of BP in 22 women (378). Patients were normotensive or had essential hypertension controlled by diuretics or angiotensin-converting enzyme (ACE) inhibitors, or both. Treatments were kept constant throughout the study. Women who were receiving beta-blockers or calcium antagonists were excluded from the study. The women (47-63 years of age and with normal BP (N=9) or treated essential hypertension (N=9)), received a three-week course of a slow-release melatonin pill (3mg) (Armonia Retard, Nathura s.r.l., Montecchio Emila RE, Italy) or placebo one hour before going to bed. They were then crossed over to the other treatment for another three weeks. Two women were excluded due to insufficient blood pressure readings, and two women did not complete the study (reasons not provided). In each woman, ambulatory BP was recorded for 41 hours at baseline at the end of each treatment period. In comparison with placebo, melatonin administration lacked an influence on diurnal BP but did decrease nocturnal systolic (-3.77 ± 1.7mmHg, p=0.0423), diastolic (-3.63 ± 1.3mmHg, p=0.0153), and mean (-3.71 ± 1.3mmHg, p=0.013) BP without modifying heart rate. The effect was inversely related to the day-night difference in BP. It is unclear as to whether the placebo was identical to the melatonin; otherwise, the study was well designed.
        • Rechciński et al. conducted a randomized controlled trial to assess the effects of melatonin before sleep in patients with coronary artery disease (CAD) and with an abnormal circadian pattern of BP (375). Subjects were sixty patients with coronary artery disease (CAD), who were "nondippers" (having an abnormal circadian pattern of blood pressure on changes in circadian blood pressure profile and heart rate variability (HRV)) between the ages of 48 and 80 (75% male). Subjects received 5mg of melatonin or placebo daily for 90 days. In three patients, the mean arterial BP at nighttime decreased by more than 20% during melatonin treatment. Two dropouts were reported prior to the 30th day of study. All patients in this study were receiving beta-adrenolytics. A potential interaction between melatonin and calcium channel blockers was noted; however, the authors found this interaction to be insignificant (p>0.05). The outcome measures for this study were as follows: the individual mean systolic (SBP) and diastolic BP (DBP) during sleep, the so-called mean pressure at night (SBPnight, DBPnight), and during waking hours, the so-called mean pressure at daytime (SBPday, DBPday), as well as the mean circadian SBP and DBP (SBP24 and DBP24). Two American Board of Preventative Medicine (ABPM) examinations were conducted for each patient, the first one at baseline and the second one on the last day of the 90-day-long treatment. Pulse wave velocity (PWV) was measured by means of an automatic device for measuring carotid-femoral transit or propagation of the pulse pressure wave (the COMPLIOR system). Thirty five percent of patients treated with melatonin and 15% taking placebo experienced a change in the circadian profile of their arterial BP (considered to be the beneficial effect of melatonin treatment). The patients in the study group also experienced a 6.4% drop in mean systolic pressure. Also, after 90 days of treatment, those in the study group experienced a significant acceleration of mean 24-hour heart rate (p=0.018), increases in daily systolic and diastolic blood pressures (p=0.05, and p=0.012, respectively), a significant change in the day to night systolic blood pressure ratio (p<0.01), and a decrease in the nightly systolic blood pressure (p=0.049). While significant differences between the responder and the nonresponder groups were lacking prior to treatment, after 90 days of melatonin treatment, patients in the responder group had a higher standard deviation of normal-to-normal intervals (SDNN), and the patients in the nonresponder group had a significant increase in their mean heart rate during 24 hours (p=0.06). Responders were defined as patients exhibiting a nocturnal decrease in averaged systolic pressure in comparison with daytime values >20% or a nocturnal increase in SBP. Limitations of the study included a small sample.
        • Scheer et al. conducted a randomized, double-blind crossover trial to determine whether enhancement of the functioning of the biological clock by repeated nighttime melatonin intake might reduce ambulatory blood pressure in 16 men with essential hypertension (377). Patients had mild or moderate, uncomplicated, essential hypertension. The men were treated with an acute (single) and repeated (daily for three weeks) controlled-release oral melatonin (2.5mg; 100% dissolved in water in 60 minutes; Terafarm, Katwijk, the Netherlands) intake one hour before sleep. Reported adverse effects were lacking. Endpoints included 24-hour ambulatory blood pressure and actigraphic estimates of sleep quality. Repeated melatonin intake significantly reduced systolic and diastolic blood pressure during sleep by 6 and 4mmHg, respectively (p<0.05). An effect of treatment was lacking on heart rate or blood pressure during the day. The day-night amplitudes of the rhythms in systolic and diastolic blood pressures were increased by 15% and 25%, respectively (approximate p=0.03). A single dose of melatonin lacked an effect on blood pressure. Repeated (but not acute) melatonin also improved sleep, but a relationship between changes in blood pressure and benefits on sleep was lacking. This study is limited by the lack of description of randomization, blinding, and control.
        • Cagnacci et al. conducted a randomized, double-blind, placebo controlled trial to examine melatonin-induced vascular reactivity and reductions in blood pressure and norepinephrine levels in 31 postmenopausal women (507). Women with high blood pressure, blood lipids, or blood sugar were excluded. The women were confirmed using hormone levels and history as being 1-4 years postmenopausal. Eighteen women were not receiving HRT, and 13 women were treated continuously with transdermal estradiol (50mcg daily) plus cyclic medroxyprogesterone acetate (5mg daily x 12 days every 28 days), oral melatonin (1mg), or placebo. A discussion of adverse effects was lacking. Endpoints included internal carotid artery pulsatility index (PI), an index of downstream resistance to blood flow, blood pressure, and catecholamine levels. In the untreated postmenopausal women, melatonin lacked an effect on blood pressure. In the HRT-treated women, studied during the only estrogenic phase, melatonin reduced systolic (-8.1 ± 9.9mmHg; p=0.054), diastolic (-5.0 ± 7.0mmHg; p=0.049), and mean (-6.0 ± 6.6mmHg; p=0.037) blood pressure within 90 minutes. Norepinephrine (-50.1 ± 66.7pg/mL; p=0.019), but not epinephrine levels, were also reduced. Similarly, resistance to blood flow in the internal carotid artery, as evaluated by the PI, decreased (-0.190 ± 0.15; p=0.0006) in a way that was linearly related to pre-existing PI values (r2=0.5; p=0.0059). The authors concluded that the circulatory response to melatonin was conserved in postmenopausal women on HRT but not in untreated postmenopausal women. This study is limited by the lack of description of randomization, blinding, and withdrawals.
        • Cagnacci et al. conducted a placebo controlled trial to examine the cardiovascular effects induced by the daytime administration of melatonin in 12 young women (506). The women were considered young and healthy and were all asked to take a fixed dose estrogen-progestin birth control pill to remove hormonal flux. Patients received melatonin (1mg) or placebo as a single dose. Endpoints included blood pressure and the pulsatility index of the internal carotid artery. Melatonin reduced the pulsatility index of the internal carotid artery (p<0.01), as well as both systolic and diastolic blood pressure (in supine position; p<0.01), within 90 minutes vs. placebo. A significant modification in supine catecholamine levels was lacking, but norepinephrine levels evaluated after five minutes of standing position were significantly reduced (p<0.02). This study is limited by the lack of description of blinding, randomization, and withdrawals. Also, the use of hormones may have affected melatonin's effect on blood pressure.

        High cholesterol

        • Summary: In early study, melatonin, when used with zinc and the diabetes drug metformin, improved diabetes-related complications, such as impaired lipid profile (387). In a poorly designed study, in patients with metabolic syndrome, melatonin treatment improved blood pressure, lipid profile, and parameters of oxidative stress (609). Melatonin in addition to lisinopril or amlopidine was suggested to normalize metabolic parameters in elderly patients with hypertension (1403). Further details are lacking. However, there is also preliminary evidence in human and animal research that melatonin both increases and decreases cholesterol levels (757;758;759;760). More research is needed to clarify these mixed results.

        HIV/AIDS

        • Summary: There is a lack of well-designed scientific evidence to recommend for or against the use of melatonin as a treatment for AIDS (544). Melatonin should not be used in place of more proven therapies, and patients with HIV/AIDS are advised to be treated under the supervision of a medical doctor.

        Memory

        • Summary: Preliminary research has suggested that melatonin may improve memory acquisition in certain stressful contexts (625). However, findings in this field are preliminary; further research is required.
        • Studies of lesser methodological strength (not included in the Evidence Table): Rimmele et al. conducted a placebo controlled, single-blind trial to evaluate the effects of melatonin on memory processing while under stress (625). Subjects were 50 healthy men, 20-35 years of age. Exclusion criteria included acute or chronic illnesses, smoking, alcohol and illicit drug use, as well as regular strenuous exercise. Patients received 3mg of oral melatonin (N=27) or placebo (N=23) one hour prior to the initiation of stress testing. The study assessed stress responses, memory encoding under stress, and memory retrieval stress responses via the stress Trier Social Stress Test (TSST), as well as blood and saliva before and after the TSST to determine stress hormone levels. To examine the influence of melatonin on memory encoding under stress, the subjects encoded objects distributed in the test room during the TSST, the memories of which were evaluated the next day. Fifteen minutes after stress, retrieval of words memorized the day prior were also tested. The influence of melatonin on attention, concentration, wakefulness, and state anxiety were assessed before and after the stressor using the d2 letter cancellation test of attention, the wakefulness scale of the Multidimensional Mood Questionnaire, and the state scale of the State-Trait-Anxiety-Inventory (STAI). Baseline measures for attention, wakefulness, and state anxiety were obtained before melatonin or placebo administration. Melatonin was observed to enhance spatial memory encoded during stress (p<0.001); however, change was lacking in recall of words to which subjects had been exposed the day before. A difference in response to stress between the melatonin and placebo groups was lacking as indicated by cortisol, norepinephrine, and epinephrine levels (p=0.47, p=0.18, and p=0.14, respectively). Wakefulness significantly decreased throughout the experiment (p<0.001), with a stronger decrease in the melatonin group (p<0.01). A difference in attentional levels was lacking between the two groups (p=0.48). Limitations of this study include the lack of double-blinding and randomization.

        Menopause

        • Summary: Evidence is mixed with respect to the use of melatonin for symptoms associated with menopause (488;626;627;628). Further study using a larger number of patients is needed before a conclusion can be made.
        • Systematic review: Kelley and Carroll conducted a systematic review to evaluate to the effects of over-the-counter (OTC) treatments for menopausal hot flashes (627). The effects of black cohosh, dong quai, evening primrose oil, ginseng, isoflavones, kava red clover, soy, vitamin E, and wild yam were assessed in this review but are excluded from this summary focusing on melatonin. One included study assessed the effects of melatonin (626). The reviewers used International Pharmaceutical Abstracts, Medline, and PubMed to conduct a search for relevant articles published before June 2010. Only clinical trials comparing a single-ingredient agent with an active treatment or placebo were included. The references of the pooled articles were manually reviewed to identify additional relevant studies. All studies included assessed vasomotor or hot flash symptoms in women with menopause. According to the reviewers, participants received 3mg melatonin daily. Information on frequency and duration of treatment, as well as standardization was lacking from the review. Participants taking melatonin had a similar incidence of adverse effects as the participants taking placebo or isoflavones. Information on toxic effects, dropouts, and interactions was lacking from the review. The primary outcome measure was the change in the number of hot flashes. Based on results from the included study, melatonin lacked a statistically significant effect on hot flashes. The reviewers concluded that melatonin had limited evidence as a therapy for hot flashes. The primary limitation of this review for melatonin was that only one study was included, and details regarding study outcomes was lacking.
        • Evidence: Secreto et al. conducted a double-blind, randomized, placebo controlled trial to evaluate the individual and additive effects of soy isoflavones and melatonin in relieving menopausal symptoms in 262 women (626). Postmenopausal women aged ≥35 years with the last menstrual flow at least six months before recruitment, with any condition for which classic HRT is not recommended, were included unless they were using breast cancer therapy or HRT during the previous three months, had overt endocrinopathy (diabetes mellitus, hyperthyroidism, etc.), or intolerance to soy. The women were randomized to receive (1) soy isoflavones + melatonin; (2) soy isoflavones alone (80mg); (3) melatonin alone (3mg); or (4) placebo, for three months. Twelve women withdrew from the study due to adverse events; however, it was even between groups. Specific adverse events related to melatonin were lacking. Endpoints included the severity of menopausal symptoms using the Greene Climacteric Scale. Although a statistical test was lacking, differences in the median percent differences between basal and final scores were lacking. It was stated that the placebo response was much higher than planned, and therefore it was meaningless to perform any statistical test. Somatic and vasomotor symptoms were also similar between the four groups. Improvement of psychological symptoms was higher in the isoflavones + melatonin group vs. the other three. Melatonin lacked an effect on menstrual flow. This study was well designed. Randomization was done using a computer program, and the supplements were identical. The study was double-dummy.
        • Kotlarczyk et al. conducted a randomized, double-blind, placebo controlled trial to assess the effects of melatonin on quality of life and bone structure of women going through menopause (N=19) (628). Women were included if they were 45 years or older with irregular periods but menstruated at least once in the previous six months. Women were excluded if they currently used hormone therapy or birth control, medications that may affect bone density, or medications for sleep, depression, or blood pressure control. Also, participants were excluded if they had used medications for bone thinning within three months of the study or had used steroids within six months of the study. Finally, participants were excluded if they were intolerant to lactose, smoked tobacco, had hypertension, chronic obstructive pulmonary disease, cancer, hepatic disease, sleep apnea, or were on medication for depression, or insomnia. Participants were randomized by computer-generated randomization (3:1 ratio) to receive 3mg melatonin or placebo, to be taken orally as capsules (supplied by Professional Compounding Centers of America) each night for six months. Both melatonin and placebo were supplied in blister packs, the capsules containing either lactose alone (placebo) or with melatonin. Information regarding adverse or toxic effects was lacking. One woman withdrew from the study. Additional information regarding this dropout was lacking. Discussion of interactions was lacking. Menopause Specific Quality of Life Intervention (MENQOL) and Pittsburgh Sleep Quality Index (PSQI) were used for quality of life and sleep assessment. Participants recorded well-being, menstruation and sleep patterns in diaries. Bone health was evaluated using calcaneal ultrasound and measurement of serum osteocalcin (bone formation (OC) and bone resorption (NTX)) as well as ratio of NTX:OC. Significant changes were lacking between groups or compared with baseline in bone density, NTX, or OC, although the NTX:OC ratio in the melatonin group was reduced while it remained unchanged in the placebo group. Significant changes were lacking in the overall PSQI score, mean number of sleeping hours, and MENQOL psychosocial, sexual or vasomotor scores, but compared to placebo, physical domain scores improved with melatonin (0.6 vs. 0.1, p<0.05). The mean number of periods over the six months was reduced in the melatonin group vs. placebo (4.3 vs. 6.5, p<0.05) and the mean days free of bleeding were greater with melatonin vs. placebo (51.2 vs. 24.1 days, p<0.05). However, the authors noted that women in the melatonin group were significantly older than those in the placebo group. The authors concluded that physical symptoms improved with melatonin, that it may prevent bone loss, and that it was well tolerated. Limitations of this study included the small participant population.
        • Bellipanni et al. conducted a randomized, placebo controlled, double-blind trial to examine the effects in 79 women of evening administration of melatonin on the level of hormones which are known to be involved in the genesis and progression of menopause (488). Perimenopausal and menopausal women from 42 to 62 years of age with no pathology or medication were selected. There were premenopausal women (N=25), perimenopausal women (N=36), and postmenopausal women (N=18). The women were divided into two age groups (42-49 and 50-62), as well as on menopausal status. Melatonin was measured in saliva to divide them into low-, medium- and high-melatonin patients. Half of the patients took 3mg of melatonin and half took placebo at bedtime (10-12 p.m.) for six months. Adverse effects were investigated, but a discussion regarding them was lacking. None of the women were using hormonal agents, so it is not clear if there would be interactions in women using these medications. At three and six months, blood was taken for determination of pituitary (LH, FSH), ovarian, and thyroid hormones I (T3 and T4). Thyroid hormone levels significantly increased in all women with low basal melatonin taking melatonin (p<0.05). Before initiation of the study, there was a negative correlation in all women between LH, FSH, and basal melatonin levels. However, after six months of melatonin use, melatonin resulted in a significant reduction of LH in the younger women (43-49 year-olds), with a lack of effect in the older women (50-62 years old). Also, in women with low basal melatonin levels, melatonin resulted in a significant decrease in FSH. Mood and depression were generally improved with melatonin use. Randomization and blinding were not adequately described in this study.

        Pain

        • Summary: In infants requiring intubation, melatonin reduced pain when measured in one, but not both, pain scales (552). Further research is required in other patient populations with various types of pain.
        • Evidence: Gitto et al. conducted a prospective, randomized, controlled trial to assess the analgesic and anti-inflammatory effects of melatonin in infants (N=60) (552). The study subjects were newborn infants of ≤32 weeks gestation who needed endotracheal intubation for mechanical ventilation. These infants had all been admitted to a Neonatal Intensive Care Unit. Infants were excluded if there were signs of infection, an inborn error of metabolism, or identifiable congenital abnormalities of the brain. All participants received standard sedation and analgesia according to the Italian Society of Neonatology. This treatment included a rapid bolus of 0.02mg/kg atropine, slow administration of 1mcg/kg fentanyl, and 0.1mg/kg vecuronium. After intubation, fentanyl was infused intravenously at the dose of 0.5-3mcg/kg/hr continuously. In addition to a standard regimen of analgesia and sedation, participants randomized to the melatonin group received 10mg/kg melatonin intravenously before intubation as a single dose. The melatonin was prepared by Helsinn Chemical Co, Biasca, Switzerland. In a chromatographically pure state, the melatonin was dissolved in a 1:90 mixture of ethanol and 0.9% saline. Discussion of adverse or toxic effects was lacking. Information on dropouts and interactions was lacking. The primary outcome of the study was pain alleviation potential of melatonin, which was assessed by the Neonatal Infant Pain Scale (NIPS) before, during and five minutes after endotracheal intubation and the Premature Infant Pain Profile (PIPP) during ventilation at 12, 24, 48, and 72 hours. The secondary outcome was inflammatory response based on plasma levels of the pro- and anti-inflammatory cytokines interleukin (IL)-6, IL-8, IL10, and IL-12. Statistically significant differences in the NIPS scores between the active treatment and control were lacking. There was significant lowering (p<0.001) of the PIPP scores in the melatonin group at every time interval compared to the control group. At 72 hours the mean scores were 7.01 ± 0.22 and 12.64 ± 0.50 for the melatonin and control groups respectively (p=0.001). With the secondary objective, all the cytokine levels were higher at each interval in the control group when compared with the melatonin group at a statistically significant level (p<0.001). Compared to the control group after 72 hours, participants administered melatonin showed significantly lower levels of IL-6 (105.6 ± 3.70 vs. 19.2 ± 1.36pg/mL, p<0.001), IL-8 (156.7 ± 2.42 vs. 13.4 ± 2.10pg/mL, p<0.001), IL-10 (18.6 ± 0.76 vs. 17.9 ± 0.83pg/mL, p<0.001), and IL-12 (156.1 ± 1.02 vs. 96.6 ± 0.67pg/mL, p<0.001). These between-group differences were still present after seven days (p<0.001). After the administration of melatonin the authors observed reduced pain scores from 12 hours after intubation compared with controls in infants during ventilation via an endotracheal tube. Limitations of this study included a lack of placebo control, lack of blinding, and lack of information on the randomization method.

        Parkinson's disease

        • Summary: There is very limited study to date for the use of melatonin as a treatment in Parkinson's disease. Papavasiliou et al. conducted a single-blind study to examine the effects of melatonin in addition to levodopa in patients with Parkinson's disease (381). All patients (N=11) were treated with melatonin (3.0-6.6g daily), and seven of these were also treated with levodopa (0.36-5.0g daily). Ten of the 11 patients had idiopathic parkinsonism. Melatonin was indicated as being well tolerated, but side effects may have included skin flushing, diarrhea, abdominal cramps, somnolence during the day, scotoma lucidum, and headaches. Melatonin lacked an effect on signs of parkinsonism or levodopa effects. In a systematic review of small studies, melatonin was found to potentially improve sleep in patients with Parkinson's disease (444). This topic has been reviewed (1406). Further research of higher methodological strength is needed before a conclusion can be made in this area
        • Systematic review: Seppi et al. performed a systematic review to update a previous evidence based (EBM) review regarding the treatment of nonmotor symptoms (including depression, fatigue, impulse dyscontrol and abnormal repetitive behaviors, dementia, psychosis orthostatic hypotension, sexual dysfunction, gastrointestinal motility problems, sialorrhea, insomnia, and tiredness) in patients with Parkinson's disease (PD) (444). The effects of dopamine agonists, tricyclic antidepressants, selective serotonin reuptake inhibitors (SSRIs), other antidepressants, omega-3 fatty acids, nonpharmacological interventions, methylphenidate, modafinil, amantadine, acetylcholinesterase inhibitors, memantine, antipsychotics, fludrocortisone/domperidone, sildenafil, macrogol, ipratropium bromide spray, glycopyrrolate, botulinum toxin B, levodopa/carbidopa, L-dopa/carbidopa, pergolide, and eszopiclone were assessed in this review but are excluded from this summary focusing on melatonin. Two new included trials assessed the efficacy of melatonin on insomnia (435;667). Relevant studies published in English from January 2002 through December 2010 were pooled from Medline and the Cochrane Library central databases. In addition, reference checking software was used to identify additional articles and clinical reports. Initially all included studies had to have a minimum of 20 patients, an established rating scale or well-defined endpoints, and a treatment duration of at least four weeks. However, for some conditions, studies that assessed the effect of treatment in patient populations <20 were included if the number of other studies was limited or if it was the only study that showed a positive result. Participants in the melatonin studies were administered 3-50mg melatonin daily before bedtime for 2-10 weeks. Information on standardization was lacking. Adverse effects were lacking in both studies included in the review. However, the reviewers mentioned that the general side effects associated with melatonin may include nausea, grogginess the next day, irritability, and headache. Information regarding toxic effects was lacking. In one study, three of 43 patients dropped out. In the other study, 18 of 20 included patients were analyzed. Information regarding why participants dropped out was lacking. Information regarding interactions was lacking. Outcome measures included sleep latency, excessive daytime sleepiness (EDS; based on daytime somnolence (ESS) and Stanford Sleepiness Scale), daytime levels of functioning, subjective sleep quality (based on the Pittsburgh Sleep Quality Index), objective sleep quality (based on polysomnography), total sleep time (TST), sleep efficiency, sleep fragmentation, and the Unified Parkinson's Disease Rating Scales (UPDRS) score. In one study, treatment with 50mg melatonin significantly increased nocturnal sleep time by 10 minutes vs. placebo (p<0.05). Also, treatment with 5mg melatonin significantly improved overall sleep quality vs. placebo (p<0.05). However, improvement in sleep quality was lacking for the 50mg melatonin group. Also, improvement in EDS was lacking based on the ESS and Stanford Sleepiness Scale. In another study, melatonin significantly improved sleep quality scores (based on Pittsburgh Sleep Quality Index) from 8.3±4.5 to 4.5±3.1 compared to placebo, which showed a change in score from 9.9±3.7 to 8.7±4.0 (p=0.03). Significant changes in other outcome measures were lacking. Limitations of this study included small sizes of the included studies.

        Periodic limb movement disorder

        • Summary: There is very limited study to date for the use of melatonin as a treatment in periodic limb movement disorder. Kunz et al. conducted a study to determine if exogenous melatonin would decrease periodic limb movement and thereby improve symptoms in patients with this disorder (629). Nine patients with a first-time diagnosis of periodic limb movement disorder without restless leg syndrome were treated over a six-week period with 3mg of melatonin, taken between 10 p.m. and 11 p.m. Melatonin improved well-being in seven of the nine patients. Further research of higher methodological strength is needed before a conclusion can be made in this area.

        REM sleep behavior disorder

        • Summary: Melatonin has been suggested as a possible treatment for REM sleep behavior disorder (1407). Limited case reports describe benefits in patients with this condition (630;631;1408) and it has been the topic of a review (1409) and a systematic review with overall evidence of benefit from mainly studies of lesser methodological strength and small studies (reviewers gave it a grade B) (441); however, more rigorous research is needed before a clear conclusion can be drawn.
        • Systematic review: Aurora et al. conducted a systematic review of 42 articles to evaluate the available therapies for REM sleep behavior disorder (RBD) to update a Best Practice Guide on RBD treatment (441). The effects of clonazepam, pramipexole, paroxetine, L-3,4-dihydroxyphenylalanine (L-DOPA), acetylcholinesterase inhibitors, zopiclone, benzodiazepines, temazepam, triazolam, alprazolam, Yi-Gan San, desipramine, clozapine, carbamazepine, sodium oxybate, and safe sleep environments were assessed in this review but are excluded from this summary focusing on melatonin. Of the 42 articles included, six trials involved melatonin specifically (630;631;1410;1411;1412;1413). The review authors first searched PubMed in February 2008 then updated the search in June 2009. The references of the pooled articles were manually reviewed to identify additional relevant studies. Articles selected for inclusion had to include a specific adult patient population presenting with a particular condition, had to provide specific details regarding the treatment, and had to specifically describe the outcome measures used to assess the treatment. All included studies were published in English. According to the reviewers, participants received 3-12mg melatonin at bedtime. Information on frequency and duration of treatment, as well as standardization, was lacking from the review. Adverse effects included headache or sleepiness in the morning, delusions, or hallucinations. Information on toxic effects, dropouts, and interactions was lacking in the review. The primary outcome measure was the clinical response to RBD therapies compared to the patient's other medications or the patient's natural history. The authors also assessed whether the RBD therapy was able to prevent falls or injuries by modifying the sleep environment. When creating the Best Practice Guide, the task force used the Oxford Centre for Evidence-based Medicine Levels of Evidence scale. Each RBD therapy received a grade from A to C depending on the quality of evidence and a task force clinical consensus. For treating RBD, the reviewers concluded that melatonin has weaker evidence than clonazepam but better evidence than the other agents reviewed. Based on results from six studies, 31 out of 38 included participants (who presented with dementia of Lewy body (DLB), Parkinson's disease (PD), multiple system atrophy (MSA), memory problems, or sleep-disturbed breathing) improved with melatonin treatment. In participants taking melatonin, polysomnography showed significant decreases in REM sleep that lacked atonia and REM movement time. The authors gave melatonin a grade B recommendation. Strengths of the review include the wide variety of RBD therapies assessed. The review was limited by the inclusion of studies of lesser methodological strength, lack of information regarding treatment frequency and duration, and lack of detailed information regarding study outcomes.
        • Evidence: Kunz et al. conducted a two-phase, randomized, double-blind, placebo controlled, crossover trial to assess the effects of melatonin on rapid eye movement (REM) sleep behavior disorder (RBD) (N=8) (632). Male individuals presenting with RBC based on International Classification of Sleep Disorders (ICSD) criteria were included in the study. Participants were excluded if they were experiencing or had experienced shiftwork within the past year, had poor sleep hygiene (which was described as having a variation in bedtime of more than two hours during the two-week recruitment period), lacked normal bedtime hours between 22:00-24:00, had traveled transmeridian within one month of trial initiation, were diagnosed with psychiatric disorders, had pathological findings in brain imaging, had recently changed medications within the last month, or took medication that would interfere with the production or secretion of melatonin or REM sleep. Participants were randomized to receive either 3mg melatonin or placebo daily between 22:00-23:00 for four weeks. Following a 3-5 day washout period, participants were crossed over to the other treatment group for an additional four weeks. The melatonin products were analyzed for purity by the Department of Pharmaceutics at Freie Universität Berlin. Bioavailability of the capsules occurred within 30 minutes of ingestion. A lack of treatment-related adverse effects was observed during the study. Information regarding toxic effects was lacking. All randomized participants completed both parts of the study. Information on interactions was lacking. Outcome measures included sleep onset latency, REM latency, sleep period time, total sleep time, sleep efficiency, wake after sleep onset, non-rapid eye movement (stages one and two), slow wave sleep, REM, REM-density, phasic muscle twitches, percentage of REM-epochs with more than 50 percent of the epoch with muscle atonia, clinical global impression-severity, and clinical global impression-change. Compared to baseline, participants treated with melatonin showed significantly reduced sleep-onset latency (-1.96±0.050 minutes, p<0.05), percentage of REM sleep epochs that lacked muscle atonia (-2.52+0.012%, p<0.05), and severity of symptoms based on CGI scores (-2.264±0.024, p<0.05). All other changes in sleep variables evaluated lacked statistical significance compared to baseline. Participants treated with placebo showed significantly reduced sleep-onset latency (-2.03±0.043 minutes, p<0.05), but lacked any other statistically significant changes. Compared to the placebo group, participants treated with melatonin showed statistically significant reduction in sleep epochs (-1.96=0.050%, p<0.05) and change in CGI (-2.157±0.031, p<0.05). The authors concluded that melatonin was effective in treating patients with RBD, although the optimal dose for treatment still needed to be determined. Limitations of this study included the small participant population.

        Restless leg syndrome

        • Summary: Preliminary research has suggested that melatonin may have a detrimental effect on motor symptoms associated with restless leg syndrome (633); however, evidence remains inconclusive. Further research is required in this area.
        • Combination studies or studies of lesser methodological rigor (not included in the Evidence Table): Whittom et al. conducted an open-label, observational study to examine whether the onset of melatonin secretion may play a role in genesis of restless leg syndrome (RLS) and worsening of the symptoms (633). Patients were included if they were drug naïve (free from any drug known to affect sleep, sensory or motor functions, or melatonin secretion), were diagnosed with primary RLS, had a severity score >20 on the International RLS Study Group Severity Scale, and had a habitual bedtime between 9 p.m. and midnight. Patients were excluded if they had a medical condition with known association to RLS (anemia, renal failure) or if another sleep disorder was present (narcolepsy, sleep apnea). Also, patients with clinical signs or a history of a psychiatric, affective, or other neurological disorder within six months preceding the study were excluded. Patients were given 3mg of oral melatonin tablet and were exposed to 3,000 lux of bright light from a light lamp. Each of the eight subjects were studied at three different times: baseline (to assess symptom severity without intervention), after oral administration of 3mg of melatonin tablet (at 7 p.m.), and during exposure to light lamp delivering 3,000 lux of bright light (7 p.m. to midnight). The dose of melatonin and bright-light exposure was separated by one week, and the order was reversed in four of the subjects. Dropouts were not specifically discussed; however, there was mention of one subject who was not included in SIT-PLM analysis, because she could not tolerate the testing during baseline and melatonin conditions. Severity of RLS symptoms were measured using the suggested immobilization test (SIT). SIT consisted of two parts: a subjective measure of motor manifestations (SIT-PLM) and a subjective mean discomfort score (SIT-MDS). The SIT PLM used surface electromyograms (EMG) on both legs to measure periodic leg movements. The SIT-MDS averaged 12 readings on a 10cm visual analog scale (VAS). SIT was administered two times at baseline, two times before oral melatonin administration, and twice for bright-light exposure. Saliva samples were also taken before and after each SIT to assess melatonin concentration. There was a statistically significant increase of the SIT-PLM index when subjects received exogenous oral melatonin compared to baseline (p=0.038) and bright-light conditions (p=0.011), but bright-light exposure lacked an effect on leg movements compared to the baseline condition. When exposed to bright light, there was a significant decrease in MDS compared to baseline (p = 0.032), a difference in MDS with oral melatonin administration was lacking when compared to both bright-light exposure and baseline condition. Also, there was a small but statistically significant (p<0.05) decrease in sensory symptoms with bright-light exposure compared to baseline. After oral melatonin administration, there was a statistically significant increase noted from baseline condition during the first and second SIT (5pg/mL at baseline to 70pg/mL, and 12pg/mL at baseline to 60pg/mL, p<0.001 for both). After exposure to bright light, there was a statistically significant suppression in melatonin secretion (mean <5pg/mL for both SIT 1 and 2, p<0.001). In addition to a small sample size, only patients who were severely affected by RLS were included in this study. It is possible that those who had more mild presentations might experience more noticeable affects than those included in the study.

        Rett's syndrome

        • Summary: Rett's syndrome is a presumed genetic disorder that affects female children, characterized by decelerated head growth and global developmental regression. There is limited research on the possible role of melatonin in improving sleep disturbance associated with Rett's syndrome (717;718;1414). Further research is needed before a conclusion can be made in this area.

        Sarcoidosis

        • Summary: A case series has examined the effects of melatonin on symptoms associated with chronic sarcoidosis (CS) and determined that it may be an effective and safe therapy for CS when other treatments fail or cause side effects (531). More well-designed studies are needed before a conclusion can be made.
        • Studies of lesser methodological rigor (not included in the Evidence Table): Pignone et al. conducted a clinical trial to examine the effects of melatonin in CS patients in whom usual treatments were ineffective or caused serious side effects (531). Eighteen patients with CS received melatonin for two years (20mg daily in the first year, 10mg daily in the second year). Side effects and disease relapse were lacking during melatonin treatment. Pulmonary function tests, chest X-rays, pulmonary computed tomography, Ga(67) scintigraphy, and angiotensin-converting enzyme (ACE) were assayed at baseline and in the follow-up. Normalization of ACE, improvement of pulmonary parameters, and resolution of skin involvement were found in the patients given melatonin. After 24 months of melatonin therapy, hylar adenopathy completely resolved in eight patients, and parenchymal lesions were markedly improved in all patients; in the five patients with reduced diffusion capacity of the lung for carbon monoxide, the values normalized after six months of therapy and remained stable until month 24. After 24 months, Ga(67) pulmonary and extrapulmonary uptake was normalized in seven patients and, at month 12, ACE was normalized in six patients, in which the values were high at the baseline. Skin lesions, present in three patients, completely disappeared at 24 months.

        Schizophrenia

        • Summary: The effect of melatonin on schizophrenia was investigated in a systematic review; two studies investigated effects on tardive dyskinesia and one on schizophrenia symptoms (634). In the one included study, schizophrenic patients treated with melatonin showed improved sleep, daytime function, and mood. Further study is needed.
        • Systematic review: Anderson and Maes conducted a systematic review to examine the association of changes in melatonin levels with the etiology and course of schizophrenia (634). While the effects of schizophrenia on endogenous melatonin levels were assessed in this review, these references have been excluded from this summary focusing on the effects of melatonin supplementation. In addition, some of the included studies assessed the effects of melatonin in healthy individuals; these studies have also been excluded from this review focusing on the effects of melatonin in patients with schizophrenia. Three published studies included in the review assessed the effects of melatonin supplementation in patients with schizophrenia (671;673;684). Articles assessing the effects of melatonin on the course and treatment of schizophrenia and its etiology were pooled from PUBMED, Google Scholar, and Scopus. In two studies, patients with schizophrenia were orally administered 2-10mg melatonin daily to treat TD. Information regarding doses and frequency were lacking in other situations. Comment on duration of therapy was lacking for all studies. Information on standardization, allergies, adverse effects, toxic effects, and dropouts was lacking. Information regarding interactions that occurred during the included clinical studies was lacking. For the clinical trials assessed in the review and included in this summary, outcome measures included mood, daytime function, psychosis, management of pneumonia and COPD, and the effects of melatonin on TD. In one study, patients with schizophrenia treated with melatonin showed improved sleep, daytime function, and mood. In other studies, participants treated with 2mg melatonin daily lacked a significant improvement in TD, while those treated with 10mg daily showed significant decrease in TD. According to the reviewers, understanding the role of melatonin in schizophrenia could assist with the management of the condition, particularly in association with vitamin D3. Details regarding doses, allergies, adverse effects, and toxic effects were lacking.

        Seasonal affective disorder (SAD)

        • Summary: There are several small brief studies of melatonin in patients with SAD (635;1415;1416;1417;1418;1419;1420;1421;1422;1423;1424;1425). This research is not well designed or well reported, and further research is necessary before a clear conclusion can be reached.
        • Evidence: Leppämäki et al. conducted a randomized, placebo controlled trial to examine the effects of melatonin on sleep, waking up, and well-being in 58 subjects with varying degrees of seasonal or weather-associated changes in mood and behavior (635). The adults exhibited subsyndromal seasonal affective disorder (s-SAD) and/or the negative or positive type of weather-associated syndrome (WAS). Patients with known hypersensitivity to melatonin, severe depression, or substance abuse were excluded. Patients were randomized to 2mg of sustained-release melatonin (Circadin®, Neurim Pharmaceuticals, Tel Aviv, Israel) or placebo tablets 1-2 hours before a desired bedtime for three weeks. Outcome measures were changes from baseline in sleep quality, sleepiness after waking, atypical depressive symptoms, and health-related quality of life by week 3. Early-morning salivary melatonin concentrations were measured at baseline and treatment cessation in all subjects. Melatonin administration significantly improved the quality of sleep (p=0.03) and vitality (p=0.02) in the subjects with s-SAD, but attenuated the improvement of atypical symptoms and physical parameters of quality of life compared to placebo in the subjects with WAS, positive type. Limitations include a lack of description of randomization and blinding.
        • Danilenko et al. conducted a randomized, double-blind, placebo controlled trial to examine the effects of melatonin on sleep patterns in 16 women with seasonal affective disorder (SAD) and compared results with 17 healthy controls (636). Patients were unmedicated, in good general health, free of major sleep disorders, and were not engaged in shift work or long-distance travel in the prior two months. The women with SAD were diagnosed according to DSM-IV criteria. Baseline waking electroencephalogram (EEG) measurements were done in all patients over 30 hours of sleep deprivation. At 5 p.m., at completion of the 30-hour baseline, patients were randomized to placebo or 0.5g of sublingual melatonin (California Health®). The dose was repeated over a six-day period. Sixteen SAD patients and 13 controls finished the study. A discussion of adverse effects was lacking. Endpoints included sleep duration, mid-sleep time, waking EEG power density, and subjective scores of mood. At baseline, the increase in EEG power density in a narrow theta-band (5-5.99 Hz, derivation Fz-Cz) during the 30-hour protocol was attenuated in patients compared with controls (p=0.037), with sleepiness (p=0.092) and energy (p=0.045) self-ratings following a similar pattern. After sleep deprivation, six patients improved (≥50% reduction on the 29-item Structured Interview Guide for the Hamilton Depression Rating Scale-Seasonal Affective Disorder Version (SIGH-SAD) score). A differential effect of melatonin or placebo was lacking on any measure; both treatments stabilized the improvement. This study is limited by a lack of description of randomization, withdrawals, and blinding.

        Seizure disorder

        • Summary: The role of melatonin in seizure disorder is controversial. There are several reported cases of children with intractable seizures or neurologic damage whose seizure-related symptoms improved with regular melatonin administration (453;454;454;455;455;456;456;457;458;721;722). Exogenous melatonin has also been suggested as a promising treatment for children with epilepsy or febrile seizures, as low serum levels of melatonin were observed in such patients (1426). Melatonin administration resulted in an increase in the activity of antioxidant enzymes, glutathione peroxidase, and glutathione reductase in epileptic children receiving valproate (720;816); it has been suggested that such activity may help protect neurons from oxidative stress and damage. In a small study, sleep quality and seizure control were improved with melatonin in children with epilepsy (744). Limited animal research has supported possible antiseizure effects (815;999;1427). However, there has also been a report that melatonin may actually lower seizure threshold and increase the risk of seizures (451). Melatonin has been suggested as a superior means (vs. chloral hydrate) of encouraging compliance with sleep EEG for the measurement of seizure activity (1428). According to a systematic review (637), better evidence is needed in this area before a clear conclusion can be drawn regarding the safety or effectiveness of melatonin in seizure disorder.
        • Systematic review: Brigo and Del Felice published a systematic review of four clinical trials (720;723;724;743) to assess the efficacy and safety of melatonin as add-on therapy for epilepsy (637). Articles assessing the effects of melatonin as add-on therapy for epilepsy that were published in any language were pooled from the Cochrane Epilepsy Group Specialized Register (May 2012), the Cochrane Central Register of Controlled Trials (CENTRAL Issue 4 of 12, The Cochrane Library 2012), and MEDLINE (1946 to April 2012). References in the articles were examined to identify additional relevant studies. In addition, conference proceedings and selected journals were hand-searched. Articles were selected if the methodology was described as being randomized and controlled. Doses used in children and adults in the included study ranged from 3-10mg daily. Duration of treatment at ranged from 2-4 weeks with a conditional extension of two months in one study. Discussion of standardization was lacking. A lack of significant treatment-related adverse effects was observed. Discussion of toxic effects, dropouts, and interactions was lacking. Primary outcomes were the reduction of seizure frequency by at least 50%, absence of seizures, and adverse events. A secondary outcome was quality of life (QOL) assessed by validated standard questionnaires. In one study, significant improvement in quality of life was lacking with melatonin treatment. In other studies, participants treated with melatonin lacked seizures; however, these participants were selected for the study because that had lacked seizures for at least six months prior to the study. Therefore the reviewers noted that the effect of melatonin on seizure frequency was unclear. According to the reviewers, sufficient evidence to conclude that melatonin reduced the frequency of seizures or affected quality of life in epilepsy was lacking. Limitations of this review included that fact that the included studies lacked sufficient information to make an assessment of seizure frequency, adverse events, or quality of life.
        • Evidence: Gupta et al. conducted a randomized, double-blind, placebo controlled trial to assess the efficacy of melatonin for improving quality of life in pediatric patients with epilepsy (N=31) (723). Subjects aged 3-12 years were included in this trial if they were diagnosed with epilepsy (with generalized or partial seizures based on the International Classification of Epileptic Seizures), lacked seizures within six months of study initiation, and were taking 10mg/kg sodium valproate (VPA) daily for six months before study initiation with blood levels of 75-125mcg/mL. Subjects were excluded from this trial if they had prior psychiatric or neurological disease, as well as if they had a disorder of the blood, heart, liver, kidney, or thyroid. Subjects were randomly assigned using a computer generator to receive either two (if age<9 years or weight<30kg) or three tablets (if age>9 years or weight>30kg) containing 3mg rapid release melatonin or an identical placebo daily for 28-32 days. The melatonin used in this study and the identical placebo were provided by Aristo Pharmaceuticals Ltd., Mumbai, India. The investigators reported a lack of adverse effects that were severe enough to cause dropouts. Further information about adverse effects was lacking. Information about toxic effects was lacking. A single subject in the control group was excluded from analysis due to loss to follow-up. Information about interactions was lacking. Outcome measures included quality of life as measured by the Quality of Life in Childhood Epilepsy (QOLCE) instrument, which included subscales of quality of life (QOL), Energy/Fatigue, Physical Restrictions, Attention/Concentration, Other Cognitive Processes, Language, Depression, Control/Helplessness, Anxiety, Social interactions, Stigma, Self-esteem, Behavior, Social activities, and General Health. A lack of significant difference was observed between groups for all outcome measures. However, within the treatment group, improvements were reported compared to baseline for the QOLCE subscales of Memory (p=0.05), Attention (p=0.001), Language (p=0.004), Other Cognitive Processes (from 4.0 to 4.7, p=0.05), Anxiety (p=0.02), and Behavior (median score 3.2 to 3.3, p=0.004). A strength of this study was the randomization and blinding while a limitation was the subjective nature of the outcome measure.
        • Studies of lesser methodological rigor (not included in the Evidence Table): Elkhayat et al. conducted a prospective study of 37 patients (N=23 with intractable epilepsy), to investigate melatonin levels, the impact of melatonin levels on sleeping patterns, and the clinical effects of melatonin therapy (380). Patients included in the study had intractable epilepsy that was uncontrolled despite treatment with antiepileptic drugs for at least two years. Patients were excluded if they had progressive neurologic or systemic disease, inborn errors of metabolism, or visual impairment. Patients received 1.5mg of oral melatonin tablets half an hour before bedtime for three months, in addition to their antiepileptic drug treatment. Patients were initiated on a dose of 3mg, but two patients experienced increased seizure frequency, leading to a decrease in dose to 1.5mg. Patients were reassessed at the end of the treatment period in terms of the frequency and severity of seizures, sleep pattern assessment, serum melatonin levels, and interictal EEG recording. Two patients experienced minor adverse effects, including an unexplained persistent headache, papular skin rash, and abdominal pain. Three patients discontinued melatonin due to increased seizure frequency. The Chalfont Seizure Severity Scale determined severity of patients' seizures and measures components of seizures, such as falling to the ground, dropping of a held object, incontinence, and injury. Children's Sleep Habits Questionnaire and the Epworth Sleepiness Scale (pediatric version) were used to assess psychometric sleep. Patients were instructed to keep sleep diaries, which were regularly checked at follow-up visits. A significant improvement in bedtime resistance, sleep duration, sleep latency, frequent nocturnal arousals, sleep walking, excessive daytime sleepiness, and Epworth sleepiness score was observed in patients with intractable seizures receiving melatonin (as compared to values prior to therapy). Nocturnal enuresis, forcible teeth grinding, and sleep apnea were also improved. A significant effect on sleep talking,