Introduction
Sleep disorders in childhood and adolescence are regarded as a common manifestation of symptoms of a disorder, mostly transitory in nature and in many cases caused by unsatisfactory sleep hygiene or maladjusted parent-child interaction during the falling asleep and sleeping through the night process. Furthermore, sleep disorders could exhibit comorbid symptoms with manifestations of psychiatric and neurological diseases [16, 17]. In these cases, they are often chronic and partially also serious in nature. In most cases, during consultation behaviorial therapeutic measues are indicated and are also sufficient. With manifestations of chronic disorders, medicinal measures play an important role [45]. Thus far, the use of an antihistamine, benzodiazepine or a neuroleptic can only be used with reservation or at least in the short term due to long-term side effects, the potential for dependency and substantial negative impacts on daytime alertness and memory functions [45]. With melatonin as an endogenous sleep-inducing hormone, for the first time a pharmacological treatment method essentially free of side effects could be offered for children. This paper summarizes the current, however still relatively narrow-based findings. The literature search is based on Medline-Search, in which substantial papers have been stored since 1985. Due to the still very provisional study status however, this could not consider exclusively randomized studies.
Physiology and Pathophysiologiy of Melaton in Secretion
Melatonin is a so-called indolamine and is produced in the pineal gland. Melatonin is synthesized from tryptophan, which for its part is converted into serotonin and then is enzymatically transformed into melatonin [31]. Melatonin synthesis is initiated by the connection of noradrenaline to adrenergic B-1 receptors. Essentially, melatonin synthesis depends on the availability of tryptophan, which is dependent on nutritional factors, as well as on medicinal influences. Fluvoxamin, a serotonin reuptake inhibitor, increases the amplitude and duration of the melatonin plasma peak [57].
Melatonin plays a significant role in the synchronization of circadian rhythms [2], starting from the second half of the first year of life, in particular the sleep/wake rhythm. Melatonin secretion reaches its high point at the age of 3-6 years old and then decreases by adulthood by around 80% [55].
Influences of sex hormons on secretion, in particular during puberty, are not known [66]. Multiple effects are attributed to the hormone: It works to facilitate falling asleep on the one hand and on the other it is a circadian pacesetter and acts as a phase shifter. Both effects are used therapeutically in sleep medicine. These also include antioxidant effects, which can be used partially therapeutically with sucess in neonatology for septic and asphytic newborns [19,23]. Furthermore, effects in the oncology area are described as inhibitory, but also partly as a having a proliferative influence on tumor growth [4] as well as immunomodulatory effects[59]. Also, melatonin is inolved in endogenous body temperature regulation [7]. Arendt and Skene [2] attibute a chronobiotic effect to the hormone, since apart from influencing the sleep/wake rhythm, it also influences other vital parameters such as body temperature and cortisol secretion.
The endogenous 24-hour melatonin profile is a reliable marker for circadian phase position. In order to examine this, it is not necessary to record the complete daily profile. It is sufficient to determine the melatonin increase in the evening with darkened light. This happens before falling asleep [35]. There is a high degree of dependence of hormone production on light conditions. Exposure to bright light during the night leads to a rapid decrease in melatonin secretion [34] and can lead in this way to a phase shift of the sleep/wake rhythm [11].
The inter-individual variation in regard to the volume of secreted melatonin is considerable and is either genetically-caused [68] or dependent on environmental factors, which can already have an intrauterine impact [30].
The melatonin secretion rhythm is generated by a endogenous clock, which is localized in the nucleus suprachaismatieus of the hypothalamus. The day/night cycle is the crucial exogenous interval timer of melatonin secretion. The most important neurotransmitter, which regulates melatonin secretion in the pineal gland, is noradrenaline, which is released at night in response to stimulating signals in the nucleus suprachiasmaticus. These assumptions are supported in human trials [6]. B-1-receptor blockers and the Alpha-2 receptor blocker clonidin as well as alpha-methyl-para tyrosine, which reduces the pre-synaptic catecholamine concentration, suppress the nocturnal melatonin secretion. Conversely, melatonin secretion is enhanced by drugs, which increase the synaptic catecholamine availability, such as MAO inhibitors or tricyclic antidepressives. In addition, animal testing indicates that other neurotransmitters such as GABA, neuro-peptide Y dopamine and glutamate suppress melatonin secretion. To what extent these findings also apply to humans is still unclear [9].
A correlation of noctural melatonin secretion to sleep stages could thus far not be determined. Only a few studies are available on the effects of exogenous melatonin administration on sleep architecture. Overall, only a little influence seems to exist unlike, for example with conventional hypnotics [72]. In two studies it was reported that Melatonin has no effect on sleep stages 2, 3 and with adults patients middle-aged and older with Insomnia [20,25]. In other studies on patients with and without sleep disorders there was found tendentially a prolongation of sleep stage 2 as well as a shortening of sleep stages 3 and 4. During the non-REM sleep stage 1 and the REM sleep stage they remain unimpaired [69]. In a subsequent study no effects on sleep architecture could be determined [71]. Rajaratnam et al. [50] found likewise no influence on REM sleep by melatonin administration with, in contrast, a decrease of sleep stage 3 and an increase in sleep stage 2. Brezinski et al. [5] examined the influence of melatonin on the sleep structure and found only small changes. A reduction in sleep latency by 4 minutes showed up as well as an increase in overall sleep by 12.8 minutes and an improvement in sleep efficiency by 2.2%.
Kunz [32] found, on the other hand, an increase in REM sleep after melatonin administration and have applied it already for treatment of a REM sleep condition disorder. Salti et al. [54] describe a connection between ultradian fluctuations of melatonin concentration and the REM sleep phases.
In the neurological area, neurodegenerative diseases such as Parkinson’s or Alzheimer’s are accompanied by a partially dramatically decreased endogenous melatonin secretion. With both diseases serious problems falling asleep are present [72].
Sleeplessness without concomitant disease appears likewise to be partly connected with decreased melatonin seretion [24].
For psychiatric illnesses, there are a few findings on changes in melatonin secretion. Thus, with inpatient treatment of depressive patients a decrease in melatonin release was found in comparison to the healthy control persons [8, 33]. Lewy et al. [33] reported high melatonin concentrations with bipolar patients in manic phases as opposed to depressive phases and assumed that the amplitude of melatonin secretion indicated illness-induced changes of the noradrenergic function.
Recent studies show, however, no changes in melatonin secretion in relation to melatonin peak as well as melatonin increase [63]. Heterogeneous results are also available from other psychiatric illnesses such as Schizophrenia and Anorexia nervosa [3].
Melatonin in the Treatment of Sleep Disorders
In the clinically therapeutic area, the administration of melatonin causes a reduction in the time to fall asleep [66]. A precisely time-controlled administration before the individual’s usual bed time can therapeutically positively initiate an earlier endogenous melatonin distribution and thus e.g. cause the readjustment of the sleep phase shifts [12].
The majority of published studies were performed in selective patient groups during childhood and under a few systematic conditions, to the extent that their significance and/or validity is limited. Jan and Freeman [29] summarize in their overview the significant treatment results.
The experiences are mainly with manifestations of pediatric neurological illneses [62]. Zhadanova et al. [70] reported, for example, on the results of melatonin treatment with 13 children with Angelman Syndrome and comorbid present sleep disorders. The main complaint picture was one of delayed falling asleep times, frequent nightly waking episodes with increased motor activity as well as early morning awakening. A six-day treatment with Melatonin (0.3 mg/kg per body weight) led to an actigraphically proven, highly-significant decrease in nocturnal movments and to an increase in total sleep time. A systematic examination of the possible influence on falling asleep time was not performed in this study.
McArthur & Budden [37] treated 9 patients with Rett Syndrome with melatonin in a double-blind, placebo-controlled crossover study. Before treatment, the majority of children exhibited actigraphically increased prolonged falling asleep times, reduced total sleep times as well as noctural sleep interruptions. Under melatonin treatment (0.1-0.2 mg/kg per body weight) over three weeks the falling asleep time was significantly reduced, not however the other variables named in comparison to the placebo. Coppola et al. [10] in a placebo-controlled study obtained through melatonin administration of an average of 3-9 mg with mentally retarded children and adolescents with seizures and sleep disorders a significant reduction in the falling asleep times in comparison with the placebo, while on the other hand there was no specific influence on nocturnal awakening. Pillar and Mitarbeiter [48] actigraphically examined the sleep behavior of psychomotor retarded children of school age under average melatonin medication of 3 mg. Under this treatment, nocturnal sleep time was prolonged by an average of approximately 1.5 hours and sleep efficiency was increased from 69% to 88%. At the same time, the daily sleep of 3.2 hours regressed to 1.7 hours so that the total sleep time remained stable over 24 hours. This was hereby in addition accompanied by a substantial improvement in the quality of life of the parents as well as the children. Jan & O’Donnell [27] demonstrated in their study mainly the positive effect of melatonin treament on children with sleep disorders in their behavior during the day. Okawa et al. [43] reported on the positive effects of melatonin treatment on adolescents with sleep disorders in their school attendance. Espezel et al. [15] likewise saw positive effects of Melatonin treatment on children and adolescents with vision problems.
The consequences of improved sleep were favorable for mood and alertness. In one earlier study of Jan et al. [26] with multiple-handicapped children with sleep disorders positive effects on memory and learning and social behavior were reported. In a recent study with craniopharyngeoma patients, silimar connections could be found between a reduced nocturnal melatonin secretion and intensified daytime fatigue [41].
In the pediatric neuropsychiatric area, there are promising approaches for melatonin treatment with vigilance-impaired and hyperkinetic children with sleep disorders if these are induced medicinally by methylphenidate. In particular, the prolonged falling asleep times with the children could be significantly reduced and the effects continued beyond the end of treatment [61, 73].
Van der Heijden et al. [65] treated 105 children ages 6-12 years old, who were afflicted by ADHD as well as a chronic falling asleep disorder, for 4 weeks within the scope of a double-blind, placebo-controlled study with melatonin. The falling asleep time decreased significantly under the verum condition as opposed to the placebo with a simultaneously significant pre-shifting of the evening endogenous melatonin secretion. In addition, the total sleep time increased significantly. Surprisingly however, there was no signficant change in the problem behaviors and cognitive dimensions.
Garstang & Wallis [21] performed a randomized, placebo-controlled, double-blind treatment study with melatonin with 11 autistic children. In the verum group there was a severe decrease in the falling asleep latencies, a significant decrease in the nocturnal waking episodes as well as a significant increase in the sleep time in comparison to the untreated condition and in comparison to the placebo. Gianotti et al. [22] examined the long-term efficacy of melatonin treatment with 25 autistic children from the ages of 2-9 years old. During the 6-month treatment, sleep problems raised with all subjects in a sleep questionnaire and a sleep log improved. After discontinuing melatonin administration, with 16 children there was a reexacerbation of the sleep problem at the starting level. The re-start of treatment however was also again effective. With children, who experienced a continuous improvement in their sleep problems due to melatonin treatment, the treatment effects could be confirmed in 12- and 24-month follow-ups. Significant side effects did not occur during the treatment.
Szeinberg et al. [60] describe positive effects of a long-term treatment over 6 months with 33 adolescents from the ages of 10-18 years old with sleep phase delay. The treatment was accompanied by an earlier sleep start and a longer total sleep duration. At the same time, there was a decrease in school difficulties. Serious side effects could not be determined.
Furthermore, reports exist on successful melatonin treatments with bipolar disorder [52], with 20 children with manifestations of various pediatric psychiatric disorders [38] as well as with Asperger Autism [44]. Smits et al. [58] finally proved sleep phase accelerated effects with 62 children with chronic idiopathic insomnia. Also their psychophysiological state of health was affected favorably.
Effect Profile and Side Effects
Pharmacokinetically, after approximately one ingestion of quick-acting, synthetic melatonin, maximum plasma concentration occurs, and then biphascially decreases again [47]. Jan et al. [27] therefore recommend ingestion approximately 20-30 minutes before going to bed. Great importance is attached however to an environmental structure promoting falling asleep, primarily darkened light conditions. The clinical effect of melatonin treatment starts independent of the dosage within a few days after the beginning of treatment. Dodge & Wilson [14] with their treatment collective with children with development delay initially determined a dosage of 5 mg as having a satisfactorily sleep-inducing effect. Paavonen et al. [44] argue that a two-week treatment attempt is sufficient to ensure that the effectiveness of a treatment can be assessed.
Tolerance of the treatment in the present, non-systematic studies in general is good and here also with prolonged treatment, no loss of effect and few or no side effects occur [40, 53]. A significant toxicity with overdosages has thus far not been observed, as well as no teratogenic effects [49]. There are still no long-term studies available on tolerance [67].
These overall positive experiences however contradict single reports on the negative effects of melatonin administration with children, such as the triggering of cerebral seizures [56, 59]. There are now also reports on an anti-convulsive effect [46]. With inappropriate dosing times, serious sleep disorders can also be triggered [39]. Also a rapid recurrence of the previous sleep disorder was reported after discontinuation of the medication [42]. Other temporary side effects may result as well from the treatment, for example disorientation, headaches, dizziness, nausea, tachycardia as well as pruritus [28, 44] and negative immunomodulatory effects on asthmatic illnesses [59]. Due to the immunomodulatory effect of melatonin, the National Sleep Foundation moreoever recommends that melatonin be used in any case with the presence of lymphoproliferative diseases [cited no. 45].
Table 1. Treatment Results with Melatonin in Childhood.
|
Study |
Illness Picture |
Complaint Picture |
Procedures |
Results |
|
|
Zhadanova et al. 1999 |
Angelman Syndrome |
< TST > Body movement in sleep > SOL |
- Actigraphy - 6 d melatonin, 0.3 mg/kg per body weight |
> TST < Body movement in sleep |
|
|
McArthur & Budden, 1998 |
Rett-Syndrome |
> SOL < TST > Sleep interruptions |
-Actigraphy < SOL - 3 weeks of melatonin - 0.1 -0.2mg/kg per body weight |
||
|
Coppola et al. 2004 |
Mentally-retarded children with cerebral seizures |
Various Sleep Disorders |
- Clinic treatment study - 3-9 mg melatonin |
< SOL |
|
|
Pillar et al. 2000 |
Psychomotor-retarded children |
Various Sleep Disorders |
- Actiraphy - 3 mg melatonin |
> TST > SEI |
|
|
Jan & O’Donnell, 1996 |
Non-retarded children |
Various Sleep Disorders |
- Clinical treatment study |
Effects on daytime behavior |
|
|
Espezel et al. 1996 |
Vision-impaired children |
Various Sleep Disorders |
- Clinical treatment study |
Effects on daytime Behavior |
|
|
Jan et al. 1999 |
Multiply-handicapped children |
Various Sleep Disorders |
- Clinical treatment study |
Effects on memory, learning and social behavior |
|
|
Zotter et al. 2001 Tion Plan G 2003 |
ADHS + Comorbidities |
Sleep disorders caused by Methylphenidate |
- Case Report |
>SOL |
|
TST Total Sleep Time;
SOL Sleep Onset Latency;
SEI Sleep Efficiency Index
The influence of sexual maturity and sex hormones was proven in animal testing [13]. Concerning this aspect, the available study results indicate for example that the sexual maturity changes between Tanner stages 1-5 are accompanied by a significant decrease in melatonin secretion.
Also the influence of the female menstrual cycle has been described. During Menopause there occurs a further decrease in melatonin secretion, which could possible correlate to changes in the gonadotropin concentrations [51]. A three-month administration of melatonin, for example, with some young, healthy volunteers resulted in massive reduction in spermatogenesis, which was also still detected 6 weeks after discontinuing the drug [36]. Application of the drug with children and adolescents should also again be critically reflected against the background of this still very uncertain study.
Table 1 gives an overview of the results of the significant treatment studies performed thus far.
Discussion
The results thus far available on the treatment of childhood sleep disorders with melatonin are unfortunately still very much incomplete and must therefore suggest an urgent need for further studies, especially treatment studies.
Therefore, neither a binding dosage recommendation can be made for children, nor is there sufficient knowledge on the short or long-term side effects or intolerances that can be expected.
Overall, the available findings are characterized by small, selected patient numbers, diagnostic uncertainty and the summary of different sleep disorders in a study.
Furthermore, it appears problematic that thus far only children with central nervous system injuries more or less were treated so that the opportunity to compare sleep disorders in children without brain injury is limited. In addition, the applied objective measurement criteria are considered by many as too non-specific, both in terms of subjective sleep and vigilance variables as well as objective measured parameters. There are, for example, almost no polysomnographic findings, which examined a relationship between a change in melatonin secretion and the sleep stage structure. Thus far, the methodical approach to actigraphic discharges is limited. Studies on the effect of successful treatment on daytime vigilance is likewise missing.
Also, there are no findings available on sleep disorders with children with manifestations of psychiatric disorders or with behavioral disorders without concomitant somatic findings. Questions on side effects or the development of tolerance must also remain open. Long-term treatment results and a review of the treatment process are also absent. Future systematic studies with precisely definied sleep disorders and somatic comorbidities must take the cited criticims into consideration in order to empirically substantiate and justify treatment with melatonin.
Van den Heuvel et al. [64] dampen the overall positive expectations that are associated with the sleep-inducing effect of melatonin. In their critical analysis, they argue that a therapeutic effect can only be expected with low serum melatonin concentrations, administration on the day no sleep-inducing effect follows and the effects are in fact motivation-dependent or are dependent on the body position as well as occur with some limitation with young people and women.
Buscemi et al. [6] also see the effects of treatment as rather limited. In a meta-analysis of 15 randomized, controlled studies, for example, no significant changes could be measured in the falling asleep latencies.
In summary, the currently available results on the use of melatonin for the treatment of childhood sleep disorders, despite the still preliminary findings, are naturally of high interest. The precise mechanism is still unclear, with high probability however both that it induces sleep as well as makes the sleep phases rhythmic. Accordingly, there is the indication of successful administration mainly with falling asleep problems and sleep phase shifts. Lower efficacy is expected with sleeping through the night problems as well as waking problems. For most children with central nervous system injuries with partly serious, chronic sleep disorders, various behavior-related or pharmacological treatments can be carried out with rapid and significant treatment success without the occurrence of significant side effects. Even with long-term treatment a decrease in the effect does not seem to occur. It is recommended that dosage determination be made individually in accordance with the present state of knowledge and that a dosage range be set between 1 and 10 mg and be given just before falling asleep. Zhadanova identifies a physiological dose of < 1 mg as clinically effective [72].
Finally, it should be noted that drug treatment approaches are to continue to be understood as subordinate and temporary therapeutic approaches or with permanent use should be reserved primarily for severe neurological illnesses with comorbid sleep disorders [18].
