Liquid Melatonin via Gastrostomy as an Alternative Hypnotic in a Child with Pervasive Developmental Disorder

To the Editor:

S leep disturbances including delayed onset of sleep, frequent night awakenings, and short duration of sleep are common among children and young adults with pervasive developmental disorders (PDDs). The generic category of PDDs includes autistic disorder, PDD not otherwise specified, and Asperger's disorder. Up to 80% of such patients are found to suffer from chronic sleep problems (Piazza et al. 1996). Although the majority of these children have long-standing, recognized disturbance in sleep, treatment of such is often neglected (Stores 1999). The impact of these sleep difficulties spreads beyond the individual patient, as parents and siblings of the affected children are often repeatedly awakened during the night and become sleep deprived as a result. Thus, sleep disturbance in children with PDDs affects the health, social, emotional, and economic functioning of the entire family (Chilcott and Shapiro 1996).

The treatment of sleep disturbance in children with PDDs involves behavioral therapies that include chronotherapy, bedtime fading, the use of extinction to decrease associated nighttime disruptive behaviors, gradual distancing for children who will not sleep in their own beds, and sleep-wake scheduling (see Didden 2001 for a review on behavioral treatments). Behavioral methods can be helpful in the treatment of sleep disturbance in children with PDDs (Richdale et al. 1999) and are the recommended approach to initial intervention. However, these methods alone frequently do not lead to normalization of sleep patterns, and adherence to these regimens is often poor, especially in families already overwhelmed by caring for a child with a prominent PDD. Modern pharmacotherapy, therefore, plays an increasingly important role in conjunction with behavioral strategies.

Benzodiazepines are among the most commonly used medications to regulate sleep. However, in patients with PDDs, studies found that behavioral side effects occurred in 13% of 446 individuals with mental retardation who were prescribed benzodiazepines; these include daytime drowsiness, sedation, drooling, and hyperactivity (Kalachnik 2002), as well as emotional and behavioral disinhibition and lability (Kalachnik et al. 2003). Chloral hydrate, which can effectively induce sleep, is ineffective in maintaining sleep throughout the night and is not intended for long-term use owing to the possibility of hepatic tumorigenicity (Salmon et al. 1995). Sedating antihistamines are frequently used; however, these medications can induce paradoxical excitation with worsening of sleeplessness, particularly in young people with atypical central nervous system development (Turk 2003).

Recent literature has suggested the potential value of melatonin for sleep induction in children. There is evidence that chronic sleep disturbance in children with PDDs is associated with disturbed melatonin secretion (Myamoto et al. 1999). In a recent double-blind, randomized, controlled crossover trial of melatonin, 22 children with PDDs who had not responded to supported behavior management strategies received placebo versus 3 months of melatonin (maximum dose of 10 mg). This study assessed acute and chronic effects of melatonin treatment over a 3 month period. Melatonin significantly improved sleep latency (by an average 47 minutes) and total sleep (by average of 52 minutes) compared with placebo; however, there were no differences in the number of nighttime awakenings, and the side-effect profile was low and not significantly different between the two arms (Wright et al. 2011). This study confirmed the results of an earlier, smaller randomized control trial of melatonin used in children with PDDs and sleep disturbance (Garstang and Wallis 2006). The clinical benefit of melatonin in children with PDDs on sleep onset and maintenance has also been demonstrated in a large case series (n = 107) using melatonin at doses ranging from 0.75 to 6 mg given before bedtime. After initiation of melatonin, 25% of parents no longer reported sleep concerns for their child at follow-up visits, 60% reported improved sleep, and 13% continued to report sleep problems as a major concern (Andersen et al. 2008). Overall, melatonin appears to be a safe and well-tolerated treatment for sleep disturbance in children with PDDs (Andersen et al. 2008; Wright et al. 2011). These studies all report use of melatonin given orally in pill, tablet, or crushed form. However, many children with PDDs have difficulty swallowing pills. In addition, children with profound neurological disabilities often require a gastrostomy tube (G-Tube) for feeding and drug administration. Therefore, medications may require a liquid preparation for dispersion.

To our knowledge, there is no literature regarding the use of liquid melatonin dispensed orally or through feeding tubes in children and adolescents with PDDs. We report herein a case of a 5-year-old child with PDD and failure to thrive who therefore required a G-Tube for feeding and medications, and who had a chronic, persistent, disruptive sleep disturbance and severe behavioral problems.

Case Presentation

The patient was a 5-year-old Caucasian male who was admitted to an inpatient pediatric medical unit for 14 days treatment because of emesis and weight loss. He had a history of shaken baby syndrome with resultant developmental delays, seizures, and failure to thrive that required a G-Tube. In addition, he had a significant sleep disturbance since 3 years of age characterized by delayed onset, frequent awakenings, and prominent early morning awakening. Behavioral approaches including ignoring disruptive behaviors at night, and the use of sleep-wake scheduling with a consistent morning wake time were attempted without success at home with the aid of a behavioral specialist. According to the history, it took the patient 2–3 hours to fall asleep every night, he awoke up to four times per night, and took up to 30 minutes to fall back asleep. The patient often awoke for the day by 5 a.m.

During the first 7 days of hospitalization, tube feeding was initiated to stabilize the patient's nutrition and electrolytes. In the hospital, the patient was initially maintained on his outpatient medication regime of clonazepam, 0.5 mg po qHS for sleep, phenobarbital 30 mg BID for seizure control, and lansoprazole, 30 mg q daily. As reported by history and evidenced by nursing observation in the hospital, the patient took 2–3 hours to fall asleep after medication administration. He awoke daily by 5 a.m. and remained fully awake throughout the day. In addition, the patient manifested physical aggression toward caregivers and self-injurious behaviors. A psychiatry consultation was requested on hospital day 2 to evaluate the patient's aggressive behaviors and sleep disturbance. The psychiatric consultant reasoned that since the clonazepam had been given orally and was followed by frequent vomiting, the amount retained was uncertain. Further, it was suspected, given the history, that clonazepam may have exacerbated the aggressive behaviors. For these reasons, it was recommended that clonazepam be discontinued. Since the patient had an underlying seizure disorder, and as clonazepam withdrawal can lower seizure threshold, clonazepam was weaned (hospital days 3–7) to minimize the potential seizure risk.

The aggressive behaviors were somewhat improved with the discontinuation of clonazepam, but problematic behaviors remained. Therefore, risperidone M-tabs, 0.5 mg qHS, was started on day 7 of the hospitalization, and the aggressive behavior improved further. However, despite treatment with a steady dose of risperidone, 0.5 mg qHS for 5 days, the patient's sleep remained problematic and was a primary concern for the family and providers. Therefore, on hospital day 12, a regimen of liquid melatonin, 2 mg qHS, was administered via the patient's G-Tube. Melatonin was chosen because of its reported beneficial use in children with PDDs. The liquid form of melatonin was chosen over crushed preparation to avoid clogging in the G-Tube. Melatonin was given without adverse effects, and the patient's response to the treatment was good over his last two nights in the hospital. Staff and mother reported that he slept continually from 10 p.m. to 7 a.m. while receiving melatonin.

On follow-up, the patient's caregiver reported that the patient had pulled out his G-Tube 2 days after discharge, and melatonin was, therefore, discontinued. The patient's sleep disturbance immediately recurred. At last report, the outpatient plan was to restart melatonin after the tube was replaced. In addition, the patient was to be referred for a sleep study to further evaluate the sleep disturbance.

Discussion

Melatonin has been released for public use by the Food and Drug Administration and is available over the counter nationwide. It has advantages over many other medications used in the treatment of sleep problems in children with developmental disabilities. Melatonin has a very low side-effect profile. Unlike antihistamines, it causes remarkably less drowsiness and has a lower risk of confusion. Unlike benzodiazepines, melatonin has fewer side effects in the short term, and there are no reports of tolerance, physical dependence, or a withdrawal phenomenon with extended use. Although there are data supporting oral melatonin for sleep disturbances in children with PDDs (Dodge and Wilson 2001; Andersen et al. 2008; Wright et al. 2011), there is no published report on the use of the liquid formulation dispensed via a G-Tube despite the fact that many children with developmental disabilities have neurodevelopmental delays, require a G-Tube, and have sleep disturbances (Smits et al. 2003).

Our patient had a limited ability to swallow oral liquid preparations and vomited if he was given large bolus volumes through the feeding tube. There are no clear guidelines for crushing melatonin tablets into a liquid dispersion; however, there are recommendations that extended release tablets of melatonin should not be crushed and dispersed. Liquid melatonin was, therefore, a palatable option in our case, because it has a small volume that decreased the risk of vomiting and does not run the risk of clogging the feeding tube. Moreover, as reported by Andersen et al. (2008), there are no increases in seizure activity after starting melatonin in children with pre-existing epilepsy and no new-onset seizures, making this a safer choice than many other medications.

Recommended doses of melatonin in preschool children range from 1 to 3 mg nightly (Gleason et al. 2007). Doses in school-age children range from 3 to 6 mg nightly, with the lower dose used for lower weight children (Smits et al. 2003). Dodge and Wilson (2001) found that children with PDDs may require higher doses of melatonin to affect sleep than those used in children with attention-deficit/hyperactivity disorder. One double-blind, placebo-controlled trial of melatonin in children with developmental delays and sleep disturbance did not show a beneficial effect on sleep with melatonin in the lower dose range of 0.5 to 1 mg on (Dodge and Wilson 2001). Given the above data, we chose to initiate melatonin at 2 mg qHS in our patient with a target dose of 5 mg if needed. Our patient, however, had a positive response to the initial dose.

Limitations are inherent in case reports and, therefore, preclude generalization. One limitation in this case is the short trial duration. Further, changes in other psychotropic medications and the use of phenobarbital for seizure management confound our ability to determine solely the impact of liquid melatonin on sleep and behavior. Phenobarbitol had been prescribed for seizure prevention in our patient and may have had an activating effect on his behavior. Therefore, a consultation with neurology to explore alternative antiseizure medications would have been a prudent first step. Finally, a sleep study to determine other potentially treatable causes for the sleep disturbance (e.g., nighttime desaturation) might have been useful.

Despite these limitations, however, the facts in this case speak to the potential value of liquid melatonin as a viable alternative for the treatment of severe insomnia in children with PDDs who have a G-Tube.

Conclusions

Practitioners should consider liquid melatonin as an alternative choice or supplement to the medications for pediatric insomnia, especially when other medications for sleep have impairing side effects or are contraindicated. Referring to the case just described, we further suggest that the use of liquid melatonin be considered in children with sleep disturbance and who have an inability to swallow and have feeding tubes. Liquid melatonin should be considered as a reasonable alternative or augmentation to benzodiazepines and sedating antihistamines in such cases. Given the limited information on the use of liquid melatonin administered through a feeding tube, we recommend administering liquid melatonin separately from other medications, as this proved an effective means of administration in the case reported herein.