In general, medications are considered when other appropriate interventions have proven inadequate. There is very little evidence for using pharmacologic interventions for pediatric insomnia, so even if a medication is selected, behavioral approaches should remain a mainstay.4 Patients and caregivers should agree to specific short-term goals ahead of time when using sleep medicine, given the limited effectiveness and recommended short duration of use. Many medications change sleep architecture, and none have been clearly shown to sustainably improve sleep quality or quantity or reduce daytime symptoms of insomnia.
Prescribing guidelines for insomnia suggest selecting an agent matched to the symptoms and relevant to any comorbidities. Melatonin may be most helpful in shifting the sleep phase rather than for direct hypnotic effects; thus adolescents or patients with ADHD whose sleep schedule has naturally shifted later may benefit from a small dose of melatonin (1-3 mg) several hours before bedtime to prime their system. Beware that melatonin is not regulated by the Food and Drug Administration and animal studies have shown significant alterations of the gonadal hormone axis, although this has not been examined in human trials. Alpha-2 agonists – such as clonidine and guanfacine – may be helpful for sleep initiation, especially in populations with comorbid ADHD, aggression, or tics, where these medications might be otherwise indicated. Prazosin, an alpha-1 antagonist, has some limited evidence as a treatment for nightmares and PTSD symptoms, so it may be a good choice for children with trauma-related hypervigilance.
In patients with depression, low doses of trazodone (12.5-50 mg) or mirtazapine (7.5-15 mg) may be effective. Although short-acting benzodiazepines may be useful in the short-term, particularly for sleep-onset difficulties, they generally are not recommended because of the risks of abuse, diversion, withdrawal, cognitive side effects, disinhibition, development of tolerance, and contraindication with such comorbidities as sleep apnea. However, the benzodiazepine receptor agonists such as zaleplon, zolpidem, and eszopiclone, while lacking evidence in the pediatric population, may be worthwhile considerations as their varying half-lives allow for specificity in treating sleep-onset vs. sleep-maintenance problems. Caregivers should be warned about the potential for sleepwalking or other complex sleep-related behaviors with this class of medicines.
Avoid tricyclic antidepressants because of the potential for anticholinergic effects and cardiotoxicity. Atypical antipsychotics generally are not worth the risk of serious and rapid side effects associated with this class of medications, which include metabolic syndrome.