Letter to the Editor: Sleep Disturbances in Selective Serotonin Reuptake Inhibitor-Treated Youth with Anxiety Disorders and Obsessive Compulsive Disorder—A Bayesian Hierarchical Modeling Meta-Analysis

To the Editor:

More than half of youth with anxiety and obsessive compulsive disorders report sleep disturbances (Alfano et al. 2007; Alfano and Kim 2011). Furthermore, the most effective psychopharmacologic interventions for these disorders—selective serotonin reuptake inhibitors (SSRIs)—may themselves increase insomnia as well as tiredness/sedation (Mills and Strawn 2020). However, the risk of sleep-related problems for specific SSRIs has not been examined meta-analytically. We sought to examine sleep-related adverse events (AEs) across specific SSRIs in clinical trials of youth with anxiety disorders as well as obsessive compulsive disorder (OCD) using Bayesian hierarchical modeling (BHM).

The search strategy for the included double-blind placebo-controlled studies has been previously described (Mills and Strawn 2020). Studies were included if they were (1) prospective, (2) randomized, (3) parallel-group, (4) placebo-controlled trials that evaluated SSRIs in social, generalized, and/or separation anxiety disorder or OCD in children or adolescents, and (5) systematically captured sleep-related AEs. Youth with OCD were included because although OCD and anxiety disorders are diagnostically distinct, youth with OCD experience sleep-related problems at similarly high rates to youth with anxiety disorders, and clinicians should consider this when treating such patients (Storch et al. 2008). Trials of SSRIs in pediatric patients with major depressive disorder were not included because these trials vary considerably in methodologic and depressive pathology may influence specific symptoms. A BHM was applied to evaluate the risk of sleep-related AEs and to allow for heterogeneity across studies. In BHMs, the degree of heterogeneity across studies is assumed to be unknown and estimated by the model. The mean probability (risk) of each AE, standard deviations, and credible intervals (CrI) were computed from the simulated posterior sample representing posterior odds against the null hypothesis of no difference and a Bayesian posterior p-value.

Data were included from studies involving five SSRIs: fluoxetine (κ = 3), fluvoxamine (κ = 3), paroxetine (κ = 1), sertraline (κ = 2), and escitalopram (κ = 1). The characteristics of the included studies have been previously described (Mills and Strawn 2020). Compared with placebo, sertraline was associated with a greater likelihood of insomnia (odds ratio [OR]: 2.72, CrI: 1.15–5.78, p = 0.02). Fluoxetine was associated with more treatment-emergent sedation (OR: 5.12, CrI: 2.01–11.30, p = 0.01). No significant sedation differences were observed between placebo and fluvoxamine, escitalopram, paroxetine, or sertraline. Also, no significant insomnia differences were observed between placebo and fluoxetine, paroxetine, fluvoxamine, or escitalopram (Fig. 1).

FIG. 1.

Somnolence/tiredness and insomnia in children and adolescents with anxiety and obsessive compulsive disorders treated with SSRIs. SSRI, selective serotonin reuptake inhibitor.

These findings suggest that sleep-related side effects are associated with specific SSRIs and have important clinical implications. Youth with insomnia typically report daytime fatigue and sedation, raising the possibility that these AEs are related. In addition, the time of medication administration and the risk of sleep-related AEs may relate to diurnal variation in the activity of metabolizing enzymes CYP2D6 and CYP2C19 (Ruben et al. 2018) and general pharmacogenetic variation that affects metabolizing activity, which may result in higher blood levels of SSRIs and greater concentration-related AEs (Ramsey et al. 2020). Finally, it is important to underscore the effectiveness of evidence-based psychotherapies on sleep in anxious youth, such as cognitive-behavioral therapies that include prolonged exposure, relaxation/meditation, and cognitive restructuring. Cognitive-behavioral therapies are effective and have few side effects. When these psychological interventions are combined with SSRIs, outcomes are enhanced. However, in cases wherein psychotherapy is not the modality of choice, clinicians may do well to consider the impact of SSRI choice on sleep.

Disclosures

Dr. Strawn has received research support from the National Institutes of Health (NIMH/NIEHS) as well as Allergan and Neuronetics. He has received material support from Assurex Health and receives royalties from the publication of two texts (Springer) and serves as an author for UpToDate and an associate editor for Current Psychiatry. Dr. Mills receives research support from the Yung Family Foundation. Dr. Ramsey has received research support from the National Institutes of Health (NICHD) and BTG, International Ltd. Dr. Skoch reports no biomedical conflicts of interest.