Abstracts

Rationale: Obstructive sleep apnea (OSA) is common, and demonstrates greater prevalence in people with epilepsy (PWE). Untreated OSA is associated with elevated morbidity and mortality. Experimental evidence suggests that reduced serotonin may be involved in the pathophysiology of OSA and seizures. However, it remains unclear whether increasing serotonin levels ameliorates OSA, and whether PWE, who already demonstrate serotonergic deficiency at baseline, would respond similarly to those without epilepsy. The goal of this study was to determine whether serotonin reuptake inhibitors (SRI) affect OSA severity, and whether a different response to SRIs exists between PWE and people without epilepsy (PWO). Methods: This was a retrospective chart review at an academic center of subjects ≥18 years of age evaluated for OSA between 2011-2016. Epilepsy and OSA were diagnosed as per the 2014 International League Against Epilepsy and International Classification of Sleep Disorders 3 criteria, respectively. Subjects were dichotomized by the presence (+SRI) or absence (-SRI) of a SRI. Outcome was measured by OSA severity (mild: AHI 5-14; moderate-to-severe: AHI ≥15; severe: AHI ≥ 30). Baseline characteristics included age, gender, body mass index, epilepsy, and medical and psychiatric comorbidities. Pearson’s Χ2, likelihood ratios and t-tests were used as appropriate. Logistic and log linear regression analysis controlled for covariates, and p < 0.05 was considered significant. Results: The total cohort consisted of 98 subjects. At baseline, hypertension (OR: 3.86, p=0.05) and depression (OR: 6.61, p=0.01) were more common in +SRI. Among the 45 PWE and 53 PWO, depression was more common in PWO (OR: 5.80, p=0.02). In PWE+SRI, depression was more common (OR: 5.22, p=0.02). In PWO+SRI, hypertension was more common (OR: 6.47, p=0.01). Compared to -SRI, +SRI was less likely to have severe OSA after unadjusted, univariate and multivariate analysis, controlling for significant baseline covariates. When epilepsy was added to each model, these associations remained. Comparing PWE to PWO, there was a difference in the relationship between +SRI and severe OSA, where PWE+SRI were less likely to have severe OSA (Χ2: 0.13, p=0.01). Likelihood ratios demonstrated significance for the unadjusted interaction of severe OSA+SRI and epilepsy (G2: 0.12, p=0.04), and after adjustment for depression (G2: 0.04, p=0.004), hypertension (G2: 0.04, p=0.01) and depression and hypertension (G2: 0.02, p=0.001). As a group, PWE+SRI were less likely to have severe OSA, with and without adjustment. In PWO+SRI, moderate-to-severe OSA and severe OSA were less likely only after adjustment for hypertension. Independently, epilepsy was associated with decreased OSA severity. Conclusions: Serotonin reuptake inhibitors are associated with reduced OSA severity, and demonstrate greater reductions in PWE. This suggests that SRIs may ameliorate OSA severity and have differential effects in PWE compared to PWO. Further investigation of SRIs as an alternative treatment for OSA is warranted, particularly in PWE, who may dually benefit from improvement of seizure-related outcomes. Funding: None.