High prevalence of low arousal threshold endotype in obstructive sleep apnea with multiple sclerosis.

PURPOSE: We sought to estimate the prevalence of low arousal threshold (LAT) in individuals with multiple sclerosis (MS).

METHODS

We conducted a retrospective non-matched study of individuals with documented MS as well as non-MS controls who were diagnosed with obstructive sleep apnea (OSA). Demographic, clinical characteristics, and polysomnographic measurements were extracted from the medical records. LAT was estimated from overnight polysomnography and defined as 2 or more of the following conditions: apnea-hypopnea index (AHI) < 30 events/hour, minimum oxygen saturation > 82.5%, or frequency of hypopnea events > 58.3%. A multivariate analysis was conducted to assess predictors of LAT.

RESULTS

LAT was identified in 87.5% of 64 patients with MS and OSA, which was significantly higher than OSA controls (71.4%). On univariate analysis, LAT was significantly associated with MS status, sex, and BMI. With logistic regression, only sex (OR 3.2, 95% CI 1.2-9.7, p = 0.03) and BMI (OR 0.87, 95% CI 0.80-0.94, p = 0.0007) were associated with LAT, although MS status trended towards significance (OR 2.64, 95% CI 0.91-8.51, p = 0.084). Without sex in analysis, MS status was significantly associated with LAT (OR 3.34, 95% CI 1.21-10.4, p = 0.026). Stratifying by OSA severity indicated that differences in LAT were even more pronounced in severe OSA, with 55.5% MS versus 20.8% non-MS exhibiting LAT.

CONCLUSION

The estimated prevalence of LAT endotype is high in MS-OSA patients, significantly exceeding rates in non-MS controls, but female sex and lower BMI are the dominant predictors. Because sex strongly confounds the relationship between MS and LAT, whether MS independently contributes to LAT beyond its demographic profile requires larger, sex-matched cohorts.