0518 Longitudinal Testing with a Home Sleep Apnea Test Shows Large Variability in Nightly Sleep Apnea Severity

Abstract Introduction Sleep apnea (SA) is traditionally diagnosed by a single-night sleep study, however research suggest that nightly variability in SA severity is common. The purpose of this study was to utilize a novel longitudinal home sleep apnea test (HSAT) to evaluate the extent of variability and identify the ideal number of tests for diagnostic accuracy. Methods This study retrospectively evaluated 149 adults across the U.S. that used a novel type-3 HSAT, capable of unlimited longitudinal testing. 1,184 tests were evaluated (Mean 7.9 tests/subject; SD 6.9). Subjects who reported using SA treatments were removed from analysis. Apnea/Hypopnea Index (AHI) was assessed to determine the presence and severity of SA. A subgroup of previously diagnosed individuals were assessed for accuracy of their original sleep study. Results On repeat testing, 69.7% of subjects had AHI scores that varied in severity. 51% had AHIs within the normal ( 5) and abnormal range (5). The average AHI point difference between a subject's best and worst test was 12.4 (SD 12.1). Of the subjects that were positive for SA, 52% had false negatives on their first test. An average of 3.2 (SD 1.6) test were completed before obtaining a positive AHI, and 3.6 (SD 2.2) tests before arriving at the highest recorded severity. 26.1% of previously diagnosed subjects had HSAT-results that were more severe than their original diagnosis. Of the subjects that reported having mild SA, 11% had HSAT scores in the severe range. Of the subjects that reported having moderate SA, 38% HSAT scores in the severe range. Factors that were found to correlate with AHI variation were variation in time spent in supine position and time spent in REM. Conclusion Longitudinal testing with an HSAT confirmed high night-to-night variability of AHI in the majority of subjects. Subjects with mild to moderate SA were at high risk for misdiagnosis or misclassification if diagnosis is based on single or double night sleep studies. This study emphasized the need to standardize longitudinal testing and suggests that 3 tests is ideal for diagnostic accuracy. Support (if any)