Variability of OSA severity measures on repeated testing

Abstract

Introduction: Obstructive sleep apnoea (OSA) is defined and graded by thresholds of obstructive respiratory events and cyclic oxygen desaturations per hour while asleep. The aim of this study was to assess night-to-night variability of the oxygen desaturation index (ODI ≥4% dips) in a sleep clinic population.

Methods: Adults referred for OSA evaluation who underwent two consecutive nocturnal pulse oximetries were eligible. Based on the mean ODI over two nights, participants were grouped into “no” (<5/h), “mild” (5-14.9/h), “moderate” (15-29.9/h), and “severe” (>30h) OSA. Outcomes were the change in ODI and coefficient of variation (CV) between repeated studies in different OSA severity categories, and changes in ODI ≥10/h, ≥15/h, and in severity. Data are reported as mean±SD.

Results: 623 of 900 included adults (58±13 years) had OSA (ODI 21.8±21.4/h). The change in ODI between nights was 6.7±7.6/h and increased with OSA severity (table 1). A CV of 27±3% indicated a high night-to-night-variability. Within the group of mild OSA, 27% changed between “no OSA” and “mild OSA”. A change in ODI ≥10/h and ≥15/h was found in up to 43% and 29%, respectively, and increased with severity as well. A severity category change was found in 35% with OSA and was most likely in moderate OSA.

Conclusion: The night-to-night variability in OSA severity, as measured by the ODI, is high and more than 1/3 of patients changed severity class within two nights. Sleep studies recorded during a single night need to be interpreted with caution and in conjunction with symptoms as part of a comprehensive sleep evaluation to allow informed treatment recommendations.