0485 Can Hypoxic Burden Predict Improvement in Sleepiness and Cognition in Response to CPAP Treatment in OSA Patients?

Abstract Introduction Obstructive sleep apnea (OSA) affects more than 10% of older Americans and is associated with future comorbid conditions. The current method for measuring OSA severity, the apnea-hypopnea index (AHI), is poorly correlated to symptom burden and clinical improvement. Hypoxic burden (HB), which quantifies and characterizes OSA-related hypoxemia, has been suggested as an alternative OSA measure. However, more research is needed to understand HB’s predictive ability. Thus, this study investigated whether HB could predict which OSA patients, prescribed CPAP, will have the greatest improvement in measures of sleepiness and cognition. Methods This is a secondary analysis from a trial that investigated the effects of CPAP versus sham treatment on sleepiness [Epworth Sleepiness Scale (ESS), psychomotor vigilance test (PVT)] and cognition [Buschke Selective Reminding Test (BSRT), Sustained Working Memory Test (SWMT), Cogscreen Pathfinder Number (PN)]. Using the SpO2 signal, HB was calculated as the area under the desaturation curve during all desaturation events (sustained >3% dip in SpO2). For categorical analyses, we classified “sleepy” and “not sleepy” in two different ways: ESS≥10 or < 10 and PVT lapses≥4 and < 4, respectively. Results Of the 426 (66% male) participants included, ESS and PVT significantly decreased after two months of treatment; 10.2±4.2 vs 8.0±4.2 (p< 0.001) and 2.5±1.6 vs 2.3±1.4 (p< 0.01), respectively. After correcting for BMI, age, sex, and AHI, baseline HB was associated with a change in sleepiness after treatment (∆ESS R2=0.04, p< 0.001; ∆PVT R2=0.01, p< 0.05). Moreover, having higher baseline HB increased odds for transitioning from “sleepy” to “non-sleepy” based on ESS (1.006 [1.001, 1.012]) or PVT (1.07 [1.003, 1.011]). HB was not associated with change in cognitive variables at follow-up (p>0.05), before correcting for any covariates. Conclusion HB had a significant yet weak effect on subjective (ESS) and objective (PVT) measures of sleepiness after correcting for covariates. HB did not significantly explain changes in any cognitive measure after two months of CPAP treatment. Whether this lack of an association is due to the absence of an overall improvement in cognition or the inability of HB to explain variability remains to be tested. Support (if any) NIH K25HL151912 NIH R01HL171813 NIH R21HL165320