0480 UTILITY OF ORDERING SPLIT STUDIES FOR DIAGNOSIS OF OBSTRUCTIVE SLEEP APNEA

Abstract Introduction: Traditionally, evaluation of obstructive sleep apnea (OSA) consists of two separate polysomnographies (PSG); a diagnostic study, followed by a CPAP titration. However, to reduce costs, many insurance provides mandate performance of split studies (SPL) in lieu of 2 separate studies. This study was done to ascertain the utility of performing SPL for diagnosis and treatment of OSA. Methods: We reviewed the laboratory records of 106 patients suspected of having OSA. Among these 51 (48.1%) were mandated to be SPL by the insurance provider. We used AASM criteria for adding CPAP to the study (apnea-hypopnea index [AHI] in the first 2 hours of sleep ≥ 40). Full in-lab PSGs were performed according to AASM standards. AHI was calculated both by AASM and CMS criteria. We calculated the sensitivity and specificity of AHI in the first 2 hours of sleep for predicting overall AHI ≥40, and ≥ 15. Results: Among 51 patients mandated for SPL, 15 (29.4%) were split. The remaining 91 patients had diagnostic PSGs. In these 91, we examined the sensitivity, specificity, positive and negative predictive values (PPV and NPV) of AHI ≥ 40 during the first 2 hours, for predicting overall AHI ≥ 40, and AHI ≥15 both using AASM and CMS criteria. For predicting overall AHI ≥40 (AASM), these were 63%, 87.5%, 41%, 95% respectively. For predicting overall AHI ≥40 (CMS), these were 40%, 96%, 57%, 93% respectively. For predicting overall AHI ≥ 15 (AASM), these were 26.5%, 100%, 100%, 36.5% respectively. For predicting overall AHI ≥ 15 (CMS) these were 29%, 98.3%, 90%, 72.8%, respectively. Conclusion: Conclusion: The minority of studies ordered as SPL actually end up being SPL according to AASM guidelines. For justification of CPAP (AHI ≥15), the sensitivity is low, but the specificity and PPV are high. Support (If Any): None