Trends in perioperative practice and resource utilization in patients with obstructive sleep apnea undergoing joint arthroplasty

BACKGROUND: Emerging evidence associating obstructive sleep apnea (OSA) with adverse perioperative outcomes has recently heightened the level of awareness among perioperative physicians. In particular, estimates projecting the high prevalence of this condition in the surgical population highlight the necessity of the development and adherence to "best clinical practices." In this context, a number of expert panels have generated recommendations in an effort to provide guidance for perioperative decision-making. However, given the paucity of insights into the status of the implementation of recommended practices on a national level, we sought to investigate current utilization, trends, and the penetration of OSA care-related interventions in the perioperative management of patients undergoing lower joint arthroplasties. METHODS: In this population-based analysis, we identified 1,107,438 (Premier Perspective database; 2006-2013) cases of total hip and knee arthroplasties and investigated utilization and temporal trends in the perioperative use of regional anesthetic techniques, blood oxygen saturation monitoring (oximetry), supplemental oxygen administration, positive airway pressure therapy, advanced monitoring environments, and opioid prescription among patients with and without OSA. RESULTS: The utilization of regional anesthetic techniques did not differ by OSA status and overall <25% and 15% received neuraxial anesthesia and peripheral nerve blocks, respectively. Trend analysis showed a significant increase in peripheral nerve block use by >50% and a concurrent decrease in opioid prescription. Interestingly, while the absolute number of patients with OSA receiving perioperative oximetry, supplemental oxygen, and positive airway pressure therapy significantly increased over time, the proportional use significantly decreased by approximately 28%, 36%, and 14%, respectively. A shift from utilization of intensive care to telemetry and stepdown units was seen. CONCLUSIONS: On a population-based level, the implementation of OSA-targeted interventions seems to be limited with some of the current trends virtually in contrast to practice guidelines. Reasons for these findings need to be further elucidated, but observations of a dramatic increase in absolute utilization with a proportional decrease may suggest possible resource constraints as a contributor.