(161) Obstructive Sleep Apnea Is Associated With Postoperative Bleeding Risk in Total Joint Arthroplasty

Introduction: Obstructive sleep apnea (OSA) is a prevalent form of sleep-disordered breathing associated with numerous adverse outcomes. Assessing its independent impact can be challenging, as OSA is frequently accompanied by other comorbidities. To the best of our knowledge, bleeding risk in total knee and hip arthroplasties (TKA & THA) has not been robustly assessed in OSA patients.

Methods: We queried TriNetX, a multicenter, federated database collecting health record and claims-derived data from over 98 million patients across the US. We assayed patients undergoing TKA & THA with and without OSA (defined as ≥2 charted instances of OSA or 0 instances, respectively) between 08/01/2003 – 06/30/2023. We then examined the 30-day incidence of unspecified anemia (UA) and posthemorrhagic anemia (PHA) (ICD-10-CM D64.9 and D62) postoperatively, before and after extensive in-platform propensity matching (in which we accounted for 23 different demographic and comorbidity attributes).

Results: Of 97,985,933 patients in the cohort, we isolated 191,915 TKA patients, of who 34,139 had OSA; further, we isolated 138,863 THA patients, 18,796 with OSA. Prior to matching, OSA was significantly associated with bleeding in both groups: in TKA patients, UA risk was 4.3 in OSA vs 2.6% in non-OSA (OR 1.7, CI 1.6 – 1.8), while in PHA it was 2.3 vs 1.5% (OR 1.5, CI 1.4 – 1.6); meanwhile, in THA, UA risk was 4.1 vs 2.7% (OR 1.5, CI 1.4 – 1.7) and PHA risk was 2.7 vs 1.9% (OR 1.4, CI 1.3 – 1.6). After matching, the association persisted: in TKA patients, UA risk was 4.3 in OSA vs 3.1% (OR 1.4, CI 1.3 – 1.5), while in PHA it was 2.3 vs 1.7% (OR 1.3, CI 1.2 – 1.5); in THA, UA risk in OSA was 4.0 vs 3.2% (OR 1.3, CI 1.1 – 1.5), while PHA risk was 2.7 vs 2.2 (OR 1.2, CI 1.03 – 1.4, p=0.014). Unless specified, all p-values were < 0.001.

Conclusions: While OSA has been associated with thrombotic events, little data is available for its impact on bleeding risk. We observed a clear excess in bleeding diagnoses in our OSA cohort, though we are limited by deidentified data and our inability to assess for positive airway pressure therapy compliance. Endothelial dysfunction with impaired vascular tone and compromised wound healing may explain these results.