Obstructive sleep apnea affects complication rates following knee arthroscopy but use of continuous positive airway pressure is not protective against complications.

PURPOSE: Obstructive sleep apnea (OSA) has not been studied as a risk factor for complications following knee arthroscopy. The goals of this study were to: (1) compare complication rates after knee arthroscopy between patients with and without OSA and (2) evaluate whether continuous positive airway pressure (CPAP) mitigated complication rates.

METHODS: A national private insurance database was queried for patients undergoing simple knee arthroscopy from 2007 to 2016. Patients with a diagnosis of OSA were then identified using ICD-9/10 codes. Patients with OSA were then subdivided into cohorts with and without a billing code for a CPAP device. Adverse events within 30 days postoperatively related to OSA were then assessed in all groups: (1) emergency room (ER) visit, (2) hospital admission, (3) pulmonary embolism (PE), (4) myocardial infarction, (5) respiratory arrest and (6) in-hospital mortality within 6 months postoperatively. Adverse event rates were compared between the control and study groups using a multivariable regression analysis.

RESULTS: 97,036 patients underwent simple knee arthroscopy with 8656 patients having a diagnosis of OSA. Of these, 3820 (44%) had orders for CPAP machines. After controlling for confounders, patients with OSA had significantly higher risk of ER visits, PE and respiratory arrest compared to controls (p < 0.05). The majority of these significant findings persisted regardless of CPAP use. There were no significant differences in complication rates between OSA patients with and without CPAP orders.

CONCLUSIONS: OSA appears to be independently associated with a higher risk for ER visits, PE and respiratory arrest following knee arthroscopy after controlling for demographic and comorbidity confounders. An order for CPAP was not associated with a significant reduction the risk for these complications. CPAP noncompliance may not be as important a factor when risk stratifying patients undergoing ambulatory knee arthroscopy compared to more significant medical comorbidities.

LEVEL OF EVIDENCE: III.