LOGIN 

 

JOIN

 

RENEW

 

CME/MOC

Sleep Matters

HomeMembersAssemblies and SectionsAssembliesSleep and Respiratory NeurobiologySleep Matters ▶ OSA and Perioperative Complications


OSA and Perioperative Complications

Undiagnosed OSA and risk for perioperative complications.

Is undiagnosed OSA a risk factor for perioperative complications? Prior retrospective cohorts have been mixed in their results, with some studies demonstrating an association of OSA with greater risk of perioperative complications and others not showing this association.  However, given the administrative nature of these cohorts, many were unable to reliably assess undiagnosed OSA.

Chan et al. recently performed a rigorous prospective cohort study among 1364 patients to shed light on this important question. The investigators enrolled patients who were scheduled to undergo non-cardiac surgeries. Each patient had a home sleep apnea testing prior to surgery, and was followed for a combined primary endpoint of cardiac death, myocardial injury (i.e. elevated troponin), congestive heart failure, new-onset thromboembolism, atrial fibrillation, or stroke. The results of home sleep apnea testing were not shared with patients or their providers until 30 days after surgery.

After adjusting for potential confounders, the authors found that severe OSA was associated with the combined endpoint. Mild and moderate OSA were not. In post-hoc analyses looking at individual endpoints, it appears that this association was driven by cardiac endpoints among those with severe OSA (e.g. greater cardiac death, myocardial injury, congestive heart failure and new onset atrial fibrillation).  Oxygen administration, type of anesthesia, and postoperative opioid use did not alter the association between OSA and the primary outcome.

Overall, these results provide greater evidence that undiagnosed OSA may portend poorer outcomes in the perioperative setting. Given the observational nature of the results, however we must be cautious in assigning a causal nature to the relationship. It is also possible that patients with undiagnosed OSA could also be at risk for subclinical heart disease that manifests in the perioperative setting. These results do provide strong rationale for an interventional trial, and suggest that providers should be cautious in the perioperative setting, pending further evidence.