Perioperative Outcomes after Regional Versus General Anesthesia for Above the Knee Amputations

Background: Nontraumatic lower extremity amputation (LEA) remains a common procedure among patients who frequently have significant comorbidities. Patients undergoing above knee amputation (AKA) have the highest rates of mortality in this cohort, yet there is little evidence to support selection between peripheral nerve block or neuraxial regional anesthesia (RA) versus general anesthesia (GA) techniques. The objective of this study was to determine whether RA (neuraxial or peripheral nerve block) techniques were associated with more favorable outcomes versus general anesthesia among patients undergoing AKA. Methods: This is a retrospective cohort study using propensity-matched groups. Patients undergoing AKA were identified in the American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) data set and grouped according to anesthetic type as either RA or GA. Patients undergoing AKA with RA were propensity matched to similar patients who had GA. Primary outcome was 30-day mortality. Secondary outcomes were numerous and included cardiac, pulmonary, infectious, and bleeding complications, as well as length of stay. Among a subset of patients for whom readmission data were available, rate of readmission was compared as a secondary outcome. Results: Nine thousand nine hundred ninety-nine patients were identified in the ACS-NSQIP database. One thousand three hundred twelve received a regional anesthetic, and the remainder had a general anesthetic. Factors significantly associated with GA included younger age (70 vs. 75 years; P < 0.001), higher body mass index (26.5 vs. 25.4; P < 0.001), and ethnically white (62.4% vs. 57%; P < 0.001). Before matching, patients receiving RA were less likely to be smokers (22% vs. 29%; P < 0.001), have a bleeding disorder (15% vs 30%; P < 0.001), or have a diagnosis of sepsis (26% vs 34%; P < 0.001). Propensity score matching produced a cohort composed of 1,916 patients equally divided between RA and GA. We found no difference with respect to the primary end point of 30-day mortality (11.7% vs 11.7%; odds ratio [OR] 1.01; P = 0.943) nor was there any difference with respect to secondary outcomes. Among patients for whom readmission data were available, there was no statistically significant difference between rates of readmission between the groups (15.6% for RA vs. 12.7% for GA; OR 1.26, confidence interval 0.87-1.828, P = 0.221). Conclusions: The present investigation did not detect any difference between regional and general anesthetic with respect to morbidity or mortality among patients undergoing AKA. This data set did not allow us to address other relevant markers including pain control or phantom limb syndrome.