A Multivariate Regression Analysis of Panniculectomy Outcomes

Abstract

Panniculectomy can improve quality of life in morbidly obese patients, but its functional benefits are counterbalanced by relatively high complication rates. The authors endeavored to determine the impact of plastic surgery training on panniculectomy outcomes. A retrospective review was performed of the prospectively maintained American College of Surgeons National Surgical Quality Improvement Program database for all patients undergoing panniculectomy from 2006 to 2010. Patient demographic details, surgeon specialty training, and 30-day outcomes were assessed. A total of 954 panniculectomies meeting inclusion criteria were identified. Plastic surgeons performed 694 (72.7 percent) of the procedures, and 260 (27.3 percent) were performed by nonplastic surgeons. Nonplastic surgeons had significantly higher rates of overall complications (23.08 percent versus 8.65 percent; p < 0.001) and wound infections (12.69 percent versus 5.33 percent; p < 0.001) than plastic surgeons. Average operative time for plastic surgeons was significantly longer than that for nonplastic surgeons (3.00 ± 1.48 hours versus 1.88 ± 0.93 hours; p < 0.001). Risk-adjusted multivariate regression showed that undergoing a panniculectomy by a nonplastic surgeon was a significant predictor of overall postoperative complications (odds ratio, 2.09; 95 percent CI, 1.35 to 3.23) and wound infection (odds ratio, 1.73; 95 percent CI, 1.004 to 2.98). Subgroup analysis of propensity-matched samples supported this finding. Multivariate regression analysis of National Surgical Quality Improvement Program data showed that panniculectomy performed by plastic surgeons results in lower rates of overall postoperative complications compared with that performed by nonplastic surgeons.