Background: Total pelvic exenterations (TPEs) for malignancies are complex operations often performed by multidisciplinary teams. The differences among primary cancer for TPE and multicentered results are not well described. We aimed to describe TPE outcomes for different malignant origins in a national multicentered sample.
Methods: Patients from the National Surgical Quality Improvement Program (NSQIP) database who underwent TPE between 2005 and 2016 for all malignant indications (colorectal, gynecologic, urologic, or other) were included. Chi square and Kruskal-Wallis tests were used to compare patient characteristics by primary malignancy. Multivariate logistic and linear regression models were used to determine factors associated with any 30-day Clavien-Dindo grade 3 or higher complication, length of hospital stay (LOS; days), 30-day wound infection, and 30-day mortality.
Results: Overall, 2305 patients underwent TPE. Indications for surgery included 33% (749) colorectal, 15% (335) gynecologic, 9% (196) other, and 45% (1025) urologic malignancies. Median LOS decreased from 10 to 8 days during the study period (p < 0.001), 36% were males, and 50% required blood transfusion. High-grade complications occurred in 15% of patients and were associated with bowel diversion [odds ratio (OR) 1.6, 95% confidence interval (CI) 1.1-2.4], disseminated cancer (OR 1.8, 95% CI 1.4-2.3), and gynecologic cancers (OR 2.9, 95% CI 1.8-4.7). Mortality was 2% and was associated with disseminated cancer (OR 2.2, 95% CI 1.1-4.3) and male sex (OR 2.4, 95% CI 1.3-4.4).
Conclusions: TPE is associated with high rates of complications, however mortality rates remain low. Preoperative and perioperative outcomes differ depending on the origin of the primary malignancy.