Propensity-score matched outcomes of minimally invasive and open pelvic exenteration in locally advanced rectal cancer

Abstract

Pelvic exenteration (PE) entails an en bloc resection of locally advanced primary or recurrent rectal cancer. This study aimed to assess the short-term and survival outcomes of minimally invasive (MI)- and open PE. A retrospective cohort analysis of patients with stage III rectal adenocarcinoma treated with PE from the National Cancer Database (2010-2019) was conducted. Open and MI-PE were matched for baseline and treatment characteristics using 2:1 propensity score matching. Primary outcomes were 30- and 90-day mortality; secondary outcomes included 30-day readmission, hospital stay, surgical margins, lymph node yield, and overall survival (OS). PE was performed in 1010 (1.9%) of 52,242 patients; 705 (69.8%) were open and 304 were (30.2%) MI procedures. After matching, 169 patients in the MI-PE group were matched to 338 patients in the open PE group. The matched cohort included 507 patients (54.2% female). Hospital stay was shorter after MI compared to open PE (median: 6 vs. 8 days, p < 0.001). MI and open PE had similar odds of 30- (OR: 0.33, p = 0.306) and 90-day mortality (OR: 0.29, p = 0.113), and comparable rates of positive surgical margins (12.7% vs. 15%, p = 0.586) and suboptimal lymph node yield (19% vs. 26%, p = 0.096). MI-PE had a similar OS to open surgery (82.5 vs. 77.5 months, p = 0.281). Robotic-assisted PE was associated with significantly lower odds of conversion to open surgery (OR: 0.15, p = 0.003) and shorter hospital stay (median: 5 vs. 7 days, p = 0.026) than laparoscopic PE. MI-PE provided similar pathologic and survival outcomes to open PE with comparable short-term mortality rates and significantly shorter hospital stays.