Prognostic Factors and Significance of Gastrointestinal Leak After Cytoreductive Surgery (CRS) with Heated Intraperitoneal Chemotherapy (HIPEC)

Background. Gastrointestinal leak (GIL) after cytoreductive surgery with heated intraperitoneal chemotherapy (CRS/HIPEC) is associated with significant morbidity and mortality. This study aimed to identify GIL prognostic factors and its impact on overall survival. Methods. A retrospective analysis of a prospectively maintained database comprising 1270 CRS/HIPEC procedures was performed. Type of GIL, functional and resection status, morbidity, mortality, and survival were reviewed. Results. Gastrointestinal leaks were identified in 8.7% (110/1270) of CRS/HIPEC procedures, including 53 anastomotic leaks (4.2%), 53 hollow viscus perforations (4.2%), and four leaks at unknown sites. The multivariate predictors of leak were Eastern Cooperative Oncology Group (ECOG) functional status (ECOG 1 vs. 0: odds ratio [OR] 2.12, p = 0.009; ECOG 2 vs. 0: OR 2.90, p = 0.004), and number of anastomoses (OR 5.34; p < 0.0001). The in-hospital mortality rate for the GIL cohort was 21.8% (24/110), with a 72% (80/110) reoperation rate. The leak cohort had a higher major morbidity rate (88.3 vs. 23.3%; p < 0.0001), a longer hospital stay (39.0 vs. 9.9 days; p < 0.0001), and a longer intensive care unit (ICU) stay (7.7 vs. 1.7 days; p = 0.0003). After surgical mortality was excluded, the overall survival periods for the leak and no-leak patients with complete cytoreduction were respectively 1.5 and 4.98 years (p = 0.0001). Clinically significant decreases in survival were observed for all primary malignancies. Conclusions. Gastrointestinal leak after CRS/HIPEC is a source of significant mortality, with a decrease in overall survival even after complete CRS. Preoperative functional status and number of anastomoses are predictors of leak for CRS/HIPEC patients.