Abstract
BACKGROUND: The aim of the study is to evaluate the risk factors of anastomotic leakage (AL) and develop a nomogram to predict the risk of AL in surgical management of primary ovarian cancer. METHODS: We retrospectively reviewed 770 patients with primary ovarian cancer who underwent surgical resection of the rectosigmoid colon as part of cytoreductive surgery between January 2000 to December 2020. AL was defined based on radiologic studies or sigmoidoscopy with relevant clinical findings. Logistic regression analyses were performed to identify the risk factor of AL, and a nomogram was developed based on the multivariable analysis. The bootstrapped-concordance index was used for internal validation of the nomogram, and calibration plots were constructed. RESULTS: The incidence of AL after resection of the rectosigmoid colon was 4.2% (32/770). Diabetes (OR 3.79; 95% CI, 1.31-12.69; p = 0.031), co-operation with distal pancreatectomy (OR, 4.8150; 95% CI, 1.35-17.10; p = 0.015), macroscopic residual tumor (OR, 7.43; 95% CI, 3.24-17.07; p = 0<001) and anastomotic level from the anal verge shorter than 10 cm (OR, 6.28; 95% CI, 2.29-21.43; p = 0.001) were significant prognostic factors for AL on multivariable analysis. Using four variables, the nomogram has been developed to predict anastomotic leakage: https://ALnomogram.github.io/ . CONCLUSION: Four risk factors for AL after resection of the rectosigmoid colon are identified from the largest ovarian cancer study cohort. The nomogram from this information provides a numerical risk probability of AL, which could be used in preoperative counseling with patients and intraoperative decision for accompanying surgical procedures and prophylactic use of ileostomy or colostomy to minimize the risk of postoperative leakage. TRIAL REGISTRATION: Retrospectively registered.