Nomogram for predicting the probability of rectal anastomotic re-leakage after stoma closure: a retrospective study

Abstract

Background

In this study, we aimed to identify the risk factors in patients with rectal anastomotic re-leakage and develop a prediction model to predict the probability of rectal anastomotic re-leakage after stoma closure.

Methods

This study was a single-center retrospective analysis of patients with rectal cancer who underwent surgery between January 2010 and December 2020. Among 3225 patients who underwent Total or Partial Mesorectal Excision (TME/PME) surgery for rectal cancer, 129 who experienced anastomotic leakage following stoma closure were enrolled. Risk factors for rectal anastomotic re-leakage were analyzed, and a prediction model was established for rectal anastomotic re-leakage.

Results

Anastomotic re-leakage after stoma closure developed in 13.2% (17/129) of patients. Multivariable analysis revealed that neoadjuvant chemoradiotherapy (odds ratio, 4.07; 95% confidence interval, 1.17–14.21; p = 0.03), blood loss > 50 ml (odds ratio, 4.52; 95% confidence interval, 1.31–15.63; p = 0.02), and intersphincteric resection (intersphincteric resection vs. low anterior resection: odds ratio, 6.85; 95% confidence interval, 2.01–23.36; p = 0.002) were independent risk factors for anastomotic re-leakage. A nomogram was constructed to predict the probability of anastomotic re-leakage, with an area under the receiver operating characteristic curve of 0.828 in the cohort. Predictive results correlated with the actual results according to the calibration curve.

Conclusions

Neoadjuvant chemoradiotherapy, blood loss > 50 ml, and intersphincteric resection are independent risk factors for anastomotic re-leakage following stoma closure. The nomogram can help surgeons identify patients at a higher risk of rectal anastomotic re-leakage.