A predictive nomogram model for low anterior resection syndrome after rectal cancer resection

摘要

Abstract Background The aim of this study was to identify risk factors associated with the low anterior resection syndrome (LARS) and to construct a nomogram capable of predicting the risk of LARS in patients who undergo rectal cancer resection. Methods About 538 patients who had undergone anterior resection were recruited as a development set. In addition, 114 patients with rectal cancer were analysed as a validation set to test the new nomogram. Patients in the development set were grouped into two separate cohorts: those with major LARS and those with minor or no LARS. Multiple logistic regression was conducted to detect risk factors for major LARS. Results The prevalence of major LARS was 40.7%, of minor LARS was 28.6% and the proportion with no LARS was 30.7% in the development set. In multivariate analysis, female gender, preoperative chemoradiation, low tumour height, diverting ileostomy, postoperative anastomotic leakage were shown to be independently associated with major LARS occurring in patients after rectal cancer resection. The area under the curve (AUC) values of the nomogram were 0.726 (95% CI: 0.682–0.769) and 0.750 (95% CI: 0.655–0.845) in the development and validation sets, respectively. The calibration curves and Hosmer‐Lemeshow goodness of fit tests showed that the model was acceptably accurate. Conclusion A nomogram model based on risk factors could be valuable as a predictor of the probability of major LARS after rectal cancer surgery, and provides a guide that clinical staff can use to take preventive measures for high‐risk patients.