A novel nomogram to predict low anterior resection syndrome (LARS) after ileostomy reversal for rectal cancer patients

Background and aim: Low anterior resection syndrome (LARS) in patients undergoing low or ultra-low anterior resection (LAR) is a common problem and significantly impacts the quality of life. Patients with an ileostomy after LAR are more likely to develop LARS. However, there hasn't been a model predicting LARS occurrence in these patients. This study aims to construct a nomogram to predict the probability of LARS occurrence in patients with temporary ileostomy and guide preventive strategies before reversal. Methods: 168 patients undergoing LAR with ileostomy from one center were enrolled as the training cohort, and 134 patients of the same inclusion criteria from another center were enrolled as the validation cohort. The training cohort was screened for risk factors for major LARS using univariate and multivariate logistic regression. The nomogram was constructed using the filtered variables, the ROC curve was used to describe the model's discrimination, and the calibration was used to describe the accuracy Results: The optimal cut-off value for stoma closure time was 128 days. Three risk factors were identified using logistic regression analysis: preoperative radiotherapy (OR = 3.038, [95%CI 1.75-5.015], P = 0.005), stoma closure time (OR = 2.298, [95%CI 1.088-4.858], P = 0.029) and pN stage (OR =1.739, [95%CI 1.235 -3.980], P = 0.001). A nomogram was constructed based on these three variables and showed good performance predicting major LARS after stoma reversal. The area under the curve (AUC) was 0.827 in the training group and 0.821 in the validation group; The calibration curve suggested good precision in both groups. Conclusions: This novel nomogram can accurately predict the probability of major LARS occurrence after ileostomy reversal for rectal cancer patients. This model can help screen ileostomy patients with high risks and guide individualized preventive strategies before stoma reversal. (c) 2022 Elsevier Ltd, BASO similar to The Association for Cancer Surgery, and the European Society of Surgical Oncology. All rights reserved.