Use of a nomogram to predict the closure rate of diverting ileostomy after low anterior resection: A retrospective cohort study.

INTRODUCTION: Although temporary ileostomy is widely used to prevent complications due to anastomotic leakage after middle and low rectal cancer surgery, some patients fail to achieve stoma closure after primary surgery. The aim of this study was to identify the risk factors for permanent stoma following low anterior resection (LAR) or intersphincteric resection (ISR) with a temporary ileostomy for rectal cancer, while focusing on the time course, to develop a nomogram that can predict the rate of unreversed ileostomy 1 year after initial surgery.

METHOD: A total of 212 consecutive rectal cancer patients who underwent LAR or ISR with or without a temporary stoma between 2012 and 2015 at the University of Tokyo Hospital were retrospectively reviewed. Data analyses were performed using JMP Pro 11.0 and R 3.0.1 with rms and Hmisc packages to identify the risk factors for diverting ileostomy resulting in un-reversed stoma, and to develop a nomogram using these factors.

RESULTS: Among 212 patients, diverting ileostomy and colostomy were performed in 116 and 11 patients, respectively, and a stoma was not created in 85 patients. Among the ileostomy cases, 94 underwent stoma reversal, and the median interval from initial surgery to stoma closure was 6.9 months. Three patients eventually underwent stoma re-creation, and hence, 25 patients had permanent stoma. The following variables were correlated with the stoma non-reversal rate and were included in the nomogram: depth of invasion (p = 0.02), presence of metastatic organs (p = 0.07), and preoperative chemoradiotherapy (p = 0.03). The nomogram C-index was 0.612, indicating moderate predictive ability.

CONCLUSIONS: The most common factors preventing stoma closure included distant metastasis or rectal cancer recurrence. The nomogram developed in the present study can help identify rectal cancer patients with high risk of stoma non-reversal.