REsults of LOw Anterior Resections With or Without Preventive Stoma in Nonemergency Department

Overview

The purpose of this multicenter randomized non-inferiority trial is to evaluate the safety of low anterior resection for rectal cancer performed with versus without a diverting stoma in patients with a low predicted risk of colorectal anastomotic leakage. The primary objective is to determine whether the rate of anastomotic leakage within 30 days after surgery in the no-stoma group is non-inferior to that in the diverting stoma group. The secondary objectives include comparison between groups regarding: Stoma rate at 1 year after surgery; Quality of life at 30 days and 1 year (EORTC QLQ-C30, EORTC QLQ-CR29, and LARS score); Short-term postoperative outcomes, including postoperative day metrics, length of hospital stay, and complications graded according to the Clavien-Dindo classification; Reoperation rates within 30 days and 1 year. Participants will include adult patients with mid- or low-rectal adenocarcinoma who are scheduled for radical minimally invasive total mesorectal excision and have a predicted risk of anastomotic leakage \<10% according to the study risk model.

This study is a multicenter randomized controlled non-inferiority trial evaluating the safety of omitting routine diverting stoma formation in low-risk patients undergoing total mesorectal excision (TME) for mid- and low-rectal cancer. Patients with a predicted risk of colorectal anastomotic leakage (AL) ≤10% will be enrolled across 7 accredited centers. Local Ethics Committee approval has been obtained at each site. The protocol follows the SPIRIT 2025 recommendations. After informed consent, patients will be randomized 1:1 to "with preventive stoma" and "without preventive stoma". Stratified block randomization (block sizes 2, 4, or 6) will be used, stratified by center and sex. Allocation will be computer-generated (R software). No blinding will be performed. Cross-over to stoma formation in the no-stoma group is permitted in case of intraoperative safety concerns. All patients will undergo minimally invasive TME according to oncological principles (sharp dissection within embryological planes, adequate vascular control, stapled tension-free colorectal anastomosis). In the stoma group, a diverting ileostomy or colostomy will be created. Closure is planned ≥2-3 months postoperatively if clinically appropriate. Postoperative care follows institutional standards. Complications will be graded according to the Clavien-Dindo classification. Patients will be followed for 1 year with clinical visits (or telephone follow-up) at 30 days and every 3 months. Imaging and colonoscopy will be performed according to clinical guidelines. Quality-of-life assessments and LARS score will be obtained at 30 days and 1 year. Risk Prediction Model. Eligibility is determined using a gradient boosted model (Bernoulli loss) developed on retrospective data. The model demonstrated AUC 0.93 with predefined cutoff risk of 10% (sensitivity 74%, specificity 91%). Model development and validation were performed using R software. Sample Size and Statistical Analysis. Assuming a baseline AL rate of 15% in the stoma group, a non-inferiority margin of 10%, one-sided α=2.5%, 80% power, and 10% dropout rate, 442 patients (221 per group) are required. The primary analysis will follow the intention-to-treat (ITT) principle. A binomial logistic regression model will estimate risk difference with 95% confidence intervals using clustered robust standard errors by center. Non-inferiority will be concluded if the upper bound of the confidence interval does not cross the predefined margin. Sensitivity analyses will include per-protocol and instrumental variable approaches. Missing data for secondary outcomes will be handled using multiple imputation (MCAR assumption). Interim Safety Analysis. A safety analysis will be conducted after enrollment of 50 patients per group. If unacceptable differences in AL rates are observed, early termination may be considered.