Unintended prolonged and permanent stoma following restorative rectal cancer surgery with defunctioning loop ileostomy: A systematic review and meta-analysis

Abstract

Background: An intentional temporary diverting loop ileostomy (DLI) is frequently part of restorative rectal cancer surgery. However, such a stoma itself risks morbidity and quality of life, especially when closure is delayed or impossible. We investigated the incidence and risk factors of such prolonged-permanent stoma (PPS), defined as stoma persistence beyond 12 months from index surgery. Methods: This systematic review and meta-analysis was conducted in accordance with PRISMA guidelines and prospectively registered (PROSPERO ID: CRD420251066362). Literature databases were searched for studies describing patients with rectal adenocarcinoma undergoing restorative resection with DLI up to August 2025. Outcomes of interest were failure to become stoma-free as expected and underlying risk factors. Study-level covariate impact was explored via meta-regression. Certainty of evidence and study quality were assessed by the GRADE and Newcastle-Ottawa scales, respectively. Results: Nineteen studies were included comprising 9932 patients of whom 18.7% (95% CI 15.3–22.5) had PPS. All were retrospective with 15 judged high quality. Seventeen potential predictors were identified and categorisable as being either tumour-related, preoperative, operative or postoperative. Local or systemic recurrence, anastomotic leak and stage IV (cM+) at diagnosis were the strongest single predictors (each conferring approximately fivefold PPS-risk). Other significant (p < 0.05) predictors were comorbidity and ASA score, open surgery, neoadjuvant/adjuvant therapy and postoperative complications (all OR < 2). Meta-regression identified study geography as a significant source of heterogeneity. GRADE certainty regarding predictors ranged from ‘very low’ to ‘moderate’ (tumour recurrence alone). Conclusion: With nearly one in five cases of temporary DLI progressing to PPS, identified risk factors must be utilised to optimise patient care. More robust clinical studies are required.