Segmental versus extended colectomy for colonic Crohn’s disease in advanced therapy era: a systematic review and Bayesian meta-analysis

Abstract Background The optimal extent of resection for colonic Crohn’s disease (CD) remains controversial. Segmental colectomy (SC) may preserve bowel and avoid permanent stoma but has been historically associated with higher recurrence than extended colectomy (EC). In the advanced therapy era, contemporary data and Bayesian methods may refine this risk–benefit balance. Methods PubMed, Embase, and Cochrane Library were searched from inception to December 1, 2025. The primary outcome was surgical recurrence; secondary outcomes were reoperation, permanent stoma, overall postoperative complications, and mortality. Frequentist random-effects models (restricted maximum likelihood) were used to pool odds ratios (ORs) with 95% confidence intervals (CIs), with heterogeneity assessed by I 2 and leave-one-out analyses. Bayesian random-effects meta-analyses (bayesmeta) were performed for surgical recurrence, reoperation, and permanent stoma using vague priors and empiric Turner priors for heterogeneity, with additional models incorporating historical evidence from the advanced therapy era. Results Five observational studies (1095 patients; 501 SC, 594 EC) were included. Surgical recurrence did not differ significantly between SC and EC (OR 1.15; 95% CI 0.53–2.50; I 2 = 74%). Reoperation (OR 1.36; 95% CI 0.71–2.61; I 2 = 68%), permanent stoma (OR 0.22; 95% CI 0.03–1.39; I 2 = 87%), and overall complications (OR 0.75; 95% CI 0.40–1.40; I 2 = 64%) were also not significantly different. Mortality was reported in one study (1 SC vs 5 EC). Bayesian models suggested a higher probability of surgical recurrence and reoperation after SC but with wide credible intervals and substantial heterogeneity, whereas SC showed a consistently high probability of reducing permanent stoma risk (posterior OR ~ 0.23; P [OR ≤ 1] > 0.95 in primary models; reinforced by historical priors). Leave-one-out analyses identified influential studies for some outcomes but did not materially alter the overall interpretation. All studies had a moderate risk of bias, mainly due to confounding and selection. Conclusion In contemporary observational evidence, SC and EC show broadly comparable outcomes for surgical recurrence, reoperation, complications, and mortality, while SC is strongly favored with respect to permanent stoma avoidance. Operative decision-making should weigh a probable increase in recurrence risk after SC against the clinically meaningful functional benefit of stoma preservation, recognizing the limitations imposed by observational designs, heterogeneity, and sparse data.