Diversion colitis (DC) is a common inflammatory complication in patients with temporary ileostomy after rectal cancer surgery, and no standardized medical treatment exists. This prospective, assessor-blinded, parallel-group, randomized controlled trial evaluated whether autologous fecal microbiota transplantation (auto-FMT) delivered through the diverting stoma ameliorates DC and improves post-reversal outcomes. Sixty-six patients with endoscopically confirmed DC were randomized 1:1 to receive daily auto-FMT (n=33) or saline irrigation (n=33) for four weeks. The primary endpoints are changes from baseline to week 4 in endoscopic (modified Harig score, 0-12) and histopathological (0-9) scores. Secondary endpoints include Wexner incontinence score, quality of life (EORTC QLQ-C30/CR29), inflammatory biomarkers, and safety. The study is designed to test whether auto-FMT produces superior improvements in endoscopic and histopathological severity compared with saline control, and leads to better functional outcomes after stoma reversal.

Background and Rationale

Diversion colitis frequently develops after fecal stream diversion, affecting most patients whose intestinal continuity remains interrupted for more than three to six months. While stoma reversal is definitive, many patients require prolonged diversion due to adjuvant chemotherapy, poor general condition, or anastomotic healing concerns. Existing medical therapies-including short-chain fatty acid enemas, 5-aminosalicylates, corticosteroids, and probiotics-lack consistent efficacy in randomized trials. Gut microbiota dysbiosis is a central driver of DC; restoring a diverse microbial community via fecal microbiota transplantation represents a rational approach. Autologous FMT using the patient's own stoma effluent avoids pathogen transmission, donor screening, and ethical concerns. However, no prospective RCT has systematically evaluated auto-FMT for DC using endoscopic and histopathological endpoints.

Study Design

Single-center, prospective, assessor-blinded, parallel-group, superiority randomized controlled trial with a 1:1 allocation ratio.

Participants

Adults aged 18-75 years with histopathologically confirmed rectal adenocarcinoma who underwent low anterior resection with temporary loop ileostomy, scheduled for reversal at 3-6 months after primary surgery, and with endoscopic DC (modified Harig score ≥4 at week 4 post-ileostomy). Key exclusion criteria: neoadjuvant chemoradiotherapy, pre-existing inflammatory bowel disease, recent antibiotic or probiotic use, severe organ dysfunction, pregnancy, or lactation.

Interventions

Auto-FMT group: Daily irrigations of autologous fecal microbiota suspension for 4 weeks. Preparation: 50-80 g of fresh stool collected from the patient's stoma bag within 2 hours of passage, homogenized with 500 mL sterile normal saline (0.9% NaCl) pre-warmed to 37°C, stirred, and filtered through two layers of sterile gauze. The filtrate was used within 30 minutes. Irrigation: a 14-16 French Foley catheter inserted 10-15 cm into the efferent limb of the loop ileostomy; suspension infused by gravity drip over 5-10 minutes; patients retained the suspension for at least 30 minutes before evacuation.

Control group: Daily irrigations of 500 mL sterile normal saline (37°C) using the same catheter and technique, with the same retention time.

Outcome Measures

Primary outcomes: Change from baseline to week 4 in endoscopic score (modified Harig score, 0-12) and histopathological score (composite of mucosal atrophy, crypt distortion, and inflammatory infiltrate, 0-9), assessed by blinded reviewers.

Secondary outcomes: Wexner incontinence score at 1, 3, and 6 months after stoma reversal; quality of life (EORTC QLQ-C30 and QLQ-CR29) at baseline, week 4, and 6 months post-reversal; serum hs-CRP, albumin, and fecal calprotectin at baseline and week 4; adverse events (CTCAE v5.0); treatment adherence (≥80% of 28 sessions).

Sample Size

33 patients per group (total 66) to detect a mean endoscopic score reduction difference of 1.5 points (assuming SD 2.0 in auto-FMT group and SD 1.8 in control group), 80% power, two-sided α = 0.05, accounting for a 20% dropout rate.

Statistical Analysis

Primary analysis was intention-to-treat. Change scores were analyzed using ANCOVA with baseline score as covariate. Secondary outcomes: Wexner scores with generalized estimating equations; quality of life with ANCOVA; biomarkers with Mann-Whitney U tests. Missing data were handled with multiple imputation. Two-tailed p < 0.05 was considered significant.

Ethical Approval

The protocol was approved by the Ethics Committee of Lin'an First People's Hospital, Hangzhou (Approval No.: Lin'an First People's Hospital Lun Yan Shen 2022 No.20, dated April 29, 2022). Written informed consent was obtained from all participants. The study followed the Declaration of Helsinki.

Inclusion Criteria:

• Histopathologically confirmed rectal adenocarcinoma.
• Low anterior resection with temporary loop ileostomy.
• Age 18-75 years.
• Scheduled for ileostomy reversal at 3-6 months after primary surgery.
• Endoscopic confirmation of diversion colitis at week 4 post-ileostomy (modified Harig score ≥4, range 0-12).
• Written informed consent.

Exclusion Criteria:

• Neoadjuvant chemoradiotherapy.
• Pre-existing inflammatory bowel disease, irritable bowel syndrome, or chronic constipation.
• Previous colorectal surgery (other than index surgery).
• Active infection requiring systemic antibiotics within 4 weeks before enrollment.
• Use of probiotics, prebiotics, or antibiotics within 4 weeks before enrollment.
• Severe organ dysfunction (Child-Pugh B/C cirrhosis, end-stage renal disease).
• Pregnancy or lactation.
• Any condition precluding protocol compliance or outcome assessment.

De-identified individual participant data underlying the results reported in this manuscript will be made available to researchers who provide a methodologically sound proposal, for the purpose of individual participant data meta-analysis or other approved research. The study protocol, statistical analysis plan, and informed consent form will also be available. Data will be accessible immediately after publication, upon reasonable request to the corresponding author.

Outcome Assessor and Statistician were blinded.