Perioperative factors during ileal pouch-anal anastomosis predict pouchitis

Pouchitis is the most common complication of IPAA. Identifying factors predictive of pouchitis may improve outcomes by modifying contributing factors and enhancing patient selection. The most objective means for confirming pouchitis is by histology because the clinical and endoscopic diagnoses rely on more subjective assessments. The importance of histological pouchitis in the absence of clinical or endoscopic findings is unknown.

Prospectively collected data on patients with IPAA and pouch surveillance were evaluated. Patients who developed pouchitis, defined as symptoms of pouchitis confirmed by endoscopic biopsy (group B) were compared with those without any episode of clinical, endoscopic, or histological pouchitis (group A) for pre- and intraoperative factors and outcomes. Asymptomatic patients with histological pouchitis on surveillance biopsies (group C) were further compared with group A. Patients with Crohn's disease were excluded.

Of the 673 patients with pouch biopsies, 422 (62.7%) were in group A, 161 (23.9%) in group B, and 90 (13.4%) in group C. Mean follow-up was 9.8 (±5.1), 12.4 (±5.4), and 13. (±4.7) years. Of the 43 preoperative factors evaluated, those associated with group B included leukocytosis (P < .001), rheumatologic extraintestinal disease (P < .001), disease proximal to splenic flexure (P = .001), pulmonary comorbidity (P = .004), prior steroid use (P = .006), and age at operation and diagnosis (P = .018 and .021). Of the 10 intraoperative factors evaluated, pouchitis was associated with S-pouch reconstruction (P < .001), transfusion (P < .001), and 2-stage instead of 3-stage operation (P = .05), all surrogates for operative complexity. On multivariate analysis, pulmonary comorbidity (OR 3.38, 95% CI 1.62-7.07), disease proximal to splenic flexure (OR 2.37, 95% CI 1.18-4.77), extraintestinal disease manifestations (OR 1.6, 95% CI 1.01-2.54), and S-pouch reconstruction (OR 1.59, 95% CI 0.99 - 2.54) were associated with pouchitis. Patients in group B had worse outcomes, including more strictures (P = .015), bowel obstructions (P = .019), fistulas (P = .18), and lower quality of life (P < .001). Group C patients had the same outcomes as those in group A and the finding was not predicted by the above-mentioned parameters.

Patients with symptomatic, biopsy-confirmed pouchitis have worse long-term outcomes than those without pouchitis. This complication is associated with specific pre- and intraoperative factors. Histological pouchitis incidentally found on surveillance biopsy in asymptomatic patients is of no clinical relevance and does not influence outcome. Identification of these preoperative factors associated with the subsequent development of pouchitis will strengthen patient counseling and may facilitate risk stratification.