Pathologic features and clinical significance of "backwash" ileitis in ulcerative colitis.

Patients with ulcerative colitis (UC) may develop inflammation in the distal ileum thought to be due to "backwash" of cecal contents ("backwash ileitis"). However, a systematic analysis of ileal changes in UC has never been performed, and the prevalence and criteria for "backwash" ileitis have not been defined. The aim of this study was to evaluate the prevalence and spectrum of inflammatory changes in the ileum in patients with UC and to correlate ileal changes with outcome after total proctocolectomy and ileal pouch-anal anastomosis. Routinely processed ileocolonic resection specimens from 200 consecutive patients with clinically and pathologically confirmed UC were evaluated for a wide variety of pathologic features in the ileum and colon. The ileal data were correlated with both the clinical features and the pathologic findings in the colon. Follow-up data were obtained to confirm absence of Crohn's disease and to evaluate outcome of ileo-anal pouches. Overall, 34 of 200 (17%) UC patients had inflammatory changes in the ileum (male/female ratio, 16/18; mean age, 42 years); 32 of 34 (94%) had pancolitis, which was significantly higher than the rate of pancolitis (39%) in patients without ileal disease (N = 166) (P < 0.001), but there were no other differences between patients with or without ileal pathology. In the colon, 22 of 34 (65%) patients had severe activity. Ileal changes included villous atrophy and crypt regeneration without increased inflammation (N = 3), increased neutrophilic and mononuclear inflammation in the lamina propria (N = 6), patchy cryptitis and crypt abscesses (N = 21) and focal superficial surface erosions (N = 4), some with pyloric metaplasia (N = 2 of 4). In general, the severity of ileal changes paralleled the severity of colonic activity. However, 2 of 4 (50%) patients with superficial erosions in the ileum had subtotal or left-sided colitis only, and had only mild colonic activity. Other cases showed only mild to moderate colonic activity and patchy or discontinuous involvement of the distal ileum. Upon follow-up of patients with erosions (mean, 48.5 months; range, 26-102 months), none developed manifestations of Crohn's disease anywhere in the gastrointestinal tract. The presence of inflammatory changes in the ileum had no effect on the prevalence of pouch complications or on the occurrence of dysplasia or cancer. Ileal changes in UC are not uncommon (prevalence, 17%), are generally mild in nature (villous atrophy, increased inflammation, scattered crypt abscesses), and are not associated with an increased rate of ileo-anal pouch complications, dysplasia, or carcinoma. In some cases, our findings are consistent with a backwash etiology. However, rarely, ileal erosions may occur in patients without cecal involvement, which may indicate that other pathogenetic mechanisms should be considered in the etiology of ileitis in UC patients.