Prevalence, detection rate and outcome of Cytomegalovirus infection in ulcerative colitis patient requiring colonic resection

OBJECTIVE: To determine the prevalence of cytomegalovirus infection in patients with steroid-refractory ulcerative colitis who required colonic resection, and to assess its possible association with the use of immunosuppressive and steroid treatment and outcome after colectomy. PATIENTS AND METHODS: The study included surgical specimens and related pre-operative endoscopic biopsy specimens of 77 consecutive ulcerative colitis patients (34 females) who underwent colectomy because of intractable steroid-refractory ulcerative colitis (55 patients), toxic megacolon (6 patients), dysplasia or cancer (7 patients) or loss of function of the colon (9 patients). Clinical features and current and past treatments were analysed. Haematoxylin and eosin and specific immunohistochemical staining for cytomegalovirus were used to detect inclusion bodies in all specimens. RESULTS: Cytomegalovirus infection was found in 15 of 55 steroid-refractory ulcerative colitis patients (27.3%) and in 2 of 22 non-refractory patients (9.1%) (p=0.123). Only six patients had positive staining for cytomegalovirus in pre-operative endoscopic biopsy specimens. Detection of cytomegalovirus inclusion in biopsy specimens was not related to the number of biopsies or to time that had elapsed since colonoscopy and index surgery. Cytomegalovirus-positive patients were more likely to be on systemic corticosteroids (p=0.03). In contrast, current use and duration of immunosuppressive treatment, number of steroid cycles since diagnosis and in the last year, as well as chronic use of steroid in the last year were not significantly related to cytomegalovirus infection. Cytomegalovirus-positive patients did not receive antiviral therapy following proctocolectomy but did not show endoscopic or histological cytomegalovirus reactivation in the ileo-anal pouch and in the remaining bowel. CONCLUSIONS: Cytomegalovirus infection is frequently found in surgical specimens of patients with steroid-refractory ulcerative colitis and is more likely in patients on corticosteroid treatment. Cytomegalovirus infection is frequently unrecognised in pre-operative biopsy specimens, thus raising concerns about the accuracy of the available diagnostic tools. Unrecognised and untreated cytomegalovirus infection does not affect the outcome of ulcerative colitis patients following proctocolectomy.