True cytomegalovirus colitis is a poor prognostic indicator in patients with ulcerative colitis flares: the 10-year experience of an academic referral inflammatory bowel disease center.

Background and aims: The impact of cytomegalovirus (CMV) colitis on long-term outcomes of ulcerative colitis (UC) flares remains controversial. Methods: A total of 257 UC patients with moderate-to-severe flares were observed for a mean follow-up of 41.2 months. CMV colitis was defined as histopathologic confirmation of CMV inclusions obtained from macroscopic endoscopic lesions in patients with UC flares. An independent gastrointestinal pathologist prospectively reviewed all specimens. A poor outcome was defined as any of hospitalization, colectomy or death during the follow-up period. Results: The prevalence of CMV colitis was 14% (36/257) over the 10-year study period (2007-2016). When compared to the controls, patients with CMV colitis were characterized by older age, higher disease activity, endoscopic deep ulcerations and more frequent use of immunosuppressive drugs (all p < .05). In total, 57 outcome events (50 hospitalizations, seven colectomies) were observed among the study population (44.7% in patients with CMV colitis vs. 18.9% in controls). The cumulative probability of a poor outcome was significantly greater in the patients with CMV colitis than in the controls (log-rank test p < .001). In a multivariable analysis, CMV colitis remained as an independent predictor of a poor outcome (hazard ratio; 2.27; 95% confidence interval: 1.12-4.60). Despite a generally favorable response to antiviral therapy (79%), the risk of recurrent CMV colitis remained quite high (57%). Most of the recurrences developed within 8 months (75%). Conclusions: True CMV colitis is a poor prognostic indicator among patients with UC flares. An effective strategy for managing recurrent CMV colitis is urgently needed (KCT0003296).