Search in the Abstract Database

Abstracts Search 2019

P139 Advance of medical therapies may improve outcome of ulcerative colitis with cytomegalovirus infection

H. Kitamoto*1, S. Yamamoto1, M. Matsuura1, Y. Honzawa1, S. Yamada1, M. Okabe1, H. Seno1

1Graduate School of Medicine, Kyoto University, Department of Gastroenterology and Hepatology, Kyoto, Japan


Cytomegalovirus (CMV) reactivation often makes ulcerative colitis (UC) refractory. Despite recent advance of medical treatment for UC, few studies evaluated whether change of UC management affected clinical course of UC with CMV infection.


A total of 140 CMV-IgG positive UC patients, who underwent colonoscopy with the polymerase chain reaction assay using colonic biopsy specimen (mucosal-PCR) to investigate CMV reactivation between October 2003 and December 2017, were enrolled in this retrospective observational study. We divided those patients into two cohorts, the early (October 2003–June 2009, n = 44) and the late period (July 2009–December 2017, n = 96), according to the timing of colonoscopy. We compared cumulative colectomy-free rate between two periods.


There was no significant difference in baseline characteristics between two groups. The 5-year cumulative colectomy-free rate in the late period was higher than that in the early period (72.4% vs. 91.2%; p < 0.05, Figure 1). Of note, while approximately 70% of CMV seropositive patients had CMV reactivation in the early cohort, less than half patients did in the late cohort (68.2% vs. 42.7%; p < 0.05). Significantly less patients in the later period received corticosteroids at enrolment compared with those in the early period (40.9% vs. 22.9%; p < 0.05). Usage of other immunosuppressant including tacrolimus, TNF-α antagonist, and thiopurine at baseline was similar between two groups. The proportion of patients with initiation or dose escalation of corticosteroids after colonoscopy was significantly lower in the late period than in the early period (27.3% vs. 12.5%; p < 0.05). Tacrolimus was also administered after colonoscopy less frequently in the late period than in the early period (47.7% vs. 30.2%; p < 0.05). Furthermore, the 5-year cumulative colectomy-free rate of patients with CMV reactivation in the late period was higher than that in the early period (66.0% vs. 92.7%; p < 0.05, Figure 1), although anti-viral therapy was more frequently performed in the early period than the late period (80% vs. 22.0%; p < 0.01). The proportion of patients who received corticosteroids after CMV reactivation were significantly lower in the late period (60% vs. 29.3%; p < 0.01).

Figure 1. Cumulative colectomy-free rate.


Our results suggest that recent medical management for UC patients, especially optimisation of corticosteroid use, could not only decrease CMV reactivation, but avoid colectomy in patients with CMV reactivation, both of which results in improvement of the long-term outcome of UC patients with CMV seropositivity.