Search in the Abstract Database

Abstracts Search 2014

* = Presenting author

P269. Characteristic endoscopic findings and risk factors for cytomegalovirus colitis in patients with active ulcerative colitis

Y. Hirayama, T. Ando, K. Ishiguro, O. Maeda, O. Watanabe, K. Morise, K. Maeda, M. Matsushita, K. Furukawa, K. Funasaka, M. Nakamura, R. Miyahara, H. Goto, Nagoya University Graduate School of Medicine, Departmant of Gastroenterology and Hepatology, Nagoya, Japan


Despite a wide range of treatment options, a significant proportion of patients with ulcerative colitis (UC) remain refractory to medical therapy. In patients with UC, cytomegalovirus (CMV) frequently plays an important role as an agent of colitis, and sometimes interferes strongly with the natural progression of UC. We conducted this study in an attempt to clarify the characteristic endoscopic findings and risk factors for CMV colitis in patients with active (moderate to severe) UC.


The records of 149 consecutive patients hospitalized for active UC from January 2004 to December 2012 were reviewed. Diagnosis of CMV colitis in patients with UC was based on clinical, endoscopic, and histologic findings. CMV positive results were determined by the presence of CMV antigenemia positive or by the presence of CMV inclusion body in the biopsy specimen. Statistical analysis included univariate and multivariate studies of clinical and biological parameters.


128 patients (M/F: 69/59) who underwent colonoscopy for active UC were enrolled (21 patients were excluded). Age at 1st hospitalization was 38.5±16.6 years. Age at diagnosis was 32.4±16.2 years. Disease duration was 6.4±7.2 years. The disease location were 83 cases (64.8%) were pancolitis, 41 cases of left-sided colitis, and 4 cases of proctitis. The clinical courses were 89 cases (69.5%) of relapse and remission, 18 cases of chronically active colitis, and 21 cases of first attack. Between the patients with CMV colitis (n = 31) and without CMV colitis (n = 97), there was a statistically significant difference in terms of disease extension (pancolitis; p < 0.05), age at diagnosis [CMV(+): 44.4±18.7, CMV(−): 28.7±13.2, p < 0.05], the amount of prednisolone for a week [CMV(+): 244.7±84.7, CMV(−): 81.0±997, p < 0.05]. The analysis of risk factors for CMV colitis with univariate and multivariate studies was performed. Pancolitis (OR 3.65, 95% CI 1.32–9.94), >35 years old at the diagnosis (OR 5.65, 95% CI 2.23–12.30), and the significant amount of prednisolone use (>120 mg/week) (OR 37.6, 95% CI 9.20–150.99) were statistically significant. Endoscopic findings of punched-out ulcer (OR 15.4, 95% CI 5.9–40.3) and deep ulcer (OR 17.7, 95% CI 5.9–52.3) were characteristic of CMV colitis.


The risk factors that correlated with CMV colitis in patients with active UC were pancolitis, >35 years old at the diagnosis, and the significant amount of prednisolone use (>120 mg/week). Characteristic endoscopic findings of CMV colitis in patients with active UC were deep ulcer, and punched-out ulcer. Therefore, we should treat these patients carefully to prevent CMV colitis.