Search in the Abstract Database

Abstracts Search 2019

P181 Cytomegalovirus infections are rare in hospitalised patients with flares of inflammatory bowel disease—a monocentre retrospective cohort study

L-V. Lorenz*1, C. Monasterio1, A-M. Globig1,2, P. Hasselblatt1

1Medical Centre – University of Freiburg, Department of Medicine II, Freiburg, Germany, 2Faculty of Medicine, University of Freiburg, Berta-Ottenstein-Programme, Freiburg, Germany


Cytomegalovirus (CMV) infection may complicate or mimic acute flares of inflammatory bowel disease (IBD). However, there are conflicting data regarding its prevalence and the optimal screening strategies in patients with severe IBD flares.


We performed a retrospective chart analysis of patients admitted to our department for IBD flares between 2010 and 2017. To identify potential risk factors for CMV infection, associations between clinical and laboratory parameters and proven CMV infection (as defined by positive CMV PCR from plasma or intestinal biopsies and/or significant expression of cytomegaloviral proteins as determined by immunohistochemistry) were analysed by univariable logistic regression analysis and calculated as odds ratios (OR) and 95% confidence intervals.


In total, 495 hospital admissions for flaring IBD were identified. CMV testing was performed in 238 patients (43% male, 52% with ulcerative colitis [UC] and 48% with Crohn’s disease [CD]). Twenty-two per cent of patients had a history of extraintestinal manifestations while 57% had previously been treated by immunomodulators or biologics. CMV infection was diagnosed in 13/238 patients (5.5%). Only 7 of these 13 patients had steroid-refractory disease. However, only 5 patients (2.1%) were considered to have clinically significant CMV infection and received antiviral therapy. Univariable regression analysis revealed that CMV infection was negatively associated with a diagnosis of CD when compared with UC (OR: 0.08 [0.01–0.6]), with increasing haemoglobin concentrations (OR: 0.47 [0.22–0.98] per increase of 3 g/dl) and serum albumin concentrations (OR: 0.25; [0.08–0.78] per increasing g/dl). Moreover, CMV infection was associated with the presence of subfebrile temperatures (37.1–38.4°C, OR: 6.31 [1.28–31.2]) or fever on admission (≥ 38.5°C; OR: 9.85 [1.71–56.6]). We did not observe significant associations of CMV infection with severity of inflammation or ulcer depth as determined by endoscopy, age or dosage of concomitant corticosteroid therapy. 2/13 patients with proven CMV infection relapsed at later time points and the percentage of CMV-infected patients requiring proctocolectomy during follow-up was moderately, albeit not significantly elevated.


Based on the results of this retrospective cohort study, the prevalence of CMV infection appears to be low in patients hospitalised for IBD flares. CMV infection was associated with a diagnosis of UC, fever on admission and blood test results indicative of malnutrition or disease severity such as anaemia or albumin deficiency. CMV testing may therefore focus on (sub-)febrile UC patients with severely active disease.