Search in the Abstract Database

Search Abstracts 2011

* = Presenting author

P066. Diagnostic tests for CMV colitis in inflammatory bowel diseases

L. Henning, P. Mundt, C. Büning, B. Siegmund, M. Zeitz, J.C. Preiss

Charité, Berlin, Germany

Aim: CMV colitis is a severe complication of inflammatory bowel diseases (IBD) that often leads to colectomy if unrecognized. On the other hand asymptomatic reactivation of CMV is also highly prevalent in IBD patients under immunosuppressive treatment. All studies investigating diagnostic tests of CMV colitis in IBD patients so far either did not take the possibility of asymptomatic reactivation into consideration or did not include true negative patients in order to calculate specificity. We wanted to calculate the statistical measures of common diagnostic tests in defining clinically relevant CMV colitis in IBD patients.

Material and Methods: We extracted data from all patients between 09/2006 and 09/2009 at the Charité Campus Benjamin Franklin and Campus Charité Mitte with ulcerative colitis, colitis Crohn or colitis unclassified, who were tested for CMV infection with at least one test. These included at our department all patients with moderate to severely active colitis, with steroid-refractory or -dependent disease and all hospitalized patients. We excluded patients with only mild symptoms from our analysis. In the absence of a true gold standard we defined clinically relevant CMV colitis using clinical outcome in addition to common diagnostic tests. This included long term course and especially response to antiviral or anti-inflammatory treatment – again taking advantage of standard procedures at our department to immediately reduce immunosuppressive treatment and to start antiviral treatment after diagnosis of CMV infection.

Results and Conclusion: We were able to include 109 patients, 9 out of those were considered to have had CMV colitis, 4 were considered to be false positive in at least one test. 2 patients with positive CMV PCR were not fully evaluable. Patients with CMV colitis were significantly more likely to be steroid-refractory, to be on immunosuppressive drugs, to have fever, and to have no leukocytosis. Calculated sensitivity and specificity for the diagnostic tests are shown in the table.

TestSensitivity (95% CI)Specificity (95% CI)
CMV PCR (750 copies/ml serum)1.00 (0.60–1.00)0.94 (0.86–0.98)
CMV PCR (2500 copies/ml serum)0.89 (0.57–0.98)0.98 (0.92–0.99)
CMV PCR (5000 copies/ml serum)0.78 (0.45–0.94)0.99 (0.95–1.00)
pp65 FACS Assay0.50 (0.03–0.97)1.00 (0.66–1.00)
Immunohistochemistry0.67 (0.24–0.94)0.94 (0.80–0.99)
Histology (H&E)0.17 (0.01–0.64)0.98 (0.90–0.99)

There is a strong correlation between the amount of CMV DNA in the serum and the diagnosis CMV colitis, yet a few patients have low systemic CMV replication nevertheless. CMV PCR with 1000 copies/ml serum seems to be just as specific as histology, so in patients without CMV colitis CMV PCR is almost always below that threshold. Therefore a patient with active colitis and a clear positive finding in either of these diagnostic tests should be treated with antiviral medication.