Prevalence of Cytomegalovirus Infection in Steroid-refractory Crohnʼs Disease

To the Editor: Previous studies have suggested an association between cytomegalovirus (CMV) infection and steroid-refractory inflammatory bowel disease (IBD).1,2 Recently, we demonstrated that the CMV DNA load determined in intestinal tissue was predictive of resistance to steroid treatment and to three-drug regimens in ulcerative colitis (UC).3 Crohn's disease (CD) may also be associated with CMV infection4,–7 but in a proportion of patients that needs to be carefully evaluated in order to refine the present recommendations regarding the diagnosis and treatment of CMV infection in IBD. The aim of the present study was therefore to assess the prevalence of CMV infection in patients with steroid-refractory CD when compared to that reported in UC patients. From January to November 2009, a single-center, prospective, observational study was carried out in 60 consecutive adult patients exhibiting moderate to severe acute flare of IBD. All patients were included after they gave written informed consent. The diagnosis of IBD was based on clinical, endoscopic, radiologic, and histological criteria. The clinical activity was appreciated by Lichtiger and Crohn's Disease Activity Index (CDAI) scores in UC and CD, respectively. In CD patients, a moderate and severe colitis was defined by a CDAI score comprised between 300 and 400, and >400, respectively. All included patients underwent a blood sample collection and colonoscopy or flexible sigmoidoscopy less then 24 hours after inclusion. Intestinal biopsies were sampled during endoscopy from the mucosal lesions. A single, trained pathologist (M.L.C.), blind for clinical, endoscopic, and virological results, screened the biopsies (10 to 20 high-power fields, depending on the biopsy size) for the presence of inclusion bodies with hematoxylin and eosin staining (H&E), and of markers of CMV infection by immunohistochemistry (IHC) with anti-CMV monoclonal antibodies (Argene Biosoft, France). A severe colitis was defined by a CDEIS score >20. Systematic virological studies included the detection of IgM and IgG antibodies to CMV in blood (Siemens, Erlangen, Germany), the measure of CMV DNA load in whole blood using the CMV R-gene kit (Argene) as previously described8 and the quantitative determination of CMV DNA by real-time polymerase chain reaction (PCR) on frozen biopsy fragments using the same kit (Argene) as previously described.3 All molecular data were interpreted by a single virologist (S.P.). The results in intestinal biopsies were expressed in copies of CMV genome per mg of tissue. Extraction and amplification steps were controlled for the absence of Taq polymerase inhibitors by amplification of the human aldolase gene as previously described.9 A total of 60 consecutive patients were enrolled in the study, including 42 UC cases and 18 CD cases. The characteristics of the UC patients have been reported.3 Those of the 18 CD patients were as follows: mean age of 34.9 years (SD: 14.2), male/female sex ratio of 0.9, mean disease duration of 77 months (SD: 68), and mean CDAI score of 360 (SD: 60). According to the Montreal Classification, the following percentages were observed for CD patients: 23 and 77 for A1 and A2 phenotypes, respectively, 44, 33, and 22 for B1, B2, and B3 phenotypes, respectively, and 22, 28, and 50 for L1, L2, and L3 phenotypes, respectively. Eight UC patients received azathioprine 12 ± 3 months prior to inclusion. IgG antibodies to CMV were present in 10 of the 18 patients for CD when compared to 25 of 42 for UC patients3 (not statistically different). CMV DNA at the rate of 10 copies/mg of tissue was detected in inflamed tissue of 18 patients (30.0%), two of them being CD patients. The prevalence of CMV infection at the tissue level was higher in UC patients (16 of 42, 38.1%)3 than in CD (2 of 18, 11.1%) (P < 0.05). Histological markers of CMV infection, analyzed by both H&E staining and immunochemistry, were detected in none of the biopsies from any patients. The clinical evolution of the two CD patients with CMV was favorable without antiviral therapy. A second set of biopsies was performed in these two patients: the CMV load was stable in one case and under the rate of 10 copies/mg of tissue in the other one. Table 1 summarizes the results of several studies (including the present one) that, by using different methods for detecting CMV at the intestinal level, reported a trend to a clear difference in the prevalence of CMV infection in IBD, which was shown to be much higher in UC than in CD.3,6,7,10,–12 The explanation of the low prevalence of CMV infection in patients with CD in comparison with UC probably relates to the differences in the immune mechanisms underlying the two diseases. It is well known that tumor necrosis factor plays an important role in reactivation of CMV in both monocytes and dendritic cells.13 Conversely, interferon-gamma (IFN-γ) productive T lymphocytes are protective against CMV infection.14 As CD is considered a Th1-type inflammatory process and thus associated with great expression levels of IFN-γ,15 this may partly explain the lower incidence of CMV infection in patients affected by that disease. Detection of Cytomegalovirus (CMV) in Intestinal Tissue of Patients with Inflammatory Bowel Disease According to Recent Publications Detection of Cytomegalovirus (CMV) in Intestinal Tissue of Patients with Inflammatory Bowel Disease According to Recent Publications The recent European recommendations for the prevention, diagnosis, and management of opportunistic infections in IBD did not distinguish UC from CD patients.16 For both types of IBD, CMV colitis should be excluded by using tissue PCR or IHC in immunomodulatory refractory cases of IBD before increasing immunomodulatory therapy. The present results together with those listed in Table 1 confirm that CMV infection may often play a role in the exacerbation of UC but not of CD. Consequently, it would be logical to improve the current recommendations with a clear indication for the detection of CMV infection in refractory UC, whereas the usefulness of CMV infection diagnosis in patients with refractory CD has yet to be determined.