The diagnosis of CMV colitis in UC patients can be confirmed by histology, which comprises either PCR or immunohistochemistry (IHC).5 However, tissue PCR for CMV is controversial because of the possibility for false positives.6 Although IHC staining may be the best method to confirm CMV colitis,7,8,9 it takes several days for this test to yield results, and could be associated with hemorrhage or perforation.10,11,12 Many clinicians have tried to identify the best method of diagnosing CMV colitis. With this in mind, I read the study of Chun et al.13 with great interest. They assessed the clinical utility of the CMV antigenemia assay for the diagnosis of CMV colitis and predicting the outcomes of patients with UC. According to their results, CMV colitis was more common in patients with positive CMV antigenemia (eight patients, 66.7%), and this association was statistically significant (P=0.001). The results were consistent with other similar studies, which showed low sensitivity and high specificity.12,14 They suggested 2 pp65-positive cells per 2×10⁵ polymorphonuclear leukocytes as a cut-off value, and this is very meaningful.

However, among patients with positive antigenemia assay results, three did not have CMV colitis and four with negative results had CMV colitis. In addition, ganciclovir was considered only in tissue-confirmed cases with steroid refractoriness. I am therefore curious as to whether these results have any clinical significance in the diagnosis of CMV colitis. Can the antigenemia assay substitute histologic evaluation? I believe that Chun et al. did not clearly demonstrate the significance of the CMV antigenemia assay without histology in the diagnosis of CMV colitis. Even in cases where the antigenemia assay was negative, they considered antiviral therapy if tissues showed CMV colitis. Therefore, I think we eventually need endoscopy and tissue confirmation in steroid refractory UC patients to decide upon antiviral therapy. One study suggested that the CMV antigenemia assay has a diagnostic role only in conjunction with endoscopic findings.14 I agree with this opinion; however, considering the high rate of false positive results of PCR for CMV, it could be helpful for diagnosis only when tissue PCR is positive and IHC staining is negative.

The role of CMV in the exacerbation of UC is still under debate.6 Several studies have reported that CMV reactivation could be improved without antiviral treatment,15,16 and others have suggested that in patients who are resistant to corticosteroids, antiviral therapy should be considered if they are positive for CMV in tissue studies.17 Chun et al. also demonstrated a significant association of CMV antigenemia with steroid refractoriness, and suggested early rescue therapy.13

CMV antigenemia could be a good predictor of clinical outcomes such as steroid refractoriness. However, it is limited in the diagnosis of CMV colitis. Despite these limitations, I think it could be very helpful in some cases. Therefore, we need to define the proper indication of the CMV antigenemia assay in the diagnosis of CMV colitis.