Clinical significance of cytomegalovirus (CMV) pp65 antigenemia in the prediction of CMV infection during immunosuppressive therapy for rheumatic disease.

To investigate the risk factors for CMV infection and to clarify the cut-off count of CMV pp65 antigenemia predicting clinical symptoms related to CMV infection in patients with rheumatic disease. We retrospectively analyzed 261 patients with rheumatic disease who were treated with immunosuppressive therapy. CMV infection was defined as positive > 1 CMV-positive cell per two slides (CMV pp65 antigenemia C10/C11). Patients with CMV infection were divided into two groups based on the presence of antiviral treatment for CMV disease. We determined a cut-off value of CMV-positive cells for the diagnosis of CMV disease. CMV infection was observed in 141 cases (54%). In a multivariate analysis, CMV infection was associated with three following factors: Age > 60 years (OR 1.87 [95% CI 1.04-3.36]); lymphocyte counts < 1000/μL (OR 3.34 [95% CI 1.88-6.05]); steroid pulse therapy (OR 2.60 [95% CI 1.27-5.55]). The cut-off level of CMV pp65 antigenemia indicating CMV disease was five positive cells average two slides by using receiver operating characteristic curve analysis (AUC 0.95, sensitivity 0.94, specificity 0.80). Age > 60 years, lymphocytopenia (< 1000/μL) and steroid pulse therapy are risk factors of CMV infection. We recommend that CMV pp65 antigenemia of > 5 cells average two slides (C10/C11) in patients with rheumatic disease should be the treated with antiviral drugs.