[Application of flow cytometry to detect PP65 antigenemia for diagnosis and monitoring of human cytomegalovirus infection].

Objectives: To evaluate the clinical significance of flow cytometry (FCM) to detect the cytomegalovirus (CMV) PP65 antigen in patients with CMV infection.

Methods: Samples from 35 patients without CMV infection were used as negative control. The definite diagnosis of CMV infection was based on the national criteria for CMV infection. All 136 patients with CMV infection were examined with the FCM to detect CMV PP65 antigen, real-time fluorescence quantitative-polymerase chain reaction assay (RFQ-PCR) to detect CMV-DNA and ELISA to measure the serum level of IgM antibody against CMV. The results of these 3 assays in 2 groups (isolated organ involvement and disseminated diseases) were compared and the significance of PP65 antigenemia was evaluated. A short-term follow-up was undertaken in 18 patients.

Results: The percentages of PP65 positivity in blood mononuclear cells (MNC) and polymorphic nuclear leukocyte (PMNL) from 35 negative control patients were 0.21% +/- 0.09% with a range of 0 - 0.41% and 0.24% +/- 0.10% with a range of 0.12% - 0.48%, respectively, which were not significantly different (t = 0.425, P > 0.05). The 95(th) percentiles (P(95)) of PP65 in MNC and PMNL were 0.39% and 0.45%, respectively, so a cutoff value of >/= 0.50% was set. Of the 136 patients with CMV infection, 118 samples from 118 patients were positive for PP65 antigenemia with a positive rate of 86.8%, which was not statistically different from that (90.4%, chi(2) = 0.91, P > 0.05) of CMV-DNA detected by RFQ-PCR assay but it was significantly higher than that (45.6%, chi(2) = 51.50, P < 0.005) of the detection by IgM measurement. PP65 detection was correlated with urine CMV DNA amplification (chi(2) = 63.78, P < 0.01) while the different detection rates between the two assays were not statistically significant (chi(m)(2) = 1.78,P > 0.05). PP65 detection was not correlated with serum IgM measurement while the detection rates between the two were significantly different (chi(m)(2) = 52.92,P < 0.01). No significant difference was found between the detection rates of CMV infection in MNC (45/53, 84.9%) and PMNL (43/53, 81.1%) (chi(m)(2) = 0.25, P > 0.05). Higher PP65 antigenemia level was correlated with systemic CMV infection, while lower level of PP65 was either in the patients with isolated organ involvement by CMV (chi(2) = 38.51, P < 0.005) or less severe in patient's situation. PP65 antigenemia of CMV infection returned to lower level or negative in recovery stage and increased when condition of patients deteriorated.

Conclusions: PP65 antigenemia detection by FCM is effective in the diagnosis of the active CMV infection. Quantitative monitoring of PP65 antigenemia is useful in the evaluation of patients with CMV infection.