Cytomegalovirus (CMV) enters latency following primary infection and can subsequently reactivate. Reinfection with a different viral strain can also occur. During these events, CMV is shed in bodily fluids. This study examined correlates of CMV shedding in specimens obtained from the HIV Epidemiology Research Study, a multicenter cohort study of US women with or at high risk for human immunodeficiency virus (HIV) infection. Among the women studied, 91.4 % (911/997) were CMV IgG seropositive. Of these women, 2.7 % (25/911) were CMV IgM seropositive. CMV DNA was detected via real-time PCR more frequently in cervicovaginal lavage (CVL) specimens (55/764, 7.2 %) than in peripheral blood mononuclear cells (PBMCs) (26/897, 2.9 %). CMV viral loads in 1 ml CVL (median 534; mean 2598; range540-74 844) were higher than in 10 6 PBMCs (median 264; mean 1287; range535-13 250). CMV DNA in PBMCs was associated with HIV seropositivity [odds ratio (OR) 13.5; 95 % confidence interval (CI) 1.8-100], increasing HIV viral load (P,0.001 for trend), decreasing CD4 cell counts (P,0.001 for trend) and CMV DNA in CVL (OR 26;. CMV DNA in CVL specimens was associated with CMV IgM seropositivity (OR 4.3; 95 % CI 1.5-12.3), HIV seropositivity (OR 7.3; 95 % CI 2.6-20), increasing HIV viral load (P,0.001 for trend) and decreasing CD4 cell counts (P,0.001 for trend). The positive predictive value of CMV IgM seropositivity for CMV DNA shedding in either PBMCs or CVL was 20 %. In summary, CMV shedding in CVL and PBMCs was highly correlated with each other and with markers of immune suppression. However, the occurrence of CMV shedding in bodily fluids is poorly understood. In particular, it is important to gain a better understanding of the frequency of CMV shedding in adults, the likelihood of shedding in different bodily fluids, the viral loads in these fluids and the factors associated with shedding. The purpose of this study was to address these questions in a large cohort of women of reproductive age.
INTRODUCTION
METHODSStudy design and population. Participants were enrolled in the HIV Epidemiology Research Study (HERS), a multicenter cohort study of women with or at high risk of HIV infection (Smith et al., 1997;Stover et al., 2003). Recruitment, screening and enrolment for the HERS took place from April 1993 to January 1995. Follow-up visits took place at 6 month intervals. Visits included an extensive interview, a general physical and pelvic examination, and biological specimen collection. For this study, participant visits were stratified on HIV status and CD4 cell count and were selected randomly, one visit per participant, with the sample enriched by a small group of participant visits from women who were pregnant. A total of 997 participant visits were included. Stored serum, peripheral blood mononuclear cells (PBMCs) and cervicovaginal lavage (CVL) specimens were retrieved subject to availability. Informed consent was obtained from all participants and the study was approved by the institutional review boards from each of the author...