75
Maternal HSV-2 Cervicovaginal Shedding Increases the Risk of Intra-partum HIV-1 Transmission
Sara Whitehead*1,2, Sara Whitehead*1,2, L Bollen1, W Leelawiwat1, P Mock1, S Asavapiriyanont3, A Chalermchokecharoenkit4, N Vanprapar4, T Chotpitayasunondh3, N Shaffer2, and R Chuachoowong1
1Thailand Ministry of Publ Hlth-CDC Collaboration, Nonthaburi; 2CDC, Atlanta, GA, US; 3Queen Sirikit Natl Inst for Child Hlth, Ministry of Publ Hlth, Bangkok, Thailand; and 4Faculty of Med, Siriraj Hosp, Mahidol Univ, Bangkok, Thailand
Background: Although the importance of herpes simplex type
2 (HSV-2) for HIV-1 sexual transmission has been extensively documented, its
role in mother-to-child transmission (MTCT) is unknown. Our study sought to
determine whether peri-partum genital HSV-2 shedding was associated with intra-partum
HIV transmission.
Methods: We evaluated stored specimens from women who
participated in an MTCT HIV transmission prevention trial in Thailand,
randomized to zidovudine (AZT) or placebo from 36 weeks’ gestation through
delivery; this was a non-breastfeeding population. Maternal plasma and
cervicovaginal lavage (CVL) specimens were collected at 38 weeks’ gestation and
infant plasma at birth and 2 months; HIV viral loads were determined using the
Amplicor version 1.5 assay. Transmission in
utero was defined as the infant’s positive results on polymerase chain
reaction (PCR) for HIV at birth (within 72 hours after delivery), and intra-partum
transmission as the infant’s negative PCR results at birth but positive PCR
results at the 2-month visit. Women who transmitted HIV in utero were excluded from this analysis. Maternal HSV-2 serology
was evaluated using an ELISA assay. HSV-2 shedding in CVL was determined with a
quantitative TaqMan-based real-time PCR assay.
Results: Among 281 eligible women with available
specimens, 207 (73.7%) were HSV-2 seropositive and 21 (7.5%) had detectable
HSV-2 DNA in CVL specimens at 38 weeks’ gestation. Women shedding HSV-2 were
more likely to transmit HIV intra-partum, with 6 (28.6%) of 21 shedders vs 22 (8.5%)
of 260 non-shedders transmitting (OR = 4.3, 95%CI 1.5 to 12.3). Median plasma HIV
at 38 weeks’ gestation was higher (4.2 vs 4.0 log10
copies/mL,
p = 0.04), and CVL detection of HIV
was more likely (62% vs 34%, p = 0.005),
among women shedding vs not shedding HSV-2. In a multivariate model adjusting
for plasma HIV viral load, CVL HIV viral load, and treatment group, HSV-2
genital shedding remained significantly associated with intra-partum
transmission (AOR = 3.2, 95%CI 1.1 to 9.3).
Conclusions: Our data show for the first time that perinatal
CVL HSV-2 shedding is associated with increased risk of intra-partum HIV
transmission, and that the effect on transmission was independent of CVL and
plasma HIV viral load. These findings suggest that interventions to control
HSV-2 shedding could further decrease intra-partum HIV transmission.
|
