HIV in Genital Secretions

Abstract & Commentary

Synopsis: The presence of HIV RNA in cervical vaginal specimens bears no relationship to menstruation; it does, however, appear to be closely related to the level of plasma HIV RNA.

Source: Goulston C, et al. Human immunodeficiency virus in plasma and genital secretions during the menstrual cycle. J Infect Dis 1996;174:858-861.

Goulston and colleagues at stanford university examined the relationship between the levels of plasma HIV RNA and the presence of HIV RNA and pro-viral DNA, as well as HIV-specific IgA antibody, in cervical and vaginal specimens. Cervical and vaginal swabs, as well as cervicovaginal lavage specimens, were obtained weekly for eight weeks in six women. None of the women had evidence of cervicitis or positive cultures for chlamydia or N. gonorrheae. Two women had CD4+ counts greater than 500, two had 300-500, and two had 100-300 cells/mm3. All but one patient was asymptomatic; three women were receiving one or more antiretroviral agents, and three were receiving no antiretroviral therapy. Quantitative plasma HIV, as measured by RT PCR, ranged from 1.48 to 5.58 log10 copies/mL.

HIV RNA or pro-viral DNA was detected at some time point in four of six women. Cervical HIVRNA was detected in three women; 100% of the weekly specimens were positive in the two women with the highest levels of plasma RNA (> 105 copies/mL), whereas only six of eight weekly specimens were positive in a third woman (who had the next highest level of plasma RNA). Pro-viral DNA was infrequently detected in three women, two of whom were also positive for cervical HIV RNA. Only 16 of 144 cell pellets from genital specimens were positive for DNA, including seven of 48 cervical swabs, eight of 48 cervicovaginal lavage specimens, and one of 48 vaginal swabs.

None of the cultures of cellular supernatant fractions of cervicolavage was positive for HIV, and HIV-specific IgA antibodies were detected infrequently.


While the presence of HIV in genital and vaginal secretions has been found to be associated with cervical mucopus, cervical inflammation ectopy, pregnancy, and oral contraceptive use, its relationship with menstruation has not been previously well-characterized. These data suggest that the presence of HIV RNA in cervical and vaginal specimens bears no relationship to menstruation. It does, however, appear to be closely related to the level of plasma HIV RNA, where it is found most reliably in women with high levels of circulating virus (> 4-5 log10/mL), and is generally 10-fold lower than that of plasma. This finding is supported by other data demonstrating a correlation between HIV RNA in plasma and vaginal secretions (R = 0.6; P < 0.001), both of which were inversely correlated with CD4+ count (Abstract #25, Fourth Conference on Retroviruses and Opportunisitic Infections, 1997).

The presence of HIVin genital secretions may be a determinant of vertical transmission of virus during pregnancy. In a recent study of 223 HIV-1 seropositive pregnant women in Nairobi, 32% of cervical and 10% of vaginal specimens were positive for HIV DNA (John GC, et al. J Infect Dis 1997;175:57-62). Cervical and vaginal shedding of virus was associated with abnormal vaginal discharge, cervical mucopus, low CD4 counts, severe vitamin A deficiency, as well as anemia, the latter of which was possibly the result of advanced HIV disease. Women with cervical shedding of HIV-1 infected cells were also more likely to shed virus in breast milk. The authors believed that such women were possibly more likely to shed virus from a greater number of locations as a result of higher total virus burden.

Increased transmission risk has also been shown to be associated with higher levels of plasma HIV RNA at the time of delivery, but even women with relatively low levels of virus in plasma could transmit virus (Sperling RS, et al. N Engl J Med 1996;335:1621-1629; see also Infect Dis Alert 1997;16:51-52). While transmission rates were as high as 41.7% for those women with the highest viral loads (> 15,700 copies/mL by RT PCR) who were not receiving antiretroviral therapy, even 2.6% vs. 7.1% of those women with the lowest levels of virus (< 1730 copies/mL), who were or were not, respectively, receiving AZT, were transmitters. The fact that women with even relatively low levels of plasma HIV RNA can vertically transmit virus suggests that factors other then this must play a significant role in cervical shedding of HIV in pregnancy.