Elsevier

The Lancet

Volume 350, Issue 9082, 27 September 1997, Pages 922-927
The Lancet

Articles
Hormonal contraception, vitamin A deficiency, and other risk factors for shedding of HIV-1 infected cells from the cervix and vagina

https://doi.org/10.1016/S0140-6736(97)04240-2Get rights and content

Summary

Background

Factors that influence shedding of HIV-1 infected cells in cervical and vaginal secretions may be important determinants of sexual and vertical transmission of the virus. We investigated whether hormonal contraceptive use, vitamin A deficiency, and other variables were risk factors for cervical and vaginal shedding of HIV-infected cells.

Methods

Between December, 1994, and April, 1996, women who attended a municipal sexually transmitted diseases (STDs) clinic in Mombasa, Kenya, and had previously tested positive for HIV-1, were invited to take part in our cross-sectional study. Cervical and vaginal secretions from 318 women were evaluated for the presence of HIV-1 infected cells by PCR amplification of gag DNA sequences.

Findings

HIV-1 infected cells were detected in 51% of endocervical and 14% of vaginal-swab specimens. Both cervical and vaginal shedding of HIV-1 infected cells were highly associated with CD4 lymphocyte depletion (p=0·00001 and p=0·003, respectively). After adjustment for CD4 count, cervical proviral shedding was significantly associated with use of depot medroxyprogesterone acetate (odds ratio 2·9, 95% Cl 1·5–5·7), and with use of low-dose and high-dose oral contraceptive pills (3·8, 1·4–9·9 and 12·3, 1·5–101, respectively). Vitamin A deficiency was highly predictive of vaginal HIV-1 DNA shedding. After adjustment for CD4 count, severe vitamin A deficiency, moderate deficiency, and low normal vitamin A status were associated with 12·9, 8·0, and 4·9-fold increased odds of vaginal shedding, respectively. Gonococcal cervicitis (3·1, 1·1–9·8) and vaginal candidiasis (2·6, 1·2–5·4) were also correlated with significant increases in HIV-1 DNA detection, but Chlamydia trachomatis and Trichomonas vaginalis were not.

Interpretation

Our study documents several novel correlates of HIV-1 shedding in cervical and vaginal secretions, most notably hormonal contraceptive use and vitamin A deficiency. These factors may be important determinants of sexual or vertical transmission of HIV-1 and are of public health importance because they are easily modified by simple interventions.

Introduction

Heterosexual transmission of HIV-1 is the predominant mode of infection among adults worldwide, while mother-to-child transmission accounts for most of the HIV-1 infection in children.1, 2 Of children who acquire HIV-1 from their mothers, 40–80% are estimated to become infected during delivery.2 For both heterosexual and mother-to-child transmission, factors that affect shedding of HIV-1 in cervical and vaginal secretions may be important determinants of transmission risk.

HIV-1 has been detected in cervical and vaginal secretions, as both the cell-associated and cell-free virus, by culture, PCR assays, and immunohistochemical techniques.3, 4, 5, 6, 7, 8 Correlates of HIV-1 shedding in the female genital tract have been addressed in only a few studies,4, 5, 6, 7 most of which have involved fewer than 100 women, and have had insufficient power to assess important potential correlates. Use of oral contraceptives, pregnancy, cervical ectopy, and cervical inflammation have been associated with detection of HIV-1 infected cells or free virus in some studies, but with conflicting results.4, 5, 6

John and co-workers'7 study of pregnant women infected with HIV-1 showed that genital-tract shedding of virus-infected cells was associated with CD4 depletion and abnormal cervical or vaginal discharge. The aim of our study was to assess whether use of oral and injectable hormonal contraceptives, immunosuppression, serum concentration of vitamin A, specific sexually transmitted diseases (STDs), and other variables were risk factors for cervical and vaginal shedding of HIV-1-infected cells.

Section snippets

Methods and patients

Between December, 1994, and April, 1996, we did a cross-sectional study of women who were seropositive for HIV-1 and attended a municipal STD clinic in Mombasa, Kenya. The clinic is a referral centre for women with symptoms of STDs and is also attended by prostitutes who receive regular government-mandated screening for STDs. Many of the women have been tested for HIV-1 antibodies in the context of another project at the clinic. Any woman not previously tested for HIV-1 was offered counselling

Microbiology and blood tests

We studied gram-stained vaginal secretions for bacterial vaginosis and scored them according to a ten-point system.9 We recorded the number of polymorphonuclear leucocytes in three non-adjacent, high-power fields of vaginal and cervical gram stains, and the presence of spermatozoa. Lactobacillus sp were cultured on Ragosa agar (Difco Laboratories, Detroit, MI, USA), and isolates were tested for production of hydrogen peroxide.10 We cultured T vaginalis in Diamonds modified medium (JRH

HIV-1 DNA detection

Cryovials that contained cervical and vaginal swabs were shipped to the University of Washington on dry ice and stored at −70°C. The samples were prepared and tested for HIV-1 DNA by nested PCR amplification of the gag gene, as described previously.7, 14 This assay is able to detect a single copy of HIV-1 DNA.14

Data analysis

Mann-Whitney U test, χ2 test, Fisher's exact test, and Spearman's correlation coefficient were used for initial univariate comparisons. Univariate tests of linear trend in the odds of HIV-1 DNA shedding with increasing or decreasing levels of exposure to HIV-1 were also done for selected variables.15 After univariate and stratified analyses, we used logistic regression to obtain adjusted estimates for key predictors—use of hormonal contraceptives, serum concentrations of vitamin A, and CD4

Results

We enrolled 318 women who were seropositive for HIV-1 (median age 28 [range 18–46] years; median number of sexual partners in the previous week 1 [0–6]). 223 (70%) women were prostitutes. There were no significant differences in mean number of sexual partners, condom use, or prevalence of STDs between prostitutes and the other women (data not shown). 182 (57%) women used no form of contraception at the time of assessment; 55 (17%) were using depot medroxyprogesterone acetate (DMPA); 25 (8%)

Endocervical shedding of HIV-1-infected cells

HIV-1 DNA was detected in 161 (51%) endocervical swabs from 315 women. Cervical swabs were unavailable from the three women who had had hysterectomies. Univariate correlates of cervical and vaginal shedding are shown in the table. Endocervical HIV-1 DNA was detected significantly more frequently in women who used oral contraceptives than in those who did not use hormonal contraceptives (78% vs 46%, p=0·0005). When stratified by non-users, use of low-dose oral contraceptives, and use of

Vaginal shedding of HIV-1 infected cells

HIV-1 DNA was detected in 44 (14%) vaginal swabs from 318 women. Vitamin A deficiency was highly associated with vaginal shedding of HIV-1 DNA; serum vitamin A concentrations less than 0·70 μmol/L, 0·70–1·04 μmol/L, 1·05–1·39 μmol/L, and 1·40 μmol/L or more were associated with 29%, 17%, 11%, and 2% prevalences of vaginal shedding, respectively (test for trend across four levels of exposure, p=0·00001). CD4 lymphocyte counts less than 200/μL, 200–499/μL, and 500/μL, or more were associated with

Discussion

The effects of hormonal contraceptives on cervical and vaginal shedding may reflect direct effects on the virus, effects on immune modulation of virus replication, or effects on local genital tract physiology. For example, steroid hormones have been shown to bind to the regulatory sequences of HIV-1 and upregulate expression of the virus.17, 18, 19 There is also growing evidence that endogenous oestrogens and progestins play a part in regulation of both hormonal and cell-mediated immunity.20, 21

References (36)

  • J Kreiss et al.

    Association between cervical inflammation and cervical shedding of HIV DNA

    J Infect Dis

    (1994)
  • GC John et al.

    Genital shedding of human immunodeficiency virus type 1 DNA during pregnancy: association with immunosuppression and abnormal cervical or vaginal discharge, and severe vitamin A deficiency

    J Infect Dis

    (1997)
  • P Van de Perre et al.

    Detection of HIV p17 antigen in lymphocytes but not epithelial cells from cervicovaginal secretions of women seropositive for HIV: implications for heterosexual transmission of the virus

    Genitourin Med

    (1988)
  • RP Nugent et al.

    Reliability of diagnosing bacterial vaginosis is improved by a standardized method of gram stain interpretation

    J Clin Microbiol

    (1991)
  • DA Eschenbach et al.

    Prevalence of hydrogen peroxide-producing Lactobacillus species in normal women and women with bacterial vaginosis

    J Clin Microbiol

    (1989)
  • A Landay et al.

    A rapid manual method for CD4+ T- cell quantitation for use in developing countries

    AIDS

    (1993)
  • A Sommer et al.

    Vitamin A deficiency: health, survival, and vision

    (1996)
  • RD Semba

    Vitamin A, immunity, and infection

    Clin Infect Dis

    (1994)
  • Cited by (0)

    View full text