Gina A. Dallabetta

Bacterial vaginosis and disturbances of vaginal flora : association with increased acquisition of HIV

Background:Cross-sectional studies suggest an association between bacterial vaginosis (BV) and HIV-1 infection. However, an assessment of a temporal effect was not possible. Objectives:To determine the association of BV and other disturbances of vaginal flora with HIV seroconversion among pregnant and postnatal women in Malawi, Africa. Design:Longitudinal follow-up of pregnant and postpartum women. Methods:Women attending their first antenatal care visit were screened for HIV after counselling and obtaining informed consent. HIV-seronegative women were enrolled and followed during pregnancy and after delivery. These women were again tested for HIV at delivery and at 6-monthly visits postnatally. Clinical examinations and collection of laboratory specimens (for BV and sexually transmitted diseases) were conducted at screening and at the postnatal 6-monthly visits. The diagnosis of BV was based on clinical criteria. Associations of BV and other risk factors with HIV seroconversion, were examined using contingency tables and multiple logistic regression analyses on antenatal data, and Kaplan–Meier proportional hazards analyses on postnatal data. Results:Among 1196 HIV-seronegative women who were followed antenatally for a median of 3.4 months, 27 women seroconverted by time of delivery. Postnatally, 97 seroconversions occurred among 1169 seronegative women who were followed for a median of 2.5 years. Bacterial vaginosis was significantly associated with antenatal HIV seroconversion (adjusted odds ratio = 3.7) and postnatal HIV seroconversion (adjusted rate ratio = 2.3). There was a significant trend of increased risk of HIV seroconversion with increasing severity of vaginal disturbance among both antenatal and postnatal women. The approximate attributable risk of BV alone was 23% for antenatal HIV seroconversions and 14% for postnatal seroconversions. Conclusions:This prospective study suggests that progressively greater disturbances of vaginal flora, increase HIV acquisition during pregnancy and postnatally. The screening and treating of women with BV could restore normal flora and reduce their susceptibility to HIV.

HIV infection and disturbances of vaginal flora during pregnancy

Disturbances of vaginal flora are common among women of reproductive age. In areas of sub-Saharan Africa where the prevalence of HIV is high, the frequency of bacterial vaginosis (BV) is also high. In this study, we assessed the association of BV and other disturbances of vaginal flora with prevalent HIV infection in two cross-sectional studies among pregnant women in urban Malawi. The prevalence of HIV-1 was 23% in 1990 and 30% in 1993. Overall, 30% of the women had BV, 59% had mild or moderate disturbance of vaginal flora, and only 11% had normal vaginal flora. Increasing prevalence of HIV was significantly associated with increasing severity of disturbance of vaginal flora (p < .00001, chi2 trend test). This trend of increased prevalence persisted after controlling for concurrent sexually transmitted diseases (STDs), sexual activity, and socioeconomic factors. After multivariate adjustment for potential confounders, the odds ratio for the association of BV with prevalent HIV infection was 3.0 (95% confidence interval [CI], 2.4-3.8), that of moderate vaginal disturbance with HIV infection was 2.2 (95% CI, 1.7-2.8), and that of mild vaginal disturbance with HIV infection was 1.6 (95% CI, 1.3-2.1). Among women with BV, HIV infection was higher among younger women than older, implying more recent infection. Although these studies were cross-sectional, our data suggest that BV could be associated with increased susceptibility to HIV infection.

Hiv, maternal death and child survival in Africa

In Blantyre Malawi two observational longitudinal studies on HIV were conducted at an urban hospital to identify risk factors linked to mother-to-infant HIV transmission. Later a third longitudinal study began at the same urban hospital but it aimed to determine the efficacy of a birth canal wash to reduce peripartum HIV transmission. The researchers conducted a multiple logistic regression analysis using data from the first two cohort studies. They used data from all three studies to conduct a matched case-control analysis. These analyses aimed to determine the association of maternal death with child survival. 56 of 2829 mothers in the two observational studies and 437 children born during the study period died. 72% of the children died during the first year of life (median age = 5.8 months). Infants of mothers who died were more likely to die than infants whose mothers were alive (48.2% vs. 14.8%; odds ratio [OR] = 5.4; p < 0.001). In the third study 19 mothers and 8 infants (42.1%) died by the end of March 1995. 90% of the dead women in the three studies were HIV infected. 46.7% of these mothers infants died. The mothers of 37.1% of these infants died first. 57.1% died before their mothers some of whom died only days before their mothers death. 5.7% died on the same day as their mother. Maternal HIV positivity increased the risk of infant death (OR = 3.66 p < 0.0001). The infants risk of death was more than 3 times greater when the mother was dead than when she was alive (p < 0.001). Thus the mothers presence is essential for child survival. Interruption of breast feeding inadequate child care or increased mother-to-infant HIV transmission in late maternal HIV disease likely contribute greatly to the excess child mortality. Care for orphans or children who have lost their main caretaker will likely reduce child mortality in the HIV endemic.

Maternal vitamin A deficiency and child growth failure during human immunodeficiency virus infection.

Although vitamin A is thought to influence growth, the relationship between maternal vitamin A deficiency during pregnancy and child growth is unknown. A longitudinal cohort study of 467 HIV-infected women and their children was conducted in Blantyre, Malawi. The childrens weight and height were measured every 3 months until they were 24 months old. Maternal vitamin A deficiency was independently related to both linear and ponderal growth after adjustment for effects of body mass index, child gender, and child HIV status. By 12 months of age, infants born to mothers who were vitamin A-deficient during pregnancy weighed approximately 8% less (p < 0.001) and were approximately 2% shorter (p < 0.001) than infants born to mothers who were not deficient. This study suggests children born to HIV-infected women who are vitamin A-deficient during pregnancy are more likely to have growth failure.

Childhood malaria parasitaemia and human immunodeficiency virus infection in Malawi

Malaria and human immunodeficiency virus (HIV) infection are major health problems in many areas in Sub-Saharan Africa. An interaction between malaria and HIV infection has been postulated, since both produce similar cellular immune responses, with a lowering of the CD4/CD8 lymphocyte ratio. The frequency of malaria parasitemia was examined in children born to HIV-seropositive and seronegative mothers attending regular postnatal visits. A prospective study on mother-to-infant transmission of HIV had been underway since 1989 in Queen Elizabeth Central Hospital, Blantyre, a major hospital in urban Malawi. Standard HIV serology was performed on pregnant women, after obtaining consent. To reduce the effect of selection bias and seasonality, HIV seropositive (case) and seronegative (control) mothers and their infants were enrolled at delivery. Children included in the study were 503 born to 494 HIV-seropositive mothers and 540 born to 536 HIV-seronegative mothers. At each 3-monthly postpartum visit a Giemsa-stained thick blood film from the child was examined for malaria parasites. Children born to HIV-seropositive mothers were tested for HIV antibodies at 12 and 18 months of age. Of the 353 children born to HIV-seropositive mothers, 82 children (23.2%) were found to be HIV seropositive by enzyme-linked immunosorbent assay and Western blotting at 12 and 18 months. No statistically significant difference was found in frequency of malaria parasitemia by maternal or infant HIV serostatus after controlling for childs age. There was, however, a significant trend of increase in high parasitemia with age, irrespective of the HIV serostatus of the mother or the child. The frequency of parasitemia was higher in the wet season than in the dry season. This study suggests that maternal or infant HIV infection does not alter susceptibility to, or the clinical course of, malaria in infants.

Maternal vitamin A deficiency and infant mortality in Malawi

The relationship between maternal vitamin A deficiency during pregnancy and infant mortality is unclear. We conducted a prospective cohort study of 377 HIV-negative women and their infants in Blantyre, Malawi. Serum vitamin A levels were measured during the second or third trimester of pregnancy and infants were followed during the first year of life. From delivery until 12 months of age, 18 infants died (47.7 per 1000). Mothers of infants who died had lower serum vitamin A levels during pregnancy (0.74 +/- 0.13 mumol/l) compared with mothers of infants who did not die (1.02 +/- 0.03 mumol/l) (p = 0.055). Infants born to women whose vitamin A levels were in the lowest quartile (< 0.32 mumol/l) had three-fold higher likelihood of mortality than infants born to women whose vitamin A levels were in the higher quartiles (p < 0.03). These results suggest that maternal vitamin A deficiency during pregnancy may contribute to higher infant mortality rates.