Pregnancy did not increase the risk of HIV infection in Ugandan, Zimbabwean women

Pregnancy or breastfeeding do not place women at an increased risk of HIV-1 infection, according to the findings of a prospective study published in the May edition of AIDS. The study recommends that antenatal and postnatal programs must continue to emphasise condom use to protect both mother and baby from HIV infection.

The potential association between pregnancy and HIV infection has serious public health implications since African countries with an escalating HIV-1 problem have high fertility rates. HIV infection in pregnant women might lead to higher rates of mother-to-child transmission of HIV and poor pregnancy outcomes.

Previous studies which suggested a link between pregnancy and HIV-1 infection did not adjust for behavioural factors. There is a need for further studies which critically examine the effect of behaviour and physiological changes in pregnancy. A team of US investigators addressed this issue by re-examining the data of a prospective study of hormonal contraception and HIV-1 acquisition in Ugandan and Zimbabwean women (Morrison et al. 2007b).

Women seeking reproductive and general healthcare services were enrolled at family planning clinics and followed between 1999 and 2004. A smaller group of 398 ‘high-risk referral’ women were recruited from sexually transmitted infections (STI) or primary healthcare clinics, sex worker networks, or military bases.

The women were 18-35 years old, were neither pregnant nor intending to get pregnant within the next year, were HIV-1 uninfected, sexually active, and had been using either no hormonal contraceptive method, combined oral contraceptive pills or depot medroxyprogesterone acetate (DMPA) for three months.

Follow-up was quarterly for 15-24 months with a physical examination, pregnancy, HIV-1 and STI testing, and completion of a standardised questionnaire.

Laboratory examinations included a pregnancy urine test for human chorionic gonadotropin complemented with clinical impressions, and specific tests for HIV-1, various STIs, and bacterial vaginosis. The date of HIV-1 infection was established by retrospective HIV-1 DNA PCR on serial visit samples.

Data analysis included participants with at least one follow-up visit with valid pregnancy and HIV-1 results. Women who were using or not using hormonal contraception were distinguished in the primary exposure variable.

Four exposure categories were defined: pregnant at the current visit, not currently pregnant but lactating (NP/L) since the last study visit, not pregnant and not lactating (NP/NL) but using hormonal contraception since the last study visit, and not currently pregnant or lactating and not using hormonal contraceptives since the last visit. The latter category was the comparison group.

A participant’s time was divided into segments corresponding to the period between study visits in order to capture changes in pregnancy and lactation status, hormonal contraceptive use, sexual risk behaviours, and the presence of STI. The study participant’s behavioural risk and the primary partner’s risk of exposure to HIV were also determined.

Study participants had a median age of 25 years, a median education of ten years, most lived with a partner, had a median of two lifetime pregnancies, and 28 % breastfed at enrolment. At baseline, 34.7 % of the women used combined oral contraceptive pills, 34.2% used DMPA, and 31.1% did not use a hormonal method. Multiple sex partners, commercial sex, or sex while using alcohol or drugs were rarely reported. Condom use was reported by less than half of the participants.

A total of 4,415 women contributed to 31,106 visit segments. The 24-month retention was 92 %, the mean follow-up was 21.9 months, and the median time between study visits was 81 days. Women were pregnant in 9.2 %, NP/L in 15.1 %, NP/NL but using hormonal contraception in 56.8 %, and NP/NL and not using hormonal contraception in 18.9 % of visit segments. Pregnancy was more common in younger women, women living with a partner, and women with less than three lifetime pregnancies at enrolment; education level had no effect on pregnancy.

Risky sexual behaviour was less in pregnant women than in the NP/NL and no hormonal contraception use group. However, the frequency of reported unprotected sex was higher in pregnant women than in the NP/NL and no hormonal contraception use group (82.0 versus 37.8 %). STIs were somewhat higher during follow up in pregnant women than in the NP/NL and no hormonal contraception use group.

There were 211 incidents of HIV-1 infections of which 13, 33, 126, and 39 occurred among pregnant women, NP/L women, NP/NL women using hormonal contraception, and NP/NL women not using hormonal contraception, respectively. Univariate and multivariate analyses demonstrated that pregnancy and lactation was not significantly associated with the risk of HIV-1 infection.

Risk factors which were significantly associated with increased HIV-1 acquisition included Zimbabwe site and Uganda high-risk referral group, not living with a partner, younger age, high participant behavioural risk, high primary partner risk, and recent alcohol use. The relationship between the risk of HIV infection and pregnancy depended on the study site and age.

In Zimbabwe pregnancy significantly protected against HIV infection while in Uganda there was an increased but non-significant risk of infection associated with pregnancy in the general population and a non-significant protective effect of pregnancy in the high-risk referral population. In older women above 25 years, pregnancy had a non-significant protective effect on HIV-1 infection (HR 0.37, 95 % CI 0.13-1.09). However, in younger women, pregnancy had no effect on HIV-1 infection (HR 1.14, 95 % CI 0.47-2.80).

Although pregnancy and lactation did not increase the risk of HIV infection, more pregnant women had unprotected sex which probably explains the higher incidence of STIs in this group. The authors point out the need for further understanding of the sexual behaviours of pregnant women and their partners which increase the risk for HIV infection.

The findings of Morrison et al. differ from previous studies. This discrepancy is probably due to differences in study design and methodology and study populations. The policy implications of this study is that antenatal and postnatal programmes must emphasise condom use and target the sex partners of pregnant women for counselling to avoid risky sexual behaviour.