Lack of understanding of treatment benefits may discourage women with HIV from having children – and could increase the risk of vertical transmission

Two papers presented at the eleventh International Congress on Drug Therapy in HIV Infection this week in Glasgow suggest that some women with HIV may not have up-to-date information about vertical transmission (HIV being passed on from mother to child) – and paradoxically the risk may be increased if they do not take antiretroviral therapy (ART) as a result.

A study from Italy found that 20% of women with HIV overestimated the likelihood of passing on HIV to their babies if they were on ART, and nearly 10% thought that transmission from a mother with HIV to a child was more than 50% likely.

The study also showed that starting ART was associated with a decrease in the desire to have a child and that fear of mother-to-baby transmission was the strongest predictor that a woman reported not wanting to have children.

Meanwhile, a study from two East London hospitals found that although the majority of women with HIV who gave birth there returned for care after giving birth, some did not return for one or two years and that 37% had a viral load over 100 copies/ml when they did return. This study was done, in part, to inform discussions about the possibility of relaxing breastfeeding guidelines in the 2012 BHIVA guidelines on treatment for pregnant women; in the end these guidelines did not change their recommendation of total avoidance of breastfeeding.

Desire to have a child

Women with HIV are often viewed as potential mothers, so it is quite unusual for a survey to ask women with HIV of childbearing age whether they actually want children. This is what the DIDI study did in Italy.

This survey asked 178 women who were in a steady relationship and under 45 years old, as part of a larger health survey that took place between November 2010 and February 2011, whether they wanted a child, or another child if they were already a mother.

Sixty-one per cent said ‘no’. This is not that surprising given that the mean age of the group was 39, so more women were nearing menopause than not, and 53% of them had a child already.

The women’s own health considerations were not an influence: this was a group with an average CD4 count of 552 cells/mm³, of whom 88% were on ART, with 79% of those (69% of all the women) with an undetectable viral load.

However, answering ‘no’ was far more likely among the 44% of women who had an HIV-negative partner. Thirty-one per cent of women were actively trying for a child, but whereas 44% of women with HIV-positive partners said they wanted a child, only 10% of women with HIV-negative partners did so.

Over half of the women trying to conceive were simply doing so by having unprotected sex with their partner (including half of the small number of women with HIV-negative partners) but about 20% used self-insemination as a technique – even with HIV-positive partners – and another 20% (50% of those with negative partners) were seeking advice from a fertility clinic.

Half of the women said their HIV diagnosis had affected their desire to have a child (or another child), and a third said it “very much” had. Strikingly, women with an undetectable viral load were 69% more likely not to want a child, although this could be because older women might be more likely to be on ART. Nearly a quarter (22%) of the women said starting ART had reduced their desire to have a child.

Women were asked: “If all necessary measures were adopted, how likely do you think your child would be born without HIV?” Most women had a reasonably accurate view of this: 35% said they thought that if everything was done, it was a certainty that their child would be born without HIV and another 35% said there would be a less than 5% chance. However 10% thought the chance was between 5% and 50% and 9.4% thought the chance was over 50% (3.4% thought it was a certainty any child of theirs would be born with HIV).

When it came to predictors of wanting to have a child, Italian-born women and former injecting drug users (who were also more likely to be native Italians) were half as likely to want a child than other women, while women on a low income (less than €800 a month) were four times more likely to want a child, though this reflects the fact that low-income women were more likely to be young and/or African immigrants.

Low income was one of only two factors that had a statistically significant association with wanting to have a child. The other was “fear of vertical transmission”: women who said they were afraid that they would pass on HIV to their child were 3.75 times more likely to not want to have a child, and those who feared they would not live long enough to raise a child were 2.85 times more likely not to want one. Fear of having to disclose their own HIV status to their child was another reason quoted by many women for not wanting a child.

Although this study serves as a useful reminder that women with HIV should not be regarded simply as potential child bearers, it also suggests that among some women, exaggerated fears of infecting either their partner or their child may be inhibiting them from becoming mothers.

Viral loads after giving birth and breastfeeding

The study from Homerton and Newham General Hospitals in London was a more straightforward survey of viral load and follow-up monitoring in pregnant women but was done with a purpose in mind: 80% of pregnant women with HIV attending these two hospitals were of African origin and many of them had expressed confusion about breastfeeding and viral load guidelines. Advice on infant feeding given to women living with HIV is very different in the UK to the advice given in many African countries.

Whereas the British HIV Association (BHIVA) guidelines on pregnancy say that women with HIV should exclusively use formula milk for infant feeding from birth and never breast feed, the international World Health Organization (WHO) guidelines say that, on the contrary, as long as women have an undetectable viral load and are in situations where that can be monitored, then they should breast feed, because there is evidence that in low-income countries it may protect maternal and child health. The hospitals therefore undertook to monitor post-partum viral load in women, partly in order to see if BHIVA could relax its guidelines.

They looked at 151 pregnancies in 140 women. The mean age of the women at delivery was 32, and 27.5% of them were diagnosed during pregnancy. Most women needed ART as treatment: only 20% had CD4 counts above 350 cells/mm³ and took ART to prevent transmission, rather than because they needed it themselves. Viral load at delivery was well controlled, with only 11% of women having a viral load over 100 copies/ml.

All but two women returned for care, but the average time to the next visit at which viral load was measured was 2.5 months – the concern being that if a woman was breastfeeding and her viral load increased in the time between viral load tests, the risk of passing on HIV to her baby would also increase. The longest gap between delivery and follow-up was 26 months.

At follow-up, quite a high proportion of women had a detectable viral load. A third of the women who were on HIV treatment for their own health had a viral load over 50 copies/ml and 21.5% over 100 copies/ml. Only 25% of women who had taken treatment specifically to prevent mother-to-child transmission (and who had then stopped taking it) had a viral load under 100 copies/ml. Altogether, 37% of the women had a viral load over 100 copies/ml at their first test after giving birth.

The authors of the paper suggest that viral load monitoring in nursing mothers needs to be considerably intensified before breastfeeding guidelines can be relaxed.