HIV transmission during pregnancy is known to occur because:

• PCR tests on some infants have only become positive between one and three months after birth, suggesting that they were infected at delivery.
• Premature babies (earlier than 34 weeks gestation) have a significantly higher chance of being infected. As the baby's mucosa are very fine at this stage, it is suggested that they are highly susceptible to infection during delivery.
• Infants who are HIV-negative by PCR testing are frequently shown to have HIV DNA in their mouths and stomachs, indicating that they were probably exposed to the virus during delivery.
• A study of twins delivered to HIV–positive mothers has shown that the second twin to be delivered is less likely to be infected, especially where both were delivered vaginally. This could be related to the first twin having more prolonged contact with the mother's cervical secretions and blood.
• Prolonged rupture of membranes prior to delivery appears to be a risk factor. If the mother's waters break more than 24 hours before delivery the risk of HIV transmission is greatly increased. However, there does not appear to be a great difference in transmission risk between women whose membranes rupture less than 24 hours prior to delivery and those whose membranes rupture less than four hours prior to delivery (Landesman).

Can the risks be reduced?

The effects of antiretroviral therapy at the time of delivery are discussed in Reducing the risk below.

A review of 8,533 births in Europe and North America established that the risk of transmission was reduced from 7.3% to 2% by the use of a planned caesarean section before the rupture of membranes and onset of labour. The rate of transmission was reduced by 87% in women who received antiretroviral therapy plus elective caesarean delivery compared to women who delivered in other ways and had no therapy (International Perinatal Group). A Swiss population study also found caesarean sections reduced transmission (Kind).

In the French Perinatal Cohort, combined use of AZT and caesarean section before the onset of labour reduced the transmission rate to just 0.8% compared with 6.6% in women on AZT who had vaginal deliveries. However, this study reported that mode of delivery did not affect the rate of transmission among women not on anti-HIV therapy (Mandelbrot).

Preliminary data from a European randomised trial showed an 11% transmission rate for vaginal births, 3% for elective caesarean births and 4% for emergency caesarean births (Semprini 1998). The French study found a higher rate of transmission among the emergency caesarean group.

In 1999, results from the European Mode of Delivery Collaborative study reported reduced rates of transmission amongst women delivering by elective caesarean section than by vaginal delivery or emergency caesarean. In this study, the majority of women received antiretroviral treatment so it's not possible to say that the benefit conferred by this mode of delivery would necessarily be observed amongst untreated women. 436 pregnant HIV-positive women were randomised to deliver their infant by either elective caesarean section at 38 weeks of pregnancy or by vaginal delivery. Infection status was confirmed for 370 infants. 3 of 170 (1.8%) born to women assigned elective c-section, and 21 of 200 (10.5%) born to women assigned vaginal delivery were HIV-infected - a reduction in the rate of transmission of 80%.

Complications following delivery were uncommon, and there were no serious adverse events in either group. The women in the trial did not breastfeed. 69.7% of women allocated elective c-section took antiretroviral therapy during pregnancy (mostly AZT according to the 076 protocol), and 58.2% of women were allocated to vaginal delivery. One of 119 (0.8%) babies delivered by elective c-section to treated mothers was infected. Maternal CD4 count was evenly distributed between the two groups, and the lower rate of transmission associated with elective caesarean delivery was observed consistently across each CD4 count stratum (Parrazini).

Elective caesareans for all?

Though the strength of these data is unquestioned by many clinicians who work with mothers and children, there is some disagreement over whether they support the use of elective caesarean delivery for all pregnancies in HIV-positive women. A leading US commentator, Dr Lisa Frenkel, has proposed that c-sections may be unnecessary in women who have a good virological response to treatment.

In the ACTG 185 study which evaluated the use of AZT plus HIV hyperimmune globulin in pregnancy, there were no HIV transmissions amongst 48 women who had viral load below 500 copies (Lambert). In two other US studies, there were no transmissions observed in 89 and 73 women receiving HAART in pregnancy (Morris, Stek). However, it should be noted that these studies were not designed to investigate the role of elective caesareans in HIV mother-to-baby transmission.

An earlier meta-analysis of 1,115 mother-infant pairs found maternal viral load to be predictive of transmission, particularly so in untreated women. 696 of the women who participated were untreated, and 419 were treated. The rates of transmission for the untreated women whose viral load was below 1,000 copies was 5%; 15% for viral load between 1,000 and 9,999 copies; and 37% for viral load over 10,000 copies. Rates in treated women were 5%, 7% and 18% respectively (Contopoulos-Ioannidis).

Analysis of 373 mother-infant pairs enrolled in the European Collaborative Study similarly found that transmission rates rose with increasing maternal viral load levels, but that there was no threshold below which transmission did not occur. The majority of women in this study were untreated, though 21% received AZT. Delivery was by elective caesarean in 25%, by emergency caesarean in 8% and 62% gave birth by vaginal delivery. Overall, elective caesarean was associated with a 79% decrease in transmission, and delivery before 37 weeks doubled the risk compared with later delivery. In a subgroup analysis, women with higher viral load were less likely to transmit if they delivered by elective caesarean rather than other methods. In women with lower viral loads, the benefit did not reach significance however, perhaps because of the small sample size; only two of sixteen infected children in this group were delivered by elective caesarean section. Hence although this study confirms the relationship between maternal viral load and risk of transmission, it adds little to the continuing debate about mode of delivery in women whose viral load is maximally suppressed by combination therapy (European Collaborative Study).

It has been argued that across-the-board elective caesarean delivery may expose some women to an increased risk of postnatal complications. In an analysis of postnatal morbidity in 1,112 women in the US Women and Infants Transmission Study, researchers found an overall rate of 19% in women who had elective caesareans. Infectious morbidity rates (due to endometritis, urinary tract infections and wound infections) were 11% with elective caesareans, 21% in non-elective caesareans, 8% in women with vaginal forceps delivery, and 4% in vaginal delivery without use of instruments (Read 1999).

Similarly, in an analysis of 497 women from ACTG 185, infectious morbidity rates were 26% with elective caesarean; 40% in non-elective caesarean; 19% with assisted vaginal deliveries; and 13% with spontaneous vaginal delivery (Watts).

Following the publication of the findings of the European Mode of Delivery Collaboration in the Lancet in 1999, these issues were debated in correspondence published in the June 26, 1999 issue (Lancet 353(9171), 1999). Similar discussion also featured in correspondence to the July 15, 1999 issue of the New England Journal of Medicine (NEJM 341(3), 1999), and in an editorial in the May 26, 1999 issue of the Journal of the American Medical Association (JAMA 281:1946-1949, 1999).

For more on the rates of mother-to-baby transmission in the UK, see Epidemiology.

References

Contopoulos-Ioannidis DG et al. Maternal cell-free viremia in the natural history of perinatal HIV-1 transmission: a meta-analysis. Journal of Acquired Immune Deficiency Syndrome and Human Retrovirology 18(2): 126-135, 1998.

European Collaborative Study. Maternal viral load and vertical transmission of HIV-1: an important factor but not the only one. AIDS 13(11): 1377-1385, 1999.

The International Perinatal HIV Group. The mode of delivery and the risk of vertical transmission of human immunodeficiency virus type 1 - a meta-analysis of 15 prospective cohort studies. NEJM 340(13): 977-987, 1999.

Lambert J et al. Risk factors for perinatal HIV transmission in women/infants receiving standard zidovudine (ZDV) prophylaxis. Twelfth World AIDS Conference, Geneva, abstract 232265, 1998.

Mandelbrot L et al. Perinatal HIV-1 transmission, interaction between zidovudine prophylaxis and mode of delivery in the French perinatal cohort. JAMA 280: 55-60, 1998.

Mandelbrot L et al. Frequent detection of HIV-1 in the gastric aspirates of neonates born to HIV-infected mothers . AIDS 13(15): 2143-2149, 1999.

Miotti PG et al. HIV Transmission Through Breastfeeding: A Study in Malawi. JAMA 282, 744-749, 1999.

Morris A et al. A review of protease inhibitors (PI) use in 89 pregnancies. Sixth Conference on Retroviruses and Opportunistic Infections, Chicago, abstract 686, 1999.

Parrazini F for The European Mode of Delivery Collaboration. Elective caesarean-section versus vaginal delivery in prevention of vertical HIV-1 transmission: a randomised clinical trial. Lancet 353: 1035-1039, 1999.

Read J et al. Mode of delivery and postpartum morbidity among HIV-infected women: The women and infants transmission study (WITS). Sixth Conference on Retroviruses and Opportunistic Infections, Chicago, abstract 683, 1999.

Semprini AE et al. An international randomised trial of mode of delivery in HIV infected women. Twelfth World AIDS Conference, Geneva, abstract 23599, 1998.

Stek A et al. Maternal and infant outcomes with highly active antiretroviral therapy during pregnancy. Sixth Conference on Retroviruses and Opportunistic Infections, Chicago, abstract 687, 1999.

Watts H et al. Complications according to mode of delivery among HIV-positive women with CD4 counts <500. Sixth Conference on Retroviruses and Opportunistic Infections, Chicago, abstract 684, 1999.

Working Group on Mother-To-Child Transmission of HIV. Rates of mother-to-child transmission of HIV-1 in Africa, America, and Europe: results from 13 perinatal studies.J Acquir Immune Defic Syndr Hum Retrovirol. 15 (8): 506-510. 1995.