| |
Prevention of Mother-to-Child Transmission of HIV Infection (and Caesarean Section Delivery)
|
| |
| |
Editorial
Katherine Luzuriaga and John L. Sullivan
Program in Molecular Medicine, University of Massachusetts Medical School, Worcester
(See the article by the European Collaborative Study below)
It has been 10 years since a study of perinatal zidovudine regimen showed that it prevented mother-to-child transmission of HIV in 67% of cases. The article from the European Collaborative Study in this issue of Clinical Infectious Diseases summarizes how far we have come over the past 10 years in the developed world. Treatment of HIV-infected pregnant women has increased such that >90% of HIV-infected women receive HAART. The overall rate of mother-to-child transmission of HIV has been dramatically reduced to <1%. The European Collaborative Study confirm that maternal viral load is the key risk factor for mother-to-child transmission of HIV and that the suppression of viral replication through administration of potent combinations of antiretroviral drugs markedly reduces the risk of mother-to-child transmission. In HAART recipients, the only other intervention that significantly reduced mother-to-child transmission of HIV was elective Caesarean section delivery. Although the European Collaborative Study article suggests that elective Caesarean section delivery and receipt of combination antiretroviral therapy may reduce mother-to-child transmission rates more than combination antiretroviral therapy alone, many physicians in the developed world will likely balance the potential risks of the procedure with its potential benefits and reserve this mode of delivery for those women with detectable viral loads (i.e., >50 copies/mL).
These are remarkable achievements that should make all persons who work to eradicate mother-to-child transmission of HIV proud. But what do these results mean for the developing world? In 2003, an estimated 2.0 million children were born to HIV-infected pregnant women in resource-poor settings, and 〜700,000 children were infected (http://www.unaids.org). This represents almost 2000 new infections per day, >90% of which currently occur in sub-Saharan Africa, where >50% of these infected infants will die by their second birthday.
Access to potent antiretroviral agents is no longer a barrier to prevention of mother-to-child transmission of HIV in limited-resource settings. Five years ago, Guay et al. demonstrated that an inexpensive simple peripartum nevirapine regimen is effective for prevention of a significant proportion of peripartum transmissions of HIV. In 2000, at the International AIDS Conference in Durban, Boehringer-Ingelheim announced that nevirapine would be offered free of charge to developing countries for the prevention of mother-to-child transmission of HIV. After this announcement, major funding was directed to putting an infrastructure in place to implement the simple 2-dose nevirapine prophylactic regimen. In spite of valiant efforts by many individuals and organizations to implement this stop-gap measure, 95% of HIV-infected pregnant women who could benefit from this prevention regimen still do not have access to it because of a continued lack of infrastructure. Although the 2-dose regimen of nevirapine reduces overall rates of perinatal mother-to-child transmission of HIV to 〜6%--12%, it has no effect on prevention of transmission of virus via breast milk, and mothers who do not receive treatment die and leave orphaned children behind. Even with access to the best 3-drug HAART regimens, the lack of medical infrastructure will continue to hamper our efforts to markedly impact mother-to-child transmission of HIV. This can only be fixed with an enormous infusion of resources from the developed world. These resources are not limited to money, but they require large numbers of trained physicians and nurses to collaborate with their colleagues in resource-limited settings to develop the most appropriate measures for prevention and treatment of HIV infection. The Infectious Diseases Society of American is among the organizations that sponsor such efforts, and we would urge all interested persons to participate to make a difference.
MAIN STUDY ARTICLE
"Mother-to-Child Transmission of HIV Infection in the Era of Highly Active Antiretroviral Therapy" & Caesarean Section Delivery
Clinical Infectious Diseases Feb 1, 2005;40:458-465
European Collaborative Studya
".....Caesarean section delivery was associated with a significant reduction (by two-thirds) in the risk of transmission, compared with vaginal delivery, independent of other variables. With the inclusion of maternal viral load, although there was a trend towards increased mother-to-child transmission risk associated with severe maternal immunosuppression, this did not reach statistical significance...."
ABSTRACT
Background. Very low rates of mother-to-child transmission (MTCT) of human immunodeficiency virus (HIV) are achievable with use of highly active antiretroviral therapy (HAART). We examine risk factors for MTCT in the HAART era and describe infants who were vertically infected, despite exposure to prophylactic MTCT interventions.
Methods. Of the 4525 mother-child pairs in this prospective cohort study, 1983 were enrolled during the period of January 1997 through May 2004. Factors examined included use of antiretroviral therapy during pregnancy, maternal CD4 cell count and HIV RNA level, mode of delivery, and gestational age in logistic regression analysis.
Results. Receipt of antenatal antiretroviral therapy increased from 5% at the start of the HAART era to 92% in 2001--2003. The overall MTCT rate in this period was 2.87% (95% confidence interval [CI], 2.11%--3.81%), but it was 0.99% (95% CI, 0.32%--2.30%) during 2001--2003.
In logistic regression analysis that included 885 mother-child pairs, MTCT risk was associated with high maternal viral load (adjusted odds ratio [AOR], 12.1; P = .003) and elective Caesarean section (AOR, 0.33; P = .04).
Detection of maternal HIV RNA was significantly associated with antenatal use of antiretroviral therapy, CD4 cell count, and mode of delivery.
Among 560 women with undetectable HIV RNA levels, elective Caesarean section was associated with a 90% reduction in MTCT risk (odds ratio, 0.10; 95% CI, 0.03--0.33), compared with vaginal delivery or emergency Caesarean section.
Conclusions. Our results suggest that offering an elective Caesarean section delivery to all HIV-infected women, even in areas where HAART is available, is appropriate clinical management, especially for persons with detectable viral loads. Our results also suggest that previously identified risk factors remain important.
Elective Caesarean section delivery in women with undetectable HIV RNA levels and/or receipt of HAART. Among 560 women with undetectable HIV RNA levels (44% with levels of <50 copies/mL), elective Caesarean section delivery was associated with a significantly reduced rate of mother-to-child transmission of HIV, compared with vaginal or emergency Caesarean section delivery, in univariable logistic regression analysis, with a 93% reduction in the risk of mother-to-child transmission (OR, 0.07; 95% CI, 0.02--0.31; P = .0004). The small number of infected children (n = 27) in this subgroup precluded the inclusion of >2 categories of maternal antiretroviral therapy in multivariable analysis; however, with inclusion of a binary antiretroviral therapy variable (none vs. any), elective Caesarean section delivery was still associated with a halving of the mother-to-child transmission risk (adjusted OR, 0.52; 95% CI, 0.14--2.03; P = .358). In a subanalysis limited to 759 mother-child pairs receiving antenatal HAART, elective Caesarean section delivery was associated with a 〜40% reduction in risk of mother-to-child transmission of HIV, compared with vaginal delivery (OR, 0.64; 95% CI, 0.08--5.37; P = .7).
Children infected despite exposure to maternal HAART. Eleven children (6 male and 4 female children) became infected, even though their mothers received HAART antenatally; however, median duration of antenatal HAART was only 38 days (range, 4--152 days) for the 10 women who started receiving HAART during pregnancy, whereas 1 woman had been receiving therapy for >1 year before she became pregnant. The 11 women had relatively advanced HIV disease, with a median maternal CD4 cell count of 209 cells/mm2 (range, 64--468 cells/mm3). Although elective Caesarean section delivery would have been the best mode of delivery for these infants in terms of prevention of mother-to-child transmission of HIV, only 5 infants were delivered by elective Caesarean section, with 4 infants delivered prematurely by emergency Caesarean section delivery and 1 delivered via spontaneous vaginal delivery at 38 weeks. All 11 neonates received antiretroviral prophylaxis (1 received zidovudine, lamivudine, and nevirapine for 6 weeks; 1 received zidovudine and lamivudine for 6 weeks; and the rest received oral zidovudine monotherapy [1 received it for 2 weeks, 1 received it for 4 weeks, and the remainder received it for 6 weeks]). Two mothers already had a child enrolled in the European Collaborative Study, and 1 of those children was also vertically infected.
AUTHOR DISCUSSION
In this large European cohort, there has been a dramatic decrease in the rate of mother-to-child transmission of HIV over time, to 〜1% in recent years, corresponding with the increasing uptake of prophylactic interventions. Nearly one-half of the women who enrolled in this study were already receiving HAART at the time that they became pregnant, and achievement of undetectable HIV RNA levels at or close to delivery was more frequent among women who started HAART before pregnancy than among women who started therapy during pregnancy. A high maternal viral load was the key risk factor for mother-to-child transmission of HIV, and the only other factor achieving statistical significance in the multivariable analysis in the HAART era was elective Caesarean section delivery. In an analysis that allowed for maternal viral load, although HAART use was associated with a reduction of two-thirds in the risk of mother-to-child transmission, this was not statistically significant, indicating that the effect of HAART is exerted through reducing the maternal plasma viral load.
Given the very small number of infections, due to widespread use of prophylactic interventions, it was not possible to confirm the statistical significance of effects such as prematurity in the HAART era. Although there is evidence to suggest that premature infants are more susceptible to vertical acquisition of HIV infection than are term infants, including the data presented here for the whole cohort, the increased risk of mother-to-child transmission of HIV associated with prematurity in the HAART era here was not statistically significant. Our rate of premature birth was 25%, reflecting the association previously reported between HAART use and duration of pregnancy. Although there was a higher rate of mother-to-child transmission among infants delivered with episiotomy and/or forceps or vacuum, this did not reach statistical significance. However, this is likely to be associated with our sample size limitations, and we cannot exclude the possibility of a real risk associated with these types of delivery, as reported elsewhere in the pre-HAART era. Our finding of an increased duration of rupture of membranes in infected infants, compared with uninfected infants, is consistent with previous reports.
The fluctuations over time in the mode of delivery seen in this cohort were caused by various factors. The temporary increase in the rate of emergency Caesarean section deliveries in 2000 was largely associated with premature labor or rupture of membranes of women booked for elective Caesarean section deliveries, which in turn resulted in a policy shift towards earlier scheduling of Caesarean section. However, the recent increase in the number of vaginal deliveries seen and the finding that women who underwent vaginal deliveries were more likely to have undetectable viral loads than were those who underwent elective Caesarean section deliveries reflect the current lack of consensus regarding the additional benefit of elective Caesarean section delivery for women with undetectable viral loads who are receiving HAART. Elective Caesarean section deliveries are associated with a higher rate of postpartum complications among HIV-infected women, compared with vaginal delivery, although HIV-infected women are generally at an increased risk of postpartum complications, compared with uninfected women, regardless of mode of delivery. In a recent study, the most common postpartum complication among infected women who underwent elective Caesarean section deliveries was anemia, which was strongly associated with receipt of antenatal antiretroviral therapy.
It is of interest to note that, even in the HAART era, we were able to show a significant reduction (two-thirds) in the risk of mother-to-child transmission of HIV associated with elective Caesarean section delivery, independent of HIV RNA load and maternal antiretroviral therapy use. In a meta-analysis involving women with plasma HIV RNA loads of <1000 copies/mL at the time of delivery, Caesarean section delivery (with elective and emergency procedures grouped together) remained protective against mother-to-child transmission of HIV (with a 70% reduction in risk) after adjustment for maternal receipt of antiretroviral therapy (mainly non-HAART) and CD4 cell count. In our cohort, the subanalyses involving women with undetectable viral loads close to delivery showed an effect of elective Caesarean section delivery in significantly reducing risk of transmission by 93%, although small numbers and the resulting lack of power precluded additional adjustment for HAART use. To show a significant effect of elective Caesarean section delivery on mother-to-child transmission of HIV among women receiving HAART, an estimated 6345 and 7217 mother-child pairs would be needed with vaginal deliveries and elective Caesarean section deliveries, respectively, with 90% power, thus precluding any single study from answering the question of what the added benefit of elective Caesarean section delivery might be for women receiving HAART.
Compartmentalization of HIV in the genital tract may explain the continuing effectiveness of elective Caesarean section delivery, even among women with undetectable plasma viral loads. Although a high correlation between plasma and genital tract viral loads has been reported in nonpregnant women, there is accumulating evidence to suggest there is a separate reservoir of the virus in the genital tract independent from plasma virus. Understanding of the correlation between plasma viral load and the genital tract viral load in pregnancy and of the differential effect of antiretroviral therapy is relatively limited, although several studies have suggested an association between genital tract viral load and mother-to-child transmission independent of plasma viral load.
Despite initial optimism regarding the potential for "elimination" of vertically acquired HIV infection in resource-rich settings, we show that a very small number of infected women continue to transmit HIV to their infants. Among the 11 infants who were infected, despite exposure to maternal HAART and neonatal prophylaxis, there was a median exposure to antenatal antiretroviral therapy of slightly more than 1 month. Despite guidelines for universal antenatal screening for HIV infection, late identification of pregnant, HIV-infected women is still a barrier to optimum application of interventions in developed country settings, in addition to lack of antenatal care in some subgroups of women.
The current strategy of offering an elective Caesarean section delivery to HIV-infected women in Europe appears to be appropriate. In our cohort, approximately one-half of the women enrolled recently had detectable viral loads close to the time of delivery, and the continuing high rate of elective Caesarean section delivery indicates that these women are being recommended to undergo an elective Caesarean section delivery to prevent mother-to-child transmission of HIV. Although, in the presence of HAART, the absolute rate of mother-to-child transmission is likely to be low (in the region of 1%--2%), our results suggest that elective Caesarean section delivery would further reduce this risk to 〜0.5%--1%. However, with these very low overall rates, a large number of elective Caesarean section deliveries would need to be performed to prevent a single vertical infection. Although elective Caesarean section delivery has been shown to be cost-effective in women receiving zidovudine monotherapy and combination therapy, such analyses need to be repeated for women receiving HAART, because previous cost-benefit analyses have indicated that results are sensitive to variations in mother-to-child HIV transmission rates. On an individual level, the decision regarding mode of delivery rests with the woman and her clinician, who should inform her of the potential risks and benefits. With adequate obstetric management, individual women may consider the benefit of a lower risk of infection in their infant to outweigh the potential disadvantages of an elective caesarean section.
|
|
| |
| |
|
|
|
