An estimated 16.4 million women worldwide are living with human immunodeficiency virus (HIV) infection; 600,000 children are infected annually, most of them by mother-to-child transmission. Interventions designed to reduce perinatal transmission of HIV in the developing world have been described elsewhere. This article focuses on the management of HIV infection in pregnant women in developed countries where antiretroviral therapy, scheduled cesarean delivery, and alternatives to breast-feeding are available. Standards of care for HIV change rapidly, and up-to-date recommendations are available on the Web (e.g., at hivatis.org). Clinicians who care for HIV-infected women of reproductive age should provide family-planning services and counseling to optimize the use of medications and health status, including updating immunizations, before a pregnancy occurs.
Interaction between HIV Infection and Pregnancy
Studies in the United States and Europe have shown no effect of pregnancy on the progression of HIV disease. The largest study included women with a known date of seroconversion and reported that the adjusted relative risk of progression from HIV infection to the acquired immunodeficiency syndrome (AIDS) associated with pregnancy was 0.7 (95 percent confidence interval, 0.4 to 1.2). Reports from developing countries suggest that progression accelerates with pregnancy, but it is difficult to interpret such reports because the samples are small and the studies are subject to selection bias related to the presence of indications for testing.
Are outcomes of pregnancy affected by HIV infection? Studies conducted in industrialized countries before antiretroviral agents became available did not show an increase in the frequency of prematurity, low birth weight, or intrauterine growth restriction associated with HIV infection. In these studies, the rates of adverse events in the control groups were relatively high, probably because of coexisting conditions such as substance abuse, although it is difficult to interpret the data. Studies in developing countries have shown greater frequencies of preterm birth, low birth weight, intrauterine growth restriction, and stillbirth among the infants of HIV-infected women than among those of similar groups of women who were HIV-negative. Higher rates of adverse outcomes were seen among women with more advanced HIV infection. Increased mortality, primarily related to HIV infection in the infant, has been reported among infants born to HIV-infected women in developing countries. No studies have indicated that there is an increase in the frequency of birth defects related to HIV infection, and the theory that there is a syndrome of malformation related to HIV infection has been disproved. The rates of factors related to preterm birth and low birth weight are similar among HIV-infected and HIV-negative pregnant women; these factors include previous adverse outcomes, hypertension, multiple gestation, smoking, bleeding during pregnancy, and Trichomonas vaginalis infection. In the absence of antiretroviral therapy, a low percentage of CD4+ cells was an additional risk factor for adverse outcomes12; but with zidovudine therapy, maternal HIV RNA levels and CD4+ cell counts were not directly correlated with adverse outcomes.
It has been suggested that antiretroviral therapy may increase the rate of adverse outcomes of pregnancy, but one randomized trial as well as cohort analyses found that zidovudine monotherapy had no negative effect on the outcomes of pregnancy. A 1998 study suggested that combination therapy may increase the rate of preterm birth; subsequent data from cohorts in the United States have not shown an increase in the rate of preterm birth with the use of dual-nucleoside or protease-inhibitor therapy, but data from the European Collaborative Study and the Swiss Mother and Child HIV Cohort Study have shown higher rates of preterm birth as the number of antiretroviral agents used has increased. Thus, clinicians should be aware of the potentially increased risk of preterm delivery among HIV-infected women who are receiving combination antiretroviral therapy but should not withhold indicated therapy.
Data on Antiretroviral Drugs during Pregnancy
Information about currently approved antiretroviral drugs is summarized in Table 3. Clinicians should report cases of exposure to antiretroviral drugs during pregnancy to the Antiretroviral Pregnancy Registry (at 1-800-258-4263 or http://www.apregistry.com) to assist in monitoring patients for teratogenic effects.
A number of studies have examined the use of antiretroviral drugs during the perinatal period. The results of additional trials of short courses of antiretroviral drugs for the prevention of perinatal transmission of HIV-1 are summarized in Figure 1.
Figure 1. Duration of Antiretroviral Therapy and Perinatal Transmission of HIV-1. The bars show the longest duration of antiretroviral therapy during pregnancy, labor, and delivery and in the newborn in a given study. Transmission rates listed are for the first analysis point reported. The first four studies listed involved women who were not breast-feeding; most of the women in the remaining studies were breast-feeding. The Pediatric AIDS Clinical Trials Group (PACTG) Protocol 076 study compared placebo with oral zidovudine starting at 14 to 34 weeks of gestation, intravenous zidovudine during labor, and oral zidovudine for 6 weeks in the infant. A short-course zidovudine study in Thailand (Thai Short) compared placebo with oral zidovudine starting at 36 weeks of gestation, oral zidovudine intrapartum, and no neonatal therapy. The Perinatal HIV Prevention Trial in Thailand (Thai 4-Group) had a factorial design and included four groups, all of which received oral intrapartum zidovudine; the short antepartum course of zidovudine started at 36 weeks of gestation, and the long course started at 28 weeks of gestation; the short course of oral zidovudine for the infant was 3 days and the long course was 6 weeks. PACTG 316 compared placebo with a single dose of nevirapine in the woman during labor and a single dose in the infant at 48 hours of age; both groups of women had been receiving ongoing antiretroviral therapy during pregnancy, including at least zidovudine. A short-course zidovudine trial in the Ivory Coast and Burkina Faso compared placebo with zidovudine beginning at 36 weeks of gestation, oral zidovudine during labor, and zidovudine for 1 week in the infant. Another short-course zidovudine trial in the Ivory Coast compared placebo with oral zidovudine beginning at 36 weeks of gestation and oral zidovudine during labor. The Perinatal Transmission (PETRA) trial involved four groups: in one, oral zidovudine–lamivudine treatment was started at 36 weeks of gestation, and was given during labor and for 1 week after delivery in mother and infant; in the second, oral zidovudine–lamivudine was given during labor and for 1 week after delivery in mother and infant; in the third, oral zidovudine–lamivudine was given during labor only; and the fourth group received placebo. The HIV Network for Prevention Trials (HIVNET) 012 Trial31 compared a single dose of nevirapine during labor and a single dose in the infant with oral zidovudine given during labor and in the infant for one week. The South African Intrapartum Nevirapine Trial (SAINT) compared a single dose of nevirapine during labor and a single dose in the infant at 48 hours with oral zidovudine–lamivudine given during labor and for one week in the infant.
Nucleoside Reverse-Transcriptase Inhibitors
The nucleoside reverse-transcriptase inhibitors are generally well tolerated and cross the placenta. These agents have not been shown to be teratogenic in animals in concentrations similar to those used in humans. The rate of birth defects among the infants born to the more than 400 women whose exposure to zidovudine or lamivudine during the first trimester of pregnancy was reported to the Antiretroviral Pregnancy Registry has been no higher than the rates among infants born to women who were exposed after the first trimester or infants in the Metropolitan Atlanta Congenital Defects Program of the Centers for Disease Control and Prevention.
Nucleoside reverse-transcriptase inhibitors bind to mitochondrial DNA polymerase gamma and can cause mitochondrial dysfunction, which may be manifested as myopathy, cardiomyopathy, neuropathy, lactic acidosis, or fatty liver. Hepatomegaly and steatosis occurred at a rate of 1.3 cases per 1000 person-years among women who received nucleoside therapy, and symptomatic hyperlactatemia occurred at a rate of 8 cases per 1000 person-years among those who received any antiretroviral drugs.
Toxic effects of long-term therapy with nucleoside reverse-transcriptase inhibitors may be enhanced in pregnant women. Three deaths and additional cases of lactic acidosis and hepatic failure have been described among pregnant women who began receiving stavudine and didanosine along with other drugs before pregnancy. Clinical findings are similar to those in acute fatty liver of pregnancy, a syndrome that is more frequent among women with heterozygous defects of mitochondrial fatty-acid metabolism who are carrying fetuses that are homozygous for the same defect. The enhancement of mitochondrial toxic effects during pregnancy is similar to that seen in mice that have significant reductions in mitochondrial fatty-acid oxidation during the late phases of pregnancy or when they are given high doses of estrogen and progesterone. Although lactic acidosis and hepatic failure have been noted most commonly with long-term use of stavudine and didanosine, the potential exists with all nucleoside reverse-transcriptase inhibitors (the binding affinity with polymerase gamma is greatest for zalcitabine, followed in descending order by didanosine, stavudine, lamivudine, zidovudine, and abacavir).
Both clinicians and their pregnant patients should be aware of the nonspecific symptoms of liver dysfunction and lactic acidosis. Routine measurement of blood lactate is not recommended because it is difficult to obtain accurate samples and because normal levels in pregnant women have not been established. However, lactate testing during pregnancy may be helpful if suggestive symptoms are present and levels are in the range that would be considered elevated in persons who were not pregnant.
It has also been suggested that mitochondrial dysfunction might develop in infants who are exposed to nucleoside reverse-transcriptase inhibitors. Eight cases of possible mitochondrial dysfunction, including two deaths, were reported in a cohort of 1754 infants exposed to zidovudine or zidovudine–lamivudine during pregnancy and the neonatal period. In a review covering more than 16,000 children in cohorts in the United States, no increase in the rate of death was reported among children exposed to nucleoside reverse-transcriptase inhibitors as compared with children with no such exposure, and no deaths were found to be definitely related to mitochondrial toxicity. Monitoring for symptoms among living children is ongoing. Among the 1798 infants in the Perinatal Transmission trial, which included a placebo group and three zidovudine–lamivudine groups, the rate of symptoms potentially related to mitochondrial toxic effects (present in a total of 5 infants [0.28 percent]) did not differ significantly among the treatment groups. Thus, the risk of mitochondrial toxic effects in an infant that are related to therapy with nucleoside reverse-transcriptase inhibitors during pregnancy appears to be small and must be balanced against the benefit in terms of a reduction in the rate of perinatal transmission.
Transplacental carcinogenesis is another issue of concern related to exposure in utero to nucleoside reverse-transcriptase inhibitors. An increase in the incidence of tumors of the liver, lungs, and reproductive system was reported in rodents exposed in utero to zidovudine at 30 times the dose used in humans, but no increase was found in a similar study using lower doses. Data from studies in humans do not suggest an increase in the incidence of tumors in children with prenatal exposure to antiretroviral drugs. Among 727 children exposed to zidovudine in utero and followed for up to six years, no tumors or deaths from cancer occurred.
Nonnucleoside Reverse-Transcriptase Inhibitors
Data on use of nonnucleoside reverse-transcriptase inhibitors during pregnancy are limited, but nevirapine and efavirenz readily cross the placenta in primates. Use of efavirenz in the early stages of pregnancy is not recommended because birth defects (anencephaly, anophthalmia, or cleft palate) occurred in 3 of 20 monkeys (15 percent) born after exposure during the first trimester of pregnancy. However, use during the later stages of pregnancy may be considered for women in whom other regimens have failed. Delavirdine at high doses has been associated with heart defects in rodents and should therefore, like efavirenz, be used only during the later stages of pregnancy. Nevirapine is the nonnucleoside reverse-transcriptase inhibitor that has been used most commonly during pregnancy, primarily close to the time of delivery.
The most common toxic effect of nonnucleoside reverse-transcriptase inhibitors is rash. Nevirapine-related rash occurs in 17 percent of users and requires an interruption of treatment in 5 to 6 percent. Hepatitis develops in about 1 percent of those who receive nevirapine and may be fatal.
Protease inhibitors are increasingly being used during pregnancy. There appears to be minimal transplacental passage in humans. No specific teratogenic effects have been noted in animals. Optimal dosing of protease inhibitors during pregnancy remains under study. Lower serum concentrations of protease inhibitors have been observed in pregnant patients than in nonpregnant patients, although in most cases, the HIV RNA levels in pregnant women have been suppressed. The toxic effects among pregnant women appear to be similar to those among nonpregnant women. There are insufficient data to support the recommendation of a specific protease inhibitor during pregnancy, although nelfinavir has been used most commonly.
From the Pediatric, Adolescent, and Maternal AIDS Branch, Center for Research on Mothers and Children, National Institute of Child Health and Human Development, Bethesda, Md.
1. UNAIDS/WHO-AIDS epidemic update — December 2000. Geneva: UNAIDS, 2000.
2. DeCock KM, Fowler MG, Mercier E, et al. Prevention of mother-to-child HIV transmission in resource-poor countries: translating research into policy and practice. JAMA 2000;283:1175-1182.
3. Saada M, Le Chenadec J, Berrebi A, et al. Pregnancy and progression to AIDS: results of the French prospective cohorts. AIDS 2000;14:2355-2360.
4. Burns DN, Landesman S, Minkoff H, et al. The influence of pregnancy on human immunodeficiency virus type 1 infection: antepartum and postpartum changes in human immunodeficiency virus type 1 viral load. Am J Obstet Gynecol 1998;178:355-359.
5. Weisser M, Rudin C, Battegay M, et al. Does pregnancy influence the course of HIV infection? Evidence from two large Swiss cohort studies. J Acquir Immune Defic Syndr Hum Retrovirol 1998;15:404-410.
6. Deschamps MM, Pape JW, Desvarieux M, et al. A prospective study of HIV-seropositive asymptomatic women of childbearing age in a developing country. J Acquir Immune Defic Syndr 1993;6:446-551.
7. Kumar RM, Uduman SA, Khurrana AK. Impact of pregnancy on maternal AIDS. J Reprod Med 1997;42:429-434.
8. Brocklehurst P, French R. The association between maternal HIV infection and perinatal outcome: a systematic review of the literature and meta-analysis. Br J Obstet Gynaecol 1998;105:836-848.
9. Bucceri A, Luchini L, Rancilio L, et al. Pregnancy outcome among HIV positive and negative intravenous drug users. Eur J Obstet Gynecol Reprod Biol 1997;72:169-174.
10. Leroy V, Ladner J, Nyiraziraje M, et al. Effect of HIV-1 infection on pregnancy outcome in women in Kigali, Rwanda, 1992-1994. AIDS 1998;12:643-650.
D. Heather Watts, M.D.
The New England Journal of Medicine