Prepared by
Lynne M. Mofenson, M.D.
Center for Research for Mothers and Children
National Institute of Child Health and Human Development
National Institutes of Health

Summary

These recommendations update the February 4, 2002, guidelines developed by the Public Health Service for the use of zidovudine (ZDV) to reduce the risk for perinatal human immunodeficiency virus type 1 (HIV-1) transmission. This report provides health-care providers with information for discussion with HIV-1--infected pregnant women to enable such women to make an informed decision regarding the use of antiretroviral drugs during pregnancy and use of elective cesarean delivery to reduce perinatal HIV-1 transmission. Various circumstances that commonly occur in clinical practice are presented, and the factors influencing treatment considerations are highlighted in this report. The Perinatal HIV Guidelines Working Group recognizes that strategies to prevent perinatal transmission and concepts related to management of HIV disease in pregnant women are rapidly evolving and will continually review new data and provide regular updates to the guidelines. The most recent information is available from the HIV/AIDS Treatment Information Service (available at http://www.hivatis.org).

In February 1994, the Pediatric AIDS Clinical Trials Group (PACTG) Protocol 076 demonstrated that a three-part--regimen of zidovudine (ZDV) could reduce the risk for mother-to-child human immunodeficiency virus type 1 (HIV-1)--transmission by nearly 70% (1). The regimen includes oral ZDV initiated at 14--34 weeks' gestation and continued throughout pregnancy, followed by intravenous ZDV during labor and oral administration of ZDV to the infant for 6 weeks after delivery. In August 1994, a U.S. Public Health Service (USPHS) task force issued recommendations for the use of ZDV for reduction of perinatal HIV-1 transmission (), and in July 1995, USPHS issued recommendations for universal prenatal HIV-1 counseling and HIV-1 testing with consent for all pregnant women in the United States (). Since the publication of the results of PACTG 076, epidemiologic--studies in the United States and France have demonstrated dramatic decreases in perinatal transmission with incorporation of the PACTG 076 ZDV regimen into general clinical practice (4--9).

Since 1994, advances have been made in the understanding of the pathogenesis of HIV-1 infection and in the treatment and monitoring of HIV-1 disease. The rapidity and magnitude of viral turnover during all stages of HIV-1 infection are greater than previously recognized; plasma virions are estimated to have a mean half-life of only 6 hours (10). Thus, current therapeutic interventions focus on early initiation of aggressive combination antiretroviral regimens to maximally suppress viral replication, preserve immune function, and reduce the development of resistance (11). New, potent antiretroviral drugs that inhibit the protease enzyme of HIV-1 are now available. When a protease inhibitor is used in combination with nucleoside analog reverse transcriptase inhibitors, plasma HIV-1 RNA levels can be reduced for prolonged periods to levels that are undetectable by current assays. Improved clinical outcome and survival have been observed among adults receiving such regimens (12,13). Additionally, viral load can now be more directly quantified through assays that measure HIV-1 RNA copy number; these assays have provided powerful new tools to assess disease stage, risk for progression, and the--effects of therapy. These advances have led to substantial changes in the standard of treatment and monitoring for HIV-1--infected adults in the United States.

Advances also have been made in the understanding of the pathogenesis of perinatal HIV-1 transmission. Most perinatal transmission likely occurs close to the time of or during childbirth (15). Additional data that demonstrate the short-term safety of the ZDV regimen are now available as a result of follow-up of infants and women enrolled in PACTG 076; however, data from studies of animals concerning the potential for transplacental carcinogenicity of ZDV affirm the--need for long-term follow-up of children with antiretroviral exposure in utero (16).

These advances have implications for maternal and fetal health. Health-care providers considering the use of antiretroviral agents for HIV-1--infected women during pregnancy must take into account two separate but related issues: 1) antiretroviral treatment of maternal HIV-1 infection, and 2) antiretroviral chemoprophylaxis to reduce the risk for perinatal HIV-1 transmission. The benefits of antiretroviral therapy for a pregnant woman must be weighed against the risk of adverse events to the woman, fetus, and newborn. Although ZDV chemoprophylaxis alone has substantially reduced the risk for perinatal transmission, antiretroviral monotherapy is now considered suboptimal for treatment of HIV-1 infection, and combination drug regimens are considered the standard of care for therapy .

This report reviews the special considerations regarding use of antiretroviral drugs for pregnant women, updates the--results of PACTG 076 and related clinical trials and epidemiologic studies, discusses use of HIV-1 RNA and antiretroviral drug resistance assays during pregnancy, provides updated recommendations on antiretroviral chemoprophylaxis for reducing perinatal transmission, and provides recommendations related to use of elective cesarean delivery as an intervention to reduce perinatal transmission.

These recommendations have been developed for use in the United States. Although perinatal HIV-1 transmission occurs worldwide, alternative strategies may be appropriate in other countries. Policies and practices in other countries--regarding the use of antiretroviral drugs for reduction of perinatal HIV-1 transmission may differ from the recommendations in this report and will depend on local considerations, including availability and cost of ZDV, access by pregnant women to facilities for safe intravenous infusions during--labor, and alternative interventions being evaluated in that area.

Treatment recommendations for pregnant women infected with HIV-1 have been based on the belief that therapies of known benefit to women should not be withheld during pregnancy unless there are known adverse effects on the mother, fetus, or infant and unless these adverse effects outweigh the benefit to the woman (17). Combination antiretroviral therapy, usually consisting of two nucleoside analog reverse transcriptase inhibitors and a protease inhibitor, is the recommended standard treatment for HIV-1--infected adults who are not pregnant (). Pregnancy should not preclude the use of optimal therapeutic regimens. However, recommendations regarding the choice of antiretroviral drugs for treatment of infected pregnant women are subject to unique considerations. These include possible changes in dosing requirements resulting from physiologic changes associated with pregnancy, potential--effects of antiretroviral drugs on the pregnant woman, and the potential short- and long-term effects of the antiretroviral drug on the fetus and newborn, which may not be known for certain antiretroviral drugs.

The decision to use any antiretroviral drug during pregnancy should be made by the woman after discussing with her health-care provider the known and unknown benefits and risks to her and her fetus.

Physiologic changes that occur during pregnancy may--affect the kinetics of drug absorption, distribution, biotransformation, and elimination, thereby also affecting requirements for drug dosing and potentially altering the susceptibility of the pregnant woman to drug toxicity. During pregnancy,--gastrointestinal transit time becomes prolonged; body water and fat increase throughout gestation and are accompanied by increases in cardiac output, ventilation, and liver and renal blood flow; plasma protein concentrations decrease; renal--sodium reabsorption increases; and changes occur in metabolic enzyme pathways in the liver. Placental transport of drugs, compartmentalization of drugs in the embryo/fetus and placenta, biotransformation of drugs by the fetus and placenta, and elimination of drugs by the fetus also can affect drug pharmacokinetics in the pregnant woman. Additional considerations regarding drug use in pregnancy are the effects of the drug on the fetus and newborn, including the potential for teratogenicity, mutagenicity, or carcinogenicity, and the pharmacokinetics and toxicity of transplacentally transferred drugs.

The potential harm to the fetus from maternal ingestion of a specific drug depends not only on the drug itself, but on the dose ingested, the gestational age of the fetus at exposure, the duration of exposure, the interaction with other agents to which the fetus is exposed, and, to an unknown extent, the genetic makeup of the mother and fetus.

Information regarding the safety of drugs in pregnancy is--derived from animal toxicity data, anecdotal experience,--registry data, and clinical trials. Data are limited for antiretroviral drugs, particularly when used in combination therapy. Drug choice should be individualized and must be based on discussion with the woman and available data from preclinical and clinical testing of the individual drugs.

Preclinical data include results of in vitro and animal--in vivo screening tests for carcinogenicity, clastogenicity/ mutagenicity, and reproductive and teratogenic effects. However, the predictive value of such tests for adverse effects in humans is unknown. For example, of approximately 1,200 known animal teratogens, only about 30 are known to be teratogenic in humans (18). In addition to antiretroviral agents, certain drugs commonly used to treat HIV-1--related illnesses demonstrate positive findings on one or more of these screening tests. For example, acyclovir is positive in some in vitro carcinogenicity and clastogenicity assays and is associated with fetal abnormalities in rats; however, data collected on the--basis of human experience from the Acyclovir in Pregnancy Registry have indicated no increased risk for birth defects in infants with in utero exposure to acyclovir . Limited data exist regarding placental passage and long-term animal carcinogenicity for the FDA-approved antiretroviral drugs  (20).

Data are conflicting as to whether receipt of combination antiretroviral therapy during pregnancy is associated with--adverse pregnancy outcomes such as preterm delivery. A retrospective Swiss report evaluated the pregnancy outcome of 37 HIV-1--infected pregnant women treated with combination therapy; all received two reverse transcriptase inhibitors and 16 received one or two protease inhibitors (21). Almost 80% of women experienced one or more typical adverse--effects of the drugs, such as anemia, nausea/vomiting,--aminotransferase elevation, or hyperglycemia. A possible--association of combination antiretroviral therapy with preterm births was noted; 10 of 30 babies were born prematurely. The preterm birth rate did not differ between women receiving combination therapy with or without protease inhibitors. The contribution of maternal HIV-1 disease stage and other covariates that might be associated with a risk for prematurity was not assessed.

The European Collaborative Study and the Swiss Mother + Child HIV-1 Cohort Study investigated the effects of combination retroviral therapy in a population of 3,920 mother--child pairs. Adjusting for CD4⁺ T-lymphocyte count (CD4⁺ count) and intravenous drug use, they found a 2.6-fold (95% confidence interval [CI] = 1.4--4.8) increased odds of preterm--delivery for infants exposed to combination therapy with or without protease inhibitors compared with no treatment; women receiving combination therapy that had been initiated before their pregnancy were twice as likely to deliver--prematurely as those starting therapy during the third trimester (22). However, combination therapy was received by only 323 (8%) women studied. Exposure to monotherapy was not associated with prematurity.

In contrast, in an observational study of pregnant women with HIV-1 infection in the United States (PACTG 367) in which 1,150 (78%) of 1,472 women received combination therapy, no association was found between receipt of combination therapy and preterm birth (23). The highest rate of preterm delivery was among women who had not received any antiretroviral therapy, which is consistent with several other reports demonstrating elevated preterm birth rates among--untreated women with HIV-1 infection (24--26). In a French open-label study of 445 HIV-1--infected women receiving ZDV who had lamivudine (3TC) added to their therapy at 32 weeks' gestation, the rate of preterm delivery was 6%, similar to the 9% rate in a historical control group of women receiving only ZDV (27). Additionally, in a large meta-analysis of seven clinical studies that included 2,123 HIV-infected pregnant women who delivered infants during 1990--1998 and had received antenatal antiretroviral therapy and 1,143 women who did not receive antenatal antiretroviral therapy, use of multiple antiretroviral drugs as compared with no treatment or treatment with one drug was not associated with increased rates of preterm labor, low birth weight, low Apgar scores, or--stillbirth (28).

Until more information is known, HIV-1--infected pregnant women who are receiving combination therapy for their HIV-1 infection should continue their provider-recommended--regimen. They should receive careful, regular monitoring for pregnancy complications and for potential toxicities.

Hyperglycemia, new-onset diabetes mellitus, exacerbation of existing diabetes mellitus, and diabetic ketoacidosis have been reported with receipt of protease inhibitor antiretroviral drugs by HIV-1--infected patients (29--32). In addition, pregnancy is itself a risk factor for hyperglycemia; it is unknown--if the use of protease inhibitors will increase the risk for--pregnancy-associated hyperglycemia. Clinicians caring for HIV-1--infected pregnant women who are receiving protease inhibitor therapy should be aware of the risk of this complication and closely monitor glucose levels. Symptoms of hyper-glycemia should be discussed with pregnant women who are receiving protease inhibitors.

Nucleoside analog drugs are known to induce mitochondrial dysfunction because the drugs have varying affinity for mitochondrial gamma DNA polymerase. This affinity can interfere with mitochondrial replication, resulting in mitochondrial DNA depletion and dysfunction (33). The relative potency of the nucleosides in inhibiting mitochondrial gamma DNA polymerase in vitro is highest for zalcitabine (ddC), followed by didanosine (ddI), stavudine (d4T), 3TC, ZDV and abacavir (ABC) (34). Toxicity related to mitochondrial dysfunction has been reported to occur in infected patients--receiving long-term treatment with nucleoside analogs and generally has resolved with discontinuation of the drug or drugs; a possible genetic susceptibility to these toxicities has been suggested (33). These toxicities may be of particular concern for pregnant women and infants with in utero exposure to nucleoside analog drugs.

During Pregnancy

Clinical disorders linked to mitochondrial toxicity include neuropathy, myopathy, cardiomyopathy, pancreatitis, hepatic steatosis, and lactic acidosis. Among these disorders, symptomatic lactic acidosis and hepatic steatosis may have a female preponderance (35). These syndromes have similarities to rare but life-threatening syndromes that occur during pregnancy, most often during the third trimester: acute fatty liver, and the combination of hemolysis, elevated liver enzymes and low platelets (the HELLP syndrome). Several investigators have correlated these pregnancy-related disorders with a recessively inherited mitochondrial abnormality in the fetus/infant that results in an inability to oxidize fatty acids (36--38). Since the mother would be a heterozygotic carrier of the abnormal gene, the risk for liver toxicity might be increased during pregnancy because the mother would be unable to properly oxidize both maternal and accumulating fetal fatty acids (39). Additionally, animal studies have demonstrated that in late gestation, pregnant mice have significant reductions (25%--50%) in--mitochondrial fatty acid oxidation and that exogeneously--administered estradiol and progesterone can reproduce these effects (40,41); whether this can be translated to humans is unknown. However, these data suggest that a disorder of--mitochondrial fatty acid oxidation in the mother or her fetus during late pregnancy may play a role in the development of acute fatty liver of pregnancy and HELLP syndrome and possibly contribute to susceptibility to antiretroviral-associated mitochondrial toxicity.

Lactic acidosis with microvacuolar hepatic steatosis is a toxicity related to nucleoside analog drugs that is thought to be related to mitochondrial toxicity; it has been reported to--occur in infected persons treated with nucleoside analog drugs for long periods (>6 months). Initially, most cases were associated with ZDV, but later other nucleoside analog drugs, particularly d4T, have been associated with the syndrome. In a report from the FDA Spontaneous Adverse Event Program of 106 patients with this syndrome (60 receiving combination and 46 receiving single nucleoside analog therapy), typical initial symptoms included 1 to 6 weeks of nausea, vomiting, abdominal pain, dyspnea, and weakness (35). Metabolic acidosis with elevated serum lactate and elevated hepatic enzymes was common. Patients described in that report were predominantly female and overweight. The incidence of this syndrome may be increasing, possibly as a result of increased use of combination nucleoside analog therapy or increased recognition of the syndrome. In a cohort of infected patients receiving nucleoside analog therapy followed at Johns Hopkins University during 1989--1994, the incidence of the hepatic steatosis syndrome was 0.13% per year (42). However, in a report from a cohort of 964 HIV-1--infected persons followed in France--for 2 years during 1997--1999, the incidence of symptomatic hyperlactatemia was 0.8% per year for all patients and 1.2% for patients receiving a regimen including d4T (43).

The frequency of this syndrome in pregnant HIV-1--infected women receiving nucleoside analog treatment is unknown. In 1999, Italian researchers reported a case of severe lactic acidosis in an infected pregnant woman who was receiving d4T-3TC at the time of conception and throughout pregnancy and who experienced symptoms and fetal death at 38 weeks' gestation (44). Bristol-Myers Squibb has reported three--maternal deaths due to lactic acidosis, two with and one without accompanying pancreatitis, among women who were--either pregnant or postpartum and whose antepartum therapy during pregnancy included d4T and ddI in combination with other antiretroviral agents (either a protease inhibitor or nevirapine) (45). All women were receiving treatment with these agents at the time of conception and continued for the duration of pregnancy; all presented late in gestation with symptomatic disease that progressed to death in the immediate postpartum period. Two cases were also associated with fetal death.

It is unclear if pregnancy augments the incidence of the--lactic acidosis/hepatic steatosis syndrome that has been--reported for nonpregnant persons receiving nucleoside--analog treatment. However, because pregnancy itself can--mimic some of the early symptoms of the lactic acidosis/ hepatic steatosis syndrome or be associated with other disorders of liver metabolism, these cases emphasize the need for physicians caring for HIV-1--infected pregnant women receiving nucleoside analog drugs to be alert for early signs of this syndrome. Pregnant women receiving nucleoside analog drugs should have hepatic enzymes and electrolytes assessed more frequently during the last trimester of pregnancy, and any new symptoms should be evaluated thoroughly. Additionally,--because of the reports of several cases of maternal mortality secondary to lactic acidosis with prolonged use of the combination of d4T and ddI by HIV-1--infected pregnant women, clinicians should prescribe this antiretroviral combination during pregnancy with caution and generally only when other nucleoside analog drug combinations have failed or have caused unacceptable toxicity or side effects.

In Utero Exposure

A study conducted in France reported that in a cohort of 1,754 uninfected infants born to HIV-1--infected women who received antiretroviral drugs during pregnancy, eight infants with in utero or neonatal exposure to either ZDV-3TC (four infants) or ZDV alone (four infants) developed indications of mitochondrial dysfunction after the first few months of life (46). Two of these infants (both of whom had been exposed to ZDV-3TC) contracted severe neurologic disease and died, three had mild to moderate symptoms, and three had no symptoms but had transient laboratory abnormalities. An association between these findings and in utero exposure to antiretroviral drugs has not been definitively established.

In infants followed through age 18 months in PACTG 076, the occurrence of neurologic events was rare; seizures occurred in one child exposed to ZDV and two exposed to placebo, and one child in each group had reported spasticity. Mortality at 18 months was 1.4% among infants given ZDV compared with 3.5% among those given placebo (47). The Perinatal Safety Review Working Group performed a retrospective--review of deaths occurring among children born to HIV-1--infected women and followed during 1986--1999 in five large prospective U.S. perinatal cohorts. No deaths similar to those reported from France or with clinical findings attributable to mitochondrial dysfunction were identified in a database of >16,000 uninfected children born to HIV-1--infected women with and without antiretroviral drug exposure (48). However, most of the infants with antiretroviral exposure had been--exposed to ZDV alone and only a relatively small proportion (approximately 6%) had been exposed to ZDV-3TC. In an African perinatal trial (PETRA) that compared three regimens of ZDV-3TC (during pregnancy starting at 36 weeks' gestation, during labor, and through 1 week postpartum; during labor and postpartum; and during labor only) with placebo for prevention of transmission, data have been reviewed relating to neurologic adverse events among 1,798 children who participated. No increased risk of neurologic events was--observed among children treated with ZDV-3TC compared with placebo, regardless of the intensity of treatment (49). Finally, in a study of 382 uninfected infants born to HIV-1--infected women, echocardiograms were prospectively performed every 4 to 6 months during the first 5 years of life; 9% of infants had been exposed to ZDV prenatally (50). No--significant differences in ventricular function were observed between infants exposed and not exposed to ZDV.

Even if the association of mitochondrial dysfunction and in utero antiretroviral exposures is demonstrated, the development of severe or fatal mitochondrial disease in these infants appears to be extremely rare and should be compared against the clear benefit of ZDV in reducing transmission of a fatal infection by nearly 70% (51). These results emphasize the importance of the existing Public Health Service recommendation for long-term follow-up for any child with in utero exposure to antiretroviral drugs.

Health-care providers who are treating HIV-1--infected pregnant women and their newborns are strongly advised to--report instances of prenatal exposure to antiretroviral drugs (either alone or in combination) to the Antiretroviral Pregnancy Registry. This registry is an epidemiologic project to collect observational, nonexperimental data regarding antiretroviral exposure during pregnancy for the purpose of assessing the potential teratogenicity of these drugs. Registry data will be used to supplement animal toxicology studies and assist clinicians in weighing the potential risks and benefits of treatment for individual patients. The Antiretroviral Pregnancy Registry is a collaborative project of pharmaceutical manufacturers with an advisory committee of obstetric and pediatric practitioners. The registry does not use patient names, and registry staff obtain birth outcome follow-up information from the reporting physician. Referrals should be directed to

In 1996, final results were reported for all 419 infants--enrolled in PACTG 076. The results concur with those initially reported in 1994; the Kaplan-Meier estimated HIV-1 transmission rate for infants who received placebo was 22.6%, compared with 7.6% for those who received ZDV, a 66% reduction in risk for transmission (52).

The mechanism by which ZDV reduced transmission in PACTG 076 participants has not been fully defined. The--effect of ZDV on maternal HIV-1 RNA does not fully--account for the observed efficacy of ZDV in reducing transmission. Preexposure prophylaxis of the fetus or infant may offer substantial protection. If so, transplacental passage of antiretroviral drugs would be crucial for prevention of transmission. Additionally, in placental perfusion studies, ZDV has been metabolized into the active triphosphate within the placenta (53,54), which could provide additional protection against in utero transmission. This phenomenon may be unique to ZDV because metabolism to the active triphosphate form within the placenta has not been observed in the other nucleoside analogs that have been evaluated (i.e., ddI and ddC) (55,56).

In PACTG 076, similar rates of congenital abnormalities occurred among infants with and without in utero ZDV--exposure. Data from the Antiretroviral Pregnancy Registry also have demonstrated no increased risk for congenital abnormalities among infants born to women who receive ZDV antenatally compared with the general population (57). Among uninfected infants from PACTG 076 followed from birth to a median age of 4.2 years (range 3.2--5.6 years), no differences were noted in growth, neurodevelopment, or immunologic status between infants born to mothers who received ZDV compared with those born to mothers who received placebo (58). No malignancies have been observed in short-term (i.e., up to age 6 years) follow-up of >727 infants from PACTG 076 or from a prospective cohort study involving infants with in utero ZDV exposure (59). However, follow-up is too limited to provide a definitive assessment of carcinogenic risk with human exposure. Long-term monitoring continues to be recommended for all infants who have received in utero ZDV exposure or in utero exposure to any of the antiretroviral drugs.

The efficacy of ZDV chemoprophylaxis for reducing HIV-1 transmission among populations of infected women with characteristics unlike those of the PACTG 076 population has been evaluated in another perinatal protocol (PACTG 185) and in prospective cohort studies. PACTG 185 enrolled pregnant women with advanced HIV-1 disease and low CD4⁺ counts who were receiving antiretroviral therapy; 24% had received ZDV before the current pregnancy (60). All women and infants received the three-part ZDV regimen combined with either infusions of hyperimmune HIV-1 immunoglobulin (HIVIG) containing high levels of antibodies to HIV-1 or standard intravenous immunoglobulin (IVIG) without HIV-1 antibodies. Because advanced maternal HIV-1 disease has been associated with increased risk for perinatal transmission, the transmission rate in the control group was hypothesized to be 11%--15% despite the administration of ZDV. At the first interim analysis, the transmission rate for the combined group was only 4.8% and did not substantially differ by whether the women received HIVIG or IVIG or by duration of ZDV use (60). The results of this trial confirm the efficacy of ZDV observed in PACTG 076 and extend this efficacy to women with advanced disease, low CD4⁺ count, and prior ZDV therapy. Rates of perinatal transmission have been documented to be as low as 3%--4% among women with HIV-1 infection who receive all three components of the ZDV regimen,--including women with advanced HIV-1 disease (6,60).

At least two studies suggest that antenatal use of combination antiretroviral regimens might further reduce transmission. In an open-label, nonrandomized study of 445 women with HIV-1 infection in France, 3TC was added at 32 weeks' gestation to standard ZDV prophylaxis; 3TC was also given to the infant for 6 weeks in addition to ZDV (27). The transmission rate in the ZDV-3TC group was 1.6% (95% CI = 0.7%--3.3%); in comparison, the transmission rate in a historical control group of women receiving only ZDV was 6.8% (95% CI = 5.1%--8.7%). In a longitudinal epidemiologic study conducted in the United States since 1990, transmission was observed in 20% of women with HIV-1 infection who--received no antiretroviral treatment during pregnancy, 10.4% who received ZDV alone, 3.8% who received combination therapy without protease inhibitors, and 1.2% who received combination therapy with protease inhibitors (61).

International Antiretroviral Prophylaxis Clinical Trials

In a trial evaluating short-course antenatal/intrapartum ZDV prophylaxis and perinatal transmission among non-breastfeeding women in Thailand, administration of ZDV 300 mg twice daily for 4 weeks antenatally and 300 mg every 3 hours orally during labor was shown to reduce perinatal transmission by approximately 50% compared with placebo (62). The transmission rate was 19% in the placebo group versus 9% in the ZDV group. A second, four-arm factorial design trial in Thailand compared administration of ZDV antenatally starting at 28 or 36 weeks' gestation, orally intrapartum, and to the neonate for 3 days or 6 weeks. At an interim analysis, the transmission rate in the arm receiving ZDV antenatally starting at 36 weeks and postnatally for 3 days to the infant was 10%, which was significantly higher than for the long--long arm (antenatal starting at 28 weeks and infant administration for 6 weeks) (63). The transmission rate in the short--short arm of this study was similar to the 9% observed with short antenatal/intrapartum ZDV in the first Thai study. The rate of in utero transmission was higher among women in the short antenatal arms compared with those receiving longer antenatal therapy, suggesting that longer treatment of the infant cannot substitute for longer treatment of the mother.

A third trial in Africa (PETRA trial) among breastfeeding HIV-1--infected women has shown that a combination regimen of ZDV and 3TC administered starting at 36 weeks' gestation, orally intrapartum, and for 1 week postpartum to the woman and infant reduced transmission at age 6 weeks by approximately 50% compared with placebo (64). The transmission rate at age 6 weeks was 15% in the placebo group versus 6% with the three-part ZDV-3TC regimen. This efficacy is similar to the efficacy observed in the Thailand study of antepartum/intrapartum short-course ZDV in non-breastfeeding women (62).

Investigators have identified two possible intrapartum/ postpartum regimens (either ZDV-3TC or nevirapine) that could provide an effective intrapartum/postpartum intervention for women for whom the diagnosis of HIV-1 is not made until near to or during labor. The PETRA African ZDV-3TC trial among breastfeeding HIV-1--infected women also--demonstrated that an intrapartum/postpartum regimen, started during labor and continued for 1 week postpartum in the woman and infant, reduced transmission at age 6 weeks from 15% in the placebo group to 9% in the group receiving the two-part ZDV-3TC regimen, a reduction of 40% (64). In this trial, oral ZDV-3TC administered solely during--the intrapartum period was not effective in lowering--transmission. Another study in Uganda (HIVNET 012), again in a breastfeeding population, demonstrated that a single--200-mg oral dose of nevirapine given to the mother at onset of labor combined with a single 2-mg/kg oral dose given to her infant at age 48--72 hours reduced transmission by nearly 50% compared with a very short regimen of ZDV given orally during labor and to the infant for 1 week (65). Transmission at age 6 weeks was 12% in the nevirapine group compared with 21% in the ZDV group. A subsequent trial in South Africa demonstrated similar transmission rates with a modified HIVNET 012 nevirapine regimen (nevirapine given to the woman as a single dose during labor with a second dose at 48 hours postpartum, and a single dose to the infant at age--48 hours) compared with the PETRA regimen of oral ZDV-3TC during labor and for 1 week after delivery to the mother and infant (66). Transmission rates at age 8 weeks were 13.3% in the nevirapine arm and 10.9% in the ZDV-3TC arm.

Two clinical trials have suggested that the addition of the HIVNET 012 single-dose nevirapine regimen to short-course ZDV may provide increased efficacy in reducing perinatal transmission. A study of nonbreastfeeding women in--Thailand compared a short-course ZDV regimen (starting at 28 weeks' gestation, given orally intrapartum, and for 1 week to the infant) with two combination regimens: short-course ZDV plus single-dose intrapartum/neonatal nevirapine, and short-course ZDV plus intrapartum maternal nevirapine only. In the short-course ZDV-only arm, enrollment was discontinued by the Data and Safety Monitoring Board at the first--interim analysis because transmission was significantly higher among those receiving ZDV alone compared with those--receiving the intrapartum/neonatal nevirapine combination regimen (67). The study is continuing to enroll to allow comparison of the two combination arms. A second open-label study in Cote d'Ivoire reported a 7.1% transmission rate at age 4 weeks with administration of short-course ZDV (starting at 36 weeks, given orally intrapartum, and for 1 week to the infant) combined with single-dose intrapartum/neonatal nevirapine. This was lower than for a nonconcurrent historical control group receiving ZDV alone (68).

In contrast to these studies, which evaluated combining single-dose nevirapine with short-course ZDV, a study in the United States, Europe, Brazil, and the Bahamas (PACTG 316) evaluated whether the addition of the HIVNET 012 single-dose nevirapine regimen to standard antiretroviral therapy (at minimum the 3-part full ZDV regimen) would provide additional benefits in lowering transmission. In this study, 1,506 pregnant women with HIV-1 infection who were receiving antiretroviral therapy (77% were receiving combination antiretroviral regimens) were randomized to receive a single dose of nevirapine or nevirapine placebo at onset of labor, and their infants received a single dose (according to the maternal randomization) at age 48 hours. Transmission was not significantly different between groups, occurring in 1.6% of women in the placebo group and 1.4% among women in the nevirapine group (69).

Certain data indicate that postexposure antiretroviral prophylaxis of infants whose mothers did not receive antepartum or intrapartum antiretroviral drugs might provide some protection against transmission. Although data from some epidemiologic studies do not support efficacy of postnatal ZDV alone, other data demonstrate efficacy if ZDV is started rapidly following birth (6,70,71). In a study from North--Carolina, the rate of infection among HIV-1--exposed infants who received only postpartum ZDV chemoprophylaxis was similar to that observed among infants who received no ZDV chemoprophylaxis (6). However, another epidemiologic study from New York State determined that administration of ZDV to the neonate for 6 weeks was associated with a significant--reduction in transmission if the drug was initiated within 24 hours of birth (the majority of infants started within 12 hours) (70,71). Consistent with a possible preventive effect of rapid postexposure prophylaxis, a retrospective case-control study of health-care workers from the United States, France, and the United Kingdom who had nosocomial exposure to--HIV-1--infected blood determined that postexposure use of ZDV was associated with reduced odds of contracting HIV-1--(adjusted odds ratio = 0.2; 95% CI = 0.1--0.6) . Several ongoing clinical trials are attempting to determine the optimal postexposure antiretroviral prophylaxis regimen for--infants.

The correlation of HIV-1 RNA levels with risk for disease progression in nonpregnant infected adults suggests that HIV-1 RNA should be monitored during pregnancy at least as often as recommended for persons who are not pregnant (i.e., every 3 to 4 months or approximately once each trimester). In addition, HIV-1 RNA levels should be evaluated at 34--36 weeks of gestation to allow discussion of options for mode of delivery based on HIV-1 RNA results and clinical circumstances. Although no data indicate that pregnancy accelerates HIV-1 disease progression, longitudinal measurements of HIV-1 RNA levels during and after pregnancy have been evaluated in only a limited number of prospective cohort studies. In one cohort of 198 HIV-1--infected women, plasma HIV-1 RNA levels were higher at 6 months postpartum than during pregnancy in many women; this increase was observed in women regardless of ZDV use during and after pregnancy (73).

Initial data regarding the correlation of viral load with risk for perinatal transmission were conflicting, with some studies suggesting an absolute correlation between HIV-1 RNA copy number and risk of transmission (74). However, although higher HIV-1 RNA levels have been observed among women who transmitted HIV-1 to their infants, overlap in HIV-1 RNA copy number has been observed in women who transmitted and those who did not transmit the virus. Transmission has been observed across the entire range of HIV-1 RNA levels (including in women with HIV-1 RNA copy number below the limit of detection of the assay), and the predictive value of RNA copy number for transmission in an individual woman has been relatively poor (73,75,76). In PACTG 076, antenatal maternal HIV-1 RNA copy number was associated with HIV-1 transmission in women receiving placebo. In women receiving ZDV, the relationship was markedly attenuated and no longer statistically significant (52). An HIV-1 RNA threshold below which there was no risk for transmission was not identified; ZDV was effective in reducing transmission regardless of maternal HIV-1 RNA copy number (52,77).

More recent data from larger numbers of ZDV-treated--infected pregnant women indicate that HIV-1 RNA levels correlate with risk of transmission even among women treated with antiretroviral agents (62,78--80). Although the risk for perinatal transmission in women with HIV-1 RNA below the level of assay quantitation appears to be extremely low, transmission from mother to infant has been reported among women with all levels of maternal HIV-1 RNA. Additionally, although HIV-1 RNA may be an important risk factor for transmission, other factors also appear to play a role (80--82).

Although there is a general correlation between viral load in plasma and in the genital tract, discordance has also been--reported, particularly between HIV-1 proviral load in blood and genital secretions (83--86). If exposure to HIV-1 in the maternal genital tract during delivery is a risk factor for perinatal transmission, plasma HIV-1 RNA levels might not--always be an accurate indicator of risk. Long-term changes in one compartment (such as can occur with antiretroviral treatment) may or may not be associated with comparable changes in other body compartments. Further studies are needed to determine the effect of antiretroviral drugs on genital tract viral load and the association of such effects on the risk of perinatal HIV-1 transmission. In the short-course ZDV trial in Thailand, plasma and cervicovaginal HIV-1 RNA levels were reduced by ZDV treatment, and each independently correlated with perinatal transmission (87). The full ZDV chemoprophylaxis regimen, alone or in combination with other antiretroviral agents, including intravenous ZDV during delivery and the administration of ZDV to the infant for the first 6 weeks of life, should be discussed with and offered to all infected pregnant women regardless of their HIV-1 RNA level.

Results of epidemiologic and clinical trials suggest that women receiving highly active antiretroviral regimens that--effectively reduce HIV-1 RNA to <1,000 copies/mL or undetectable levels have very low rates of perinatal transmission (27,61,69,88). However, since transmission can occur even at low or undetectable HIV-1 RNA copy numbers, RNA levels should not be a determining factor when deciding whether to use ZDV for chemoprophylaxis. Additionally, the efficacy of ZDV is not solely related to lowering viral load. In one study of 44 HIV-1--infected pregnant women, ZDV was effective in reducing transmission despite minimal effect on HIV-1 RNA levels (89). These results are similar to those observed in PACTG 076 (52). Antiretroviral prophylaxis reduces transmission even among women with HIV-1 RNA levels <1,000 copies/mL (90). Therefore, at a minimum, ZDV prophylaxis should be given even to women who have a very low or--undetectable plasma viral load.

Many women infected with HIV-1 (nearly 60% in some centers) enter pregnancy with a known diagnosis, and nearly half of these women enter the first trimester of pregnancy--receiving treatment with single or multiagent antiretroviral therapy (91). Additionally, as many as 40% of women who have begun antiretroviral therapy before their pregnancy might--require adjustment of their therapeutic regimen during their pregnancy course.

The American College of Obstetrics and Gynecology advocates extending to all women of childbearing age the opportunity to receive preconception counseling as a component of routine primary medical care. It is recognized that >40% of pregnancies may be unintended and that the diagnosis of pregnancy most frequently occurs late in the first trimester when organogenesis is nearly completed. Preconception care can identify risk factors for adverse maternal or fetal outcome (e.g., age, diabetes, hypertension), provide education and counseling targeted to the patient's individual needs, and treat or stabilize medical conditions before conception to optimize maternal and fetal outcomes (92).

For women with HIV-1 infection, preconception care must also focus on maternal infection status, viral load, immune status, and therapeutic regimen as well as education regarding perinatal transmission risks and prevention strategies,-- expectations for the child's future, and, where desired, effective contraception until the optimal maternal health status for pregnancy is achieved.

The following components of preconception counseling are recommended for HIV-1--infected women:

• selection of effective and appropriate contraceptive methods to reduce the likelihood of unintended pregnancy;
• education and counseling about perinatal transmission risks, strategies to reduce those risks, and potential effects of HIV-1 or treatment on pregnancy course and outcomes;
• initiation or modification of antiretroviral therapy:
  --- avoid agents with potential reproductive toxicity for the developing fetus (e.g., efavirenz, hydroxyurea) (20),
  --- choose agents effective in reducing the risk of perinatal HIV-1 transmission,
  --- attain a stable, maximally suppressed maternal viral load,
  --- evaluate and control for therapy-associated side effects that may adversely affect maternal/fetal health outcomes (e.g., hyperglycemia, anemia, hepatic toxicity),
• evaluation and appropriate prophylaxis for opportunistic infections and administration of medical immunizations (e.g., influenza, pneumococcal, or hepatitis B vaccines) as indicated;
• optimization of maternal nutritional status;
• institution of the standard measures for preconception evaluation and management (e.g., assessment of reproductive and familial genetic history, screening for infectious diseases/sexually transmitted diseases, and initiation of folic acid supplementation);
• screening for maternal psychological and substance abuse disorders; and
• planning for perinatal consultation if desired or indicated.

HIV-1--infected women of childbearing potential receive primary health-care services in various clinical settings, e.g., family planning, family medicine, internal medicine, obstetrics/ gynecology. It is imperative that primary health-care providers consider the fundamental principles of preconception counseling an integral component of comprehensive primary health care for improving maternal/child health outcomes.

Medical care of the HIV-1--infected pregnant woman--requires coordination and communication between the HIV specialist caring for the woman when she is not pregnant and her obstetrician. Decisions regarding use of antiretroviral drugs during pregnancy should be made by the woman after discussion with her health-care provider about the known and--unknown benefits and risks of therapy. Initial evaluation of an infected pregnant woman should include an assessment of HIV-1 disease status and recommendations regarding antiretroviral treatment or alteration of her current anti-retroviral regimen.

This assessment should include the following:

• evaluation of the degree of existing immunodeficiency--determined by CD4⁺ count;
• risk for disease progression as determined by the level of plasma RNA;
• history of prior or current antiretroviral therapy;
• gestational age; and
• supportive care needs.

Decisions regarding initiation of therapy should be the same for women who are not currently receiving antiretroviral therapy and for women who are not pregnant, with the additional consideration of the potential impact of such therapy on the fetus and infant . Similarly, for women currently receiving antiretroviral therapy, decisions regarding alterations in therapy should involve the same considerations as those used for women who are not pregnant. The three-part ZDV chemoprophylaxis regimen, alone or in combination with other antiretroviral agents, should be discussed with and--offered to all infected pregnant women to reduce the risk for perinatal HIV-1 transmission.

Decisions regarding the use and choice of antiretroviral drugs during pregnancy are complex; several competing factors--influencing risk and benefit must be weighed. Discussion--regarding the use of antiretroviral drugs during pregnancy should include the following:

• what is known and not known about the effects of such drugs on the fetus and newborn, including lack of long-term outcome data on the use of any of the available antiretroviral drugs during pregnancy;
• what treatment is recommended for the health of the HIV-1--infected woman; and
• the efficacy of ZDV for reduction of perinatal HIV-1--transmission.

Results from preclinical and animal studies and available clinical information about use of the various antiretroviral agents during pregnancy also should be discussed (20). The hypothetical risks of these drugs during pregnancy should be placed in perspective with the proven benefit of antiretroviral therapy for the health of the infected woman and the benefit of ZDV chemoprophylaxis for reducing the risk for HIV-1 transmission to her infant.

Discussion of treatment options should be noncoercive, and the final decision regarding use of antiretroviral drugs is the responsibility of the woman. Decisions regarding use and choice of antiretroviral drugs for persons who are not pregnant are becoming increasingly complicated as the standard of care moves toward simultaneous use of multiple antiretroviral drugs to suppress viral replication below detectable limits. These decisions are further complicated in pregnancy because the long-term consequences for the infant who has been exposed to antiretroviral drugs in utero are unknown. A woman's decision to refuse treatment with ZDV or other drugs should not result in punitive action or denial of care. Further, use of ZDV alone should not be denied to a woman who wishes to minimize exposure of the fetus to other antiretroviral drugs and therefore, after counseling, chooses to receive only ZDV during pregnancy to reduce the risk for perinatal transmission.

A long-term treatment plan should be developed after discussion between the patient and the health-care provider and should emphasize the importance of adherence to any prescribed antiretroviral regimen. Depending on individual circumstances, provision of support services, mental health services, and drug abuse treatment may be required. Coordination of services among prenatal care providers, primary care and HIV-1 specialty care providers, mental health and drug abuse treatment services, and public assistance programs is essential to ensure adherence of the infected woman to antiretroviral treatment regimens.

General counseling should include what is known regarding risk factors for perinatal transmission. Cigarette smoking,--illicit drug use, and unprotected sexual intercourse with multiple partners during pregnancy have been associated with risk for perinatal HIV-1 transmission (93--97), and discontinuing these practices might reduce this risk. In addition, CDC recommends that infected women in the United States refrain from breastfeeding to avoid postnatal transmission of HIV-1 to their infants through breast milk ; these recommendations also should be followed by women receiving antiretroviral therapy. Passage of antiretroviral drugs into breast milk has been evaluated for only a few antiretroviral drugs. ZDV, 3TC, and nevirapine can be detected in the breast milk of women, and ddI, d4T, abacavir, delavirdine, indinavir, ritonavir, saquinavir and amprenavir can be detected in the breast milk of lactating rats. Limited data are available regarding either the efficacy of antiretroviral therapy for the prevention of postnatal transmission of HIV-1 through breast milk or the toxicity of long-term antiretroviral exposure of the--infant through breast milk.

Women who must temporarily discontinue therapy because of pregnancy-related hyperemesis should not resume therapy until sufficient time has elapsed to ensure that the drugs will be tolerated. To reduce the potential for emergence of resistance, if therapy requires temporary discontinuation for any reason during pregnancy, all drugs should be stopped and--reintroduced simultaneously.

The following recommendations for use of anti-retroviral chemoprophylaxis to reduce the risk for perinatal transmission are based on situations that may be commonly encountered in clinical practice , with relevant considerations highlighted in the subsequent discussion sections. These recommendations are only guidelines, and flexibility should be exercised according to the patient's individual circumstances. In the 1994 recommendations , six clinical situations were delineated on the basis of maternal CD4⁺ count, weeks of gestation, and prior antiretroviral use.--Because current data indicate that the PACTG 076 ZDV regimen also is effective for women with advanced disease, low CD4⁺ count, and prior ZDV therapy, clinical situations based on CD4⁺ count and prior ZDV use are not presented. Additionally, because data indicate that most transmission occurs near the time of or during delivery, ZDV chemoprophylaxis is recommended regardless of weeks of gestation; thus, clinical situations based on weeks of gestation also are not--presented.

The antenatal dosing regimen in PACTG 076 (100 mg--administered orally five times daily)  was selected on the basis of the standard ZDV dosage for adults at the time of the study. However, recent data have indicated that administration of ZDV three times daily will maintain intracellular ZDV triphosphate at levels comparable with those observed with more frequent dosing (99--101). Comparable clinical response also has been observed in some clinical trials among persons receiving ZDV twice daily (102--104). Thus, the current standard ZDV dosing regimen for adults is 200 mg three times daily, or 300 mg twice daily. Because the mechanism by which ZDV reduces perinatal transmission is not known, these dosing regimens may not have equivalent efficacy to that--observed in PACTG 076. However, a regimen of two or three times daily is expected to increase adherence to the regimen.

The recommended ZDV dosage for infants was derived from pharmacokinetic studies performed among full-term infants (105). ZDV is primarily cleared through hepatic glucuronidation to an inactive metabolite. The glucuronidation metabolic enzyme system is immature in neonates, leading to prolonged ZDV half-life and clearance compared with older infants (ZDV half-life: 3.1 hours versus 1.9 hours; clearance: 10.9 versus 19.0 mL/minute/kg body weight, respectively). Because premature infants have even greater immaturity in hepatic metabolic function than full-term infants, further prolongation of clearance may be expected. In a study of 15 premature infants who were at 26--33 weeks' gestation and who received different ZDV dosing regimens, mean ZDV half-life was 7.2 hours and mean clearance was 2.5 mL/minute/kg body weight during the first 10 days of life (106). At a mean age of 18 days, a decrease in half-life (4.4 hours) and increase in clearance (4.3 mL/minute/kg body weight) were found. The--appropriate ZDV dosage for premature infants has not been defined but is being evaluated in a phase I clinical trial among premature infants <34 weeks' gestation. The dosing regimen being studied is 1.5 mg/kg body weight orally or intravenously every 12 hours for the first 2 weeks of life; for infants aged--2 to 6 weeks, the dose is increased to 2 mg/kg body weight every 8 hours.

Recommendation. Pregnant women with HIV-1 infection must receive standard clinical, immunologic, and virologic evaluation. Recommendations for initiation and choice of antiretroviral therapy should be based on the same parameters used for persons who are not pregnant, although the known and unknown risks and benefits of such therapy during pregnancy must be considered and discussed. The three-part ZDV chemoprophylaxis regimen, initiated after the first trimester, should be recommended for all pregnant women with HIV-1 infection regardless of antenatal HIV-1 RNA copy number to reduce the risk for perinatal transmission. The combination of ZDV chemoprophylaxis with additional antiretroviral drugs for treatment of HIV-1 infection is recommended for infected women whose clinical, immunologic, or virologic status--requires treatment or whose HIV-1 RNA is >1,000 copies/mL regardless of their clinical or immunologic status. Women who are in the first trimester of pregnancy may consider--delaying initiation of therapy until after 10--12 weeks'--gestation.

Discussion. When ZDV is administered in the three-part PACTG 076 regimen, perinatal transmission is reduced by approximately 70%. Although the mechanism by which ZDV reduces transmission is not known, protection is likely multifactorial. Preexposure prophylaxis of the infant is provided by passage of ZDV across the placenta so that inhibitory levels of the drug are present in the fetus during the birth process.--Although placental passage of ZDV is excellent, that of other antiretroviral drugs is variable . Therefore, when combination antiretroviral therapy is initiated during pregnancy, ZDV should be included as a component of antenatal therapy whenever possible. Because the mechanism by which ZDV reduces transmission is not known, the intrapartum and newborn ZDV components of the chemoprophylactic regimen should be administered to reduce perinatal HIV-1 transmission. If a woman does not receive ZDV as a component of her antenatal antiretroviral regimen, intrapartum and newborn ZDV should still be recommended.

Because of the evolving and complex nature of the management of HIV-1 infection, a specialist with experience in the treatment of pregnant women with HIV-1 infection should be involved in their care. Women should be informed that--potent combination antiretroviral regimens have substantial benefit for their own health and may provide enhanced protection against perinatal transmission. Several studies have indicated that for women with low or undetectable HIV-1 RNA levels (e.g., <1,000 copies/mL) rates of perinatal--transmission are extremely low, particularly when they have received antiretroviral therapy (61,78,79). However, there is no threshold below which lack of transmission can be assured, and the long-term effects of in utero exposure to multiple antiretroviral drugs are unknown. Decisions regarding the use and choice of an antiretroviral regimen should be individualized based on discussion with the woman about the following factors:

• her risk for disease progression and the risks and benefits of delaying initiation of therapy;
• possible benefit of lowering viral load for reducing perinatal transmission;
• potential drug toxicities and interactions with other drugs
• the need for strict adherence to the prescribed drug schedule to avoid the development of drug resistance;
• unknown long-term effects of in utero drug exposure on the infant; and
• preclinical, animal, and clinical data relevant to use of the currently available antiretroviral agents during pregnancy.

Because the period of organogenesis (when the fetus is most susceptible to potential teratogenic effects of drugs) is during the first 10 weeks of gestation and the risks of antiretroviral therapy during that period are unknown, women in the first trimester of pregnancy might wish to delay initiation of therapy until after 10--12 weeks' gestation. This decision should be carefully considered by the health-care provider and the--patient; a discussion should include an assessment of the woman's health status, the benefits and risks of delaying initiation of therapy for several weeks, and the fact that most perinatal HIV-1 transmission likely occurs late in pregnancy or during delivery. Treatment with efavirenz should be avoided during the first trimester because significant teratogenic--effects in rhesus macaques were seen at drug exposures similar to those representing human exposure(20). Hydroxyurea is a potent teratogen in a variety of animal species and should also be avoided during the first trimester.

When initiation of antiretroviral therapy is considered--optional on the basis of current guidelines for treatment of nonpregnant persons , infected pregnant women should be counseled regarding the potential benefits of standard combination therapy and should be offered such therapy, including the three-part ZDV chemoprophylaxis regimen. Although such women are at low risk for clinical disease progression if combination therapy is delayed, antiretroviral therapy that successfully reduces HIV-1 RNA to levels <1,000 copies/mL may substantially lower the risk of perinatal HIV-1 transmission and lessen the need for consideration of elective cesarean--delivery as an intervention to reduce transmission risk.

When combination therapy is administered, the regimen should be chosen from those recommended for nonpregnant adults . Dual nucleoside analog therapy without the--addition of either a protease inhibitor or nonnucleoside--reverse transcriptase inhibitor is not recommended for nonpregnant adults because of the potential for inadequate viral suppression and rapid development of resistance (107). For pregnant women not meeting the criteria for antiretroviral therapy for their own health, and receiving antiretroviral drugs only for prevention of perinatal transmission (e.g., those with HIV-1 RNA <1,000 copies/mL), dual nucleoside therapy may be considered in selected circumstances. If combination therapy is given principally to reduce perinatal transmission and would have been optional if the woman were not pregnant, consideration may be given to discontinuing therapy postnatally, with the option to reinitiate treatment according to standard criteria for nonpregnant women. If drugs are discontinued postnatally, all drugs should be stopped simultaneously. Discussion regarding the decision to continue or stop combination therapy postpartum should occur before beginning therapy during pregnancy.

Antiretroviral prophylaxis has been beneficial in preventing perinatal transmission even for infected pregnant women with HIV-1 RNA levels <1,000 copies/mL. In a meta-analysis of factors associated with perinatal transmission among women whose infants were infected despite the women's having HIV-1 RNA <1,000 copies/mL at or near delivery, transmission was only 1.0% among women receiving antenatal antiretroviral therapy (primarily ZDV alone) compared with 9.8% among those receiving no antenatal therapy (90). Therefore, use of antiretroviral prophylaxis is recommended for all pregnant women with HIV-1 infection regardless of antenatal HIV-1 RNA level.

The time-limited use of ZDV alone during pregnancy for chemoprophylaxis against perinatal transmission is controversial. Standard combination antiretroviral regimens for treatment of HIV-1 infection should be discussed and should be offered to all pregnant women with HIV-1 infection regardless of viral load; they are recommended for all pregnant women with HIV-1 RNA levels >1,000 copies/mL. Some women may wish to restrict exposure of their fetus to antiretroviral drugs during pregnancy and still reduce the risk of transmitting HIV-1 to their infant. Additionally, for women with HIV-1 RNA levels <1,000 copies/mL, time-limited use of ZDV during the second and third trimesters of pregnancy is less likely to--induce the development of resistance because of the limited viral replication existing in the patient and the time-limited exposure to the antiretroviral drug. For example, the development of ZDV resistance was unusual among the healthy population of women who participated in PACTG 076 (108). The use of ZDV chemoprophylaxis alone (or, in selected circumstances, dual nucleosides) during pregnancy might be an--appropriate option for these women.

2. HIV-1--infected women receiving antiretroviral therapy during the current pregnancy

Recommendation. HIV-1 infected women receiving antiretroviral therapy whose pregnancy is identified after the first trimester should continue therapy. ZDV should be a component of the antenatal antiretroviral treatment regimen after the first trimester whenever possible, although this may not always be feasible. Women receiving antiretroviral therapy whose pregnancy is recognized during the first trimester should be counseled regarding the benefits and potential risks of antiretroviral administration during this period, and continuation of therapy should be considered. If therapy is discontinued during the first trimester, all drugs should be stopped and reintroduced simultaneously to avoid the development of drug resistance. Regardless of the antepartum antiretroviral--regimen, ZDV administration is recommended during the--intrapartum period and for the newborn.

Discussion. Women who have been receiving antiretroviral treatment for their HIV-1 infection should continue treatment during pregnancy. Discontinuation of therapy could lead to an increase in viral load, which could result in decline in--immune status and disease progression as well as adverse--consequences for both the fetus and the woman.

Although ZDV should be a component of the antenatal antiretroviral treatment whenever possible, there may be circumstances, such as the occurrence of significant ZDV-related toxicity, when this is not feasible. Additionally, women--receiving an antiretroviral regimen that does not contain ZDV but who have HIV-1 RNA levels that are consistently very low or undetectable (e.g., <1,000 copies/mL) have a very low risk of perinatal transmission (61), and there may be concerns that the addition of ZDV to the current regimen could--compromise adherence to treatment.

The maternal antenatal antiretroviral treatment regimen should be continued on schedule as much as possible during labor to provide maximal virologic effect and to minimize the chance of development of drug resistance. If a woman has not received ZDV as a component of her antenatal therapeutic antiretroviral regimen, intravenous ZDV should still be--administered during the intrapartum period whenever feasible. ZDV and d4T should not be administered together because of potential pharmacologic antagonsim; options for women receiving oral d4T as part of their antenatal therapy include either continuation of oral d4T during labor without intravenous ZDV or withholding oral d4T during the period of--intravenous ZDV administration during labor. Additionally, the infant should receive the standard 6-week course of ZDV.

For women with suboptimal suppression of HIV-1 RNA (i.e., >1,000 copies/mL) near the time of delivery despite having received prenatal ZDV prophylaxis with or without combination antiretroviral therapy, it is not known if administration of additional antiretroviral drugs during labor and delivery provides added protection against perinatal transmission. In the HIVNET 012 study among Ugandan women who had not received antenatal antiretroviral therapy, a 2-dose nevirapine regimen (single dose to the woman at the onset of labor and single dose to the infant at age 48 hours) significantly reduced perinatal transmission compared with a very short intrapartum/1 week postpartum ZDV regimen (65). For women in the United States, Europe, Brazil, and the Bahamas receiving antenatal antiretroviral therapy, addition of the--2-dose nevirapine regimen did not result in lower transmission rates (69). Given the lack of further reduction of transmission with nevirapine added to one of the standard antepartum regimens used in developed countries and the potential development of nevirapine resistance (See Antiretroviral Drug Resistance and Resistance Testing in Pregnancy), addition of nevirapine during labor for women--already receiving antiretroviral therapy is not recommended in the United States.

Women receiving antiretroviral therapy may realize they are pregnant early in gestation and want to consider temporarily stopping antiretroviral treatment until after the first trimester because of concern for potential teratogenicity. Data are--insufficient to support or refute the teratogenic risk of antiretroviral drugs when administered during the first--10 weeks of gestation; certain drugs are of more concern than others (20). The decision to continue therapy during the first trimester should be carefully considered by the clinician and the pregnant woman. Discussions should include considerations such as gestational age of the fetus; the woman's clinical, immunologic, and virologic status; and the known and unknown potential effects of the antiretroviral drugs on the fetus. If antiretroviral therapy is discontinued during the first trimester, all agents should be stopped and restarted--simultaneously in the second trimester to avoid the development of drug resistance. No data are available to address whether temporary discontinuation of therapy is harmful for the woman or fetus.

Health-care providers might consider administering ZDV in combination with other antiretroviral drugs to newborns of women with a history of prior antiretroviral therapy, particularly in situations in which the woman is infected with HIV-1 with documented high-level ZDV resistance, has had disease progression while receiving ZDV, or has had extensive prior ZDV monotherapy. The efficacy of this approach is unknown but would be analogous to the use of multiple agents for postexposure prophylaxis for adults after inadvertent--exposure. However, the appropriate dosage and short- and long-term safety of many antiretroviral agents in the neonate has not been established. The half-lives of ZDV, 3TC, and nevirapine are prolonged during the neonatal period because of immature liver metabolism and renal function, requiring specific dosing adjustments when these agents are administered to neonates. Optimal dosages for protease inhibitors in the neonatal period are still under study. The infected woman should be counseled regarding the theoretical benefit of combination antiretroviral drugs for the neonate, potential risks, and available data on appropriate dosing. She should also be informed that using antiretroviral drugs in addition to ZDV for prophylaxis of newborns is of unknown efficacy in reducing risk of perinatal transmission.

3. HIV-1--Infected Women in Labor Who Have Had No Prior Therapy

Recommendation. Several effective regimens are available for intrapartum therapy for women who have had no prior therapy

• a single dose of nevirapine at onset of labor followed by a single dose of nevirapine for the newborn at age 48 hours;
• oral ZDV and 3TC during labor, followed by 1 week of oral ZDV-3TC for the newborn;
• intrapartum intravenous ZDV followed by 6 weeks of ZDV for the newborn; and
• the 2-dose nevirapine regimen combined with intrapartum intravenous ZDV and 6 weeks of ZDV for the--newborn.

In the immediate postpartum period, the woman should have appropriate assessments (e.g., CD4⁺ count and HIV-1 RNA copy number) to determine whether antiretroviral therapy is recommended for her own health.

Discussion. Although intrapartum antiretroviral medications will not prevent perinatal transmission that occurs--before labor, most transmission occurs near to or during labor and delivery. Preexposure prophylaxis for the fetus can be provided by giving the mother a drug that rapidly crosses the placenta to produce systemic antiretroviral drug levels in the fetus during intensive exposure to HIV-1 in maternal genital secretions and blood during birth.

Several intrapartum/neonatal antiretroviral prophylaxis regimens are applicable for women in labor who have had no prior antiretroviral therapy . Two regimens, one--using 2 doses of nevirapine (one each for the mother and--infant) and the other a combination of ZDV and 3TC, were shown to reduce perinatal transmission in randomized clinical trials among breastfeeding women, and available epidemiologic data suggest the efficacy of a third, ZDV-only regimen. The fourth regimen, combining ZDV with nevirapine, is based upon theoretical considerations.

In the HIVNET 012 trial, conducted in Uganda, a regimen consisting of a single dose of oral nevirapine given to the woman at onset of labor and a single dose to the infant at age 48 hours was compared with oral ZDV given to the woman every 3 hours during labor and postnatally to the infant for--7 days . At age 6 weeks, the rates of transmission were 12% (95% CI = 8%--16%) in the nevirapine arm versus 21% (95% CI = 16%--26%) in the ZDV arm, a 47% reduction (95% CI = 20%--64%) in transmission (65). No serious short-term toxicity was observed in either group. Because no placebo group was included, no conclusions can be drawn regarding the efficacy of the intrapartum/1-week neonatal ZDV regimen versus no treatment.

In the PETRA trial, conducted in Uganda, South Africa, and Tanzania, ZDV and 3TC were administered orally intrapartum and to the woman and infant for 7 days postnatally. Oral ZDV and 3TC were administered at the onset of labor and continued until delivery  Postnatally, the woman and infant received ZDV and 3TC every 12 hours for 7 days. At age 6 weeks, the rates of transmission were 9% in the ZDV-3TC arm versus 15% in the placebo arm, a 40% reduction in transmission (64). However, no differences in transmission were observed when oral ZDV and 3TC were administered only during the intrapartum period (transmission of 14% in the ZDV-3TC arm versus 15% in the placebo arm), indicating that some postexposure prophylaxis is needed, at least in breastfeeding settings.

These clinical trials were conducted in Africa, where the majority of women breastfeed their infants. Because HIV-1 can be transmitted by breast milk and the highest risk period for such transmission is the first few months of life (109), the absolute transmission rates observed in the African trials may not be comparable to what might be observed with these regimens in HIV-1--infected women in the United States, where breastfeeding is not recommended. However, comparison of the percentage of reduction in transmission at early timepoints (e.g., 4--6 weeks) may be applicable. In the effective arms of the PETRA trial, antiretroviral drugs were administered postnatally to both the mother and the infant to reduce the risk of early transmission through breast milk. In the United States, administration of ZDV-3TC to the mother postnatally in addition to the infant would not be required for prophylaxis against transmission because HIV-1--infected women are--advised not to breastfeed their infants (although ZDV-3TC might be indicated as part of a combination postnatal treatment regimen for the woman).

Epidemiologic data from New York State indicate that--intravenous maternal intrapartum ZDV followed by oral ZDV for 6 weeks to the infant may significantly reduce trans-mission compared with no treatment . Transmission rates were 10% (95% CI = 3%--22%) with intrapartum and neonatal ZDV compared with 27% (95% CI = 21%--33%) without ZDV, a 62% reduction in risk (95% CI = 19%--82%) (70,71). Similarly, in an epidemiologic study in North Carolina, intravenous intrapartum and 6-week oral neonatal ZDV treatment was associated with a transmission rate of 11%, compared with 31% without therapy (6). However, intrapartum ZDV combined with very short-term ZDV administration to infants postnatally, e.g., the 1-week postnatal infant ZDV course in HIVNET 012 (65), has not proved effective to date. This underscores the necessity of recommending a full 6-week course of infant treatment when ZDV alone is used.

No data are available to address the relative efficacy of these three intrapartum/neonatal antiretroviral regimens for prevention of transmission. In the absence of data to suggest the superiority of one or more of the possible regimens, choice should be based upon the specific circumstances of each woman. The 2-dose nevirapine regimen offers the--advantage of lower cost, the possibility of directly observed therapy and increased adherence compared with the other two regimens. In a clinical trial (SAINT) in South Africa, which compared the 2-dose nevirapine and the intrapartum/ postpartum ZDV-3TC regimens, no significant differences were observed between the two regimens in terms of efficacy in reducing transmission or in maternal and infant toxicity (66).

It has not been determined if combining intravenous intrapartum/6-week neonatal oral ZDV with the 2-dose nevirapine regimen will provide additional benefit over that observed with each regimen alone. Clinical trial data have established that combination therapy is superior to single-drug therapy for treatment of persons with established infection and that--infants born to women in labor who have not received any antiretroviral therapy are at high risk for infection. The--2-dose nevirapine regimen had no serious short-term drug-associated toxicity in the 313 mother--infant pairs exposed to the regimen in the HIVNET 012 trial. Nevirapine and ZDV are synergistic in inhibiting HIV-1 replication in vitro (110), and both nevirapine and ZDV rapidly cross the placenta to achieve drug levels in the infant nearly equal to those in the mother. In contrast to ZDV, nevirapine can decrease plasma HIV-1 RNA concentration by at least 1.3 log by 7 days after a single dose (111) and is active immediately against intracellular and extracellular virus (112). However, nevirapine resistance can be induced by a single mutation at codon 181, whereas high-level resistance to ZDV requires several mutations. Nevirapine resistance mutations were detected at 6 weeks postpartum in 19% of antiretroviral naive women and 15% of women receiving antiretroviral drugs during pregnancy who received single-dose nevirapine during labor (See Antiretroviral Drug Resistance and Resistance Testing in--Pregnancy).

A theoretical benefit of combining the intrapartum/neonatal ZDV and nevirapine regimens would be the efficacy of this combination if the woman had acquired infection with HIV-1 that is resistant to either ZDV or nevirapine. Perinatal transmission of antiretroviral drug-resistant virus has been reported but appears to be unusual (6,113,114). Virus with low-level ZDV resistance may be less likely to establish infection than wild-type virus, and transmission may not occur even when maternal virus has high-level ZDV resistance (114--117). Since the prevalence of drug-resistant virus is an evolving phenomenon, surveillance is needed to determine this prevalence in pregnant women over time and the risk of transmission of resistant viral strains. The potential benefits of combination prophylaxis with intrapartum/neonatal nevirapine and ZDV must be weighed against the increased cost, possible problems with nonadherence, potential short- and long-term toxicity, including the risk of emergence of nevirapine-resistant virus, and the lack of definitive data to show that combining the two intrapartum/postpartum regimens offers any additional benefit for prevention of transmission over the use of either drug alone.

Recommendation. The 6-week neonatal component of the ZDV chemoprophylactic regimen should be discussed with the mother and offered for the newborn. ZDV should be--initiated as soon as possible after delivery, preferably within 6--12 hours of birth. Some clinicians may use ZDV in combination with other antiretroviral drugs, particularly if the mother is known or suspected to have ZDV-resistant virus. However, the efficacy of this approach for prevention of transmission is unknown, and appropriate dosing regimens for neonates are incompletely defined. In the immediate postpartum period, the woman should undergo appropriate assessments (e.g., CD4⁺ count and HIV-1 RNA copy number) to determine if antiretroviral therapy is required for her own health. The--infant should undergo early diagnostic testing so that if he or she is HIV-1 infected, treatment can be initiated as soon as--possible.

Discussion. Definitive data are not available to address whether ZDV administered only during the neonatal period would reduce the risk of perinatal transmission. Epidemiologic data from a New York State study indicate a decline in transmission when infants were given ZDV for the first 6 weeks of life compared with no prophylaxis (70,71). Transmission rates were 9% (95% CI = 4.1%--17.5%) with ZDV prophylaxis of newborns only (initiated within 48 hours after birth) versus 18% (95% CI = 7.7%--34.3%) with prophylaxis initiated after 48 hours, and 27% (95% CI = 21%--33%) with--no ZDV prophylaxis (70). Epidemiologic data from North Carolina did not demonstrate a benefit of ZDV for newborns only compared with no prophylaxis (6). Transmission rates were 27% (95% CI = 8%--55%) with prophylaxis of newborns only and 31% (95% CI = 24%--39%) with no prophylaxis. The timing of initiation of infant prophylaxis was not defined in this study. Data from a case-control study of postexposure prophylaxis of health-care workers who had nosocomial percutaneous exposure to blood from HIV-1--infected persons indicate that ZDV administration was associated with a 79% reduction in the risk for HIV-1 seroconversion following exposure  Postexposure prophylaxis also has prevented retroviral infection in some studies involving animals (118--120).

The interval during which benefit can be gained from postexposure prophylaxis is undefined. When prophylaxis was delayed beyond 48 hours after birth in the New York State study, no efficacy could be demonstrated. For most infants in this study, prophylaxis was initiated within 24 hours (71). Data from studies of animals indicate that the longer the delay in institution of prophylaxis, the less likely that infection will be prevented. In most studies of animals, antiretroviral prophylaxis initiated 24--36 hours after exposure has usually not been effective for preventing infection, although later administration has been associated with decreased viremia (118--120). In cats, ZDV treatment initiated within the first 4 days after challenge with feline leukemia virus afforded protection, whereas treatment initiated 1 week postexposure did not (121). The relevance of these animal studies to prevention of perinatal HIV-1 transmission in humans is unknown. HIV-1 infection is established in most infected infants by age 1--2 weeks. In a study of 271 infected infants, HIV-1 DNA polymerase chain reaction (PCR) was positive in 38% of samples from infants tested within 48 hours of birth. No substantial change in diagnostic sensitivity was observed within the first week of life, but detection increased rapidly during the second week of life, reaching 93% by age 14 days (122). Initiation of postexposure prophylaxis after age 2 days is not likely to be efficacious in preventing transmission, and by age 14 days, infection would already be established in most infants.

When the mother has received neither the antenatal nor intrapartum parts of the three-part ZDV regimen, administration of antiretroviral drugs to the newborn provides chemoprophylaxis only after HIV-1 exposure has already occurred. Some clinicians view this situation as analogous to nosocomial postexposure prophylaxis and may wish to provide ZDV in combination with one or more other antiretroviral agents. Such a decision must be accompanied by a discussion with the woman of the potential benefits and risks of this approach and the lack of data to address its efficacy and safety.

The development of antiretroviral drug resistance is one of the major factors leading to therapy failure in HIV-1--infected persons. Resistant viral variants emerge under selective pressure, especially with incompletely suppressive regimens,--because of the inherent mutation-prone process of reverse transcription with viral replication. The administration of combination antiretroviral therapy with maximal suppression of viral replication to undetectable levels limits the development of antiretroviral resistance in both pregnant and nonpregnant persons. Some have raised concern that using non-highly--active antiretroviral regimens, such as ZDV monotherapy, for prophylaxis against perinatal