Our ability to prevent perinatal transmission of HIV infection has prompted legislation requiring healthcare providers to encourage all pregnant women to learn their HIV status -- and to offer counseling and treatment all those who test positive for the virus. The purpose of this treatment is two-fold: to decrease the likelihood of mother-to-infant transmission, and to provide optimal antiretroviral therapy to seropositive women, who are all too frequently underdiagnosed and undertreated.
We have known since 1994, when the landmark study ACTG 076 was published, that the administration of zidovudine (ZDV, AZT, Retrovir®) to pregnant women reduces HIV transmission rates by 67.5% (see "AZT Diminishes Transmission of HIV-1 from Mother to Their Infants," Vol. 1, No. 2, pages 31-35). Since that time antiretroviral therapy has evolved dramatically, and ZDV monotherapy -- which was once the standard of care for all seropositive patients -- has been supplanted by multidrug regimens that generally include two nucleoside analogs and a protease inhibitor or a non-nucleoside reverse-transcriptase inhibitor.
The accumulated data from numerous clinical trials clearly demonstrate that combination therapy in adults reduces the frequency of hospitalizations and opportunistic infections and confers a significant survival benefit (see "Update: The Protease Inhibitors," Vol. 2, No. 3, pages 52-61). And while ZDV monotherapy remains the standard of care for the prevention of vertical transmission of HIV, it can no longer be regarded as optimal therapy for the treatment of pregnant seropositive women.
Unfortunately, we do not yet have efficacy data on antiretroviral therapies other than ZDV monotherapy in this population. We do know that women who are not pregnant respond well to combination therapy,(7, 8) so the real question is: Will the use of more potent antiretroviral therapies in pregnant women have deleterious effects on their unborn children? Studies now being conducted will eventually provide us with answers to that crucial question. In the interim, clinicians should bear in mind that no one knew for sure, at the time ACTG 076 was enrolling patients, what the effect of ZDV monotherapy would be on unborn children. It took a full analysis of the follow-up data to establish that, except for a transient lowering of hemoglobin in infants born to treated mothers, no significant adverse events were attributable to ZDV therapy in the first three years of life.
In the absence of definitive data on the safety and efficacy of various combination therapies in pregnant women and their fetuses, clinicians will find themselves making decisions about antiretroviral treatment based on personal experience, anecdotal evidence, data from Phase II/III clinical trials, the preliminary findings of large-scale studies, and other factors. This is simply a fact of life in the field of HIV treatment, where therapeutic advances have consistently outstripped the ability of researchers to test these new agents in large, well-designed, randomized, double-blind clinical trials.
Such clinical trials have established the superiority of multidrug antiretroviral regimens over monotherapy.(5-7) As a result, decisions about therapy for pregnant women should no longer be based solely on the need to prevent perinatal transmission of HIV but should also recognize the therapeutic advantages of these newer therapies in pregnant women.(8) We now know that the virologic and clinical benefits of ZDV monotherapy are modest when compared to the effects of combination therapy, and monotherapy with any antiretroviral agent can actually hasten the emergence of drug-resistant viral strains -- not only in the mother but in her infant as well.(9, 10) It may therefore be inappropriate to withhold the newer combination regimens from women simply because they are pregnant.
Since 1994, when ACTG 076 was published, the U.S. Public Health Service has recommended HIV testing and counseling for all pregnant women,(11) and the prophylactic use of ZDV to reduce the rate of perinatal transmission of the virus (Table 1). Drug registry information shows that there has been no increase in birth defects in infants born to mothers taking ZDV, but a handful of animal studies indicate an increase in embryo resorption.(12) In addition, one recent report suggests that there may be an increase in cancer rates in the offspring of pregnant mice treated with high-dose ZDV during pregnancy.(13)
100 mg ZDV by mouth 5 times a day
Intravenous ZDV at a dose of 2 mg/kg, infused over one hour, then 1 mg/kg by continuous infusion until delivery.
Oral ZDV syrup at a dose of 2 mg/kg every 6 hours for 6 weeks, beginning 8-12 hours after birth. ZDV may be given IV if neonate is unable to take oral medication.
While these findings are of interest, it is difficult to know how to extrapolate the data to human beings -- none of whom are receiving comparably high doses of ZDV. At present, the manifest benefit of therapy -- which reduces the risk of viral transmission by two thirds -- outweighs any theoretical risk to the fetus. The Public Health Service therefore continues to recommend that ZDV therapy be initiated in all HIV-positive pregnant women at the end of the first trimester of gestation (i.e. 14 weeks) at a dose of 100 mg five times a day (or 200 mg t.i.d.).
It is important to recognize that the benefits of ZDV prophylaxis and therapy may not be as great in practice as they have proven to be in clinical trials, where compliance is generally higher. Indeed, a recently completed study in eastern Virginia found that transmission of HIV was reduced when the women in this study were given ZDV, but the reduction was not as great as that seen in ACTG 076.(14) On the other hand, a comparable study conducted in North Carolina found that ZDV prophylaxis reduced rates of maternal-to-infant transmission to levels as low or lower than those attained in ACTG 076.(15) In the end, of course, reduction is reduction -- and no clinician should counsel a pregnant patient against ZDV therapy simply because it may not result in a 70% reduction in the likelihood of transmission.
There are three components to the ZDV prophylaxis protocol: antenatal, intrapartum, and neonatal ZDV therapy (Table 1). We do not yet know which of these components provides the greatest benefit in preventing the perinatal transmission of HIV. Therefore, ZDV therapy is recommended by the Public Health Service for all pregnant women, including those whose HIV status is unknown until late in their pregnancy and those who receive no prenatal care -- on the grounds that some benefit may be derived from even a partial course of ZDV therapy (see the PULL OUT AND SAVE feature in this issue, "U.S. Public Health Service Recommendations for the Use of Antiretroviral Drugs During Pregnancy").
For this same reason, seropositive women who present in labor with no prior ZDV therapy should be offered intravenous intrapartum ZDV and should be encouraged to permit their neonates to receive ZDV therapy. As Dr. Catherine M. Wilfert points out in her editorial in this issue, the evidence we do have leads to the hypothesis that "it is the presence of ZDV in the neonate's blood and tissues around the time of delivery, when infection is most likely to occur, that accounts for the drug's high rate of success in interrupting transmission of HIV." Or, to put it another way: in the vertical transmission of HIV, timing may be everything.(16)
If a pregnant patient has been exposed to antiretroviral agents, viral resistance may have emerged, especially if that exposure has lasted for six months or more -- and this may decrease the benefits of ZDV prophylaxis. There have been reports of ZDV-resistant HIV in newborns,(9) so this concern is more than theoretical.
For women with CD4 counts below 200 cells/mm3, ZDV should be offered for maternal as well as fetal benefit. At present we have little data on the benefits of using other antiretroviral agents, alone or in combination, during pregnancy. By the same token, we have no reports as yet that any of these agents pose particular dangers during pregnancy, especially if they are used in the same way that ZDV is used.
If the clinician does decide to recommend combination therapy for a pregnant patient, based on the efficacy of such regimens for the treatment of maternal disease, he should include ZDV as one component of that regimen, based on the established efficacy of this agent in interrupting perinatal transmission of HIV. In time, studies may establish that other drugs are as effective as ZDV -- or even superior to it -- in preventing infection of infants born to HIV-positive mothers, but for now ZDV remains the mainstay of prophylactic therapy. Pharmacologic studies indicate that high levels of ZDV do reach the fetal compartment during gestation,(17) which may account for its efficacy in interrupting transmission.
While it is clear from the available data that antiretroviral therapies that include a protease inhibitor delay disease progression and thereby prolong life,(18) no information is yet available regarding the safety or efficacy of these agents during pregnancy. Nor do we know if these agents reduce the risk of perinatal transmission of HIV. Nelfinavir, the newest of the protease inhibitors, is also the first of these agents to be made available in pediatric formulation and to receive F.D.A. approval for use in this population (see "At last, a protease inhibitor for children as well as adults," Vol. 3, No. 1, pages 18-19). The F.D.A. has since approved a pediatric formulation of ritonavir for use in infants and children (see "Antiretroviral Therapy in Children" in this issue).
The use of protease inhibitors during pregnancy is a matter of controversy and contention. At a major national conference that was convened earlier this year to consider this clinical question, no consensus was reached regarding whether pregnant seropositive women should be encouraged or discouraged from taking protease inhibitors. Further research will be necessary to establish the safety and efficacy of this class of agents in pregnant women. In the interim, clinicians should be guided by the U.S. Public Health Service recommendations found elsewhere in this issue.
The development of powerfully suppressive antiretroviral regimens has been enhanced by use of HIV RNA assays to measure therapeutic efficacy. The data suggest that patients with low viral loads (<2,500 particles/mL) are at lower risk of disease progression than those with high (2,500-20,000 particles/mL) or very high (>20,000 particles/mL) viral burdens.(19) This information has been of use to clinicians in developing treatment regimens for asymptomatic seropositive adults.
In pregnant women, there appears to be a strong correlation between increasing viral load and risk of vertical transmission, but this relationship is not absolute. For one thing, there seems to be no threshold below which transmission does not occur; women with no detectable viral burden have given birth to seropositive infants. And conversely, mothers with very high viral loads have given birth to seronegative offspring.(20) As a result, viral load assays are of limited utility in predicting which pregnant women are at greatest risk of transmitting HIV infection.
Treatment of pregnant seropositive women should therefore include ZDV for the prevention of perinatal transmission regardless of viral load. HIV RNA assays can be used to guide therapy in individual patients, however, and they will be of particular usefulness in deciding when to initiate or change therapy in women with less severe immunosuppression (CD4 counts >200 cells/mm3), in order to tailor therapy for maximum maternal benefit.
Any therapeutic regimen developed for a pregnant seropositive woman must take into account not only the presence of HIV but the patient's CD4 count and HIV RNA level. For patients with a CD4 count above 500 cells/mm3, clinicians have a number of therapeutic options, one of which is to initiate only the three-part ZDV prophylaxis regimen shown in Table 1. However, these women should still be followed carefully throughout pregnancy, and any significant drop in CD4 count or significant rise in viral load, if confirmed by a repeat test, should occasion a change in therapy.
For women with CD4 counts between 200 and 500 cells/mm3, lamivudine (3TC, Epivir®) should be added to this regimen -- especially of the patient's viral burden exceeds 20,000 particles/mL, and consideration should be given to three-drug therapy. While less is known about the safety and efficacy of other agents in pregnant women, the current standard of care for these patients embraces combination therapy when the CD4 count is low and/or the viral burden is high (Table 2).
Table 2: Recommendations for Multidrug Antiretroviral Therapy in HIV-Infected Pregnant Women
1 or 2
2 or 3
Note: These are conservative recommendations, and clinicians, in consultation with their patients, may wish to initiate therapy earlier and/or may choose to begin combination therapy earlier. In any case, all pregnant seropositive women should be offered the three-part ZDV prophylaxis protocol, and patients should be strongly encouraged to begin, stop, and resume all their drugs at the same time.
Women with CD4 counts below 200 cells/mm3 should receive combination therapy with three agents, one of them a protease inhibitor, irrespective of viral titers. Whether protease inhibitors used during pregnancy remains controversial, which is why the clinician should involve the patient in this decision process. All patients in this category also require prophylaxis against PCP and other opportunistic infections.
In cases of severe immunosuppression or acute opportunistic infection during pregnancy, maternal health must be the clinician's primary concern. Pregnant women with low CD4 counts and/or symptomatic HIV disease should therefore be considered candidates for combination antiretroviral therapy. So should women with CD4 counts above 200 cells/mm3 and asymptomatic disease, if these patients have viral loads in excess of 20,000 particles/mL, because we now know that individuals with high viral burdens need to begin antiretroviral therapy much earlier than previously thought.
All of these women are candidates for combination therapy, and those with active disease or a diagnosis of AIDS should receive a protease inhibitor in addition to ZDV and whatever other nucleoside analog or non-nucleoside reverse transcriptase inhibitor they have been assigned. Here both clinician and patient are moving through uncharted waters, without hard data from large-scale clinical trials to guide their decision-making. This makes it all the more important that the physician and patient reach decisions in concert, since the outcome cannot be predicted with any real assurance.
There is now considerable optimism for families affected by, and infected by, HIV. The widespread adoption of combination therapy has stabilized a remarkable number of infected individuals, and that, in turn, has shifted the clinical approach to HIV disease away from acute care and toward chronic management. Still, many questions remain about the optimal care of HIV-infected pregnant women -- and few of those questions will be answered until data from studies currently under way are made available to practitioners.
In particular, we do not yet know if combination antiretroviral therapies, especially those that contain a protease inhibitor, will further reduce perinatal transmission of HIV beyond the dramatic reductions achieved with ZDV alone. And we do not know if the use of agents other than ZDV, alone or in combination, will pose threats to the unborn child or the pregnant woman.
In the absence of any indication that combination therapy does pose a threat to the woman or her fetus, clinicians should be guided by the just-released recommendations of the U.S. Public Health Service Task Force on the use of antiretroviral agents in pregnancy (see the PULL OUT AND SAVE feature in this issue). The members of that task force have declared that "therapies of known benefit to HIV-positive women should not be withheld during pregnancy unless there are known adverse effects on the mother, the fetus, or the infant -- and these adverse effects outweigh any potential benefit to the woman."
1. Rosenberg PS. Scope of the AIDS epidemic in the United States. Science 1995; 270: 1372-5.
2. Update: AIDS among women, United States 1994. MMWR 1994; 43: 155-60.
3. Centers for Disease Control and Prevention. U.S. Public Health Service recommendations for human immunodeficiency virus counseling and voluntary testing for pregnant women. MMWR 1995; 44 (RR-7): 1-15.
4. Connor EM, Sperling RS, Gelber R, Kiseler P, Scott G, O'Sullivan MJ, et al. Reduction of maternal-infant transmission of human immunodeficiency virus type 1 with zidovudine treatment. N Engl J Med 1994; 331: 1173-80.
5. Volberding PA. The case for combination therapy. HIV Newsline 1995; 1: 19.
6. Stephenson J. Other AIDS drug regimens beat AZT alone to reduce clinical progression and mortality. JAMA 1995; 274: 1183-4.
7. Choo W. Combination superior to zidovudine in the Delta trial. Lancet 1995; 346: 895.
8. Human Immunodeficiency Virus Infection in Pregnancy. Am Coll Obstet Gynecol; January 1997, No. 232.
9. Frankel LM, Wagner LE, Demeter LM, et al. Effects of zidovudine use during pregnancy on resistance and vertical transmission of human immunodeficiency virus type 1. Clin Infec Dis 1995; 20: 1321-6.
10. Larder BA, Darby G, Richman DD. HIV with reduced sensitivity to zidovudine (AZT) isolator during prolonged therapy. Science 1989; 243: 1731-4.
11. Recommendation of the U.S. Public Health Service Task Force on the use of zidovudine to reduce perinatal transmission of human immunodeficiency virus. MMWR 1994; 43 (RR-11): 1.
12. Toltzis P, Morton T, Magnuson T. Effect of zidovudine on preimplantation murine embryos. Antimicrob Agents Chemother 1993; 37: 1610-13.
13. Katz D, et al. NCI ongoing study press release. January 1997.
14. Mitchell DK, Treichler DB, Dattel BJ, Rubio TT. Maternal-fetal transmission of HIV in southeastern Virginia: Impact of perinatal zidovudine (ZDV) use. Soc Ped Res; Washington, D.C., 1996.
15. Fiscus SA, Ademora AA, Schoenbach VJ, Lim W, et al. Perinatal HIV infection and the effect of zidovudine therapy on transmission in rural and urban counties. JAMA 1996; 275: 1483-8.
16. Pizzo PA, Butler KM. In the vertical transmission of HIV, timing may be everything. N Engl J Med 1991; 325: 652-4.
17. O'Sullivan MJ, Boyer PJ, Scott GB, Parks WP, Weller S, Blum MR, et al. The pharmacokinetics and safety of zidovudine in the third trimester of pregnancy for women infected with human immunodeficiency virus and their infants: Phase I ACTG Study (Protocol 082). Am J Obstet Gynecol 1993; 168: 1510-16.
18. Carpenter CJ, Fischl MA, Hammer SM, et al. for the International AIDS Society U.S.A. Antiretroviral therapy for HIV infection in 1996: Recommendations of an international panel. JAMA 1996; 276: 146-54.
19. O'Brien WA, Fisher AE, Holodniy M, Johnson V, Kubota MK. Patient monitoring and markers of disease progression. In: Comprehensive Management of HIV Disease: HIV Speakers Forum. 6th Annual Update Meeting, Part I; Amelia Island, Florida, 1995.
20. Sperling RS, et al. Maternal viral load, ZDV treatment, and risk of transmission of HIV type I from mother to infant. N Engl J Med 1996; 335: 1621-8.
Bonnie J. Dattel, M.D., Professor of Obstetrics and Gynecology and Associate Director, Division of Maternal-Fetal Medicine, Eastern Virginia Medical School, Norfolk, VA.