As Dr. Catherine M. Wilfert observes in her introduction to this special issue of AIDS Care, the availability of potent combination antiretroviral therapy has not only increased the life-spans of people living with HIV, it has considerably enhanced the quality of their lives. These improvements have allowed HIV-positive individuals to contemplate the future with greater optimism and confidence. As a result, people living with HIV are now daring to make long-term plans. With increasing frequency, these plans include the possibility of conceiving, bearing, and raising children. Significant numbers of women with HIV -- and women whose partners are seropositive -- have begun to weigh the opportunities and risks of childbearing, and more and more of them are asking their primary-care providers to advise them about this all-important decision.
Can I have children? What an HIV-positive woman means when she asks this question is: What is the likelihood that I can carry a child to term and deliver it safely -- without infecting it with the human immunodeficiency virus? There are other questions, of course. Does pregnancy impose particular risks for HIV-positive women? Will I live long enough to see this child grow up? Will I be able to ensure that my child is adequately provided for, if something should happen to me? But all of these questions are of secondary importance. The paramount question is: How likely am I to infect my child?
There are no easy answers to any of these questions, but we now know a good deal more than we did even a few years ago about mother-to-infant transmission rates -- and we also know a good deal more about how to interrupt transmission. In the United States, the likelihood that an HIV-positive mother will give birth to an infected child ranges from less than 5% to as much as 30%. Several factors contribute to this broad range in transmission risk: whether the mother is in the early, middle, or late stage of HIV disease; whether the mother is taking antiretroviral medications; and how effective those medications are in suppressing viral replication in the mother's body. The mode of delivery -- vaginal versus cesarean section -- is also a factor, but we know less about its role in mother-to-infant transmission than we do about the other three.
A number of recent studies have concluded that the mother's viral burden correlates directly with the likelihood that HIV will be transmitted to her fetus. Mothers who have high viral loads are more likely to transmit the virus in utero or during labor and delivery. This is clearly not the only factor affecting mother-to-infant transmission, however, because we now know that women with a wide range of viral loads -- some exceedingly high, some exceptionally low -- have delivered healthy, uninfected children. Unfortunately, the converse is also true: women with undetectably low viral loads have given birth to HIV-infected children.
What these findings tell us is that viral load is a good -- but by no means a perfect -- predictor of how likely a seropositive woman is to transmit HIV to her fetus. Other factors -- among them the mother's CD4 count and disease stage -- also play roles in this process. The data suggest that low maternal CD4 count and advanced stage of disease appear to increase the risk of transmission. What this means is that a woman with a relatively high CD4 count, an undetectable viral load, and asymptomatic disease is much less likely to pass the virus to her newborn than a woman with a low CD4 count, a high viral load, and a diagnosis of AIDS.
The trouble is that there are no absolute guarantees that an HIV-positive woman -- even one with a near-normal CD4 count and no detectable virus in her body -- will deliver an HIV-negative child. Even in the best of circumstances, some risk remains. However, we have known since 1994, when the results of the AIDS Clinical Trials Group's study 076 were published, that antiretroviral therapy markedly reduces the risk of mother-to-infant HIV transmission. This landmark study demonstrated a decrease in transmission from 28% to 8% with use of a three-step regimen: oral AZT during the second and third trimesters of pregnancy, intravenous AZT during labor and delivery, and oral AZT for the infant during the first six weeks of life.
The size of this decrease in transmission -- a two-thirds reduction in infection rates in treated patients -- came as something of a surprise to the researchers who conducted the study, because they knew that AZT monotherapy results in only modest decreases in maternal viral load. The fact that this single drug proved so effective in interrupting mother-to-infant transmission despite its limited effectiveness in suppressing viral replication suggests that it works in ways we do not yet fully understand.
Now that we have potent combination antiretroviral therapies at our disposal, it is reasonable to think that we can reduce rates of maternal transmission still further -- although there are, as yet, no data from large clinical trials to confirm the superiority of any of these regimens over the AZT-based regimen used in ACTG 076. The current recommendation is that all pregnant women should be treated with combination antiretroviral therapy appropriate for their level of disease (see "Optimal Antiretroviral Therapy for HIV Infection: Updated recommendations of the International AIDS Society - U.S.A.," the eight-page Pull Out and Save section of the December 1997 issue of HIV Newsline, and "HIV Disease in Women and Children," the special issue of HIV Newsline devoted to the prevention and treatment of pregnant women and their offspring that was published in August 1997).
One additional factor that may bear on mother-to-infant transmission of HIV is the method of delivery. We now know that more than 60% of these infections occur during delivery, possibly as a result of fetal exposure to maternal blood and secretions during passage through the birth canal. Several studies have suggested that delivery by cesarean section, rather than vaginal delivery, may lessen the likelihood of mother-to-infant transmission. This issue requires more investigation, however, and routine C-section has not yet been recommended for all HIV-positive mothers. (For more on this subject, read Dr. Wilfert's comments on the advisability of C-sections for pregnant, HIV-positive women. You will find those remarks in this issue, under the rubric What This Means for You.)
Finally, researchers have identified several factors that are thought to increase the risk of mother-to-infant transmission of HIV. These factors include premature rupture of the placental membranes, slow progression of labor, use of invasive monitoring devices during delivery, and breast-feeding. In the United States, HIV-positive mothers are advised against breast-feeding their newborns for this reason.
It is also important to minimize the chances that a HIV-positive individual's sexual partner will become infected as an indirect result of the couple's desire to conceive a child. In cases where the female partner is infected but the male partner is not, fertilization can be accomplished safely though the use of artificial-insemination techniques. Unfortunately, insemination is expensive and it is rarely covered by insurance companies, which limits its accessibility.
In cases where a male partner is infected but the female partner is not, the risk of transmission to the woman can be substantial -- but there may be hope for seropositive would-be fathers. Dr. Simon Marina and his colleagues recently reported that they have developed a technique for "washing" semen specimens to extract the HIV virus (see "Positive New Fathers: Refinement in insemination process may allow HIV-positive men to safely father children" in the Newsline section of this issue of AIDS Care). In this study, which was conducted in Barcelona, Spain, 107 semen samples were collected from the 63 HIV-positive participants. After the sperm samples were washed, all of them were tested for the presence of HIV. The six samples that tested positive were discarded; the remaining 101 samples were used to inseminate the HIV-negative partners of these men.
In all, 31 pregnancies resulted from this insemination process, and because several of the women conceived twins, a total of 37 healthy, uninfected children were ultimately delivered -- and none of the women in the study seroconverted. Although this study is too small to tell us if this method of purging HIV from semen is in fact effective in reducing transmission rates -- which are very small to begin with for what is, in effect, a single act of vaginal intercourse -- these results are certainly encouraging. At the very least, the Spanish data suggest that "washing" semen specimens in this way may reduce the risk that an HIV-positive man will infect his HIV-negative partner when his semen is used to impregnate her through intrauterine insemination.
It is reasonable to anticipate that the HIV transmission rate will be 5% or less among seropositive pregnant women who are receiving AZT monotherapy -- and this modest risk may eventually be eliminated altogether in women who respond well to combination therapy, whether they elect to deliver by C-section or not. It is also reasonable to anticipate that women with HIV who experience a sharp drop in viral load and a significant rebound in immune function as a result of potent combination therapy may live symptom-free for a decade or longer.
Under these altered circumstances, it is no longer unreasonable for HIV-positive women to entertain the prospect of having -- and rearing -- healthy, HIV-negative children. What every prospective parent must recognize is that although the risk of HIV transmission can now be minimized, it cannot be completely eliminated by any of the techniques and therapies currently available to us. This means that any HIV-positive individual who is contemplating parenthood must carefully consider this sobering fact before embarking on the extraordinary journey that is parenthood.
Kimberly Y. Smith, M.D., M.P.H., is from Rush-Presbyterian-St. Luke's Medical Center, Chicago, IL.