TOPICS > Health

Extended Interview: Dr. Mary Glenn Fowler

November 6, 2003 at 12:00 AM EDT


SUSAN DENTZER: The decline in maternal to child transmission in the U.S. is really a success story in the epidemic in the United States … What has happened over the last decade?

DR. MARY GLENN FOWLER: It’s been one of our really strong successes in the HIV epidemic. In the early ’90s we had approximately between 1,000 and 2,000 women delivering infected babies each year, and we have really reduced that down to close to 280 to 370 babies being infected.

So it’s almost greater than an 80 percent decline, and it’s related to use of anti-retrovirals during pregnancy as well as to obstetrical intervention such as scheduled C-section prior to the onset of labor.

SUSAN DENTZER: And just to draw a comparison between the transmission rates that we’re experiencing here in the United States versus what is seen in some of the hardest hit countries around the world, what’s the difference?

DR. MARY GLENN FOWLER: Again, it really highlights the impact of our new interventions that we have. But in the U.S., we can tell a mother that her risk of transmission is reduced down to about 1 to 2 percent. In most of the world, without the new interventions with anti-retrovirals that we have available, transmission rates are anywhere between 25 to 40 percent.

So in the U.S. we really are very fortunate because with these interventions, we can really tell a mother that she has a very, very low risk of transmission.

SUSAN DENTZER: You started to talk about this a moment ago, but let’s go back to the reasons that we’ve experienced this large decline in the United States.

DR. MARY GLENN FOWLER: One of the first things that happened was a major trial in the United States that used AZT, which was one of our first drugs for HIV, and that study showed that we could reduce transmission from mother to child just with AZT, if it was given prenatally at labor and delivery, and then to the newborn for six weeks, you could reduce transmission by about 70 percent.

Since that time, we’ve really built on this knowledge, and we’ve added other very potent anti-retrovirals so that usually most women during pregnancy will actually receive three or sometimes four anti-retrovirals during the pregnancy, and that plus using a scheduled C-section at delivery, if needed, can reduce transmission to the 1 to 2 percent that we see.

SUSAN DENTZER: You said a C-section if needed. Under what circumstances?

DR. MARY GLENN FOWLER: What obstetricians really try to do is look very carefully at viral load at delivery, right around the time of labor and delivery. We know that the mother’s virus, how much she carries around in her body right around the time the baby is about to be born is probably the most important risk factor. And if the mother’s viral load is very low, or what we call nondetectable, then potentially there is less need to have a C-section.

If the mother had a high viral load around labor or delivery, then very often the obstetrician would recommend to the mother that she have a scheduled C-section before labor.

SUSAN DENTZER: What do we know about why transmission occurs and when transmission occurs from a pregnant mother to her infant, and what still remains unknown as a consequence of the science not having advanced enough?

DR. MARY GLENN FOWLER: We know a lot about when transmission actually occurs. Most transmission, based on a number of different studies in the U.S., Africa and Europe, would indicate that around labor and delivery probably about two-thirds of transmission occurs.

In addition, in settings where there’s breast feeding, there’s probably another 12 to 14 percent of absolute transmission that can occur after delivery, during probably over a two-year period of breast feeding, which is common in many settings. And before labor and delivery, probably only about 20 percent overall of all transmission would occur.

SUSAN DENTZER: What do we think explains that, given that a baby does develop a separate blood supply from the mother? What’s happening, at least as far as we suspect?

DR. MARY GLENN FOWLER: What I think is going on is that even before we had anti-retrovirals to protect the baby, the placenta is actually a very good protection, sort of a natural protection, and it’s around labor and delivery as the baby is going through the birth canal that there’s much more exposure to maternal blood.

So that’s probably one of the main things. The other key thing, as I mentioned, though, is the amount of virus that the mother is carrying in her blood is very, very important. So a woman with a high viral load is much more at risk for transmission than a woman who has a non-detectable viral load.

SUSAN DENTZER: One issue obviously is making sure that more women are tested [for HIV]. What is the state of play with respect to how broadly now women are being offered HIV testing in pregnancy?

DR. MARY GLENN FOWLER: It’s one of our areas that we’ve seen since 1995, when we first began recommending that all women during pregnancy be offered HIV testing, that there has been a lot of progress. But it really does vary by location, by center, and we think there’s still more work that needs to be done.

In some settings, about 50 percent of women will actually end up getting testing, and in other settings, it’s greater than 90 percent. So we really need to strive to make sure that all settings can get a universal offering of HIV testing to pregnant women in the prenatal setting ideally, but if not then, at labor and delivery for those women whose status is still not known.

SUSAN DENTZER: And when health providers contend that women may not be at risk for HIV and therefore don’t necessarily warrant being tested, what’s the response to that?

DR. MARY GLENN FOWLER: The fact that a woman is pregnant probably does mean that she’s had unprotected sex, and that in itself is a risk factor for HIV. So we would say that it’s better not to try to make a decision ahead of time by the health care provider whether the woman is at risk, and just to routinely offer HIV testing as part of the standard battery, a prenatal test, just as we do for syphilis. Just go ahead and offer that test to all women. And we’re trying to really implement that as part of what we call an opt out strategy where HIV testing is part of the prenatal battery of tests.

The woman is told that, and she’s also told she can decline it if she wishes to, and we believe that that approach based on data that we have both from the U.S. as well as other settings such as Canada, that that approach of routine testing, unless the woman declines, will lead to extremely high rates of prenatal HIV testing.

SUSAN DENTZER: So in terms of lowering the maternal child transmission rates further, is testing women the No. 1 thing that should be done, or are there other steps that should be taken as well?

DR. MARY GLENN FOWLER: There clearly are a number of things, but as we look in the United States today, one of the main factors and probably about 50 percent of the cases where we see HIV infection in this year and just the prior years, it’s probably related to the woman not being offered HIV testing, or that somehow in the system the test results were lost. And in that case, that’s probably one of the most important things that we can do, and CDC is really pushing the routine voluntary testing in the prenatal period.

The other factors that we clearly have to look at are for those women who did appear to at least have gotten the prescription for anti-retrovirals to find out if they were able to take the drugs, and in some cases to find out if the baby [was] infected with what we would call a resistant virus, and that’s an area that we’re really trying to look into.

I think we’re viewing each new case of HIV infection in a baby as [an event] that needs to be looked at extremely carefully to better understand how to intervene for these few remaining cases of new baby infections in the U.S. of HIV.

SUSAN DENTZER: The leading suspicion [for cases of mother-to-child transmission that occur even when the woman was taking AIDS drugs] is that there may have been an exceptionally resistant strain of HIV involved?

DR. MARY GLENN FOWLER: It’s one of the things that may have gone on. The other potential possibility is that adherence to the medications may not have been as good as one would hope for, and the other factors are that despite everything, there will be some babies who become infected.

We know that some women with non-detectable virus can transmit. So nothing is perfect, but we also know that making sure the woman is tested, making sure that she’s on the optimal anti-retroviral treatment for herself and for her unborn baby is probably the best way to reduce her risk of transmission to one to two perfect.

SUSAN DENTZER: Let’s talk about some of the results that we know of from the trials done overseas…

DR. MARY GLENN FOWLER: We’ve had a number of very important studies that were done internationally since the 076 AZT trial [on pregnant, HIV-positive women] in the United States, and one of these studies was using a single dose or a single tablet of nevirapine, which was given to women in Uganda who were HIV infected, and they were told to take it at home when they first started labor. And then they came to the hospital and the babies were also given a single dose of the drug before they left the hospital of the nevirapine, and the results of that study which followed about 600 mother-infant pairs showed that giving that regimen to the women could reduce the risk of transmission to their baby by about 42 percent, or almost half. And this was true when we also followed up throughout to 18 months that the risk reduction compared to the other arm of the study, which gave a very short course of AZT to the mother during labor and delivery, and to the baby in the first week, it did not work as well. And so this was about a 50 percent reduction, both short and long term, for women who received the nevirapine.

So based on the simplicity of that regimen, as well as the fact that it’s one that looks like it’s deliverable in many settings, it’s being widely implemented in Africa as well as some parts of Asia.

SUSAN DENTZER: Is it not considered plausible to recommend to women in very hard hit countries not to breast feed?

DR. MARY GLENN FOWLER: …I believe that the WHO guidance is probably the best advice that women can be offered. If it is safe and feasible not to breast feed in a resource limited setting, then the women are encouraged not to breast feed.

For the majority of women, where the water supply is unsafe in many of these resource limited settings, where they cannot afford to use other supplies such as formula, and where culturally it would be stigmatizing and also just not accepted, most women will choose to breast feed, and for that situation then they encouraged to wean as soon as is feasible and safe to do so, and for many settings, that would be between four and six months where there would be alternative food supplies that were available and affordable, and were safe for the baby to take at that point.

So I would say there are some settings and some women in Africa, for example, or other resource limited settings who could choose to use formula and do it safely, and for those women yes, they are being encouraged to not breast feed. But for the majority of women, that’s not often an option, and those women do make a decision after hearing the WHO sort of guidance, and do decide to breast feed. And for those women we really are struggling to get research that will provide answers as to whether there are interventions that will help reduce their risk of transmission during the period that they do breast feed.

SUSAN DENTZER: I want to go back and ask you about C-section practice because I understand there also are international differences there even among the industrialized countries as to how much people believe that a C-section is paramount versus not. What is the evidence there?

DR. MARY GLENN FOWLER: I think we have very good evidence that a scheduled C-section, based on both a randomized trial as well as a very large study that looked individually at different studies, that a scheduled C-section can reduce the risk of transmission for women by about 50 percent.

The caveats on that are that if the woman’s virus load is very low, such as less than 1,000, we have very little data to know whether for that group of women who have an extremely low risk of transmission anyway, whether giving that woman a scheduled C-section will have much impact, or whether it’s just increasing her risk of surgery and morbidity from the scheduled C-section without adding much benefit.

It’s going to be difficult to really get good data on that because we are striving very hard to make sure women are nondetectable, or have very low viral loads at delivery. So as more and more women achieve that goal in the U.S., a number of obstetricians may feel less the need to offer the scheduled C-section.

In Europe … they have a more likelihood of offering scheduled C-section just routinely to the women. So there’s sort of a difference of philosophy based on the same data, and that’s not surprising. Obstetricians, though, will offer the scheduled C-section to women with varying degrees of sort of enthusiasm, probably less in the U.S. if the woman’s viral load is nondetectable or very low, than in Europe where they may well offer it even in the presence of a very low viral load.

SUSAN DENTZER: I want to go back and ask you a question about testing because some people have raised the question about whether such broad testing of all pregnant women for HIV is cost effective. What does the evidence suggest?

DR. MARY GLENN FOWLER: When we’ve tried to look at it very carefully, it does appear that in almost all settings in the United States that universal offering is cost-effective, that the actual cost of taking care of an infected baby over their lifetime, which now will vary, extend into at least adolescence and we hope much longer, that there are lifetime costs that build up in terms of medical care. So it is clearly much better in terms of cost-effectiveness as well as just for the family unit to insure that the mother is tested, and that the baby who was born is not infected…

SUSAN DENTZER: Another area of concern that I know CDC has raised questions about, as has NIH, is what is the long-term impact on those children who are paranatally infected, of being on anti-retroviral drugs for so long. What evidence do we have about that now and what does it show?

DR. MARY GLENN FOWLER: It’s a very important question, and starting in ’94 and ’95 with the recommendations for using initially AZT and then are reinforced with our revisions to our recommendations. We’ve said it’s very important to follow up on the babies long-term to find out if there are late consequences.

The trade-off is you have the potential to avoid a fatal infection in a child, and given that the side effects would have to be extremely high to not make it in terms of the benefit/risk ratio, to not use the drugs. But it’s also very important to follow up.

Some of the concerns that have been raised about whether there might be problems long-term to the sort of inner workings of individual cells, the factories of the cells called mitochondria, and so there have been studies that are trying to look at that in babies.

We so far have very reassuring news in the United States. We’ve followed babies up through about eight years who were in the initial AZT trial that was reported in 1994, and half of those babies received placebo and their mothers received placebo, meaning no active drug, and the other half of the babies and mothers got AZT, and so far through about seven to eight years, their growth has been normal, their development, meaning how they perform in school and how they do is normal, and both groups are not different. And also, their immune function is normal.

They will also try to follow up on a risk of cancer, and again, although the numbers clearly from that trial are not huge numbers — there were about 400 mother-infant pairs in that study — there have been no reported cases of cancer, which was another theoretical concern with the use of these drugs.

So I think at this point we are feeling that to date the findings are encouraging, that we are not seeing any important side effects following out into middle school years, up to about age eight or so, but it’s also extremely important to continue to follow up the mothers and their babies to make sure that there aren’t long-term effects that we don’t know about.

So those things are underway through studies both at NIH as well as studies in CDC.

SUSAN DENTZER: Stepping back from all of this now, to put the success [in] stemming from maternal to child transmission rate, in the context of other things that have happened over the course of the HIV pandemic, how big a deal has this been, what we’ve managed to achieve in lowering the rate of maternal-to-child transmission?

DR. MARY GLENN FOWLER: I think that in the United States, it’s been a really big deal. I think that it’s really shown that we can potentially eliminate any new cases of paranatal HIV infection in the U.S., and as a proof of concept, it may also have important implications in trying to understand preventing transmission from adult to adult. If we understand how important is viral load in paranatal HIV prevention, it probably can be studied very carefully in adult transmission also.

So I think it’s a big deal as both an intervention that works for paranatal prevention, but also helping us understand at least conceptually approaches that might work on preventing transmission between adults.

In addition, the information we got in the United States from the paranatal trials clearly translated internationally. We had, as I think I mentioned, about 1,000 to 2,000 infected babies in the U.S. before our very good interventions now available. We’re now down between 270 infant infections. Internationally, there are well over 600,000 babies who are born infected each year, mostly in resource-limited areas. We know that even with the single dose nevirapine regimen, we could probably cut that transmission in half, which would translate into about 300,000 babies who would have become infected not becoming infected.

If the new interventions … are being tested to prevent breast feeding transmission, either using infant drugs with anti-retrovirals or the maternal drugs during breast feeding, if they are highly effective, as we hope they would be, we may be able to get down to rates of less than 5 percent transmission internationally, which again, if you have 600,000 babies or more, it actually increases each year the numbers infected, if you could really substantially reduce that in resource limited settings with the trials that are ongoing, if we prove they work, then I would say on a global level yes, it would be a big deal.

SUSAN DENTZER: What do you think is the single most important thing for people, the general public, to understand about this area of maternal-to-child transmission?

DR. MARY GLENN FOWLER: I think one of the important things that we have learned from some recent work in talking with pregnant women is that many pregnant women actually don’t know much about HIV. They also don’t know that there are good ways to actually prevent mother-to-child transmission. So part of our need right now, both at CDC and in private settings, is to get the message out to pregnant women that HIV is something that they should be tested for during pregnancy, that all women should be tested. They also should know that interventions that are highly effective are available to protect their baby from getting infected, and also interventions that will help them long-term with their own health care. So I think that’s probably one of the most important messages.

For health care providers, I believe we’ve done a pretty good job about notifying them about how to prevent HIV. I think we still need to put efforts into make sure that they understand the importance of testing all pregnant women, and not doing a risk assessment themselves to decide who should get tested.

I think that we need to remove the health care provider from being a barrier to the woman being offered HIV testing, and that’s why we’re really moving to what we’re terming opt out approach, or routine offering of testing to all pregnant women as part of the normal battery, a standard test that they would receive, but with an awareness that the woman is told she can decline the testing if she wants it.

I think those two things, making the women more aware of HIV and how to prevent transmission, and making providers absolutely aware that they should offer it to all pregnant women, the HIV test.