Correspondence

Ethical Issues in the Use of Zidovudine to Reduce Vertical Transmission of HIV

N Engl J Med 1995; 332:891-892March 30, 1995

Article

To the Editor:

Despite the strong evidence reported by Connor et al. (Nov. 3. issue)1 that zidovudine therapy administered to pregnant woman with human immunodeficiency virus (HIV) infection can significantly reduce the risk of maternal–infant transmission of the virus, in an accompanying editorial, Bayer maintains that compulsory testing and treatment of pregnant women for HIV disease violate the principle of consent and so are unjustified.2 We believe this claim is too strong. Compulsory testing, although not treatment, may be justified.

Bayer errs in conflating compulsory unblinded testing of pregnant women with compulsory treatment of competent adults. The arguments he offers, which are directed against programs requiring both testing and treatment, do not apply with the same strength to programs that require only testing and leave women to decide with their physicians whether or not to be treated.

Compulsory testing for HIV infection would be minimally invasive and virtually free of risk. Such testing would not involve the practical and ethical difficulties of forced treatment but would ensure that pregnant women and their physicians had the information they needed to make responsible decisions about treatment.

Information can change people's attitudes and behavior. Even though a woman might not have wished to know her HIV status, once she knows she is infected with the virus and has been told the chances of transmitting it to her child and the relative effectiveness of zidovudine therapy in reducing those chances, she will be in a position to act responsibly. Given this information, she is likely to opt for the therapy. Hence, it can be expected that compulsory testing will, over time, prevent the suffering and preserve the lives of thousands of children who would otherwise have died of AIDS.

Christopher A. Hoffman, Ph.D.
Ronald Munson, Ph.D.
University of Missouri–St. Louis, St. Louis, MO 63121-4499

2 References

1. 1

   Connor EM, Sperling RS, Gelber R, et al. Reduction of maternal-infant transmission of human immunodeficiency virus type 1 with zidovudine treatment. N Engl J Med 1994;331:1173-1180
   Full Text | Web of Science | Medline

2. 2

   Bayer R. Ethical challenges posed by zidovudine treatment to reduce vertical transmission of HIV. N Engl J Med 1994;331:1223-1225
   Full Text | Web of Science | Medline

To the Editor:

Prolongation of the life of an HIV-infected child should not be the only benefit to consider in weighing the real cost of compromising the mother's privacy rights. Even if life were not meaningfully extended by the early identification and treatment of HIV-infected children, substantial suffering could be avoided by the prevention or early recognition of serious illness in these children.

Bayer states that “mandatory screening of children could become justifiable if therapeutic interventions could substantially extend the lives of infected children, because treatment, regardless of parental objections, would be imperative.” The prevention of perinatal HIV infection by maternal use of zidovudine is the most effective means to extend the lives of children who are at risk and would make mandatory screening of mothers imperative, according to Bayer's argument.

Donald M. Thea, M.D.
New York City HIV Perinatal Transmission Collaborative Study, New York, NY 10013

To the Editor:

Bayer states, “Whether the wealthy nations of the world will be able and willing to make zidovudine available to the poorest nations will determine the course of the global epidemiology of pediatric AIDS.” We strongly disagree.

For the moment there is no evidence that zidovudine can substantially reduce the number of cases of pediatric AIDS in resource-poor countries. Making zidovudine available to pregnant women may even have the opposite effect, increasing the number of women who want to become pregnant and the number of pregnant women who decide not to consider abortion. The total number of HIV-seropositive children may actually increase, and the number of orphans certainly will. It is also not known whether zidovudine will have the same effect of reducing the rate of HIV transmission from mother to child in resource-poor countries, where the reported rates of HIV transmission are generally higher than in more wealthy nations.1 In addition, pregnant women in resource-poor countries are often anemic. Because zidovudine has potentially hematotoxic effects and hematologic monitoring of pregnant women is often not possible, a higher frequency of side effects associated with the use of zidovudine can be expected. Finally, the resources used for zidovudine treatment and HIV testing among pregnant women may be diverted from other, more useful interventions that have been proved effective in preventing the transmission of HIV.

A. De Roo, M.D.
R. Colebunders, M.D., Ph.D.
Institute of Tropical Medicine, B-2000 Antwerp, Belgium

1 References

1. 1

   Ryder RW, Behets F. Reasons for the wide variation in reported rates of mother-to-child transmission of HIV-1. AIDS 1994;8:1495-1497
   CrossRef | Web of Science | Medline

Author/Editor Response

Dr. Bayer replies:

To the Editor: Compulsory screening of pregnant women for HIV infection continues to provoke sharp dispute, as evidenced by the letters from Hoffman and Munson and from Thea. Crucial matters of a pregnant woman's privacy, the right of self-determination, and the interests of the future child are involved. A long and unhappy record of mandatory treatment of pregnant women — in most cases imposed on those least able to resist the coercive power of the state and of physicians — serves as a backdrop to this dispute.

I oppose mandatory screening of pregnant women for HIV, because I believe that forcing women to undergo treatment with zidovudine for the benefit of their future children is neither ethical nor practical. The evidence provided by ACTG protocol 076 provides good reasons for urging such treatment and for despairing when it is rejected, but not for forcing treatment on unwilling women. Hoffman and Munson seek to distinguish between coerced therapy, which they also reject, and coerced screening, which they find morally acceptable. I disagree.

Hoffman and Munson characterize the intrusion on privacy that coerced screening would entail as “minimally invasive and virtually free of risk.” Although true in a technical sense, this statement disregards the extent to which the imposition of knowledge about a woman's HIV status is psychologically burdensome. The results of an HIV test could, after all, tell a woman that she has a lethal condition. More important, I reject the proposition that such coerced screening can be justified because it would set the stage for a freely chosen and fully informed decision about treatment. The freedom to elect or reject therapy includes the right to determine whether to be informed of the condition that would warrant such treatment. This is true not only for ethical reasons but also because it is a mistake to begin discussing with a woman a potential course of zidovudine treatment on the basis of a test that she did not elect. The mere possibility that compulsory testing would enhance the prospect of a choice to act “responsibly” is not sufficient warrant.

The issues raised by De Roo and Colebunders are different. They suggest that zidovudine therapy might actually increase the number of babies born with HIV infection in poorer countries. their unstated analogy is to the potential influence of a less-than-perfect vaccine on the extent of unsafe sexual practices. There is no evidence that infected women in poor nations “choose” not to become pregnant because of concern about HIV transmission. If there were a treatment that made a radical reduction in HIV transmission possible, would it not be a great boon to women in the world's poorer nations to have the option of such care? They will not have that opportunity without a singular commitment on the part of the wealthier nations of the world. Limitations on children's health are imposed by the conditions of poverty in which so many of the world's women now live. That said, it is clear that efforts to provide zidovudine to pregnant women should never be made at the expense of efforts to prevent HIV infection.

Ronald Bayer, Ph.D
Columbia University School of Public Health, New York, NY 10032

Citing Articles (1)

Citing Articles

1. 1

   Claire Thorne, Marie-Louise Newell, David Dunn, Catherine Peckham, . (1995) The European Collaborative Study: clinical and immunological characteristics of HIV 1-infected pregnant women. BJOG: An International Journal of Obstetrics and Gynaecology 102:11, 869-875
   CrossRef