Is peripartum zidovudine ‘safe and highly beneficial to African children?’

AIDS, 2001 November 9, 15(16): 2204-5

[Note: AZT is the first AIDS drug ever approved, and the main drug given to pregnant women to prevent transmission of HIV to their fetus. The fact that the drug is very likely carcinogenic and mutagenic (among other side effects) is ignored by most AIDS researchers. The study to which this letter is responding is just one example of sloppy or biased thinking when this drug is discussed.]

The strong conclusion `Peripartum exposure to zidovudine [AZT] is safe and highly beneficial to African children, whether or not they are infected' that was made by Dabis et al. [1] does not appear to be warranted, and when considered carefully, the opposite conclusion can be drawn.

The paper's conclusion is supported by the lower overall 18 month mortality rate in children whose HIV-positive mothers received zidovudine, compared with those whose mothers received placebo. One of the pieces of evidence arguing against the paper's conclusion is the higher mortality rate in HIV-positive children whose mothers received zidovudine. It is hard to explain why the children expected to benefit from zidovudine (HIV-positive) did not, and why the children who were not expected to benefit (HIV-negative) did. Children whose mothers received placebo and yet who were born HIV-negative should have a health advantage over children of mothers given zidovudine, by not being exposed to the side-effects of this drug, which is known to cross the placenta [2]. However, in that study, the opposite was apparently found.

A possible explanation for this conundrum is that a methodological Øaw in the research resulted in healthier mothers being assigned to the zidovudine arm of the trial.

Earlier papers from the study [3,4] supported the view that there may be methodological problems in the research:

• Although women with severe anaemia were excluded from participating in the study, there were a significant number of women in the placebo arm who had severe anaemia or neutropenia at day 8 or 45 of the study (eight out of 206), almost as many as in the zidovudine arm (11/201).
• There was a much higher number of stillbirths in the placebo arm (seven versus one).
• Adherence to the trial protocol was claimed to be fairly good, but reliance on pill counts has to assume that participants are not biased towards telling researchers what they probably assume that they want to hear - that all the pills are being consumed. Even so, 14% of mothers admitted taking fewer than 50% of the cumulative treatment.
• Another source of methodological problems could be the partial unblinding of the trial, as a result of the recognition of zidovudine side-effects or the availability of information on the mean corpuscular volume.

It is also of grave concern that mortality in the children of zidovudine-treated mothers was greater from 9 to 18 months. Although the excess mortality was not that great compared with the first 8 months, these children's lives do not end at 18 months. Over the next few years, if this trend continues, the excess mortality in the zidovudine arm of the trial could easily exceed the reduced mortality in the first 8 months.

It is true that the information published by Dabis et al. [1] does not directly support this pessimistic view, but it does not support the optimistic view stated in the conclusions, either. Surely, researchers have a responsibility to state conclusions that are only solidly supported by the evidence in their papers, and not the conclusions that they might like to believe.

Even if the conclusions were justified on the basis of the information gathered by the study, it obviously cannot be generalized to the entire African population. By excluding anaemic/neutropenic women from the study (presumably because of concerns over the exacerbation of these conditions by zidovudine), the conclusions cannot be retroactively applied to this large group of African women.

David R. Crowe
Alberta Reappraising AIDS Society
Box 61037, Kensington Postal Outlet
Calgary, Alberta, T2N 4S6, Canada.

Received: 18 June 2001; accepted: 2 July 2001.

References

[1] Dabis F, Elenga N, Meda N, et al. 18-Month mortality and perinatal exposure to zidovudine in West Africa. AIDS 2001, 15:761-770.

[2] Olivero OA, Parikka R, Poirier MC, Vahakangas K. 3'-Azido-3'- deoxythymidine (AZT) transplacental perfusion kinetics and DNA incorporation in normal human placentas perfused with AZT. Mutat Res Fundam Mol Mech Mutagen 1999, 428:41-47.

[3] Dabis F, Msellati P, Meda N, et al. 6-Month efficacy, tolerance, and acceptability of a short regimen of oral zidovudine to reduce vertical transmission of HIV in breastfed children in Cote d'Ivoire and Burkina Faso: a double-blind placebo-controlled multicentre trial. Lancet 1999, 353:786-792.

[4] DITRAM ANRS 049 Study Group. 15-Month efficacy of maternal oral zidovudine to decrease vertical transmission of HIV-1 in breastfed African children. Lancet 1999, 354:2050-2051.