Our article "Further Reduction of Vertical Transmission with the Use of Elective Cesarean Section" (RITA!, 5:1, p. 16, 1999) reviewed the data presented in an article that was published in The New England Journal of Medicine regarding the use of elective cesarean delivery as a method of further reducing the risk of transmission from mother to child.

The question of elective cesarean delivery has been addressed again in The Journal of the American Medical Association. In an article entitled "Prophylactic cesarean delivery for the prevention of perinatal Human Immunodeficiency Virus transmission/The case for restraint," (281:1946, 1999) Stringer, et al. argue that the evidence for the use of elective delivery as a method of reducing vertical transmission is outdated and weak. These researchers argue that the meta-analysis of the different cohorts was not in fact a meta-analysis of treatment trials, but rather it was a meta-analysis of observational trials. As such, the results are merely suggestive and "represents evidence of insufficient quality by which to make broad clinical recommendations."

Furthermore, they argue that due to rapidly evolving standards of therapy the women in the meta-analysis were receiving monotherapy and not maximally suppressive therapy. Thus, the more pertinent issue of prophylactic cesarean delivery in women on combination therapy is not addressed.

Finally, they point out that there is an increased risk of operative morbidity from subjecting immunocompromised patients to "major abdominal surgery." The irony, they state, is that "the women whose infants theoretically will benefit most from cesarean delivery—women with high viral loads and low CD4 cell counts, who have the highest risk of transmission—are also the women who are most likely to experience operative morbidity."

These authors point out that women who are pregnant are in an especially vulnerable position and may be willing to take almost any risk that will improve the odds against vertical transmission. With this in mind, they caution against adopting prophylactic cesarean delivery as a public policy before more data about the advantages, if any, and the risks of such a procedure are available.

This and other observations were reflected in a letter from Hunter Hammill, MD of Baylor College of Medicine. Portions of that letter reflecting Dr. Hammill's personal experiences follow.

Letter to the editor:

[The following letter is written in response to our review of The New England Journal of Medicine article "The mode of delivery and the risk of transmission of Human Immunodeficiency Virus type-1-a meta-analysis of 15 prospective cohort studies." RITA! 5:1, p. 16, 1999. Due to space consideration only portions of this letter have been reproduced.]

In Houston, Texas we've had the benefit of large medical centers, a multitude of doctors and access to highly active antiretroviral therapy. Since I do not have the luxury of having patients who are HIV-uninfected, I've taken the route of offering pregnant women optimum therapy. In my opinion, this involves offering combination antivirals and following their viral load test results and CD4 T cell counts. The most common combination I currently use is Combivir [zidovudine and lamivudine] and Crixivan [indinavir] or Combivir and Viracept [nelfinavir]. Surprisingly, in one hundred deliveries in my own practice in 1997 and 1998-84 vaginal deliveries, 16 cesarean deliveries-only two infants have been HIV-infected. All these cases involved the use of AZT monotherapy. To date sixty-eight women have received AZT intrapartum [during delivery] prior to the rupture of membranes-58 vaginal deliveries and 10 cesarean deliveries. Only one child is HIV-infected among those born to this group. In my experience the vertical transmission rate using combination therapy is less than 2%. My conclusion from my clinical practice, which would be equivalent in numbers to the clinical practice of some small European countries, is that vertical transmission is radically decreased with close clinical supervision and combination therapy.

The mystery is not solved and there maybe a subset of patients who get infected at a higher rate going through the birth canal even when on combination therapy. This subset has not been defined. We also have some pieces of the puzzle that cloud the issue, such as the data from aborted fetuses showing incorporation of the virus in the thymus even in early gestation. Cesarean sections would not protect these infants.

In summary, the meta-analysis The New England Journal of Medicine published with much fanfare, even on the Internet for early dissemination, is probably a year and half out of date.

It would be tragic to follow a policy that puts a stop to further optimum therapy for women and further diminishes vertical transmission. To paraphrase Oliver W. Holms, "We must provide the optimum approach for women when their lives are doubly precious at the eventful period when they're pregnant." I hope this article does not add confusion, but only give the readers of Research Initiative/Treatment Action! a better understanding of the state of the art.

Hunter Hammill, MD
Board Certified Ob/Gyn
Associate Professor of Ob/Gyn
Department of Community and Family Medicine
Baylor College of Medicine