The recommendations on trial of labor versus elective cesarean delivery for women with a previous cesarean delivery, summarized in this issue of American Family Physician,¹ represent the work of a policy team appointed by the American Academy of Family Physicians (AAFP) Commission on Clinical Policies and Research. In formulating these recommendations, the policy team followed the evidence reviewed in an evidence-based report² from the Agency for Healthcare Research and Quality (AHRQ), which was supplemented by additional literature reviewed by the policy team. The recommendations made by the policy team were based on the best available evidence at the time of the review.

Recommendations based on the best evidence available, however, can fall short when information is not available. For example, the policy team could find no credible evidence that examined the optimal facilities or personnel who should be present when a woman with a previous cesarean delivery attempts a vaginal birth. A recent position statement³ from the American College of Obstetricians and Gynecologists (ACOG) acknowledged this lack of information, yet made strong recommendations that vaginal birth in these circumstances be limited to facilities where immediate cesarean delivery could be performed and where physicians capable of performing the surgical delivery were immediately available. These recommendations have been interpreted to mean that the obstetrician, surgeon, and anesthesiologist must be in the hospital at all times during the patient’s labor. This arrangement may be feasible in hospitals replete with surgeons or trainees, but it has not been proven necessary and is not practical in all hospitals that offer maternity care. The result of this recommendation would be to restrict the availability of a trial of labor for many women who might wish to attempt a vaginal birth.

Given this lack of evidence, the AAFP policy team took a different approach. Members thought that without evidence demonstrating the benefit of an immediately available surgical team, it was inappropriate to limit women’s access to vaginal birth services. Moreover, there was some concern that such availability had the potential for harm (i.e., early surgical intervention for women who might otherwise have had successful vaginal births with expectant management). Because uterine rupture is a devastating complication requiring specific and prompt attention, it is clearly prudent that physicians observe patients closely and have a plan available to deal with this situation if the need arises. However, the policy team felt that a recommendation requiring certain services to be available went beyond what could be supported by the evidence. As a result, the AAFP recommendation conflicts with the ACOG recommendation that surgical attendants be immediately available, and it may create some controversy as physicians and institutions debate whether or how to offer services to women with a previous cesarean delivery. Ultimately, we hope that this difference will stimulate better research to demonstrate the optimal conditions for monitoring women with a previous cesarean delivery without denying the availability of this option to many women who reside outside large cities.

An unrelated issue, but one germane to any evidence-based review, is that the recommendations are limited to the information available at the time that the report was prepared. The current guideline reflects a comprehensive review of the literature through March 2004. A recently published and widely advertised large study⁴ examined the outcomes of women undergoing a trial of labor compared with planned repeat cesarean delivery in a collaborative group of 19 large U.S. teaching institutions. This cohort study showed that the symptomatic rupture rate during a trial of labor was 0.7 percent, which was consistent with the literature reviewed in the AHRQ evidence-based report. The only other maternal outcome that differed between groups was the rate of endometritis, which was higher in the trial-of-labor group (odds ratio [OR], 1.62). In examining neonatal outcomes, researchers found that infants born after a trial of labor were more likely to have hypoxic-ischemic encephalopathy (OR, 0.08) than those born after an elective cesarean delivery (OR, zero).

The results of this trial are important, but the study has several limitations that should be noted. Nearly all low birth weight deliveries occurred in the trial-of-labor group. Similarly, women in the trial-of-labor group were nearly three times as likely to deliver prematurely (i.e., before 37 weeks’ gestation). This raises questions about the comparability of this group with the elective repeat cesarean delivery group. Another limitation is that hypoxic-ischemic encephalopathy is more common in premature infants and therefore may account for many of the observed neonatal differences. Also, during the time that this cohort was studied, prostaglandin was still used for labor induction in women with a prior cesarean delivery. The rate of uterine rupture when prostaglandins were used was twice the rate when no prostaglandins were used, which limits the applicability of the information because of the current recommendation that prostaglandins be avoided. The results of this study were not included in the evidence reviewed by the AAFP policy team. However, this study does show that the landscape of information is constantly changing. Any evidence-based conclusion is only as good as the evidence that was available when the recommendations were made. Consequently, we urge readers interested in this topic to closely monitor the literature for new developments that may help clarify the many areas of this issue that remain unclear.