The rate of cesarean delivery in the United States is at a record high and continues to increase. Conversely, rates of vaginal birth after cesarean delivery (VBAC) have declined (to 12.7 percent in 2002 from 28.3 percent in 1996). Childbirth accounts for about one quarter of hospital admissions for U.S. women. Guise and colleagues conducted an extensive review to evaluate the benefits and harms of VBAC and elective repeat cesarean delivery.

The authors searched electronic databases for relevant published articles, systematic reviews, clinical trials, and reports. They also sought evidence from reference lists and contacted experts for recommendations. Relevant evidence arising since the 1980 National Institutes of Health Consensus Development Conference on VBAC was sought. Clinical studies conducted in developed countries were eligible for inclusion if they contained at least 10 patients and provided data about uterine rupture, VBAC rates, and maternal and fetal outcomes, and were not focused on specific subgroups of patients such as those with specific medical or obstetric problems. Two independent investigators rated the quality of each study. Data were abstracted from each study by two independent reviewers.

From more than 6,800 citations, the researchers identified only two randomized controlled trials and 18 observational studies. In all studies, women attending tertiary care centers were more likely to attempt VBAC. On average, 80 percent of mothers who had spontaneous onset of labor delivered vaginally compared with 68 percent of those who received oxytocin, but the difference was not statistically significant. Maternal hemorrhage requiring transfusion was reported in 1.1 percent of the VBAC group and 1.3 percent of women in the repeat cesarean group in a large population-based study. Although this difference was not statistically significant, such hemorrhage was significantly less common in the trial-of-labor group in a prospective cohort study (0.72 compared with 1.72 percent).

Hysterectomy was performed in 0.2 percent of women in the largest studies, with no difference between groups. Information on the reasons for hysterectomy were not well documented. Documentation problems and differences in definitions limited estimates of maternal infection. Data also were poorly and inconsistently recorded on infant outcomes. Uterine rupture was more common in the VBAC group.

The authors conclude that the deficiencies in the scientific literature about the safety of VBAC are striking and that research is needed to help physicians select mothers with the optimal chance of successful VBAC.