The most award winning
healthcare information source.
TRUSTED FOR FOUR DECADES.
Abstract & Commentary
A recent report in the Lancet could have a huge impact on the management of breech deliveries. Hannah and colleagues undertook what seemed like an impossible task to construct and carry out a randomized trial with enough numbers to put to rest the question of whether to deliver frank and/or complete breeches vaginally.
The study involved 2088 women from 121 centers in 26 countries. These women, presenting with complete (breech below the feet) or frank (in the "piked" position) breeches, were randomized to have either a vaginal delivery by an "experienced" clinician (1042 patients) or a planned cesarean section (1041 patients). Of those scheduled to have a vaginal birth, 591 (56.7%) delivered vaginally. Of those assigned to elective cesarean section, 941 (90.4%) had this operation, the others either changed their minds about the route of delivery, or more frequently delivered vaginally before the caesarean section could be accomplished. Follow-up data were available for up to six weeks post delivery.
The results were sobering. Combined perinatal mortality, neonatal mortality, and serious perinatal morbidity were significantly lower in the planned cesarean section group vs. the planned vaginal delivery group (1.6% vs 5%, risk ratio [RR] 0.33; 95% confidence interval [CI] 0.19-0.56). Serious maternal morbidity was not statistically different (3.9% vs 3.2%, P = 0.35). These results appeared to be independent of the experience of the operators or the perinatal mortality rate of the countries in which the deliveries occurred. Perhaps the most striking difference was in the cord pH’s of less than 7.0, where there were 0.4% in the cesarean section group and 2.6% in the vaginal breech group (Hannah ME, et al. Lancet 2000;356:1375-1383).
Comment by John C. Hobbins, MD
Hannah et al commented in the discussion section that many experienced obstetricians would be "disappointed" by the results of the study. This reader was nonplussed by the findings. My bias was that the route of delivery in experienced hands would be no different between groups with regard to perinatal outcome. I also expected more maternal morbidity for those patients having a major operation. Therefore, I began sifting through the study to find holes that would support my bias. Sadly, I could find no major flaws.
How did Hannah et al document the experience of the operators doing the vaginal delivery? The four categories of experience included: 1) Those who considered themselves experienced and were vouched for by the heads of their departments; 2) those with more than 10 years experience with vaginal breech delivery; 3) those with more than 20 years experience with vaginal breech deliveries; and 4) licensed obstetricians (80% of the vaginal breeches were delivered by obstetricians). When broken down, none of these categories had any effect on outcome between the two randomized groups.
I suspected that the increased perinatal mortality and morbidity was only manifested in those countries with high perinatal mortality rates. This also did not pan out because the differences in perinatal mortality between groups in countries with a low perinatal mortality by World Health Organization (WHO) standards was even more striking than in countries with high perinatal mortality rates (0.4% vs 5.7%, RR 0.07, compared with 2.9% vs 4.4%, RR 0.66, NS).
The good news from the study is that perinatal morbidity and serious neonatal morbidity was not as high as one would expect in this multi-center, multi-country study. For example, birth trauma in both groups was less than 15 in 1000 neonatal seizures and were less than eight in 1000. Apgar scores less than 5 at 10 minutes occurred in one in 1000 cases in the cesarean section group and nine in 1000 in the vaginal breech groups.
The higher morbidity in vaginal breeches may not be a function of the delivery itself, but more related to the nature of labor in breech presentations. For example, the incidence of cord blood base in excess of 15 was 2.8% in the vaginal breech group, compared with 0.9% in the planned cesarean section group. A base deficit in this range would point towards predelivery events being responsible for potential fetal compromise.
Up until this point, the only other two breech randomized trials in the literature did not show a difference in the outcome between vaginally delivered breeches and those delivered by cesarian section.1,2 However, the small numbers involved in these studies precluded any valid conclusion. This Lancet study represents the only solid evidence of an advantage to planned caesarean section for breech deliveries, and sets up a major dilemma for teaching institutions. Today, residents in obstetrics get little enough experience in breech delivery, and the results of this study could deliver the final blow to any chance for this type of resident training. The problem is that soon nobody will have the experience to safely deliver the 5.2% of patients in this study who, because of mitigating circumstances, never got their scheduled cesarean sections.
1. Collea JV, et al. Am J Obstet Gynecol 1990;137: 235-244.
2. Gimovsky ML, et al. Am J Obstet Gynecol 1983;146: 34-40.