Peter S. Bernstein, MD, MPH


July 14, 2004

Recently, one of our graduating residents was bemoaning the fact that she had not ever had the opportunity to deliver a singleton, breech-presenting baby vaginally. I found this extremely disturbing. The delivery of breech babies is rapidly becoming a lost skill largely based on the results of a single randomized trial. That trial by Hannah and colleagues,[1] which was published in The Lancet in 2000, led the Committee on Obstetric Practice of the American College of Obstetricians and Gynecologists (ACOG) to issue an opinion that effectively made it outside the standard of care to offer a trial of labor to a patient with a persistent breech presentation in a singleton pregnancy at term.[2] The Committee Opinion states, "Patients with persistent breech presentation at term in a singleton gestation should undergo cesarean delivery." Although ACOG Committee Opinions are not meant to dictate a specific mode of management, any obstetric care providers who deviate from this recommended course of treatment place themselves at substantial legal risk.

Even if it becomes widely accepted that singleton, breech-presenting babies at term should be delivered via cesarean, there will still be vaginal breech deliveries that need to be done because some are only recognized in the late stages of labor when it is too late to perform a cesarean delivery. Within a generation, these patients may be left in the care of physicians, such as our graduating resident, who have extremely limited experience in performing these sorts of deliveries. From a public health perspective, this may result in larger numbers of poor neonatal outcomes than were potentially averted by the policy recommended by the ACOG Committee on Obstetric Practice.

But should the results of the study by Hannah and colleagues be the ones that guide practice? Although the study was very well done in that it was a multicenter, prospective, randomized trial of more than 2000 women at term with breech-presenting fetuses, there are some significant issues that limit its generalizability. The study was performed at 121 centers in 26 countries. Many of the patients were recruited at medical centers in countries with high perinatal morbidity. The authors go to admirable lengths to stratify their analysis between centers with high and low national perinatal mortality rates; this does not, however, entirely lay to rest the issue of the generalizability of the results to other centers. Thus, although the authors found that there were fewer cases of perinatal mortality in the group assigned to vaginal delivery, many questions arise when looking at their reported cases of perinatal deaths. For example, why was there a twin gestation included when the trial was only to involve singleton pregnancies? Why is a baby who was discharged home well but died in its sleep included? Or one that was dysmorphic? Or one with a myelomeningocele? Or one that died of diarrheal disease after being discharged home well? Although the authors find similar improvement in rates of perinatal morbidity in the group assigned to planned cesarean deliveries, we are left wondering whether there are similar irregularities and violations of the inclusion criteria among the subjects with reported morbidities.

Thus, the study by Hannah and colleagues should not be the last word on vaginal breech deliveries. Fortunately, other investigators are still publishing their research on this topic. Most recently, Alarab and colleagues[3] published their retrospective review of 641 women with singleton, breech-presenting fetuses at term, nearly 300 of whom had vaginal deliveries. Once those with lethal anomalies were excluded, there were no perinatal deaths or serious morbidities among any of the newborns.

Until the pendulum swings back toward the opinion that vaginal delivery of selected patients carrying otherwise uncomplicated pregnancies with breech presentations is appropriate, what are we to do to ensure that the skills to perform these sorts of vaginal deliveries survive? One possibility may be to use birthing simulators. The technology for these devices is improving, and reports of using them to train physicians in other scenarios are beginning to be published, such as the one by Deering and colleagues,[4] who report their experience using them to train residents to manage shoulder dystocia. Another option may be to have 1 or 2 providers at each center available at all times to perform vaginal breech deliveries for women seeking that option or for those who present too late to undergo a cesarean delivery.

If we wait too long to address this issue, we may not be able to best serve our patients.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.