Research contradicts the Term Breech Trial (TBT)

A study of 13,000 breech and headfirst births at a Parisian Hospital resulted in a surprisingly high rate of vaginal breech delivery (79%). This is of the highest rates recorded in the literature. The most surprising finding was the absence of significant differences in neonatal morbidity between breech and vertex babies regardless of mode of delivery.

Additionally, there were no significant differences in neonatal morbidity between breech babies delivered vaginally and those delivered by planned cesarean. The use of strict obstetric protocols for breech presentation---including pelvimetry, continuous cervical dilation, very limited use of oxytocin or induction, and specific extraction techniques were held responsible for the excellent neonatal outcomes.

Among both headfirst and breech babies, those delivered by planned cesarean were MORE likely to be intubated than those delivered vaginally. Vertex babies delivered by cesarean were eight times more likely to be intubated and six times more likely to be admitted to the NICU than those born vaginally. See Pubmed Abstract.

These latter findings confirm recent studies showing the link between cesarean delivery and increased risk for respiratory distress syndrome in near term infants.  A 2006 study looking at over 8,000 breech births in France and Belgium concluded that there were no significant differences between planned vaginal and planned cesarean deliveries. The study used similar outcome measures as the TBT, but far stricter criteria for obstetric management. Thus, the outcomes for vaginal deliveries far surpassed those of the TBT and several similar studies. This study showed vaginal breech delivery to be as safe as cesarean delivery, as long as careful selection criteria are followed. The TBT, by contrast, did far less to ensure the safety of the vaginal deliveries it interrogated.

Unlike the TBT, all babies in the vaginal arm were in frank breech position, with no hyper-extension (the star-gazing position). There were no infants with excessively high or low birthweights and all were subjected to continuous electronic fetal heart-rate monitoring. Moreover, pelvimetry was used for over 80% of the candidates in this study, unlike the TBT where only 10% of all women received any pelvic measurement. In this study, roughly 75% of all breech babies were delivered by cesarean. Pubmed Abstract here.

A study of over 600 births in Ireland concluded that vaginal delivery is as safe as cesarean, as long as strict protocol were followed including: appropriate fetal weight (between 2500 -3800 gms), sufficient amniotic fluid, normal fetal morphology, absence of hyperextension of the head, and absence of complete or frank breech presentation. There was no induction nor augmentation of labor other than amniotomy, and women were supposed to dilate 1 cm per hour and limited to no more than one hour of pushing. All women were attended by an experienced obstetrician, using Løvset's maneuver to deliver shoulders and the Mariceau-Smellie-Veit maneuver, if required. The authors state that both obstetric protocols and the avoidance of oxytocin for induction/augmentation are responsible for the favorable outcomes. Pubmed Abstract here.

Recent Studies in Support of the TBT

Several well-designed research trials have confirmed and improved upon the findings of the TBT, using better obstetric protocols and management. They are:

A 2005 study of nearly a million births in Sweden found higher rates of mortality among breech babies versus headfirst babies, as well as higher rates among breech babies delivered vaginally than those delivered by cesarean. Essentially, planned cesareans reduced the risk of mortality by half. Yet the study did admit that one third of all the deaths may have been independent of the mode of delivery, due to high prevalence of malformations. While the results were not strong enough to recommend that all twins be delivered by cesarean, nearly 90% of all singleton term breech babies already are sectioned in Sweden. Here is the Pubmed Abstract.

A 2005 study of term breech babies from the Netherlands confirms a dramatic rise in the cesarean rate and accompanying decline in perinatal mortality and birth trauma. Looking at more than 33,000 breech infants, the study noted a twofold decrease in perinatal death and a fourfold decrease in neonatal trauma in the two years following implementation of the Term Breech Trial. Within two months of the trial's publication, the cesarean rate for term breeches shot up from 50% to 80%, where it has remained since. Here is the Pubmed Abstract.

A retrospective study from California looked at over 100 ,000 term breech babies born over a decade. It concluded that full-term singleton breech babies delivered vaginally had much higher rates of mortality and morbidity than babies those delivered by planned cesarean. However, more than 95% of the breech babies were delivered by cesarean, suggesting a very low threshold of tolerance (or skill?) for vaginal delivery. Here is the full article.

Critiques of the Term Breech Trial (TBT)

Marek Glazerman published a scathing critique of the TBT in ACOG 's flagship journal, American Journal of Obstetrics & Gynecology . Here is the Pubmed Abstract. The argument notes:

• Most cases (8 out of 13 in the vaginal delivery group) of perinatal mortality were unrelated to the mode of delivery.
• Nearly half of all perinatal deaths arose from infants who should not even have been in the study---ie. undiagnosed twins, anenchephaly, & stillbirths.
• There were far more fetuses with birthweights over 4000 gms in the group intended for vaginal delivery than in the planned cesarean group. Again, all of these fetuses should have been excluded from the study.
• The standard of care at two thirds of participating hospitals was so low as to put vaginal deliveries at a definite disadvantage---no facilities for immediate resuscitation, bag/mask, or immediate intubation of newborns.
• Roughly one third of all infants with significant morbidity failed to have an experienced obstetrician in attendance.
• In a two- year followup, study authors themselves found no differences in neurological or developmental outcomes between planned cesarean and planned vaginal deliveries. In fact, infants from the cesarean group had slightly worse outcomes than those from the vaginal delivery group.
• This last point effectively negates the trial's major findings. Rather audaciously, Glazerman suggests that the Trial Recommendations be withdrawn. Yet even if this were to happen, he thinks it is too late to rescue the rapidly dwindling option of vaginal breech delivery. Indeed, the Trial has has been overwhelmingly adopted in most of the developed and developing world.
  

MJ Turner discusses the further flaws of the TBT in the Journal of Obstetrics and Gynaecology (Aug 2006):

• The findings may apply largely to primigravidas or women giving birth the first time. There were twice as many perinatal deaths among primigravidas as among multigravidas.
• The slight differences in immediate perinatal outcomes were not related to delivery, but to labor, especially augmented or induced labor. The adverse effects of synthetic oxytocin and other uterine stimulants on breech deliveries is well known, but was not adressed in the study's conclusions. Even more disconcertingly, the study abandoned a pre-trial consensus to use oxytocin sparingly, if at all.
• Turner concludes that the trial's findings should be reconsidered, although he,too, agrees it is probably too late.

In an early critique published in Birth (March 2002), Marc Keirse claimed hat study guidelines doomed vaginal delivery from the start. Why so?

• A large majority (87%) of births without an obstetrician were in the planned vaginal group.
• The study included 2 stillbirths, two cases of twins, and two with lethal malformations---many of which died, but none of whom should have been included in the trial.
• One third of the babies who died were low birthweight babies---whose death may have had litttle to do with mode of delivery.
• Finally, those countries with higher rate of successful vaginal deliveries had lower rates of overall morbidity and mortality. This implies that a familiarity with vaginal deliveries may be crucial for the ultimate outcomes of those deliveries. As intuitive but often overlooked point that is the focus of Kotaska's brilliant critique.

Andrew Kotaska expresses deep reservations about the TBT and randomized control trials more generally in the British Medical Journal (2004; vol 329):

• The study achieved a remarkable 57% successful vaginal delivery rate only by asking those centers with lower vaginal birth rates to withdraw from the trial or increase their rate.
• As a result, centers may have increased vaginal delivery far beyond their ordinary comfort or skill level, thereby reducing the success of vaginal delivery outcomes.
• More disturbingly, they unwittingly compromised the safety of infants and mothers in the trial.
• The delivery as well as selection of breech babies for a safe vaginal delivery requires considerable skill and finesse---not randomization, which may take the choice out of the physicians hands.
• The questionable results of one randomized trial are now dictating a new standard of care worldwide. This does little justice to the complex variables that influence the safety of a vaginal breech delivery.
  

News flash--- second twins at higher risk

A new study has just been published in the British Medical Journal (BMJ) that claims that second-born term twins are twice as like to die at birth or shortly after birth as first born twins. This finding reflects a relative risk, not an absolute risk which remains very small. It also does not hold for premature twins.

The Atlanta Journal Constitution (3/2/07) reports
" 'I would hate to see this study used as an excuse to have every mother with twins have a Caesarean section,' said Dr. F. Sessions Cole, the head of the neonatal intensive care unit at St. Louis Children's Hospital."

Yet this is precisely my fear. How many doctors will use this study to push mothers with twins into having cesareans? The author of the BMJ study, Gordon Smith, says that cesareans lower the risk for second twins and the research study concludes: "the findings of this and other studies suggest that planned cesarean section may be beneficial for all twins." However the key word is MAY.

This conclusion runs head into an ongoing debate about the best way to deliver twins. Medical opinion is divided on whether cesareans are protective for all twin deliveries . A rather large debate remains, especially when it comes to premature twins or twins that are not headfirst. Yet that debate rarely makes it into the news. when it does, an expert is usually called up to "interpret" the findings.

Our expert here is Dr. Cole, who is quoted as saying that the study applies more to Britain than the US, because we monitor twins better in the US. I think a few British obstetricians might object to this generalization. Dr. Cole also states that most doctors are happy to deliver twins vaginally in the US. This is hardly the case. Between 50% and 90% of twins are delivered by cesarean, depending on their orientation. I'm not talking about sexual preference here. In the twin literature, orientation refers to the twin's position in the uterus: headfirst ( vertex), breech, or transverse. The latter two are likely candidates for a cesarean according to some studies, while this is debated by others.

It is good that Dr. Cole is questioning automatic cesareans for all mothers with twins. Yet the rate of obligatory cesareans for twins is rising in the US and elsewhere. In the US, roughly 60% of all twin pregnancies will include one twin who is not headfirst. This amounts to more than 79 thousand babies a year in the US. Many women with twins where one is headfirst will have little choice but to opt for the 'planned' cesarean their doctor recommends. They won't find a doctor willing to risk a vaginal delivery.

This is what happened to me. My second twin was transverse. I went to the four best and most vaginal-friendly obstetric providers I could find in a 100 mile radius. Not a single OB even entertained the thought of a vaginal delivery. Not the OBs in the hip, feminist, pro-lesbian town of Northampton, nor the crunchy, earth-mother practice in southern Vermont. Luckily, I delivered accidentally into the hands of one of the few doctors on the East Coast who was comfortable with a vaginal breech delivery at 26 weeks.

Studies disagree about the safest mode of delivery in my case---when only twin B is breech and preterm. A French study shows that vaginal delivery is as safe if not safer for twin deliveries, regardless of presenataion or gestational age. Here is the Pubmed abstract of the 2006 study.

So why are so many doctors afraid to deliver breech twins vaginally? Because of the hostile legal climat they labor under, but also because the newer generation of obstetricians has less and less experience delivering twins or breeches (and god forbid, both combined!) vaginally. The public, too, is to blame for wanting a perfect birth and not understanding the very real risks of breech and twin deliveries.

My brilliant obstetrician gives me some hope. But what happens as the younger generation of OBs take over, few of whom have seen or done a vaginal delivery of a breech or twin birth?

Mothers of twins, esp. breech twins----read those scientific abstracts. It may be that a cesarean is safer in your case. It may be that a vaginal birth is safer. It may be that nobody really knows. But don't take one doctor or one expert as the final word.

PS. My second twin actually did better than my first twin. Yeah, yeah, only anecdote. But guess how many cases were responsible for one of Gordon Smith's earlier conclusions? Nine births!!