First Amsterdam Breech Conference, Day 2
Ken Johnson & Betty-Anne Daviss
Rethinking the Physiology of Breech Birth:
A Cohort Study in Frankfurt, Germany, 2004-2011
Betty-Anne Daviss opened the session by remarking that this study has been a long time coming; she’s been working on it since 2008. It is a collaboration between Frankfurt and Ottawa involving Frank Louwen, Anke Reitter, herself, and her epidemiologist husband Ken Johnson.
|from Spinning Babies|
B-AD: Research over the last several decades has focused largely on comparison between vaginal birth and elective cesarean section (ECS), and almost no focus on how to improve vaginal breech birth (VBB). She finds that sad. Most of the large registry studies (such as the ones done in the Netherlands, Canada, or the U.S.) appear to have a higher neonatal mortality and/or morbidity with VBB than with ECS. But the registry studies do not capture the details that the cohort studies do.
There are other problems. In Canada, when Lyons et al published their registry study, the conclusions read that the neonatal mortality and morbidity rates were higher with vaginal breech birth. What the abstract did not make clear was that because the outcome measure was reported as a composite variable. Although the two outcomes were reported together as "higher," it was only the morbidity, not the mortality, that was higher. In fact, when she and Ken went to the actual table, the neonatal mortality (NNM) was clearly reported as "0" and the morbidity was, no doubt, not long-term (as in the Term Breech Trial). But if you only look at the abstract and can't wade through the real meaning of the study, you get terribly fearful of vaginal breech birth. And that fear is difficult to undo.
Betty-Anne suggested that we look at cohort studies done in units, like in France, Belgium, Dublin, Newcastle, Norway, and Frankfurt. In all of these places--with skilled attendants, good screening, and protocols--almost invariably the difference in NNM is very negligible.
Today she and Ken are presenting what it looks like to compare two kinds of vaginal birth. It wasn’t an intention-to-treat study; rather, it compares what actually happened. For more understanding of concerns about relying only on RCTs such as the Term Breech Trial and the history of some of the breech research, refer to Evolving Evidence Since the Term Breech Trial: Canadian Response, European Dissent, and Potential Solutions.
KJ: (Next, Ken presented some information on the premature breeches, which they excluded from the study, but were interesting nevertheless.)
The Frankfurt study included 750 term breeches. 42% were scheduled cesareans; half of those cesareans were by the mother’s choice. The Frankfurt cohort had a high number of primips. Most of the vaginal breech births ended with the mothers upright. They also looked just at the last 2.5 years at the clinic, since they were almost exclusively doing upright births at that point. With mothers exclusively upright, they saw slightly higher success rates.
B-AD: This is an observational cohort study, not a randomized controlled trial. We are looking at what is, not at what’s planned. That is, the cohort study describes what has happened at each birth in the natural process of a particular delivery unit, without instigating or removing parameters, as with the randomized controlled trial. Observational data in a unit can thus be very useful and has some merit of itself that can be more useful than randomization.
But it does raise the question: how do people decide what position they end up in? We explain that in the study.
KJ: Having a woman upright resulted in fewer maneuvers. Forceps and episiotomies were never needed in any of the vaginal breech births.
B-AD: We didn’t collect information about fundal pressure in the database, which is actually used frequently, so that would be useful to do in the future.
KJ: Upright maternal positioning resulted in fewer neonatal injuries and a shorter 2nd stage of labor. How do they define 2nd stage in Germany? It starts at full dilation--not at the onset of spontaneous maternal pushing--so it includes a latent stage. This explains some of the longer 2nd stages recorded in the Frankfurt study.
The Frankfurt study used the definitions of fetal and neonatal mortality & morbidity in the PREMODA study. This allows us to compare the Frankfurt data to the PREMODA study and to the TBT (upon which PREMODA was based).
B-AD: This database is incredibly useful. We need to have more of these databases to amplify this area of knowledge. Observational data in cohort studies is really valuable to individual hospitals so they know what is going on and to compare notes with other units. Collect your data in your unit!
(I had to leave right as they started the Q&A)