Vaginal birth after caesarean for women with three or more prior caesareans: assessing safety and success


Authors’ Reply


We appreciate the interest in our recently published retrospective cohort study1 expressed by Drs Spencer and Pakarian2, as well as the opportunity to clarify some of the misconceptions put forth by their letter regarding our study as well as the available published literature on vaginal birth after caesarean (VBAC). First, the authors have expressed concern that the counselling of individual women in our study was not accounted for. Scientifically, this is referred to as confounding by indication, which as we mention in the manuscript, is an inherent potential weakness that should be considered when evaluating our results. We are sure the authors also realise that all published evidence to date on VBAC have this inherent weakness, as it is all observational by nature, and this cannot be accounted for by statistics. We would offer that arguing that our data should not be used to impact clinical practice based on this, in fact applies to the entire body of literature on the subject of VBAC, and is unsound. While the evidence is not perfect, it is the best evidence we have. The alternative, which seems implied by the approach of Drs Spencer and Pakarian, is to practice based on anecdote, opinion and misinformation. That has certainly led us down the wrong path in obstetrics on multiple occasions.

Second, the authors discuss the risk of placenta accreta in the setting of placenta praevia and multiple uterine scars, which we did not discuss in our manuscript because it had nothing to do with the hypothesis tested. However, we are certainly aware of the relationship between multiple uterine scars and abnormal placentation,3 and it would seem to further argue for consideration of strategies that reduce the number of uterine scars in women who have not completed childbearing.4 Although not the direct subject of our study, a casual reader may misinterpret the out-of-context quote of ‘61% risk of placenta accreta’ for women with three prior caesareans, which we feel obligated to address. In the paper by Dr Silver and colleagues,5 a cohort study of women undergoing caesarean without labour, the incidence of placenta praevia in the group undergoing their fourth caesarean (that is, with a history of three) had a placenta praevia rate of 2.27% (33 of 1452); of those 33, 20 had a placenta accreta (which is the source of the ‘61%’).

Third, the assumption that all women who failed their VBAC attempt underwent ‘emergency caesarean sections’ is simply incorrect.

Finally, to the broader point made by Drs Spencer and Pakarian that the results of our study may lead to ‘inappropriate obstetric management’, we would respectfully ask what data these physicians use to counsel, for example, a woman who presents in active labour with a history of three prior low-transverse caesareans as well as a previous vaginal birth, because we are not aware of any. We would respectfully suggest that the authors consider carefully the impact of letters such as these, and refrain from writing opinion pieces that misrepresent the actual evidence on VBAC-associated morbidity and mortality and let the science speak for itself.