Traditionally, obstetricians have taken an apparently easy way out and used the presentation combination of twins to decide how to deliver them. Generally, vertex-vertex twins were always appropriate candidates for vaginal birth, whereas nonvertex presentation of the leading twin indicated cesarean birth. The main controversy existed for the subgroup of vertex/nonvertex twins, roughly comprising 30% of all twin sets. This combination of presentations was subject to countless studies and was always considered 1 of the highest dexterity challenges in obstetrics. Despite the plethora of studies, no robust evidence exists to guide clinical advice regarding the best method of birth in these circumstances.

The question of how to deliver a vertex-first twin gestation appeared to be settled following the publication of a large randomized control multicenter study after a very long recruitment period. This Twin Birth Study showed that, between 32 + 0 and 38 + 6 weeks gestation, the rate of cesarean delivery was 90.7% in the planned-cesarean-delivery group (ie, some 10% delivered vaginally before the planned cesarean delivery) and was 43.8% in the planned-vaginal-delivery group (ie, only 56% of planned vaginal births ended vaginally). Planned cesarean delivery did not significantly decrease or increase the risk of fetal or neonatal death or serious neonatal morbidity, as compared with planned vaginal delivery. On the other hand, planned cesarean delivery did not significantly increase the risk of maternal morbidity, as compared with planned vaginal delivery. Thus, this Twin Birth Study appears to support those who advocate vaginal birth and those who believe in planned abdominal delivery.

The article by Asztalos et al in this issue is a subsequent analysis of the Twin Birth Study, in which the authors were able to follow a substantial number (83%) of the twins born in the randomized trial for 2 years and to assess neurodevelopment and survival of these twins. The study appropriately used the generalized estimating equations approach to reduce the effect of twinning on the analysis (ie, if twin A is affected, then twin B is more likely to be affected as well). The authors found no significant difference in the outcome of death or neuro-developmental delay at 2 years of age between twin infants born by planned cesarean delivery compared with those born by planned vaginal birth. The authors reiterate the conclusions from the Twin Birth Study, namely that a policy of planned cesarean delivery provides no benefit for the mother and her twins compared with planned vaginal birth in vertex-first uncomplicated twin pregnancies between 32 and 39 weeks gestation. Conversely, one might also conclude that a policy of planned cesarean delivery had no significant disadvantage for the mother and her twins compared with planned vaginal birth in vertex-first uncomplicated twin gestations.

Despite some significant limitations of the Twin Birth Study, this trial and the present secondary outcome study are the best we have to date. Thus, one might wonder how these analyses could change our practice in delivering a nonvertex twin vis-à-vis the proven advantage of a planned cesarean delivery for the term breech documented by the same institution. As a result of the 2000 Term Breech Trial, many centers worldwide no longer perform vaginal breech deliveries and have extrapolated the conclusions of the Term Breech Trial and applied the conclusion for twins.

Consequently, the newer generations of residents might lack the training, experience, and manual dexterity required for breech delivery, in both singletons and twins. Indeed, a secondary analysis of the Term Breech Trial cited the presence of “an experienced clinician at delivery” among the significant factors that reduced the risk of adverse perinatal outcome among vaginal breech deliveries. Also, the reason that vaginal breech delivery is safer for twin B than for a singleton is unclear. The common explanation that twin A opened the passage for twin B simply ignores the common observation of cervical clamping after the birth of twin A, the rather common situation of high station breech at the beginning of decent of twin B, and the more frequent use of total breech extraction required in twin pregnancies. These considerations might explain the reason why a direct relationship exists between cesarean delivery rates in twins and the rate of combined twin delivery (twin A vaginally, twin B abdominally), which suggests that those who perform more abdominal births in twins and are probably less experienced in vaginal deliveries of twins are more likely to decide on an emergent cesarean delivery for the second twin, thus explaining the higher incidence of combined twin deliveries. A vicious circle might be created whereby the lack of manual dexterity that is required in vaginal births of malpresented twins leads to reduced confidence, which increases the cesarean delivery rate, which, in turn, further reduces manual dexterity, and so forth. Hence, many clinicians who lack the manual dexterity in potentially complicated deliveries, intentionally or not, prefer to deliver twins by the abdominal route for subtle reasons rather than clear-cut indications.

The present study is about neurodevelopment of the twins. However, neurologic morbidity of twins is related infrequently to the mode of delivery, unless birth was traumatic or associated with intrapartum fetal hypoxia. In contrast, brain damage in twins is related clearly to very preterm birth (<32 weeks gestation) and its complications. This risk is especially accentuated by the fact that 11.3% of twins in the United States are born very preterm. Regrettably, this vulnerable group of twins was excluded from the Twin Birth Study, which comprised approximately 90% of twin pairs born after 34 + 0 weeks gestation. It is thus not surprising and rather expected that the incidence of neurologic morbidity will be low and practically the same in both modes of delivery.

Historic US data showed that cesarean birth rate for twins delivered at >22 weeks gestation who weighed >500 g increased by 13% between 1989–1991 and 1997–1999. Although the rates increased to a greater extent at earlier than at later gestational ages, the absolute number of cesarean deliveries was much higher at later gestational ages. More recently, Lee et al used data from the US National Center of Health Statistics to show that abdominal birth rates for twins increased steadily from 53.4% in 1995 to 75.0% in 2008. Interestingly, rates rose for the breech and for the vertex twin category. After risk adjustment, the authors identified an average increase in cesarean births of 5% each year for twins during the study period. One might assume that the increased incidence of cesarean births among twins is due to their increased incidence in the general population. However, Barber et al found that the indication for cesarean delivery in the United States for multiple gestations increased at a much faster rate than the incidence of multiples in the population, with an estimate of 200% higher than would be expected based on population figures. Indeed, multiples contributed just 5.3% of all primary cesarean deliveries in 2003, but their contribution increased to 16.3% in 2009 to the total increase in primary cesarean deliveries. Nevertheless, because twin pregnancies comprised a rather small proportion among all births, the accomplished vaginal births are definitely unable to reduce the general cesarean delivery rate significantly. Thus, reduction of the overall cesarean delivery rate is insignificant (and more so in centers with an overall high cesarean delivery rate) and cannot be an argument against abdominal births in twins.

One may wonder whether we are too late to try to change the fact that the majority of twins are born by cesarean delivery. Cesarean delivery data for twins, from both sides of the Atlantic Ocean suggest that the horses already have left the barn. The reasons for these global trends might be the proportion of twin pregnancies after infertility treatment in older women, consisting the so-called (but never defined) “premium” pregnancies group. In such circumstances, the attending obstetrician may consider a planned cesarean delivery as more controlled and without “hidden surprises” that might unpredictably pop up during vaginal birth of twin B. A planned cesarean delivery also might be the approach to complicated pregnancies, whereby caregivers might follow the cliché that “no high-risk pregnancy should end with a high-risk delivery.” Trial of labor after cesarean delivery also is undertaken infrequently in a twin gestation, irrespective of presentation combination. On top of everything, there is an obvious effect of potential litigation on the obstetric care that leads to higher cesarean rates among complicated births. Ultimately, the results of the Twin Birth Study and the subsequent neurodevelopmental analysis seem to be consequential to a diminishing proportion of twin gestations.

One may also ask whether we are satisfied with the present situation and, if not, whether the reassuring findings of both studies should initiate a renaissance for vaginal birth of twins? The Twin Birth Study did not recommend either mode of delivery but merely states that the vaginal route is permissible and as safe as the abdominal route for both twins and their mothers. If so, why should we bother to change the present trend when even the Twin Birth Study described that 44% of planned vaginal births ended anyway by an unplanned (probably emergent) cesarean delivery? Moreover, one might be interested in neurodevelopmental outcomes of those twin gestations that required an emergent abdominal birth during a planned vaginal delivery, as compared with those who had a planned, probably daytime, elective cesarean delivery. Unfortunately, the data for the last question was not documented by the present study.

Based on the premise that centers who participated in the Twin Birth Study were skilled enough in the art of vaginal twin birth, one should counsel candidates for a vaginal twin birth that, even in such centers, the chance to complete a vaginal twin birth was roughly 50%. Even so, some women might enjoy the freedom of choice between vaginal and cesarean births that is provided by the results of these studies.