Vaginal term breech delivery – a time for reappraisal?

Last October, the results of the ‘Term breech trial’ (TBT) were published in The Lancet (1). This multicenter trial recruited 2088 women with term breech fetuses for randomization to either planned cesarean section (CS, n=1043) or planned vaginal delivery (n=1045). The women were eligible for the inclusion: (i) if the fetus at >-37 weeks was in a frank or complete breech; (ii) the fetal weight, assessed clinically or by ultrasound, was below 4000g; (iii) if there were no signs of fetopelvic disproportion when estimated either clinically or with x-ray pelvimetry; and (iv) the fetal head was flexed. Women with fetal anomaly or contraindication to vaginal labor were excluded. After the women had given informed consent they were centrally randomized at the University of Toronto, by a computerized program accessible by telephone. If randomized to elective CS, the operation was performed at or after 38 weeks of gestation. When randomized to vaginal birth, management was expectant until an obstetric indication for induction of labor or CS developed. Of the 1042 women randomized to vaginal birth, a total of 591 delivered vaginally. Perinatal mortality, neonatal mortality, and serious perinatal morbidity were significantly lower (1.5%) in the planned CS group compared to the planned vaginal birth group (5%, p<0.0001). Serious maternal complications were similar in both groups. Although the results of the TBT tell us only about the first 6 weeks follow up of infants and women, despite the absence of an adequate control group of vaginally born cephalic babies, this large and prospective study will have an impact on the management of a term breech baby. The main conclusion of the TBT was to recommend an elective CS as mode of delivery for term breech fetuses. It is, therefore, possible that all breech fetuses of primiparas and multiparas, not only in the Nordic countries with a low perinatal mortality but also in countries with a high perinatal mortality, will, in the future, be born abdominally. If so, this will have remarkable consequences on maternal obstetric outcome, not only in the increased occurrence of risky trials of labor with a uterine scar and an increasing number of uterine ruptures, but also an increasing number of repeated elective CS. For example, in The Netherlands, the risk of maternal death in association with CS is seven times that associated with vaginal birth (2). In addition, there would be a growing numbers of both intra-operative and long-term post-operative complications; infections, pain and adhesions being the most common but not the most serious ones associated with CS (3–7). Now it is time to discuss and evaluate the TBT, especially regarding its ‘standard of care’, the methods and definitions of normality, as well as having a closer look at the short term analyses and presentation of the results. First of all, the standard of care in the 121 centers and 26 countries of the TBT study varied from ‘high’ to ‘usual’. The standard of care was ‘usual’ in 65% of the centers, meaning in practice that in labor, these centers were not able to undertake a CS within 10 minutes of making the decision to do so and not usually having a person available to resuscitate a depressed baby by giving oxygen by bag and mask immediately. Neither were these hospitals able to provide on-site resuscitation or ventilation for the baby for more than 24 hours (1). Prenatally, the size of the pelvis and the deflexion-flexion of the head was estimated clinically in 91% and 31% of cases, respectively, and the fetal weight was estimated by ultrasound in only 60% of cases. Also 51% of cases were born in countries with a very high perinatal mortality rate of >20/1000. This means that, in a significant number of the cases, the real evaluation of the suitability of the patient for vaginal delivery was done on a basis which differs significantly from that used in Scandinavian countries, and they were born in circumstances not common in western delivery units. During labor, fetal heart rate was accepted to be monitored intermittently every 15 minutes in the first stage, and continuous electronic fetal heart rate monitoring was not required even during the second stage of the labor. In the second stage, the descent of the presenting part, the breech, to the pelvic floor was allowed to take 2 hours from full dilatation, and an additional one hour was allowed with active pushing before the delivery had to be imminent (1). These limits of ‘safe’ vaginal course of labor are so broad and wide that, without doubt, even a cephalic baby born vaginally according to this protocol would certainly be more likely to have lower pH and Apgar scores, and be more likely to be neurologically abnormal and even die, than one born by elective CS (8, 9). The main result of the TBT was that when vaginal birth was accepted to take place within this protocol, it was unsafe for the baby. Without doubt, in the setting of the TBT, and in view of the results, we all can agree with this conclusion. However, it does not mean that term vaginal birth would not be safe in selected cases: (i) followed with continuous electronic monitoring of the fetal heart rate and blood sampling when indicated; (ii) acceptance of comparable definitions of adequate course of labor for cephalic and breech babies; and (iii) with a CS performed early enough both in the first and second stage of labor if vaginal delivery is not imminent (9, 10). Several earlier reports have shown that vaginal term breech delivery is safe in selected cases, with no long term consequences or handicap for the infant (9, 11–13). A large Swedish register study of 6542 singleton infants with breech presentation at age of >34 weeks found no difference in the intra- or early neonatal mortality of infants born either vaginally or by CS (14). From this data it can be calculated that in a period of 2 years, the lives of three babies can be saved in Sweden if all full term breech babies are delivered by CS (14). This number is far from the calculations of the TBT, which concluded that for every 14 CSs done, one baby will avoid death or serious morbidity (1). In a study of 1212 breech babies from Norway with careful selection, including a strict protocol of vaginal delivery, continuous fetal cardiotocography (CTG) and possibility of immediate CS if needed, 57% of babies were delivered safely by vaginal route with no long term sequelae (9). Furthermore, during the study period the rate of CS for failure of vaginal breech delivery was decreased from 21% to 6% (9). The results are in accordance with another study from the USA, which showed that of 1021 women with full-term breech babies, 71% of those selected for vaginal delivery actually managed it successfully (15). Large register data show that babies born in breech have a three times higher risk of intrauterine growth retardation (IUGR) and fetal anomaly than those born in cephalic presentation (14). Vaginally born breech babies have lower Apgar scores than those born by CS at age of 5 minutes (1, 16, 17). However, the risk of cerebral palsy in relation to breech presentation at term is not related to the mode of delivery, but linked to the higher rate of IUGR among breech infants (17). In the TBT, 16 of the 2078 babies (0.8%) with no lethal anomalies were stillborn or died neonatally. However, only three of these were from countries with perinatal mortality (PNM) below 20/1000, and in fact one of these babies was a twin with birthweight of 1150 g, being severely growth retarded and probably (?!) dead already before enrolment (1). Thus, deaths of only two babies of the 511 (0.4%) with planned vaginal birth in countries with PNM rate below 20/1000 were actually associated with the vaginal breech delivery. The first had a trial of vaginal delivery with heart rate abnormalities, and a too late CS, and the heart tones of the other disappeared during the second stage of labor. Most likely also these babies could have been saved by continuous CTG monitoring, early fetal blood sampling for pH and base deficit, and an emergency CS provided in a hospital with a high standard of care. Thus, the PNM rate of 4/1000 calculated from these numbers for vaginally born term breech babies is not high, and could have been even lower with better standard of care. As regards the morbidity of the vaginally born babies, long term consequences cannot be evaluated yet. Of the short term morbidity, a significant difference was seen e.g. in the tonus of the babies; those with planned vaginal birth being more hypotonic. Furthermore, the vaginally born babies had lower level of consciousness, lower Apgar scores and cord blood base deficits and pHs. All these parameters are associated with the mode of delivery being vaginal, not only with the fetal presentation in labor (8, 18–23), and should not be compared with those of babies born by elective CS. Unfortunately, there was not a control group of vaginally born cephalic babies in the TBT (1). In fact, the setting of the TBT, randomization of breech babies to either CS or vaginal birth, is equal to a hypothetic study of randomization of cephalic babies to either elective CS or vaginal birth. However, at present, most clinicians regard such a study unethical. In Stockholm, on the 15th of December, 2000, the perinatal working party of the Swedish Association of Obstetrics and Gynecology had an extra meeting focusing on term breech delivery. The meeting was open to delegates from all Nordic societies of obstetrics and gynecology for discussion and sharing opinions. However, no uniform conclusions or statements were made. Instead, it was decided that the Swedish perinatal registers will be scrutinized thoroughly to find out possible long-term consequences of the mode of breech delivery on neonatal welfare. In addition, the possibility to make a new randomized study in Scandinavian delivery units with high and equal standard of care was discussed. However, there would emerge several problems, not only ethical, but also practical e.g. how to recruit a sufficient number of patients accepting randomization. In addition, comparisons would also have to be made between cephalic babies randomized to either vaginal or abdominal delivery, a procedure which, as mentioned above, may be regarded as unethical. In my opinion, willing and compliant multiparous patients should at present have the possibility of delivering a breech baby vaginally, with continuous CTG, possibility of blood sampling from the fetal buttock if needed, and the possibility of emergency CS when necessary. Therefore, after careful selection, an experienced obstetrician should have the possibility of supporting such a choice by the patient to deliver vaginally. However, the generation of obstetricians who are able to regard themselves as experienced in vaginal breech delivery will slowly diminish, and thereafter the patient will have no choice (24). Additional problems connected to inexperience in breech deliveries may arise: How and by whom should cases of nondiagnosed emergency breech deliveries be taken care of, and will, for example, all twins be sectioned in the future?

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.