There is a higher risk for perinatal morbidity and mortality with a breech presentation, even when the risk is adjusted for preterm gestation. The factors that cause breech presentation are often more important than the presentation itself. For example, the severe congenital anomalies that predispose to breech presentation (e.g., hydrocephalus, anencephaly) significantly contribute to neonatal morbidity and mortality. Relative perinatal mortality rates (calculated from data for linked siblings from the Medical Birth Registry of Norway) confirm that breech presentation is a marker of perinatal risk, regardless of the mode of delivery.⁷ Both nonreassuring fetal heart rate (FHR) tracings and dystocia occur more commonly in patients with a term breech presentation, even those who have undergone successful external cephalic version.⁸

Vaginal breech delivery entails a higher risk for maternal morbidity (e.g., infection, perineal trauma, hemorrhage) than vertex delivery.⁴ However, among women with a fetal breech presentation, maternal outcomes are similar between women who had a planned cesarean delivery and those who had a trial of labor. At 2 years postpartum, maternal morbidity assessed by questionnaire (917 responses for a 79% return) was not different for urinary incontinence, breast-feeding, pain, depression, menstrual problems, fatigue, and distressing memories of the birth experience.⁹ In a single-center study of 846 singleton breech deliveries, Schiff et al.¹⁰ also did not find a higher risk for maternal morbidity in women who underwent cesarean delivery during labor than in women who underwent planned cesarean delivery.

External Cephalic Version

The process of external cephalic version converts a breech or shoulder presentation to a vertex presentation. The average success rate for this procedure is 58%, with a wide range reported in published studies.^11,12 External cephalic version is most likely to be successful if (1) the presenting part has not entered the pelvis, (2) amniotic fluid volume is normal, (3) the fetal back is not positioned posteriorly, (4) the patient is not obese, (5) the patient is parous, and (6) the presentation is either frank breech or transverse.¹³ Early labor does not preclude successful external cephalic version, but external cephalic version is rarely successful when the cervix is fully dilated or when the membranes have ruptured. No scoring system has been developed that reliably predicts which candidates will have a successful version attempt, although the variables just listed can be used when obtaining informed consent.

The optimal timing of external cephalic version is after 36 or 37 weeks’ gestation, for the following reasons.^11,14 First, if spontaneous version to a vertex presentation is going to occur, it will likely happen by 36 or 37 weeks’ gestation, and successful performance of external cephalic version after 37 weeks’ gestation decreases the likelihood of reversion from a vertex to a breech presentation. Second, if complications occur during external cephalic version performed after 37 weeks’ gestation, emergency delivery will not result in delivery of a preterm infant.

Successful external cephalic version helps reduce the risk for perinatal morbidity and mortality associated with breech delivery. The American College of Obstetricians and Gynecologists (ACOG)¹¹ has suggested that “because the risk of an adverse event occurring as a result of external cephalic version is small and the cesarean delivery rate is significantly lower among women who have undergone successful version, all women near term with breech presentations should be offered a version attempt.” Labor and vaginal delivery occur in the majority of patients who have undergone successful external cephalic version, albeit with an increased risk for intrapartum cesarean delivery because of dystocia or a nonreassuring FHR tracing.¹⁵ A meta-analysis found that the intrapartum cesarean delivery rate was 27.6% after successful external cephalic version versus 12.5% in women with a spontaneous cephalic presentation.¹⁵

External cephalic version is associated with a low rate of morbidity in contemporary obstetric practice, although placental abruption and preterm labor have been reported.¹¹ Safe external cephalic version requires FHR monitoring and access to cesarean delivery services. In a systematic review of 84 studies that involved 12,955 women, complications included transient (6.1%) and persistent (0.22%) FHR abnormalities, vaginal bleeding (0.30%), placental abruption (0.08%), emergency cesarean delivery (0.35%), and stillbirth (0.19%).¹² Fetal-maternal hemorrhage is another potential complication of external cephalic version.¹² In one study, 16 of 89 (18%) patients undergoing external cephalic version had Kleihauer-Betke stains that signaled the occurrence of fetal-maternal hemorrhage.¹⁶

Obstetricians usually administer a tocolytic agent (e.g., terbutaline) before performing external cephalic version. A randomized placebo-controlled trial found that the success rate of version was doubled when terbutaline was given rather than placebo.¹⁶ Several studies have shown a benefit to tocolysis only in nulliparous women.¹¹ A randomized controlled trial of intravenous nitroglycerin for tocolysis (100- to 300-µg bolus doses, up to a maximum total dose of 1000 µg) found that the success rate was 24% in nulliparous women who received nitroglycerin versus 8% in the placebo group.¹⁷ The success rate was higher in parous women (43%), but it did not differ between the nitroglycerin and placebo groups.¹⁷ Interestingly, the rates of hypotension were similar between groups. A Cochrane Review¹⁸ found that tocolytic therapy for external cephalic version increases the number of women with a cephalic presentation at the onset of labor (relative risk [RR], 1.38; 95% confidence interval [CI], 1.03 to 1.85) and reduces the number of cesarean deliveries (RR, 0.82; 95% CI, 0.71 to 0.94).

Several studies have described the use of epidural or spinal analgesia or anesthesia for external cephalic version.¹⁹ Maternal discomfort may be significant during external cephalic version; greater pain during the procedure is associated with a lower chance of success.²⁰ Some obstetricians argue that the absence of anesthesia limits the force that the obstetrician can apply during the procedure. They contend that administration of anesthesia may encourage the obstetrician to use excessive force, possibly increasing the risk for perinatal morbidity and mortality, but that concern is not supported by published evidence. In fact, spinal anesthesia reduces the force required for successful version.²¹

Weiniger et al.²² randomly assigned 70 nulliparous women to receive either spinal anesthesia with bupivacaine 7.5 mg or no anesthesia for external cephalic version. The success rate was 67% in those receiving spinal anesthesia, 32% in those without analgesia, and 42% in those who did not consent to enroll in the study. A randomized controlled trial in parous women using similar methodology also found an increased success rate with spinal anesthesia (87% versus 58%).²³ In contrast, Sullivan et al.²⁴ randomized 96 parturients to receive either intrathecal bupivacaine 2.5 mg with fentanyl 15 µg as part of a combined spinal-epidural (CSE) technique or intravenous fentanyl 50 µg before attempted external cephalic version. There was no difference between groups in the rate of successful external cephalic version (47% with CSE analgesia and 31% with intravenous fentanyl) or vaginal delivery (36% versus 25%), although pain scores were lower and satisfaction scores were higher with CSE analgesia.

A meta-analysis of seven studies using neuraxial blockade to facilitate external cephalic version concluded that administration of an anesthetic dose of local anesthetic doubles the success rate of external cephalic version (RR, 1.95; 95% CI, 1.46 to 2.60), whereas an analgesic dose does not have any effect (RR, 1.18; 95% CI, 0.94 to 1.49).²⁵ A subsequent meta-analysis confirmed that neuraxial blockade increases the rate of successful version (60%) compared with no neuraxial block (38%); when the authors calculated the number needed to treat, they determined that five women must receive a neuraxial block to achieve one additional successful version.¹⁹ There was no difference between groups in the rate of cesarean delivery; however, this second meta-analysis failed to distinguish between neuraxial anesthesia and analgesia, and it included at least one trial that allowed both vaginal breech delivery and spinal anesthesia for a subsequent version attempt among control patients who had a failed version.¹⁹ Several investigators have reported successful outcomes with neuraxial analgesia or anesthesia in women in whom the first attempt at external cephalic version without neuraxial analgesia had been unsuccessful.^22,26 These patients elected to undergo another version attempt with neuraxial analgesia. Weiniger et al.²² found that failure of external cephalic version was attributed to pain in 15 women in their control group. Eleven of those 15 (73%) women subsequently had successful external cephalic version with spinal analgesia. Cherayil et al.²⁶ reported successful version in 13 of 15 (87%) repeat procedures performed with neuraxial blockade.

In our practices, we do not routinely provide spinal or epidural analgesia during external cephalic version. However, emerging evidence suggests that it may help facilitate successful version and vaginal delivery. If a neuraxial technique is used, administration of an anesthetic dose of local anesthetic appears to result in higher success rates than use of an analgesic dose.

Mode of Delivery

A substantial number of obstetricians recommend the routine performance of cesarean delivery in patients with a breech presentation. The publication of the Term Breech Trial in 2000 changed clinical practice around the world.⁶ In contemporary obstetric practice in the United States, most parturients with a breech presentation are delivered abdominally.

The Term Breech Trial Collaborative Group⁶ enrolled 2088 women from 26 countries with a singleton fetus in a frank or complete breech presentation. These women were randomly assigned to undergo planned cesarean delivery or planned vaginal delivery. Using an intent-to-treat analysis, the investigators noted that perinatal and neonatal mortality rates, and serious neonatal morbidity, were significantly lower in the planned cesarean delivery group (1.6% versus 5%). This difference was greatest in those countries with a low perinatal mortality rate (e.g., Canada, United Kingdom, United States).⁶ Secondary analysis of perinatal outcomes demonstrated that the lowest risk for adverse outcome occurred when a pre-labor cesarean delivery was performed at term gestation. The risk for adverse outcome progressively increased with cesarean delivery performed during early labor and active labor and was highest with a vaginal birth. Labor augmentation and a longer time between the start of pushing and delivery were associated with an increased risk for adverse perinatal outcome, whereas the presence of an experienced clinician at delivery was associated with a reduced risk for adverse perinatal outcome.²⁷ Interestingly, in a 2-year follow-up study in some centers that participated in the Term Breech Trial there was no difference in the risk for death or neurodevelopmental delay between children delivered by planned cesarean and those delivered by planned vaginal delivery.²⁸ The following two factors may have contributed to the lack of significant differences in outcome at 2 years of age: (1) the sample size may have been inadequate, and (2) measures of early neonatal morbidity have a low predictive value for long-term outcomes (i.e., most children with early neonatal morbidity survive and develop normally).²⁸

In the Term Breech Trial,⁶ maternal morbidity and mortality did not differ between the two groups for the first 6 postpartum weeks. Women who underwent planned cesarean delivery were less likely to report urinary incontinence at 3 months²⁹; however, there was no difference at 2 years.⁹ As assessed by questionnaire, maternal outcomes at 2 years after delivery were similar after planned abdominal delivery and after planned vaginal delivery for singleton breech infants born at term.⁹

As a result of the Term Breech Trial, the number of planned vaginal breech deliveries has decreased in many regions of the world. For example, in Denmark, the proportion of singleton term breech infants delivered vaginally decreased abruptly from 20% before 1999 to 6% after 2001.³⁰ At the same time, intrapartum or early neonatal mortality among all term breech infants decreased from 0.13% to 0.05% (RR, 0.38; 95% CI, 0.15 to 0.98).³⁰ These kinds of trends are self-reinforcing. As the number of practitioners with experience in performing vaginal breech delivery has decreased, the number of vaginal breech deliveries available to teach obstetric residents may no longer be adequate.³¹ Results from an Australian survey suggest that “few of the next generation of …obstetricians plan to offer vaginal breech delivery to their patients.”³²

Nevertheless, obstetricians in selected regions of the world retain a strong tradition of offering vaginal breech delivery for selected patients. Published in 2006, the PREsentation et MODe d ‘Accouchement (PREMODA; presentation and mode of delivery) study³³ described birth outcomes for all term breech deliveries in 2001 through 2002 in 174 centers in France and Belgium. The study included 5579 women who planned cesarean breech delivery and 2526 women who planned vaginal breech delivery, of whom 1796 actually delivered vaginally.³³ The primary outcome captured a composite of fetal and neonatal mortality and serious morbidity and was not different between women who planned to undergo vaginal delivery (1.60%; 95% CI, 1.14 to 2.17) and women who planned to undergo cesarean delivery (1.45%; 95% CI, 1.16 to 1.81), with an overall odds ratio of 1.10 (95% CI, 0.75 to 1.61).³³ The authors suggested that rigorous adherence to protocols for patient selection, intrapartum fetal surveillance, and second-stage management contributed to improved outcomes for women attempting vaginal breech delivery.

In 2006, in recognition of results from the PREMODA study and other single-center descriptions of excellent outcomes for vaginal breech delivery, the American College of Obstetricians and Gynecologists (ACOG) made the following recommendations about mode of singleton breech delivery at term³¹:

Although a planned trial of labor and vaginal breech delivery occurs uncommonly in most hospitals in North America and the United Kingdom, vaginal breech delivery still occurs because some patients present in advanced labor. Selection criteria such as those listed in Box 35-2 are used by advocates of a trial of labor and vaginal delivery.⁴ The availability of personnel experienced in obstetric anesthesia and neonatal resuscitation are prerequisites for a trial of labor. Hyperextension of the fetal head remains an absolute contraindication to a trial of labor in the patient with a breech presentation.⁴