Fetal malpresentation occurs in approximately 3–4% of term pregnancies1 and is one of the three most common indications for a cesarean delivery.2,3 External cephalic version (ECV) has been shown to reduce the frequency of malpresentation at term and thus the number of cesarean deliveries performed for this indication.1,4 The rapid increase in the frequency of cesarean delivery in the past 20 years now means that one in three women giving birth in the United States will undergo a cesarean delivery.5 Thus, ECV is an important obstetric intervention geared toward the reduction of cesarean delivery, and an attempt is recommended by the American College of Obstetricians and Gynecologists in eligible women without contraindications.6 However, this technique remains underutilized with an estimated 20–40% of eligible women not being offered or not choosing to proceed with an ECV attempt.7–9
Although the potential benefit of a successful ECV with regard to cesarean delivery is evident, the possibility that an ECV attempt, which may or may not be successful, may be related to adverse perinatal outcomes has been understudied.10 Indeed, it remains unclear whether an attempt at ECV results in different perinatal outcomes than foregoing ECV and proceeding with a planned cesarean delivery for fetal malpresentation (ie, “expectant management”). Outcome data regarding this clinically relevant comparison are lacking. Thus, our objective was to compare perinatal outcomes between women with fetal malpresentation at term who underwent an ECV attempt and those who were expectantly managed.
MATERIALS AND METHODS
This is a retrospective cohort study of women who either underwent an ECV procedure at 37 weeks of gestation or greater or did not undergo an ECV and delivered a neonate in a noncephalic presentation at 37 weeks of gestation or greater at Northwestern Memorial Hospital in Chicago, Illinois, from January 2006 to April 2016. Women were included in the cohort if they were at least 18 years of age, had a singleton nonanomalous gestation, and received prenatal care at our institution before 36 weeks of gestation. Women were excluded if they had a history of a prior cesarean delivery or if they had a known placenta previa or cavity-entering myomectomy. Routine clinical policy at our institution during the study period was to offer an ECV attempt in eligible women at approximately 37 weeks of gestation. All ECV procedures are performed by or under the direct supervision of obstetric attending physicians. In addition, it is typical institutional practice for a patient to have neuraxial analgesia, receive one dose of 0.25 mg terbutaline subcutaneously for uterine relaxation before the ECV attempt, and have continuous fetal heart rate monitoring after an attempted ECV until a reactive tracing is observed and the patient is discharged.
All potentially eligible women were identified through a query of the hospital electronic medical records. We systematically identified those who underwent an ECV attempt by searching billing charges and procedure notes for ECV. We identified those who did not undergo an ECV attempt by searching specific templated terms “breech, footling,” “breech, complete,” “breech, frank,” and “transverse” on standardized labor and delivery admission notes. There were no significant changes to the electronic medical records during the study period. Charts were abstracted for demographic and baseline clinical information and for obstetric and perinatal outcomes. Demographic and baseline clinical data collected included maternal age, race and ethnicity, body mass index (BMI, calculated as weight (kg)/[height (m)]2) at delivery, parity, and any pre-existing comorbidities (ie, chronic hypertension, pregestational or gestational diabetes mellitus). Obstetric data abstracted included gestational age at delivery, mode of delivery, and the primary indication if a cesarean delivery was performed.
Women who underwent attempted ECV were compared with those who did not undergo an ECV attempt and delivered a malpresenting neonate at 37 weeks of gestation or greater. Fetal malpresentation was defined as a fetus who was in a complete breech, frank breech, footling breech, or transverse presentation. The primary outcome was a composite of perinatal morbidity and mortality including stillbirth, neonatal death within 72 hours after birth, Apgar score less than 5 at 5 minutes, umbilical artery pH less than 7.0, or base deficit 12 mmol/L or greater, or neonatal therapeutic hypothermia. The decision to perform neonatal therapeutic hypothermia was made at the discretion of the attending neonatologist, but typically occurred in the setting of both neonatal acidemia and a suppressed 10-minute Apgar score. The secondary outcomes were neonatal intensive care unit admission and neonatal anemia (a nadir hemoglobin value of less than 13.5 g/dL in the first 28 days of life).
An a priori power calculation was performed using an α of 0.05 and β of 0.2. With an estimated risk for the composite perinatal morbidity and mortality outcome of 1.0% in the expectant management group based on risks reported in the existing literature,11–13 and using a one-to-one exposed/nonexposed ratio, 4,638 women would be needed to detect at least a twofold increase in the primary perinatal composite outcome associated with an ECV attempt. We projected that a study period of 10 years would allow us to obtain the targeted sample size.
Bivariable comparisons were performed using the Student t test for continuous variables or either χ2 analysis or Fisher exact tests for categorical variables, as appropriate. Multivariable logistic regression analyses were performed to assess whether an attempted ECV procedure was independently associated with the frequency of composite perinatal morbidity and mortality. Variables that significantly differed by exposure (P<.05) in the bivariable analyses were included in the multivariable logistic regression equations.
To account for women who were not documented to have a nonvertex presentation at 37 weeks of gestation but still delivered a vertex fetus after spontaneous version at term, a sensitivity analysis was done to ensure that any observed difference in the frequency of composite perinatal morbidity and mortality was not attributable to the exclusion of these women from the main analysis. This was performed using estimates of the prevalence of breech presentation by gestational age reported in the literature.14–16 We used the highest reported spontaneous version rate of 3.2% in a population-based study by Hickok et al16 to calculate the number of additional women to be included in the expectant management group. We assumed that none of the women who experienced spontaneous version would deliver a neonate with a composite perinatal morbidity and mortality outcome. This assumption biases away from the null. Furthermore, because we anticipated that cord gases may not be obtained for all cases and to account for bias that such missing data may introduce, another sensitivity analysis was done after multiple imputation was used.17,18
All hypotheses tests were two-tailed and P<.05 was used to define statistical significance. All statistical analyses were performed using Stata 14.2. Approval for this study was obtained from the Northwestern University institutional review board with a waiver of informed consent before its initiation.
Of the 4,117 women who met eligibility criteria during the study period, 1,263 (30.7%) underwent attempted ECV and 2,854 (69.3%) were expectantly managed (Fig. 1). Among women who underwent attempted ECV, 1,206 (95.5%) received neuraxial analgesia and 1,198 (94.9%) received terbutaline for uterine relaxation before their procedure.
Women who underwent attempted ECV were more likely to be non-Hispanic white and multiparous and had a lower mean BMI at delivery compared with women who did not undergo an ECV attempt (Table 1). Women who underwent attempted ECV delivered at a slightly earlier gestational age. Of women who underwent attempted ECV, 23 (1.8%) underwent cesarean delivery at the time of their ECV for the indication of nonreassuring fetal status. An additional 816 (64.6%) ultimately underwent cesarean delivery at a later time for a variety of indications (Table 2). Still, women who did not undergo attempted ECV were more likely to deliver by cesarean for any indication (99.2% vs 66.4%, P<.001). Additionally, although the distribution of the primary indication for cesarean delivery differed between the two groups, fetal malpresentation was still the most common indication in both groups.
In bivariable analyses, there were no significant differences between the two groups in the frequencies of the composite perinatal morbidity and mortality outcome or its individual components. There were also no significant differences in the secondary outcomes of neonatal intensive care unit admission and neonatal anemia (Table 3).
Outcomes were similar in the sensitivity analysis that included an additional estimated 94 women who may have had an unknown malpresentation at 37 weeks of gestation and thus did not undergo ECV attempt but delivered a vertex neonate after experiencing spontaneous version at term. Specifically, the composite perinatal morbidity and mortality outcome did not significantly differ between those who underwent ECV attempt and those who did not (2.9% vs 2.5%, P=.46). Similarly, after multiple imputation of missing data in the umbilical artery gases (24% missing), the composite perinatal morbidity and mortality outcome did not significantly differ between those who underwent ECV attempt and those who did not (P=.65).
Multivariate logistic regression was performed in which nulliparity, race and ethnicity, and BMI at delivery were included as independent variables and the primary composite outcome was the dependent variable. After adjusting for these potential confounders, ECV attempt showed no significant association with the composite perinatal morbidity and mortality outcome (Table 4).
In this observational study, fewer than one third of eligible women underwent an ECV attempt for fetal malpresentation at term. Of those who did undergo an ECV attempt, 40% experienced a successful version at the time of their procedure, 79% of whom ultimately delivered vaginally. Of those who did not undergo an ECV attempt and had persistent nonvertex fetal presentation at time of delivery, fewer than 1% delivered vaginally. The composite perinatal morbidity and mortality outcome and secondary outcomes of neonatal intensive care unit admission and neonatal anemia did not differ significantly between those delivered after an ECV attempt and those delivered without an ECV attempt. The absolute frequencies of adverse perinatal outcomes were low in both groups, and there was no evidence of an association between an ECV attempt at term and the risk of adverse perinatal outcomes even after adjusting for potential confounders.
Our findings are consistent with previous studies. In a systematic review by Collaris et al,19 transient pathologic cardiotocography patterns, vaginal bleeding, placental abruption, cord complications, emergency cesarean delivery, and fetal death were reported among women who underwent ECV. However, the absolute rates of these outcomes were low, the complications reported from the included studies were heterogeneous and poorly defined, and adverse outcomes were only assessed among women who underwent an ECV attempt without any comparison group. Similarly, in a systematic review and meta-analysis by Grootscholten et al,10 the pooled complication rate after ECV was 6.1%, but only 0.24% for serious complications. This review also included studies only of women who underwent an ECV attempt and compared outcomes between those who had a successful ECV and those in whom the ECV attempt failed. However, this comparison does not inform the clinical decision of whether to attempt ECV. Our study expands on prior findings by evaluating more detailed and clinically significant perinatal outcomes and comparing these outcomes after an ECV attempt with the more appropriate comparison group of women with a nonvertex presentation who did not undergo an ECV attempt (ie, the expectant management group).
Hofmeyr et al4 performed a Cochrane review of four randomized trials of 1,308 women, comparing women who underwent an ECV attempt with those who were eligible for but did not undergo ECV. They found no statistical differences in Apgar scores, umbilical vein pH less than 7.20, neonatal admissions, or perinatal death. However, these were secondary outcomes in these studies (the primary outcome was mode of delivery), and aside from perinatal death (assessed among all 1,308 deliveries), the other neonatal outcomes were assessed only in smaller subsets (428 neonates for Apgar score, 52 neonates for umbilical vein pH less than 7.20, and 368 neonates for neonatal admission). Our study, although it is observational, augments their findings by focusing primarily on more comprehensive perinatal outcomes among a larger number of women.
Limitations of our analysis should be noted. First, the 40% success rate of ECV observed in our study is on the lower end of the aggregate range reported in the literature.10 In addition, only 31% of the study cohort underwent an ECV attempt. Thus, generalizability to other sites, particularly those with differing ECV protocols, is not assured. Second, the absolute risks of adverse perinatal outcomes are low. Even so, although our a priori sample size estimates were not met, the actual outcome rate was higher than expected; based on the 2.5% rate that was actually observed, the study had 80% power to detect a minimum relative risk between the groups in the composite perinatal morbidity and mortality of 1.7. Third, given the observational nature of the study, there is possible selection bias because women who were thought to be better candidates may have been more likely to undergo an ECV attempt. This is reflected in the baseline characteristics in which women who underwent an ECV attempt were more likely to be multiparous and of non-Hispanic white race and had a lower mean BMI at delivery compared with those who underwent expectant management. Although these potential confounders were accounted for in multivariable regression analysis, the potential for unmeasured confounding remains.
In conclusion, an ECV attempt at term is not associated with increased perinatal morbidity or mortality compared with expectant management. Moreover, although success can be difficult to predict reliably, an ECV attempt does result in a significantly reduced chance of cesarean delivery when compared with expectant management. As such, these data support the existing recommendations that an ECV attempt should be encouraged when fetal malpresentation exists at term.
1. Cluver C, Gyte GM, Sinclair M, Dowswell T, Hofmeyr GJ. Interventions for helping to turn term breech babies to head first presentation when using external cephalic version. The Cochrane Database of Systematic Reviews 2015, Issue 2. Art. No.: CD000184. DOI: 10.1002/14651858.CD000184.pub4.
2. Barber EL, Lundsberg LS, Belanger K, Pettker CM, Funai EF, Illuzzi JL. Indications contributing to the increasing cesarean delivery rate. Obstet Gynecol 2011;118:29–38.
3. Lee HC, El-Sayed YY, Gould JB. Population trends in cesarean delivery for breech presentation in the United States, 1997–2003. Am J Obstet Gynecol 2008;199:59.e1–8.
4. Hofmeyr GJ, Kulier R, West HM. External cephalic version for breech presentation at term. The Cochrane Database of Systematic Reviews 2015, Issue 4. Art. No.: CD000083. DOI: 10.1002/14651858.CD000083.pub3.
5. Safe prevention of the primary cesarean delivery. Obstetric Care Consensus No. 1. American College of Obstetricians and Gynecologists. Obstet Gynecol 2014;123:693–711.
6. External cephalic version. Practice Bulletin No. 161. American College of Obstetricians and Gynecologists. Obstet Gynecol 2016;127:e54–61.
7. Caukwell S, Joels LA, Kyle PM, Mills MS. Women's attitudes towards management of breech presentation at term. J Obstet Gynaecol 2002;22:486–8.
8. Say R, Thomson R, Robson S, Exley C. A qualitative interview study exploring pregnant women's and health professionals' attitudes to external cephalic version. BMC Pregnancy Childbirth 2013;13:4.
9. Vlemmix F, Rosman AN, te Hoven S, van de Berg S, Fleuren MA, Rijnders ME, et al. Implementation of external cephalic version in the Netherlands: a retrospective cohort study. Birth 2014;41:323–9.
10. Grootscholten K, Kok M, Oei SG, Mol BW, van der Post JA. External cephalic version-related risks: a meta-analysis. Obstet Gynecol 2008;112:1143–51.
11. Signore C, Klebanoff M. Neonatal morbidity and mortality after elective cesarean delivery. Clin Perinatol 2008;35:361–71, vi.
12. Hansen AK, Wisborg K, Uldbjerg N, Henriksen TB. Risk of respiratory morbidity in term infants delivered by elective caesarean section: cohort study. BMJ 2008;336:85–7.
13. Hannah ME, Hannah WJ, Hewson SA, Hodnett ED, Saigal S, Willan AR. Planned caesarean section versus planned vaginal birth for breech presentation at term: a randomised multicentre trial. Term Breech Trial Collaborative Group. Lancet 2000;356:1375–83.
14. Scheer K, Nubar J. Variation of fetal presentation with gestational age. Am J Obstet Gynecol 1976;125:269–70.
15. Hill LM. Prevalence of breech presentation by gestational age. Am J Perinatol 1990;7:92–3.
16. Hickok DE, Gordon DC, Milberg JA, Williams MA, Daling JR. The frequency of breech presentation by gestational age at birth: a large population-based study. Am J Obstet Gynecol 1992;166:851–2.
17. Eekhout I, de Vet HC, Twisk JW, Brand JP, de Boer MR, Heymans MW. Missing data in a multi-item instrument were best handled by multiple imputation at the item score level. J Clin Epidemiol 2014;67:335–42.
18. Simons CL, Rivero-Arias O, Yu LM, Simon J. Multiple imputation to deal with missing EQ-5D-3L data: should we impute individual domains or the actual index? Qual Life Res 2015;24:805–15.
19. Collaris RJ, Oei SG. External cephalic version: a safe procedure? A systematic review of version-related risks. Acta Obstet Gynecol Scand 2004;83:511–8.