Enhancing external cephalic version success: insights from an Israeli tertiary center

Introduction

Breech presentation occurs when a fetus is positioned with its buttocks or lower extremities entering the pelvis first [1]. This occurs in about 3–4 % of pregnancies at term and is typically detected via ultrasound [1], 2]. Due to the well-documented complications breech presentation can pose to both the fetus and mother during delivery, cesarean deliveries (CD) are traditionally indicated, despite their own associated risks [3], 4]. A recent study highlighted that the rising global CD rates are contributing to an increase in maternal mortality, with a rate of eight per 1,000 procedures in low- and middle-income countries and 16 per 100,000 births in more developed regions [5]. This underscores the need for improved management of breech presentations to help reduce these figures.

One potential approach is external cephalic version (ECV), a manual procedure aimed at repositioning the fetus to a head-down (cephalic) orientation, thereby decreasing the need for cesarean delivery [5]. Although Israel has one of the lowest CD rates among Organization for Economic Co-operation and Development (OECD) countries, with 143.4 CD per 1,000 births [6], there remain significant barriers to the widespread use of ECV in breech cases. These barriers, as noted by Rosman et al., include practitioners lacking sufficient knowledge to effectively counsel patients, physician inexperience, and concerns from patients about potential harm to the fetus or pain associated with the procedure [7].

In recent years, the safety and effectiveness of the ECV procedure have been extensively studied and confirmed by numerous research studies [8], 9]. Factors such as parity, previous vaginal birth, and a high Amniotic Fluid Index have been consistently linked to increased success rates [10]. However, the impact of other factors, including estimated fetal weight, maternal body mass index, and placental location, remains a topic of debate in the literature [10], [11], [12].

In this context, our study aims to contribute new insights into the factors that influence the success of the ECV procedure, while also expanding the existing evidence on maternal and fetal outcomes associated with the procedure.

Materials and methods

This observational study was conducted between November 2013 and March 2020 at Soroka University Medical Center in Beer Sheva, Israel. It received approval from the Medical Ethics Committee (approval number [0020-22-SOR]), and informed consent was obtained from all participants prior to the procedure.

The study included women with breech presentations from 36 weeks of gestation for nulliparous and 37 weeks for multiparas, who were candidates for ECV and had no contraindications for vaginal delivery. Patients with conditions such as oligohydramnios, ruptured membranes, suspected placental abruption, severe hypertension/preeclampsia, multiple gestations, non-reassuring fetal heart rate tracings, or vaginal bleeding were excluded from the study.

Data on maternal characteristics (age, gravidity, parity, body mass index or BMI, blood type, and history of cesarean delivery) were collected from the hospital’s computerized database. Ultrasound assessments of fetal weight, fetal presentation, amniotic fluid index, and placental location were performed before the procedure and included in the analysis.

The ECV procedures were performed at the outpatient day care unit by experienced obstetricians, each with over 10 years of experience. Fetal monitoring was conducted for at least 1 h before and after the procedure to assess fetal well-being. Rh-negative women were given Rh-Ig (Anti-D) following the procedure. During ECV, ultrasound was used to monitor fetal position and condition, including heart rate, tone, breathing, and body movements.

Most procedures were conducted without the use of tocolytics, although subcutaneous injections of Ventolin (0.2 mg) were introduced towards the end of the study for nulliparous patients with increased uterine tone or after a failed ECV attempt. The procedure was stopped if significant maternal discomfort or fetal bradycardia occurred, classifying it as a failed ECV. Due to the lack of anesthesia facilities at the outpatient department, anesthesia was not used.

A successful ECV was defined as achieving a cephalic presentation by the end of the procedure, with no more than two attempts per patient. The failure group included women who still had a non-cephalic presentation at the end of the procedure. No patient was included twice during the same pregnancy.

Data on delivery mode, gestational age at delivery, and perinatal outcomes such as Apgar score, umbilical pH, birth weight, and gender were collected from patients’ computerized records.

To explore factors affecting labor outcomes after a successful ECV, two groups were compared: women who had a vaginal birth and those who underwent a CD after a successful ECV. Basic characteristics, obstetric outcomes, and neonatal outcomes were analyzed using chi-square tests for categorical variables and one-way ANOVA for continuous variables. Categorical data were presented as counts and percentages, and continuous data were shown as mean ± SD Statistical analysis was performed using SPSS software (version 23), with a p-value<0.05 considered statistically significant.

Results

During the study period, 262 women underwent ECV attempts. The overall success rate was 60 % (159/262), with significantly higher success rates among multiparous women compared to nulliparous (73 % [125/171] vs. 37 % [34/91], p<0.001).

Women with successful ECVs were significantly older than those with failed procedures (30.06 ± 5.78 vs. 28.03 ± 5.17 years, p=0.01). While gestational age at ECV was comparable between groups, parity was significantly higher in the successful group (2.11 ± 2.13 vs. 1.03 ± 1.64, p<0.001). Notably, among the 19 women with a previous cesarean delivery, the success rate was not statistically different compared to those without (9.4 vs. 3.9 % respectively, p=0.09). Maternal demographic and clinical characteristics of the study group can be found in Table 1.

No severe complications or stillbirths were reported following ECV procedures. The success of the procedure was associated with longer gestational ages at delivery (39.8 ± 1.13 vs. 39.02 ± 0.76, p<0.001, Table 2). The incidence of minor complications (vaginal bleeding, fetal heart rate changes, or onset of contractions) was similar between successful and failed groups. No cases of membrane rupture occurred during the procedures. Placental location did not influence ECV success rates (Table 2).

Successful ECV was strongly associated with vaginal delivery (86 vs. 3.9 % in failed ECV, p<0.001).

While most neonatal outcomes were comparable between groups, infants born after successful ECV showed slightly lower umbilical cord pH values compared to those after failed procedures (7.29 ± 0.09 vs. 7.33 ± 0.06, p<0.001, Table 3). Likewise, they showed a higher birthweight than babies born after failed ECV (3,354.47 ± 484.48 vs. 3,234.41 ± 371.90, p=0.02, Table 3). Estimated fetal weight (EFW) at the time of the procedure and birth weight were compared using Dollberg growth curves, and these results closely matched actual birth weights.

Among cases with successful ECV, we analyzed factors potentially influencing the subsequent mode of delivery. No significant differences were found in maternal demographics, clinical characteristics, or post-procedure complications between women who delivered vaginally and those requiring cesarean delivery. Placental location also showed no association with delivery mode (Tables 4 and 5).

However, among successful ECV cases, neonates delivered by CD showed less favorable outcomes, including lower 5-min Apgar scores (4.5 vs. 27.3 %, p=0.01) and lower umbilical cord pH values (7.23 ± 0.11 vs. 7.30 ± 0.08, p=0.01, Table 6).

Discussion

This observational study of 262 women who underwent ECV aimed to clarify the factors that influence the success of the procedure and to thoroughly assess maternal and fetal clinical outcomes after ECV. Our analysis yielded several noteworthy findings that address existing gaps in the literature. First, we identified multiparity as a strong positive predictor of ECV success. Additionally, we found that failed ECV attempts were not associated with immediate complications or adverse neonatal outcomes. Another key finding was that women with a history of cesarean delivery experienced success rates similar to those without prior cesareans, countering concerns about ECV attempts in this group. Notably, in contrast to some earlier reports [13], 14], neither placental location nor maternal BMI proved significant in influencing ECV success rates. Together, these findings offer valuable insights to refine patient selection and counseling for ECV procedures, shedding light on several previously uncertain aspects of this intervention.

This study demonstrated an ECV success rate of approximately 60 %, aligning with success rates reported in current literature [11], 15], 16]. Interestingly, while Londero et al. found a U-shaped risk profile for maternal age, where younger (<25 years) and older (>40 years) maternal ages were protective against ECV failure [11], our findings revealed only a slight age difference between the success and failure groups, which we believe lacks clinical significance. This also contrasts with a recent study from Saudi Arabia, which reported better outcomes for women over 30 [15]. Given that multigravidity is associated with higher success rates, both in the literature [17], 18] and in our study, we suspect the slight age difference observed in ECV success may be due to a greater number of younger nulliparouss in the failed ECV group, potentially skewing our results.

BMI has been central to debates regarding its effect on ECV success. While some studies have found that increased BMI correlates with a lower ECV success rate [19], others have reported no significant impact [14], 20]. Although our relatively small sample size may influence our findings, our results align with the latter studies, showing no significant effect of BMI on ECV success. Controversy also surrounds the role of placental location in ECV outcomes. Svensson et al. recently reported higher success rates with posterior placenta positioning, suggesting that it may facilitate ECV by enhancing maneuverability and reducing uterine resistance [14]. Although we acknowledge that a posterior placenta may offer these benefits, our study does not show a significant clinical impact of placental location on ECV success. Rather, our findings align with studies like that of Cobec et al. [21], which describes no influence of placenta positioning in ECV success. We believe that regardless of the BMI and placental location, the expertise of thephysicians involved in our study might have had an impact on maintaining higher success rate of ECV.

In their analysis, Pinto et al. highlight that common reasons for not offering ECV include a lack of experience and insufficient conditions to perform it safely [22]. They also note that, at the national level in Portugal, most centers (87.8 %) expressed interest in hands-on ECV training [22]. While data on Israeli obstetricians’ self-efficacy in performing ECV in indicated cases is lacking, we believe there is a pressing need to assess this to enhance training programs and build self-efficacy among residents and attending physicians across the country.

This is particularly relevant as our study’s results not only align with numerous studies supporting the safety of ECV for mothers but also for fetuses born after ECV attempts, regardless of whether the procedure was successful. Consistent with findings by Impey et al. in their prospective study, we observed no cases of uterine rupture, scar dehiscence, or adverse feto-maternal outcomes throughout our study period [22]. There was no significant difference between the successful and failed ECV groups regarding 1- and 5-min Apgar scores or cord pH values, echoing previous research [22]. Notably, the intrapartum cesarean delivery rate following successful ECV was 14 %, which is lower than the average cesarean rate of 17 % in our labor and delivery unit, based on departmental data. While further statistical analysis is needed to confirm this, we believe, similar to findings reported by other authors [23], 24], that ECV can effectively reduce the overall cesarean delivery rate for breech presentations at term, with low complication rates and favorable maternal and fetal outcomes.

Therefore, extensive training in ECV is essential, particularly within residency training programs, to enhance the self-efficacy of physicians and their comfort in performing this technique when indicated, as recommended by guidelines. This is critical as a prospective cohort study from the Netherlands found that 26 % of women with a breech presentation did not undergo ECV, and in 48 % of these cases, the decision not to proceed was made by the obstetrician, contrary to guidelines [29]. Therefore, mastering the proper technique, as outlined in the literature, is crucial for minimizing potential complications and achieving higher success rates [7]. Focusing on comprehensive training in the correct technique, ideally conducted near a Labor and Delivery unit to address potential premature deliveries, may provide an optimal environment for trainees to build confidence and enhance their ability to perform the procedure successfully. Certain training modalities, such as simulation techniques, have been implemented by previous authors and shown positive outcomes [30, 31]. With our current technological advancements, we strongly advocate for the incorporation and improvement of these simulations into the preliminary training of resident doctors, enabling them to develop a skill that not only improves women’s reproductive health but may also reduce costs for healthcare systems globally [24].

Our study has several strengths. Firstly, we utilized accurate computerized documentation of labor processes and outcomes at our institution. Additionally, Soroka Medical Center is the only tertiary hospital in the area, serving over a million people and handling nearly 17,000 births per year. Due to the absence of other maternity clinics nearby, all pregnant women included in the study delivered at our hospital, which minimized patient drop-out from the study. However, our study also has limitations. The relatively small sample size restricted our ability to analyze certain data, such as the use of tocolytics, which were administered in only a few cases toward the end of the study period. Furthermore, we did not collect data on the type of breech presentation, as certain types, like frank breech, may influence the success rate of ECV. Lastly, cultural and traditional influences that encourage Israeli women to have larger families may contribute to a potential bias, wherein women might tolerate more pain during ECV to reduce the risk of recurrent cesarean deliveries due to breech presentation. Although we lack data to support this hypothesis, further research is needed to better understand this phenomenon.

In conclusion, our study offers valuable insights into the outcomes of ECV in Israel and underscores the critical importance of comprehensive physician training to successfully perform this safe and beneficial procedure. Based on these findings, we advocate for incorporating physician proficiency as a variable in predictive models and for the widespread adoption of simulation-based training in residency programs. These steps would not only improve adherence to current research and guidelines but also contribute to enhanced patient care and more effective resource management within healthcare systems globally.