The influence of advanced maternal age on the success rate, delivery mode and maternal and fetal outcomes of pregnant women undergoing external cephalic version: a retrospective cohort study

Abstract

Background

Advanced maternal age (AMA) is associated with a higher rate of cesarean sections. Non-cephalic presentation is an important cause of cesarean delivery, and external cephalic version (ECV) offers an opportunity for vaginal trial of labor for these women. However, the influence of advanced age on the success rate of ECV and the ensuing obstetric outcomes remains to be elucidated.

Methods

We conducted a retrospective, single-center cohort study of women with singleton pregnancies who underwent an ECV for breech or shoulder presentation at a tertiary, university-affiliated medical center from January 2016 to September 2023. Data were extracted from routine hospital records. The success rates of ECV and subsequent delivery outcomes were compared between the AMA group and controls.

Results

A total of 448 ECV procedures were analyzed in this study, including 81 AMA cases and 367 controls. The success rate of converting to a cephalic presentation was 75.31% (61/81) for the AMA group and 74.93% (275/367) for controls, with no significant difference (p = 0.943). Of the women who underwent ECV, 214 delivered at our hospital. The vaginal delivery rate was 68.18% (30/44) for the AMA group and 73.53% (125/170) for controls, showing no significant difference (p = 0.479). Further, a comparative analysis of maternal and neonatal outcomes following post-ECV vaginal delivery revealed no significant differences between the AMA group and controls.

Conclusions

In this retrospective cohort, ECV success and short-term obstetric outcomes were comparable between AMA and younger women. Generalisability is limited by the single-centre design, exclusion of scarred uteri, and loss of follow-up for deliveries outside our hospital. Larger, multicentre prospective studies—including women with prior caesarean—are required before definitive recommendations can be made.

Background

Advanced maternal age (AMA) has become a defining feature of contemporary obstetrics [1]. AMA is linked to a higher incidence of non-cephalic presentations, which is a major indication for cesarean section, accounting for 12.6% to 17.6% of all cases and showing an upward trend [2, 3]. Non-cephalic presentation is mostly managed by elective cesarean section; external cephalic version (ECV) is offered selectively, whereas vaginal breech delivery is now rarely attempted because of perceived fetal-maternal risks [4, 5]. Advanced maternal age further increases cesarean uptake [6, 7], and a growing proportion of older mothers present with a previous uterine scar, compounding morbidity in subsequent pregnancies [8–10]. For women who desire for vaginal delivery, current guidelines recommend ECV as first-line intervention [4, 11]. ECV converts the fetus to cephalic presentation and increase the likelihood of vaginal birth, but still entails a risk of failure, additional costs, and potential, albeit minimal, maternal or fetal complications [12, 13].

However, data on ECV success and subsequent obstetric outcomes in AMA women remain insufficient and inconsistent. Several studies have reported that univariate analysis showed successful ECV cases to be older than failures, yet multivariate regression finds no independent association between age and ECV success [14, 15]. Londero AP and colleagues observed a significant difference in age distribution between successful and failed ECVs, with both younger (< 25 years) and older (>40 years) maternal age linked to ECV success [16], whereas another study detected no age-related difference [17]. Regarding mode of delivery after successful ECVs, vaginal birth was associated with older age and multiparity, yet only parity remained independently predictive [18]. None of the existing studies have specifically examined the impact of AMA (≥ 35 years) on ECV outcomes or delivery route; therefore, the influence of maternal age ≥ 35 years on ECV success and on maternal–fetal outcomes warrants further investigation.

In this retrospective cohort study, we evaluated women who underwent ECV at our center. ECV success, delivery mode and maternal-fetal outcomes were compared between AMA and younger women to inform clinical management of non-cephalic pregnancies in older mothers.

Methods

Study design

This was a retrospective, single-center cohort study of women with singleton pregnancies who underwent an ECV for breech or transverse presentation at a tertiary, university-affiliated medical center from January 2016 to September 2023. The study conformed to the Declaration of Helsinki for Medical Research involving Human Subjects. It was approved by the Institutional Review Board of the Tenth Affiliated Hospital, Southern Medical University (KYKT2024-010).

Participants

The inclusion criteria for the participants were as follows: (1) women with singleton pregnancies ≥ 36 weeks of gestation; (2) pregnancies characterized by a breech or shoulder presentation; (3) patients who voluntarily consented to undergo an ECV; (4) ≤ two umbilical cord loops around the fetal neck. Specifically, we excluded pregnant women with more than two loops according to our institutional protocol, based on the evidence that a cord entanglement of ≥ 3 loops is associated with a higher risk of fetal distress [19], which markedly increase the risk of ECV procedure as well as the risk of cesarean delivery even after a successful ECV.

The exclusion criteria for the study participants were detailed as follows: (1) cases with suspected or confirmed placental abruption, placenta previa, active labor, premature rupture of membranes, fetal distress, major fetal anomalies, uterine rupture, abnormal fetal Doppler on ultrasound, and antepartum hemorrhage within seven days; (2) any conditions mandating a cesarean section, regardless of fetal presentation; (3) women with a history of uterine scarring, including those with previous cesarean sections, myomectomy, uterine perforation, or other forms of uterine surgery; (4) instances where patient data were incomplete or insufficient for study inclusion. Specifically, although some studies concluded that ECV is safe and successful for women with one previous cesarean delivery, available data are not sufficient for the low risk of uterine rupture [4, 11]. Therefore, women with a preexisting uterine scar were excluded in this study. The inclusion and exclusion criteria were consistently applied over time.

Women aged ≥ 35 years were included in the AMA group, while those under 35 years of age were included in the control group.

ECV intervention

Patients considered for an ECV were routinely scheduled for a thorough ultrasonographic assessment by skilled sonographers prior to the procedure. On the day of the procedure, these women were admitted to the hospital. Their suitability for ECV was reconfirmed through history, physical examination and bedside ultrasound. A nonstress test (NST) was administered to assess fetal well-being and monitor for uterine contractions. Intravenous tocolysis (ritodrine 50 mg in 150 mL normal saline) was administered 30 min before manipulation if contractions were detected clinically or on the NST.

During ECV, fetal heart rate and position were monitored continuously with bedside ultrasound. As previously described [20], the ECV was executed with the following steps (Fig. 1): The operator applied firm pressure to the fetal hip via the patient’s abdomen, lifting it out of the pelvis and laterally displacing it. Constant, firm pressure was maintained on the fetal hip and head until the fetus was successfully rotated to a cephalic presentation. The obstetrician typically began with a forward roll maneuver. If this initial attempt failed, a backward roll was attempted. When rotation was difficult, the operator instructed the woman to perform slow abdominal breathing, using the rise and fall of the maternal abdomen as an aid. Generally, the procedure was stopped after three unsuccessful attempts, unless the operator, with maternal consent, considered further attempts reasonable. In accordance with previous studies [12, 14] and our institutional experience, the ECV attempt was discontinued under any of the conditions: (1) successful conversion of the fetus to a cephalic position; (2) the passage of 30 min of fetal manipulation; (3) the patient’s request to halt the procedure due to pain or other reasons; (4) the occurrence of vaginal bleeding, premature rupture of membranes, significant abdominal pain, or if bedside ultrasonography detected non-reversible fetal heart rate changes, umbilical cord prolapse, or placental abnormalities; (5) the obstetrician’s determination that further attempts would be futile. Sedation or analgesia was not routine; neuraxial block was used only when specifically requested. All ECVs were performed by the same team led by Dr. Zhongjun Li in our center.

Fig. 1.

Schematic diagram of the ECV procedure

Following the procedure, an ultrasound scan was conducted to confirm fetal presentation and another NST was performed. Provided the patient’s condition remained stable, she was discharged the following day to await spontaneous labor or induction for obstetrical indications.

Data extraction

Data were extracted from the hospital’s electronic medical record system. The collected data included:

1. Maternal characteristics at the time of ECV: age, education level, ethnicity, gravidity, parity, gestational weeks, body mass index (BMI), blood pressure, fundal height, abdominal circumference, and the engagement status of the fetal presentation.

2. Preoperative ultrasonic findings: placental location, any morphological abnormalities of the placenta and umbilical cord, umbilical artery blood flow spectrum, amniotic fluid volume (AFV) and amniotic fluid index (AFI), fetal presentation, nuchal cord loops, and fetal growth parameters such as biparietal diameter (BPD), head circumference (HC), abdominal circumference (AC), and femur length (FL). Additionally, the thickness of the maternal abdominal wall and detailed fetal position were recorded.

3. ECV details: success or failure, use of tocolytic regimen, number of attempts, direction of rotation (clockwise/counter-clockwise), reasons for termination, and interval from preoperative ultrasound to the ECV.

4. Maternal outcomes: mode of delivery, interval from ECV to birth, gestational age at birth, indications for cesarean delivery (if applicable), characteristics and volume of the amniotic fluid, placental and cord abnormalities. For women undergoing vaginal delivery, the following details were further recorded: duration of labor, postpartum hemoglobin decrease, estimated blood loss, episiotomy use, perineal or cervical lacerations, intrapartum fever, premature rupture of membranes, and length of hospital stay.

5. Neonatal outcomes: birth weight, gender, body length, Apgar scores, and admission to the neonatal intensive care unit.

Statistical analysis

The statistical analyses were conducted using SPSS Version 26.0 (IBM Corp., USA). For continuous variables that were normally distributed, the data were presented as mean ± standard deviation (SD), and comparisons between groups were performed using a two-tailed Student’s t-test. In cases where continuous variables were not normally distributed, the data were depicted using median and interquartile range, with group comparisons carried out using the Mann-Whitney U test. Categorical variables were expressed as frequency (percentage), n (%), and inter-group comparisons were executed using the chi-square test or Fisher’s exact test, depending on the situation. A multivariable logistic regression analysis (reported as odds ratios and 95% confidence intervals) was performed to assess factors independently associated with ECV outcome and delivery mode, respectively. All possible predictive factors from univariate analysis were taken in the multivariate model. Mantel–Haenszel stratification analysis was further adopted to ensure robustness of the findings. Statistical significance was defined as a p-value < 0.05.

Results

Demographic and clinical characteristics associated with maternal age

In this study, a total of 448 ECV procedures were analyzed, including 81 patients of advanced age and 367 controls (Fig. 2). Among all ECV procedures, 336 were successful and 112 were unsuccessful, with an overall success rate of 75.00% (336/448). Among the 448 procedures, two complications (0.45%) were recorded. In the AMA group, one intra-operative vaginal bleeding episode occurred; in the control group, one case exhibited persistent funic presentation before, during, and after the procedure. Both pregnancies were delivered by cesarean section.

Fig. 2.

Data enrollment in this study

As shown in Table 1, the median age of the AMA group was significantly higher than that of the controls. The AMA group demonstrated a pronounced increase in both the gravidity and parity, coupled with a markedly higher incidence of multiparous women, all of which were statistically significant. But AMA women did not differ from controls in BMI, blood pressure, fundal height or fetal engagement.Preoperative ultrasound findings were summarized in Table 2. The results marginally greater fetal BPD and HC in the AMA group, but all other sonographic parameters were comparable.

Table 1.

Demographic and clinical characteristics associated with maternal age

	AMA group (n = 81)	Control group (n = 367)	p-value
Age (year)	36.00 [35.00, 38.00]	29.00 [26.00, 31.00]	< 0.001
Gravidity	3.00 [2.00, 4.00]	2.00 [1.00, 3.00]	< 0.001
Parity	1.00 [1.00, 2.00]	0.00 [0.00, 1.00]	< 0.001
Nulliparous			< 0.001
Nulliparous	10 (12.35%)	190 (51.77%)
Multiparous	71 (87.65%)	177 (48.23%)
BMI (kg/m2)	26.70 ± 3.07	25.92 ± 3.18	0.051
Systolic pressure (mmHg)	116.00 [109.00, 120.00]	113.00 [107.00, 121.00]	0.609
Diastolic pressure (mmHg)	75.00 [70.00, 79.00]	73.00 [69.00, 78.00]	0.283
Uterus height (cm)	33.00 [32.00, 34.00]	33.00 [32.00, 34.00]	0.861
Abdomen circumference (cm)	99.00 [95.00, 103.00]	96.00 [93.00, 100.00]	< 0.001
Engagement of fetal presentation			0.534
Yes	1 (1.23%)	12 (3.27%)
No	80 (98.77%)	355 (96.73%)

AMA advanced maternal age, BMI body mass index, ECV external cephalic version

Table 2.

Ultrasonic characteristics associated with maternal age

Ultrasonic parameters	AMA group (n = 81)	Control group (n = 367)	p-value
Fetal growth parameters
BPD (mm)	92.00 [89.00, 95.00]	91.00 [89.00, 93.00]	0.024
HC (mm)	335.00 [326.00, 341.00]	331.00 [325.00, 338.00]	0.022
AC (mm)	329.00 [320.00, 340.00]	327.00 [319.50, 335.00]	0.09
FL (mm)	70.00 [68.00, 71.00]	70.00 [68.00, 71.00]	0.669
Fetal presentation			0.428
Breech	66 (81.48%)	312 (85.01%)
Shoulder	15 (18.52%)	55 (14.99%)
Placental location			0.707
Anterior	46 (56.79%)	200 (54.50%)
Non-anterior	35 (43.21%)	167 (45.50%)
Battledore placenta			0.685
Yes	3 (3.70%)	8 (2.18%)
No	78 (96.30%)	359 (97.82%)
Umbilical cord around fetal neck (round)			0.711
0	49 (60.49%)	211 (57.49%)
1	30 (37.04%)	149 (40.60%)
2	2 (2.47%)	7 (1.91%)
AFV (mm)	46.00 [39.00, 55.00]	45.00 [38.00, 53.00]	0.391
AFI (mm)	114.00 [98.00, 143.00]	110.00 [93.00, 136.00]	0.275
Thickness of maternal abdominal wall (mm)
3 o’clock direction	13.00 [11.00, 16.00]	13.00 [11.40, 15.00]	0.921
6 o’clock direction	12.00 [11.00, 15.00]	13.00 [11.00, 15.00]	0.492
9 o’clock direction	12.00 [10.00, 15.00]	13.00 [11.00, 15.00]	0.191
12 o’clock direction	13.00 [11.00, 15.00]	12.00 [11.00, 14.00]	0.662
Location of fetal spine			0.061
Umbilical region	5 (6.17%)	14 (3.81%)
First quadrant	32 (39.51%)	166 (45.23%)
Second quadrant	11 (13.58%)	67 (18.26%)
Third quadrant	1 (1.23%)	22 (5.99%)
Forth quadrant	32 (39.51%)	98 (26.70%)
Location of fetal breech			0.258
First quadrant	2 (2.47%)	7 (1.91%)
Second quadrant	29 (35.80%)	167 (45.50%)
Third quadrant	49 (60.49%)	191 (52.04%)
Forth quadrant	1 (1.23%)	2 (0.54%)
Location of fetal head			0.158
First quadrant	39 (48.15%)	207 (56.40%)
Second quadrant	2 (2.47%)	2 (0.54%)
Third quadrant	0 (0.00%)	1 (0.27%)
Forth quadrant	40 (49.38%)	157 (42.78%)
Days from preoperative ultrasound to ECV	1.00 [1.00, 2.00]	1.00 [1.00, 3.00]	0.588

AMA advanced maternal age, BPD biparietal diameter, HC head circumference, AC abdominal circumference, FL femur length, AFV amniotic fluid volume, AFI amniotic fluid index, ECV external cephalic version

ECV outcomes associated with maternal age

The outcomes of ECV and procedure-related conditions were presented in Table 3. The proportion of women who were successfully converted to a cephalic presentation in the AMA group and the controls were 75.31% and 74.93%, respectively, showing no significant difference in ECV success rates (p = 0.943). AMA women received tocolysis less frequently, but gestational weeks at ECV, amniotic fluid volume, anesthesia use, direction of rotation, and number of ECV attempts were comparable between groups.

Table 3.

ECV outcomes and associated information

	AMA group (n = 81)	Control group (n = 367)	p-value
ECV outcome			0.943
Success	61 (75.31%)	275 (74.93%)
Failure	20 (24.69%)	92 (25.07%)
Gestational weeks on ECV day			0.202
36–36 + 6 weeks	21 (25.93%)	79 (21.53%)
37–37 + 6 weeks	29 (35.80%)	186 (50.68%)
38–38 + 6 weeks	19 (23.46%)	75 (20.44%)
39–39 + 6 weeks	10 (12.35%)	21 (5.72%)
≥ 40 weeks	2 (2.47%)	6 (1.63%)
AFV on during ECV (mm)	44.00 [35.00, 53.00]	42.00 [35.00, 50.00]	0.18
AFI on during ECV (mm)	110.00 [90.00, 134.00]	106.00 [89.00, 130.00]	0.559
Preoperative tocolysis			0.025
Yes	76 (93.83%)	362 (98.64%)
No	5 (6.17%)	5 (1.36%)
Anesthesia			0.991
Yes	4 (4.94%)	21 (5.72%)
No	77 (95.06%)	346 (94.28%)
Direction of ECV attempts			0.058
Clockwise	24 (29.63%)	97 (26.43%)
Counter-clockwise	45 (55.56%)	237 (64.58%)
First clockwise, then counter-clockwise	3 (3.70%)	19 (5.18%)
First counterclockwise, then clockwise	9 (11.11%)	14 (3.81%)
Times of ECV attempts			0.774
Once	47 (58.02%)	195 (53.13%)
Twice	13 (16.05%)	84 (22.89%)
Three times or more	21 (25.93%)	88 (23.98%)

AMA advanced maternal age, AFV amniotic fluid volume, AFI amniotic fluid index, ECV external cephalic version

To address potential confounding, we constructed a multivariable logistic regression model that included all univariately significant variables (maternal age, gravidity, parity, maternal abdominal circumference, fetal BPD and HC) and clinically relevant factors previously linked to ECV success (maternal BMI, fetal position, engagement, placental location, nuchal cord, AFV and AFI). After adjustment, AMA was not independently associated with ECV success (p = 0.231; adjusted OR = 1.51; 95% CI 0.77–2.96). In contrast, parity and AFI emerged as significant predictors: multiparity nearly doubled the odds of success (p = 0.002; adjusted OR 2.39; 95% CI 1.38–4.14), and each 1-cm increase in amniotic fluid index modestly but significantly raised the likelihood of success (P = 0.001; adjusted OR 1.02; 95% CI 1.01–1.03).

To explore whether parity modified the association between AMA and ECV success, we performed a Mantel–Haenszel stratification analysis by parity. Results showed no evidence of effect modification. Among 200 nulliparous women, AMA was not associated with ECV outcomes (OR = 1.97, 95% CI 0.55–7.05), and among 248 multiparous women the association remained non-significant (OR = 1.43, 95% CI 0.71–2.87). The Breslow-Day test confirmed homogeneity across strata (χ² = 0.19, p = 0.663). The pooled common OR was 1.54 (95% CI 0.83–2.83, p = 0.169), indicating that, after adjustment for parity, AMA does not independently influence the likelihood of successful ECV. However, the wide 95% CI includes clinically relevant effects in both directions, so a type II error due to limited power cannot be excluded.

Mode of delivery following ECV

Since a successful ECV does not guarantee a vaginal birth [18], we further studied the mode of delivery in two groups (Fig. 2). Among all women who undergoing a successful ECV, a total of 214 delivered in our hospital (Table 4). The vaginal delivery rates were similar between two groups: 68.18% (30/44) in AMA and 73.53% (125/170) in controls (p = 0.479).

Table 4.

Mode of delivery associated with maternal age

Mode of delivery	AMA group (n = 44)	Control group (n = 170)	p-value
Vaginal delivery	30 (68.18%)	125 (73.53%)	0.479
Cesarean section	14 (31.82%)	45 (26.47%)

AMA advanced maternal age

Multivariable logistic regression showed that both BMI and parity, but not AMA, independently influenced the delivery mode after successful ECV. Each 1 kg m⁻² increase in BMI raised the odds of vaginal delivery by 24% (p = 0.010; adjusted OR = 1.24, 95% CI 1.05–1.47). Nulliparity reduced the likelihood of vaginal delivery (p = 0.001; adjusted OR = 0.27, 95% CI 0.11–0.65). AMA was not significantly associated with the delivery mode (p = 0.154; adjusted OR = 0.49, 95% CI 0.18–1.31), indicating that age alone does not affect the route of delivery once ECV is successful.

To evaluate whether parity modifies the association between AMA and mode of delivery after successful ECV, we performed Mantel-Haenszel stratification. Among the 105 nulliparous women the odds of vaginal delivery did not differ between AMA and control subjects (crude OR = 1.19, 95% CI 0.10–13.54). In contrast, among the 149 multiparous women AMA was associated with a markedly lower rate of vaginal delivery (OR = 0.30, 95% CI 0.12–0.71). The Breslow-Day test for homogeneity (p = 0.011) indicated heterogeneity across strata, justifying a pooled Mantel-Haenszel estimate. After adjustment for parity, AMA was independently associated with a 64% reduction in the odds of vaginal delivery (Mantel-Haenszel OR = 0.36, 95% CI 0.16–0.81; p = 0.013). Thus, advanced maternal age significantly impairs the chance of vaginal birth after successful ECV, but only in multiparous women.

Maternal and fetal outcomes following ECV

Subsequently, we compared maternal and neonatal outcomes for women who experienced vaginal delivery after ECV. Our maternal outcome assessment encompassed a range of factors including labor duration, amniotic fluid levels, hemoglobin reduction, postpartum hemorrhage volume, soft birth canal laceration, fever during childbirth, and hospital stay duration. Concurrently, we investigated neonatal outcomes encompassing birth weight, gender, body length, head circumference, and Apgar scores. As shown in Tables 5 and 6, all indices were comparable between groups.

Table 5.

Maternal outcomes of women with vaginal delivery following ECV

	Advanced maternal age (n = 22)	Control (n = 113)	p-value
Age (year)	36.86 ± 1.93	28.59 ± 3.40	< 0.001
Duration of labor (h)	4.38 [3.48, 6.52]	5.92 [4.00, 7.75]	0.121
Volume of amniotic fluid (ml)	500.00 [400.00, 600.00]	400.00 [400.00, 500.00]	0.159
Decrease in hemoglobin after delivery (g/L)	3.00 [-2.25, 7.25]	6.00 [0, 15.00]	0.065
Vaginal bleeding within 2 h (ml)	150.00 [150.00, 200.00]	150.00 [150.00, 200.00]	0.957
Vaginal bleeding within 24 h (ml)	250.00 [235.75, 281.25]	250.00 [220.00, 275.00]	0.556
Perineal condition			0.106
Contact	3 (13.64%)	9 (7.96%)
Laceration	18 (81.82%)	79 (69.91%)
Episiotomy	1 (4.55%)	25 (22.12%)
Cervical laceration
Yes	0	0
No	22 (100.00%)	113 (100.00%)
Meconium staining amniotic fluid			0.849
Yes	2 (9.09%)	15 (13.27%)
No	20 (90.91%)	98 (86.73%)
Postpartum hemorrhage			> 0.999
Yes	1 (4.55%)	6 (5.31%)
No	21 (95.45%)	107 (94.69%)
Premature rupture of membranes			0.772
Yes	5 (22.73%)	29 (25.66%)
No	17 (77.27%)	84 (74.34%)
Fever during labor			> 0.999
Yes	0 (0.00%)	4 (3.54%)
No	22 (100.00%)	109 (96.46%)
Length of hospital stay (day)	3.00 [2.00, 4.00]	3.00 [2.00, 4.00]	0.366

AMA advanced maternal age, ECV external cephalic version

Table 6.

Neonatal outcomes of women with vaginal delivery following ECV

	AMA group (n = 22)	Control group (n = 113)	p-value
Birth weight (g)	3278.63 ± 282.7	3271.15 ± 385.14	0.931
Gender			0.881
Male	8 (36.36%)	43 (38.05%)
Female	14 (63.64%)	70 (61.95%)
Body length (cm)	50.50 [50.00, 51.25]	51.00 [50.00, 52.00]	0.92
Head circumference (cm)	34.00 [33.00, 34.00]	34.00 [33.00, 34.00]	0.963
Apgar score − 1 min	9.00 [9.00, 9.00]	9.00 [9.00, 9.00]	0.659
Apgar score − 5 min	10.00 [10.00, 10.00]	10.00 [10.00, 10.00]	0.253
Cord around neck (round)			0.4
0	19 (86.36%)	98 (86.73%)
1	2 (9.09%)	14 (12.39%)
2	1 (4.55%)	1 (0.88%)

AMA advanced maternal age, ECV external cephalic version

Discussion

Compared to younger women, AMA women exhibit a notably higher rate of cesarean sections. This elevation in cesarean rates not only heightens the risks for both mother and child during the current pregnancy, but also poses additional challenges for future pregnancies [8, 9]. Therefore, it is necessary to manage and reduce the cesarean section rates among this group of women. Non-cephalic presentation is an important cause of cesarean delivery, especially in women of advanced age, as a higher maternal age is associated with a higher risk of non-cephalic presentation [3, 21]. ECV offers an opportunity for vaginal trial of labor for these women, thereby serving as a vital strategy to lower cesarean rates and improve maternal and fetal outcomes [4, 11]. However, the influence of AMA on the success rate of ECV and the ensuing obstetric outcomes remains to be revealed.

To our knowledge, this is the first study to comprehensively analyze the impact of AMA on ECV success, as well as the subsequent mode of delivery, and maternal and fetal outcomes. The findings from this study revealed that the success rate of ECV was comparable between the AMA group and the controls. In a further exploration of the delivery modes following ECV, our results demonstrated equivalent vaginal delivery rates across two groups. Additionally, the maternal and fetal outcomes associated with vaginal delivery subsequent to ECV were found to be indistinguishable between the two groups. Therefore, these data indicated that AMA does not exert a significant influence on the efficacy of ECV or the subsequent obstetric outcomes.

Previous studies have examined the correlation between maternal age and the success or failure of ECV, but these investigations have been insufficiently profound, and the findings across various studies were not entirely consistent. Some publications suggested a positive correlation between increased maternal age and the likelihood of successful version. Notably, the mean maternal age was observed to be higher in the successful ECV group compared with the unsuccessful group [14]. Women who experienced a successful ECV were found to be slightly older and more frequently multiparous [15, 22]. However, other studies have reported no discernible correlation between maternal age and the outcomes of ECV [17, 23]. This discrepancy underscores the need for more comprehensive and methodologically rigorous research to clarify the role of maternal age in the success of ECV procedures. In our cohort, the success rates for the AMA group and the controls were 75.31% and 74.93%, respectively, with no significant difference between groups. As the reported success rate of ECV ranges from approximately 50% to 73% [4, 14, 22–24], depending on the patient characteristics, as well as the use of tocolysis and anesthesia. Compared with the results reported in the literature, the success rate in our cohort is encouraging. We attribute this primarily to operator experience: at our centre, all ECV procedures were performed by a single, dedicated team of physicians who have long focused on this technique and therefore possess extensive expertise. Our findings also indicate that AMA does not compromise the likelihood of a successful ECV.

Previous studies have suggested a positive correlation between parity and the success of ECV [4, 11, 15]. In our study, despite the AMA group having higher gravidity and parity compared to the control group, the ECV success rates remained statistically indistinguishable between two groups. Existing data has highlighted that heavier estimated fetal weight tends to increase the likelihood of ECV success [16, 22]. Interestingly, in our study, the AMA group exhibited larger fetal measurements for BPD and HC as determined by ultrasound, yet the ECV success rates were found to be comparable with the controls. This discrepancy may be attributed to the interplay and potential neutralization of the effects of various factors. To adjust for confounders, we also performed multivariable logistic regressions. The results showed that AMA was not significantly related to ECV outcome, while multiparity is independently associated with increased rates of both ECV success.

Literature has indicated that the rate of vaginal delivery following a successful ECV typically ranged from 69.1% to 90.9%, depending on the included populations [15, 25]. In our research, the AMA group and the controls reported vaginal delivery rates of 68.18% and 73.53%, respectively, aligning well with the literature’s findings. The influence of AMA on the mode of delivery and the outcomes for both mother and child post-ECV has been understudied in the past. One study reported that vaginal delivery after successful ECV was associated with multiparity and older age, but only parity was an independent predictor [18]. Interestingly, we report for the first time that the influence of AMA on mode of delivery after successful ECV is confined to multiparous women: AMA reduced the odds of vaginal birth by almost two-thirds in this subgroup, whereas no effect was observed among nulliparous parturients. This novel, parity-dependent association has not been documented in previous series that simply reported overall vaginal-delivery rates after version. We propose two, non-mutually exclusive explanations. Biologically, repeated uterine stretching superimposed on maternal ageing may lead to impaired myometrial contractility, slower cervical remodelling and a lower threshold for labour dystocia. Clinically, older multiparous women—often satisfied with existing family size—may request caesarean delivery at the first sign of labour deviation, while obstetricians, anticipating reduced fetal reserve and precipitate complications, likewise adopt a lower operative threshold. Together these factors inflate the caesarean rate in this specific subset. Importantly, once vaginal delivery was achieved maternal and neonatal outcomes did not differ by age, confirming that ECV remains a safe intervention for AMA women.

This study shows AMA does not impair overall ECV success or maternal-fetal outcomes after vaginal delivery, which have potential implications for counselling and clinical management. First, obstetricians can reassure women ≥ 35 years with a non-cephalic singleton that the probability of achieving cephalic presentation by ECV comparable to that of their younger counterparts. Second, our data support offering ECV as a component of intrapartum care in older gravidae, rather than proceeding directly to planned cesarean on the basis of age alone. Third, the absence of excess maternal or neonatal morbidity underscores that ECV is safe in this cohort. Based on our findings, systematically offering ECV to suitable advanced-age women with non-cephalic presentations might be an effective strategy to reduce unnecessary cesarean delivery rates in this population.

Still, we acknowledge several limitations to the current study. The most important limitation is that patients with a scarred uterus (previous caesarean or any uterine surgery) were excluded in this study. As many AMA women may have had prior cesareans, excluding these population impairs the generalizability of the results. The findings therefore cannot be extrapolated to the large—and rapidly growing—subgroup of older mothers with a scarred uterus, which substantially curtails the clinical applicability of the study. We are now launching a multicentre prospective cohort that will specifically include women with one prior low-transverse caesarean section to fill this evidence gap. Secondly, roughly one-third of successful ECVs were performed in our unit but the patient returned to her referring hospital for delivery. The direction of the resulting bias is unpredictable. Women who remain under our care may be those who anticipate complications (higher caesarean likelihood) or, conversely, those motivated for vaginal birth (lower caesarean likelihood); either scenario distorts the true effect of AMA on final delivery mode. Because we cannot quantify this self-selection, the external validity of the reported obstetric end-points is weakened. Future prospective protocols must mandate linkage of ECV and delivery datasets across centres. Thirdly, the retrospective, single-center design inevitably introduced selection and information biases, unmeasured confounders, and limits external validity, thereby potentially distorting the observed associations between AMA and ECV outcomes. Consequently, these findings should be regarded as hypothesis-generating rather than definitive. Their validity and generalisability can only be strengthened through rigorously designed, multi-centre prospective cohort studies or randomized controlled trials that employ standardised protocols, comprehensive data collection, and advanced statistical methods such as propensity-score matching or instrumental-variable analyses. Fourthly, prior literature offers limited data on the impact of AMA on ECV outcomes, precluding precise sample-size estimation. Exploiting our center’s retrospective cohort, we observed that age alone did not influence ECV success or delivery mode. But the confidence intervals around the AMA effect were broad, compatible with either modest benefit or substantial harm. This imprecision means we cannot rule out a clinically meaningful association, and the non-significant p-value may simply reflect insufficient statistical power rather than a true absence of effect.

Conclusion

In this single-centre, retrospective cohort, ECV success and short-term obstetric outcomes were statistically similar between AMA and younger women after adjustment for parity, BMI and other relevant confounders. These observations are hypothesis-generating rather than conclusive: they suggest that maternal age alone may not preclude ECV, but they require validation in larger, multicentre prospective studies—particularly populations with scarred uteri—before any clinical recommendations can be made.

Abbreviations

AMA

advanced maternal age

ECV

external cephalic version

NST

nonstress test

BMI

body mass index

AFV

amniotic fluid volume

AFI

amniotic fluid index

BPD

biparietal diameter

HC

head circumference

AC

abdominal circumference

FL

femur length

SD

standard deviation

OR

odds ratio

CI

confidential interval

Authors’ contributions

YTX, HXL, MRT and ZJL devised the study plan; YTX, HXL, RCC, DDZ, WMM and HBL helped with data acquisition; QJZ, DDZ, YJO, YCZ and SJZ processed and analyzed the data; SRH and ZJL supervised the research. YTX and HXL wrote the draft manuscript, and QJZ substantially revised the manuscript. XYT created the initial versions of the figures, and QJZ revised them. All authors read the draft manuscript and made important intellectual contributions to the final version. The authors read and approved the final manuscript.

Funding

This study was supported by the Foundation for Basic and Applied Basic Research of Guangdong Province (2024A1515140124, 2025A1515011024), Medical Scientific Research Foundation of Guangdong Province (C2022125), and Dongguan Science and Technology of Social Development Program (20221800906392, 20221800905872).

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study conformed to the Declaration of Helsinki for Medical Research involving Human Subjects. It was approved by the Institutional Review Board of the Tenth Affiliated Hospital, Southern Medical University (KYKT2024-010). Informed consent was exempted due to its retrospective nature.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.