Comparison one: ECV before term (from 32 weeks with one repeat attempt) compared with no ECV

Mensink 1980 included women in early third trimester (as early as 32 weeks' gestation) in a randomised controlled trial undertaken in Gronigen, The Netherlands. Allocation was undertaken using randomised sealed envelopes, stratified by parity. Breech was verified by ultrasound. Women with a singleton breech presentation before term (from 32 weeks) were included. Women with any contraindication to external version were excluded. External cephalic version (ECV) was attempted without tocolysis by an assistant in training (n = 50) compared with no ECV attempt (n = 52). If the ECV failed, a further attempt was made by an obstetrician one week later. Outcomes included: non‐cephalic births; caesarean section; one minute Apgar score less than seven; umbilical vein pH less than 7.2; neurological deficit in newborn; and perinatal mortality; neonatal morbidity at the time of delivery was reported, but this was not defined.

Comparison two: ECV commencing before term (33 to 40 weeks with repeated attempts) compared with no ECV

Van Veelen 1989 enrolled 180 healthy white Dutch women with uncomplicated pregnancy at 33 to 40 weeks' gestation and a live singleton breech fetus attending antenatal clinic of Ikazia Hospital, Rotterdam, The Netherlands. Random allocation of women used sealed envelopes, and was stratified by parity. Repeated ECV was performed between 33 and 40 weeks' gestation up to four times with no tocolysis, analgesia or anaesthesia compared with no ECV. The outcomes included: presentation at delivery; mode of delivery; neonatal outcome including perinatal death.

Comparison three: ECV commencing before term compared with ECV commencing after term (37 weeks' gestation)

In this update three studies are now included in this comparison (Akhtar 2013; Hutton 2003; Hutton 2011).

Hutton 2003 is an international multicentre randomised controlled trial (n = 233). All nulliparous women with any breech presentation and multiparous women with a frank breech presentation were eligible for the trial if they had a live singleton fetus and a gestational age of between 34 weeks, 0 days and 36 weeks 0 days. Women were excluded if they had a parity greater than four, if they planned to move to a non‐trial centre, or if there was any contraindication to labour or vaginal birth (such as placenta previa, or previous classical caesarean section), to ECV (such as fetal heart rate abnormalities, abruptio placenta, fetal anomalies, uterine anomalies, oligohydramnios, rupture of membranes, over distended uterus) or to early ECV (such as fetus engaged in the pelvis, an increased risk of preterm labour, increased risk of abruptio placenta). ECV was begun between 34 weeks 0 days and 36 weeks 0 days in the early group (n = 117); and between 37 weeks 0 days and 38 weeks 0 days in the delayed group (n = 116). Tocolysis was recommended to be used either routinely or selectively in both groups; analgesia was permitted. The primary outcome was presentation at delivery; other outcomes included: caesarean section; serious fetal complication; preterm birth less than 37 weeks; women's views about ECV. The study was funded by Canadian Institutes of Health Research.

The Hutton 2011 study included 68 centres in 21 countries with ECV carried out by clinicians who were experienced in the procedure and with birth facilities that were deemed to meet Canadian standards. One‐thousand, five‐hundred and forty‐three women were randomised. The study recruited women with a singleton fetus in a breech presentation, between gestation ages of 33 weeks 0 days and 35 weeks six days. Women with contraindications to ECV (e.g. fetal heart rate abnormalities, placental abruption, major life‐threatening fetal anomalies, uterine anomalies, hyper‐extended fetal head, rupture of fetal membranes, severe oligohydramnios or hydramnios); contraindications to early ECV (e.g. increased risk of preterm labour or placental abruption); or contraindications to labour or vaginal birth (e.g. placenta praevia, previous classical caesarean section); or if they had been prior participants in the trial; were at increased risk of unstable lie (such as grand multiparity); or if they planned to give birth by caesarean section even if the fetus turned to a cephalic position, or if they planned a vaginal birth if the fetus remained breech were excluded. In the early ECV group (n = 767), ECV carried out between 34 weeks 0 days and 35 weeks six days gestation, and within seven days of randomisation. In the delayed ECV group (n = 774) ECV carried out at or after 37 weeks' gestation. In both groups fetal presentation was confirmed by ultrasound, fetal heart rate was monitored before, during and after the procedure. The use of tocolytics and analgesia was left to the discretion of the clinician, and they were directed to use the same approach for women in both arms of the trial. If the procedure was unsuccessful, or if a fetus reverted to non‐cephalic, a repeat ECV procedure could be performed at a later date at the discretion of the care provider in consultation with the woman.

The study by Akhtar 2013 is a single‐centre, parallel‐group randomised controlled trial carried out in a hospital in Pakistan. The study included women with a singleton fetus with breech presentation between 33 and 35 weeks' gestation (n = 123 women). Women with contraindications to ECV, contraindications to early ECV or contraindications to labour or vaginal birth (e.g. fetal heart rate abnormalities, vaginal bleeding, rupture of membranes, placental abruption, fetal growth restriction, previous CS, low amniotic fluid index, fetal weight greater than 4 kg) or women unwilling to undergo ECV were excluded. In the early ECV group, ECV was carried out between 34 (238 days) and 35 weeks of gestation (n = 63). In the delayed ECV group ECV was carried out at or after 37 weeks. No tocolytics were used in either group and women were monitored for three hours before and one hour after the procedure. Up to two‐three attempts were allowed. The procedure was discontinued if there was excessive maternal discomfort or fetal heart rate irregularities (n = 60).

Ongoing studies

We have limited information on the study by Belizan 1989; it is not clear whether this study was completed, more information is set out in Characteristics of ongoing studies.

Primary outcomes

The rate of non‐cephalic presentation at birth in the ECV group was 40% and in the no ECV group was 39% (risk ratio (RR) 1.04, 95% confidence interval (CI) 0.64 to 1.69) Analysis 1.1. (The trial authors ascribe the low success rate to the gentleness with which ECV was attempted.) There was no clear difference between groups for failure to achieve vaginal cephalic birth (caesarean section plus breech vaginal birth) (RR 1.04, 95% CI 0.67 to 1.62) Analysis 1.2. (The rate of caesarean section was 14% in the ECV group and 8% in the no ECV group (RR 1.82, 95% CI 0.57 to 5.84) Analysis 1.3. The number of women undergoing vaginal breech delivery was comparable in the two groups (RR 0.87, 95% CI 0.49 to 1.52) Analysis 1.4.

Secondary outcomes

There was no clear evidence of differences between groups for other outcomes.

• The rate of one minute Apgar scores less than seven (RR 0.62, 95% CI 0.25 to 1.59) Analysis 1.5.

• The rate of stillbirth or neonatal mortality less than seven days (RR 0.35, 95% CI 0.04 to 3.22) Analysis 1.6.

Rates of premature delivery were similar in the two groups (two premature deliveries in the intervention group and three in the control group). "Perinatal morbidity" at the time of delivery was also reported although this was not defined (with one event in the intervention group and three in the control group).

Primary outcomes

The ECV group had 44% non‐cephalic presentation at birth compared to 74% in the no ECV group (RR 0.59, 95% CI 0.45 to 0.77); this difference between groups was statistically significant Analysis 2.1 . Women in the ECV group were at reduced risk of failing to achieve cephalic vaginal birth (RR 0.62, 95% CI 0.49 to 0.80) Analysis 2.2. The rate of caesarean section delivery was 11% in the ECV group compared to 14% in the no ECV group (RR 0.62, 95% CI 0.27 to 1.43) Analysis 2.3. The frequency of vaginal breech delivery was reduced in the ECV group (RR 0.63, 95% CI 0.46 to 0.85) Analysis 2.4.

Secondary outcomes

There was insufficient information on other outcomes.

• The rate of five minute Apgar scores less than seven (one event in the intervention group) Analysis 2.5.

• The rate of stillbirth or neonatal mortality less than seven days (one event in the control group) Analysis 2.6.

The authors reported no "major complications" in either group.

Primary outcomes

The rate of non‐cephalic presentation at birth was lower when ECV was started before term (RR 0.81, 95% CI 0.74 to 0.90; participants = 1906; studies = three; I² = 0%, evidence graded high quality) Analysis 3.1. Women who had early ECV were at slightly less risk of failing to achieve a cephalic vaginal birth (RR 0.90, 95% CI 0.83 to 0.97; participants = 1888; studies = three; I² = 0%, evidence graded high quality) Analysis 3.2. The rate of caesarean section was reduced when ECV was started before 37 weeks' gestation although the difference between groups did not reach statistic significance (RR 0.92, 95% CI 0.85 to 1.00; participants = 1888; studies = three; I² = 0%, evidence graded high quality) Analysis 3.3. Women who were randomised to early ECV were at a considerably reduced risk of having a vaginal breech birth; the difference between groups for this outcome was statistically significant (RR 0.44, 95% CI 0.25 to 0.78; participants = 1888; studies = three; I² = 0%, evidence graded high quality) Analysis 3.4.

Secondary outcomes

The rate of preterm birth less than 37 weeks was increased in the early ECV group (RR 1.51, 95% CI 1.03 to 2.21; participants = 1888; studies = three; I² = 0%, evidence graded high quality) Analysis 3.7.

There was no strong evidence of differences between groups identified for:

• the rate of five minute Apgar scores less than seven (RR 1.16, 95% CI 0.39 to 3.44; participants = 1759; studies = two; I² = 0%, evidence graded low quality due to imprecision) Analysis 3.5;

• the rate of stillbirth or neonatal mortality less than seven days (RR 0.23, 95% CI 0.04 to 1.34; participants = 1887; studies = three; I² = 0%, evidence graded low quality due to imprecision) Analysis 3.6;

• The studies by Hutton 2003 and Hutton 2011 reported several outcomes relating to neonatal outcome but these were not mutually exclusive and so a single composite outcome was reported: one or more serious fetal complication (RR 0.87, 95% CI 0.42 to 1.79; participants = 1761; studies = two; I² = 0%) Analysis 3.8.

• There no clear difference between groups for NICU stay for four days or longer (RR 2.50, 95% CI 0.49 to 12.63; participants = 232; studies = 1) Analysis 3.9.

Non‐prespecified outcome

One study reported maternal pain scores following the ECV attempt; pain scores were lower when ECV was commenced before term (mean difference (MD) ‐4.60, 95% CI ‐7.74 to ‐1.46; participants = 1533).

Seesummary of findings Table for the main comparison.