Hong?Kong?Med?J??Vol?13?No?1?# February?2007?# www.hkmj.org
Department of Obstetrics and
Gynaecology, Tuen Mun Hospital, Tuen
Mun, Hong Kong
SPY Yong, MB, BS, MRCOG
Correspondence to: Dr SPY Yong
Hong Kong Med J 2007;13:40-5
Stephen PY Yong
Breech presentation and, less commonly, oblique and transverse lie occur in 3 to 4% of
pregnancies at term.1-3 Planned caesarean section is safer for the baby than planned vaginal
breech delivery for term breech foetuses in extended or flexed presentations.4 Compared to
vaginal breech delivery, caesarean section reduced perinatal mortality, late neonatal mortality,
and serious neonatal morbidity by two thirds.4 Hence, caesarean section is the preferred and
more commonly used mode of delivery for otherwise uncomplicated breech presentations at
term. However, caesarean section is associated with higher maternal morbidity and mortality as
well as financial costs and long-term complications than vaginal delivery per se.5
term. A meta-analysis of six randomised controlled trials has found it effective in reducing the
number of vaginal breech deliveries by 87% and caesarean sections by 64%.2 No significant
increase in foetal or maternal mortality or morbidity following ECV has been found, though
numbers may have been too small to reliably detect changes in perinatal morbidity or mortality.6
The American College of Obstetricians and Gynecologists7 and Royal College of Obstetricians
External cephalic version (ECV) is another option for foetuses with breech presentation at
Objective To assess the outcome of external cephalic version for routine
management of malpresenting foetuses at term.
Design Prospective observational study.
Setting Tertiary teaching hospital, Malaysia.
Patients From September 2003 to June 2004, a study involving 41 pregnant
women with malpresentation at term was undertaken. An external
cephalic version protocol was implemented. Data were collected
for identifying characteristics associated with success or failure of
external cephalic version.
Main outcome measures Maternal and foetal outcome measures including success rate of external
cephalic version, maternal and foetal complications, and characteristics
associated with success or failure; engagement of presenting part,
placental location, direction of version, attempts at version, use of
intravenous tocolytic agent, eventual mode of delivery, Apgar scores,
birth weights, and maternal satisfaction with the procedure.
Results Data were available for 38 women. External cephalic version was
successful in 63% of patients; the majority (75%) of whom achieved
a vaginal delivery. Multiparity (odds ratio=34.0; 95% confidence
interval, 0.67-1730) and high amniotic fluid index (4.9; 1.3-
18.2) were associated with successful external cephalic version.
Engagement of presenting part (odds ratio=0.0001; 95% confidence
interval, 0.00001-0.001) and a need to resort to backward somersault
(0.02; 0.00001-0.916) were associated with poor success rates.
Emergency caesarean section rate for foetal distress directly resulting
from external cephalic version was 8%, but there was no perinatal
or maternal adverse outcome. The majority (74%) of women were
satisfied with external cephalic version.
Conclusions External cephalic version has acceptable success rates. Multiparity,
liquor volume, engagement of presenting part, and the need for
backward somersault were strong predictors of outcome. External
cephalic version is relatively safe, simple to learn and perform,
and associated with maternal satisfaction. Modern obstetric units
should routinely offer the procedure.
Introducing external cephalic version in a Malaysian
Breech presentation; Obstetric surgical
procedures; Patient satisfaction;
Pregnancy outcome; Version, fetal
? Hong?Kong?Med?J??Vol?13?No?1?# February?2007?# www.hkmj.org?
and Gynaecologists1 recommend that ECV be offered to
all suitable women at term with breech presentation.
pregnant women having foetuses with breech presentation
at term were not offered ECV or were not made aware
of this option by their obstetric carers.8-10 This was also
noted in our hospital before this study was carried out,
with some consultants performing ECV occasionally. In
our setting most junior staff were not familiar with the
technique and had never performed it, which was similar
to findings from a recent survey in England.11
Despite these recommendations, recent surveys of
safety of ECV for the routine management of malpresenta-
tions (mainly breech) at term and assess factors that
influence success as well as women’s views about the
procedure in general. This study was also designed to
enable junior staff to learn how to perform ECV.
This study aimed to assess the effectiveness and
This was a prospective observational study conducted over
the period September 2003 and June 2004 inclusive.
Patients attending the antenatal clinic at the Tenku
Ampuan Afzan Hospital, Kuantan, Pahang, Malaysia
were recruited. External cephalic version was performed
at or after 37 weeks. The Research Centre of the Interna-
tional Islamic University of Malaysia including its Ethics
Committee approved and funded the study. An ECV
protocol was created and implemented, based on the
author’s prior experience in other units and from publish-
ed protocols and guidelines.7
ing foetuses at term were offered ECV. Malpresentation in-
cluded all forms of breech, oblique, and/or transverse lie.
Standard exclusion criteria were as proposed by the
American College of Obstetricians and Gynecologists7
All suitable patients with uncomplicated malpresent-
presentation and exclude contra-indications to ECV.
Informed consent was obtained after counselling each
patient about the diagnosis and risks of malpresentation,
the nature and risks of ECV, its timing, predicted success
rate (50%), and alternative options (elective caesarean
section or assisted vaginal breech delivery) if ECV
failed. The patient was admitted after fasting overnight,
intravenous access was secured, blood was typed and
screened, and operating theatre personnel were placed
on standby. A cardiotocogram (CTG) and an ultrasound
were performed, and if findings from these tests were
non-reassuring or revealed contra-indications to ECV, the
procedure was abandoned in favour of caesarean section.
If not, tocolysis using intravenous terbutaline (250 μg
diluted in 5 mL of normal saline infused over 30 seconds)
was given selectively to patients with a tense uterus. Blood
pressure and pulse were checked before and after ECV.
A detailed ultrasound scan helped confirm mal-
Myerscough.12 Forward somersault was tried first and
then backward somersault if version was difficult. If the
version did not occur within 15 minutes, the procedure
was abandoned. Ultrasound was used selectively for
cases requiring a pause during the version to check
on foetal heart rate. Otherwise ECV was completed in
one continuous torque without loss of momentum. The
procedure was also abandoned if ECV was: (i) causing
unbearable pain to the patient, (ii) could not be achieved
easily, or (iii) foetal bradycardia was noted.
The technique of ECV was as described by
negative individuals after ECV. Patients with a successful
version were discharged if they had a satisfactory CTG to
await natural labour or were offered a stabilising induc-
tion (if they had an unstable lie). Failed ECV patients were
offered either an emergency caesarean section or a trial
of assisted vaginal breech delivery (not chosen by any
patient). An emergency section was also performed for
any foetal or maternal complications resulting from ECV.
Anti-D immunoglobulin was given to Rhesus-
Analysis of data and outcome measures
Non-parametric tests were used for statistical analysis
Hong?Kong?Med?J??Vol?13?No?1?# February?2007?# www.hkmj.org
including the Mann-Whitney U test for descriptive data,
Pearson’s Chi squared and Fisher’s exact tests for univariate
analyses, and logistic regression for multivariate analysis
of factors associated with successful ECV. All statistical
analyses were performed using the Statistical Package
for the Social Sciences (Windows version 13; SPSS Inc,
Chicago [IL], US).
success rate of ECV, maternal and foetal complications,
and characteristics associated with success or failure in-
cluding maternal age, parity, amniotic fluid index (AFI),
engagement of presenting part, placental location, direc-
tion of version, attempts at version, use of intravenous toco-
lytic agent, eventual mode of delivery, Apgar scores, birth
weights, and maternal satisfaction with the procedure.
Maternal and foetal outcome measures included
There were 6570 deliveries during the study period.
Among these, 228 (3.5%) patients had malpresentations,
of which 41 (18%) consented to undergo ECV, but three
were excluded from the analysis due to irretrievable
data. Details regarding the numbers of women with
malpresentation who were offered ECV, trial of vaginal
breech delivery, or elective caesarean section were
unavailable. Of the 228 women, 177 (78%) delivered
by caesarean section, 33 (14%) had breech vaginal de-
liveries, and 18 (8%) had cephalic vaginal deliveries due
to successful ECV. Initially all ECVs were performed by
the author; subsequently 60% cases of ECV were done
by junior trainees who had mastered the technique after
witnessing or being supervised on about six cases. There
was no overall difference in success rates between the
author and the trainees once the latter had mastered the
technique. All trainees regarded ECV as relatively easy to
learn and practise.
Patient and antenatal characteristics
The median maternal age was 29 (interquartile range
[IQR], 25-33) years and median parity was 1 (IQR, 0-3).
Approximately two thirds of our patients were multipara.
All were beyond 37 weeks’ gestation with one post-dates
at 41 weeks and 3 days. The majority (92%) had breech
presentations including 20 that were extended, 14 flexed,
and one footling. Two (5%) were oblique breech lies and
one (3%) had a transverse lie. The majority of patients
had an unremarkable antenatal course. Two patients had
a history of bronchial asthma (in remission), and three
had gestational diabetes mellitus controlled by diet. One
patient had mild pregnancy-induced hypertension and
another had mild nutritional anaemia. All patients were
Rhesus blood group positive.
maternal age or birth weights in women with a successful
or failed attempt at ECV. For patients who failed ECV,
their AFI ranged from 8.3 to 13.7 cm; for patients who
successfully underwent ECV, their AFI ranged from 9.1
to 14.6 cm. Thus, those in whom it succeeded had a
higher median value than those in whom it failed (Table
1), though all 38 patients had normal AFIs (reference
range of AFI at term: 6.8-19.6 cm13) with a median of
11. Multiparity, non-engagement of the presenting
part, a fundal or posterior upper segment placenta, and
need for forward somersault alone were all significantly
associated with success of ECV. Conversely, the type of
malpresentation and number of ECV attempts were not
significant factors (Table 2).
There were no significant differences between
External cephalic versionP value*
Maternal age (years)29 (27-33)27 (24-30)0.118
Birth weight (kg)3 (2.9-3.4)2.9 (2.8-3.3)0.513
Amniotic fluid index (cm)12 (11-13) 10.3 (10-10.8)0.002
TABLE 1. Characteristics of patients in whom external cephalic version was successful
Mann-Whitney U test
† IQR denotes interquartile range
Chi squared test
† Fisher’s exact test
Anterior upper segment placenta5/15 (33)
Fundal or posterior upper
Forward only19/23 (83)
Backward and forward5/15 (33)
Not used9/9 (100)
Type of malpresentation0.410*
Extended breech12/20 (60)
Yes 23/28 (82)
TABLE 2. Factors associated with the success of external
? Hong?Kong?Med?J??Vol?13?No?1?# February?2007?# www.hkmj.org?
Effect of tocolysis
The number of women who received tocolysis was
29; they included 12/13 nullipara of which four (33%)
achieved successful ECV and 17/25 multipara of which
11 (65%) achieved success. Among the remaining
nullipara and multipara who did not receive tocolysis,
all nine (100%) achieved successful ECV. Thus, use
of tocolysis was significantly associated with failure
(P=0.014, Table 2).
statistically significant variables further singled out the
most significant factors to high AFI and multiparity,
which strongly predicted successful ECV. In contrast,
engagement of presenting part and need for backward
somersault strongly predicted failure (Table 3). Resort
to a backward in addition to a forward somersault was
associated with a 40% lower success rate. Nor did the
addition of tocolysis to this procedure have any obvious
impact. Backward somersault was not performed for
some patients for whom the forward manoeuvre failed,
because the operator felt it was not feasible or the patient
was unwilling to tolerate further pain due to version.
Multivariate logistic regression analyses of the
Maternal and foetal outcomes of external cephalic
Twenty-four (63%) of the 38 patients achieved
successful ECV, with lower success rates in nulliparas
than multiparas (38 vs 76%). The overall caesarean
section rate was 53% (20/38). Of those with successful
version, 75% (18/24) achieved a vaginal delivery. The
remaining six cases underwent caesarean section due to:
failure of labour to progress (n=2), cord prolapse (n=1),
and foetal distress (n=3; two with foetal bradycardia
lasting >5 minutes immediately after ECV and one with
intrapartum bradycardia). None of the patients in whom
ECV failed elected to undergo assisted vaginal breech
delivery though one (2.6%) actually achieved a vaginal
delivery following spontaneous cephalic version. There
was no reversion to malpresentation after a successful
ECV. Indications for caesarean section in those in whom
ECV failed were: persistent malpresentation (n=12) and
foetal distress (n=1) manifesting as foetal bradycardia
immediately after the procedure.
after a successful ECV which led to labour and vaginal
delivery. Seven patients complained of pain during the
One patient had spontaneous rupture of membranes
procedure, leading to abandonment in one. There was no
maternal or perinatal mortality. The numbers of male and
female infants were equal. All babies had normal Apgar
scores of >7 at 5 minutes. One infant was admitted to
the neonatal intensive care unit for suspected meconium
aspiration but was subsequently discharged uneventfully.
Twenty-eight (74%) women were satisfied with the
attempt at ECV and would choose it again in the future
if needed; in 24 the procedure succeeded. Two thirds
of the women who were satisfied with ECV went on to
achieve vaginal delivery. On the contrary, in nine of the
10 women dissatisfied with the ECV procedure, it had
failed and so they underwent caesarean section (Table 2).
The remaining patient was dissatisfied despite successful
ECV, she underwent emergency caesarean section for
foetal distress immediately after the procedure.
This small study was undertaken in a tertiary hospital in
Malaysia, to assess the feasibility and outcome of ECV as
part of the routine management of malpresenting foetuses
at term. The majority of our patients were of low socio-
economic status, of high parity (highest was para 7), of
Malay ethnicity, and considered to have relatively high
pain thresholds. In our hospital, assisted vaginal breech
delivery was still considered a reasonable mode of
delivery and offered as a management option for breech
presentation at term. Increasingly however, patients
were advised to undergo planned elective caesareans in
view of the term breech trial. Prior to this study, ECV was
seldom performed by senior staff and most of our trainees
had no experience with it. Initial recruitment of patients
was therefore suboptimal; only 18% of women with
malpresentation participated. However as preliminary
results were good, the procedure was gradually accepted
and increasingly offered to suitable women. Overseas
research has found that women’s uptake rate of ECV can
be improved by education of staff.14 Most of the results
from this study were consistent with experience in ECV
for term breech presentations obtained elsewhere in
South-East Asia and overseas.6,15-17
Our ECV success rate of 63% is similar to those
reported by others; quoted at approximately 50% with
a range of 35 to 86%.1,7,15-17 In keeping with others,2,14,18
multiparity, AFI, and non-engagement of the presenting
part were strongly associated with successful ECV. A
placenta in the anterior upper segment was associated
with a higher chance of failure than if it was in the
posterior upper segment or fundally, because in the
former situation the head is directly beneath it and
therefore less accessible for version.19 The favourable
ECV success rate we achieved could be attributed to
the fact that most patients were multiparous. It may
also reflect the importance of adhering to strict patient
Based on logistic regression analysis
Predictive factors Coefficient OR95% CI
Amniotic fluid index1.588 4.91.31-18.23
Engagement -12.450.0001 0.00001-0.001
Backward somersault -3.890.020.00001-0.916
TABLE 3. Factors predictive of success or failure of external
Hong?Kong?Med?J??Vol?13?No?1?# February?2007?# www.hkmj.org
selection criteria for the procedure. The use of ultrasound
facilitated ECV and is highly recommended.1
In contrast to other reports,1,6,20 in our study selective
use of tocolysis was significantly associated with failure.
In a few of our patients with tense uteri, administering a
tocolytic agent immediately converted impending failed
ECV to a successful version, but the proportion in whom
this occurred was small (52%). In contrast, all succeeded
in those who did not receive tocolysis (100%), the
difference being statistically significant. This apparently
paradoxical finding was probably due to a sampling
bias, as tocolysis was used selectively in women with
tense uteri, the majority of whom were nullipara. As
mentioned, nulliparity strongly predicts failure of ECV
in our study. Thus, using multiple logistic regression,
it was clear that tocolysis per se was not a significant
independent predictor of ECV success or failure.
If a forward somersault failed, a backward somer-
sault could be tried.7,12 The low success rate of a
backward somersault after a failed forward attempt is
understandable, as the need to resort to a backward flip
means that ECV was quite difficult to begin with. In other
words, easy ECV usually succeeds with the first (forward)
somersault, and this was also reflected in our logistic
Whereas 75% of patients having successful ECV
subsequently achieved vaginal delivery, 25% were de-
livered by caesarean, which was higher than the 18% an-
nual baseline caesarean section rate for our hospital. Our
results were similar to those of another study in Hong
Kong which reported vaginal delivery in 83% of women
having a successful ECV, and recourse to emergency
caesarean section in 17% of patients.16 Caesareans were
resorted to mainly for non-reassuring CTG findings or
poor progress, which was 2.25 fold higher than the base-
line rate.16 The higher rate of caesarean section after a
successful ECV is a recognised but unexplained phenom-
enon, very likely related to foetal and maternal factors.16,21
Regarding patients who failed ECV, all except
one were delivered by caesarean section. Repeating
ECV again at a later date after a failed first attempt,
increases the overall success rate by another 17%.14
However, this was not feasible in our setting because
many of our patients were from distant villages, such that
repeated travel to and from hospital would have been
unaffordable. Hence, if ECV failed they were offered
an emergency caesarean section or allowed to await
spontaneous labour and assisted vaginal breech delivery,
though none actually took up the latter option.
There were three (8%) instances of ECV-related
foetal distress, which is higher than the 0.37 to 1%
rates reported by others,16,22 but eventually all had
favourable neonatal outcomes. Two systematic reviews
recently found that the most frequently reported foetal
complication of ECV was a transiently abnormal CTG
pattern (ranging from 1-47% with a mean incidence of
5.7%).16,22 Transient foetal bradycardias usually last 5
minutes but can be as long as 1 hour.17 Arguably, if our
three foetuses in distress had been observed for longer
than 5 minutes, their heart rates may have recovered and
the need for caesarean section precluded.
reported in a recent review.17 However, we encountered
one such patient, which nevertheless resulted in a good
neonatal outcome following an emergency caesarean
section. In our series there were no significant perinatal
or maternal complications. Uncommon complications
reported in the literature are very rare and include:
foetomaternal haemorrhage (3.7%), vaginal bleeding
(0.5%), persisting pathological CTG readings (0.4%),
and placental abruption (0.1-0.4%).16,22 Therefore ECV
can be considered a safe procedure.1,2,7,14,16,17,22
No case of cord prolapse after an ECV was
with ECV and would have it again if needed. In most of
these women (86%) the procedure had succeeded, but
some in whom it had failed also held this view. In 90%
of those who were dissatisfied, the procedure had failed.
Thus, patient satisfaction with the procedure appeared
linked to having a good chance of achieving vaginal
delivery and avoiding caesarean section. In addition,
the opportunity to take an active part in management
decisions provided a sense of control and satisfaction,
even if ECV failed (as in four of our patients). Consistent
with our experience, a review of the literature on
maternal perception of ECV suggests that women would
likely be satisfied with it, so long as it was tolerable,
safe, efficacious, and associated with a reasonably good
chance of vaginal delivery.9,23,24
The majority (74%) of our patients were satisfied
size and missing records. Also, racial factors have been
shown to influence ECV success rates,14,18 which inevit-
ably limits the generalisability of our findings to other
populations. Unlike other published reports,2 this study
did not demonstrate a significant reduction in caesarean
section rates after ECV. This could be partially due to
the low (18%) recruitment rate, whether due to women
with malpresentations being undiagnosed, uninformed,
unsuitable, or unwilling. The other reason may have
been the perceived availability of vaginal breech delivery
in our unit. We estimated that vaginal breech delivery
alone would reduce the caesarean section rate by 15%,
which is greater than the 8% reduction associated with
ECV alone. Hence, any benefits of ECV in terms of
reduced caesarean section rates were more than offset
by vaginal breech deliveries. However, no data are
available regarding the outcome of neonates resulting
from vaginal breech delivery. Hence, a second clinical
audit is to be conducted to compare corresponding
outcomes to allow the full impact of a universal ECV
policy to be evaluated.
Major limitations of our study were the small sample
randomised trials on the management for transverse or
oblique lie.6 These latter cases were included in our
study because we assumed that ECV can be applied
Unlike for breech presentation, there are no
? Hong?Kong?Med?J??Vol?13?No?1?# February?2007?# www.hkmj.org?
to all suitable malpresentations. We encountered only
three such patients, all of whom had successful ECV, but
no conclusions can be drawn regarding this issue owing
to the very small numbers.
External cephalic version was successfully introduced
in a Malaysian hospital; its efficacy was comparable to
that in other countries. Multiparity and high AFI were
strong predictors of a success, whereas engagement
of the presenting part and the need for backward
somersault were strong predictors of a failure. It is a
relatively safe procedure, simple to learn and perform,
and it is associated with a high maternal satisfaction rate.
All modern obstetric units should offer ECV to suitable
women at term with malpresentation.
I wish to thank the following doctors from Hospital Tenku
Ampuan Afzan for their assistance: Drs Suhaidin Che Ngah
(drafting ECV protocol), Jalilah (data collection), and Ahmad
Murad Zainuddin (comments). I am also grateful to the
following staff from Tuen Mun hospital: Ms Willy Sung
(statistical analyses) and Dr Kai-bun Cheung (comments). I
also wish to thank the referees and the Editors of this Journal
for their valuable comments and suggestions. This study
was funded by the Research Centre of the International
Islamic University of Malaysia (IIUM/504/022/3/C 31).
1. Royal College of Obstetricians and Gynaecologists. The
management of breech presentation. Greentop guideline
No. 20. London: RCOG press; 2001:1-8.
2. Thorpe-Beeston JG. Management of breech presentation at
term. In: Studd JW, editor. Progress in obstetrics and gynae-
cology. Vol 13. Edinburgh: Churchill Livingstone; 1999:87-100.
3. Johanson R. Malposition, malpresentation and cephalopelvic
disproportion. In: Edmonds DK, editor. Dewhurst’s textbook
of obstetrics and gynaecology for postgraduates. 6th ed.
London: Blackwell Science; 1999:277-90.
4. 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.
5. van Roosmalen J, Rosendaal F. There is still room for
disagreement about vaginal delivery of breech infants at
term. BJOG 2002;109:967-9.
6. Hofmeyr GJ. External cephalic version facilitation for breech
presentation at term (Cochrane Review). In: The Cochrane
Library, Issue 2. Oxford: Update Software; 2002.
7. ACOG practice patterns. External cephalic version. Number
4, July 1997. American College of Obstetricians and
Gynecologists. Int J Gynaecol Obstet 1997;59:73-80.
8. 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.
9. Raynes-Greenow CH, Roberts CL, Barratt A, Brodrick B, Peat
B. Pregnant women’s preferences and knowledge of term
breech management, in an Australian setting. Midwifery
10. Coltart T, Edmonds DK, al-Mufti R. External cephalic
version at term: a survey of consultant obstetric practice in
the United Kingdom and Republic of Ireland. Br J Obstet
11. Edmonds S. Management of breech deliveries [Letter]. The
Obstetrician Gynaecologist 2002;4:239-43.
12. Myerscough P. The practice of external cephalic version. Br
J Obstet Gynaecol 1998;105:1043-5.
13. Moore TR, Cayle JE. The amniotic fluid index in normal
human pregnancy. Am J Obstet Gynecol 1990;162:1168-73.
14. Burr RW, Johanson RB. Breech presentation: is external
cephalic version worthwhile? In: J Studd, editor. Progress
in obstetrics and gynaecology. Vol 12. Edinburgh: Churchill
15. Devendra K. Introducing routine external cephalic version
for the management of the malpresenting fetus near term.
Med J Malaysia 2002;57:454-9.
16. Lau TK, Lo KW, Rogers M. Pregnancy outcome after
successful external cephalic version for breech presentation
at term. Am J Obstet Gynecol 1997;176:218-23.
17. Nassar N, Roberts CL, Barratt A, Bell JC, Olive EC, Peat B.
Systematic review of adverse outcomes of external cephalic
version and persisting breech presentation at term. Paediatr
Perinat Epidemiol 2006;20:163-71.
18. Lau TK, Lo KW, Wan D, Rogers MS. Predictors of successful
external cephalic version at term: a prospective study. Br J
Obstet Gynaecol 1997;104:798-802.
19. Guyer CH, Heard MJ. A prospective audit of external
cephalic version at term: are ultrasound parameters
predictive of outcome? J Obstet Gynaecol 2001;21:580-2.
20. Nor Azlin HI, Haliza H, Mahdy ZA, Anson I, Fahya MN, Jamil
MA. Tocolysis in term breech external cephalic version. Int J
Gynaecol Obstet 2005;88:5-8.
21. Chan LY, Leung TY, Fok WY, Chan LW, Lau TK. High
incidence of obstetric interventions after successful external
cephalic version. BJOG 2002;109:627-31.
22. 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.
23. Leung TY, Lau TK, Lo KW, Rogers MS. A survey of pregnant
women’s attitude towards breech delivery and external
cephalic version. Aust N Z J Obstet Gynaecol 2000;40:253-9.
24. Fok WY, Chan LW, Leung TY, Lau TK. Maternal experience
of pain during external cephalic version at term. Acta Obstet
Gynecol Scand 2005;84:748-51.