See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/221833441

Cervical cerclage: A review of current evidence

Article in Australian and New Zealand Journal of Obstetrics and Gynaecology February 2012
DOI: 10.1111/j.1479-828X.2012.01412.x Source: PubMed

CITATIONS READS

14 604

3 authors, including:

Andrew Shennan
King's College London
511 PUBLICATIONS 9,187 CITATIONS

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

EQUIPP View project

CLIP: Community Level Interventions for Pre-eclampsia View project

All content following this page was uploaded by Andrew Shennan on 18 February 2014.

The user has requested enhancement of the downloaded file.

Australian and New Zealand Journal of Obstetrics and Gynaecology 2012; 52: 220223 DOI: 10.1111/j.1479-828X.2012.01412.x

Review Article

Cervical cerclage: A review of current evidence

Danielle ABBOTT1, Meekai TO2 and Andrew SHENNAN1
1
Division of Womens Health, Kings College London, Womens Health Academic Centre KHP, and 2Harris Birthright Research Centre
for Fetal Medicine, Kings College Hospital, London, UK

Cervical cerclage is commonly used in the management of women considered to be at high risk of second-trimester loss
and spontaneous preterm birth. Insertion is dictated by factors such as multiple pregnancy, uterine anomalies, a history of
cervical trauma through destructive procedures or forced dilatation, and cervical shortening seen on transvaginal
ultrasound examination. However, its use and efficacy in these different groups is highly controversial as there is
contradiction in the results of individual studies and meta-analyses. This review examines the contemporary evidence on
cervical cerclage and its current role in obstetrics.
Key words: cerclage, cervical suture, cervix, preterm.

Introduction pathophysiology, for example, prior cervical surgery, or

ultrasound indicated. This review will address the
Preterm birth is the leading cause of neonatal morbidity indications for cerclage, including the controversies
and mortality. In Europe and other developed countries, surrounding this subject and the latest evidence base. The
reported preterm birth rates are around 59% and this Cochrane Library, DARE, EMBASE, TRIP, Medline and
appears to have increased in some countries.1 The PubMed (electronic databases) were searched for relevant
corresponding healthcare costs to the public sector are randomised controlled trials, systematic reviews and meta-
significant, estimated at $4.5 billion annually in the analyses from 1980 to 2009 to generate this review.
United States.2 Few interventions have improved
outcome, and management remains an important
challenge in modern obstetrics. Cervical cerclage is a History-Indicated Cerclage
common prophylactic intervention performed by A history-indicated cerclage is inserted as a result of risk
obstetricians in the management of women with preterm factors related to a womans obstetric or gynaecological
deliveries or second-trimester losses, despite the lack of a history. It is inserted as a prophylactic measure in
well-defined population for whom there is clear evidence asymptomatic women, usually at 1214 weeks gestation,
of benefit. The efficacy of cerclage is uncertain, with an preferably following assessment of viability and
estimated beneficial effect in one in 25 women.3 There is chromosomal risk. The largest randomised controlled trial
little consensus on the optimal procedure or technique (eg comparing history-indicated cerclage with expectant
low/high vaginal, abdominal, tape/nylon, single/multiple, management, coordinated by the Medical Research
endocervical/purse string) or timing of insertion (elective, Council and RCOG,3 was an international multicentre
ultrasound indicated, pre-conceptual). Furthermore, the trial which recruited 1292 women whose obstetrician was
mechanism of action is not understood; cerclage may uncertain as to whether a cerclage would be of benefit. In
offer a degree of structural support, but also plays a role practice, this equated to only 71% of the study population
in maintaining a biochemical barrier protecting having a history of second-trimester loss or preterm birth
membranes against exposure to ascending pathogens. before 37 weeks. However, 28% of this population
Mechanisms may be different depending on underlying delivered preterm, suggesting they were a high-risk group
suitable for intervention. Overall there were 25% fewer
deliveries prior to 33 weeks in the cerclage group
Correspondence: Danielle Abbott, Womens Academic Health
compared with the controls, but even without cerclage,
Centre, Kings College London and Kings Health Partners,
St Thomas Hospital, 10th Floor North Wing, Westminster 83% of women delivered after this primary endpoint. This
Bridge Road, London SE1 7EH, UK. equates to the prevention of one delivery before 33 weeks
Email: [email protected] for every 25 cerclage insertions, but could be as few as 12
or as many as 300, given the 95% confidence limits. No
Received 12 October 2011; accepted 25 December 2011. other outcomes, including perinatal death, favoured

220 2012 The Authors

ANZJOG 2012 The Royal Australian and New Zealand College of Obstetricians and Gynaecologists

The Australian and

New Zealand Journal
of Obstetrics and
Gynaecology
 Cervical cerclage review

cerclage, and some outcomes, including infection, were results were reported from a meta-analysis that included
worse in the cerclage group. The effect was only 607 pregnancies from four RCTs of ultrasound-indicated
significant in women with three prior losses, where their cerclage.9 This discrepancy of ultrasound-indicated
risk of preterm birth reduced by more than half. cerclage benefit between high- and low-risk women is
In the United States, one in 300 women receive a difficult to explain. However, there was a beneficial trend
cerclage,4 but this is higher for multiple pregnancies, in the low-risk trial (ie 22 vs 26%), which is consistent
reaching 10% of triplets. In other populations, rates are with the benefit in the high-risk trials, and may represent
high and increasing, usually over 1%.5 In the UK, a type 2 statistical error.
therefore, it is likely that more than 5000 women undergo There have been no studies evaluating ultrasound-
an unnecessary procedure and <300 women benefit. Even indicated cerclage performed solely on the presence of
this small benefit disappears when these data are meta- funnelling. However, whilst it appears that funnelling is a
analysed with two smaller trials addressing the same function of cervical shortening, it may not independently
issue.6 The evidence that a history-indicated cerclage is of add to the risk of preterm birth associated with cervical
benefit to women with three or more previous preterm length.10,11
births may justify its use in this subgroup of high-risk
women. However, the desire for the clinician to help
coupled with the desperate need of the woman to receive
Multiple Pregnancy
a beneficial intervention understandably leads to non- The insertion of a history- or ultrasound-indicated
evidence-based practice. cerclage in multiple pregnancies is not advisable, as there
is some evidence to suggest that it may be detrimental
and is associated with an increase in preterm delivery and
Ultrasound-Indicated Cerclage
pregnancy loss. In a meta-analysis, subgroup examination
In an attempt to better target women who will genuinely of 39 twin pregnancies demonstrated a doubling in
benefit from cerclage, ultrasound-indicated procedures delivery before 35 weeks with the use of ultrasound-
were introduced. This involves the insertion of cerclage as indicated cerclage when compared with expectant
a therapeutic measure in the presence of cervical length management in pregnancies with a cervical length
shortening (a potent predictor of preterm birth). <25 mm (RR 2.15; 95% CI 1.154.01).4 There is only
Detection is by transvaginal ultrasound, usually in one RCT of history-indicated cerclage in twin
asymptomatic women before 24 weeks gestation. To et al. pregnancies, which examined the effect of cerclage
screened 47 123 women at 2224 weeks gestation using (N = 25) versus no cerclage (N = 23) in twins conceived
transvaginal ultrasound. In 470 women (1%), the cervix following ovulation induction. This study demonstrated
was 15 mm or less. Of these women, 54% agreed to that cerclage was not effective in prolonging gestation or
participate in a randomised study comparing Shirodkar improving fetal outcome, but has limited power.12
cerclage (n = 127) to expectant management (n = 126). The existing published studies are either inadequately
The incidence of preterm delivery before 33 weeks was controlled or include insufficient numbers to be able to
similar in both groups, 22% in the cerclage group versus make evidence-based recommendations on the role of
26% in the control, with no significant differences in cerclage in other high-risk groups such as women with
perinatal or maternal morbidity or mortality.7 Thus, the Mullerian anomalies, previous cervical surgery (cone
insertion of an ultrasound-indicated cerclage is not biopsy, large loop excision of the transformation zone).
deemed beneficial in women who have an incidentally
identified short cervix, in the absence of a history of
Rescue Cerclage
spontaneous preterm delivery or second-trimester loss.
The role of cerclage in women with other risk factors There has been one RCT evaluating rescue cerclage and
such as previous cervical surgery, with or without a short bed rest against bed rest alone.13 This trial included only
cervix, is not clear. 23 women (16 singleton and seven twins) who were
Prior risk may determine who will benefit from confirmed to have cervical dilatation and prolapse of the
ultrasound-indicated cerclage. In an RCT involving 302 membranes on speculum examination at a mean gestation
singleton pregnancies with a history of spontaneous of 2223 weeks. No information was given on degree of
preterm birth and a cervical length of <25 mm (16+0 and cervical dilatation. All women, irrespective of random
21+6 weeks), cerclage reduced pre-viable birth allocation, were hospitalised and on bed rest until
(<24+0 weeks) (6.1 vs 14%, P = 0.03). Perinatal death 30 weeks gestation and received 1 week of broad-
was reduced by about a half (8.8 vs 16%, P = 0.046), spectrum antibiotics. In addition, those undergoing
when secondary perinatal outcomes were analysed. cerclage received peri-operative indomethacin. Women in
However, the reduction in birth <35 weeks (which was the cerclage group delivered on average 4 weeks later than
the primary outcome, 32 vs 42%; OR = 0.67, CI 0.42 those in the bed rest group (mean interval between
1.07) did not reach significance, although a secondary randomisation and delivery 54 vs 20 days), and there was
analysis when cervical lengths were <15 mm showed a a significant reduction in delivery before 34 weeks (53 vs
significant and substantial reduction (P = 0.006).8 Similar 100%, P = 0.02). There was a trend towards

2012 The Authors 221

ANZJOG 2012 The Royal Australian and New Zealand College of Obstetricians and Gynaecologists
D. Abbott et al.

improvement in neonatal survival (56 vs 28%) and a comparing the effectiveness of transabdominal cerclage
significant reduction in compound neonatal morbidity with expectant management or with transvaginal cerclage;
(defined as neonatal admission to intensive care unit and/ and this is certainly an area which would benefit from
or neonatal death) (71 vs 100%, RR 1.6, 95% CI 1.1 further research. A systematic review reported a lower risk
2.3). However, the authors did not describe the incidence of perinatal death/delivery before 24 weeks (6 vs 12.5%)
of chorioamnionitis or neonatal morbidity. Therefore, it in those women who had undergone transabdominal
appears that insertion of a rescue cerclage may delay cerclage (N = 117) compared with those who had a
delivery when compared with expectant management/bed repeat insertion of transvaginal cerclage (N = 40).18
rest alone. It may also be associated with a twofold However, there was a higher incidence (3.4% vs none) of
reduction in the chance of delivery before 34 weeks, but serious operative complications (bleeding requiring
there are only limited data to support an associated transfusion, injury to bladder/bowel/uterine artery,
improvement in neonatal mortality or morbidity. More anaesthetic problems (4/117). A non-randomised study in
research is needed. women with a prior failed transvaginal cerclage reported
Currently, there is insufficient evidence to recommend that the incidence of delivery before 33 weeks was lower
routine amniocentesis or genital tract screening prior to in the 40 women with a transabdominal suture when
rescue or ultrasound-indicated cerclage as there are no compared with 24 women with a transvaginal suture
clear data demonstrating that it improves outcome. insertion (10 vs 38%, P = 0.01).19 Consequently, with a
lack of compelling evidence to convince clinicians of its
superiority and the small, albeit significant risk of
Cerclage Technique morbidity associated, transabdominal cerclage should be
There is little consensus on the optimal cerclage technique. reserved for judicious use in the appropriate patient.
A transvaginal cerclage (MacDonald) is a purse-string Laparoscopic insertion of abdominal cerclage has been
suture placed at the cervicovaginal junction, without described placing a 5-mm Mersilene band at the level of
bladder mobilisation. A high transvaginal cerclage the cervical isthmus, yet choice of material, number of
(Shirodkar) is usually a transvaginal purse-string suture sutures or techniques are not evidence based and are
placed following bladder mobilisation, to allow insertion largely dictated by personal experience.20 Furthermore,
above the level of the cardinal ligaments. It may require there are no published data at present to suggest a
regional anaesthesia for removal at 37 weeks, given that the laparoscopic approach is superior, in terms of efficacy, to
technique often involves burial of the suture. In a secondary laparotomy for an abdominal cerclage.
analysis of singleton pregnancy data from four randomised Pre-conceptual insertion is becoming a more popular
trials of cervical cerclage in women with a short cervix, alternative when possible because of the technical
there was no significant difference in the rate of delivery advantage of operating on a non-pregnant uterus. There
before 33 weeks in those with a McDonald cerclage is no evidence to date that pre-conceptual transabdominal
compared with those with a Shirodkar suture, once cerclage has any detrimental impact on fertility or
adjusted for confounding factors (OR 0.55; 95% CI management of early miscarriage.
0.21.3).14 However, this trial, like several others in this
area,15 was underpowered to detect a statistically significant
Cervical Occlusion
difference in this outcome. There is only one study that
looked at suture material for cerclage, and this found no Total cervical occlusion involves closure at the external os
difference in spontaneous preterm birth rate when during the first trimester, in addition to the primary
Mersilene, Tevdek and Prolene were compared.16 cerclage, with a hypothesised action of preserving the
Consequently, the choice of transvaginal cerclage technique mucus plug to maintain an immunological barrier against
(ie Shirodkar vs McDonald) and suture material is usually vaginal pathogens. Whilst anecdotal evidence may be
at the discretion of the surgeon. The authors are aware that optimistic, there have been no controlled studies on the
certain practitioners may insert vaginal cerclage in high- use of a cervical occlusion suture in addition to the
risk women during the pre-conception period. The primary cerclage, and this is the subject of an ongoing
disadvantage of this is removing the suture in the event of randomised trial.21 The authors of this trial have reported
aneuploidy or other abnormalities that require termination. on-line that the occlusion suture does not add significantly
This should be weighed up against the theoretical to the efficacy of the standard suture (www.
possibility that an early or pre-conception suture may be cervicalocclusion.com/page001.html).
more effective, and more studies are needed.
In 1965, Benson and Durfee described the placement
of a cerclage at the cervical isthmus, positioned in the
Fetal Fibronectin
avascular space above the cardinal and uterosacral There is now substantial evidence of the benefit of fetal
ligaments following laparotomy, known as transabdominal fibronectin (fFN) tests in the prediction of preterm birth
cerclage.17 A transabdominal cerclage is usually inserted independent of cervical length. There are recent data
following a failed vaginal cerclage or extensive cervical confirming the utility of fFN in the presence of cerclage;
surgery. There are at present no randomised studies whilst there is a higher false-positive rate in these women,

222 2012 The Authors

ANZJOG 2012 The Royal Australian and New Zealand College of Obstetricians and Gynaecologists
 Cervical cerclage review

the high negative prediction is maintained and it remains a cerclage for prevention of preterm delivery in women with
valid and reassuring test in this high-risk group of women.22 short cervix: randomised controlled trial. Lancet 2004; 363:
18491853
8 Owen J, Hankins G, Iams JD et al. Multicenter randomized
Progesterone trial of cerclage for preterm birth prevention in high-risk
The impact of progesterone in the presence of cerclage women with shortened midtrimester cervical length. Am J
and whether it independently affects risk of preterm birth Obstet Gynecol 2009; 201: 375 e1375 e8.
9 Berghella V, Odibo AO, To MS et al. Cerclage for short
are not well described. A secondary analysis of 58 women
cervix on ultrasonography: meta-analysis of trials using
receiving 17a-hydroxyprogesterone caproate with
individual patient-level data. Obstet Gynecol 2005; 106: 181
ultrasound-indicated cerclage did not point to a beneficial
189.
effect of progesterone in the presence of cerclage.23 This
10 Owen J, Yost N, Berghella V et al. Mid-trimester endovaginal
question remains to be addressed by future research. sonography in women at high risk for spontaneous preterm
birth. JAMA 2001; 286: 13401348.
Comment 11 To MS, Skentou C, Liao AW et al. Cervical length and
funneling at 23 weeks of gestation in the prediction of
In summary, clinicians must rely on evidence-based spontaneous early preterm delivery. Ultrasound Obstet Gynecol
practice wherever possible when faced with decisions 2001; 18: 200203.
involving cerclage. It is difficult not to act in women with 12 Dor J, Shalev J, Mashiach S et al. Elective cervical suture of
adverse histories, but most women will have a successful twin pregnancies diagnosed ultrasonically in the first trimester
pregnancy regardless of intervention. This may propagate following induced ovulation. Gynecol Obstet Invest 1982; 13:
unnecessary surgery, sometimes with risk, when surgery is 5560.
performed without clear evidence of benefit. We must 13 Althuisius SM, Dekker GA, Hummel P, van Geijn HP.
continue to further our knowledge on this complex Cervical incompetence prevention randomized cerclage trial:
subject through robust research trials. All clinicians should emergency cerclage with bed rest versus bed rest alone. Am J
be brave enough to take part. Obstet Gynecol 2003; 189: 907910.
14 Odibo AO, Berghella V, To MS et al. Shirodkar versus
McDonald cerclage for the prevention of preterm birth in
Acknowledgements women with short cervical length. Am J Perinatol 2007; 24:
5560.
Professor Andrew Shennan is the Principal Investigator of
15 Harger JH. Comparison of success and morbidity in cervical
the MAVRIC trial, which is comparing abdominal with
cerclage procedures. Obstet Gynecol 1980; 56: 543548.
vaginal cerclage and funded by Moulton Charitable
16 Pereira L, Levy C, Rust O, Berghella V. Effect of suture
Foundation.
material on the outcome of McDonald cerclage in singleton
pregnancies. Obstet Gynecol 2005; 105: 33S.
References 17 Benson RC, Durfee RB. Transabdominal cervicouterine
cerclage during pregnancy for the treatment of cervical
1 Lawn JE, Gravett MG, Nunes TM et al. GAPPS Review
incompetency. Obstet Gynecol 1965; 25: 145155.
Group. Global report on preterm birth and stillbirth (1 of 7):
18 Zaveri V, Aghajafari F, Amankwah K, Hannah M.
definitions, description of the burden and opportunities to
Abdominal versus vaginal cerclage after a failed transvaginal
improve data. BMC Pregnancy Childbirth 2010; 10 (Suppl 1):
cerclage: a systematic review. Am J Obstet Gynecol 2002;
S1.
187: 868872.
2 Mangham LJ, Petrou S, Doyle LW et al. The cost of preterm
19 Davis G, Berghella V, Talucci M, Wapner RJ. Patients with a
birth throughout childhood in England and Wales. Pediatrics
prior failed transvaginal cerclage: a comparison of obstetric
2009; 123: e312e327.
outcomes with either transabdominal or transvaginal cerclage.
3 Final report of the Medical Research Council/Royal College
Am J Obstet Gynecol 2000; 183: 836839.
of Obstetricians and Gynaecologists multicentre randomised
20 Lesser KB, Childers JM, Surwit EA. Transabdominal
trial of cervical cerclage. MRC/RCOG Working Party on
cerclage: a laparoscopic approach. Obstet Gynecol 1998; 91(5
Cervical Cerclage. Br J Obstet Gynaecol 1993; 100: 516523.
Pt 2): 855856.
4 Menacker F, Martin J. National vital statistics reports.
21 Secher NJ, McCormack CD, Weber T et al. Cervical occlusion
Expanded Health Data From the New Birth Certificate 2005,
in women with cervical insufficiency: protocol for a
Vol. 56, Number 13. Hyattsville, MD: National Center For
randomised, controlled trial with cerclage, with and without
Health Statistics, 2008.
cervical occlusion. BJOG 2007; 114: 649, e1e6.
5 Al-Azemi M, Al-Qattan F, Omu A et al. Changing trends in
22 Duhig KE, Chandiramani M, Seed PT et al. Fetal fibronectin
the obstetric indications for cervical cerclage. J Obstet
as a predictor of spontaneous preterm labour in asymptomatic
Gynaecol 2003; 23: 507511.
women with a cervical cerclage. BJOG 2009; 116: 799803.
6 Drakeley AJ, Roberts D, Alfirevic Z. Cervical stitch (cerclage)
23 Rafael TJ, Mackeen AD, Berghella V. The effect of 17a-
for preventing pregnancy loss in women. Cochrane Database
hydroxyprogesterone caproate on preterm birth in women
Syst Rev, 2003; 1: CD003253.
with an ultrasound-indicated cerclage. Am J Perinatol 2011;
7 To MS, Alfirevic Z, Heath VC et al. Fetal Medicine
28: 389394.
Foundation Second Trimester Screening Group. Cervical

2012 The Authors 223

ANZJOG 2012 The Royal Australian and New Zealand College of Obstetricians and Gynaecologists

View publication stats