https://orcid.org/0000-0003-1804-1882
Figures
Abstract
Introduction
Spontaneous preterm birth (sPTB) occurs in 0.5–1% of pregnant women and is commonly attributed to cervical insufficiency. Cervical cerclage can reduce the rate of spontaneous preterm birth in high-risk women with a shortened cervix. Management options when primary cerclage fails are uncertain. This review aims to synthesise the evidence for repeat cervical cerclage in the same pregnancy following primary cerclage failure, to understand outcomes and aid decision making for patients and clinicians.
Materials and methods
Databases were searched according to a prospective protocol registered with PROSPERO (CRD42025638147). Included studies reported outcomes for pregnant women with a cervical cerclage in situ that failed and compared repeat cervical cerclage with expectant management. The primary outcome was a composite outcome of pregnancy loss; to include miscarriage, stillbirth, neonatal death and termination of pregnancy. Secondary outcomes included preterm birth less than 37 and less than 34 weeks, miscarriage and previable neonatal death less than 24 weeks, birthweight and gestational age at delivery. Random effects meta-analysis was performed using Review Manager software (RevMan) and risk of bias was assessed using the Robins-I tool.
Results
Database and citation searching retrieved 1006 titles and abstracts. There were 20 papers that underwent full text review. Six retrospective cohort studies met inclusion criteria for meta-analysis. There was no significant difference in pregnancy loss (odds ratio (OR) 1.65, 95% confidence interval (CI) 0.23–11.62), preterm birth less than 34 weeks (OR 1.11 95% CI 0.14–8.70) or preterm birth less than 37 weeks (OR 1.88 95% CI 0.74–4.80) for repeat cervical cerclage compared to expectant management, with a trend towards improved outcomes with expectant management.
Citation: Emms A, Vaughan M, Man R, Morris RK, Hodgetts-Morton V, Pilarksi N (2025) Should a repeat cervical cerclage be inserted when the primary cerclage fails, to prevent pregnancy loss and preterm birth? A systematic review and meta-analysis. PLoS One 20(11): e0329427. https://doi.org/10.1371/journal.pone.0329427
Editor: Burak Bayraktar, Ankara Etlik City Hospital, TÜRKIYE
Received: July 16, 2025; Accepted: October 20, 2025; Published: November 7, 2025
Copyright: © 2025 Emms et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting Information files.
Funding: The author(s) received no specific funding for this work.
Competing interests: The authors have declared that no competing interests exist.
1. Introduction
Prematurity is the leading cause of child mortality and morbidity worldwide, and accounts for approximately 10% of all births [1]. The aetiology of preterm birth (PTB) is multifactorial; however premature dilatation of the cervix, described as cervical insufficiency, weakness or incompetence, is commonly attributed as the primary causative factor in spontaneous preterm birth (sPTB). This occurs in 0.5–1% of pregnant women [2].
For decades, insertion of a cervical cerclage (either low vaginal, high vaginal or transabdominal) has been used as a management strategy to prevent this process. Cervical cerclage has been shown to reduce the rate of sPTB in women at high risk of preterm birth with a shortened cervix demonstrated on ultrasound [2–4]. However, the optimal management after primary cerclage failure in remains uncertain in women who are asymptomatic for threatened preterm labour. At present it is unclear whether a repeat cerclage should be performed if the woman is eligible for one.
There is no consensus nationally in the UK to define primary cerclage failure: either a short or progressively shortening cervix or prolapse of membranes beyond the level of the primary cerclage is generally used. The diagnosis is usually made by transvaginal cervical length ultrasound surveillance, or by physical speculum examination of the cervix. Once failure is diagnosed there is no guidance for the most appropriate management strategy for prevention of pregnancy loss and preterm birth [5,6]. Clarfield et al have published the largest study to date comparing repeat cervical cerclage with expectant management [7]. This study included 40 women who had reinforcing cerclage after ultrasound evidence of failure and 40 women who received expectant management. Their data showed no significant differences in pre-viable or preterm delivery, however those who had repeat cerclage had increased rates of placental infection and chorioamnionitis.
This review aims to synthesise the current body of evidence investigating outcomes of repeat cervical cerclage compared with expectant management, following asymptomatic primary cerclage failure. In addition, this review aims to draw conclusions that can guide clinical decision making when considering a repeat cerclage or not.
2. Methods
2.1 Study design and registration
This is a systematic review and meta-analysis comparing the pregnancy loss and preterm birth rates of repeat cervical cerclage with expectant management in women with asymptomatic failure of a primary cervical cerclage. It was conducted in accordance with the ‘Preferred Reporting Items for Systematic reviews and Meta-Analyses’ (PRISMA) guidelines 2009 [8,9]. Ethical approval was not required. It was prospectively registered with the International prospective register of systematic reviews (PROSPERO) prior to full data extraction and analyses. PROSPERO registration number: CRD42025638147 [10].
2.2 Inclusion and exclusion criteria
Given the rare incidence of this condition, and even rarer clinical scenario of considering a repeat cervical cerclage when the primary cerclage fails, all types of studies were included in the review. These included: randomised controlled trials, quasi-randomised controlled trials, cohort studies, case-control studies and case series where five or more cases were reported. Systematic reviews were excluded. Excluded from the meta-analysis were papers reporting outcomes for patients who received repeat cervical cerclage after primary cerclage failure but did not include a comparator group receiving expectant management. These papers were included in the qualitative overview only.
2.3 Outcomes
The primary outcome was pregnancy loss. This is a composite outcome that encompasses: miscarriage or previable neonatal death (before 24 weeks gestation), stillbirth (intrauterine death after 24 weeks gestation), termination of pregnancy (medically indicated termination due to prematurity and its complications, or consequence of cerclage insertion) and early neonatal death (up to seven days post-delivery). This was chosen as our primary outcome because of the importance for patients and clinical significance.
Our pre-defined secondary outcomes encompassed a wide range of both maternal and neonatal outcomes, including those detailed in the core outcome set for interventions to prevent preterm birth, published by van Hooft et al. [11] These are listed in full in Appendix 1. Included studies reported: gestational age at delivery including less than 28 weeks, less than 32 weeks, less than 34 weeks and less than 37 weeks in live births more than 24 weeks gestation, antenatal steroid use, preterm pre-labour rupture of membranes (PPROM) and chorioamnionitis, early and late neonatal death, birthweight in grams, neonatal unit admission following live birth and neonatal unit length of stay and whether respiratory support was required.
2.4 Search strategy
A systematic search of the four main databases, MEDLINE, Embase, CINAHL and Cochrane, from inception (1946, 1974, 1981 and 1996 respectively) to 13th January 2025 was undertaken. Keywords and variants of “pregnancy”, “cervical cerclage” and “reoperation” were used, including exploded medical subject headings (MeSH), in addition to searches within the paper title, abstract and keyword fields. An example of the full search strategy is detailed in Appendix 2, which was adapted to each database searched. There were no restrictions on language, country of origin, publication date, study design or follow-up period. Apart from conference abstracts, no other grey literature was included in the searches.
2.5 Study selection and data extraction
Two reviewers independently reviewed the titles and abstracts in the retrieved papers (AE and MV) and any conflicts were resolved by a third independent reviewer (NP). This process was repeated for full text screening and data extraction. Conflicts with data extraction were resolved by the third reviewer for three outcomes, two outcomes in the same paper, and one outcome in a second paper. Covidence software was used for study screening. Data from the studies that met inclusion criteria were extracted using a piloted data entry form and then transferred onto an electronic spreadsheet.
2.6 Risk of bias assessment
Two review authors independently assessed risk of bias for each included study (AE and MV) with conflicts resolved by a third reviewer (NP). Risk of bias was assessed using the ROBINS-I tool for non-randomised studies, reported at individual domain level [12].
2.7 GRADE certainty of evidence assessment
Certainty of evidence was assessed using the Grading of Recommendations Assessment, Development and Evaluation tool (GRADE) [13,14].
2.8 Publication bias
Funnel plots were not used to assess possible publication bias as there were only six studies included in the meta-analysis [15].
2.9 Data analysis
Data was synthesised using Cochrane Review Manager software (RevMan Web version 9.3.0). For assessment of dichotomous outcomes, odds ratios (ORs) were calculated, with 95% confidence intervals. Means and standard deviations were extracted for continuous outcomes. Where continuous outcomes were expressed as the median and interquartile range, these were converted to mean and standard deviation estimations according to Wan et al’s method [16]. A random effects model meta-analysis was performed, using the Mantel-Haenszel method for dichotomous outcomes, and the DerSimonian and Laird method for continuous outcomes. Heterogeneity in the estimates of effect was assessed using I2. As per the Cochrane Handbook for Systematic Reviews of Interventions, these categories were used to define the degree of statistical heterogeneity: 0% to 40%: might not be important, 30% to 60%: may represent moderate heterogeneity, 50% to 90%: may represent substantial heterogeneity, and 75% to 100%: considerable heterogeneity [17]. A sensitivity analysis including only studies at low or moderate risk of bias was planned provided the number of eligible studies was adequate.
3. Results
3.1 Included studies
Following database searches, and three papers identified via citation searching, 1006 titles and abstracts were screened. Twenty papers were retrieved for full text review. Six retrospective cohort studies were eligible for inclusion in the meta-analysis [7,18–22]. including 238 participants. Fig 1 demonstrates the search and selection process generated using Covidence software (aligned to PRISMA 2020). Conference abstracts were eligible for inclusion if they reported the primary outcome for both comparator groups, however none of the abstracts identified met eligibility criteria. There were no unpublished data identified that were eligible for inclusion. There were no non-English language studies eligible for inclusion.
Characteristics of included studies are demonstrated in Table 1. Two studies were conducted in the United States of America (USA) (Baxter et al, Contag et al), one in Canada (Clarfield et al), one in the United Kingdom (UK) (Simcox et al), one in Korea (Song et al), one in China (Tang et al). Three studies defined primary cerclage failure as a short cervix less than 25 mm in length (Baxter, Clarfield and Contag et al [7,18,19]) and three defined failure as prolapse of fetal membranes through the level of the cerclage, either diagnosed by transvaginal ultrasound or by physical examination (Simcox, Song and Tang et al [20–22]). Studies used varying combinations of indications for primary cervical cerclage: either history, ultrasound-indicated or emergency cerclage and included various surgical techniques: low vaginal, high vaginal or transabdominal. One study, Song et al, included one set of twins in each study group. [21] All of Tang et al’s participants were twin pregnancies. [22] The remainder of the studies excluded multiple pregnancies. There was limited and variable reporting of baseline characteristics between the studies (Supplementary Fle 1).
3.2 Risk of bias
Risk of bias was assessed using the Risk of Bias in Non-randomised Studies – of Interventions (ROBINS-I) tool for observational studies. Appendix 3 demonstrates the quality of each study according to individual ROBINS-I domain assessment. None of the studies were deemed at low risk of bias for confounding. Overall, two studies were deemed at moderate risk of bias (Contag, Tang [19,22]), two studies at serious risk of bias (Baxter, Song [18,21]) and two studies at critical risk of bias (Clarfield, Simcox [7,20]). Due to the limited available evidence, all six studies were included in the meta-analysis including those deemed to be at high risk of bias.
3.3 GRADE certainty of evidence assessment
The level of certainty of evidence was assessed using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) assessment tool and applied to outcomes with three or more studies included in the meta-analysis: pregnancy loss as composite, preterm birth less than 37 weeks gestation, preterm birth less than 34 weeks gestation, miscarriage and previable neonatal death less than 24 weeks gestation, birthweight and gestational age at delivery. All outcomes were graded at very low certainty of evidence, except from miscarriage and previable neonatal death less than 24 weeks which was graded at low certainty. The full GRADE assessments with explanations for the grading are displayed in Table 2. The studies were deemed at either moderate, serious or critical risk of bias, and there was considerable heterogeneity in the estimates of effect with all outcomes except miscarriage and previable neonatal death less than 24 weeks. [17] The variation between study populations (singleton pregnancies, multiple pregnancies or both singletons and multiples and the indication for primary cerclage) contributed to downgrading for indirectness.
3. Data analysis
Meta-analyses were performed using RevMan software for each outcome that was reported by three or more studies: pregnancy loss as composite, preterm birth less than 37 weeks, preterm birth less than 34 weeks, miscarriage and previable neonatal death less than 24 weeks, birthweight in grams and gestational age at delivery in weeks. A summary of the meta-analyses are presented in Table 3.
For pregnancy loss as composite, including four studies with 153 participants, [7,18,20,21] the odds ratio (OR) was 1.65 with a 95% confidence interval (CI) of 0.23–11.62 and a p-value of 0.62. This demonstrates no effect on pregnancy loss rate with or without repeat cervical cerclage when the primary cerclage fails (Fig 2).
Total number of participants used as denominator. Created using RevMan. Abbreviations: M-H = Mantel-Haenszel, CI = confidence interval.
For preterm birth less than 37 weeks, including three studies with 103 participants, [19,20,22], the OR was 1.60 with a 95% CI of 0.28–9.21 and a p-value of 0.60. This demonstrates no effect on preterm birth less than 37 weeks with or without repeat cervical cerclage (Fig 3).
Live birth rate used as denominator. Created using RevMan. Abbreviations: M-H = Mantel-Haenszel, CI = confidence interval.
For preterm birth less than 34 weeks, including three studies with 103 participants, [19,20,22], the OR was 1.11 with a 95% CI of 0.14–8.70 and a p-value of 0.92. This demonstrates no effect on preterm birth less than 34 weeks with or without repeat cervical cerclage (Fig 4).
Live birth rate used as denominator.
For miscarriage and previable neonatal death less than 24 weeks, including three studies with 129 participants, [7,18,20], the OR was 4.27 with a 95% CI of 1.86–9.78 and a p-value of 0.0006 (Fig 5). This demonstrates a statistically significant association of repeat cerclage with increased odds of miscarriage and previable neonatal death less than 24 weeks.
Total number of participants used as denominator. Created using RevMan. Abbreviations: M-H = Mantel-Haenszel, CI = confidence interval.
For birthweight, including four studies with 227 participants, [7,19,21,22], the mean difference was 35.09g with a 95% CI of −533.70–603.89 and a p-value of 0.90. This demonstrates no effect on birthweight with either repeat cervical cerclage or expectant management (Fig 6).
Total births used as denominator. Created using RevMan. Abbreviations: CI = confidence interval.
For gestational age at delivery in weeks, including six studies with 240 participants, [7,18–22], the mean difference was −2.86 with a 95% CI of −7.70–1.99 and a p-value of 0.25. This demonstrates no effect on gestational age at delivery with either repeat cervical cerclage or expectant management (Fig 7).
Total births used as denominator. Created using RevMan. Abbreviations: CI = confidence interval.
3.6 Other secondary outcomes.
Secondary outcomes were variably reported by our included studies. There was paucity in secondary neonatal outcome data, and therefore we were unable to meta-analyse these. Contag reported 8 out of 20 (40.0%) patients who received repeat cervical cerclage experienced PPROM, compared with 7 out of 31 (22.6%) from their expectant management group [19]. Tang conversely reported 3 out of 12 (25.0%) participants who received repeat cerclage experienced PPROM compared with 9 out of 24 (37.5%) who received expectant management [22]. Clarfield reported rates of placental infection being 92.9% in the repeat cerclage group and 66.7% in the expectant management group, although the clinical significance of this and the method of diagnosis was unclear [7].
3.7 Subgroup and sensitivity analyses.
During data extraction a clear divide was found between studies defining primary cerclage failure as a short cervix and those that used prolapse of membranes beyond the level of the cerclage. Therefore, a post-hoc subgroup analysis was performed to examine any differences for the two definitions. This was confined to the primary outcome of pregnancy loss only to ensure adequate studies in each subgroup.
Only 4 studies were suitable for inclusion in the subgroup analysis and there was no evidence of a statistically significant difference between the subgroups (p = 0.14). However, amongst those with a short cervix, repeat cerclage was associated with increased odds of pregnancy loss with an OR of 5.94 with a 95% CI of 1.25–28.27 and a p-value of 0.03 (Fig 8).
Total number of participants used as denominator. Created using RevMan. Abbreviations: M-H = Mantel-Haenszel, CI = confidence interval.
The planned sensitivity analysis including only studies at low or moderate risk of bias was not possible owing to the limited number of eligible studies.
4. Discussion
4.1 Summary of key findings
Overall repeat cervical cerclage insertion following failure of the primary cerclage has no association with pregnancy loss rate as composite, preterm birth less than 37 weeks and less than 34 weeks, birthweight and gestational age at delivery. However, the spread of effect, particularly for pregnancy loss rate and preterm birth, may suggest that there is an association between repeat cerclage and worse outcomes, compared to expectant management. For miscarriage and previable neonatal death less than 24 weeks, receiving a repeat cervical cerclage could be associated with increased likelihood of pregnancy loss compared with those receiving expectant management (OR 4.27,95% CI 1.86–9.78). This was the only odds ratio in the analysis to reach statistical significance with a p-value of 0.0006. There was an association between receiving a repeat cerclage for a short cervix on ultrasound follow-up, with greater odds of pregnancy loss, compared to those receiving expectant management. There was no association with repeat cerclage and rates of pregnancy loss when failure was defined as prolapse of fetal membranes through the level of the cerclage. Overall, the available low-quality evidence does not demonstrate benefit of routine repeat cervical cerclage when the primary cerclage fails.
4.2 Review of existing literature
Three additional papers were identified during screening that reported case series’ of repeat cervical cerclage following primary cerclage failure without a comparator group. Ru et al. reported the perinatal outcomes of 55 patients who underwent repeat cervical cerclage for: prolapse of membranes beyond the level of the cerclage but without cervical dilatation and cervical dilatation up to 6 cm on physical examination [23]. 38 of 55 patients (69.1%) had a preterm birth less than 37 weeks and neonatal survival rate was 72.7%. Park et al. reported 35 women who underwent repeat cervical cerclage for prolapsed membranes on speculum examination [24]. Out of 35 patients, 8 had a miscarriage or previable neonatal death, 4 experienced ‘perinatal death’, described as death less than 72 hours following birth, and 5 had an early neonatal death, equating to a 48.6% rate of pregnancy loss as composite.
Fox et al. reported 12 patients who underwent repeat cervical cerclage. Their pregnancy loss rate (calculated as the inverse of live birth rate) was 1 out of 12 (8.4%) and 3 (25%) delivered prematurely before 37 weeks gestation.
These three studies do not offer comparison data and therefore, while descriptive of potential outcomes following repeat cerclage, do not aid counselling or decision-making for whether repeat cerclage is beneficial compared to expectant management.
4.3 Strengths and limitations
This systematic review and meta-analysis was performed according to a prospectively published protocol, using robust methodology in line with the PRISMA 2020 guidelines [8,9]. Comprehensive searches were carried out across the four main databases to ensure all possible studies eligible for inclusion were identified. To our knowledge there are no other published systematic reviews and meta-analyses of repeat cervical cerclage when the primary cerclage fails in asymptomatic patients.
The authors acknowledge that the quantity and quality of evidence used in this review is poor and therefore no recommendations can be made on this data alone. There are only retrospective cohort studies included, each with varying degrees of bias; two at moderate risk, two at serious risk and two at critical risk. The certainty of evidence is low or very low, according to the GRADE assessments performed, and there is substantial variation in the estimates of effect. It does, however, provide an important contribution to the body of evidence to guide and support clinical decision-making in this small subset of patients, where there are no pre-existing guidelines or robust data sets. Reporting of adverse maternal outcomes, such as PPROM, sepsis and chorioamnionitis was limited in the studies included in this review, and therefore there may be additional harm associated with repeat cervical cerclage that are not yet known. The sub-group analysis, grouped by definition of cerclage failure, identified no significant difference between the groups. This may reflect no true difference, or the analysis may be underpowered given only two studies contributed to each subgroup. It is however noteworthy that there was an association between having repeat cerclage for a short cervix and increased odds of suffering a pregnancy loss compared to expectant management. Although this was a highly significant odds ratio (p 0.0006), the poor quality evidence that this is based upon and the non-significant difference between the subgroups means conclusions are limited.
Given the retrospective nature of the studies, selection bias is likely to be significant, whereby patients with clinically more severe failure (e.g., greater dilatation of the cervix at the point of diagnosis of failure, or greatest risk according to their previous history) may have been selected to have a repeat cervical cerclage over expectant management. Clinicians may try everything possible to avoid pregnancy loss or extreme prematurity, particularly for those at greatest risk, and repeat cerclage may have been viewed as the last resort option. Although this is not demonstrated in the baseline population characteristics reported, it is clinically plausible and therefore may account for the trend seen towards worse outcomes for those who received repeat cervical cerclage compared to those who received expectant management.
4.4 Clinical application and recommendations
Routine ultrasound follow-up in the second trimester after cervical cerclage insertion has become more routine practice in UK Preterm Prevention Clinics in secondary and tertiary care. In terms of national and international guidance for ultrasound follow-up, the Society for Maternal-Fetal Medicine (SMFM) do not recommend routine ultrasound surveillance after cerclage insertion, as there is no evidence that repeat cervical cerclage improves outcomes [25]. The Royal College of Obstetricians (RCOG) similarly advise that routine ultrasound surveillance after history-indicated cervical cerclage is not recommended, although it can be offered to selected individuals where appropriate [6].
The studies included in this review differ in their criteria for primary cerclage failure, with two main definitions used: either a short cervix or membranes prolapsing beyond the level of the cerclage. Standardising the criteria would be beneficial for clinicians in practice. This review suggests that using the definition of a short cervix as primary cerclage failure may be associated with poorer outcomes with repeat cerclage, compared to if the membranes are prolapsing beyond the level of the primary cerclage. Clinically, a short but closed cervix beyond the cerclage suggests that it is still functioning adequately enough to maintain pregnancy. Therefore, inserting a repeat cerclage may do more harm by unnecessarily introducing surgical risks, risks of membrane rupture and repeat disruption of the cervical mucosa leading to infection and inflammatory processes which may cause pregnancy loss or preterm birth [26]. In comparison, when membranes prolapse through the level of the cerclage, the primary cerclage appears to be loose and therefore unlikely to be working adequately. In these cases, it is less clear whether repeat cerclage insertion would be beneficial for preventing pregnancy loss and preterm birth. Overall, ensuring the primary cerclage is of good quality is key to offering the best outcomes for patients, as repeat attempts may be associated with poorer outcomes. Addressing this with consensus internationally for optimal surgical technique and training is required. As demonstrated by Stirrat et al. on a cervical cerclage training simulator, even amongst experienced clinicians there was considerable variation in technique, height and tension of cerclage placement [27].
The authors of this systematic review suggest that using the definition of membranes prolapsing beyond the level of the cerclage is more clinically relevant and should be used preferentially to a short cervix.
The possible association between poorer outcomes and repeat cerclage as demonstrated in this review may be attributed to several factors. Procedural risks with a repeat cerclage in the same pregnancy, potentially days or weeks after the initial insertion, may be increased. These could include a greater chance of cervical trauma or the introduction of infection, particularly with shorter intervals between procedures. Selection bias is also likely to be a significant contributory factor, whereby the highest risk patients are selected for a repeat cerclage as clinicians are trying to do everything possible to prevent pregnancy loss or preterm birth. There may also be underlying uterine malformations or cervical pathology that increases the likelihood of the primary cerclage failing and repeating the cerclage does not compensate for the underlying pathology [28].
As this review shows, the evidence is insufficient to support repeat cervical cerclage following primary cerclage failure. However, timely identification of cerclage failure provides the opportunity for risk assessment and antenatal preparation. Particularly considering perinatal optimisation strategies with antenatal corticosteroids or in utero transfer to a centre with appropriate neonatal expertise, as per the British Association of Perinatal Medicine (BAPM) framework. [29] Identification of cerclage failure, particularly at gestations on the cusp of viability can guide clinicians to consider antenatal hospital admission, perform fetal growth assessment and provide the opportunity for neonatal counselling for parents. Without early identification of cerclage failure through routine ultrasound follow-up there is the risk that these patients attend symptomatic of preterm labour requiring emergency removal of their cerclage. They may also miss the window for optimally timed antenatal corticosteroids and in utero transfer to an appropriate neonatal intensive care centre.
4.5 Further research
This review was limited by the retrospective nature of the studies included. Future prospective cohort studies would provide much needed robust data to support our findings. An agreed definition for cerclage failure, or clear reporting of the indication for repeat cerclage is important for future meta-analysis. Although a randomized controlled trial would likely not be feasible due to the rare incidence of this clinical situation and therefore recruitment would be extremely challenging. The authors of this study aim to collaborate within the Preterm Clinical Network in the UK to agree a standardised definition of primary cerclage failure and establish a multi-centre registry of cerclage data, including long term neonatal outcome data. This dataset can be used to examine outcomes from patients with cerclage failure, with or without repeat cervical cerclage insertion.
4.6 Conclusions
In conclusion, repeat cervical cerclage insertion when the primary cerclage fails, has no association with rates of pregnancy loss, preterm birth less than 37 weeks, preterm birth less than 34 weeks, birthweight or gestational age at delivery. There is a trend towards worse outcomes with repeat cerclage compared to expectant management, however it is unclear if the underlying risk of preterm birth at the time of primary cerclage failure differed between groups. The quantity and quality of the evidence base is poor and therefore patients should be informed of this during counselling. Prospective research studies in this area are needed to form more robust conclusions for optimal management following primary cerclage failure.
Supporting information
S1 Appendix 1. Secondary outcomes.
Full list of secondary outcomes.
https://doi.org/10.1371/journal.pone.0329427.s001
(DOCX)
S2 Appendix 2. Search strategy.
Example of search strategy used for Medline and adapted for each database according to their search functionality.
https://doi.org/10.1371/journal.pone.0329427.s002
(DOCX)
S3 Appendix 3. Risk of bias assessment.
Risk of bias assessment table according to each domain within the Robins-I tool.
https://doi.org/10.1371/journal.pone.0329427.s003
(DOCX)
S4 Supplementary File 1. Population characteristics supplementary file.
https://doi.org/10.1371/journal.pone.0329427.s004
(XLSX)
Acknowledgments
The authors would like to thank Derick Yates, Evidence Specialist at Birmingham Women’s and Children’s NHS Foundation Trust Library and Knowledge Services, for his assistance with the search strategy and database searches. The authors would also like to thank the team at Birmingham Women’s and Children’s NHS Foundation Trust for their help retrieving full text articles.
References
- 1. Ohuma EO, Moller A-B, Bradley E, Chakwera S, Hussain-Alkhateeb L, Lewin A, et al. National, regional, and global estimates of preterm birth in 2020, with trends from 2010: a systematic analysis. Lancet. 2023;402(10409):1261–71. pmid:37805217
- 2. Hodgetts Morton V, Toozs-Hobson P, Moakes CA, Middleton L, Daniels J, Simpson NAB, et al. Monofilament suture versus braided suture thread to improve pregnancy outcomes after vaginal cervical cerclage (C-STICH): a pragmatic randomised, controlled, phase 3, superiority trial. Lancet. 2022;400(10361):1426–36. pmid:36273481
- 3. Owen J, Hankins G, Iams JD, Berghella V, Sheffield JS, Perez-Delboy A, et al. Multicenter randomized trial of cerclage for preterm birth prevention in high-risk women with shortened midtrimester cervical length. Am J Obstet Gynecol. 2009;201(4):375.e1-8. pmid:19788970
- 4. Alfirevic Z, Stampalija T, Medley N. Cervical stitch (cerclage) for preventing preterm birth in singleton pregnancy. Cochrane Database Syst Rev. 2017;6(6):CD008991. pmid:28586127
- 5.
Excellence NIfHaC. Preterm Labour and Birth. London, UK: NICE; 2022.
- 6. Shennan AH, Story L, Royal College of Obstetricians G. Cervical cerclage. BJOG: An International Journal of Obstetrics & Gynaecology. 2022;129(7):1178–210.
- 7. Clarfield L, Shapiro J, Cherniak R, Szadkowski L, Huszti E, Whittle W. Short cervix following initial cerclage-reinforcing cerclage versus expectant management. J Obstet Gynaecol Can. 2024;46(4):102286.
- 8. Moher D, Liberati A, Tetzlaff J, Altman DG, PRISMA Group. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. BMJ. 2009;339:b2535. pmid:19622551
- 9. Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 2021;372:n71. pmid:33782057
- 10.
Emms A PN, Vaughan M, Hodgetts-Morton V. Pregnancy loss rate and perinatal outcomes in women with repeat cervical cerclage, following failure of the primary cerclage, compared with expectant management: a systematic review and meta-analysis PROSPERO. 2024. https://www.crd.york.ac.uk/PROSPERO/view/CRD42025638147
- 11. van ’t Hooft J, Duffy JMN, Daly M, Williamson PR, Meher S, Thom E, et al. A Core Outcome Set for Evaluation of Interventions to Prevent Preterm Birth. Obstet Gynecol. 2016;127(1):49–58. pmid:26646133
- 12. Sterne JA, Hernán MA, Reeves BC, Savović J, Berkman ND, Viswanathan M, et al. ROBINS-I: a tool for assessing risk of bias in non-randomised studies of interventions. BMJ. 2016;355:i4919. pmid:27733354
- 13.
Schünemann HJ, Vist GE, Glasziou P, Akl EA, Skoetz N, Guyatt GH. Completing ‘summary of findings’ tables and grading the certainty of the evidence. In: Higgins JPT, Chandler J, Cumpston M, Li T, Page MJ, Welch VA, editors. Cochrane Handbook for Systematic Reviews of Interventions. Cochrane: Cochrane; 2024.
- 14. Prasad M. Introduction to the GRADE tool for rating certainty in evidence and recommendations. Clinical Epidemiology and Global Health. 2024;25:101484.
- 15.
Schünemann HJ, Vist GE, Glasziou P, Akl EA, Skoetz N, Guyatt GH. Considering bias and conflicts of interest among the included studies. In: Higgins JPT, Chandler J, Cumpston M, Li T, Page MJ, Welch VA, editors. Cochrane Handbook for Systematic Reviews of Interventions version 65. Cochrane; 2024.
- 16. Wan X, Wang W, Liu J, Tong T. Estimating the sample mean and standard deviation from the sample size, median, range and/or interquartile range. BMC Med Res Methodol. 2014;14:135. pmid:25524443
- 17.
Schünemann HJ, Vist GE, Glasziou P, Akl EA, Skoetz N, Guyatt GH. Analyses data and undertaking meta-analyses. In: Higgins JPT, Chandler J, Cumpston M, Li T, Page MJ, Welch VA, editors. Cochrane Handbook for Systematic Reviews of Interventions version 65. Cochrane; 2024.
- 18. Baxter JK, Airoldi J, Berghella V. Short cervical length after history-indicated cerclage: is a reinforcing cerclage beneficial?. Am J Obstet Gynecol. 2005;193(3 Pt 2):1204–7. pmid:16157138
- 19. Contag SA, Woo J, Schwartz DB, Arrabal PP, O’Reilly G, Harper M. Reinforcing cerclage for a short cervix at follow-up after the primary cerclage procedure. J Matern Fetal Neonatal Med. 2016;29(15):2423–7. pmid:26429503
- 20. Simcox R, Shennan A. Reinforcing cerclage in the prevention of preterm birth in women at high risk: a retrospective case-controlled study. Aust N Z J Obstet Gynaecol. 2012;52(3):224–8. pmid:22533341
- 21. Song J-E, Lee K-Y, Jun H-A. Repeat cerclage prolongs pregnancy in women with prolapsed membranes. Acta Obstet Gynecol Scand. 2011;90(1):111–3. pmid:21275924
- 22. Tang Y, Bao Y, Jiang X, Xie H, Cao Z, Qu X, et al. Cervical dilation after initial cervical cerclage in twin pregnancies: is repeat cerclage beneficial?. Am J Obstet Gynecol MFM. 2024;6(3):101305. pmid:38301998
- 23. Ru P, Ni X, Xu W, Liu Y, Meng L, Yuan W, et al. Perinatal outcomes in patients undergoing repeat cerclage: A retrospective case series study. Int J Gynaecol Obstet. 2024;165(1):343–9. pmid:37899709
- 24. Park S, Lee K-Y, Song J-E. Postoperative cervical length to predict success of repeat cerclage in singleton pregnancies with prolapsed membranes after prior cerclage. Front Med (Lausanne). 2023;10:1248321. pmid:37671397
- 25. McIntosh J, Feltovich H, Berghella V, Manuck T. The role of routine cervical length screening in selected high- and low-risk women for preterm birth prevention. Am J Obstet Gynecol. 2016;215(3):B2-7. pmid:27133011
- 26. Lin Y, Fang J, Ni R, Zhang L, Zhao J, Jiang X, et al. Dynamic Change of Novel Systemic Inflammation Markers to Predict Maternal-Neonatal Prognosis After Cervical Cerclage. J Inflamm Res. 2023;16:1745–56. pmid:37113628
- 27. Stirrat LI, Tydeman F, Suff N, Hall M, English WJ, Shennan AH, et al. Cervical cerclage technique: what do experts actually achieve?. Journal of Obstetrics and Gynaecology. 2023;1(1):1–10.
- 28. Wang S, Wang K, Hu Q, Liao H, Wang X, Yu H. Perinatal outcomes of women with Müllerian anomalies. Arch Gynecol Obstet. 2023;307(4):1209–16. pmid:35426514
- 29.
Medicine BAOP. Perinatal management of extreme preterm birth before 27 weeks gestation. A framework for practice. UK: BAPM; 2019.