Sutura cervical (cerclaje) para la prevención del parto prematuro en el embarazo múltiple
Resumen
Antecedentes
El cerclaje cervical es una intervención quirúrgica que incluye la colocación de una sutura alrededor del cuello uterino. El material de sutura tiene como objetivo prevenir el acortamiento y la abertura cervical, y así reducir el riesgo de parto prematuro. La efectividad y seguridad de este procedimiento en embarazos múltiples todavía son polémicas.
Objetivos
Evaluar si el uso de un cerclaje cervical en las gestaciones múltiples, en presencia de alto riesgo de pérdida del embarazo debido solamente a la gestación múltiple (cerclaje indicado por los antecedentes), a los hallazgos de la ecografía de "cuello uterino corto" (cerclaje indicado por la ecografía) o a los cambios detectados mediante el examen del cuello uterino (cerclaje indicado por el examen físico), mejora los resultados obstétricos y perinatales. Los resultados primarios evaluados fueron muertes perinatales, morbilidad neonatal grave y muertes perinatales y morbilidad neonatal grave.
Métodos de búsqueda
Se hicieron búsquedas en el registro de ensayos del Grupo Cochrane de Embarazo y Parto (Cochrane Pregnancy and Childbirth Group) (30 de junio de 2014) y en las listas de referencias de los estudios recuperados.
Criterios de selección
Todos los ensayos controlados aleatorios (ECA) de cerclaje cervical en embarazos múltiples. Se consideró la inclusión de ensayos cuasialeatorios y ECA que emplearan un diseño aleatorio grupal (pero no se identificaron). Los estudios que utilizaran un diseño cruzado (crossover) y los que se presentaran solamente como resúmenes no fueron elegibles para inclusión.
Se incluyeron los estudios que compararon cerclaje cervical con ningún cerclaje cervical en embarazos múltiples.
Los estudios que compararan la sutura cervical versus cualquier otro tratamiento preventivo (p.ej. progesterona) en los embarazos múltiples y los estudios que incluyeron comparaciones entre diferentes protocolos de cerclaje (cerclaje indicado por los antecedentes versus indicado por la ecografía versus indicado por el examen físico) también fueron elegibles para inclusión pero no se identificaron.
Obtención y análisis de los datos
Dos revisores evaluaron de forma independiente los ensayos para la inclusión y el riesgo de sesgo. Dos revisores extrajeron los datos. Se verificó la exactitud de los datos.
Resultados principales
Se incluyeron cinco ensayos con 1577 pacientes asignadas al azar que incluyeron gestaciones únicas y múltiples. Después de excluir las gestaciones únicas, el análisis final incluyó a 128 pacientes, de las cuales 122 tenían gestaciones gemelares y seis tenían gestaciones de trillizos. Dos ensayos (n = 73 pacientes) evaluaron el cerclaje indicado por los antecedentes, mientras tres ensayos (n = 55 pacientes) evaluaron el cerclaje indicado por la ecografía. Los cinco ensayos se calificaron de calidad promedio a por encima del promedio, y tres de los ensayos tuvieron riesgo incierto con respecto a los sesgos de selección y de detección.
Con respecto a los resultados primarios, cuando los resultados del cerclaje se agruparon juntos para todas las indicaciones y se compararon con ningún cerclaje, no hubo diferencias estadísticamente significativas en las muertes perinatales (19,2% versus 9,5%; cociente de riesgos [CR] 1,74; intervalo de confianza [IC] del 95%: 0,92 a 3,28; cinco ensayos, n = 262), la morbilidad neonatal grave (15,8% versus 13,6%; CR promedio 0,96; IC del 95%: 0,13 a 7,10; tres ensayos, n = 116), o el resultado compuesto muerte perinatal y morbilidad neonatal (40,4% versus 20,3%; CR promedio 1,54; IC del 95%: 0,58 a 4,11; tres ensayos, n = 116).
Entre los resultados secundarios no hubo diferencias significativas entre los grupos de cerclaje y ningún cerclaje. Para mencionar algunos, no hubo diferencias significativas entre los siguientes: parto prematuro con menos de 34 semanas (CR promedio 1,16; IC del 95%: 0,44 a 3,06; cuatro ensayos, n = 83), parto prematuro con menos de 35 semanas (CR promedio 1,11; IC del 95%: 0,58 a 2,14; cuatro ensayos, n = 83), bajo peso al nacer menor de 2500 g (CR promedio 1,10; IC del 95%: 0,82 a 1,48; cuatro ensayos, n = 172), muy bajo peso al nacer menor de 1500 g (CR promedio 1,42; IC del 95%: 0,52 a 3,85; cuatro ensayos, n = 172) y síndrome de dificultad respiratoria (CR promedio 1,70; IC del 95%: 0,15 a 18,77; tres ensayos, n = 116). Tampoco hubo diferencias significativas entre los grupos de cerclaje y ningún cerclaje cuando se examinó la cesárea (programada y de urgencia) (CR 1,24; IC del 95%: 0,65 a 2,35; tres ensayos, n = 77) y los efectos secundarios maternos (CR 3,92; IC del 95%: 0,17 a 88,67; un ensayo, n = 28).
Al examinar las diferencias entre subgrupos preespecificados, el cerclaje indicado por la ecografía se asoció con un mayor riesgo de bajo peso al nacer (CR promedio 1,39; IC del 95%: 1,06 a 1,83; Tau² = 0,01; I² = 15%; tres ensayos, n = 98), muy bajo peso al nacer (CR promedio 3,31; IC del 95%: 1,58 a 6,91; Tau² = 0; I² = 0%; tres ensayos, n = 98) y síndrome de dificultad respiratoria (CR promedio 5,07; IC del 95%: 1,75 a 14,70; Tau² = 0; I² = 0%; tres ensayos, n = 98). Sin embargo, debido al escaso número de ensayos, la heterogeneidad significativa y las diferencias de subgrupos, estos datos se deben interpretar con precaución.
Ningún ensayo informó resultados del neurodesarrollo del lactante a largo plazo. No hubo cerclajes indicados por el examen físico disponibles para la comparación entre los estudios incluidos.
Conclusiones de los autores
Esta revisión se basa en los datos limitados de cinco estudios pequeños de calidad promedio a por encima del promedio. Para las gestaciones múltiples no existen pruebas de que el cerclaje sea una intervención eficaz para prevenir los partos prematuros y reducir las muertes perinatales o la morbilidad neonatal.
PICO
Resumen en términos sencillos
Sutura cervical para la prevención del parto prematuro en pacientes con un embarazo múltiple
Antecedentes
Gestar más de un feto aumenta el riesgo de la mujer de tener un parto prematuro. Los riesgos aumentan con el número de fetos gestados. Los neonatos nacidos prematuramente tienen mayores probabilidades de presentar resultados deficientes que incluyen enfermedades graves y muerte. El cerclaje cervical es un procedimiento quirúrgico realizado durante el embarazo para tratar de prevenir el parto prematuro al limitar el acortamiento y la abertura del cuello uterino. Se realiza al colocar material de sutura alrededor del cuello uterino, al que se accede por la vagina o a través del abdomen de la madre. La efectividad y seguridad de este procedimiento para las gestaciones múltiples todavía son inciertas. La probabilidad de parto prematuro espontáneo se puede evaluar al analizar los antecedentes obstétricos de la madre, el examen físico o el examen ecográfico transvaginal en el segundo trimestre.
Pregunta de la revisión
Evaluar si el uso de una sutura cervical en las pacientes con gestaciones múltiples con alto riesgo de pérdida del embarazo mejora los resultados obstétricos y perinatales.
Características de los estudios
Se incluyeron cinco ensayos con 1577 pacientes que compararon cerclaje con ningún cerclaje en pacientes con gestaciones únicas y múltiples. Después de excluir las gestaciones únicas, el análisis final incluyó a 128 pacientes, de las cuales 122 estaban embarazadas de gemelos y seis estaban embarazadas de trillizos. El cerclaje estaba indicado por los antecedentes obstétricos en dos ensayos (n = 73 pacientes) y por la ecografía transvaginal en tres ensayos (n = 55 pacientes).
Principales hallazgos
Cuando el cerclaje se comparó con ningún cerclaje en las pacientes con gestaciones múltiples no hubo diferencias en las muertes perinatales o la enfermedad neonatal, ni en las tasas de parto prematuro. Sin embargo, el número de pacientes incluidas en los cinco estudios no fue suficiente para proporcionar conclusiones significativas.
No fue posible calcular el efecto a largo plazo del cerclaje sobre los resultados del neurodesarrollo en los lactantes supervivientes ni la infección y los efectos secundarios maternos. Por lo tanto, no estuvo claro si el cerclaje en las pacientes con embarazos múltiples pone la salud de las madres o los neonatos en algún riesgo.
Calidad de la evidencia
En general los cinco estudios incluidos se consideraron de calidad promedio a por encima del promedio, pero tres de los estudios fueron difíciles de evaluar completamente debido a la falta de información metodológica.
No se encontraron estudios que compararan indicaciones diferentes para el cerclaje (cerclaje indicado por los antecedentes obstétricos versus cerclaje indicado por la ecografía) o que compararan el cerclaje con otra intervención (como la progesterona).
Authors' conclusions
Background
Description of the condition
The twin birth rate in the USA has risen over the last three decades, from 18.9 per 1000 births in 1980 to 33.3 per 1000 births in 2009. Although twin pregnancies currently represent only 1% to 3% of live births in most countries, they represent a substantial portion of preterm deliveries, with 10% of all preterm births in the USA attributable to twins (Martin 2012). In 2006 in the USA, the mean age at delivery for twins was 35.3 weeks compared with 38.8 weeks for singletons (Martin 2009), with 12.1% of twins delivering before 32 weeks compared to only 1.6% in singletons. Although less than 1% of singletons deliver before 28 weeks' gestation, 5% of twins are born before that extremely early gestational age. Similar rates are observed elsewhere worldwide, including Europe and Asia, where twin preterm birth rates (less than 37 weeks) range from 42% to 68% (Blondel 2006; Ooki 2010), with as many as 8% of twins delivering at less than 32 weeks (Ooki 2010; Papiernik 2010).
According to recent data from the USA, one in four very low birthweight infants (less than 1500 g), and one of every six infant deaths (within the first six months of life), come from twin pregnancies (Martin 2009; Mathews 2006). The burden of prematurity is worse for triplets and higher order gestations, with the median gestational age at delivery for triplets being 32 weeks. With premature infants being at higher risk of poor outcomes including death, preterm birth is, in effect, a surrogate for mortality and morbidity. Twins are five times, and triplets nearly 15 times more likely than singletons to die within one month of birth (Martin 2008), and those born preterm are at significantly greater risk of incurring serious neonatal morbidities (Refuerzo 2010).
The mechanism for early preterm birth in multiple gestations is unclear. While preterm birth usually represents the final common pathway of at least four distinct pathophysiologic processes (maternal and/or fetal stress, inflammation, abruption or decidual bleeding, and excessive mechanical stretching of uterus), it seems the latter, or overdistension of the uterus, is the most common aetiology causing twin preterm birth (Hodgson 2010). Given the different incidence of preterm birth and the different mechanisms leading to it in singleton and multiple pregnancies, we felt that safety and efficacy of interventions aimed at preventing preterm birth should be evaluated separately. The role of the cervix and/or cervical insufficiency, if any, is unclear in the aetiologies of preterm birth in multiple gestations.
Over the last 50 years, much has been learned about the cervix and its complex role involving both normal and abnormal parturition. In normal pregnancies approaching term, the cervix undergoes remodelling, ripening, and eventual dilatation, leading to normal labour and delivery. These events can occur prematurely, leading to either second trimester losses (e.g. between 16 and 24 weeks of gestation), or preterm births between 24 + 0 and 36 + 6 weeks of gestation. In the absence of uterine contractions, painless dilatation of the cervix leading to recurrent second trimester losses had been the classic definition of cervical insufficiency, previously referred to as cervical incompetence.
Transvaginal ultrasound examination of the cervix has emerged as an effective screening method for predicting pregnancies which will spontaneously deliver preterm. This is especially true for women with a prior preterm birth (Owen 2001). The earlier the short cervical length is detected, the higher the risk of preterm birth. Given this new information, the definition of cervical insufficiency needs to be updated to encompass women with prior preterm births with cervical shortening before 24 weeks in the current pregnancy (Berghella 2010). While the true incidence of cervical insufficiency is unknown, several risk factors for the condition have been reported in the literature, such as cervical excision surgery (loop electrosurgical excision procedure, cone biopsy) (Berghella 2004a; Jakobsson 2007; Noehr 2009), greater than one surgical termination of pregnancy (Liao 2011 ; Visintine 2008), diethylstilbestrol exposure (Ludmir 1987), and collagen tissue disorders.
Description of the intervention
Cervical cerclage represents one of the most well‐known surgical interventions in obstetrics. It is performed by securing suture material around the cervix to prevent cervical shortening and opening. Both the transvaginal and transabdominal methods for cervical cerclage have been reported.
While there are various methods of placing a cerclage, the two most popular include the McDonald and Shirodkar methods. Usually these procedures require regional anaesthesia in the form of a spinal or epidural block. General anaesthetic remains an option. In 1955, Shirodkar reported the insertion of a cervical stitch (suture) at around 14 weeks of pregnancy, placing a purse string stitch around the cervix, requiring dissection of the bladder and rectum from the cervix prior to stitch placement (Shirodkar 1955). Two years later, McDonald described a simpler technique, whereby the stitch is inserted around the body of the cervix present in the vagina in three or four bites (McDonald 1957). The McDonald procedure is technically easier to perform, with less bleeding, and the stitch is easier to remove. These techniques were described as indicated just for women with both a prior preterm birth and a cervix that is shortening and/or dilating in the second trimester. Several other variations for cerclage technique have been described.
Stitches are normally inserted via the vaginal route, although transabdominal cerclage has also been proposed for women when vaginal stitches have failed to prevent a preterm birth, or when a woman has an extremely short, scarred cervix, making vaginal stitch insertion technically difficult (Anthony 1997; Gibb 1995). The transabdominal procedures can be carried out in early pregnancy around 12 weeks of gestation, or are scheduled before pregnancy. Regardless of the timing, during laparotomy, the suture is placed at the cervicoisthmic portion of the uterus. Recently a laparoscopic approach to transabdominal cerclage has been described as a safe and effective alternative approach to laparotomy (Carter 2009).
Vaginally inserted cervical stitches are usually taken out at 36 to 37 weeks' gestation, or when the woman presents in labour. Abdominal cervical stitches are left in place and the baby is delivered by caesarean section.
Risks associated with cerclage placement, albeit rare, include bleeding, infection, cervical lacerations, and iatrogenic rupture of membranes.
Definitions of cervical cerclage types, including those placed for multiple pregnancies and those considered to be at high risk for pregnancy loss, vary depending on indication:
-
history‐indicated cerclage ‐ a planned procedure around 12 to 15 weeks based on previous obstetric history;
-
ultrasound‐indicated cerclage ‐ a procedure carried out following the discovery of a shortened cervical length on transvaginal ultrasound examination;
-
physical exam‐indicated cerclage ‐ a procedure carried out following detection of advanced cervical dilation during vaginal examination. This can been either an incidental finding or a finding in women with some symptoms of threatened preterm labour.
The safety and efficacy of cerclage should be evaluated separately for each of these groups, as they are different clinical situations and results would be expected to vary.
How the intervention might work
No matter when the cerclage is placed, the aim is to prevent or halt the process of cervical shortening that leads to established labour and eventual preterm birth. A history‐indicated cerclage is placed early in pregnancy, typically at the end of the first trimester (12 to 15 weeks). These are placed in women with poor pregnancy histories (e.g. two to three second trimester losses, prior failed ultrasound‐indicated cerclage, etc), regardless of any events in the current pregnancy, in an attempt to provide preemptive mechanical support before the cervix becomes 'insufficient' (Berghella 2007).
More recently, transvaginal ultrasonography measuring cervical length has shown that the finding of a short cervix is one of the best predictors for preterm birth (Berghella 1999; Iams 1996; Owen 2001), and ultrasound‐indicated cervical cerclage is gaining popularity.
Rather than prevent cervical shortening prior to its commencement (history‐indicated cerclage), an ultrasound‐indicated cerclage looks to halt initial cervical length shortening by providing mechanical support to the cervix and lower uterine segment, and therefore prolonging pregnancy.
Interestingly, cerclage was originally described as being indicated for women with both second trimester loss/preterm birth and a dilated cervix (McDonald 1957; Shirodkar 1955).
Why it is important to do this review
While there is emerging evidence supporting the use of cerclage in women with both prior preterm birth and current mid‐trimester cervical shortening, this evidence applies solely to singleton gestations.
In a previously published Cochrane review, Drakeley and colleagues showed a lack of reduction in total pregnancy loss, early pregnancy loss, or preterm delivery before 28 and 34 weeks in women receiving cerclage, compared with no cerclage (Drakeley 2003).
In their meta‐analysis of individual patient data, Berghella et al concluded that cerclage does not prevent preterm birth in all women with a short cervix at ultrasound examination, but could be beneficial in singleton pregnancies with short cervix and prior preterm birth (Berghella 2005). This benefit in singleton pregnancies with short cervix and prior preterm birth was confirmed in a more recent randomised trial (Owen 2009) and a meta‐analysis (Berghella 2011). Two meta‐analyses (Berghella 2005; Jorgensen 2007) showed no benefit for multiple gestation pregnancies, with Berghella and colleagues demonstrating an increased risk for preterm birth before 35 weeks in those women with twin gestations who received an ultrasound‐indicated cerclage (risk ratio 2.15, 95% confidence interval 1.15 to 4.01).
Despite published meta‐analyses and other studies demonstrating the lack of efficacy of cerclage in multiple gestations (Rebarber 2005; Roman 2005), recent data from the USA indicate that roughly 10% of triplets, and 1.3% of twins are still receiving cerclages (Menacker 2008).
A separate Cochrane review (Alfirevic 2012) looks at singleton pregnancies. As the incidence of preterm birth and mechanisms that bring it about are so different in singleton and multiple pregnancies, we believe that safety and efficacy of interventions aimed at preventing preterm birth should be evaluated separately. This review and Alfirevic 2012 update the previously published review by Drakely (Drakeley 2003).
Objectives
To assess whether the use of a cervical stitch in multiple gestations at high risk of pregnancy loss based on woman's history, ultrasound findings of 'short cervix', or physical exam changes in the cervix, improves obstetrical and perinatal outcomes.
Methods
Criteria for considering studies for this review
Types of studies
All randomised controlled trials (RCTs) of cervical cerclage in multiple pregnancies. Quasi‐RCTs and RCTs using a cluster‐randomised design were eligible for inclusion (but none were identified). Studies using a cross‐over design and those presented only as abstracts were not eligible for inclusion.
We included studies comparing cervical cerclage with no cervical cerclage in multiple pregnancies.
Studies comparing cervical stitch versus any other preventative therapy (e.g. progesterone) in multiple pregnancies, and studies involving comparisons between different cerclage protocols (history‐indicated versus ultrasound‐indicated versus physical exam‐indicated cerclage) were also eligible for inclusion but none were identified.
Types of participants
All women with multiple gestations were included. We planned also to carry out a priori subgroup analyses (see Subgroup Analysis section).
Types of interventions
Cervical stitch (cerclage) inserted before or during pregnancy irrespective of the indication.
Comparisons
-
Cervical stitch versus no stitch
-
Cervical stitch versus any other preventative therapy (e.g. progesterone)
-
Any comparison of different cerclage protocols (history‐indicated versus ultrasound‐indicated versus physical exam‐indicated cerclage)
Types of outcome measures
We selected outcome measures with the help of a proposed core data set of outcome measures (Devane 2007).
Primary outcomes
-
Perinatal deaths
-
Serious neonatal morbidity (defined by trialists)
-
Perinatal deaths and serious neonatal morbidity
We acknowledge that it is unusual to include a composite as the primary outcome when individual components may have significantly different consequences for families and healthcare providers. However, both perinatal deaths and severe neonatal morbidity are rare events even in these high‐risk populations, and therefore meta‐analysis may fail to detect clinically important differences.
As there are no internationally agreed definitions for severe neonatal morbidity, we will accept any reasonable definition by trialists, as long as it is applied consistently across the whole study population in an unbiased manner. The same applies for perinatal deaths. Although perinatal mortality is clearly defined (stillbirths and neonatal deaths within first week of life), some premature babies may die after the first week of life and it is important to include this information where available.
It may seem unusual not having preterm birth as the primary outcome. We believe that, in the context of this review, preterm birth is a merely a surrogate for mortality and morbidity. More importantly, there is a real possibility that prolongation of pregnancy may be misinterpreted as benefit, when in fact, it may be harmful to keep a baby in what can sometimes be a hostile uterine environment. Preterm birth (at various gestations) remains an important secondary outcome.
Secondary outcomes
Neonatal
-
Stillbirth (fetal demise after 20 weeks' gestation, prior to delivery)
-
Neonatal death (after birth, and before 29 days of neonatal life or discharge from hospital)
-
Preterm birth (as defined by trialists, e.g. less than 28, 32, 34, 35, 37 weeks)
-
Gestational age at delivery
-
Low birthweight defined as less than 2500 g
-
Very low birthweight defined as less than 1500 g
-
Respiratory distress syndrome (defined by trialists)
-
Intraventricular haemorrhage (defined by trialists)
-
Necrotising enterocolitis (defined by trialists)
-
Sepsis (defined by trialists)
-
Neonatal intensive care unit admission
-
Long‐term infant neurodevelopmental outcomes
Maternal
-
Caesarean section (planned and emergency)
-
Maternal infection requiring intervention, e.g. antibiotics or delivery (including chorioamnionitis and endometritis)
-
Maternal side‐effects (vaginal discharge, bleeding, pyrexia not requiring antibiotics)
Search methods for identification of studies
Electronic searches
We searched the Cochrane Pregnancy and Childbirth Group’s Trials Register by contacting the Trials Search Co‐ordinator (30 June 2014).
The Cochrane Pregnancy and Childbirth Group’s Trials Register is maintained by the Trials Search Co‐ordinator and contains trials identified from:
-
monthly searches of the Cochrane Central Register of Controlled Trials (CENTRAL);
-
weekly searches of MEDLINE;
-
weekly searches of Embase;
-
handsearches of 30 journals and the proceedings of major conferences;
-
weekly current awareness alerts for a further 44 journals plus monthly BioMed Central email alerts.
Details of the search strategies for CENTRAL, MEDLINE and Embase, the list of handsearched journals and conference proceedings, and the list of journals reviewed via the current awareness service can be found in the ‘Specialized Register’ section within the editorial information about the Cochrane Pregnancy and Childbirth Group.
Trials identified through the searching activities described above are each assigned to a review topic (or topics). The Trials Search Co‐ordinator searches the register for each review using the topic list rather than keywords.
Searching other resources
We searched the reference lists of the studies identified.
We did not apply any language restrictions.
Data collection and analysis
Selection of studies
Two review authors (Zarko Alfirevic and Timothy Rafael) independently assessed for inclusion all the potential studies we identified as a result of the search strategy. Disagreements were resolved through discussion.
Data extraction and management
We used a pre‐designed form to extract data. For included studies, two review authors extracted the data using the agreed form. We resolved discrepancies through discussion. We entered data into Review Manager software (RevMan 2012) and checked for accuracy. When information regarding any of the above was unclear, we attempted to contact authors of the original reports to provide further details.
Along those lines, in studies that included both singleton and multiple gestations, we made every effort to extract the data and results specific to the multiple gestation. Where outcomes were not reported specifically by subgroup, but rather as outcomes for singleton and multiple gestations combined, we attempted to contact the authors of the original reports to provide further details, e.g. original patient level data.
Assessment of risk of bias in included studies
Two review authors independently assessed risk of bias for each study using the criteria outlined in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We resolved any disagreement by discussion or by involving a third assessor.
(1) Random sequence generation (checking for possible selection bias)
We described for each included study the method used to generate the allocation sequence in sufficient detail to allow an assessment of whether it should produce comparable groups.
We assessed the method as:
-
low risk of bias (any truly random process, e.g. random number table; computer random number generator);
-
high risk of bias (any non‐random process, e.g. odd or even date of birth; hospital or clinic record number);
-
unclear risk of bias.
(2) Allocation concealment (checking for possible selection bias)
We described for each included study the method used to conceal allocation to interventions prior to assignment and assessed whether intervention allocation could have been foreseen in advance of, or during recruitment, or changed after assignment.
We assessed the methods as:
-
low risk of bias (e.g. telephone or central randomisation; consecutively numbered sealed opaque envelopes);
-
high risk of bias (open random allocation; unsealed or non‐opaque envelopes, alternation; date of birth);
-
unclear risk of bias.
(3.1) Blinding of participants and personnel (checking for possible performance bias)
We described for each included study the methods used, if any, to blind study participants and personnel from knowledge of which intervention a participant received. We considered that studies were at low risk of bias if they were blinded, or if we judged that the lack of blinding would be unlikely to affect results. We assessed blinding separately for different outcomes or classes of outcomes.
We assessed the methods as:
-
low, high or unclear risk of bias for participants;
-
low, high or unclear risk of bias for personnel.
(3.2) Blinding of outcome assessment (checking for possible detection bias)
We described for each included study the methods used, if any, to blind outcome assessors from knowledge of which intervention a participant received. We assessed blinding separately for different outcomes or classes of outcomes.
We assessed methods used to blind outcome assessment as:
-
low, high or unclear risk of bias.
(4) Incomplete outcome data (checking for possible attrition bias due to the amount, nature and handling of incomplete outcome data)
We described for each included study, and for each outcome or class of outcomes, the completeness of data including attrition and exclusions from the analysis. We stated whether attrition and exclusions were reported and the numbers included in the analysis at each stage (compared with the total randomised participants), reasons for attrition or exclusion where reported, and whether missing data were balanced across groups or were related to outcomes. Where sufficient information was reported, or could be supplied by the trial authors, we re‐included missing data in the analyses which we undertook.
We assessed methods as:
-
low risk of bias (e.g. no missing outcome data; missing outcome data balanced across groups);
-
high risk of bias (e.g. numbers or reasons for missing data imbalanced across groups; ‘as treated’ analysis done with substantial departure of intervention received from that assigned at randomisation);
-
unclear risk of bias.
We discussed whether missing data greater than 20% might impact on outcomes, acknowledging that with long‐term follow‐up, complete data are difficult to attain.
(5) Selective reporting (checking for reporting bias)
We described for each included study how we investigated the possibility of selective outcome reporting bias and what we found.
We assessed the methods as:
-
low risk of bias (where it was clear that all of the study’s pre‐specified outcomes and all expected outcomes of interest to the review have been reported);
-
high risk of bias (where not all the study’s pre‐specified outcomes have been reported; one or more reported primary outcomes were not pre‐specified; outcomes of interest are reported incompletely and so cannot be used; study failed to include results of a key outcome that would have been expected to have been reported);
-
unclear risk of bias.
(6) Other bias (checking for bias due to problems not covered by (1) to (5) above)
We described for each included study any important concerns we had about other possible sources of bias.
We assessed whether each study was free of other problems that could put it at risk of bias:
-
low risk of other bias;
-
high risk of other bias;
-
unclear whether there is risk of other bias.
(7) Overall risk of bias
We made explicit judgements about whether studies were at high risk of bias, according to the criteria given in the Handbook (Higgins 2011). With reference to (1) to (6) above, we assessed the likely magnitude and direction of the bias and whether we considered it was likely to impact on the findings. We explored the impact of the level of bias through undertaking sensitivity analyses ‐ seeSensitivity analysis.
Measures of treatment effect
We used as the denominator the number of babies of women who were randomised, even though some babies could not have attained the outcome; for example, if there was a stillbirth then this baby would not have been able to attain the outcome of 'admission to special care baby unit'. For outcomes dealing with the pregnancy as a whole (e.g. preterm birth less than 28 weeks, less than 32 weeks, etc), the unit of analysis was the pregnancy, and if the outcome occurred in either the fetus or neonate, we considered the pregnancy to have met the outcome.
Dichotomous data
For dichotomous data, we presented results as a summary risk ratio with 95% confidence intervals.
Continuous data
For continuous data, we used the mean difference if outcomes were measured in the same way between trials. We planned to use the standardised mean difference to combine trials that measured the same outcome, but used different methods.
Unit of analysis issues
Cluster‐randomised trials
We did not identify any cluster‐randomised trials for inclusion in this review. However, if we identify any cluster‐randomised trial in future updates of this review we will include them in the analyses along with individually‐randomised trials. We will adjust their standard errors using the methods described in the Handbook using an estimate of the intracluster correlation co‐efficient (ICC) derived from the trial (if possible), from a similar trial or from a study of a similar population. We will meta‐analyse effect estimates and their standard errors from correct analyses of cluster‐randomised trials using the generic inverse‐variance method in RevMan. If we use ICCs from other sources, we will report this and conduct sensitivity analyses to investigate the effect of variation in the ICC. If we identify both cluster‐randomised trials and individually‐randomised trials, we plan to synthesise the relevant information. We will consider it reasonable to combine the results from both if there is little heterogeneity between the study designs and the interaction between the effect of intervention and the choice of randomisation unit is considered to be unlikely.
We will also acknowledge heterogeneity in the randomisation unit and perform sensitivity analyses to investigate the effects of the randomisation unit.
Cross‐over trials
We considered cross‐over designs inappropriate for this research question.
Dealing with missing data
For included studies, we planned to note levels of attrition. We planned to explore the impact of including studies with high levels of missing data in the overall assessment of treatment effect by using Sensitivity analysis.
For all outcomes, we planned to carry out analyses, as far as possible, on an intention‐to‐treat basis, i.e. we attempted to include all participants randomised to each group in the analyses, and analysed all participants in the group to which they were allocated, regardless of whether or not they received the allocated intervention. The denominator for each outcome in each trial was the number of babies of women who were randomised.
We planned to exclude data on outcomes where there was greater than 20% missing data on short‐term outcomes.
Assessment of heterogeneity
We assessed statistical heterogeneity in each meta‐analysis using the Tau², I² and Chi² statistics. We regarded heterogeneity as substantial if the Tau² was greater than zero and either an I² was greater than 30% or there was a low P value (less than 0.10) in the Chi² test for heterogeneity.
In cases of substantial heterogeneity, we planned to explore the causes of it by pre‐specified subgroup analysis provided that at least 10 studies contributed to the meta‐analysis. We anticipated that for outcomes with fewer than 10 studies it would be difficult to assess subgroup effects with adequate power.
Assessment of reporting biases
In future updates of this review, if there are 10 or more studies in the meta‐analysis, we will investigate reporting biases (such as publication bias) using funnel plots. We will assess funnel plot asymmetry visually. If asymmetry is suggested by a visual assessment, we will perform exploratory analyses to investigate it.
Data synthesis
We carried out statistical analysis using the Review Manager software (RevMan 2012).
We used a fixed‐effect meta‐analysis for combining data where it was reasonable to assume that studies were estimating the same underlying treatment effect: i.e. where trials were examining the same intervention, and the trials’ populations and methods were judged sufficiently similar. If there was clinical heterogeneity sufficient to expect that the underlying treatment effects differed between trials, or if we detected substantial statistical heterogeneity, we used random‐effects meta‐analysis to produce an overall summary, if an average treatment effect across trials was considered clinically meaningful. We treated the random‐effects summary as the average range of possible treatment effects and discuss the clinical implications of treatment effects differing between trials.
Where we used random‐effects analyses, we presented the results as the average treatment effect with its 95% confidence interval, and the estimates of Tau² and I².
Subgroup analysis and investigation of heterogeneity
If we identified substantial heterogeneity, we investigated it using subgroup analyses and sensitivity analyses. We considered whether an overall summary was meaningful, and if it was, used random‐effects analysis to produce it.
We planned to carry out the following a priori subgroup analyses in the overall cerclage versus no cerclage analysis. It was felt that these subgroups specify five different clinical scenarios where the effect of cerclage may differ in the direction and effect size.
-
Twin only‐indicated cerclage, e.g. the only indication for the cerclage was the twin pregnancy
-
Twin and history‐indicated cerclage, e.g. a woman deemed to be at an increased risk for a preterm birth based on prior Ob/Gyn history: prior cone biopsy, previous termination of pregnancy or first trimester miscarriage, prior preterm birth, cervical abnormality on physical examination (known prior to pregnancy, or early during pregnancy ‐ not referring to dilation), or uterine abnormality
-
Ultrasound‐indicated cerclage, e.g. carried out following the discovery of a shortened cervical length on transvaginal ultrasound examination
-
Physical exam‐indicated cerclage, e.g. carried out following the detection of advanced cervical dilation during vaginal examination, which could be either an incidental finding or a finding in women with symptoms of threatened preterm labour
-
Triplet cerclage
Outcomes to be used in the subgroup analyses are as follows.
-
Perinatal deaths
-
Serious neonatal morbidity
-
Composite ‐ perinatal deaths and serious neonatal morbidity
-
Stillbirth (fetal demise after 20 weeks' gestation, prior to delivery)
-
Neonatal death (after birth and before 29 days of neonatal life or discharge from hospital)
-
Preterm birth less than 28 weeks
-
Preterm birth less than 32 weeks
-
Preterm birth less than 34 weeks
-
Preterm birth less than 35 weeks
-
Preterm birth less than 37 weeks
-
Mean gestational age at delivery
-
Low birthweight defined as less than 2500 g
-
Very low birthweight defined as less than 1500 g
-
Respiratory distress syndrome (defined by trialists)
-
Intraventricular haemorrhage (defined by trialists)
-
Necrotising enterocolitis (defined by trialists)
-
Sepsis (defined by trialists)
-
Neonatal intensive care unit admission
-
Long‐term infant neurodevelopmental outcomes
-
Caesarean section (elective and emergency)
-
Maternal infection requiring intervention, e.g. antibiotics or delivery (including chorioamnionitis and endometritis)
-
Maternal side‐effects (vaginal discharge, bleeding, pyrexia not requiring antibiotics)
We assessed subgroup differences by interaction tests available within RevMan (RevMan 2012). We reported the results of subgroup analyses quoting the Chi² statistic and P value, and the interaction test I² value. Should data from different subgroups come from the same trial (e.g. triplets who also had a short cervical length), the data were split in the mutually exclusive groups. This approach could cause some problems in a random‐effects analysis because there may appear to be more trials than there actually are, which affects the estimate of the between‐study variation, and hence the results. We addressed this issue in additional analysis.
Sensitivity analysis
We planned to perform sensitivity analysis on the primary outcomes based on trial quality, separating high‐quality trials from trials of lower quality. 'High quality' was, for the purposes of this sensitivity analysis, defined as a trial having 'low risk of bias' for sequence generation and allocation concealment. While this sensitivity analysis was not carried out due to the low numbers of studies involved, this analyses will be carried out in future updates of this review as more data become available.
Results
Description of studies
Results of the search
(See: Figure 1). The search of the Cochrane Pregnancy and Childbirth Group's Trials Register retrieved 25 trial reports, of which five trials (22 reports) have been included, which in total randomised 1577 women, encompassing both singleton and multiple gestations. For the purposes of this review, after excluding singletons, the final analysis included 128 women (of which 122 women had twin gestations, and six women had triplet gestations). Three trials have been excluded. For further details of trial characteristics, please refer to the tables of Characteristics of included studies and Characteristics of excluded studies.
Study flow diagram.
Included studies
All of the included studies specified a cerclage versus no cerclage comparison (Althuisius 2001; Berghella 2004; Dor 1982; MRC/RCOG 1993; Rust 2001). Three of these studies required women to undertake some form of bedrest both in the intervention (cerclage) and control (no cerclage) groups (Althuisius 2001; Berghella 2004; Rust 2001).
In one trial, twins conceived via ovulation induction were included, with cerclage placement shortly at or after 13 gestational weeks (Dor 1982). This therefore represents assessment of cerclage based solely for the risk factor of multiple gestation (twin only‐indicated cerclage).
In another trial, increased risk for preterm birth was based on prior obstetric history (MRC/RCOG 1993), with subsequent cervical cerclage placement occurring "as soon as possible". This study therefore assessed the effect of twin and history‐indicated cerclage.
Three studies assessed women at high risk for preterm birth and identified those at higher risk based on short cervical length via transvaginal ultrasound (Althuisius 2001; Berghella 2004; Rust 2001), thus having cerclage placement in the early to mid second trimester of pregnancy. These studies therefore assessed the effect of ultrasound‐indicated cerclage.
One of these studies involved twins only (Dor 1982), while the remaining four studies contained both singleton and multiple pregnancies (Althuisius 2001; Berghella 2004; MRC/RCOG 1993; Rust 2001).
There were no studies comparing different indications for cerclage (e.g. history‐indicated versus ultrasound‐indicated cerclage). There were no studies comparing cerclage to another intervention (e.g. progesterone).
We were able to obtain the original databases from three of the five trials (Althuisius 2001; Berghella 2004; Rust 2001).
For further information, please refer to tables of Characteristics of included studies.
Excluded studies
Three studies were excluded. Two of these studies excluded multiple gestations (Blair 2002; Lazar 1984). The third trial (Nicolaides 2001) had only methodology and design published ‐ the current status of this trial is unknown.
Risk of bias in included studies
The quality of the five included studies was difficult to assess due to lack of information in some studies regarding randomisation, selection bias, and detection bias.
Please see Figure 2 and Figure 3 for summary of 'Risk of bias' assessments.
'Risk of bias' graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.
'Risk of bias' summary: review authors' judgements about each risk of bias item for each included study.
Allocation
Two studies had both adequate random sequence generation and concealment allocation (Berghella 2004; Rust 2001). In the remaining three studies (Althuisius 2001; Dor 1982; MRC/RCOG 1993), both sequence generation and allocation concealment were unclear.
Blinding
Blinding of both participants and personnel was not possible given the nature of the intervention. Regarding detection bias, in one trial (Berghella 2004), the outcome assessors were not blinded; none of the remaining four trials were clear if the outcome assessors were blinded.
Incomplete outcome data
Four studies adequately addressed the issue of incomplete outcome data assessment (attrition bias) (Althuisius 2001; Berghella 2004; MRC/RCOG 1993; Rust 2001), while in the remaining study (Dor 1982), the quality of outcome data assessment (and incomplete outcome data) was unclear.
Selective reporting
None of the five study protocols were available, but in three of the five studies (Althuisius 2001; Berghella 2004; Rust 2001), the original database was supplied by the primary first authors, thus those three seem to be free of selective reporting. In one of the studies (MRC/RCOG 1993), while the study protocol was not available, the authors provided the individual data on an extraction form, and thus was deemed low risk for reporting bias. There is an unclear risk of reporting bias in one of the studies (Dor 1982), as the full study protocol was not available.
Other potential sources of bias
Three studies were judged to be free of other sources of bias (Althuisius 2001; Berghella 2004; MRC/RCOG 1993), while the remaining two studies (Dor 1982; Rust 2001) were judged as unclear.
Effects of interventions
Comparison 1 ‐ Cerclage versus no cerclage
Primary outcomes
When cerclage was compared with no cerclage in all trials, there was no statistically significant difference in perinatal deaths (19.2% versus 9.5%; risk ratio (RR) 1.74, 95% confidence intervals (CI) 0.92 to 3.28, five studies, n = 262) (Analysis 1.1) or serious neonatal morbidity (15.8% versus 13.6%; average RR 0.96, 95% CI 0.13 to 7.10, Tau² = 1.88, I² = 60%, three studies, n = 116) (Analysis 1.2). There was a higher rate of composite perinatal death and serious neonatal morbidity in the cerclage group compared with the no cerclage group (40.4% versus 20.3%, respectively), but the difference was also not statistically significant (average RR 1.54, 95% CI 0.58 to 4.11, Tau² = 0.41, I² = 44%, three studies, n = 116) (Analysis 1.3). The CIs for the primary outcomes are wide, thus it is not possible to exclude either benefit or harm to the liveborn baby from cerclage.
Secondary outcomes
Among the prespecified secondary outcomes, there were no significant differences between the cerclage and the no cerclage groups.
-
Stillbirth (fetal demise after 20 weeks' gestation, prior to delivery) ‐ RR 0.26, 95% CI 0.01 to 5.26, four studies, n = 188 (Analysis 1.4).
-
Neonatal death (after birth, and before 29 days of neonatal life or discharge from hospital) ‐ RR 1.60, 95% CI 0.69 to 3.74, four studies, n = 188 (Analysis 1.5).
-
Preterm birth less than 28 weeks ‐ RR 1.54, 95% CI 0.63 to 3.81, five studies, n = 128 (Analysis 1.6).
-
Preterm birth less than 32 weeks ‐ RR 1.43, 95% CI 0.72 to 2.83, four studies, n = 83 (Analysis 1.7).
-
Preterm birth less than 34 weeks ‐ average RR 1.16, 95% CI 0.44 to 3.06, Tau² = 0.67, I² = 58%, four studies, n = 83 (Analysis 1.8).
-
Preterm birth less than 35 weeks ‐ average RR 1.11, 95% CI 0.58 to 2.14, Tau² = 0.28, I² = 52%, four studies, n = 83 (Analysis 1.9).
-
Preterm birth less than 37 weeks ‐ RR 1.13, 95% CI 0.89 to 1.43, five studies, n = 128 (Analysis 1.10).
-
Mean gestational age at delivery ‐ Mean difference (MD) ‐0.95, 95% CI ‐2.64 to 0.75, four studies, n = 83 (Analysis 1.11).
-
Low birthweight less than 2500 g ‐ average RR 1.10, 95% CI 0.82 to 1.48, Tau² = 0.07, I² = 60%, four studies, n = 172 (Analysis 1.12).
-
Very low birthweight less than 1500 g ‐ average RR 1.42, 95% CI 0.52 to 3.85, Tau² = 0.73, I² = 66%, four studies, n = 172 (Analysis 1.13).
-
Respiratory distress syndrome ‐ average RR 1.70, 95% CI 0.15 to 18.77, Tau² = 3.13, I² = 70%, three studies, n = 116 (Analysis 1.14).
-
Intraventricular haemorrhage ‐ RR 0.88, 95% CI 0.25 to 3.12, three studies, n = 116 (Analysis 1.15).
-
Necrotising enterocolitis ‐ Not estimable, three studies, n = 116 (Analysis 1.16).
-
Sepsis ‐ RR 0.27, 95% CI 0.03 to 2.31, three studies, n = 116 (Analysis 1.17).
-
Neonatal intensive care unit admission ‐ average RR 0.35, 95% CI 0.06 to 2.12, Tau² = 0.99, I² = 48% , two studies, n = 42 (Analysis 1.18).
-
Long‐term infant neurodevelopmental outcomes ‐ Not estimable, no studies (Analysis 1.19).
-
Caesarean section (elective and emergency) ‐ RR 1.24, 95% CI 0.65 to 2.35, three studies, n = 77 (Analysis 1.20).
-
Maternal infection requiring intervention, e.g. antibiotics or delivery (including chorioamnionitis and endometritis) ‐ Not estimable, two studies, n = 45 (Analysis 1.21).
-
Maternal side‐effects (vaginal discharge, bleeding, pyrexia not requiring antibiotics) ‐ RR 3.92, 95% CI 0.17 to 88.67, one study, n = 28 (Analysis 1.22).
These results remained non‐significant when the Rust 2001 triplet cohort was combined with the ultrasound‐indicated cohort, as above.
No trials reported on long‐term infant neurodevelopmental outcomes.
Subgroup analysis
Examining the differences between prespecified subgroups, ultrasound‐indicated cerclage was associated with an increased risk of low birthweight (average RR 1.39, 95% CI 1.06 to 1.83, Tau² = 0.01, I² = 15%, three studies, n = 98 (Analysis 1.12)), very low birthweight (average RR 3.31, 95% CI 1.58 to 6.91, Tau² = 0, I² = 0%, three studies, n = 98 (Analysis 1.13), and respiratory distress syndrome (average RR 5.07, 95% CI 1.75 to 14.70, Tau² = 0, I2 = 0%, three studies, n = 98 (Analysis 1.14)). However, given the low number of trials, as well as substantial heterogeneity and subgroup differences (specifically among Analysis 1.12 and Analysis 1.13), these data must be interpreted cautiously.
We have tested the impact of analysing two subgroups (triplets; ultrasound‐indicated cerclage) in the Rust 2001 as if they are coming from two separate studies. Our main analyses treated these two subgroups as different studies in order to graphically depict the subset of triplets as an analysis group. For completeness, the triplet and ultrasound‐indicated subgroup from Rust 2001 were also combined given that these subgroups were in fact from a single trial. This was done so as not to affect the estimate of the between‐study variation, and hence the results.
Neither the results for serious neonatal morbidity nor for composite perinatal death and serious neonatal morbidity were significantly altered, as the relative risks for the primary outcomes after combining the subgroups were:
-
Perintal deaths ‐ RR 1.75, 95% CI 0.93 to 3.29.
-
Serious neonatal morbidity ‐ RR 1.26, 95% CI 0.53 to 2.97.
-
Composite perinatal deaths and serious neonatal morbidity ‐ RR 1.70, 95% CI 0.95 to 3.05.
Cervical stitch versus any other preventative therapy (i.e. progesterone)
We did not identify any trials comparing cervical stitch versus any other preventative therapy (e.g. progesterone) in multiple gestations.
Comparisons between different cerclage protocols
We did not identify any trials comparing different cerclage protocols (history‐indicated versus ultrasound‐indicated versus physical exam‐indicated cerclage) in multiple gestations.
Discussion
Summary of main results
Examining the evidence from the five included randomised trials, in women with multiple gestations, placing a cerclage is not associated with a significant difference regarding perinatal deaths or neonatal morbidity. As only 128 women were randomised in the five included studies, the data are insufficient for meaningful conclusions, and this area deserves further study.
Unfortunately, none of the trials examined long‐term neurodevelopmental outcomes among surviving neonates, and therefore the effect of cerclage could not be estimated for this outcome. Hence, the question of whether prolonging a pregnancy in what could be a 'hostile' intrauterine environment actually does more harm than good, remains unanswered. In addition, while among three studies (Berghella 2004; Dor 1982; MRC/RCOG 1993), it does not appear that cerclage has any effect on caesarean section; while in one study (MRC/RCOG 1993) cerclage did not appear to have an effect on maternal side‐effects, maternal infection could not be estimated, so it's unclear if cerclage in multiple gestations puts maternal health at any degree of risk.
It was also not possible to evaluate cerclage versus other modalities (e.g. progesterone), nor was it possible to examine the effect of one cerclage modality versus another (e.g. history‐indicated versus ultrasound‐indicated cerclage), as there are no trials evaluating these comparisons.
As cerclage may have a different effect depending on the indication for the procedure, it is probably more clinically meaningful to evaluate the data according to the indication of cerclage in multiples, rather than analysing the totality of the data. Therefore, we prespecified five clinical scenarios (subgroups) based on the indications for cervical cerclage in current clinical practice.
We were only able to obtain the original databases from three of the five included studies (Althuisius 2001; Berghella 2004; Rust 2001), making these subgroup analyses somewhat limited. Given the low number of trials, as well as significant heterogeneity it is not possible to draw any meaningful conclusions from the currently available data. While an ultrasound‐indicated cerclage appears to be associated with an increased risk for the composite adverse neonatal outcome, including respiratory distress syndrome, low and very low birthweight, given there is no evidence of a consistent subgroup effect, the observed results can most likely be attributed to chance.
Without adequately powered randomised clinical trials, questions will remain regarding any theoretical benefit (or harm) involving the indication of cerclage placement in multiple gestations.
The issue of prevention of preterm birth, especially among multiple gestations, is definitely a hot topic in the field of obstetrics today. Progesterone, in its various forms, has been shown to prevent preterm birth in singleton gestations, both in its synthetic caproate form among women with a prior preterm birth (Meis 2003), as well as in its micronised form among women with shortened cervical lengths (Hassan 2011). This prevention of preterm birth with progesterone use has not yet translated to those women with multiple gestations (Klein 2011; Rouse 2007). However, a recent individual patient data meta‐analysis, although limited by small numbers of twin pregnancies, suggests that vaginal progesterone may have some role in multiple pregnancies with short cervix (Romero 2012).
Overall completeness and applicability of evidence
Given the limited numbers of trials, the overall numbers are small, and subgroup effects were difficult to assess with adequate power, which is a weakness of this meta‐analysis. An additional limitation is the lack of long‐term neonatal neurodevelopmental data, an important outcome to consider when implementing an antenatal intervention, regardless of the timing of this intervention. As there is considerable anxiety concerning preterm birth in multiple gestations, there is a pressure on clinicians to "do everything," which results in placing a cerclage, or mechanical support, as this seems intuitive. Thus far, however, despite the small numbers and potential biases in this analysis, there does not appear to be a benefit in multiple gestations with cerclage placement. More evidence needs to be obtained through properly conducted prospective trials to further address this clinical question.
Quality of the evidence
Overall, the included trials were of average to above average quality. Only two of the five included studies (Berghella 2004, Rust 2001) were judged to be at low risk for selection bias, with three of the trials at unclear risks regarding both selection and detection biases (Althuisius 2001; Dor 1982; MRC/RCOG 1993). Selective reporting of results is a concern when the trial protocols are not available for review. While this concern was alleviated with the providing of the original databases by three of the five study authors (Althuisius 2001; Berghella 2004; Rust 2001), the original databases were not available for review for the remaining two trials (Dor 1982; MRC/RCOG 1993). Additional data were entered into the collection sheets by the MRC/RCOG group.
Performance bias (blinding of both participants and personnel) is always (and will always be) an issue regarding cerclage trials, in that it is impossible to blind. The outcomes of interest, however, are unlikely to be affected by this inability to blind.
Potential biases in the review process
None identified.
Agreements and disagreements with other studies or reviews
Cervical cerclage in multiple gestations remains controversial, with no definitive evidence to date of its benefit at prolonging gestation or reducing neonatal mortality/morbidity. The current analysis demonstrates that overall cervical cerclage has no significant effect on perinatal death, neonatal death, or preterm birth. Interestingly, composite perinatal death and serious neonatal morbidity was significantly increased when limiting the analysis to the subgroup of ultrasound‐indicated cerclage.
Concerning preterm birth, while another meta‐analysis examining short cervical length and cerclage (Berghella 2005) demonstrated a 215% increase in the risk of preterm birth less than 35 weeks in twins receiving an ultrasound‐indicated cerclage, the current meta‐analysis did not demonstrate a significant difference in preterm birth comparing ultrasound‐indicated cerclage with no cerclage. A retrospective analysis (Roman 2005) also did not demonstrate a reduction of spontaneous preterm delivery less than 28, 30, 32, or 34 weeks with the use of cerclage in multiple gestations (both twins and triplets). An additional large retrospective analysis of triplet gestations (Rebarber 2005) concluded that a prophylactic cerclage did not result in improved pregnancy or neonatal outcomes in triplet pregnancies without a history of cervical insufficiency.
A prospective, non‐randomised trial (thus not included in this review) (Newman 2002), looked at cerclage versus no cerclage among twin pregnancies with a shortened mid‐trimester cervical length, and did not find differences in length of gestation, birthweight, delivery at less than 34 weeks, preterm premature rupture of fetal membranes, or very low birthweight.
These current data, given the low numbers of participants, need to be interpreted with caution.
Study flow diagram.
'Risk of bias' graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.
'Risk of bias' summary: review authors' judgements about each risk of bias item for each included study.
Comparison 1 Cerclage versus no cerclage, Outcome 1 Perinatal deaths.
Comparison 1 Cerclage versus no cerclage, Outcome 2 Serious neonatal morbidity (defined by trialists).
Comparison 1 Cerclage versus no cerclage, Outcome 3 Composite – Perinatal deaths and serious neonatal morbidity.
Comparison 1 Cerclage versus no cerclage, Outcome 4 Stillbirth (fetal demise after 20 weeks’ gestation, prior to delivery).
Comparison 1 Cerclage versus no cerclage, Outcome 5 Neonatal death (after birth, and before 29 days of neonatal life or discharge from hospital).
Comparison 1 Cerclage versus no cerclage, Outcome 6 Preterm birth less than 28 weeks.
Comparison 1 Cerclage versus no cerclage, Outcome 7 Preterm birth less than 32 weeks.
Comparison 1 Cerclage versus no cerclage, Outcome 8 Preterm birth less than 34 weeks.
Comparison 1 Cerclage versus no cerclage, Outcome 9 Preterm birth less than 35 weeks.
Comparison 1 Cerclage versus no cerclage, Outcome 10 Preterm birth less than 37 weeks.
Comparison 1 Cerclage versus no cerclage, Outcome 11 Mean gestational age at delivery.
Comparison 1 Cerclage versus no cerclage, Outcome 12 Low birthweight defined as less than 2500 grams.
Comparison 1 Cerclage versus no cerclage, Outcome 13 Very low birthweight defined as less than 1500 grams.
Comparison 1 Cerclage versus no cerclage, Outcome 14 Respiratory distress syndrome (defined by trialists).
Comparison 1 Cerclage versus no cerclage, Outcome 15 Intraventricular hemorrhage (defined by trialists).
Comparison 1 Cerclage versus no cerclage, Outcome 16 Necrotising enterocolitis (defined by trialists).
Comparison 1 Cerclage versus no cerclage, Outcome 17 Sepsis (defined by trialists).
Comparison 1 Cerclage versus no cerclage, Outcome 18 Neonatal intensive care unit admission.
Comparison 1 Cerclage versus no cerclage, Outcome 20 Caesarean section (elective and emergency).
Comparison 1 Cerclage versus no cerclage, Outcome 21 Maternal infection requiring intervention, e.g. antibiotics or delivery (including chorioamnionitis and endometritis).
Comparison 1 Cerclage versus no cerclage, Outcome 22 Maternal side‐effects (vaginal discharge, bleeding, pyrexia not requiring antibiotics).
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Perinatal deaths Show forest plot | 5 | 262 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.74 [0.92, 3.28] |
| 1.1 Twin only‐indicated | 1 | 90 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.19 [0.47, 3.02] |
| 1.2 Twin and history‐indicated (prior Ob/Gyn history) | 1 | 56 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.33 [0.20, 8.80] |
| 1.3 Ultrasound‐indicated | 3 | 98 | Risk Ratio (M‐H, Fixed, 95% CI) | 2.66 [0.83, 8.54] |
| 1.4 Physical exam‐indicated | 0 | 0 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 1.5 Triplets | 1 | 18 | Risk Ratio (M‐H, Fixed, 95% CI) | 3.0 [0.38, 23.68] |
| 2 Serious neonatal morbidity (defined by trialists) Show forest plot | 3 | 116 | Risk Ratio (M‐H, Random, 95% CI) | 0.96 [0.13, 7.10] |
| 2.1 Twin only‐indicated | 0 | 0 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] |
| 2.2 Twin and history‐indicated (prior Ob/Gyn history) | 0 | 0 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] |
| 2.3 Ultrasound‐indicated | 3 | 98 | Risk Ratio (M‐H, Random, 95% CI) | 2.59 [0.85, 7.86] |
| 2.4 Physical exam‐indicated | 0 | 0 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] |
| 2.5 Triplets | 1 | 18 | Risk Ratio (M‐H, Random, 95% CI) | 0.11 [0.01, 1.80] |
| 3 Composite – Perinatal deaths and serious neonatal morbidity Show forest plot | 3 | 116 | Risk Ratio (M‐H, Random, 95% CI) | 1.54 [0.58, 4.11] |
| 3.1 Twin only‐indicated | 0 | 0 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] |
| 3.2 Twin and history‐indicated (prior Ob/Gyn history) | 0 | 0 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] |
| 3.3 Ultrasound‐indicated | 3 | 98 | Risk Ratio (M‐H, Random, 95% CI) | 2.52 [1.20, 5.30] |
| 3.4 Physical exam‐indicated | 0 | 0 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] |
| 3.5 Triplets | 1 | 18 | Risk Ratio (M‐H, Random, 95% CI) | 0.6 [0.20, 1.79] |
| 4 Stillbirth (fetal demise after 20 weeks’ gestation, prior to delivery) Show forest plot | 4 | 188 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.26 [0.01, 5.26] |
| 4.1 Twin only‐indicated | 1 | 90 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 4.2 Twin and history‐indicated (prior Ob/Gyn history) | 1 | 56 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.26 [0.01, 5.26] |
| 4.3 Ultrasound‐indicated | 2 | 42 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 4.4 Physical exam‐indicated | 0 | 0 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 4.5 Triplets | 0 | 0 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 5 Neonatal death (after birth, and before 29 days of neonatal life or discharge from hospital) Show forest plot | 4 | 188 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.60 [0.69, 3.74] |
| 5.1 Twin only‐indicated | 1 | 90 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.19 [0.47, 3.02] |
| 5.2 Twin and history‐indicated (prior Ob/Gyn history) | 1 | 56 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 5.3 Ultrasound‐indicated | 2 | 42 | Risk Ratio (M‐H, Fixed, 95% CI) | 5.57 [0.44, 70.55] |
| 5.4 Physical exam‐indicated | 0 | 0 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 5.5 Triplets | 0 | 0 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 6 Preterm birth less than 28 weeks Show forest plot | 5 | 128 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.54 [0.63, 3.81] |
| 6.1 Twin only‐indicated | 1 | 45 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.52 [0.05, 5.36] |
| 6.2 Twin and history‐indicated (prior Ob/Gyn history) | 1 | 28 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.33 [0.09, 19.23] |
| 6.3 Ultrasound‐indicated | 3 | 49 | Risk Ratio (M‐H, Fixed, 95% CI) | 2.62 [0.72, 9.51] |
| 6.4 Physical exam‐indicated | 0 | 0 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 6.5 Triplets | 1 | 6 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.0 [0.10, 9.61] |
| 7 Preterm birth less than 32 weeks Show forest plot | 4 | 83 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.43 [0.72, 2.83] |
| 7.1 Twin only‐indicated | 0 | 0 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 7.2 Twin and history‐indicated (prior Ob/Gyn history) | 1 | 28 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.33 [0.04, 2.61] |
| 7.3 Ultrasound‐indicated | 3 | 49 | Risk Ratio (M‐H, Fixed, 95% CI) | 2.48 [0.96, 6.37] |
| 7.4 Physical exam‐indicated | 0 | 0 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 7.5 Triplets | 1 | 6 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.0 [0.32, 3.10] |
| 8 Preterm birth less than 34 weeks Show forest plot | 4 | 83 | Risk Ratio (M‐H, Random, 95% CI) | 1.16 [0.44, 3.06] |
| 8.1 Twin only‐indicated | 0 | 0 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] |
| 8.2 Twin and history‐indicated (prior Ob/Gyn history) | 1 | 28 | Risk Ratio (M‐H, Random, 95% CI) | 0.27 [0.04, 1.99] |
| 8.3 Ultrasound‐indicated | 3 | 49 | Risk Ratio (M‐H, Random, 95% CI) | 2.19 [0.72, 6.63] |
| 8.4 Physical exam‐indicated | 0 | 0 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] |
| 8.5 Triplets | 1 | 6 | Risk Ratio (M‐H, Random, 95% CI) | 0.71 [0.31, 1.66] |
| 9 Preterm birth less than 35 weeks Show forest plot | 4 | 83 | Risk Ratio (M‐H, Random, 95% CI) | 1.11 [0.58, 2.14] |
| 9.1 Twin only‐indicated | 0 | 0 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] |
| 9.2 Twin and history‐indicated (prior Ob/Gyn history) | 1 | 28 | Risk Ratio (M‐H, Random, 95% CI) | 0.22 [0.03, 1.61] |
| 9.3 Ultrasound‐indicated | 3 | 49 | Risk Ratio (M‐H, Random, 95% CI) | 1.63 [0.88, 3.02] |
| 9.4 Physical exam‐indicated | 0 | 0 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] |
| 9.5 Triplets | 1 | 6 | Risk Ratio (M‐H, Random, 95% CI) | 0.71 [0.31, 1.66] |
| 10 Preterm birth less than 37 weeks Show forest plot | 5 | 128 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.13 [0.89, 1.43] |
| 10.1 Twin only‐indicated | 1 | 45 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.95 [0.51, 1.78] |
| 10.2 Twin and history‐indicated (prior Ob/Gyn history) | 1 | 28 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.33 [0.71, 2.51] |
| 10.3 Ultrasound‐indicated | 3 | 49 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.18 [0.91, 1.53] |
| 10.4 Physical exam‐indicated | 0 | 0 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 10.5 Triplets | 1 | 6 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.0 [0.59, 1.69] |
| 11 Mean gestational age at delivery Show forest plot | 4 | 83 | Mean Difference (IV, Fixed, 95% CI) | ‐0.95 [‐2.64, 0.75] |
| 11.1 Twin only‐indicated | 0 | 0 | Mean Difference (IV, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 11.2 Twin and history‐indicated (prior Ob/Gyn history) | 1 | 28 | Mean Difference (IV, Fixed, 95% CI) | 0.70 [‐3.50, 4.90] |
| 11.3 Ultrasound‐indicated | 3 | 49 | Mean Difference (IV, Fixed, 95% CI) | ‐1.24 [‐3.13, 0.66] |
| 11.4 Physical exam‐indicated | 0 | 0 | Mean Difference (IV, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 11.5 Triplets | 1 | 6 | Mean Difference (IV, Fixed, 95% CI) | ‐2.0 [‐11.23, 7.23] |
| 12 Low birthweight defined as less than 2500 grams Show forest plot | 4 | 172 | Risk Ratio (M‐H, Random, 95% CI) | 1.10 [0.82, 1.48] |
| 12.1 Twin only‐indicated | 0 | 0 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] |
| 12.2 Twin and history‐indicated (prior Ob/Gyn history) | 1 | 56 | Risk Ratio (M‐H, Random, 95% CI) | 0.91 [0.62, 1.34] |
| 12.3 Ultrasound‐indicated | 3 | 98 | Risk Ratio (M‐H, Random, 95% CI) | 1.39 [1.06, 1.83] |
| 12.4 Physical exam‐indicated | 0 | 0 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] |
| 12.5 Triplets | 1 | 18 | Risk Ratio (M‐H, Random, 95% CI) | 0.79 [0.54, 1.16] |
| 13 Very low birthweight defined as less than 1500 grams Show forest plot | 4 | 172 | Risk Ratio (M‐H, Random, 95% CI) | 1.42 [0.52, 3.85] |
| 13.1 Twin only‐indicated | 0 | 0 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] |
| 13.2 Twin and history‐indicated (prior Ob/Gyn history) | 1 | 56 | Risk Ratio (M‐H, Random, 95% CI) | 0.19 [0.03, 1.45] |
| 13.3 Ultrasound‐indicated | 3 | 98 | Risk Ratio (M‐H, Random, 95% CI) | 3.31 [1.58, 6.91] |
| 13.4 Physical exam‐indicated | 0 | 0 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] |
| 13.5 Triplets | 1 | 18 | Risk Ratio (M‐H, Random, 95% CI) | 1.0 [0.52, 1.92] |
| 14 Respiratory distress syndrome (defined by trialists) Show forest plot | 3 | 116 | Risk Ratio (M‐H, Random, 95% CI) | 1.70 [0.15, 18.77] |
| 14.1 Twin only‐indicated | 0 | 0 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] |
| 14.2 Twin and history‐indicated (prior Ob/Gyn history) | 0 | 0 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] |
| 14.3 Ultrasound‐indicated | 3 | 98 | Risk Ratio (M‐H, Random, 95% CI) | 5.07 [1.75, 14.70] |
| 14.4 Physical exam‐indicated | 0 | 0 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] |
| 14.5 Triplets | 1 | 18 | Risk Ratio (M‐H, Random, 95% CI) | 0.11 [0.01, 1.80] |
| 15 Intraventricular hemorrhage (defined by trialists) Show forest plot | 3 | 116 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.88 [0.25, 3.12] |
| 15.1 Twin only‐indicated | 0 | 0 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 15.2 Twin and history‐indicated (prior Ob/Gyn history) | 0 | 0 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 15.3 Ultrasound‐indicated | 3 | 98 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.13 [0.27, 4.74] |
| 15.4 Physical exam‐indicated | 0 | 0 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 15.5 Triplets | 1 | 18 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.33 [0.02, 7.24] |
| 16 Necrotising enterocolitis (defined by trialists) Show forest plot | 3 | 116 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 16.1 Twin only‐indicated | 0 | 0 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 16.2 Twin and history‐indicated (prior Ob/Gyn history) | 0 | 0 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 16.3 Ultrasound‐indicated | 3 | 98 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 16.4 Physical exam‐indicated | 0 | 0 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 16.5 Triplets | 1 | 18 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 17 Sepsis (defined by trialists) Show forest plot | 3 | 116 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.27 [0.03, 2.31] |
| 17.1 Twin only‐indicated | 0 | 0 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 17.2 Twin and history‐indicated (prior Ob/Gyn history) | 0 | 0 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 17.3 Ultrasound‐indicated | 3 | 98 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.23 [0.01, 4.58] |
| 17.4 Physical exam‐indicated | 0 | 0 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 17.5 Triplets | 1 | 18 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.33 [0.02, 7.24] |
| 18 Neonatal intensive care unit admission Show forest plot | 2 | 42 | Risk Ratio (M‐H, Random, 95% CI) | 0.35 [0.06, 2.12] |
| 18.1 Twin only‐indicated | 0 | 0 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] |
| 18.2 Twin and history‐indicated (prior Ob/Gyn history) | 0 | 0 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] |
| 18.3 Ultrasound‐indicated | 2 | 42 | Risk Ratio (M‐H, Random, 95% CI) | 0.35 [0.06, 2.12] |
| 18.4 Physical exam‐indicated | 0 | 0 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] |
| 18.5 Triplets | 0 | 0 | Risk Ratio (M‐H, Random, 95% CI) | 0.0 [0.0, 0.0] |
| 19 Long‐term infant neurodevelopmental outcomes | 0 | 0 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 19.1 Twin only‐indicated | 0 | 0 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 19.2 Twin and history‐indicated (prior Ob/Gyn history) | 0 | 0 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 19.3 Ultrasound‐indicated | 0 | 0 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 19.4 Physical exam‐indicated | 0 | 0 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 19.5 Triplets | 0 | 0 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 20 Caesarean section (elective and emergency) Show forest plot | 3 | 77 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.24 [0.65, 2.35] |
| 20.1 Twin only‐indicated | 1 | 45 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.34 [0.61, 2.98] |
| 20.2 Twin and history‐indicated (prior Ob/Gyn history) | 1 | 28 | Risk Ratio (M‐H, Fixed, 95% CI) | 1.07 [0.36, 3.14] |
| 20.3 Ultrasound‐indicated | 1 | 4 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 20.4 Physical exam‐indicated | 0 | 0 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 20.5 Triplets | 0 | 0 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 21 Maternal infection requiring intervention, e.g. antibiotics or delivery (including chorioamnionitis and endometritis) Show forest plot | 2 | 45 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 21.1 Twin only‐indicated | 0 | 0 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 21.2 Twin and history‐indicated (prior Ob/Gyn history) | 1 | 28 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 21.3 Ultrasound‐indicated | 1 | 17 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 21.4 Physical exam‐indicated | 0 | 0 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 21.5 Triplets | 0 | 0 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 22 Maternal side‐effects (vaginal discharge, bleeding, pyrexia not requiring antibiotics) Show forest plot | 1 | 28 | Risk Ratio (M‐H, Fixed, 95% CI) | 3.92 [0.17, 88.67] |
| 22.1 Twin only‐indicated | 0 | 0 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 22.2 Twin and history‐indicated (prior Ob/Gyn history) | 1 | 28 | Risk Ratio (M‐H, Fixed, 95% CI) | 3.92 [0.17, 88.67] |
| 22.3 Ultrasound‐indicated | 0 | 0 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 22.4 Physical exam‐indicated | 0 | 0 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
| 22.5 Triplets | 0 | 0 | Risk Ratio (M‐H, Fixed, 95% CI) | 0.0 [0.0, 0.0] |
