Find synonyms Find exact match
INTRODUCTION — Classically, the term “cervical insufficiency” was used to describe a disorder in which painless cervical dilation led to recurrent second trimester pregnancy losses/births of otherwise normal pregnancies. Structural weakness of cervical tissue was thought to cause or contribute to these adverse outcomes. The diagnosis has also been applied to women with one or two such losses/deliveries or at risk for such a loss/delivery.
The American College of Obstetricians and Gynecologists (ACOG) define cervical insufficiency as the inability of the uterine cervix to retain a pregnancy in the absence of the signs and symptoms of clinical contractions, or labor, or both in the second trimester . The ultrasonographic finding of a short cervical length in the second trimester is not sufficient for the diagnosis of cervical insufficiency.
Although structural cervical weakness is the source of some preterm losses/births, most are caused by other disorders, such as decidual inflammation/infection, hemorrhage, or uterine overdistension. These disorders can initiate biochemical changes in the cervix that lead to premature cervical shortening and, often, pregnancy loss or preterm birth. These disorders usually do not recur in successive pregnancies. (See "Pathogenesis of spontaneous preterm birth".)
Past obstetrical history — Historical factors suggestive of cervical insufficiency include a history of second trimester pregnancy losses/deliveries, especially in the setting of short labors or progressively earlier deliveries in successive pregnancies.
Cervical risk factors — Risk factors for cervical insufficiency can be congenital or acquired; acquired factors are more common.
●Collagen abnormalities — Alterations in the regulation of type I collagen expression may influence cervical ripening [2,3], and genetic disorders affecting collagen (eg, Ehlers-Danlos syndrome) have been associated with increased risk of preterm birth . This may explain familial aggregation of cervical insufficiency . As an example, in one study, 34 of 125 (27 percent) women with cervical insufficiency had a first degree relative with the same diagnosis, but none of the 165 unaffected women had a family history of cervical insufficiency .
●Uterine anomalies — Congenital structural uterine abnormalities may involve the cervix and be associated with cervical insufficiency . The risk of second trimester preterm birth is increased with uterine anomalies, including canalization defects (eg, septate uterus), unification defects (eg, bicornuate uterus), and even arcuate uterus . (See "Clinical manifestations and diagnosis of congenital anomalies of the uterus".)
●Diethylstilbestrol (DES) exposure — In utero DES exposure has been linked to subsequent cervical insufficiency when the DES daughter became pregnant . Pregnancy in DES daughters is now uncommon, as almost all of these women are over 40 years old. (See "Outcome and follow-up of diethylstilbestrol (DES) exposed individuals".)
●Biologic variation — The length of the cervix during the second trimester in an obstetric population is distributed in a bell-shaped curve. (See "Transvaginal ultrasound assessment of the cervix and prediction of spontaneous preterm birth", section on 'Short cervical length'.) The wide range in normal cervical length (the 10th and 90th percentiles are 25 and 45 mm, respectively) during this period is due, in part, to biologic variation, but may also result from premature cervical effacement. Although a short cervix is predictive of preterm birth, it is not diagnostic of cervical insufficiency and many women who have a congenitally short cervix deliver at term.
●Obstetric trauma — A cervical laceration may occur during labor or in the process of delivery, including spontaneous, forceps, vacuum, or cesarean birth . This might weaken the cervix, and contribute to cervical insufficiency.
●Mechanical dilation — Cervical insufficiency has also been attributed to rapid mechanical dilation of the cervix during gynecologic procedures (eg, dilation and curettage [D&C], dilation and evacuation [D&E], pregnancy termination, hysteroscopy) [10,11]. In women with a short cervical length and no prior preterm birth, prior cervical mechanical dilatation is one of the most common associated risk factors. In a meta-analysis, an increasing number of voluntary pregnancy terminations was associated with an increasing risk of spontaneous preterm birth . Trauma to the cervix during gynecologic procedures may be prevented if the cervix is gradually ripened with laminaria or misoprostol, or if pregnancy termination is performed medically [11-13]. (See "Overview of pregnancy termination", section on 'Future pregnancies'.)
●Treatment of cervical intraepithelial neoplasia — Treatment of cervical intraepithelial neoplasia has been associated with an increased risk of subsequent pregnancy loss/preterm birth. This topic is discussed in more detail separately. (See "Cervical intraepithelial neoplasia: Reproductive effects of treatment".)
Symptoms — Women with cervical insufficiency may be asymptomatic or may present with mild symptoms such as pelvic pressure, premenstrual-like cramping or backache, and increased vaginal discharge. These symptoms have usually been present over several days or weeks, typically beginning around 14 to 20 weeks of gestation. A slight change in the color (from clear, white, or light yellow to pink, tan, or spotting) and consistency (thinner) of vaginal discharge is also consistent with cervical ripening from any cause.
Contractions are absent or mild.
Physical examination — The initial clinical examination may reveal a soft somewhat effaced cervix, with minimal, if any, dilation . Application of suprapubic or fundal pressure, or Valsalva, as a provocative maneuver rarely reveals membranes in the endocervical canal or vagina; this is always abnormal. Tocodynamometry usually reveals no or infrequent contractions at irregular intervals.
Late clinical presentation is characterized by advanced dilation and effacement (eg, ≥4 cm dilated and ≥80 percent effaced), spotting, unprovoked grossly prolapsed membranes or ruptured membranes, or contractions that seem inadequate to explain the advanced effacement and dilation.
The presence of infection, bleeding from placental causes (eg, previa or abruption), or multiple gestation, which are associated with preterm birth independent of cervical insufficiency, suggests an alternative diagnosis.
Imaging — Before 28 weeks of gestation, cervical length by transvaginal ultrasound below the 10th percentile (25 mm) is consistently associated with an increased risk of spontaneous preterm birth. Other sonographic features suggestive of increased risk of preterm birth include separation of the membranes and debris (sludge) in the amniotic fluid. These findings are described in detail separately. (See "Transvaginal ultrasound assessment of the cervix and prediction of spontaneous preterm birth".)
Laboratory — There are no laboratory abnormalities characteristically associated with uncomplicated cervical insufficiency.
In some cases, cervical insufficiency is caused by, or leads to, subclinical intraamniotic infection, which can be diagnosed by examination of amniotic fluid from amniocentesis. (See "Intraamniotic infection (chorioamnionitis)", section on 'Diagnosis of clinical chorioamnionitis'.)
Cervico-vaginal fetal fibronectin may be positive. (See "Fetal fibronectin for prediction of preterm labor and delivery".)
DIAGNOSIS — The diagnosis of cervical insufficiency is either based on historic factors or, preferably, by a combination of historic factors and transvaginal ultrasound (TVU) measurement of cervical length.
●Using historic factors alone, cervical insufficiency is defined as painless cervical dilatation leading to recurrent second trimester pregnancy losses/births. This definition precludes diagnosis of cervical insufficiency until at least two pregnancy losses/births before 28 weeks of gestation have occurred.
●A preferable definition allows the diagnosis of cervical insufficiency to be made in primigravidas or in multigravidas without multiple prior pregnancy losses. Using this definition, cervical insufficiency is defined by TVU cervical length <25 mm and/or advanced cervical changes on physical examination before 24 weeks of gestation in women with either:
•One or more prior pregnancy losses or preterm births at 14 to 36 weeks, and/or
•Other significant risk factors for cervical insufficiency (see 'Cervical risk factors' above).
The diagnosis of cervical insufficiency is usually limited to singleton gestations because the pathogenesis of delivery at 14 to 28 weeks in multiple gestations is usually unrelated to a weakened cervix. In addition, preterm labor, infection, abruptio placenta, and bleeding placenta previa should be excluded, as these disorders could account for biochemically mediated cervical ripening leading to second trimester pregnancy loss or preterm delivery independent of structural/anatomic cervical weakness . Although the rate of decrease in cervical length over time  and shortening before 20 weeks [17-19] are associated with an increased risk of early preterm birth, these findings do not absolutely distinguish cervical insufficiency from other causes of preterm delivery.
Can cervical insufficiency be diagnosed before pregnancy? — The diagnosis of cervical insufficiency cannot be made or excluded outside of pregnancy by any test. Evaluation of cervical function with dilators, balloons, or hysteroscopy is not helpful. Ultrasound, magnetic resonance imaging, or hysterosalpingography may reveal a uterine anomaly, which is a risk factor for cervical insufficiency, but is not diagnostic.
APPROACH TO MANAGEMENT
Women with prior pregnancy losses or preterm births
Candidates for history-indicated cerclage — A minority of recurrent second trimester losses/births are primarily, and perhaps exclusively, caused by congenital or acquired structural weakness of the cervix, and can be treated effectively with support by a “history-indicated” cerclage. Anatomical, biochemical, and clinical evidence from observational studies and randomized trials consistently support this hypothesis [20-23].
●Two or more consecutive prior second trimester pregnancy losses or three or more early (<34 weeks) preterm births.
●Risk factors for cervical insufficiency. As discussed above, these risk factors include a history of cervical trauma and/or short labors or progressively earlier deliveries in successive pregnancies (see 'Past obstetrical history' above and 'Cervical risk factors' above).
●Other causes of preterm birth (eg, infection, placental bleeding, multiple gestation) have been excluded.
The technique for cerclage is described separately. (See "Transvaginal cervical cerclage".)
We also treat these women with 17-alpha-hydroxyprogesterone caproate weekly from 16 to 36 weeks of gestation [24-26]. Although randomized trials support the benefit of history-indicated cerclage  and the benefit of progesterone supplementation [24-26] in this population, no trials have evaluated the efficacy of combination therapy (both history-indicated cerclage and 17 alpha hydroxy-progesterone caproate).
Candidates for ultrasound surveillance and possible ultrasound indicated cerclage — The majority of women with suspected cervical insufficiency do not meet the above criteria for history-indicated cerclage. For these women, we usually initiate TVU cervical length screening (table 2), administer 17-alpha-hydroxyprogesterone caproate prophylaxis, and apply a cerclage if cervical length decreases to <25 mm . The rationale for this approach is:
●Women with a short cervix on transvaginal ultrasound examination are at increased risk of spontaneous preterm birth .
•In women with a history of spontaneous preterm birth, a systematic review of controlled studies showed that measurement of cervical length in the second trimester, especially before 24 weeks, predicted the risk of recurrent preterm birth . The use of a TVU cervical length <25 mm at <24 weeks to predict preterm birth at <35 weeks yielded sensitivity of 65.4 percent, specificity of 75.5 percent, positive predictive value of 33.0 percent, and negative predictive value of 92.0 percent. The shorter the cervical length, the higher the positive likelihood ratio for spontaneous preterm birth <35 weeks.
●In randomized trials, progesterone prophylaxis with 17 alpha hydroxy-progesterone caproate starting at 16 to 20 weeks in women with a history of spontaneous preterm birth and continuing until 36 weeks reduced the risk of recurrent preterm birth [24,25]. (See "Progesterone supplementation to reduce the risk of spontaneous preterm birth".)
●Placement of cerclage upon identification of a short cervix (“ultrasound-indicated cerclage”) is effective in reducing preterm birth , results in pregnancy outcomes comparable to those with history-indicated cerclage , and avoids cerclage in about 60 percent of patients with a suggestive history . The benefit of ultrasound-indicated cerclage may derive from bolstering cervical strength , preventing membranes from being exposed, and retention of the mucus plug.
•In a meta-analysis of randomized trials of women with singleton gestation and prior spontaneous preterm birth and short cervical length <25 mm before 24 weeks, treatment with ultrasound-indicated cerclage significantly lowered total neonatal morbidity and mortality (15.6 versus 24.8 percent without cerclage; RR 0.64, 95% CI 0.45-0.91), presumably because cerclage significantly reduced the frequency of preterm birth (delivery <35 weeks RR 0.70, 95% CI 0.55-0.89; 28.4 percent versus 41.3 percent in women without cerclage) .
•In another meta-analysis of randomized trials of women with singleton gestations and prior preterm birth managed either by (1) cervical length screening with cerclage for short cervical length or (2) history-indicated cerclage, patients with ultrasound-indicated versus history-indicated cerclage had similar rates of preterm birth before 37 weeks (31 versus 32 percent, RR 0.97, 95% CI 0.73-1.29), preterm birth before 34 weeks (17 versus 23 percent, RR 0.76, 95% CI 0.48-1.20), and perinatal mortality (5 versus 3 percent, RR 1.77, 95% CI 0.58-5.35), and only 42 percent developed a short cervical length and received cerclage .
We usually initiate cervical length screening at 14 weeks, but may screen as early as 12 weeks in women with early second trimester losses, recurrent second trimester losses, or prior large cold knife conization (table 2) . In women with prior preterm birth at 28 to 36 weeks, we initiate screening at 16 weeks. Ultrasound examination is generally repeated every two weeks until 24 weeks as long as the cervical length is ≥30 mm, and increased to weekly if cervical length is 25 to 29 mm, with the expectation that preterm cervical changes will precede overt preterm labor or membrane rupture symptoms by three to six weeks . Transvaginal ultrasound screening is usually discontinued at 24 weeks of gestation, as cerclage is not usually performed after this time. (See "Transvaginal ultrasound assessment of the cervix and prediction of spontaneous preterm birth".)
We continue intramuscular progesterone supplementation until 36 weeks, whether or not a cerclage is placed. Although no studies have reported a statistically significant improvement in pregnancy outcome with combined therapy [31,34,35], these studies are retrospective analyses and underpowered. One post-hoc analysis of data from a randomized trial observed that the rate of early preterm birth trended lower in women who received both 17-alpha-hydroxyprogesterone caproate and ultrasound-indicated cerclage than in those who received either therapy alone (preterm birth <28 weeks: both interventions: 9 percent versus cerclage alone: 17 percent and progesterone alone: 15 percent; preterm birth <32 weeks: both interventions: 17 percent versus cerclage alone: 25 and progesterone alone: 21 percent) .
No randomized controlled trial has directly compared vaginal progesterone, intramuscular 17-alpha-hydroxyprogesterone caproate, and cervical cerclage for the prevention of preterm birth in women with a sonographic short cervix in the mid trimester, singleton gestation, and previous preterm birth. An indirect comparison meta-analysis concluded vaginal progesterone and cerclage were equally efficacious in the prevention of preterm birth in this population . Based on evidence from the direct comparisons in the randomized trials discussed above, we treat women with prior preterm birth with intramuscular 17-alpha-hydroxyprogesterone caproate and then perform cerclage if the cervical length becomes less than 25 mm. (See 'Women with prior pregnancy losses or preterm births' above.)
Women with singleton pregnancy, no prior birth, but risk factors for cervical insufficiency — Studies of women with risk factors for cervical insufficiency, such as uterine anomaly, prior minor cervical surgery, or pregnancy termination, observed a correlation between risk of preterm birth and short cervical length, but data are limited [38-40]. A wide range of recurrent preterm birth rates has been reported in these studies, depending on the cervical length threshold used and the gestational age at the time of measurement . No intervention has been evaluated by randomized trials specifically in these populations.
We perform a single TVU cervical length measurement at 18 to 24 weeks in women with risk factors for cervical insufficiency and no prior delivery (table 2), and treat those with a short cervix (≤20 mm) with vaginal progesterone supplementation (table 2). In a meta-analysis of five trials, administration of vaginal progesterone to women with a short cervix reduced the rate of spontaneous preterm birth and composite neonatal morbidity and mortality , and it appears to be cost-effective [43,44]. (See "Progesterone supplementation to reduce the risk of spontaneous preterm birth", section on 'Short cervix in current pregnancy' and "Cervical intraepithelial neoplasia: Reproductive effects of treatment", section on 'Obstetric management in subsequent pregnancies'.)
If the patient delivers preterm, subsequent pregnancies are managed as described above (see 'Women with prior pregnancy losses or preterm births' above). If she delivers at term, we again perform a single cervical length measurement at 18 to 24 weeks and give vaginal progesterone if the cervix is short.
Short cervix at <24 weeks, no prior preterm birth, no risk factors for preterm birth — These women do not meet our definition for diagnosis of cervical insufficiency (see 'Diagnosis' above). The management of their pregnancies is discussed separately. (See "Transvaginal ultrasound assessment of the cervix and prediction of spontaneous preterm birth", section on 'Singleton, no prior preterm birth'.)
Physical examination reveals a dilated cervix and visible membranes before 24 weeks — Rarely, a woman presents before 24 weeks with minimal or no symptoms and physical examination reveals a dilated cervix. Occasionally, such findings follow the identification of a very short cervical length (eg, <5 mm) on TVU. The management of these cases is governed primarily by whether a condition that would likely lead to prompt delivery is present, such as overt infection, ruptured membranes, or significant hemorrhage. In the absence of indications for delivery, the gestational age and degree of cervical dilation are the next considerations. Prior to viability, management is aimed at prolonging the pregnancy; some parents may choose pregnancy termination, especially for preterm premature rupture of membranes (PPROM) or advanced (>4 cm) cervical dilation. After viability, the goal is to both prolong the pregnancy and improve neonatal outcome in the likely event of preterm birth.
Data from several studies suggest that a grossly dilated cervix with visible membranes may be an appropriate criterion for placement of a “physical exam-indicated cerclage” in some cases (also called “rescue cerclage” or “emergency cerclage”). Placement of a physical exam-indicated cerclage when a dilated cervix and visible membranes are detected on digital examination at <24 weeks appeared to prolong pregnancy and improve pregnancy outcome compared to expectant management in a small randomized trial, a prospective study, and retrospective cohort studies [45-54]. Due to differences in patient populations, actual outcomes varied among these studies. Physical exam-indicated cerclage in women with visible bulging membranes should only be considered in the absence of infection, labor, and vaginal bleeding (abruption). (See "Transvaginal cervical cerclage".)
In women without clinical signs of infection, we offer amniocentesis to check for subclinical infection when the cervix is ≥2 cm dilated and on a case-by-case basis when there are ultrasound findings suggestive of inflammation: membrane edema, separation of membranes from the decidua, or debris in the amniotic fluid (“sludge”) (image 1). Some studies have reported an association between presence of intraamniotic debris and intraamniotic inflammation, preterm delivery, and other pregnancy complications [55-58]. (See "Intraamniotic infection (chorioamnionitis)", section on 'Amniocentesis' and "Transvaginal ultrasound assessment of the cervix and prediction of spontaneous preterm birth".)
Prior successful outcome after cerclage — Prolongation of pregnancy after cerclage does not prove a diagnosis of cervical insufficiency because many pregnancies with premature cervical effacement have good outcomes in the absence of surgical intervention. As discussed above, in randomized trials and controlled studies, about 60 percent of women with a history of early preterm birth or recurrent late miscarriage maintain cervical length above 25 mm and have low rates of recurrent preterm birth/loss without placement of a cerclage [31,45,59-63]. Therefore, repeat cerclage in subsequent pregnancies is not mandatory.
In women who received a cerclage in a prior pregnancy without an appropriate indication, especially those who, after removal of cerclage at 36 to 37 weeks, did not go into labor in the subsequent two weeks, the risk of preterm birth in a subsequent pregnancy probably does not warrant a history-indicated cerclage; instead we suggest TVU cervical length screening .
Prior unsuccessful outcome after cerclage — Transabdominal cerclage may be successful in women who deliver very preterm despite placement of a transvaginal cerclage. (See "Transabdominal cervical cerclage".)
Pessary — Vaginal pessaries are intended to alter the axis of the cervical canal and displace the weight of the uterine contents away from the cervix. By changing the angle of the cervix in relation to the uterus, the pessary also obstructs the internal os and thus may provide protection against ascending infection.
Before 2012, only a few observational studies suggested that pessaries may have efficacy equivalent to cerclage and/or progesterone [64-67]. In 2012, a multicenter trial randomly assigned 385 pregnant women with cervical length ≤25 mm at 20 to 23 weeks to use of a cervical pessary or expectant management (no pessary) [68,69]. The majority of these patients (89 percent) had no history of previous preterm birth and none were treated with progestogens or cerclage. The pessary group had a significant reduction in spontaneous preterm <28 weeks (4/190 [2 percent] versus 16/190 [8 percent] OR 0.23, 95% CI 0.06-0.74) and <34 weeks (12/190 [6 percent] versus 51/190 [27 percent] OR 0.18, 95% CI 0.08-0.37).
No adverse effects were noted, other than vaginal discharge and mild discomfort upon insertion. Use of a pessary to prolong pregnancy in women with a short cervical length may be an effective, inexpensive, and easy to implement intervention; however, given the open-label design with lack of a placebo control and the relatively small number of preterm births accounting for the dramatic results, we do not recommend implementation of this approach instead of cerclage or progestogens until these data are confirmed by additional randomized and placebo-controlled trials.
A review of the different models of pessary used for prevention of preterm birth, as well as sizing, insertion, changing/removing, follow-up, and patient education, has been published by Arabin and Alfirevic .
Indomethacin — A review of data from three randomized cerclage trials found that indomethacin therapy for asymptomatic women with a short cervix (<25 mm) at 14 to 27 weeks who did not receive a cerclage did not reduce the risk of spontaneous preterm birth <35 weeks, but appeared to reduce the risk of preterm birth <24 weeks . Further research including larger numbers and a randomized trial design is necessary to further clarify the effectiveness, as well as the risks, of this therapy.
Antibiotics — There is insufficient evidence to recommend antibiotics for women with cervical insufficiency, based on poor obstetrical history, short TVU cervical length, or dilated cervix on physical exam.
Lifestyle interventions — Clinicians should consider the available evidence and the patient's individual circumstances when making lifestyle recommendations as there are social, psychological, financial, and medical side effects associated with these interventions. However, lifestyle interventions (cessation of work and exercise, abstinence from coitus, bedrest/limited activity) have not been adequately evaluated by well-designed studies. Although coitus is not a risk factor for onset of labor in women at term who are scheduled for induction , there are inadequate data on the safety of coitus in women at risk for preterm birth because of previous preterm birth or preterm cervical ripening . In our practice, we advise women with both a prior preterm birth and a short cervix to avoid coitus.
SUMMARY AND RECOMMENDATIONS
●Cervical insufficiency describes a presumed physical weakness of cervical tissue that causes or contributes to the early delivery of an otherwise healthy pregnancy, usually in the second trimester. (See 'Introduction' above.)
●Cervical insufficiency may be congenital or acquired. Women at risk include those with a history of cervical trauma (eg, dilation and curettage, dilation and evacuation), collagen disorders, or congenital anomalies of the uterus/cervix. (See 'Cervical risk factors' above.)
●The diagnosis of cervical insufficiency is either based on historic factors or on a combination of historic factors and transvaginal ultrasound (TVU) measurement of cervical length. (See 'Diagnosis' above.)
Using historic factors alone, cervical insufficiency is defined as painless cervical dilatation leading to recurrent second trimester pregnancy losses/births.
A preferable definition allows the diagnosis of cervical insufficiency during any pregnancy. Using this definition, cervical insufficiency is defined by TVU cervical length <25 mm and/or advanced cervical changes on physical examination before 24 weeks of gestation in women with either:
•One or more prior pregnancy losses/births at 14 to 36 weeks, and/or
•Other significant risk factors for cervical insufficiency.
●For women with two or more consecutive prior second trimester pregnancy losses or three or more early preterm births who have risk factors for cervical insufficiency and in whom other causes of preterm birth have been excluded, we recommend history-indicated cerclage (Grade 1B). (See 'Candidates for history-indicated cerclage' above.)
●For women with suspected cervical insufficiency and prior early preterm birth who do not meet criteria for history-indicated cerclage, sonographic surveillance should be started early in pregnancy (eg, 14 to 16 weeks). We suggest cerclage for women who develop a short cervix (<25 mm) before 24 weeks (Grade 2B). (See 'Candidates for ultrasound surveillance and possible ultrasound indicated cerclage' above.) Vaginal progesterone administration or placement of a pessary are alternative approaches. (See "Progesterone supplementation to reduce the risk of spontaneous preterm birth", section on 'Short cervix in current pregnancy' and 'Pessary' above.)
●We recommend progesterone supplementation for women with singleton gestations and a history of spontaneous preterm birth (Grade 1B). (See "Progesterone supplementation to reduce the risk of spontaneous preterm birth".)
All topics are updated as new information becomes available. Our peer review process typically takes one to six weeks depending on the issue.