Archives of Gynecology and Obstetrics

, Volume 287, Issue 4, pp 621–628

Short cervix syndrome: current knowledge from etiology to the control

Authors

  • Hélio Antonio Guimarães Filho
    • Imaging Training Center (CETRIM)
    • Department of ObstetricsFederal University of São Paulo (UNIFESP)
  • Claudio Rodrigues Pires
    • São Paulo Ultrasound Training Center (CETRUS)
  • Luciano Marcondes Machado Nardozza
    • Department of ObstetricsFederal University of São Paulo (UNIFESP)
  • Antonio Fernandes Moron
    • Department of ObstetricsFederal University of São Paulo (UNIFESP)
Maternal-Fetal Medicine

DOI: 10.1007/s00404-013-2740-0

Cite this article as:
Guimarães Filho, H.A., Araujo Júnior, E., Pires, C.R. et al. Arch Gynecol Obstet (2013) 287: 621. doi:10.1007/s00404-013-2740-0

Abstract

Background

Preterm delivery is one of the most serious public health problems and is the most important factor relating to neonatal morbidity and mortality. The strategies for preventing it include understanding the risk factors, with specific interventions. Recently, uterine cervix measurements using ultrasonography and vaginal administration of progesterone have gained importance in predicting and secondarily preventing spontaneous preterm delivery.

Objective

To describe the short cervix syndrome, including its etiology, diagnosis, and possible therapies.

Methods

Research in ISI, Pubmed, and Scielo database using the words short cervix, preterm delivery, sludge, cervical funneling, cervical gland area, progesterone, cerclage, and pessary.

Results

We found a lot of articles about this topic, including randomized controlled trials. The etiology is multifactorial, being the diagnosis based in a cervix shortening at 20–24 weeks. The history and measurement of cervix length by transvaginal ultrasound have been shown to be effective to select the high risk pregnancies. The progesterone, cervical cerclage, and cervical pessary showed to be effective to reduce the preterm delivery in pregnant women with short cervix.

Conclusion

The successful management of pregnant women presenting a short cervix depends on the understanding that cervical shortening is the final common path for several causes of preterm delivery. The best approach should be individualized to each patient.

Keywords

Short cervixPreterm deliveryTransvaginal ultrasoundProgesteroneCervical cerclageCervical pessary

Introduction

Preterm delivery, which is defined as termination of pregnancy before 37 weeks have been completed, constitutes one of the most serious public health problems and is the most important factor relating to neonatal morbidity and mortality. The morbidity and mortality rates become significantly higher particularly when delivery takes place before the 32nd week. Although many maternal and fetal characteristics have been correlated with preterm delivery, its etiology often cannot be defined, and idiopathic causes account for 20–40 % of the cases [1, 2]. Many published papers have highlighted clinical methods and biological, biochemical, and ultrasound markers that seek to predict spontaneous preterm delivery separately or in combinations [3, 4].

A statistically significant inverse relationship between cervical length seen on transvaginal ultrasonography in the second trimester and the risk of preterm delivery has been demonstrated in both in low-risk asymptomatic single pregnancies and in high-risk patients, including among these with previous preterm delivery.

Etiology and epidemiology

Cervical shortening is a syndrome that may have several etiologies, acting separately or in combination. Shortening and effacement of the uterine cervix is thus a common final path for many, but not all of the causes of premature delivery [3]. At the beginning of pregnancy, the etiology of a short cervix differs from what occurs in the middle and at the end of pregnancy. A cervix that starts to diminish at the beginning of pregnancy may be a sign of previous damage to the cervix, such as through surgery (conization or excision using a LEEP loop) or due to uterine malformations [5, 6]. In the second trimester of pregnancy, isthmocervical incompetence is the most important cause of cervical shortening, thereby causing late abortions and spontaneous preterm delivery. This may be congenital (primary) or be acquired (secondary) through trauma to the isthmocervical region, such as untimely cervical dilatation, application of a high forceps or amputation [3, 7].

Cervical shortening diagnosed between the 20th and 24th weeks is an important etiological factor for preterm delivery. Its causes and possible control methods have been targeted in many studies over the last 15 years. The start of delivery labor involves two interdependent processes: remodeling, i.e. effacement and possible dilatation of the uterine cervix; and triggering of frequent rhythmic uterine contractions. This process of cervical effacement begins some weeks before the labor of premature delivery and can be recognized from the shortening of the uterine cervix in the transvaginal ultrasound examination performed between the 20th and 24th weeks. However, its etiology is variable and may be related to several causes. Among these are idiopathic causes, intra-amniotic or cervical inflammation/infection and premature uterine contractions [8, 9].

The risk of preterm delivery varies as a function of the gestational age at the time of diagnosing the cervical shortening. Women with a short cervix who are diagnosed before the 20th week of pregnancy often have a short interval between the diagnosis and the delivery and a higher rate of preterm delivery than shown by women diagnosed between the 20th and 24th weeks. Almost two-thirds of the women diagnosed with a short cervix before the 20th week of pregnancy have their delivery at the threshold of viability (< 24 weeks of gestation), and less than one quarter of them will still be pregnant when the 28th week of gestation is completed. As well as the gestational age at which the diagnosis is made, the degree of shortening is also an independent risk factor for the time interval between diagnosis and delivery, and for the preterm delivery rate. The greater the shortening is, the smaller the interval between diagnosis and delivery and the higher the preterm delivery rate will be [10, 11].

Around half of the cases of spontaneous preterm delivery occur among nulliparous women. In any given group of women with spontaneous preterm delivery, around 15 % will have had a previous history of preterm delivery. On the other hand, around 85 % will have had an obstetric history without major events. For some authors, this provides justification for making efforts towards finding effective solutions in universal screening programs using transvaginal ultrasonography. However, the sensitivity and positive predictive value of cervical length, as shown by transvaginal ultrasonography for detecting preterm delivery, are low in the general population of pregnant women [3]. The highest sensitivity rates have been found in studies on populations presenting risk factors, such as obstetric antecedents of prematurity. The clinical history of risks is capable of predicting, with a false-positive rate of 5 %, around 30 % of the cases that will evolve to preterm delivery. Therefore, prediction of preterm delivery starts with a good directed anamnesis, albeit with limited application [4, 10].

Most authors who have investigated preterm delivery make a distinction between early and late preterm delivery, with the aim of selecting the high-risk preterm cases. One cutoff point that is commonly used is birth before 32 weeks of gestational age. Prediction of preterm delivery by means of screening for a short cervix has the main aim of identifying women presenting a risk of this outcome, and therefore offering the possibility of prophylaxis and interventions. In a study conducted at King’s College, London, with a large sample, the patients with a history of preterm delivery at between the 24th and 32nd weeks in a previous pregnancy presented a 3.7-fold greater chance of preterm delivery in the subsequent pregnancy. If the delivery in the previous pregnancy occurred between the 33rd and 36th weeks, the risk was elevated less, by around 2.7-fold. The same study revealed that an antecedent of cervical surgery was a factor that raised the risk by 1.3-fold [12].

Investigation solely of the risk factors described so far has not provided any significant reduction in the incidence of spontaneous preterm delivery. Predictive indices have been drawn up with this aim, based on data from the previous history and on factors indicative of risk that were observed during the pregnancy. However, the final results from applying these indices have not been encouraging. The presence of multiple factors causing preterm delivery limits the success of prematurity prevention programs, which usually focus on just one aspect within the multifactorial complex situation [1, 1012].

Diagnosis

Gynecological examination

The uterine cervix can be examined by means of vaginal palpation. However, this procedure, which has the aim of investigating the characteristics of the cervix (dilatation, effacement, and position), has been shown to present low sensitivity and low positive predictive value for detecting preterm delivery. For this reason, this examination is not useful for diagnosing and following up pregnant women with shortening of the uterine cervix [10, 13].

Complementary examinations

Level of cytokine and inflammatory mediators in the amniotic fluid

The inflammation/infection has been shown to be a significant contributor to the preterm parturition syndrome and frequently is associated with mid-trimester short cervix [14]. Levels of inflammatory mediators were significantly higher in patients who experienced cervical shortening, especially in those who delivered at 34 weeks’ gestation [15]. Proinflammatory cytokines have been identified in increased concentrations in cervical tissue, cervicovaginal secretions, myometrium, amniotic fluid, and fetal membranes of patients during parturition, stimulating prostaglandin production, uterine activity, and cervical remodeling [14, 15].

Some studies have evaluated cytokines in the amniotic fluid, such as matrix metalloproteinase-8, interleukin (IL)-6, IL-8, or monocyte chemotactic protein-1 as predictors of spontaneous preterm birth. Keeler et al. [16] assayed 25 cytokines and inflammatory mediators in the amniotic fluid of pregnant women diagnosed with short cervix in the second trimester. They observed a significant increase only of monocyte chemotactic protein-1 that had 60 % sensitivity, 100 % specificity, 100 % positive predictive value, and 70 % negative predictive value for delivery below of 34 weeks’ gestation. This suggests that an early elevation of monocyte chemotactic protein-1 in asymptomatic mid-trimester short cervix patients may be an initial factor in the preterm delivery syndrome. Recently, this same group developed a fluid amniotic inflammatory score that presented higher accuracy compared to the dosage of isolated cytokines; with a high sensitivity (87 %) without modify the specificity (100 %) [17].

Level of fetal fibronectin in the cervicovaginal fluid

Cervicovaginal fetal fibronectin (fFN) is a glycoprotein presents in trace quantities that is usually undetectable in the cervicovaginal fluid. A higher quantity has been purported to be an indication of imminent labor onset [18]. The test is readily available in the form of a commercial rapid test kit. The result is either positive (fFN is present), or negative (fFN is not present). The presence of fFN is a predictor of preterm birth in asymptomatic and symptomatic pregnancies that are evaluated for preterm labor [18]. Yoneda et al. [19] reported that the level of fFN in cervicovaginal fluid is highly predictive marker for preterm delivery before of 34 weeks’ gestation.

In a recent systematic review for the predicting of accuracy of spontaneous preterm birth before of 34 weeks, including the reviewing of 22 tests, only few tests had positive likelihood ratio (LR +) > 5, one of them being the fFN [20].

Ultrasonography

Transvaginal ultrasonography has been shown to be the most accessible, objective, and reliable method for assessing cervical biometry and morphology during pregnancy, with the advantage that it can be done in a timely manner during the fetal morphological examination of the second trimester. The uterine cervix as an assessment route during pregnancy has already been the subject of many published papers in the medical literature. Today, there is a consensus that that the most appropriate route for studying the cervix is vaginal and, as a second option, transperineal examination (particularly in situations such as intense vaginal bleeding or premature amniorrhexis). The abdominal route should not be used, because of the high numbers of false-positives and false-negatives consequent to segmental contractions and bladder compression, respectively, and also because viewing the cervix using this route is usually impaired by the fetal positioning [21].

Echographic examination for cervical biometric and morphological assessments is simple but requires technical care and observation of some prior conditions. Patients have to completely empty their bladder immediately before the examination is started. The echographic probe has to be positioned such that it does not compress the cervix; the focus should be positioned over the cervical canal; and the image magnification should be adjusted until the cervix occupies at least three quarters of the screen. With these conditions fulfilled, the physician can start the biometric and morphological assessment of the cervix to investigate signs of risk of preterm delivery [9].

Biometric ultrasound markers

The main sign that should be investigated using transvaginal ultrasonography in screening for preterm delivery is cervical length. Other measurements such as anteroposterior diameter of the cervix do not have their deserved emphasis in the literature with regard to prediction of prematurity.

After identification of the internal os, external os and the entire extent of the cervical canal, the cervical length is measured in a straight-line manner, even when the cervix presents an arched shape [2, 9]. Although the cervical length measurement may be significantly longer than the straight-line measurement in cases of a curved cervix, this difference does not present clinical significance, since short cervixes invariably present a straightened appearance. The examination should last for at least 3 min, with at least three measurements, and the shortest of these should be used. During the examination, a funneling sign can be observed, which subsequently disappears (dynamic changes), with consequent changes to cervical length.

At the beginning of the 1990 s, the first studies revealed a relationship between shortening of the cervix and the risk of premature delivery, and indicated that the normal minimum limit was a length of 35 mm [22]. Since then, there have been hundreds of published papers on this topic, and the cervical length indicative of a risk of preterm delivery has progressively become smaller. Currently, the uterine cervical length limit representing a risk of spontaneous preterm delivery is still a matter of controversy in the literature. However, despite the controversy regarding the cutoff point, many studies have indicated that women with cervical lengths <15–25 mm presented a higher risk of preterm delivery [9, 12, 2325]. Most published papers present one conclusion in common: the shorter the cervical length is, the greater the risk of preterm delivery is. The paper with the largest sample so far included more than 58,000 pregnant women and suggested that a cervical length of 15 mm at between the 20th and 24th week of pregnancy was the lower limit for normalcy for singleton pregnancies [25]. According to these authors, screening done using an integrated model with the cervical length and obstetric history can detect 75 % of preterm deliveries before the 28th week, with a sensitivity of 57 % for preterm deliveries before the 32nd week and a false-positive rate of 10 %.

Morphological ultrasound markers

Cervical funneling
The cervical funneling (also known as the “glove finger” or “feeding bottle teat” sign) was the first ultrasound marker denoting a risk of preterm delivery (Fig. 1). However, the clinical significance of this cervical morphological modification seen on transvaginal ultrasonography remains a topic of major medical discussion, since it presents great variability between observers and even for the same observer. Thus, there is controversy in the literature regarding its capacity for predicting preterm delivery. While some studies have indicated that this funneling is strongly correlated with preterm delivery, others have been unable to demonstrate any significant association. On the other hand, some researchers have suggested that the cervical funneling constitutes an early marker of isthmocervical insufficiency [3, 9, 11, 26].
https://static-content.springer.com/image/art%3A10.1007%2Fs00404-013-2740-0/MediaObjects/404_2013_2740_Fig1_HTML.jpg
Fig. 1

Longitudinal section through the uterine cervix showing cervical funneling. A Width of funneling (opening of the internal os of the cervix); B extent of funneling; C length of remainder of cervix (functional length)

The sign needs to be described in detail regarding the diameter of the internal os (minimum of 5 mm for the diagnosis) and the depth of the funneling, as well as the length of the remaining cervix, also known as the functional length. Since there is no consensus in the literature, caution is recommended in basing the approach only on the presence of cervical funneling.

Cervical gland area
Starting at the end of the 1990 s, some papers from a single research group emphasized the importance of another morphological marker for the uterine cervix that could be seen on transvaginal ultrasonography: the “cervical gland area”. If this was not detected, it would indicate an increased risk of preterm delivery [2729] (Fig. 2). It was observed that disappearance of the endocervical gland area (a band peripheral to the cervical canal) on the transvaginal ultrasound examination presented a correlation with the cervical maturation process and could be considered to be predictive of an unfavorable prognosis for patients with preterm delivery labor [27]. Some researchers correlated failure to identify the cervical gland area with preterm delivery before the 32nd week, with significant elevation of the sensitivity and positive predictive value, compared with the parameter of shortened cervical length [29].
https://static-content.springer.com/image/art%3A10.1007%2Fs00404-013-2740-0/MediaObjects/404_2013_2740_Fig2_HTML.jpg
Fig. 2

Longitudinal section through the uterine cervix. The white arrows indicate the cervical gland area, peripheral to the cervical canal

In Brazil, Pires et al. translated the “cervical gland area” to “echo-glandular endocervical” (EGE) and also found a correlation between disappearance of this sign between the 20th and 24th weeks and spontaneous preterm delivery, in a low-risk population. According to these authors, multivariate logistic regression analysis indicated that absence of the EGE was the variable most associated with preterm delivery, in relation to cervical length and the funneling sign [30, 31].

Although the first studies indicated that the cervical gland area was an important marker for the risk of preterm delivery, collaborative multicenter studies should be concluded in order to incorporate this definitively in screening for prematurity. Up to the present day, measurement of cervical length remains the principal echographic sign to be investigated for these purposes.

Sludge from the amniotic fluid
Identification of echographic images from condensation of echogenic particles from the amniotic fluid, such that an agglomerate resembling mud or sludge is formed adjacent to the internal os of the uterine cervix, was described for the first time in 2005, in patients presenting a high risk of preterm delivery [32, 33]. This sign has been presented as an independent variable for preterm delivery, premature amniorrhexis, microbiological invasion of the amniotic cavity, and chorioamnionitis, among asymptomatic women at high risk of preterm delivery (Fig. 3). The combination of sludge and a short cervix conferred greater risk of preterm delivery before the 28th and 32nd weeks, compared with short cervix alone. However, there is a need for more studies in order to implement management based on this parameter, since so far, there have not been any published papers on adequately sized cohorts [8, 34].
https://static-content.springer.com/image/art%3A10.1007%2Fs00404-013-2740-0/MediaObjects/404_2013_2740_Fig3_HTML.jpg
Fig. 3

Longitudinal section through the uterine cervix. Note the reduced cervical length, significant protrusion of the membranes into the interior of the cervical canal and absence of the cervical gland area. The white arrows indicate the sludge from the amniotic fluid

Treatment

The importance of diagnosing cases of short cervix in the middle of pregnancy derives from the evidence that such patients, independent of whether they might have any history of previous premature delivery, may be candidates for therapeutic and control interventions to reduce the preterm delivery rates.

Hospitalization

Physical rest, not necessarily in bed, avoidance of domestic chores and professional duties, and sexual abstinence are the control measured usually prescribed for preventing preterm delivery. On the other hand, hospitalization has not been shown to be beneficial in reducing the risk of preterm delivery. In a retrospective cohort study on women who had a singleton pregnancy with cervical length < 25 mm measured between the 16th and 28th weeks, conducted expectantly, hospital admission was an independent risk factor for delivery before the 34th week, for younger gestational age at birth and also for a shorter time interval from diagnosis to delivery. This was also true for women with cervical length < 15 mm [35].

Clinical treatment

Over the last few years, a return to progesterone use as an effective medication for reducing the risk of prematurity among patients at risk of preterm delivery has been observed. The mechanism of action of progesterone for prolonging pregnancy is not fully known. It has been suggested that progesterone relaxes the smooth muscle tissue of the myometrium, blocks the action of oxytocin and inhibits the formation of gap junctions [10]. However, the mechanism of action for preventing preterm delivery mainly involves the uterine cervix. Both synthetic and natural progesterone have the effect of lengthening the uterine cervix. Clinical studies have demonstrated that antiprogestins induce cervical maturation, both in the first trimester and at term [12]. Progesterone may be considered to be a hormone that produces cervical reinforcement, and its removal will result in release of pro-inflammatory cytokines, with consequent softening and effacement of the uterine cervix. Therefore, treatment with progesterone has a local effect on the uterine cervix, thus controlling the release of pro-inflammatory cytokines and preventing degradation of the extracellular matrix of the uterine cervix [10, 36].

Although intramuscular and oral treatments generate higher levels of progesterone in the blood, several studies have shown higher progesterone concentrations in the endometrium of women who received progesterone vaginally, thus indicating that there is a direct transportation mechanism between the vagina and the uterus. Consequently, vaginal progesterone treatment has been associated with lower serum levels of metabolites and fewer side effects. Going from this observation, progesterone has started to be indicated for patients who present histories of preterm delivery and/or short cervix on transvaginal ultrasonography, at a dose of 100–200 mg vaginally, once or twice a day, or 100–200 mg orally, three times a day, from the 16th week to the 34th or 36th week of pregnancy [37].

In a clinical trial, use of progesterone vaginally between the 24th and 34th weeks in patients with a short cervix (≤15 mm) reduced the frequency of premature deliveries by 50 %. Since progesterone use vaginally does not have any significant adverse effects and is capable of halving the frequency of preterm delivery, this should be regarded as an important means of reducing the risk of prematurity among pregnant women with a single fetus who present a short cervix between the 20th and 24th weeks [37].

Surgical treatment

The association between short cervix and prematurity has given rise to indications of cerclage as a therapeutic measure. However, after a sharp increase in the number of surgical procedures performed, some clinical trials demonstrated that this procedure in patients with a short cervix diagnosed between the 20th and 24th weeks, without any condition of isthmocervical incompetence, did not diminish the prematurity rates. Subsequently, other studies demonstrated reductions in preterm delivery in specific groups of women [38, 39]. In three meta-analysis studies, it was demonstrated that among women with a history of preterm delivery who presented singleton pregnancies and a cervical length < 25 mm at between the 20th and 24th weeks, cerclage significantly reduced the incidence of preterm delivery, along with the neonatal morbidity and mortality [3, 10, 39]. Reinforcement cerclage has also been shown to be associated with unfavorable outcomes, and is currently contraindicated [40].

Cervical pessary

The cervical pessary is silicone device that was used by first time by Arabin et al. [41] to prevent the preterm delivery. They reported a preterm birth rate before 34 weeks of zero when they inserted pessaries into women with short cervical lengths at 22 weeks compared with nearly 50 % in a matched control group. Sieroszewski at al. [42] described a case series of 54 pregnant women. Insertion of a pessary in those with cervical lengths between 15 and 30 mm resulted in an incidence of preterm birth before 29 weeks of 1.9 % and a birth-at-term rate of 83.3 %.

Recently, Goya et al. [43] realized a controlled randomized trial to investigate the efficacy of cervical pessary in women with short cervix to prevent the preterm delivery (< 34 weeks). These authors randomized 393 pregnant women (18–43 years old) with cervical length ≤ 25 mm in a 1:1 ratio to the cervical pessary and the expectant management (192 and 193, respectively). The preterm delivery (< 34 weeks) was significantly less frequent in the pessary group than the expectant management group (6 vs. 27 %).

Preterm delivery in twin/multiple pregnancies

Preterm labor and delivery occurs more frequently in twin pregnancies than singleton pregnancies. The prediction and prevention of preterm delivery is a major concern for clinicians managing twin pregnancies.

The shortening of cervical length in the second trimester is one of the best indicators for predicting preterm delivery in singleton as well as twin pregnancies [44]. Sequential measurements of cervical length in the second trimester starting at < 20 weeks may be a suitable parameter to predict preterm labor and delivery in twin pregnancies [44].

Cervicovaginal fetal fibronectin provides moderate to minimal prediction of preterm birth in women with multiple pregnancies. The test is most accurate in predicting spontaneous preterm birth within 7 days of test in women with twin pregnancies and threatened preterm labor [45]. The combination of cervical length and fFN test between 24 and 28 weeks’ gestation was also studied as a predictor of preterm birth in twin pregnancies by the Preterm Prediction Study. Both positive fFN test and a short cervical length were associated with the highest rates of preterm birth at 32, 35, and 37 weeks’ gestation [46]. In asymptomatic twin pregnancies, Fox et al. [47] showed that fFN and cervical length between 22 and 32 weeks’ gestation could identify pregnancies that were significantly increased risk for preterm birth, including deliveries at 28 weeks’ gestation.

To prevent the preterm delivery, the progesterone administration has not been associated with decreasing of preterm delivery rates among women with multiple pregnancies and a short cervix [4850]. In twin pregnancies, cerclage has been associated with significantly higher incidence of preterm delivery, so this method is not indicated to prevent the preterm delivery [51].

Therefore, there is evidence that sonographic cervical length assessment and fetal fibronectin test can identify twin pregnancies at risk for preterm delivery, but, to date, there are not proven interventions for prevention of preterm delivery in twin/multiple pregnancies [51].

Conclusion

Prophylactic progesterone use, cervical cerclage, and cervical pessary in selected cases may reduce the risk of preterm delivery. However, the preterm delivery rate has not diminished over the last 30 years and has increased in developed countries, mainly because of failure to identify the group presenting a high risk of spontaneous preterm delivery during routine prenatal assessments. Transvaginal ultrasonography is an efficient and highly reliable method for assessing cervical biometry and morphology during pregnancy. However, cervical shortening is a syndrome that may have several etiologies. It is becoming increasingly clear that identifying women who would benefit from an intervention, including cerclage, should not be done solely on the basis of the obstetric history or cervical ultrasound findings. Thus, successful management of patients presenting a short cervix during pregnancy depends on the understanding that cervical shortening is the final common path for several causes of preterm delivery, such as isthmocervical incompetence, intra-amniotic or cervical infection/inflammation, and premature uterine contractions. Hence, the best approach towards short cervix syndrome needs to be adapted to its etiology, according to the patient in question.

Conflict of interest

None.

Copyright information

© Springer-Verlag Berlin Heidelberg 2013