High-resolution ultrasonography allows the most important imaging signs of cervical incompetence to be determined. The parameters assessed include the cervical length and the appearance of the internal os. [1, 2, 3, 4, 5, 6, 7, 8, 9, 10] Previously, hysterosalpingography (HSG) was used to diagnose an incompetent cervix. Evidence of a funnel-shaped cervix or a cervix appearing as an inverted sac with an increased diameter of internal uterine orifice were seen on HSG.
See the ultrasound images below regarding cervical incompetence.
The radiologist should be able to recognize these imaging findings and report them to the referring clinician. The clinician may then select patients who should undergo serial ultrasound studies from the start of the second trimester of pregnancy or determine suitable treatment based on ultrasound findings suggestive of incompetence before clinical examination. Anticipated clinical and technological improvements in 3-dimensional ultrasonography and MRI may hopefully depict those changes in the cervical connective structures that are responsible for incompetence. 
Ultrasonography remains an operator-dependent modality, and many pitfalls are possible with regard to imaging technique or interpretation.  Ultrasonographic results correlate well with findings from digital examination. However, the consistency of the cervix (soft or firm), an important sign of cervical 'change' that is suggestive of impending miscarriage, can be assessed only digitally.
Ultrasound measurement of the cervix should be performed by a standardized method. The method described by Colombo and Iams has been widely accepted and used in obstetrics.  The examination is performed with the bladder empty. A transvaginal probe (5-7 MHz) is introduced into the anterior vaginal fornix, taking care to avoid undue pressure on the cervix to avoid distorting the anatomy. The whole length of the sonolucent endocervical mucosa is identified in the sagittal section, and the image is magnified to occupy 75% of the screen. Calipers are placed from the triangular echodense area marking the external os to the V-shaped indentation that marks the internal os; this distance is measured in a straight line.
MRI is expensive and currently has somewhat limited availability. In addition, MRI poses potential hazards in patients with ferromagnetic foreign bodies, in those with some cardiac pacemakers, and in those with claustrophobia. [12, 13]
In patients with a history of pregnancy loss, it is prudent to establish a baseline cervical-length measurement at 12-14 weeks of gestation. In addition, follow-up ultrasonograms should be obtained every 1-2 weeks, depending on the findings on ultrasound or from recent examination or depending on the patient's symptoms.
In patients at risk for pregnancy loss, placement of cervical cerclages in response to ultrasound-detected shortening of the endocervical canal is an acceptable alternative to elective cerclage. [14, 15, 16, 5, 6] Elective cervical cerclage is offered to women with a history of previous mid-trimester loss due to possible cervical incompetence at 14-16 weeks. An emergency or ‘rescue’ cerclage is sometimes offered to women who present with bulging membranes and no evidence of infection. Thus, in clinical practice, there are 3 scenarios where cervical cerclages could be used:
- Elective cerclage at 14-16 weeks: in high risk cases, as identified from the history of previous mid-trimester pregnancy losses
- Emergency or rescue cerclage: can be used up to 25-26 weeks, in response to an open cervix with bulging membranes; clearly, there are risks of premature rupture of membranes in this situation, and the operation is often unsuccessful in prolonging the pregnancy for a considerable period of time, but it may allow for time to administer steroids and transfer to a neonatal ICU center
- Cervical cerclage in response to shortened cervical length (as detected by ultrasound): there seems to be increasing evidence that this may be a reasonably successful option
The Society for Maternal-Fetal Medicine recommends that routine transvaginal cervical length screening be performed for women with singleton pregnancy and history of prior spontaneous preterm birth and that routine transvaginal cervical length screening not be performed for women with cervical cerclage, multiple gestation, preterm premature rupture of membranes, or placenta previa. 
In a retrospective study of 88 singleton pregnancies after cerclage, cervical length measurements performed perioperatively by transvaginal ultrasound showed that in women with delivery at less than 35 weeks, cervical length declined from the 16th to the 22nd weeks of gestation. Cervical length measurements were performed perioperatively and at weeks 16 + 0, 18 + 0, 20 + 0, and 22 + 0 by transvaginal ultrasound. In univariate analysis, all cervical length measurements were predictive of delivery at less than 35 weeks in women who underwent ultrasound-indicated cerclage and in women who received a history-indicated cerclage, whereas in multivariate analysis, only cervical length 3 to 6 days after cerclage remained significant. In women with ultrasound-indicated cerclage, optimized cut-off was 20 mm or less (specificity 83.8%, sensitivity 84.2%). 
Papanna et al conducted a multicentric, retrospective cohort study with 163 patients with a short cervix before fetoscopic laser photocoagulation (FLP) for twin-to-twin transfusion syndrome. The purpose of this study was to evaluate the benefit of cervical cerclage for cervical length of at least 25 mm at the time of FLP for twin-to-twin transfusion syndrome. Seventy-nine of the patients (48%) had cerclage placement at the surgeon's discretion. The outcome measures that were compared were gestational age at delivery and perinatal mortality rates for patients with cerclage and those who were treated conservatively. The authors concluded that the benefit of cerclage for patients with short cervix before FLP remains questionable. 
In a Danish study, the recurrence rate of second trimester miscarriage or extreme preterm delivery was 28% in women with cervical insufficiency without cerclage; prophylactic cerclage was associated with a significant reduction in recurrence. 
Normal cervix in pregnancy
In the nongravid state and in the first trimester (see the images below), the cervix is usually difficult to distinguish from the lower uterine segment. From the middle trimester onward, the amniotic sac clearly defines the internal os.
The length of the cervix is the distance between the internal os and the external os. First-trimester cervical length is greater than 5 cm. The cervical length decrease to less than 34 mm at 28 weeks' gestation. In patients who have had multiple gestations, the cervix is notably shortened. [21, 7, 8]
Nomograms have been developed for cervical length in twin and triplet pregnancies; these help in identifying patients at high risk for preterm labor.
Cervical length, as determined on ultrasonograms, is inversely proportional to the risk of preterm labor. Rates of preterm labor increase fourfold among women with a cervical length of less than 25 mm at 24 weeks' gestation. [22, 23, 9, 10]
The cervix at term
Several groups of researchers have used ultrasonography to evaluate the favorability and inducibility of cervix, and a scoring system has been proposed. Parameters such as cervical length, cervical diameter, and the appearance of the internal os are scored. Among these indices, the presence of funneling and a short cervix is associated with a shortened duration of induced labor.
Cervical incompetence is defined as cervical dilatation without uterine contractions. It is a functional condition associated with a history of recurrent and usually painless, spontaneous second-trimester abortions. The typical clinical scenario is that of ‘silent’ or ‘painless’ dilatation of the cervix, leading to bulging fetal membranes extruding through the external os.
Cervical incompetence is a frequent cause of inevitable miscarriage in the second trimester, and it is often associated with a poor fetal outcome. In a multiparous woman, a history of previous miscarriages may raise the suspicion for cervical incompetence. However, in a primigravida without previous risk factors, early diagnosis may not be possible. Nevertheless, early diagnosis in such primigravida women would allow for prompt treatment with an improved prognosis.
Pregnant women with cervical incompetence usually present with silent cervical dilatation between 16 and 24 weeks of gestation. In addition, cervical incompetence may occur with clinically significant cervical dilation of 2 cm or more but with minimal symptoms. When the cervix reaches 4 cm or more, active uterine contractions or rupture of membranes may ensue.
The diagnosis is usually based on the patient's history and findings from physical examination, with the aid of ultrasonography.
Preterm labor is a frequent cause of admission to the hospital, but most cases of preterm labor are diagnosed on the basis of only clinical criteria, and only a few patients admitted to the hospital with suspected preterm labor ultimately have a preterm delivery. Therefore, sensitive methods for identifying patients who are genuinely at high risk for preterm birth are needed.
Developments such as ultrasound measurement of cervical length and fetal fibronectin (FFN) measurement may improve prediction of the risk of actual preterm delivery. FFN is a glycoprotein found in amniotic fluid, placental tissue and the deciduas basalis. It is normally found in the cervicovaginal secretions before 16 weeks of pregnancy, a time at which the fusion between the amnion and the decidua is not yet completed. It is not normally present in the cervicovaginal secretions after 22 weeks. It appears in the secretions due to disruption of the chorionic-decidual interface and is often secondary to infection.
In women being screened for preterm labor, FFN provides a good negative predictive test (ie, a negative FFN rules out preterm labor). However, FFN cannot be used as a test for cervical incompetence because, typically, the pregnancy loss in these cases is around 16-20 weeks, a time when FFN may be found in the cervicovaginal secretions in normal conditions.
Future developments are likely to be based on anatomic changes as well as functional criteria. New techniques are being developed to evaluate cervical distensibility, to permit noninvasive measurement of cervical collagen content (eg, by using light-induced fluorescence of cervical collagen), or even to enable direct assessment of the changes with cervical water-content MRI. Although correlations have been found between these parameters and the risk of preterm labor, widespread clinical studies are required to determine true predictive values for these new methods in clinical practice.
A dynamically changing cervix is a dramatic finding that can be seen during ultrasonography. The cervix shows varying degrees of funneling of the internal os over time in association with vigorous closing and opening of the cervix. These cervical changes occur in the absence of uterine contractions.
Although most patients with a spontaneously changing cervix deliver preterm, this is not a sign of imminent delivery. With conservative management, delivery may be delayed in some patients. Measurements obtained when the cervix appears most abnormal are most predictive of early delivery.
During transvaginal ultrasonography in twin gestations, a cervical measurement of greater than 35 mm at 24-26 weeks helps in identifying patients who are at low risk for delivery before 34 weeks of gestation. [21, 24]
In twin gestations, the use of ultrasound and clinical criteria to select patients for cerclage helped prevent the birth of the youngest and smallest of the twins and significantly decreased perinatal mortality.  Routine cerclage is not recommended for twin gestations, but multifetal gestations may benefit from sonographic surveillance for cervical incompetency. 
Magnetic Resonance Imaging
Magnetic resonance imaging findings suggestive of cervical incompetence may be summarized as follows:
Shortening of the endocervical canal to less than 3 cm
Widening of the internal os to greater than 4 mm
Asymmetric widening of the endocervical canal
Thinning or absence of low signal intensity in the cervical stroma
MRI is not vulnerable to the technical (technologist- and/or transducer-related) considerations that affect ultrasonography. It is vulnerable to the need for established and correct protocols from the point of view of both time of the imaging exam and its correct interpretation.
MRI may demonstrate more soft tissue contrast than ultrasonography in depicting the uterine anatomy. In some instances, MRI may be more accurate than other studies in depicting cervical incompetence in the gravid patient. Pitfalls in the diagnosis of cervical incompetence have not yet been established. However, metallic objects in the bony pelvis can produce artifacts.
Powell et al assessed the usefulness of MRI in demonstrating the pregnant mother's anatomy. They examined the signal intensity of different maternal tissues, using T1- and T2-weighted sequences. The bony pelvis was depicted with sufficient clarity for MRI to be an alternative to conventional radiographic pelvimetry. The placenta and cervix had a distinctive appearance that facilitated the diagnosis of placenta previa. The authors suggested that the demonstration of cervical morphology may offer the potential for investigating currently ill-understood conditions of cervical dystocia and cervical incompetence.
Hricak et al prospectively studied the ability of MRI to demonstrate cervical incompetence in 41 nonpregnant female volunteers. These included 20 normal women, 11 with a clinical history suggestive of cervical incompetence, and 10 with a clinically small cervix (DES exposure in utero). MRI findings of a cervix shorter than 3.1 mm, an internal cervical os wider than 4.2 mm, and abnormal signal intensity in the cervical stroma were highly suggestive of incompetent cervix. However, with the advent of high-resolution ultrasonography, MRI is not likely to be used clinically in the diagnosis of cervical incompetence, as it offers no major advantage over ultrasonography.
Ultrasonography is currently the principal imaging modality for diagnosing cervical incompetence during pregnancy (see the images below). However, various technical factors related to the patient, the technologist, and the transducer may limit its usefulness in evaluating cervical incompetence. [1, 2, 3, 4, 5, 6, 7, 8, 10]
Ultrasonography has advantages over digital examination in that the internal os can be visualized by means of transvaginal ultrasonography; thus, early changes of cervical incompetence can be detected. Moreover, compared to digital examination, ultrasound examination decreases the risk of infection and cervical irritation.
Transvaginal ultrasound can be performed with ruptured membranes and/or vaginal bleeding as long as aseptic precautions are taken. It is important to adhere to a standardized technique to avoid errors. A cervical length of less than 3 cm is considered abnormally short.
Ultrasonographic features of cervical incompetence may be summarized as follows:
Cervical effacement, dilatation, or shortening in the absence of labor in the second trimester
Funneling of the cervix with changes in the forms of T, Y, V, or U
Shortened cervical length in association with a gaping internal os
Membranes protruding into the endocervical canal
Fetal parts in the cervix or vagina
Early changes can be elicited by performing transvaginal ultra sonography combined with the application of transfundal pressure or by scanning the patient in a standing position. However, clearly, excessive pressure should be avoided.
Uterine contractions are differentiated by observing a thickened myometrium in association with uterine contractions, with palpation of the uterus, and with the use of a tocodynamometer. On questioning, the patient may report having uterine contractions.
A narrowed lower uterine segment can mimic funneling. This condition is clarified by observing the cervix over time. An overdistended urinary bladder involving the lower uterine segment may produce the impression of cervical lengthening. Cystic lesions, such as nabothian cysts and vaginal cysts, can mimic a dilated internal os.
Clinicians should exercise caution when interpreting cervical lengths because of the potential anatomic and technical pitfalls that may lead to false negative or false positive diagnoses. Anatomic pitfalls that may hamper identification of the internal os include undeveloped lower uterine segments, focal myometrial contractions, rapid and spontaneous cervical changes, and endocervical polyps. Technical pitfalls include incorrect interpretation of dilatation of the internal os because of the orientation of the vaginal probe and because of artificial lengthening of the endocervical canal due to cervical distortion by the transducer.