With its accuracy, scope, and reproducibility, might transvaginal cervical imaging help us not only predict preterm labor but understand its mechanisms? The early and timely diagnosis of preterm labor (PTL) continues to elude even the most accomplished obstetrician. The diagnosis of PTL is usually based on the presence of regular uterine contractions between 20 and 37 weeks gestation in conjunction with progressive cervical change, cervical dilatation of 2 cm or more, or cervical effacement of 80% or more.1

Palpation
For many generations, digital palpation of the pregnant cervix has been the cornerstone of evaluating the gravid patient. The classic digital examination assesses the position, effacement, softness, and dilatation of the cervix. All of these parameters are subjective and have high intra- and interobserver variability.2,4 Furthermore, palpation is limited to the vaginal portion of the uterine cervix. Even the most experienced clinician cannot digitally evaluate the upper half of the cervix (i.e., the area immediately adjacent to the internal os), and this is the significant region for detecting incipient labor. Effacement and shortening of the cervix start here but may go undetected by digital palpation regardless of the examiner's experience.

Ultrasonography
Ultrasonography has changed the practice of obstetrics and gynecology by allowing the practitioner to visualize the pelvic organs and developing fetus. Although there is a large volume of literature on the ultrasonographic appearance of the pregnant and nonpregnant uterus, fewer articles deal exclusively with the gravid cervix. In my opinion, ultrasonography especially transvaginal ultrasonography (TVS) is a safe, accurate, objective, and reliable method for evaluating the gravid cervix. The first articles describing the ultrasonographic appearance of the cervix used transabdominal ultrasonography, which depends on the presence of a full bladder. Visualization of the cervix is poor if the bladder is not sufficiently full, but an overfull bladder can bias the ultrasonographic measurement of cervical length upward.

Currently, transvaginal and transperineal ultrasonography are the premier cervical imaging modalities. Transperineal ultrasonography produces adequate cervical visualization, but the images lack the high resolution and clarity of the TVS images. Transvaginal ultrasonography has high resolution, which in turn translates into high reproducibility. In addition, it produces a record of the cervical image, which can be reviewed and remeasured later if necessary.

Cervical Measurement
A technically correct cervical image produced by TVS can minimize intra- and interobserver variability. The resulting cervical measurement is more accurate and covers the entire length of the cervix including the segment above the portio vaginalis, which is inaccessible to palpation. Both the cervical image and the measurements can be standardized and stored for future evaluations5. TVS permits evaluation of the internal os and its adjacent areas before cervical dilatation occurs. If the internal os is dilated, TVS can detect whether the amniotic membranes have herniated into the endocervical canal.

TVS permits real-time imaging of the cervix, allowing for dynamic changes that may occur during scanning to be observed and recorded. This requires a somewhat longer observation time (i.e., several minutes) in the longitudinal plane because the uterine contractions that modulate the shape of the internal os and cervical length may be infrequent. Observation of these dynamic changes can furnish valuable clinical data for patient management.

Transvaginal ultrasonographic cervical findings that have been found to correlate positively with preterm delivery include decreased length,5,9 funneling,10,12 and positive stress test results (e.g., fundal pressure).13 A large, multicenter study of 2915 women scanned at 24 weeks gestation and 2531 scanned at 28 weeks found that cervical length is normally distributed (mean 35, standard deviation 8.3 mm; and mean 33.7, standard deviation 8.5 mm, respectively).9 This study agrees with others in the literature in concluding that the shorter the length of the cervix, the greater the risk of preterm delivery. In addition, this study goes a step further by calculating the relative risk of preterm delivery at different cervical lengths.

The ultrasonographic observation of funneling or wedging translates in the digital examination as effacement. TVS permits visualization of the progression of effacement from the internal to the external os. Effacement can be imaged with TVS even in the absence of cervical dilation. In a study of 70 women with singleton pregnancies and no known risk factors admitted to the hospital with contractions between 20 and 35 weeks gestation, the TVS finding of wedging or funneling was predictive of preterm labor.12 Results indicated sensitivity of 100%, specificity of 74.5%, positive predictive value of 59.4%, and negative predictive value of 100%. Another study confirmed these findings and suggested that the proportion of the cervix affected by funneling correlates directly with the rate of preterm labor.10

Conclusion
The era of relying on the digital examination as the only means of assessing the gravid cervix is drawing to a close. Currently, most hospital labor and delivery units, physician s offices, and outpatient clinics have ultrasound machines and transvaginal probes.

Cervical changes represent the final stage of a cascade of different triggering mechanisms operating via numerous avenues, and we do not yet fully understand how these changes occur. However, these changes are readily detected with TVS, and it is hoped that such screening will result in earlier initiation of therapy in patients who really need it. Only through further observation of cervical changes, prospective clinical studies, basic research, and the incorporation of testing for biochemical markers such as fetal fibronectin14 can we finally begin to gain an understanding of why some women go into labor before term.

READ THE OPPOSING VIEW: