Sonographic examination of the cervix is gaining enthusiasm in obstetric practice for the prediction of preterm birth.1 One proponent suggested that ultrasonic cervical measurements “will soon join fetal sonography as a standard part of obstetric diagnostic imaging.”2 The likelihood of spontaneous preterm birth increases as the cervical length shortens and the internal os begins to dilate. In a population-based study of women at low risk for preterm delivery, Iams et al3 found that the incidence of spontaneous preterm birth increased as cervical length at 24–28 weeks decreased. Others have observed a similar association between increased risk of short gestation and short cervical length or funneling of the internal os when viewed by transvaginal ultrasound.4
Although a clear association between cervical findings on transvaginal ultrasound and preterm birth has been described, using this knowledge effectively to modify pregnancy outcomes presents two main problems. First, the majority of women (75%) with shortened cervices do not deliver prematurely,3,5 and second, randomized trials of intervention to prevent preterm births in women with short cervices have not been done. Faced with these dilemmas in the clinical application of transvaginal cervical ultrasound, the practitioners at our institution chose to measure cervical length prospectively in women at risk for preterm birth before adopting this approach for routine clinical management. While implementing this program, we found early on that measuring the cervix using transvaginal ultrasound was difficult and more prone to pitfalls than we expected. This article describes the various anatomic and technical difficulties that we encountered with transvaginal ultrasound imaging of the cervix during pregnancy.
Three groups of women were selected for cervical ultrasound examinations: 1) those with histories of spontaneous or induced preterm births at 36 weeks' gestation or earlier, 2) those diagnosed with incompetent cervices based upon histories of painless second-trimester pregnancy losses, verified by their medical records, and 3) those admitted for preterm labor that did not progress. The women in the latter group presented with contractions and cervical dilation between 2 and 4 cm, which arrested spontaneously. Women with multiple gestations or ruptured membranes were excluded. Clinicians were blinded to the ultrasound results.
In accordance with a protocol approved by the Institutional Review Board of the University of Texas Southwestern Medical Center and Parkland Memorial Hospital, eligible women who gave written consent had a digital cervical examination followed by transvaginal ultrasound imaging of the cervix. We used Acuson-XP10 (Acuson Corporation, Mountain View, CA) ultrasound equipment with 5-MHz or 7-MHz vaginal probe transducers. The ultrasound examinations were standardized and consisted of three image acquisitions before and three after fundal pressure. The measurements included endocervical canal length, funneling length, and dilation of the internal os. Images were obtained in the sagittal plane with the patient recumbent and immediately after voiding. Fundal pressure was standardized by having one research nurse press on the uterine fundus for 3–5 minutes during this phase of the ultrasound examination. All images were reviewed by a single sonologist (DMT), who also performed the examinations that were found to be problematic. Each examination took approximately 10 minutes, which permitted observation of dynamic changes in the cervix that might have occurred spontaneously or from fundal pressure.
Optimum imaging in the sagittal plane was defined according to the criteria reported by Iams et al.3 Briefly, the appropriate view was identified by locating the triangular area of echodensity at the external os, a V-shaped notch at the internal os, and a faint line of echodensity or echolucency between the two, representing the endocervical canal. We avoided undue pressure on the cervix that might artificially increase its length by first obtaining a satisfactory image, then withdrawing the probe until the image blurred, and finally reapplying only enough pressure to restore the image.
Sixty consecutive women had obstetric ultrasound examinations for assessment of cervical length. Most of the women were Hispanic (n = 29) or black (n = 28); the remaining three were white (n = 2) and Asian (n = 1). The primary indication for cervical ultrasound examination was a history of preterm birth, accounting for 77% of the examinations in this series (Table 1). Of these preterm births, 41 were spontaneous and five were induced. Five and nine women, respectively, had ultrasound examinations for incompetent cervices or arrested preterm labor. Gestational age at examination varied according to the indication for ultrasound.
Of the 60 women at risk for preterm birth, 27% (95% confidence interval 16%, 40%) presented difficulties during the examination that made interpretation of cervical length potentially problematic. Six types of problems were encountered while attempting to view the cervix (Table 1). These pitfalls can be divided into those due to anatomic or technical considerations. The most common anatomic difficulty was a poorly developed lower uterine segment that obscured adequate visualization of the internal os. We encountered this problem in examinations done during the early weeks of the second trimester. At this point in pregnancy, the gestational sac is often high in the uterus, and the ultrasonic interface between amniotic fluid and the internal cervical os has not yet formed. Identification of the internal os was also obscured by a focal myometrial contraction in one woman (Figure 1) and by an endocervical polyp in another. Another anatomic pitfall involved spontaneous and rapid changes in the appearance of the internal os from no apparent dilation to dilation with funneling within 3.5 minutes (Figure 2).
Technical difficulties occurred predominantly in women being evaluated for preterm labor that had not progressed. All nine of these women were dilated at least 2 cm by digital examination; however, six did not have dilated cervices by ultrasound. For example, the woman whose cervix is shown in Figure 3 was dilated 2 cm by digital examination, yet the endocervical canal was long and undilated when imaged in the sagittal plane. Fundal pressure did not modify the ultrasound appearance of the cervix. When the same cervix was viewed in the transverse plane, however, the digital and ultrasonic examination findings correlated. Another technical difficulty encountered was due to pressure exerted by the vaginal probe on the anterior lip of the cervix, resulting in apparent elongation of the endocervical canal (Figure 4). Such distortion of the cervix from the transducer can occur with even minimal manipulation.
Approximately one in four cervical length examinations presented anatomic or technical difficulties that made interpretation of cervical length potentially problematic. The main problem was inability to identify consistently the internal os of the cervix. Reasons included early gestational age, uterine contractions, fluctuating cervical length during the brief examination, and technical difficulties in introducing and orienting the transvaginal ultrasound probe. Because of these experiences, we conclude that measurement of the cervix using ultrasound for the prediction of preterm birth is potentially complicated by pitfalls.
The most common pitfall we encountered was the inability to reconcile the discrepancy between digital and sonographic cervical findings. The appearance of an undilated cervix in the sagittal plane, the traditional view for cervical sonography, was found in six of nine women who were evaluated for arrested preterm labor but who were clinically dilated at least 2 cm. The appearance of a closed internal os in these six women was not influenced by fundal pressure. As illustrated in Figure 3, we found a better correlation between digital and ultrasonic findings when the vaginal probe was oriented in the transverse plane, the same plane in which digital examinations are conventionally performed. We suspect that there might be a greater correlation with digital examination if the cervix is imaged in the transverse rather than the sagittal plane.
Some of the anatomic pitfalls we encountered have been noted by others. For example, Sonek et al6 described an inability to view and measure the cervix adequately because of a large endocervical polyp. Unexplained minute-to-minute fluctuations in the internal os, which may significantly alter the appearance of the length of the cervix, have also been described. These fluctuations may occur spontaneously without fundal pressure or uterine contractions.7,8 This dynamic feature of the cervix is analogous to attempting to take a snapshot of a moving target. Not only is the cervix dynamic, but its appearance may also be altered by placing the transvaginal probe too far in the vagina. Burger et al9 described how excessive pressure on the anterior lip of the cervix can artificially elongate the cervical canal.
Because of our experience, and that of others, with potential anatomic and technical pitfalls of ultrasonic cervical length examinations, we caution practitioners who are interested in adding cervical examinations to their ultrasound practice to be wary of falsely reassuring findings.
1. Gomez R, Galasso M, Romero R, Mazor M, Sorokin Y, Goncalves L, et al. Ultrasonographic examination of the uterine cervix is better than cervical digital examination as a predictor of the likelihood of premature delivery in patients with preterm labor and intact membranes. Am J Obstet Gynecol 1994;171:956–64.
2. Iams JD. Cervical ultrasonography. Ultrasound Obstet Gynecol 1997;10:156–60.
3. Iams JD, Goldenberg RL, Meis PJ, Mercer BM, Moawad A, Das A, et al. The length of the cervix and the risk of spontaneous premature delivery. N Engl J Med 1996;334:567–72.
4. Berghella V, Kuhlman K, Weiner S, Texeira L, Wapner RJ. Cervical funneling: Sonographic criteria predictive of preterm delivery. Ultrasound Obstet Gynecol 1997;10:161–6.
5. Craigo SD. Cervical incompetence and preterm delivery. N Engl J Med 1996;334:595–6.
6. Sonek JD, Iams JD, Blumenfeld M, Johnson F, Landon M, Gabbe S. Measurement of cervical length in pregnancy: Comparison between vaginal ultrasonography and digital examination. Obstet Gynecol 1990;76:172–5.
7. Hertzberg BS, Kliewer MA, Farrell TA, DeLong DM. Spontaneously changing gravid cervix: Clinical implications and prognostic features. Radiology 1995;196:721–4.
8. Parulekar SG, Kiwi R. Dynamic incompetent cervix uteri: Sonographic observations. J Ultrasound Med 1988;7:481–5.
© 1999 The American College of Obstetricians and Gynecologists
9. Burger M, Weber-Rossler T, Willmann M. Measurement of the pregnant cervix by transvaginal sonography: An interobserver study and new standards to improve the interobserver variability. Ultrasound Obstet Gynecol 1997;9:188–93.