Grand Rounds

The short and funneled cervix:
What do I do now?

By Vincenzo Berghella, MD

Measuring cervical length and funneling with transvaginal ultrasound can help ob/gyns predict which patients will deliver preterm. But how do you interpret U/S results and put them to use in patient care? An expert in the field offers practical guidelines for meeting the challenge.

Many clinicians consider transvaginal ultrasound (TVU) a blessing; some would say it's the most important development in obstetrics and gynecology in the last 25 years.¹ Its ability to accurately detect a short or funneled cervix often provides ob/gyns with a way to "see the future" and predict preterm birth (PTB). In fact, measuring cervical length (CL) with TVU allows far earlier and better prediction of PTB than the traditional digital exam of the cervix, making TVU the gold standard for evaluating the cervix in practically all clinical settings, including vaginal bleeding and rupture of membranes.²

By way of contrast, transabdominal ultrasound (TAU) is less accurate because fetal parts can obscure visualization of the cervix, especially after 20 weeks, and because the TAU transducer is further away from the cervix than with TVU. In addition, the full urinary bladder usually required for TAU can elongate the cervix, masking funneling. And while translabial, or transperineal, U/S is not obstructed by fetal parts, and does not require bladder filling, it too has a major shortcoming: Gas in the rectum or vagina can hamper visualization of the cervix.

Of course, while the reliabilty of TVU of the cervix in pregnancy is well supported by clinical research, its accuracy in predicting PTB and its usefulness in preventing prematurity depends on proper technique, the clinical setting, and judicious use of cerclage. My goal here is to first outline the best technique for measuring cervical length and funneling, and then to provide practical advice on interpreting the findings and putting them to use in planning patient care.

Make sure to use proper technique—consistently

Current recommendations for performing TVU of the cervix are outlined in "How to assess cervical length with TVU". For best results, the whole length of the cervix should be visualized, a symmetric image of the external os should be obtained, and the distance from the surface of the posterior lip to the cervical canal should be equal to the distance from the surface of the anterior lip to the cervical canal. There should not be any increased echogenicity in the cervix, that would be a sign of excessive pressure.

While TVU imaging of the cervix is usually straightforward, some anatomic and technical difficulties are encountered in about one fourth of patients. For this reason, some experts recommend that sonographers be supervised for their first 50 procedures so that they can acquire the needed skills.

Knowing what to measure

Different cervical parameters have been evaluated to see how well they can predict preterm birth. As Figures 1 and 2 show, CL is measured from the internal to the external os along the endocervical canal. If the cervical canal is curved—a deviation of the canal of more than 5 mm from a straight line connecting the external and internal os—the CL can be either traced or you can use the sum of two straight lines that essentially follow the curve.⁵ Keep in mind that a short CL is mostly straight, and the presence of a curved cervix usually signifies a CL that is longer than 25 mm and is therefore a reassuring finding. If the cervical canal is closed, CL is the only parameter you need to measure.

The internal os is open in about 25% to 33% of high-risk patients.^4,5 In this case, the open portion of the cervix (the funnel length) and the internal os diameter (the funnel width) can be measured. Funneling is defined as funnel length (B in Figure 2) divided by total CL. Total CL in this case refers to the sum of A, the closed portion of the endocervical canal (the CL), and B, the length of the open cervix. Because funneling can be difficult to measure and there's often variability in these measurements among different examiners and different centers, CL has emerged as the gold standard cervical measurement to predict PTB.⁶

However, even in the presence of a normal (>25 mm) CL, funneling above a normal CL will likely progress to a short CL and eventual PTB.⁵ CL may change dynamically during the course of a 5- to 10-minute examination in less than 5% of patients who have had TVU, and in some cases funneling of the upper cervical canal may appear and resolve. Similarly, the cervix may shorten in response to transfundal pressure in about 5% of cases. When changes occur, the shortest best CL should be recorded. In most cases in which the cervix shortens spontaneously or in response to fundal pressure, it is already abnormal at baseline. At least one researcher has shown that dynamic changes, either spontaneous or after applying transfundal pressure, significantly improved the predictive accuracy of TVU for PTBs.⁵ None of the other parameters of cervical change including funnel width, funnel length, anterior and posterior cervical width, cervical angle, cervical position (horizontal vs. vertical), lower uterine segment thickness, vascularity, and visibility of chorionamnion at internal os, cervical index (funnel length +1/functional length) has proven more reliable or predictive than CL.

Know what's normal, and what's abnormal

In low-risk women, CL during pregnancy has a mean of 35 to 40 mm from 14 to 30 weeks, the lower 10th percentile being 25 mm and the upper 10th (90th percentile) 50 mm.^6,7 A progressive and natural shortening of CL is noticed after 30 weeks even in patients who deliver at term.⁸

Some investigators have reported a very slight progressive linear reduction of CL even before 30 weeks.⁹ A CL less than 25 mm is abnormal between 14 and 24 weeks. The shorter the cervix, the higher is the risk of PTB.

When you do detect a short cervix, the patient's obstetrical history determines the precise risk of PTB. Table 1 describes how accurate the CL measurement is in predicting PTB in different asymptomatic populations. The better studies provide details on the measurement technique used and blind physicians to U/S results.^6,7,10

You'll notice that a CL below 25 mm has been linked to very different positive predictive values (PPV) (probability of PTB <35 weeks) in different populations: For instance, when a CL below 25 mm is found in low-risk women with an overall incidence of PTB of 4%, the PPV is only 18%; when the same 25 mm cut-off is used in high-risk patients with a history of PTB at less than 32 weeks with an overall incidence of PTB of 26%, the PPV jumps to 55%. Put another way, 82% of low-risk patients who were found to have a short CL at 24 weeks delivered at or after 35 weeks without intervention.

Several studies have been performed in women at high risk for PTB, with high risk usually based on obstetric history, including, but not limited to, prior PTB, müllerian anomaly, two or more D&Es, DES exposure, cone biopsy, and twin gestation. While many women would be expected to have a PTB based solely on their histories, a CL at or above 35 mm between about 18 and 24 weeks was correlated with preterm delivery risk of only 4% in both high-risk singleton and twin gestations.^11,12

These data can help obstetricians to avoid bed rest and other interventions commonly used in these pregnancies. On the other hand, a CL below 25 mm in patients with a history of PTB at less than 32 weeks achieves sensitivity and PPVs of 69% and 55%, respectively.⁵ The higher the baseline risk of PTB, the higher the PPV of TVU of the cervix will be. TVU of the cervix also predicts PTB in women with triplets, a cerclage in place, preterm labor, and preterm premature rupture of membranes, and possibly the length of labor and incidence of cesarean section in women induced at term.

Linking cervical assessment to gestational age

Almost all patients, even those at the highest risk, have a normal CL in the first and early second trimesters. One study that used TVU to assess CL in high-risk women, for example, found only 5% had a CL below 25 mm between 10 and 14 weeks.¹³ So CL is not very useful for predicting PTB at this point during the pregnancy. An additional drawback of very early screening is the fact that the lower uterine segment is difficult to distinguish from the true cervix in the late first and early second trimesters.

The most common gestational age at which a short cervix or funneling develops is 18 to 22 weeks.^3-5 So if a screening program is to only include one CL assessment, perform it during this interval. Moreover, the earlier the short CL is detected, the higher the risk of PTB. For example, a CL below 25 mm in high-risk women is associated with a 70% risk of PTB at less than 35 weeks if detected at 14 to 18 weeks but a 40% risk if detected at 19 to 24 weeks.³ Therefore, it may be that patients with the highest risk of PTB will benefit from a 14- to 18-week U/S examination to determine their need for intervention. Unfortunately the ideal interval for repeating TVU has not been clearly established.

If a clinician wants to develop a screening program for relatively low-risk women, one TVU of the cervix at around 18 to 22 weeks would probably be most effective. One normal TVU CL between 14 and 18 weeks and another between 18 and 22 weeks is reassuring in most high-risk women. In women at very high risk for PTB, such as those with a prior second-trimester loss or very early spontaneous PTB, some have advocated cervical TVU every 2 weeks, at least between 14 and 24 weeks.

As you might expect, one's ability to accurately predict a PTB is greater with serial TVU, as opposed to just one.⁵ The fact that TVU at 16 to 22 weeks is at least as predictive of PTB as TVU after 22 weeks is important, since interventions to prevent PTB are most effective when changes leading to PTB are detected early in the process.

Avoid cerclage if TVU is normal

Many investigators believe the high negative predictive value of TVU of the cervix is important, since physicians can reassure patients with a normal CL that they don't need any medical intervention. On the other hand, three randomized studies on prophylactic cerclage without the use of TVU have not been able to show that cerclage prevents PTB in high-risk singleton gestations, except in a subgroup of women with three or more prior second-trimester losses or PTBs.¹⁴ With that in mind, only these kinds of patients should be offered prophylactic cerclage at 12 to 14 weeks, based solely on obstetric history.

One randomized and three non-randomized trials have evaluated whether women with suspected incompetent cervices can be safely followed with TVU, with placement of a cerclage only in those with TVU-documented cervical change, (in other words, therapeutic cerclage).^15-18 At least 60% of these very high-risk women maintain a normal CL until after 24 weeks, deliver at term, and can be spared any intervention. About 40% develop a short CL, placing them at true risk of PTB, and can be offered intervention. While management with serial TVU of the cervix appears to be a safe alternative to traditional prophylactic cerclage, we still need larger randomized trials to prove that this management approach is appropriate. In fact, the question of what to do after you discover that the patient's cervix is significantly shortened or funneled is the most important one to answer. A high negative predictive value alone is not sufficient to justify a screening program. For TVU to be judged cost-effective, there must be an effective intervention to prevent PTB if the test is positive. For asymptomatic patients with a short CL, the only intervention studied so far has been cervical cerclage.

Staying current on the cerclage research

Five nonrandomized studies have looked at the benefit of therapeutic cerclage in singleton gestations when short CL is identified on TVU. Three used screening in a cross-section of low- and high-risk patients, and two screened only high-risk women.^19-21 While all these studies have some methodological problems, two of the three cross-sectional and one of the two high-risk studies reported no benefit from cerclage.^11,20,21

Randomized trials of cerclage are extremely difficult to perform because many doctors and patients believe the procedure is beneficial. Randomized trials of therapeutic cerclage to prevent PTB in women with a short CL have produced varying results, depending on the type of patients studied. One trial in an unselected population of 113 women with either CL less than 25 mm or funneling greater than 25% showed cerclage didn't prevent PTB.²³ In contrast, a trial involving 35 patients that included only patients at high risk for PTB—many of whom were suspected of having cervical incompetence—showed cerclage did reduce preterm births, along with neonatal morbidity and mortality.²⁴ Interestingly, when Rust and associates re-analyzed a subgroup of their study population that had a prior PTB, they still couldn't find a benefit from cerclage.²⁵

A recent meta-analysis concluded that "the role of cerclage in women whose U/S reveals short cervix remains uncertain."²⁶ Similarly ACOG states that the 'management of women who have U/S findings of a short cervix or funneling remains uncertain.'²⁷

Offering practical recommendations

Using TVU to assess CL is an effective way to predict PTB and "incompetent cervix," now better named cervical insufficiency.²⁷ It's safe and patients accept the examination well. A CL of below 25 mm between 16 and 24 weeks' gestation clearly increases the threat of PTB. In fact, the shorter the cervix, the higher the risk of PTB, and the earlier in gestational age at which the shortening occurs, the higher the risk.

The role of TVU of the cervix has been studied in a wide variety of settings and with different patient populations showing good prediction of PTB in all. Screening frequency should depend on severity of obstetric history, with serial TVU of the cervix having a better predictive accuracy than one, especially in high-risk populations.

I do not recommend screening with TVU of the cervix for women with singleton pregnancies who are not at risk of PTB. If a short or funneled cervix is detected in such a patient, the probability of delivering at or beyond 35 weeks is 80% or more without intervention. It would be difficult to find an intervention to improve on this prognosis, and it would require a very large population and a powerful intervention. In women with obstetric risk factors for PTB, the finding of a short cervix is an ominous one. Monitoring for contractions, counseling, and bed rest are frequently used but remain unproven. The efficacy of tocolytics, antibiotics, and progesterone hasn't been studied in this clinical setting. There is controversy about whether therapeutic cerclage is beneficial for women with a short cervix at less than 24 weeks. Probably the more high risk the patient is by obstetric history, the more beneficial the cerclage will be. Until this is proven by properly conducted, large randomized trials, however, screening high-risk women with TVU of the cervix and placement of a cerclage for the short or funneled cervix should not be considered standard of care.

REFERENCES

1. Mishell DR Jr. What is the most significant development in your area of expertise in the last 25 years? Contemporary OB/GYN. 1998;(May Special Anniversary Issue)43:30.

2. Carlan SJ, Richmond LB, O'Brien WF. Randomized trial of endovaginal ultrasound in preterm premature rupture of membranes. Obstet Gynecol. 1997;89:458-461.

3. Berghella V, Tolosa JE, Kuhlman K, et al. Cervical ultrasonography compared with manual examination as a predictor of preterm delivery. Am J Obstet Gynecol. 1997;177;723-730.

4. Berghella V, Kuhlman K, Weiner S, et al. Cervical funneling: sonographic criteria predictive of preterm delivery. Ultrasound Obstet Gynecol. 1997;10:161-166.

5. Owen J, Yost N, Berghella V, et al. Mid-trimester endovaginal sonography in women at high risk for spontaneous preterm birth. JAMA. 2001;286:1340-1348.

6. Iams JD, Goldenberg RL, Meis PJ, et al. The length of the cervix and the risk of spontaneous premature delivery. National Institute of Child Health and Human Development Maternal-Fetal Medicine Unit Network. N Engl J Med. 1996;334:567-572.

7. Andersen HF, Nugent CE, Wanty SD, et al. Prediction of risk for preterm delivery by ultrasonographic measurement of cervical length. Am J Obstet Gynecol. 1990;163:859-867.

8. Andersen HF. Transvaginal and transabdominal ultrasonography of the uterine cervix during pregnancy. J Clin Ultrasound. 1991;19:77-83.

9. Gramellini D, Fieni S, Molina E, et al. Transvaginal sonographic cervical length changes during normal pregnancy. J Ultrasound Med. 2002;21:227-232.

10. Goldenberg RL, Iams JD, Miodovnik M, et al. The preterm prediction study: risk factors in twin gestations. National Institute of Child Health and Human Development Maternal-Fetal Medicine Units Network. Am J Obstet Gynecol. 1996;175:1047-1053.

11. Berghella V, Daly SF, Tolosa JE, et al. Prediction of preterm delivery with transvaginal ultrasonography of the cervix in patients with high-risk pregnancies: does cerclage prevent prematurity? Am J Obstet Gynecol. 1999;181:809-815.

12. Yang JH, Kuhlman K, Daly S, Berghella V. Prediction of preterm birth by second trimester cervical sonography in twin pregnancies. Ultrasound Obstet Gynecol. 2000;15:288-291.

13. Berghella V, Talucci M, Desai A. Does transvaginal sonographic measurement of cervical length before 14 weeks predict preterm delivery in high-risk pregnancies? Ultrasound Obstet Gynecol. 2003;21:140-144.

14. Final report of the Medical Research Council/Royal College of Obstetricians and Gynaecologists multicentre randomised trial of cervical cerclage. MRC/RCOG Working Party on Cervical Cerclage. Br J Obstet Gynaecol. 1993;100:516-523.

15. Althuisius SM, Dekker GA, van Geijn HP, et al. Cervical incompetence prevention randomized cerclage trial (CIPRACT): study design and preliminary results. Am J Obstet Gynecol. 2000;183:823-829.

16. To MS, Palaniappan V, Skentou C, et al. Elective cerclage vs. ultrasound-indicated cerclage in high-risk pregnancies. Ultrasound Obstet Gynecol. 2002;19:475-477.

17. Berghella V, Haas S, Chervoneva I, et al. Patients with prior second-trimester loss: prophylactic cerclage or serial transvaginal sonograms? Am J Obstet Gynecol. 2002;187:747-751.

18. Kelly S, Pollock M, Maas B, et al. Early transvaginal ultrasonography versus early cerclage in women with an unclear history of incompetent cervix. Am J Obstet Gynecol. 2001;184:1097-1099.

19. Heath VC, Souka AP, Erasmus I, et al. Cervical length at 23 weeks of gestation: the value of Shirodkar suture for the short cervix. Ultrasound Obstet Gynecol. 1998;12:318-322.

20. Hibbard JU, Tart M, Moawad AH. Cervical length at 16-22 weeks' gestation and risk for preterm delivery. Obstet Gynecol. 2000;96:972-978.

21. Hassan SS, Romero R, Maymon E, et al. Does cervical cerclage prevent preterm delivery in patients with a short cervix? Am J Obstet Gynecol. 2001;184:1325-1331.

22. Guzman ER, Benito CW, Yeo L, et al. Bed rest versus cervical cerclage in the treatment of cervical incompetence manifested by ultrasound around the time of fetal viability. Am J Obstet Gynecol. 1999;180:S78. Abstract 249.

23. Rust OA, Atlas RO, Reed J, et al. Revisiting the short cervix detected by transvaginal ultrasound in the second trimester: why cerclage therapy may not help. Am J Obstet Gynecol. 2001;185:1098-1105.

24. Althuisius SM, Dekker GA, Hummel P, et al. Final results of the Cervical Incompetence Prevention Randomized Cerclage Trial (CIPRACT): therapeutic cerclage with bed rest versus bed rest alone. Am J Obstet Gynecol. 2001;185:1106-1112.

25. Rust O, Atlas R, Fischl S, et al. Does cerclage therapy improve perinatal outcome in patients with a history of previous preterm birth and cervical changes on 2nd trimester transvaginal ultrasound? Am J Obstet Gynecol. 2002;187:S58. Abstract 13.

26. Drakeley AJ, Roberts D, Alfirevic Z. Cervical cerclage for prevention of preterm delivery: meta-analysis of randomized trials. Obstet Gynecol. 2003;102:621-627.

27. American College of Obstetricians and Gynecologists Practice Bulletin No. 48. Cervical insufficiency. Obstet Gynecol. 2003;102:1091-1099.

Dr. Berghella is Associate Professor, and Director, Division of Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, Jefferson Medical College of Thomas Jefferson University, Philadelphia, Pa.

How to assess cervical length with TVU

• Have the patient empty her bladder;

• Insert the clean, condom-covered ultrasound probe (the probe can be inserted by the patient or sonographer);

• Gently place the probe in the anterior fornix of the vagina;

• Obtain a sagittal view of the cervix, with the long axis view of echogenic endocervical mucosa along the length of the canal;

• Withdraw the probe until the image is blurred and reapply just enough pressure to restore the image (avoid excessive pressure on the cervix, as that can elongate it);

• Enlarge the image so that the cervix occupies at least 2/3 of the image, and both the external and internal os are well seen;

• Measure the cervical length from the internal to the external os along the endocervical canal (Figure 1);

• Obtain at least three measurements of the closed portion of the cervical length, and record the shortest best measurement in millimeters; and

• Apply transfundal pressure for 15 seconds, and record any changes in cervical length or funneling.

Key points

• When a cervical length (CL) below 25 mm is found in low-risk women with an overall incidence of PTB of 4%, the positive predictive value (PPV) of CL is 18%. When the same 25 mm cut-off is used in high-risk patients with a history of PTB at less than 32 weeks with an overall incidence of PTB of 26%, PPV jumps to 55%.

• The most common gestational age at which a short cervix or funneling develops is 18 to 22 weeks. So if a screening program is to only include one CL assessment, perform it during this interval.

• While many women would be expected to have a PTB based solely on their histories, a CL at or above 35 mm between about 18 and 24 weeks was correlated with preterm delivery risk of only 4% in both high-risk singleton and twin gestations.

Vincenzo Berghella. Grand Rounds: The short and funneled cervix: What do I do now?

Contemporary Ob/Gyn

Apr. 1, 2004;49:26-34.