SEARCH

SEARCH BY CITATION

Keywords:

  • cervical insufficiency;
  • cervical length;
  • preterm birth

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. References

Objective

To study if a repeat cervical length (CL) measurement in the patient already diagnosed with a short cervix has any additional value in the prediction of preterm delivery.

Study design

This was a retrospective study of singleton pregnancies with cervical lengths 1–25 mm at a gestational age of 16–28 weeks seen in our institution between 2002 and 2005. Patients who were managed expectantly and had a follow-up CL measurement within 3 weeks were included. Delivery data were obtained from the patients' computerized medical records.

Results

Sixty-eight patients met the inclusion criteria. 37% of the patients had a shorter CL on the second measurement. These patients delivered at an earlier gestational age (36 + 4 vs. 38 + 2 weeks, P = 0.031) and were more likely to deliver at < 37 weeks (60% vs. 26%, P = 0.009). The change in the CL correlated with earlier gestational age at delivery and delivery at < 37 weeks.

Conclusion

In patients diagnosed with a short cervix, follow-up CL measurement is a strong predictor of preterm delivery. Greater change in the CL correlates with an earlier gestational age at delivery. In the patient diagnosed with a short cervix, a repeat measurement of CL gives additional predictive value. Copyright © 2006 ISUOG. Published by John Wiley & Sons, Ltd.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. References

Cervical length (CL) measurements are commonly performed in women undergoing second- and third-trimester ultrasound scans. There is an inverse correlation between cervical length and risk of preterm delivery1. Since there are no interventions proven to reduce the risk of spontaneous preterm birth in the low-risk patient found to have a short cervix, routine assessment of cervical length is not currently recommended2. However, despite this, many providers still measure CL in the second trimester. Some low-risk patients found to have a short cervix receive a cerclage. Others are managed expectantly and undergo serial ultrasound CL measurements. Sometimes these repeat measurements are used to devise a management plan (cerclage, bedrest, steroid administration etc.), but there are limited data to support this strategy. It is currently unknown if repeated CL measurements are useful in the patient already diagnosed with a short cervix. Our objective was to determine whether a repeat CL measurement in the patient already diagnosed with a short cervix has additional predictive value.

PATIENTS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. References

A retrospective study was conducted from a chart review of pregnancies seen at our ultrasound unit between 2002 and 2005; approval of the Weill Medical College of Cornell University Institutional Review Board was obtained prior to conducting the study. Women with singleton pregnancies between 16 and 28 weeks' gestation found to have a CL measurement of 1–25 mm on transvaginal ultrasonography who had a follow-up CL measurement within 3 weeks were included in the analysis. Gestational age was based on the last known menstrual period and was confirmed by first- or second-trimester ultrasonography. All CL measurements were taken using transvaginal ultrasound with an empty bladder, with the optimal image defined according to the criteria reported by Iams et al.1. The shortest functional CL was used as this has been found to be the most reproducible measurement3. Women with a cerclage in place at the time a short cervix was detected, and women who subsequently received a cerclage, were excluded from the analysis. Women without follow-up data were also excluded from the analysis.

The primary outcome measured was gestational age at delivery. Secondary outcomes were delivery at < 34 weeks' gestation and delivery at < 37 weeks' gestation. All ultrasound data were retrieved from our computerized ultrasound database. Hospitalization and delivery data were obtained from the patients' computerized medical records. Chi-square analysis, the Mann–Whitney U-test, and Spearman correlation were used for analysis (SPSS 12.0 for Windows© 1989–2003, Chicago, IL, USA).

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. References

Sixty-eight patients met the inclusion criteria over the study period, two of whom did not deliver at our institution, but were seen in our ultrasound unit after 34 weeks' gestation (secondary outcome). The median initial CL was 19.5 mm; the median gestational age was 22 weeks; the median time to follow-up CL measurement was 7 days; the median follow-up CL was 18 mm; and the median gestational age at delivery was 37 + 6 weeks. Baseline and delivery characteristics are shown in Table 1.

Table 1. Baseline and delivery characteristics
ParameterMedian (10%, 90% centiles) or frequency (%)
  • *

    Information was missing for one woman. US, ultrasound.

Age (years)32 (25, 39)
Prior births28/68 (41.2)
Prior preterm births10/68 (14.7)
Cervical length at initial US scan (mm)19.5 (7, 25)
Gestational age at initial US scan (weeks)22 (19 + 0, 27 + 0)
Time from 1st to 2nd US scan (days)7 (5, 15)
Cervical length at second US scan (mm)18 (6, 25)
Shorter cervical length on second measurement25/68 (36.8)
Gestational age at delivery37 + 6* (28 + 3, 40 + 2)
Delivery at < 37 weeks' gestation26/67* (38.8)
Delivery at < 34 weeks' gestation12/68 (17.6)

Twenty-five patients (37%) had a follow-up CL that was shorter than the first measurement. For these 25 patients, the median CL on follow-up measurement was 12 mm (interquartile range 5–18 mm), and the median change in CL was 5 mm (interquartile range 1.5–7.5 mm) shorter. These 25 patients who had a shorter CL on follow-up measurement delivered at an earlier gestational age (36 + 4 vs. 38 + 2 weeks, P = 0.031) and were more likely to deliver at < 37 weeks (60% vs. 26%, P = 0.009). Though the rate of delivery at < 34 weeks' gestation was higher in those with a shorter CL on follow-up measurement, this did not reach statistical significance (28% vs. 12%, P = 0.108; Table 2).

Table 2. Delivery outcome based on whether the second cervical length (CL) was shorter than the first
OutcomeCL shorter on 2nd ultrasound scan (n = 25)CL not shorter on 2nd ultrasound scan (n = 43)P
Median gestational age at delivery (weeks)36 + 438 + 20.031
Delivery at < 37 weeks' gestation15/25 (60.0%)11/42 (26.2%)0.009
Delivery at < 34 weeks' gestation7/25 (28.0%)5/43 (11.6%)0.108

We analyzed the correlation between continuous variables and our primary outcome, gestational age at delivery (Table 3). There was a significant correlation between older maternal age and earlier gestational age at delivery (P = 0.030). There was a significant correlation between change in CL and earlier gestational age at delivery. Increased shortening of the CL correlated with an earlier gestational age at delivery. This was seen when the change was measured absolutely (mm), as a rate of change (mm/day), and as a percentage of the original CL ((1st CL measurement—2nd CL measurement)/1st CL measurement)× 100.

Table 3. Correlation of continuous variables with gestational age (GA) at delivery
ParameterCorrelation to GA at delivery (Spearman's rho)P
  1. US, ultrasound.

Maternal age−0.2670.030
Cervical length at initial US scan0.1200.338
Gestational age at initial US scan0.0410.741
Time from 1st to 2nd US scan0.1450.244
Cervical length at second US scan0.1810.145
Change in cervical length (mm)−0.3000.014
Rate of change in cervical length (mm/day)−0.3180.009
Change in cervical length (%)−0.3310.007

When considering the secondary outcome groups, the median maternal age was significantly greater in those who delivered at < 34 weeks (P = 0.013; Table 4). There was a trend towards greater shortening of the CL on follow-up ultrasound in those delivering before 34 weeks, though the difference was not statistically signficant (Table 4). Those who delivered at < 37 weeks' gestation had significantly greater shortening of their CL on follow-up ultrasound (Table 5).

Table 4. Characteristics of patients delivered before and after 34 weeks' gestation
ParameterDelivered < 34 weeks (n = 12)Delivered ≥ 34 weeks (n = 56)P
  1. CL, cervical length; US, ultrasound.

Median maternal age (years)37310.013
Median CL at initial US scan (mm)16.5200.815
Gestational age at initial US scan (weeks)22 + 621 + 50.607
Time from 1st to 2nd US scan (days)770.355
Median CL at second US scan (mm)17180.406
Median change in CL (mm)−100.08
Median rate of change in CL (mm/day)−0.0800.068
Median change in CL (%)−600.054
Table 5. Characteristics of patients delivered before and after 37 weeks' gestation
ParameterDelivered < 37 weeks (n = 26)Delivered ≥ 37 weeks (n = 41)P
  1. CL, cervical length; GA, gestational age; US, ultrasound.

Median maternal age (years)33310.120
Median CL at initial US scan (mm)18200.363
GA at first US scan (weeks)22 + 1220.964
Time from 1st to 2nd US scan (days)780.081
Median CL at second US scan (mm)16180.095
Median change in CL (mm)−1+ 10.004
Median rate of change in CL (mm/day)−0.1+ 0.10.003
Median change in CL (%)−6+ 40.003

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. References

In our population of patients with a short cervix, the follow-up CL measurement gave useful additional information on the likelihood of preterm delivery. Compared to the patient with a stable CL, the patient with a shorter cervix on follow-up ultrasound delivered almost 2 weeks earlier and was more than twice as likely to deliver at < 37 weeks' gestation. The amount of change in CL also correlated with an earlier gestational age at delivery and delivery at < 37 weeks. This could be used to help counsel the patient. In our patients, the majority of women (63%) did not have a shorter CL on repeat measurement. Only 26% of them delivered at < 37 weeks and only 12% delivered at < 34 weeks. Of the patients with a shorter cervix on follow-up ultrasound scan, these percentages were 60% and 28%, respectively. It is possible that patients with a short cervix in the second trimester can be subdivided into two separate groups: a low-risk group with a stable, short cervix, and a high-risk group with a short cervix that is shortening.

Our findings could also influence recommendations for intervention. The largest randomized trial studying prophylactic cerclage in high-risk patients found little benefit except in the highest-risk patients with three or more prior second-trimester losses or preterm deliveries4. Because of this, many patients who may have had a cerclage in the past are now followed with ultrasound CL measurements. Although it is not recommended by the American College of Obstetricians and Gynecologists (ACOG), many low-risk patients are also routinely screened for CL in the second trimester2.

There are multiple randomized studies of cerclage placement versus expectant management for the patient found to have a short cervix in the second trimester5–8. In all of these studies randomization occurred following the initial diagnosis of a short cervix. A meta-analysis of these studies found that cerclage was beneficial in patients with historical risk factors, singleton pregnancies and a short cervix9. To identify a higher-risk population among women with a short cervix, future studies could focus on those with a shorter cervix on follow-up examination.

Based on our observations, it may be reasonable to carry out studies to see if it is useful to obtain more than one CL measurement prior to deciding whether to place a cerclage or not. It is possible that a high-risk patient with a short but stable cervix would not benefit from a cerclage. It is also possible that a low-risk patient with a cervix that is shortening over time would be more likely to benefit from a cerclage. Prospective studies are needed to support or refute these hypotheses.

Limitations to our study include its retrospective design. The decisions to measure the CL initially and whether to repeat the CL measurement were not standardized. In addition, although we excluded those patients who received a cerclage, the decision to place a cerclage was also not standardized. Had these patients not received a cerclage, and been included in our analysis, the results may have been different. Finally, our population was mostly a low-risk population (only 14% with prior preterm births). It is possible that results obtained from a high-risk population would differ.

References

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. References
  • 1
    Iams JD, Goldenberg RL, Meis PJ, Mercer BM, Moawad A, Das A, Thom E, McNellis D, Copper RL, Johnson F, Roberts JM. The length of the cervix and the risk of spontaneous preterm delivery. N Engl J Med 1996; 334: 567572.
  • 2
    ACOG Practice Bulletin. Cervical Insufficiency. Obstet Gynecol 2003; 102: 10911099.
  • 3
    Yost NP, Bloom SL, Twickler DM, Leveno KJ. Pitfalls in ultrasound cervical length measurements for predicting preterm birth. Obstet Gynecol 1999; 93: 510516.
  • 4
    MacNaughton MC, Chalmers JG, Dubowitz V, Dunn PM, Grant AM, McPherson K, Pearson JF, Peto R, Turnbull AC. Final report of the Medical Research Council/Royal College of Obstetricians and Gynaecologists multicentre randomized trial of cervical cerclage. Br J Obstet Gynaecol 1993; 100: 516523.
  • 5
    Althuisius SM, Dekker GA, Hummel P, Bekedam DJ, van Geijn HP. 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: 11061112.
  • 6
    Rust OA, Atlas RO, Reed J, van Gaalen J, Balducci J. Revisiting the short cervix detected by transvaginal ultrasound in the second trimester: why cerclage therapy may not help. Am J Obstet Gynecol 2001; 185: 10981105.
  • 7
    To MS, Alfirevic Z, Heath VC, Cicero S, Cacho AM, Williamson PR, Nicolaides KH. Cervical cerclage for prevention of preterm delivery in women with short cervix; randomised controlled trial. Lancet 2004; 363: 18491853.
  • 8
    Berghella V, Odibo AO, Tolosa JE. Cerclage for prevention of preterm birth in women with a short cervix found on transvaginal ultrasound examination: a randomized trial. Am J Obstet Gynecol 2004; 191: 13111317.
  • 9
    Berghella V, Odibo AO, To MS, Rust OA, Althuisius SM. Cerclage for short cervix on ultrasonography: meta-analysis of trials using individual patient-level data. Obstet Gynecol 2005; 106: 181189.