Human Reproduction, Vol. 16, No. 5, 956-960,
May 2001
© 2001 European Society of Human Reproduction and Embryology
Transvaginal sonographic assessment of cervical length changes during triplet gestation
1 Department of Obstetrics and Gynecology, Assaf Harofe Medical Center, Zerifin 70300 (affiliated with the Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv), Israel and 2 Academic Department of Obstetrics and Gynaecology, Royal Free and University College, UCL Campus, London, UK
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
The current study aimed to evaluate the contribution of transvaginal sonography (TVS) for monitoring cervical changes during the second half of triplet gestation. Forty-five pregnant women with triplets pregnancies were prospectively scanned by TVS from ~26 weeks gestation and were longitudinally followed-up until delivery. Based on a receiver-operating curve it was found that a cervical length of 25 mm is the most accurate parameter (94% sensitivity and 45% specificity) for predicting premature delivery
33 gestational weeks. Thus, a single cervical length measurement of 25 mm at 26 weeks gestation correlated well with premature delivery at 33 weeks (
2; P = 0.002). Using the linear regression model, a mathematical equation [(Week of delivery = 27.4 + 1.6xcervical length; R2 = 0.46; P = 0.01)] for predicting the gestational age of delivery (dependent variable) was determined based on mid-gestation cervical measurements (predictors). In parturient women with triplet gestation, TVS assessment of the uterine cervix offers insight into the cervical status and provides valuable information for prenatal care. This includes both monitoring the cervical changes throughout third trimester as well as predicting the likelihood of premature delivery.
Key words: cervical length/premature delivery/transvaginal sonography/triplet pregnancy
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
Over the past few decades, both maternal age and the application of assisted reproduction technology have risen dramatically. Iatrogenic interventions have made a remarkable contribution to the rate of multi-fetal pregnancies which have resulted from fertility treatment (Seoud, 1992). A population-based analysis of all live births and fetal deaths in the United States between 1983–1988 reported 9523 live births of triplets (Luke, 1996
), whereas 6000 infants in sets of triplets or more were delivered there during 1996 alone (Cefalo, 1998). Although the perinatal mortality rate of multi-fetal gestations has improved dramatically, the fetal death rate for triplets is still five times higher than that for singletons (Luke, 1996), and the risk of neurological or other serious handicaps increases almost geometrically from singletons to higher orders of multiplicity (Keith et al., 1991). The major cause of those complications is due to premature delivery, since the average duration of triplet gestation remains between 32–34 weeks gestation (Luke, 1994), and about 70% of triplets are born before 35 weeks gestation (Kiely, 1998).
It has been reported that pre-labour uterine activity in multi-fetal pregnancies, starting from as early as 20 weeks gestation, is significantly greater than that in singleton pregnancies (Garite et al., 1990). Thus, the onset of labour any time during pregnancy may not necessarily be a sudden event, but rather a culmination of many silent uterine and cervical changes. A transvaginal sonographic (TVS) assessment of the uterine cervix provides insight into the cervical status, including the likelihood of preterm delivery (Maymon et al., 1996; Roberts and Morrison, 1998). Therefore, TVS for cervical length measurement has been proposed as providing valuable information in prenatal care for singletons (Heath et al., 2000), in twin pregnancies (Michaels et al., 1991; Kushnir et al., 1995; Imseis et al., 1997; Souka et al., 1999; Yang et al., 2000) as well as in triplet gestation (Bianco et al., 2000; Guzman et al., 2000a,b). In addition, Ramin et al. (1999) have reported their promising data using transperineal sonographic technique for assessing the cervical length in triplet pregnancies (Ramin et al., 1999).
No definitive cervical landmarks have been established for predicting labour in triplet pregnancies and information on this subject is still lacking (Roberts and Morrison, 1998). The objective of the current study was to evaluate the contribution of TVS for monitoring cervical changes in triplet pregnancies during the second half of gestation.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
This observational study was carried out prospectively between May 1996 and May 2000 in two fetal medicine units, the Assaf Harofe Medical Center, Zrifin, Israel and the University College London Hospital (UCL), London, UK, both using the same diagnostic protocol. Each has a tertiary unit to which pregnant women with multi-fetal pregnancies are referred for target scanning. Gestational age was determined from the date of ovulation in cases of controlled ovarian stimulation or by the date of embryo transfer in cases of assisted reproductive techniques, and thereafter confirmed by the crown-rump length (CRL) measurement (Robinson et al., 1975). Specifically, the gestational age was calculated on the basis of the largest CRL (Sebire et al., 1997
). All patients were cared for according to a management protocol published by Peaceman et al. (1992) and Ramin et al. (1999). Caesarean delivery was performed in those pregnancies that progressed to 36 weeks. Starting from late mid-gestation, those women were offered the opportunity of having periodic TVS assessment of their cervices. The initial TVS for cervical measurement was carried out at ~26 weeks gestation. Another two assessments were planned to take place at around 29 and 32 weeks gestation. Patients with multi-fetal pregnancies which were reduced to triplets (one case), those with cerclage (one case) and those with uterine contraction on the first examination (three cases) were excluded from this study. (A series of ultrasound images of a representative premature spontaneous delivery is presented in Figure 1.) The data on subject characteristics, including patients' age, previous obstetric history, and infertility treatment were registered in a computerised database. All patients but one delivered in our institutions. All the relevant data on their pregnancy outcomes were obtained from the departments' databases. This includes week of delivery, indications and mode of delivery, as well as the newborns' weights and their outcome.
|
In the current study, the indications of delivery were divided into three categories: elective (at 36 completed weeks), spontaneous, and by intervention in which delivery was induced because of an underlying obstetric complication(s). The entire study cohort was divided into two groups, one comprised the women whose labour was induced or who underwent a Caesarean section (group 1, n = 24) and those who had spontaneous onset of labour (group 2, n = 21).
Cervical assessment
After the sonographic procedure was thoroughly explained, and consent had been obtained from the parturient women, TVS scanning was carried out. The subject lay in a dorsal lithotomy position with an empty bladder. A transvaginal probe (Aloka 1600, Tokyo, Japan and HDI 3000; Advanced Technology Laboratories, Seattle, WA, USA) with a dynamic focus of 5–7 MHz was placed in the anterior fornix of the vagina. We aimed to obtain a sagittal view of the cervix and of the endocervical mucosa marking the cervical canal. Whenever the cervical canal appeared curvilinear, the cervical length was the sum of two linear segments. The distance between the triangular area of echogenicity at the caudal tip close to the posterior wall of the upper vagina (the external os) and the end of endocervical mucosa (the internal os) was measured. Callipers were used to measure the distance between the triangular area of echodensity at the external os and the `V'- or `U'-shaped notch at the internal os (Andersen et al., 1990). It was important for the entire cervical canal to be visualized and that the two cervical lips were similarly widened. Both the external and internal ostium were identified at the two sides of the echogenic line. Care was taken to avoid pressure on the cervix. Only the closed portion of the cervical canal was measured, similar to our previous reported images (Maymon et al., 1996). The shortest of three repeated measurements was recorded. All the measurements were performed by experienced sonographers, and were printed on a hard copy. A measurement was considered inadequate if the above sonographic criteria could not be identified, whereupon re-examination was requested.
Statistical analysis
Data are presented as mean ± SD. Demographic parameters, cervical length measurement and week of delivery were compared between groups 1 and 2 using one way ANOVA test, followed by Tukey's method for multiple comparison. Student's t-test was applied to compare the cervical length measurement at 26 weeks gestation with other subject characteristics. Receiver operating characteristic curve analysis was used to determine the best sonographic cervical length measurement at ~26 weeks gestation for predicting delivery >33 gestational weeks. The 2 test was used to test the relation between the cervical length and the gestational week of delivery for the last group of patients. A linear regression model was applied to predict the time of delivery (dependent variable), using the cervical length and the gestational week in which this test was performed. A probability value of < 0.05 was considered significant. Statistical analysis was performed in the statistical department of the Tel Aviv University, using SPSS computer software.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
Forty-five pregnant women carrying triplets and for whom all the relevant information was available were included in this study. None of them was lost to follow up. Thirty-eight (84%), 6 (13%), and 1 (2%) conceived following IVF, ovarian stimulation and spontaneously respectively. The mean maternal age was 29 ± 4 years and 39 (67%) parturient women were primiparous. Twenty-one (47%) women had spontaneous onset of labour, 13 (29%) had elective Caesarean sections at 36 weeks gestation and 11 (24%) had induced delivery because of an obstetric complication(s). A total of 40 women (90%) delivered by Caesarean section. The mean gestational age at delivery for the entire study cohort was 32 gestational weeks. The mean ± SD newborn weight was 1600 ± 500 g. There were three cases of third trimester intrauterine fetal death, as well as three early and two late neonatal deaths, one from a sacrococcygeal teratoma which had been diagnosed prenatally.
The median cervical length was found to decrease with advancing gestation. This applied to both study groups (Table I). There was no case of severe maternal morbidity. There were no significant differences in the analysed demographic parameters between the two groups, including maternal age, parity, mode of conception and mode of delivery. They all were similar to the mean ± SD of the entire group, which was previously reported. However, group 2 of spontaneous deliveries tend to deliver about two weeks earlier (Table I) and therefore their infant weight was smaller 1300 g ± 500 versus 1900 g ± 300 from group 1 (one way ANOVA test; P = 0.001).
|
The receiver operating characteristic curve analysis was applied to determine the best sonographic cervical length for predicting delivery after 33 completed gestational weeks. This threshold was adapted from a previous study (Ramin et al., 1999
) in which triplet gestation was divided between early and late deliveries at a threshold of 33 gestational weeks. Thus, a cervical length of 25 mm was found to be the most accurate parameter with 94% sensitivity and 45% specificity for predicting premature delivery. Those results were similar both for the study population as a whole as well separately for groups 1 and 2 (Figure 2). From all the analysed subject characteristics, only a cervical length measurement at 26 weeks gestation was found to correlate significantly with delivery 33 weeks gestation. Thus, 25 women with a mean cervical length of 27 mm recorded at the first TVS measurement delivered up to 33 gestational weeks, whereas 20 parturient women with a mean cervical length of 33 mm delivered beyond 33 weeks gestation (Student's t test; P = 0.027).
|
In Table II
we present the correlation of cervical length at week 26 of gestation versus the week of delivery among 34 cases which had either spontaneous onset of labour or elective Caesarean delivery. As seen, 90% of the women which had cervical length 25 mm delivered 33 completed weeks, whereas 70% of the cases which had cervical length >25 mm delivered up to >33 gestational weeks. In this study group it was found that 11 of 34 cases (32%) or one in three triplet pregnancies has a short (<25 mm) cervix at 26 weeks gestation. This could be described as `mid gestation TVS test positive' results.
|
A linear regression model showed a significant correlation (R2 = 0.22; P = 0.01) between cervical length measurements (CX1) at ~26 weeks gestation (predictor) and week of delivery (dependent variable):
 |
Based on a linear regression curve (Figure 3) we hereby present a mathematical equation for expressing the relation between the cervical length (cm) and the gap time (weeks) to delivery for cases with spontaneous onset of labour:
 |
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
The current study demonstrates that cervical length measurement by TVS can predict premature delivery in triplet gestation. This is in agreement with other studies on singletons (Andersen et al., 1990
; Iams et al., 1996; Heath et al., 2000), twins (Michaels et al., 1991; Kushnir et al., 1995; Imseis et al., 1997; Souka et al., 1999; Yang et al., 2000) and triplets (Ramin et al., 1999; Bianco et al., 2000; Guzman et al., 2000a,b). The current study found that the mean cervical length decreased with advancing gestation, from 26 mm at ~26 weeks to 17 mm at ~32 weeks, in the group of patients whose onset of labour was spontaneous. Some authors (Roberts and Morrison, 1998) have questioned `whether the cervical length measurement of <25 mm should be modified for the multifetal gestations and what would be the test efficiency if such adjustments were performed'. Our study addresses exactly those questions, and proposes a mathematical equation for predicting the interval time between TVS examination and delivery. This dependent variable can be calculated based on the two predictors: cervical length measurement and the relevant week in which TVS was performed. In agreement with other studies on singletons (Iams et al., 1996), twins (Souka et al., 1999) and triplets (Bianco et al., 2000), the current study found that cervical length at a threshold value of 25 mm at around 26 weeks gestation is a sensitive sonographic marker for predicting preterm delivery of triplets. Similar data were reported by Ramin et al. (1999) who found a cervical length of 26 ± 10 mm at 25 weeks gestation and 21 ± 7 mm at 30 weeks and by Yang et al. (2000) who reported a cervical length of 30 mm at 18–26 weeks. In singletons, a cervical length <25 mm had a maximum test efficiency corresponding to the 10th centile in cervical length when TVS was performed at 24 weeks, showing a sensitivity of 37% and a specificity of 92% for predicting preterm delivery (Iams et al., 1996). In twins at 28 weeks gestation a cervical length of 25 mm had a 100% sensitivity for predicting preterm delivery (Souka et al., 1999). However, less promising results were reported for twins assessed at 18–26 weeks gestation in which a cervical length of 30 mm was found. This was associated with 53% sensitivity and 62% specificity for predicting premature delivery (Yang et al., 2000). As for triplet pregnancies in which a cervical length of 27 mm was measured at an interval of 26–31 weeks: TVS could predict delivery <34 weeks with 30% specificity and 30% sensitivity (Bianco et al., 2000). Among our patients, 90% of the cases in which the cervical length measured 25 mm at approximately week 26 of gestation (`test positive') delivered prematurely. Thus, a single TVS examination at 26 weeks gestation appears adequate to screen those patients who are at risk for preterm delivery in otherwise uncomplicated triplet pregnancies. At this stage, cervical lengths of 25 mm require further clinical evaluation and, in agreement with a recent study on twins, (Yang et al., 2000) no additional information emerged from repeating the examination if the cervical length at this gestational age was 
33 mm. The fact that one in three triplet pregnancies has a short cervix (<25 mm) at 26 weeks and that 90% of them delivered preterm is interesting. However, one potential limitation of cervical assessment in multiple pregnancies compared to singletons is the larger number of cases presented with short cervix at mid-gestation. When analysing those `test positive' results and knowing that this kind of information will always increase the sensitivity, complementary tests contributing to the specificity for detecting premature delivery among triplets are still needed.
By defining the spontaneous preterm delivery rate at 33 weeks, it was reported that 1.2% of singletons (Kiely, 1998), 8–11% of twins (Kiely, 1998; Souka et al., 1999) and 31% (Kiely, 1998) of triplets delivered earlier.
Similar to the study design by Ramin et al. (1999), we compared the cervical length in two groups of women, i.e., those who delivered before and those who delivered after 33 weeks gestation. For patients who delivered prematurely, our findings agree with their findings indicating a relatively short cervical length throughout the third trimester of gestation. However, for the group of patients who delivered at >33 weeks gestation, we found less consistency between the results and consider that further studies using a uniform scanning technique are required.
We propose that the present study might aid the perinatologist in decision making when treating pregnant women with triplet pregnancies. Decision involving various intervention modalities (e.g. when to start tocolytic drugs or steroids for enhancing lung maturation and according to which protocol), or when to recommend maternal transfer to a tertiary referral centre with facilities for premature babies could be supported by TVS assessment of the cervical length.
Although some authors have argued that `it is doubtful that ultrasonographic measurement of cervical dilatation will replace the time-honoured digital examination' (Bowes, 1995), this sonographic technique offers additional valuable information (Maymon et al., 1996) both for physicians and their patients who are involved with assisted reproductive techniques. Since TVS for cervical length measurement is a simple and reproducible technique, with an inter-observer variability rate of only 5–10% (Kushnir et al., 1995), we believe that it should be more commonly integrated into antenatal care. This is especially true when the patient is found to be in a high-risk category for preterm delivery, as in triplet pregnancy.
|
Acknowledgements |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
We thank Esther Eshkol for editorial assistance.
|
Notes |
|---|
3 To whom correspondence should be addressed. E-mail: intposgr{at}post.tau.ac.il
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
Andersen, H.F., Nugent, C.E., Wanty, S.D. et al. (1990) Prediction of risk for preterm delivery by ultrasonographic measurement of cervical length. Am. J. Obstet. Gynecol., 163, 859–867.[Web of Science][Medline]
Bianco, A., Bhandarkar, A., Kuczynski, E. et al. (2000) Cervical length and the risk of preterm delivery among multiple gestation (SMFM Abstract). Am. J. Obstet. Gynecol., 182, S121.
Bowes, W.A.J.R. (1995) Editorial. Obstet. Gynecol. Surv., 50, 18–19.
Cefalo, R.C. (1998) Editorial. Obstet. Gynecol. Surv., 53, 674–675.
Garite, T.J., Bentley, D.L., Hamer, C.A. et al. (1990) Uterine activity characteristics in multiple gestations. Obstet. Gynecol., 76, 56S-59S.
Guzman, E.R., Walters, C., Meirowitz, N.B. et al. (2000a) A comparison of sonographic cervical parameters in predicting spontaneous preterm birth in triplet gestations (SMFM Abstract) Am. J. Obstet. Gynecol., 182, S49.
Guzman, E.R., Ananth, C.V., Meirowitz, N.B. et al. (2000b) A comparison of sonographic cervical length between 15 and 25 weeks gestations in singleton, twin and triplet (SMFM Abstract) Am. J. Obstet. Gynecol., 182, S114.
Heath, V.C.F., Southall, T.R., Souka A.P. et al. (2000) Cervical length at 23 weeks of gestation: prediction of spontaneous preterm delivery. Ultrasound Obstet. Gynecol., 12, 312–317.
Iams, J.D., Goldberg, R.L., Meis, P.J. et al. (1996) The length of the cervix and the risk of spontaneous premature delivery. N. Engl. J. Med., 334, 567–572.
Imseis, H.M., Albert, T.A. and Iams, J.D. (1997) Identifying twin gestations at low risk for preterm birth with transvaginal ultrasonographic cervical measurement at 24 to 26 weeks' gestation. Am. J. Obstet. Gynecol., 177, 1149–1155.[Medline]
Keith, L.G., Papiernik, E. and Luke, B. (1991) The costs of multiple pregnancy. Int. J. Obstet. Gynaecol., 36, 109–114.
Kiely, J.L. (1998) What is the population-based risk of preterm birth among twins and other multiples? Clin. Obstet. Gynecol., 41, 3–11.[Web of Science][Medline]
Kushnir, O., Izquierdo, L.A., Smith, J.F. et al. (1995) Transvaginal sonographic measurement of cervical length. Evaluation of twin pregnancies. J. Reprod. Med., 40, 380–382.[Medline]
Luke, B. (1994) The changing patterns of multiple births in the United States: maternal and infant characteristics 1973 and 1990. Obstet. Gynecol., 84, 101–106.[Web of Science][Medline]
Luke, B. (1996) Reducing fetal deaths in multiple births: optimal birthweights and gestational ages for infants of twin and triplet births. Acta Genet. Med. Gemellol. Roma, 45, 333–348.[Medline]
Maymon, R., Shulman, A., Ariely, S. et al. (1996) Sonographic assessment of cervical changes during pregnancy and delivery: current concepts. Eur. J. Obstet. Gynecol. Reprod. Biol., 67, 149–155.[Medline]
Michaels, W.H., Schreiber, F.R., Padgett, R.J. et al. (1991) Ultrasound surveillance of the cervix in twin gestations: management of cervical incompetency. Obstet. Gynecol., 78, 739–744.[Medline]
Peaceman, A.M., Dooley, S.L., Tamuru, R.K. et al. (1992) Antepartum management of triplet gestation. Am. J. Obstet. Gynecol., 167, 1117–1120.[Medline]
Ramin, K.D., Ogburn, P.L. Jr., Mulholland, T.A. et al. (1999) Ultrasonographic assessment of cervical length in triplet pregnancies. Am. J. Obstet. Gynecol., 180, 1442–1445.[Medline]
Roberts, W.E., Morrison, J.C. (1998) Has the use of home monitors, fetal fibronectin, and measurement of cervical length helped predict labor and/or prevent preterm delivery in twins? Clinical Ob/Gynecol., 41, 95–102.
Robinson, H.P., Fleming, J.E.E. (1975) A critical evaluation of sonar crown-rump length measurements. Br. J. Obstet. Gynecol., 82, 702–710.[Web of Science][Medline]
Sebire, N.J., D'Ercole, C., Hughes, K. et al. (1997) Increased nuchal translucency thickness at 10–14 weeks of gestation as a prediction of severe twin to twin transfusion syndrome. Ultrasound Obstet. Gynecol., 10, 86–89.[Web of Science][Medline]
Seoud, M.A., Toner, J.P., Kruithoff, C. et al. (1992) Outcome of twin, triplet, and quadruplet in vitro fertilization pregnancies: the Norfolk experience. Fertil. Steril., 57, 825–834.[Web of Science][Medline]
Souka, A.P., Heath, V., Flint, S. et al. (1999) Cervical length at 23 weeks in twins in predicting spontaneous preterm delivery. Obstet. Gynecol., 94, 450–454.[Medline]
Yang, J.H., Kuhlman, K., Daly, S. et al. (2000) Prediction of preterm birth by second trimester cervical sonography in twin pregnancies. Ultrasound Obstet. Gynecol., 15, 288–291.[Medline]
Submitted on August 4, 2000; accepted on February 12, 2001.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  S. Maslovitz, J. Hartoov, I. Wolman, A. Jaffa, J. B. Lessing, and G. Fait Cervical Length in the Early Second Trimester for Detection of Triplet Pregnancies at Risk for Preterm Birth J. Ultrasound Med., September 1, 2004; 23(9): 1187 - 1191. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Prapas, Y. Prapas, Y. Panagiotidis, S. Prapa, P. Vanderzwalmen, and G. Makedos Cervical dilatation has a positive impact on the outcome of IVF in randomly assigned cases having two previous difficult embryo transfers Hum. Reprod., August 1, 2004; 19(8): 1791 - 1795. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||