ABSTRACT

Conclusion:

In our population the 5^th precentile value of cervical length which is 30 mm in singletons and 10^th percentile cervical length which is 31 mm in twins can be used to follow-up and treat pregnant women at risk for preterm delivers.

Results:

A total of 4621 consecutive asymptomatic pregnant women admitting for advanced obstetric ultrasound screening were evaluated. Of these 4340 (93.9%) were second trimester singleton pregnancies and 281 (6.1%) were twin pregnancies and were included. Mean cervical length measurements of singleton and twin pregnancies were 38.2±6.5 mm and 37.6±7.2 mm respectively (p=0.17). Overall, the 5^th percentile of cervical length measurement after analysing singleton and twin pregnancies together was 29.4 mm at 16 weeks, 30 mm at 17 weeks, 30 mm at 18 weeks, 30 mm at 19 weeks, 30 mm at 20 weeks, 30 mm at 21 weeks, 30 mm at 22 weeks, 31 mm at 23 weeks, 29 mm at 24 weeks.

Material and Methods:

This study was conducted by retrospective analysis of mid-trimester transvaginal cervical measurements of women with singleton and twin pregnancies that were examined by a single perinatologist in a single center.

Objective:

The aim of this study was to determine the standard mid-trimester cervical lengths of singleton and twin pregnancies.

Introduction

In the last few decades, developments in obstetric care has led to a considerable decrease in both maternal and perinatal mortality and morbidity. Among these obstetric complications, preterm deliveries create an increased level of burden on the individual pregnant woman, on family and on even on society as a whole, by causing long-term consequences, such as growth retardation, mental retardation, chronic diseases and cognitive impairments (1,2,3). Nevertheless, preterm delivery constitutes between 7-11% of all deliveries and so remains one of the leading causes of neonatal morbidity and mortality in developed countries (4,5).

Short cervix is defined as a cervical measurement lesser than 25 mm before 24^th week of pregnancy (2^nd-3^rd percentile) (6,7). A short cervix diagnosed before the 24^th gestational week could predict preterm deliveries in later weeks of pregnancy. Risk of preterm delivery significantly increases in presence of a short cervix in women with a history of a previous preterm delivery (8,9,10,11). Therapeutic interventions, such as cervical cerclage placement, or medical measures, such as progesterone administration, initiated at early phases of pregnancy have been demonstrated to prevent preterm delivery in 30% to 40% of high risk patients (6,7,12). Globally, various values of cervical lengths have been proposed as a threshold level to initiate these therapeutic or prophylactic interventions throughout. Different clinical guidelines suggest cervical length shorter than 15 mm (0.5 percentile) or shorter than 20 mm (1^st percentile) as threshold levels. Nonetheless, the American College of Gynecology recommends initiation of therapeutic or prophylactic interventions in case of a cervical length measurement shorter than 20 mm women without a history of preterm birth. Moreover, initiation of therapy was suggested for patients with a cervical length shorter than 25 mm in women with a previous history of preterm delivery before 34^th week of pregnancy (13). These variations in optimal cut-off values of cervical lengths in predicting preterm delivery were primarily dependant on methodological discrepancies and different population of patients in the various studies. The level of evidence about models involving prediction of preterm delivery depending on cervical length in twin pregnancies is low. Therefore various modalities of management are applied among clinicians. Cervical length is routinely evaluated trans-abdominally in every pregnant women as a part of detailed fetal anomaly screening sonography that is usually performed between the 18^th-22^nd gestational week. Inadequate imaging of cervix or a short cervical measurement in this examination warrants transvaginal cervical length measurement that would be repeated once for every two weeks between 16^th and 24^th weeks of pregnancy (14).

The aim of this study was to evaluate the cervical length of a high number of asymptomatic, consecutive, pregnant women between 16 and 24 weeks of gestation and produce a histogram and percentile charts of cervical length in this patient population.

Material and Methods

In this study, trans-vaginal cervical measurement of women, consecutively admitted to a single clinic, were retrospectively evaluated. A single perinatologist (EC) carried out all of the measurements between 2016 and 2021 using transvaginal ultrasound (Voluson E8 4Mhz probe, GE company). Cervical measurements were performed in the lithotomy position following the emptying of the maternal bladder by urinary catheterization to standardize all patients. The cervix was visualized in sagittal axis and endocervical length, appearing as a weak linear echodensity between the internal and external cervical ostia, was measured. All of the measurements were performed without applying excess pressure to cervix and a mean of three measurements obtained from a single patient was recorded as cervical length.

Independent risk factors such as age, body mass index, ethnicity, parity, conception via assisted reproduction, history of previous surgeries, chronic diseases, and secondary obstetrical complications were not taken into account. Mean ± standard deviation (SD), median (range) and percentile values of cervical lengths were reported, by gestational age in weeks.

This study was following approval by Alanya Alaaddin Keykubat University Faculty of Medicine Clinical Research Ethics Committee (approval number: 2022/12, date: 25.05.2022). Informed consents were obtained from all participants.

Results

Overall 4,647 patient records were evaluted. Twenty-one triplet pregnancies and five women with a history of cervical conization or loop electrosurgical excision procedure were excluded from the study. Thus, 4340 singleton pregnancies and 281 twin pregnancies that were eligible were included in the study. The demographic data of the study population is given in Table 1. Women with twin pregnancies were significantly younger, more frequently nulliparous and had in vitro fertilization treatment for the present pregnancy compared with the singleton pregnancy group.

Table 1

Overall cervical length measurements of singleton pregnancies were 38.3±6.5 (0-67) mm and 37.6±7.3 (9-59) mm respectively (p=0.17). Comparison of cervical length measurement between singleton and twin pregnancies across 16-24 weeks are presented in Table 2. The percentiles of cervical length in singleton pregnancies is given in Table 3. The 5^th percentile of cervical length was between 29-31 mm throughout theperiod 16-24 weeks of gestation. The percentiles of cervical length in twin pregnancies is given in Table 4. The 5^th percentile of cervical length was 27-30 mm between 16-18 weeks and 21-26 mm between 18-24 weeks. Chart analysis and and histogram of cervical length in singleton and twin pregnancies are given in Figure 1. Overall cervical length percentiles, including singleton and twin pregnancies together, is given in Figure 2. The 5^th percentile was 29.4 mm at 16 weeks, 30 mm at 17 weeks, 30 mm at 18 weeks, 30 mm at 19 weeks, 30 mm at 20 weeks, 30 mm at 21 weeks, 30 mm at 22 weeks, 31 mm at 23 weeks, 29 mm at 24 weeks. When all 4340 singleton pregnancies in the period 16-24 gestational weeks was considered, the 2.5^th percentile was 28 mm, the 5^th percentile was 30 mm and the 10^th percentile was 31 mm. When 281 twin pregnancies during the same period were evaluated, the 2.5^th percentile was 24 mm, the 5^th percentile was 26 mm and the 10^th percentile was 31 mm. Distribution of cervical lengths in singleton and twin pregnancies are shown in Figure 3. Incidence of short cervix increased as the gestation progressed.

Table 2

Table 3

Table 4

Figure 1

Figure 2

Figure 3

Discussion

Contemporary predictive models for preterm delivery includes a comparison of each patients cervical measurement with standard normograms. Reliability of these models partially depends on accurate determination of cervical length percentile and, as normal values for cervical length could be expected to vary in different populations, in this study we aimed to evaluate the normal ranges of cervical lengths in a Turkish population and establish a standard cervical length normogram for this specific population.

To avoid the limitations of a trans-abdominal approach in cervical length measurement, such as difficulty in cervical visualization and lengthening of cervix due to a filled bladder, in this study we preferred to assess records of patients that were evaluated solely by a trans-vaginal approach. Although cervical measurements could be performed from the 14^th week of pregnancy, measurements made between 16^th and 18^th gestational weeks at the time that the cervix separates from the lower uterine segment are considered more consistent and accurate (15). In normal conditions cervical length remains stable between the 14^th and 28^th gestational weeks and thus changes in cervical length in this period of pregnancy are described with a bell-curve.

The published definitions of short cervix are somewhat obscure. A wide variety of cervical lengths, from 10-35 mm, were studied as empirical cut-off values for prediction of spontaneous preterm birth in singleton pregnancies (16). The same cut-off values were empirically used for twin pregnancies, while a limited number of studies have reported the value of cervical shortening as an important predictor of spontaneous preterm delivery in mathematical models.

Normograms prepared by using specific populations may demonstrate variations and might have a negative impact on prediction of preterm delivery. For instance, in a study conducted in North America, mean cervical length of women in the 24^th gestational week was estimated to be 34.0±7.8 mm (17). In contrast, a similar study conducted in Iran demonstrated that mean cervical length was 38.3±5.61 mm and another conducted in Switzerland was found mean cervical length in the 24^th week to be 39.1±5.6 (18,19). Considering the discrepancies in these data, normograms prepared based on North American data could be expected to have higher false negativity in prediction of preterm delivery for a patient from Switzerland or Iran. In the present study, the mean cervical length for Turkish women in the 24^th gestational week was 37.9±6.5 mm.

Singleton normograms enable comparison of 5^th percentile values between the different populations. The 5^th percentile for cervical length between 16-24 weeks of gestation was found to be 27 mm in Chicago, 30 mm in an American Hispanic population, 30.6 mm in a Thai population, 28 to 29 mm in South Africa, 25 mm in Switzerland, 32-25 mm in Paris, 30 mm in Iran, and between 30-33 mm in Brasil (9,20,21,22,23,24,25).  In our large cohort, the overall 5^th percentile of cervical length was 30 mm between 16-24 weeks of gestation, including both singleton and twin pregnancies. Small variations in different populations result from the sample sizes and the weight of pregnant women between 16-19 weeks or 20-24 weeks. The cervical length tend to be a few milimeters shorter as the pregnancy approaches the 24^th week of gestation.

Data involving normal range of cervical length in twin pregnancies is scarce, in comparison to that for singleton pregnancies. A study conducted on 172 patients in Brazil found median cervical length of women with twin gestation to be 39 mm (37-40) (26). In the present study, the estimated median value of cervical length in twin pregnancies was 36 mm (24-50). Normogram data for cervical length in twin gestations are also rare. As preterm delivery rate of twins are higher, studies tend to take the 10^th percentile of cervical length as a cut-off for short cervix (27). The 10^th percentile of cervical length was reported to be 30 mm in Canada and 25 mm in Poland (27,28). The 5^th percentile of cervical length was 21-25 mm at 18-22 weeks in Brasil (29). In the present study in a Turkish population these values were 26 mm and 31 mm for cervical length at the 5^th and 10^th percentile in one of the largest twin cohorts so far reported.

Cervical length is dynamic throughout the pregnancy. Studies have shown that multiple variable can affect cervical length dynamics and affect preterm delivery (30). Both 25 mm and 30 mm cut-off values may be used in our population to test interventions for preventing preterm births in twin pregnancies.

Conclusion

The 5^th precentile value of cervical length was 30 mm in singletons, while the 10^th percentile cervical length was 31 mm in twin pregnacies in this Turkish population. These values can be used to follow-up and treat pregnant women at risk for preterm delivers.

Statistical analysis

Statistical analysis was performed using IBM SPSS, version 23.0 (SPSS Inc., Chicago, IL, USA). Descriptive statistics are expressed as mean ± SD for normally distributed data and as median (minimum-maximum) for non-normally distributed data. Categorical variables are expressed as numbers and percentages (%).

Study Limitations

The limitation of this study is that not all patients were followed-up until delivery, so the predictive value of the present data was not calculated. However, in two prior studies from our center an empirical 30 mm cervical length cut-off value was used to prevent early preterm delivery <32 weeks of gestation (31,32). In our population weekly follow-up of singleton pregnant women by cervical length, and emergency cerclage when indicated, prevented 62.5% of preterm deliveries earlier than 32 weeks (31). Vaginal progesterone treatment in singleton pregnant women with cervical length less than 30 mm and emergency or ultrasound-indicated cerclage was used when indicated postponed 98.7% of the deliveries beyond 34 weeks of gestation (32). In this context, comparing an individual patients’ cervical length with the population-appropriate normal values may assist in increasing the accuracy of preterm delivery prediction.