Low uterine segment thickness in the prediction of cesarean delivery after induction of labor

Introduction

Induction of labor (IOL) is a frequently used method in obstetrics to start labor in cases such as post-term pregnancies, maternal diabetes, preeclampsia, fetal growth restriction (FGR) and/or other maternal or fetal diseases that require prompt delivery. For this purpose, methods such as oxytocin, dinoprostone, misoprostol, or intracervical balloon catheter placement are usually used [1], 2]. Numerous studies have been conducted to determine whether IOL will be successful and result in vaginal delivery [3], [4], [5], [6], [7]. Many ultrasonographic parameters such as uterocervical angle, cervical length, and cervical elastography have been used to predict the success of IOL. Even though these parameters have been tried for this purpose, they have not been able to replace Bishop scores in daily practice [8], [9], [10], [11], [12]. However, despite its frequent use, the predictive ability of Bishop scores was found low [13], [14], [15].

The lower uterine segment (LUS) is the isthmic region of the uterus, which thins as pregnancy progresses [16]. It has been frequently used in previous studies to predict the risk of uterine rupture in vaginal birth after cesarean delivery(CD) [17], [18], [19]. There are several techniques to measure LUS thickness. Measuring the full thickness of the peritoneum, bladder wall and myometrium together (LUS) or only the myometrium (MLUS) are two well-defined techniques [20]. In 2023, Del Campo et al. showed that LUS thickness measurements before IOL were reproducible and significantly related in the prediction of IOL outcomes. Their findings supported the idea that pregnancies with a thinner LUS would be more likely to result in vaginal delivery [21]. However, in LUS measurement, the peritoneum, bladder wall and myometrium are measured as a whole, and there is limited data in the literature regarding the predictive value of myometrium measurement alone (MLUS) in vaginal delivery. To the best of our knowledge, the current study is the first to compare the predictive value of LUS and MLUS for successful IOL.

The aim of our study was to examine the relationship between LUS thickness and CD and compare the capacity of two measurement methods (LUS and MLUS) in the prediction of successful IOL.

Materials and methods

This was a prospective cross-sectional study of consecutive pregnancies undergoing IOL between 2022 and 2023 at our hospital. The study was approved by the ethics committee of our hospital and conducted in accordance with the Declaration of Helsinki. The inclusion criteria were women with singleton pregnancies with cephalic presentation whose Bishop scores were below six undergoing IOL after 37 weeks of gestation with vaginal application of dinoprostone. Patients with a history of CD, fetal anomalies, multiple gestations, and any contraindication to vaginal delivery were excluded from the study group. Descriptive data of the pregnant women such as age, weight, weight gain, height, gravidity, parity, conceiving method, gestational age at the time of IOL, indication of IOL, mode of delivery, and Bishop scores were recorded.

According to our institution protocol, if there is an indication for IOL but the cervix is unfavorable with a Bishop score of <6, a vaginal pessary containing 10 mg dinoprostone is placed in the posterior fornix. When cervical ripening occurs, we remove the dinoprostone and administer oxytocin if needed. Dinoprostone is maintained for a maximum of 24 h. If the Bishop remains unfavorable, another dinoprostone pessary is inserted. The dinoprostone is removed if uterine tachysystole (>5 uterine contractions in 10 min), or non-reassuring fetal heart rate occur. Fetal surveillance and uterine activity were monitored through external continuous electronic fetal heart rate monitoring and tocodynamometry. Failed induction was defined as the failure of the second dose of dinoprostone with a Bishop score of <6 or active phase arrest or prolonged second stage not suitable for instrumental delivery. Balloon dilation of the cervix is not routinely used in our hospital; dinoprostone is the preferred route of IOL. Successful IOL was defined as vaginal delivery.

In this study, after the decision for IOL was taken, patients were informed about the study and consent was obtained. Participants were asked to empty their bladder and bowels, and they were placed in the dorsal-lithotomy position. An ultrasound probe was covered with a sterile lubricant condom. A transvaginal probe was inserted into the vagina without excessive compression and when the area of the cervix was identified, the image was magnified. LUS was measured as the full thickness of the peritoneum, bladder wall, and myometrium together (LUS) and as only the myometrium (MLUS) (Figure 1). The calipers were placed at the thinnest part of the LUS and only one measurement was taken. Each patient contributed one measurement to the study.

Figure 1:

Ultrasonographic measurement of low uterine segment (LUS) thickness and myometrial LUS (MLUS) thickness a transvaginal probe was inserted into the vagina without excessive compression and when the area of the cervix was identified, the image was magnified. LUS was measured as the full thickness of the peritoneum, bladder wall, and myometrium together (LUS) (A) and as only the myometrium (MLUS) (B).

All the ultrasound scans were performed by an experienced operator and a trained resident using a Samsung RS85 ultrasound system (Samsung, South Korea) with a transvaginal probe. The sonographers who performed measurements were not in charge in the labor and delivery unit; the obstetricians in charge were blinded to measurements.

Time interval until delivery, time interval to birth after IOL, and indication if delivery occurred by CD were recorded. The study group was divided into two groups to compare demographic and clinical characteristics: pregnancies delivered by CD and pregnancies delivered vaginally. To examine the effectiveness of LUS measurements, the study group was also divided into different groups in terms of patient age (<25 years vs. >25 years), height (<160 cm vs. >160 cm), and body mass index (BMI) (<30 vs. >30 kg/m²), and nulliparity (nulliparous vs. parous).

Results

A total of 140 patients met the eligibility criteria. Of these, 16 were excluded because of missing data. The demographic and obstetric characteristics of the study group are summarized in Table 1. When we examined the ultrasound parameters, the mean cervical length was 28.6 ± 8.1 mm, the mean LUS was 7.47 ± 1.98 SD, the mean MLUS was 4.29 ± 1.35, and funneling was present in 20 (16.1 %) patients.

Table 2 represents a comparison of clinical and demographic characteristics of women with and without CD. LUS (10.13 ± 1.98 mm vs. 7.07 ± 1.66 mm, p<0.001), MLUS (5.63 ± 1.48 mm vs. 4.10 ± 1.22 mm, p<0.001), and cervical length (33.4 ± 9.3 vs. 27.9 ± 7.7, p=0.011) in the CD group were significantly higher than in the non-CD group. Height in the CD group was significantly shorter than in the non-CD group (157.4 ± 6.4 cm vs. 162.5 ± 6.1 cm, p=0.006). Age, weight, weight gain, gravidity, parity, nulliparity, gestational age at IOL, presence of funneling, time interval to the onset of active labor, time interval to the birth, Bishop scores, spontaneous conception rate, and newborn weights did not differ between the groups with and without CD.

The results of the logistic regression analyses are summarized in Table 3. In the univariate model, a significant effect of height (OR:0.87, 95 % CI: 0.79–0.95), LUS (OR:2.76, 95 % CI: 1.72–4.45), MLUS (OR:2.17, 95 % CI: 1.45–3.24), and cervical length (OR:1.10, 95 % CI: 1.02–1.18) was observed in the prediction of CD after IOL. In the multivariate model, significant independent effects of height (OR:0.88, 95 % CI: 0.79–0.98) and LUS (OR:2.71, 95 % CI: 1.66–4.43) values were observed in the prediction of CD.

ROC analysis was performed to evaluate optimal cut-off values of LUS and MLUS to predict CD. Both LUS (AUC 0.882, p<0.001) and MLUS (AUC 0.756, p<0.001) significantly predicted CD with an optimal cut-off value of 9.8 mm (75.0 % sensitivity, 92.6 % specificity) for LUS, and 5.1 mm (68.8 % sensitivity, 82.4 % specificity) for MLUS (Figure 2).

Figure 2:

ROC analysis of LUS thickness and MLUS thickness to predict the duration of CD. ROC analysis was performed to evaluate optimal cut-off values of LUS and MLUS to predict CD. Both LUS (AUC 0.882, p<0.001) and MLUS (AUC 0.756, p<0.001) significantly predicted CD with an optimal cut-off value of 9.8 mm (75.0 % sensitivity, 92.6 % specificity) for LUS, and 5.1 mm (68.8 % sensitivity, 82.4 % specificity) for MLUS. AUC, area under the curve; CD, caesarean delivery; LUS, low uterine segment; PC, predictive value; ROC, receiver operating characteristics.

Because LUS was evaluated as an independent variable in logistic regression analysis, its ability to predict the probability of CD was also studied. Half of the pregnancies resulted in CD with an LUS value over 10 mm and 80 % when it was over 12 mm (Figure 3).

Figure 3:

Probability of CS at different LUS thickness values ​​after IOL half of the pregnancies resulted in CS with an LUS value over 10 mm and 80 % when it was over 12 mm. CS, caesarean section; IOL, induction of labor; LUS, low uterine segment.

We also assessed the predictive value of LUS measurements across different patient subgroups. We found that LUS values were significantly effective in predicting CD in patients aged under and over 25 years. It was also effective in patients with a BMI under and over 30 kg/m². However, LUS measurements were more effective in patients with a height≤160 cm in the prediction of CD compared with patients taller than 160 cm (p<0.05). LUS measurements were also found to be more effective in parous women than in nulliparous women (p<0.05) (Figure 4).

Figure 4:

Effectiveness of LUS thickness value in different subgroupsLUS values were significantly effective in predicting CS in patients aged under and over 25 years. It was also effective in patients with a BMI under and over 30 kg/m². However, LUS measurements were more effective in patients with a height≤160 cm in the prediction of cesarean delivery compared with patients taller than 160 cm (p<0.05). LUS measurements were also found to be more effective in parous women than in nulliparous women (p<0.05). BMI, body mass index; CS, caesarean section; IOL, induction of labor; LUS, low uterine segment.

Discussion

Our study showed that both LUS and MLUS thickness before IOL were significantly associated with prediction of CD. We provide cut-off points for LUS and MLUS that are useful in clinical practice to predict if successful IOL will occur. However, when we compared the prediction accuracy of these two measurements, we found that LUS had a better predictive ability than MLUS. LUS thickness was also successful in predicting CD in all age groups and all BMI groups, but was better in the prediction of CD in women shorter than 160 cm and multiparous women.

Conclusions

Ultrasonographic evaluation of LUS before IOL is a promising ultrasound parameter that can be easily applied to predict CD. Although LUS and MLUS thickness measurements can be used to predict CD, LUS thickness has better sensitivity and specificity compared to MLUS thickness. Assessment of LUS prior to IOL may provide valuable information regarding the risk of CD for both practitioners and pregnant women. Prospective studies with larger numbers of participants are needed to determine the cut-off values ​​of LUS thickness values ​​specific to gestational weeks.