The predictive role of serial transperineal sonography during the first stage of labor for cesarean section

Introduction

Labor dystosia, also defined as prolonged labor, is characterized by slow progression of labor [1]. It is tightly related to adverse perinatal outcomes such as instrumental delivery, fetal distress, increased cesarean section rates, postpartum bleeding, infection and mortality [2]. Digital vaginal examination is the gold standard way to evaluate labor progress. Fetal head descent, fetal head station, cervical effacement and dilatation could be assessed by digital examination. Unfortunately, it is a subjective and uncomfortable methods for patients and it increases the risk of ascending infection especially in premature rupture of membrane cases [3]. In a study of Rizzo et al. ultrasound examination was found to be a better tolerated method than vaginal examination for labor progress [4]. Considering this, intrapartum transperineal sonography has been investigated for the evaluation of labor progress. Although the use of sonography is not a standard practice, it could provide unbiased data about the progression of labor.

The abnormality in fetal head descent results in cesarean delivery. Thus, head to perineum distance (HPD) has claimed to be a marker for delivery mode and labor progress [5]. Head to perineum distance is defined as the distance between fetal skull and the perineum [6]. Another transperineal sonography marker is angle of progression (AOP). It is the angle from the maternal symphysis pubis to the leading part of the skull [1], 7]. In the literature, there has been studies searching the role of AOP and HPD in labor progress. In these studies, HPD and AOP has found to be correlated with fetal station and cervical dilatation findings in vaginal examination [8], [9], [10], [11]. These studies have also emphasized high interobserver correlation for HPD and AOP. However, they generally focused on single measurement, second stage of labor, and patient groups not classified as nulliparous and multiparous. In the present study, we aimed to evaluate the role of serially measured AOP and HPD in labor progress, assess the correlation of these markers with digital examination findings and determine these role both in nulliparous and multiparous patients. Also, we firstly compared the predictive role of HPD and AOP for labor progress.

Materials and methods

This was a prospective longitudinal study conducted at a university affiliated research and training hospital between November 2019 and January 2020. Institutional Ethical and Research Committee approved the present study with a decision number of 2011-KAEK-252019/10-16. The study was in accordance with the declaration of Helsinki. Written informed consent was obtained from all study participants.

Results

The sociodemographic and perinatal characteristics of patients were presented in Table 1. There was no significant difference between vaginal delivery and cesarean groups in terms of age, BMI, median gestational week, first and fifth minutes Apgar scores. The number of parity, the ratio of multiparous patients, BISHOP scores and duration of active phase were higher in vaginal delivery group. Estimated fetal weight, birth weight and head circumference were statistically significantly lower in vaginal delivery group.

The serial measurements of HPD and AOP were presented in Table 2. Significant differences were found for each mesurement of HPD and AOP between vaginal delivery and cesarean groups. The chance in fetal head station, cervical dilatation, HPD and AOP were demonstrated in Figure 1. Delta values of head to perineum distance and angle of progression with the respect of first examination were presented in Table 3. There was significant difference between all measurements except ΔHPD1-2.

Figure 1:

The chance in fetal head station, cervical dilatation, head to perineum distance and angle of progression. ***AOP, angle of progression; CS, cesarean section; HPD, head to perineum distance; NVD, vaginal delivery.

In the correlation analysis using Spearman correlation coefficient, HPD was found to be inversely correlated with AOP (r=−0.839, p<0.001), cervical dilatation (r=−0.761, p<0.001) and head station (r=−0.761, p<0.001). Additionally, AOP was positively correlated with cervical dilatation (r=0.759, p<0.001) and head station (r=0.756, p<0.001) while it was negatively correlated with HPD (r=−0.839, p<0.001).

The predictive role of sonographic AOP and HPD for cesarean delivery in all patients was evaluated with receiver operating curve analysis and presented in Figure 2. Head perineum distance >38 mm was found to predict cesarean section with 86.9 % sensitivity and 59.2 % specificity (AUC=0.782, p<0.001). Angle of progression ≤117° was found to predict cesarean section with 93.5 % sensitivity and 65.1 % specificity (AUC=0.877, p<0.001). The predictive role of HPD and AOP in all patients was compared and AOP had superior predictive role than HPD (p<0.001).

Figure 2:

Receiver operating curve analysis evaluating the predictive role of angle of progression and head to perineum distance for cesarean delivery in all patients. AOP, angle of progression; HPD, head to perineum distance.

The predictive role of sonographic AOP and HPD for cesarean delivery in nulliparous patients was evaluated with receiver operating curve analysis and presented in Figure 3. Head perineum distance >37 mm was found to predict cesarean section with 84.2 % sensitivity and 55.2 % specificity (AUC=0.763, p<0.001). Angle of progression ≤110° was found to predict cesarean section with 82.7 % sensitivity and 73.3 % specificity (AUC=0.862, p<0.001). The predictive role of HPD and AOP in nulliparous patients was compared and AOP had superior predictive role than HPD (p<0.001).

Figure 3:

Receiver operating curve analysis evaluating the predictive role of angle of progression and head to perineum distance for cesarean delivery in nulliparous patients. AOP, angle of progression; HPD, head to perineum distance.

The predictive role of sonographic AOP and HPD for cesarean delivery in multiparous patients was evaluated with receiver operating curve analysis and presented in Figure 4. Head perineum distance >39 mm was found to predict cesarean section with 95.5 % sensitivity and 62.5 % specificity (AUC=0.824, p<0.001) in multiparous patients while AOP ≤109° was found to predict cesarean section with 87.9 % sensitivity and 78.2 % specificity (AUC=0.909, p<0.001). The predictive role of HPD and AOP in multiparous patients was compared and AOP had superior predictive role than HPD (p<0.001).

Figure 4:

Receiver operating curve analysis evaluating the predictive role of angle of progression and head to perineum distance for cesarean delivery in multiparous patients. AOP, angle of progression; HPD, head to perineum distance.

Discussion

Integration of sonography on the labor ward for routine obstetric practice may be an alternative way to overcome various obstacles in obstetrics. Mappa et al. claimed that intrapartum sonography was related to decreased rate of incorrect diagnosis of fetal head station and position [15]. The first approach known as sonopartogram was first introduced in 2014 by Hassan et al. [8]. After this, many studies searching the role of sonography in labor progress have been performed. Sonographic cervical dilatation, AOP, HPD and head to symphysis distance were some of the sonographic parameters used in the evaluation of labor progress.

HPD is the simple method evaluating the fetal head descent [16]. In a study of Horst et al. higher HPD was related to cesarean section and a unit increase in HPD was found to be associated with 39.63 % decrease in vaginal delivery rate [17]. In a European study searching the role of HPD in vacuum extraction, patients with HPD ≤25 mm had significantly shorter extraction time and patients with HPD ≤35 mm had significantly lower cesarean section rates [18]. In another study, HPD 55 mm predicted vaginal delivery with 97 % sensitivity and 88.1 % specificity [19]. Kasbaoui et al. claimed that HPD ≥40 mm in the second stage of labor is associated with difficult instrumental delivery [20]. Moreover, Dall’Asta et al. reported that short HPD obtained in the second stage of labor is related to spontaneous vaginal delivery [21].

Another parameter showing the fetal head descent is AOP. When the AOP increases, fetal head descent increases during labor. In previous studies, AOP was shown to be a predictor for successful vacuum delivery in occiput anterior position and forceps delivery for non-occiput posterior fetuses [14], 22], 23]. Furthermore, AOP was significantly higher in spontaneous vaginal delivery group as compared to vacuum extraction or cesarean delivery group [24]. The studies searching the cut-off levels for AOP in the literature could be classified according to the parity and stage of labor. Bulut et al. evaluated AOP in first stage of labor in multiparous and nulliparous patients and found that AOP >132.3° predicted vaginal birth with 70 % sensitivity and 75 % specificity [25]. Similarly Vinod et al. reported a degree of AOP as ≥89 with 79.3 % sensitivity and 65.6 specificity [26]. Kandil et al. evaluated AOP in the first stage of labor in only nulliparous patients and found that AOP >104° predicted vaginal birth with 90 % sensitivity and 86 % specifity [27]. Likewise, Elkadi et al. and Ibrahim et al. reported an AOP level ≥97 and >107.5° in the first stage of labor respectively [28], 29].

Barbera et al. evaluated AOP in the second stage of labor in both multiparous and nulliparous patients and reported that all patients delivered vaginally when AOP exceeds 120° [7]. Yonetani et al. claimed that second stage AOP ≥160° predicted vaginal birth in nulliparous patients with 42.6 % sensitivity and 88.1 % specificity while AOP ≥150° predicted vaginal birth in multiparous patients with 51.8 % sensitivity and 78.6 % specificity [30]. Malik et al. assessed only nulliparous patients and claimed that AOP >116° predicted vaginal delivery with 96.49 % sensitivity and 96.43 % specificity [31]. Similarly, Hadad et al. reported 71.6 % sensitivity and 83.7 % specificity for AOP levels above 138° in nulliparous patients at second stage of labor [32]. In a systematic review searching eight studies reported that AOP between 108 and 119° at the beginning of second stage predicted vaginal delivery with 94 % sensitivity and 47 % specificity [33].

Many authors have focused on using AOP and HPD in previous studies. In a study of Eggobo et al., half of the women who had HPD >40 mm or AOP <110° delivered vaginally. This study reported 0.68 AUC (95 % CI, 0.55–0.80) for AOP and 0.67 AUC (95 % CI, 0.55–0.80) for HPD for spontaneous delivery. The likelihood of cesarean delivery was found to be 7 % when HPD is <40 mm whereas it increases up to 82 % when HPD is >50 mm. Likewise, the likelihood of cesarean delivery was 12 % for AOP >110° while it rose up 62 % for AOP <100° [34]. Hadad et al. found wider AOP and shorter HPD levels in the second stage of labor in spontaneous vaginal delivery [32]. Carvalho Neto showed that second stage AOP levels 129.9° and HPD levels 43 mm predicted vaginal delivery [5]. In contrast to these studies, Mitta et al. found no difference according to AOP levels between vaginal delivery, instrumental delivery and cesarean section groups. HPD values were different between vaginal and cesarean delivery groups and operative deliveries and cesarean delivery groups. Also, this study clarified that negative correlation was present between AOP and parity while HPD was positively correlated with BMI [14].

In the literature, only a few studies have focused on serial measurements of sonographic parameters for evaluating labor progress. In a study of Lee et al. 100 women between 37 and 41⁺⁶ weeks of gestation were assessed and slower change in parasagittal AOP and HPD were reported in cesarean section group. Moreover, cesarean group had higher estimated marginal mean of HPD and lower parasagittal AOP levels. In this study, measurements were taken in the first stage of labor serially but no cut-off value for cesarean section was obtained [35]. Another study using serial transperineal sonography in early labor included 183 nulliparous women and reported the feasibility of transperineal sonography for obstructed labor in early active phase [36]. Tse et al. evaluated 315 women in the first stage of labor. In this study, parity and obesity were found to effect ultrasound determined labor progress. Serial measurements of parasagittal AOP achieved a sensitivity of 80 % and specificity of 80 % for the prediction of cesarean delivery due to labor dystosia. This study evaluated parasagittal AOP and HPD per unit change of head station and cervical dilatation [1]. In the present study, we performed serial measurement of HPD and AOP. Consistent with the literature, we demonstrated slower changes in HPD and AOP values in cesarean section group. In our study, HPD >38 mm predicted cesarean delivery with 86.9 % sensitivity and 59.2 % specificity while AOP ≤117° predicted with 93.5 % sensitivity and 65.1 % specifity. Different from other studies, we compared the predictive role of these markers and found that AOP was superior than HPD in predicting cesarean delivery.

Another issue about sonographic parameters is the correlation of sonographic parameters with vaginal digital examination findings. In a study of Sweed et al. strong association were reported for cervical dilatation, moderate correlation for fetal head position and head station determined by HPD [3]. Similarly, Tse et al. reported positive correlation between parasagittal AOP and fetal head station or cervical dilatation whereas negative correlation was present between HPD and fetal head station or cervical dilatation [1]. HPD has known to be moderately correlated with digital vaginal examination head descent [8]. We found negative strong correlation between HPD and cervical dilatation or head station while AOP had strong positive association between cervical dilatation or head station.

The factors effecting fetal head descent determined by HPD and AOP was evaluated in the study of Chaemsaithong et al. [37]. Multiparity was found to be associated with faster head descent while obesity was related to slower fetal descent and higher HPD values. In our study, we classified according to parity and found different cut-off levels for multiparous and nulliparous patients supporting this finding. A HPD >39 mm predicted cesarean delivery with 95.5 % sensitivity and 62.5 % specificity while AOP ≤109° predicted with 87.9 % sensitivity and 78.2 % specifity in multiparous patients. In nulliparous patients, HPD >37 mm predicted cesarean delivery with 84.2 % sensitivity and 55.2 % specificity while AOP ≤110° predicted with 82.7 % sensitivity and 73.3 % specificity. The predictive role of AOP was superior than HPD in predicting cesarean delivery both in multiparous and nulliparous patients.

Strengths of the study include longitudinal design with serial sonographic measurements. Although interobserver variability is low, single sonographer performed all measurements in the present study. The study has larger sample size as compared to other repeated measurement studies. Blindness of clinicians and researchers minimized the bias. We classified patients as multiparous and nulliparous and suggested cut-off levels for cesarean section for each group.

This study also has some limitations. First, the findings could not be generalizable. Second, only term cases were included. Last, the results were limited to only Turkish patients.

In conclusion, we suggest that differences in labor progress could be determined by using serial intrapartum transperineal sonography. However, more research is needed for the implementation of sonopartogram to the clinical obstetric practice.