To evaluate the clinical significance of fetal head progression distance (HPD), measured by transperineal ultrasound, during prolonged second stage of labor.
To evaluate the clinical significance of fetal head progression distance (HPD), measured by transperineal ultrasound, during prolonged second stage of labor.
In this prospective study, a single operator, who was blinded to the results of the digital examination, assessed using transperineal ultrasound women at ≥ 37 weeks of gestation with failure to progress in the second stage of labor. Patients had an empty urinary bladder and the examination was performed during maternal pushing. HPD was defined as the length of the line perpendicular to the infrapubic line that would connect it to the lowest part of the fetal bony skull. We analyzed associations between HPD and digital examination of fetal head station, fetomaternal characteristics, mode of delivery and perinatal outcome.
Sixty-five patients in prolonged second stage of labor participated in the study. The overall mean HPD was 6.50 (± 1.35; 95% CI, 6.16–6.83) cm. No correlation was found between HPD and head position or mode of delivery, but HPD was positively correlated with fetal head station and neonatal head circumference measured after delivery. Logistic regression and receiver–operating characteristics curve analysis demonstrated no significant predictive value of HPD with respect to mode of delivery.
Although HPD in prolonged second stage of labor could not predict mode of delivery, it may have a role as an ancillary tool for fetal head station assessment. Copyright © 2013 ISUOG. Published by John Wiley & Sons, Ltd.
Digital assessment of fetal head station during labor is a subjective tool, prone to considerable disagreement among examiners[1-3], especially during protracted labor, when molding and caput succedaneum make fetal head position interpretation more difficult. As a consequence, serious management errors in determining the most appropriate mode of mode of delivery may result. A number of studies, therefore, have investigated the use of objective transperineal ultrasound (TPU) parameters, at the second stage of labor, to predict mode of delivery[4, 6]. The head progression distance (HPD) represents a line connecting the most distal part of the fetal head to the infrapubic line (IPL, a vertical line drawn from the lower edge of the pubic symphysis). When this sonographic parameter was used for the assessment of fetal head engagement in 139 nulliparous pregnant women who were not in labor, it was found that the method had high reproducibility and that HPD correlated well with fetal head station, as assessed by both abdominal palpation and vaginal digital examination. Though assessments of HPD during the second stage of labor have not been documented, the value of HPD during this stage has been suggested in several publications[7-9].
Prolonged second stage represents a critical point of labor and prompts a decision as to the optimal mode of delivery. The objective of our study was to assess the clinical significance and utility of HPD in predicting mode of delivery during prolonged second stage of labor.
During a 12-month period (January 2010–December 2010), we evaluated prospectively 65 women at term (≥ 37 weeks' gestation) with a singleton pregnancy and failure to progress in the second stage of labor. Sixty-two of the women were included in a sister study on pubic arch angle and delivery mode, in this issue of the Journal. Fetal head station below the ischial spine and normal reassuring fetal heart rate were inclusion criteria of the study. Prolonged second stage was defined for nulliparous women as full cervical dilatation for > 3 hours if regional anesthesia was administered or > 2 hours in the absence of regional anesthesia, and for parous women as > 2 hours with regional anesthesia or > 1 hour without anesthesia. Following our normal practice in cases of prolonged second stage, we prepared patients for delivery in the lithotomy position and, when spontaneous vaginal delivery did not occur, an operative delivery was considered. Mode of delivery (instrumental vs Cesarean section (CS)) was determined by the physician, who was blinded to the results of HPD measurement. Similarly, the decision as to whether to use vacuum or forceps extraction was left to the discretion of the physician. Patients with prolonged second stage and head station higher then + 2 (on scale of − 5 to + 5) underwent evaluation in the operating room. A short trial of operative delivery was considered followed by CS if this failed.
Ultrasound examinations for the sonographic assessment of HPD were performed using a Voluson I portable ultrasound system (GE Medical Systems, Zipf, Austria) with a 3–5-MHz convex transabdominal transducer. The examination was performed on women with an empty urinary bladder. The ultrasound transducer, covered with a glove and ultrasound gel, was placed on the perineum in a sagittal position (Figure 1) and measurements were made during maternal pushing. The HPD was measured as the perpendicular distance from the IPL (a line drawn vertically from the inferior edge of the echogenic core of the symphysis pubis) to the lowest part of the fetal bony skull (Figure 2).
The ultrasound images obtained were stored in the system's hard drive. Clinical data were collected from patients' medical charts. The managing obstetricians were blinded to the ultrasound findings, including the measurements of HPD, and the sonographer (Y.G.) was blinded to the results of the digital examination.
Statistical analysis, using SPSS 17 statistical package (SPSS, Inc, Chicago, IL, USA) and Microsoft Excel 2007 software (Microsoft Corp, Richmond, CA, USA), included the ANOVA test (or Kruskal–Wallis test for non-normally distributed variables) for the evaluation of possible differences in fetomaternal clinical characteristics between various modes of delivery, and of HPD in various fetal head stations, head positions and modes of delivery. Bivariate correlation was assessed between HPD and fetomaternal characteristics, and between HPD and mode of delivery.
In order to compare all possibilities of mode of delivery vs HPD we classified mode of delivery as follows: MOD1 included spontaneous vaginal delivery (SVD), vacuum extraction, forceps extraction and CS; MOD2 included two groups: SVD, and all operative deliveries (OD) combined as one group (vacuum extraction, forceps extraction and CS); MOD3 included three groups: SVD, instrumental vaginal deliveries (including vacuum extraction and forceps extraction deliveries), and the CS group. Binary logistic regression analysis was performed to evaluate the predictive value of HPD vs mode of delivery as a dichotomous dependent variable (SVD and OD). The univariate odds ratio for operative delivery was calculated according to the formula: eb1, where b1 is the estimated increase in the log odds per unit increase in the value of HPD. The same formula: eb1+b2+b3+bn was used for calculation of the multivariate odds ratio. Statistical significance was defined as P < 0.05. Informed consent was obtained from all patients, and approval was given by the ethics committee of Sheba Medical Center.
Sixty-five patients in prolonged second stage of labor participated in the study. The mean maternal age was 30.6 (± 4.9) years. The mean gestational age was 39.5 (± 1.4) weeks. Forty (61.5%) women were nulliparous, 11 (16.9%) had delivered once previously and 14 (21.5%) had two previous deliveries, range of parity, 0–2; median, 0). Five (7.7%) women had undergone a previous CS. Thirty-eight (58.4%) fetuses were in an occipitoanterior (OA) position, 15 (23.1%) were occipitoposterior (OP), and 12 (18.5%) were occipitotransverse (OT). Seventeen (26.1%) patients who were managed expectantly underwent SVD and 37 (56.9%) underwent instrumental vaginal delivery, 32 (49.2%) by vacuum extraction and five (7.7%) by forceps extraction (including two who had failed vacuum extraction). Eleven (16.9%) were delivered by CS (including one who had failed vacuum extraction). Table 1 describes the fetomaternal clinical characteristics according to mode of delivery. The only parameters with statistically significant differences were a higher mean fetal head station in the CS group and shorter duration of the second stage of labor in the SVD group.
|Characteristic||Mode of delivery|
|Spontaneous vaginal delivery (n = 17)||Vacuum extraction (n = 32)||Forceps extraction (n = 5)||Cesarean section (n = 11)||P|
|Maternal age (years)||29.4 ± 4.7||31.4 ± 4.7||28 ± 2.6||31.2 ± 4.1||0.284|
|Maternal BMI before delivery (kg/m2)||27.3 ± 5.3||26.3 ± 3.5||27.4 ± 5.1||28.9 ± 3.9||0.437|
|GA at delivery (weeks)||39.1 ± 1.2||39.6 ± 1.3||38.6 ± 2.2||39.9 ± 1.5||0.216|
|Parity||0 (0–2)||0 (0–2)||0 (0–0)||0 (0–2)||0.496a|
|Birth weight (g)||3289.0 ± 375.3||3232.8 ± 375.5||3000.0 ± 296.9||3421.3 ± 378.5||0.199|
|Neonatal HC (mm)||325.4 ± 32.2||330.7 ± 22.9||326.6 ± 18.9||334.1 ± 16.8||0.863|
|Fetal head station||2 (0–3)||2 (0–3)||3 (2–3)||0 (0–2)||0.01a|
|Occipitoanterior||9 (23.7)||14 (36.8)||5 (13.2)||10 (26.3)|
|Occipitoposterior||4 (26.7)||10 (66.7)||0||1 (6.7)|
|Occipitotransverse||4 (33.3)||8 (66.7)||0||0|
|Duration of second stage of labor (min)||118.5 ± 47.4||179.0 ± 55.5||216.0 ± 72.2||193.9 ± 47.0||0.001|
|(P = 0.008)c||(P = 0.01)c||(P = 0.02)c|
The overall mean HPD was 6.50 cm (± 1.35; 95% CI, 6.16–6.83) cm, the range was 3–9 cm and the median was 6.6 cm. A statistically significant difference (Figure 3) and a positive correlation (correlation coefficient = 0.486) was found between mean HPD and fetal head station (P < 0.001). HPD also correlated positively with neonatal head circumference, measured after delivery (Table 2). There was no statistically significant difference (P = 0.384) between mean HPD in various head positions (6.38 ± 1.3, 6.49 ± 1.6 and 7.01 ± 0.8 cm for OA, OP, and OT positions, respectively), and no correlation (P = 0.195) between HPD and fetal head station (Table 2). There was also no significant correlation when only fetal OA presentation was evaluated (P = 0.153).
|Fetal head position||0.17||0.195|
|Fetal head station||0.486||< 0.001a|
|Second stage duration||0.127||0.34|
No statistically significant differences between the mean HPDs according to the various modes of delivery were found using ANOVA overall comparison (P = 0.179) or on post-hoc analysis between SVD and the other modes of delivery (Figure 4). A non-statistically significant difference was observed between the mean HPD in the SVD group and that in the operative delivery (instrumental vaginal delivery and CS) group (6.6 ± 0.97 cm vs 6.46 ± 0.46 cm, P = 0.658). There was no correlation between HPD and the three classifications of mode of delivery (Table 3).
Binary logistic regression was used to examine relationships between HPD and mode of delivery as a dichotomous outcome (SVD or operative delivery). Univariate analysis revealed a non-significant (P = 0.708) odds ratio of 0.923 (95% CI, 0.606–1.406) for operative delivery, for each cm increase of HPD. A non-significant (P = 0.586) odds ratio of 0.874 (95% CI, 0.537–1.421) for operative delivery was also found when only fetuses in the OA position were considered.
Multivariate analysis adjusted for maternal age, gestational age, maternal body mass index, parity, previous CS, fetal gender, neonatal head circumference and fetal head position demonstrated a non-statistically significant relationship between HPD and mode of delivery (P = 0.901), with an odds ratio of 0.951 (95% CI, 0.434–2.08). None of these fetomaternal variables yielded a statistically significant odds ratio for mode of delivery. The predicted probability for operative delivery by receiver–operating characteristics curve yielded a non-statistically significant (P = 0.771) area under the curve of 0.48 (95% CI, 0.61–0.90) (Figure 5), and the predicted probability was practically unchanged with increased HPD (Figure 6). No spontaneous deliveries occurred if HPD was below 4.4 cm (sensitivity of 87.5% and specificity of 100% for operative delivery).
The assessment of fetal head station at prolonged second stage of labor is cardinal to selecting the safest mode of delivery. Obstruction in the second stage of labor poses particular problems in fetal head station assessment, due to head compression in the birth canal that results in fetal head molding and caput succenedeum. As direct assessment of the ischial spines is not feasible by ultrasound, other reference landmarks for evaluation of the fetal head station have been suggested. Our findings confirm a correlation between clinical assessment of fetal head station and HPD during labor, supporting a study that demonstrated such a relationship in women not in labor.
Dupuis et al. reported a 12% error rate in the assessment of fetal head engagement using a birth simulator, even by experienced attending physicians. Such oversight may lead to erroneous application of instrumental delivery at unfavorable fetal head station, increasing the risk for neonatal and maternal morbidity[12, 13]. Thus, bedside ultrasound examination may be important as an objective tool for evaluating fetal head station.
Henrich et al. assessed, by three-dimensional computer tomography, the spatial relationship between the IPL and a parallel line at the level of the ischial spines. They marked the ischial spine level at a distance of 3 cm distal to the IPL. This measurement concurs with our clinical assessment of fetal head station at or below the ischial spines, since all HPD measurements were above 3 cm. The observation that no spontaneous vaginal deliveries occurred when HPD measured < 4.4 cm, which corresponds to fetal head station 0–2 (on scale of −5 to +5), further supports concordance between clinical and sonographic fetal head station assessment.
The vertical line IPL, used for the measurement of HPD, was traced from the lower margin of the echogenic core of the pubis and not from the hypoechogenic margin as proposed by Dietz and Lanzarone. We decided to use this echogenic landmark as it is better defined, in comparison to the lower pubic symphysis margin, during prolonged second stage of labor. Consequently, our HPD measurements were larger in comparison to theirs.
Despite the positive correlation of HPD with fetal head station during the second stage of labor, there was no statistically significant correlation or predictive value between HPD and mode of delivery. Thus, from a practical point of view, the usefulness of HPD in labor management could not be confirmed. A possible explanation for this is that, for the high fetal head stations, the linear portion of the birth canal concurs with HPD, since the latter is a linear measurement. However, the measurement may be less accurate with low stations, for which the caudal birth canal is curved, and the angle of progression may be more appropriate[6, 14]. An alternative explanation lies in the fact that the HPD measurement involves the lower part of the skull, below the IPL, but provides no information about the proportion of the head circumference that has passed this level of the birth canal. The linear positive correlation between head circumference and HPD observed in our study further supports the limited value of HPD in monitoring head descent during the second stage of labor, especially in fetuses with large head circumferences.
Henrich et al. reported an increased chance for successful instrumental delivery once the widest part of the skull has passed the IPL. This observation, together with our results, may encourage us to seek a sonographic landmark that will enable correlation between the relative portion of the fetal head that has passed the IPL and the mode of delivery.
The main drawback of our study was its small sample size, due to the relatively low number of women who reach prolonged second stage, and lack of comparison to women in the first stage of labor. A larger prospective study, including patients evaluated during the first stage of labor as well as the second, is required to further assess HPD measurement in the clinical management of labor.
Although our results do not confirm a predictive value for HPD measurement in determining mode of delivery, the finding that HPD correlates well with fetal head station suggests its usefulness as an ancillary tool for assessing fetal head descent during labor.