A retrospective cohort study was conducted in a single tertiary, university-affiliated medical center, between January 2012 and December 2018.
Study Population: All women presenting with an unfavorable cervix (Bishop score < 6) (3), who were admitted for cervical ripening with PROPESS due to prolonged gestation were included in the study.
The cohort was divided into three study groups according to evaluation at 24 hours of treatment: group A) responders - women who achieved cervical ripening (Bishop score ≥ 8); group B) partial responders - women who had some cervical response but did not reach a Bishop score ≥ 8.
Or those, who were diagnosed with membrane rupture during this timeframe; and group C) non responders - women who required an additional ripening method, because they did not achieve cervical ripening (Bishop score < 6).
Eligibility was limited to women between 18 and 45 years of age with singleton gestations, who delivered a live birth neonate. Women with prior cesarean delivery, premature rupture of membranes at admission or known fetal anomaly were excluded from the study. Women who underwent cesarean delivery (due to non-reassuring fetal heart rate/ opted for CD), or those in whom cervical ripening was terminated prior to 24 hours evaluation (due to tachysystole, meconium stained amniotic fluid, intractable pain) were excluded as well.
Procedures and definitions
Gestational age was calculated according to last menstrual period and confirmed by a 1st trimester sonographic fetal crown-rump length measurement.
Postdate was defined as gestational age over 40 + 0 weeks (12).
According to our departmental protocol for postdate pregnancy, women carrying a low risk pregnancy are managed expectantly with non-stress test (NST) and biophysical profile twice weekly. When no contraindication for vaginal delivery exists and active labor is not observed, induction of labor is offered. Women opting for induction of labor undergo vaginal examination to assess cervical status - parameters of cervical dilatation, effacement, consistency, location and vertex position are determined to establish the Bishop score (4), and if 6 or below, cervical ripening is indicated. Cervical ripening by prostaglandins is performed in our center by a 10mg slow release vaginal insert of PGE2 Dinoprostone (PROPESS, Ferring Pharmaceuticals, Saint-Prex Switzerland). The vaginal insert is introduced into the posterior vaginal fornix and the woman is hospital admitted, for the following 24 hours. Women are evaluated upon regular contractions, rupture of membranes, suspected fetal compromise and routinely every 12 hours. The assessment includes NST and, if indicated, digital vaginal examination. The vaginal insert is removed upon the following scenarios: 1) Bishop score above 8, which denoted successful cervical ripening; 2) non reassuring fetal heart rate; 3) tachysystole, defined as more than 5 contractions in 10 minutes; 4) and upon completion of 24 hours from insertion.
Women who achieve cervical ripening (subgroup A, responders) or partial cervical ripening or rupture of membranes (subgroup B, partial responders) at or prior to completing 24 hours of treatment are transferred to labor and delivery floor for further labor augmentation by oxytocin infusion and/or amniotomy, following removal of the vaginal insert. Women who do not achieve cervical ripening at 24 hours undergo a second line ripening method with an EAB (subgroup C, non-responders).
Cervical ripening by EAB is performed by transcervical insertion of a 22F Foley catheter under visualization, inflated with 80ml saline. The catheter is attached to the women’s thigh with traction and removed after 24h, or earlier if spontaneous rupture of membranes or spontaneous EAB expulsion had occurred. After EAB removal or expulsion, the woman is transferred to labor and delivery floor, where oxytocin infusion and/or amniotomy are commenced if active labor had not occurred.
Data collection
Data was obtained from the medical electronic records including perinatal comprehensive database, maternal-fetal clinic records and delivery ward charts.
The following demographic and medical-obstetrical variables were recorded: maternal age, gravidity, parity, mode of conception, initial Bishop score, gestational age at delivery; and maternal co-morbidities including chronic hypertension, any thrombophilia, pre- gestational diabetes, and gestational diabetes. Body mass index (BMI) was calculated to account for overweight and obesity (BMI > 25kg/m2).
Maternal and neonatal outcomes were collected from the medical records. Maternal measures included: mode of delivery, intrapartum fever, Obstetric Anal Sphincter Injuries (OASI), post-partum hemorrhage (PPH) and blood products transfusion. Neonatal measures included: gender, birthweight, 1st- and 5th-minute Apgar scores, neonatal intensive care unit (NICU) admission status, sepsis, asphyxia, seizures, hypoxic ischemic encephalopathy (HIE), transient tachypnea of the newborn (TTN), respiratory distress syndrome (RDS), oxygen enrichment, mechanical ventilation and meconium aspiration syndrome (MAS). Labor characteristics of epidural anesthesia or meconium stained amniotic fluid were recoded as well.
Outcome measures: Primary outcome was defined as the rate of intrapartum cesarean delivery due to labor dystocia. Secondary outcomes were defined as maternal composite adverse outcomes including any of the following: intrapartum fever, OASI, PPH and blood products transfusion and neonatal composite adverse outcomes including any of the following: TTN, RDS, oxygen enrichment, mechanical ventilation, and MAS, 5-minutes Apgar score < 7, asphyxia, seizures or HIE.
Statistical analysis: Statistical analysis was performed using the SAS software (Version 9.4, SAS Institute, North Carolina, USA). Comparison between continuous variables was performed with Student's t-test and categorical data was compared using χ2 test. A probability value < 0.05 was considered significant. A multivariate logistic regression was performed to detect independent risk factors for cesarean delivery after controlling for possible confounders including maternal age, overweight and obesity and primiparity. We constructed a predictive model accounting for the following: responsiveness to prostaglandins, age, overweight and obesity and primiparity to predict the risk of cesarean delivery, using a receiver operating characteristic (ROC) analysis.