Preinducción del parto con misoprostol frente a dinoprostona vía vaginal
- QUIJADA CAZORLA, MARÍA ASUNCIÓN
- Eduardo Cazorla Amorós Director
- María Luisa Sánchez Ferrer Director
- Juan Carlos Martínez Escoriza Director
Defence university: Universidad de Murcia
Fecha de defensa: 20 April 2018
- Anibal Nieto Díaz Chair
- Josefa Marcos Sanmartín Secretary
- Ana María Palacios Marqués Committee member
Type: Thesis
Abstract
INTRODUCTION. Induction of labor (IOL) is one of the most common procedures performed in obstetrics, accounting for about the 25% of deliveries. Currently, there are two prostaglandins for IOL: dinoprostone and misoprostol. When the cervix is unfavorable, prostaglandins reduce the cesarean rate and they improve maternal-fetal results. However, the available studies are heterogeneous and contradictory in terms of drugs, doses and outcomes. In fact, there is not any study that demonstrates the superiority of one prostaglandin over the other. OBJECTIVES. The aim of this study is to compare misoprostol versus dinoprostone in terms of effectiveness (with vaginal deliveries as the main variable), safety, obstetric and perinatal outcomes and costs, as well as to perform a cost-effectiveness study. Secondly, it is intended to analyze the variables that influence the IOL and, through a mathematical model, to develop a computer application that allows to predict the probability of preinduction success. MATERIAL AND METHODS. A retrospective analytical observational study of cohorts has been developed in preinduced women with prostaglandins in the Hospital General Universitario de Alicante between 2012 and 2015. There were included: normal single pregnancies, greater or equal to 36 weeks, with unfavorable cervix (Bishop Score ?6) and medical indication to end the pregnancy. Pregnant women with contraindication of IOL and those ones with a previous cesarean section were excluded. RESULTS. A total of 1312 pregnant women were included: 890 with vaginal misoprostol (25 µg/4 hours) and 422 with vaginal dinoprostone 10 mg. The misoprostol group had a higher proportion of vaginal deliveries (VD), although without statistically significant differences [79.8% vs. 73.9%; pa 0.109; ORa 1.28 (0.95-1.74)]. No differences were found between both groups in VD without a serious side effect nor in VD before 12, 24 and 36 hours. Patients treated with misoprostol showed a statistically significant higher percentage of the beginning of active labor [52,6% vs. 41%; pa 0,004; ORa 1,44 (1,12-1,85)]. With the use of dinoprostone was observed a greater proportion of events (tachysystole, tachysystole with pathological fetal heart rate and pathological fetal heart rate) and higher percentage of emergency caesarean section (0.4% with misoprosol versus 1.9% with dinoprostone). With misoprostol was observed a greater use of epidural analgesia (85.1% vs. 76.8%; p<0,001) and shorter time until VD and until the beginning of labor. There were no differences in intrapartum or postpartum maternal complications, neonatal Apgar Score, umbilical artery pH<7.00, neonatal resuscitation or perinatal complications. The final cost of the process has been significantly higher in the dinoprostone cohort. Misoprostol has presented a better cost-effectiveness ratio. During the study period, a total cost saving of 154,175 euros was estimated when misoprostol was used instead of dinoprostone. We have found association of higher probability of VD and onset of labor with the lowest maternal age, the higher parity and previous deliveries, the lowest weight at the end of pregnancy and the lowest weight gain during pregnancy, the highest score in Bishop Test and the use of misoprostol versus dinoprostone. Finally, a mathematical model with a satisfactory predictive capacity (ROC curve = 0.78) has been developed that, integrated in a computer app, allows to predict the result of the IOL. CONCLUSIONS. According to the results of this study, misoprostol presents a better cost-effectiveness ratio and higher proportion of the beginning of labor, without significant differences in maternal-fetal safety. The computer application that we have developed could be useful in clinical practice as predictive tool to estimate a priori the probability of success of the IOL.