Prediction of spontaneous onset of labor at term (PREDICT study): Research protocol

Abstract

Background Recent studies have shown that elective induction of labor versus expectant management after 39 weeks of pregnancy result in lower incidence of perinatal complications, while the proportion of cesarean deliveries remains stable, or even decreases. Still, evidence regarding collateral consequences of the potential increase of induction of labor procedures is still lacking. Also, the results of these studies must be carefully interpreted and thoroughly counter-balanced with women’s thoughts and opinions regarding the active management of the last weeks of pregnancy. Therefore, it may be useful to develop a tool that aids in the decision-making process by differentiating women who will spontaneously go into labor from those who will require induction. Objective To develop a predictive model to calculate the probability of spontaneous onset of labor at term. Methods We designed a prospective national multicentric observational study including women enrolled at 39 weeks of gestation, carrying singleton pregnancies. After signing an informed consent form, several clinical, ultrasonographic, biophysical and biochemical variables will be collected by trained staff. If delivery has not occurred at 40 weeks of pregnancy, a second visit and evaluation will be performed. Prenatal care will be continued according to current hospital guidelines. Once recruitment is completed, the information gathered will be used to develop a logistic regression-based predictive model of spontaneous onset of labor between 39 and 41 weeks of gestation. A secondary exploration of the data collected at 40 weeks, as well as a survival analysis regarding time-to-delivery outcomes will also be performed. A total sample of 429 participants is needed for the expected number of events. Conclusion This study aims to develop a model which may help in the decision-making process during follow-up of the last weeks of pregnancy. Trial registration NCT05109247 (clinicaltrials.gov).

Labor induction at 41+0 gestational weeks or expectant management for the nulliparous woman: The Finnish randomized controlled multicenter trial

“…The rates of CS (16.7% [n = 31] vs. 24.1% [n = 47], RR 0.7 [95% CI: 0.5-1.0], p = 0.07) and a composite of adverse neonatal outcomes (9.7% [n = 18] vs. 14.4%…”

“…In the exploratory analyses, these women had the lowest rate of CS, 12.4%, although the difference to 16.4% in women in the early induction group was not statistically significant. Predicting the probability of spontaneous onset of labor is difficult and research on the subject is needed, 24 but in our opinion, cervical status at the screening visit may serve as a useful basis for discussion with the pregnant woman.…”

Current status of the position on labor progress prediction for contemporary pregnant women using Friedman curves: An updated review

“…Recently, machine learning algorithms have been developed to predict various aspects of spontaneous childbirth, including premature birth, [33][34][35] mode of delivery, 36 fetal growth management, 37 pregnancy complications, 38 perinatal health management, 39 labor onset and delivery, 40 birth outcomes, 41 and labor progression. 6,42,43 While several systematic reviews have indicated the potential of AI and machine learning approaches in prenatal care (Table 1), they also highlight the need for further research for widespread clinical implementation.…”