A new study shows that a widely available cesarean prediction tool accurately predicts delivery outcomes and highlights maternal risks, helping patients and providers make better-informed choices about induction.
Study: External validation of calculator for cesarean delivery during induction of labor. Image credit: Tatiana Diuvbanova/Shutterstock.com
A team of US-based researchers externally validated a risk calculator that predicts the probability of cesarean delivery during labor induction. The findings, published in the International Journal of Gynecology and Obstetrics, reveal that the calculator works satisfactorily in predicting cesarean delivery risk and maternal adverse outcomes.
Background
The rate of labor induction, which is primarily carried out to achieve safe vaginal delivery, has increased significantly in recent times. The probability of cesarean delivery during induction varies between 9% and 59%.
Various predictive models have been developed to calculate the risk of cesarean delivery for women undergoing induction. These models are expected to facilitate clinical decision-making about induction and predict the risk of adverse pregnancy outcomes. However, only a few of these models have been validated in an external population or linked with adverse outcomes.
In the current study, researchers validated the predictive capacity of a published risk calculator developed by Rossi and colleagues in the U.S.
Study design
The study population included a total of 548 women with singleton pregnancy (pregnancy involving a single baby) who were undergoing labor induction at 32 weeks or more of pregnancy.
The risk calculator under examination was used to calculate cesarean delivery risk scores for each participant, which were categorized as less than 10%, 10% to less than 30%, and 30% or greater. These scores were used to predict the probability of cesarean delivery during induction and its association with maternal and neonatal adverse outcomes.
Key findings
Among 548 pregnant women who underwent labor induction, 29% had a cesarean delivery and 71% had a vaginal delivery.
The external validation of the risk calculator in the study population revealed that the calculator's risk prediction is equivalent to the observed events of cesarean delivery, indicating a good discriminative efficacy. Specifically, the area under the receiver operating characteristic curve (AUC) was 0.77, demonstrating good predictive performance, and calibration results showed predicted probabilities aligned with observed outcomes.
The risk calculator-derived cesarean delivery risk scores placed 33.8%, 34.7%, and 31.6% of participants in the ‘less than 10%’, ‘10% to less than 30%’, and ‘30% or greater’ categories, respectively.
The rate of cesarean delivery was significantly different between these categories, with a higher probability of having a cesarean delivery among participants with a score of 10% to less than 30% and 30% or greater. The participants belonging to these two categories also exhibited a significantly higher risk of adverse pregnancy outcomes in mothers. However, no significant differences in the risk of neonatal adverse outcomes were observed between the categories.
Study significance
The study demonstrates the applicability of a publicly available cesarean delivery risk calculator originally proposed by Rossi et al. According to the findings, the calculator-predicted cesarean delivery score of 10% or higher is associated with a higher risk of cesarean delivery and pregnancy-related adverse outcomes in mothers.
The calculator's external validation has been done at a single large level IV academic hospital in the U.S., which serves as a tertiary referral center and manages a significant proportion of high-risk pregnancies that frequently require labor induction. Existing evidence indicates that most pregnant women accept labor induction when medically indicated.
Given this high induction rate, a risk calculator is necessary to facilitate shared decision-making about the risks and chances of successful induction. The calculator validated in this study does not require physical or ultrasonographic examinations and thus could be used by patients as part of counseling or telemedicine consultations. However, the authors emphasize that its primary role is to support clinician-patient discussions rather than self-directed decision-making.
The study findings highlight the calculator's predictive accuracy and its association with maternal adverse outcomes. This can help pregnant women assess the risks and benefits of induction and better allow them to give informed consent.
Notably, the study highlights that the calculator's predictive accuracy reduces at very high predicted cesarean delivery rates. This finding suggests that over-reliance on calculators may lead to more frequent and potentially avoidable cesarean deliveries. Therefore, it is crucial to understand that risk calculators have been designed to guide counseling and shared decision-making, not to replace clinical judgment.
Based on the predictions made by the calculator, women with higher cesarean delivery risk inherently present with elevated obstetric risk factors that predispose them to both cesarean delivery and a higher incidence of complications like postpartum hemorrhage, which is the primary significant maternal adversity identified in the study.
Some risk factors substantially contribute to these pregnancy adversities, including prolonged labor, uterine atony, and chorioamnionitis. Therefore, calculator-based predictions should not be used to avert medically indicated inductions. These predictions should rather be regarded as vital information highlighting the need for heightened clinical vigilance and proactive management in high-risk pregnancies.
Because of the observational study design, the study could not determine the causality of observed associations. The researchers excluded women with previous cesarean deliveries from the study population, as their institution routinely uses the already-validated TOLAC calculator for this specific group. Therefore, the predictive power of this calculator in this specific group remains unclear.
The calculator validated in this study uses socially constructed variables of race and ethnicity, which may increase health disparities. This highlights the need for developing a calculator that does not use race and ethnicity as a predictor.
Download your PDF copy now!
Aston Medical School secures NIHR funding for 12 academic clinical fellows