Developing a concise multivariable predictive model for cesarean delivery following neuraxial analgesia during labor: a prospective observational cohort study.

BACKGROUND

Sometimes, planned vaginal deliveries with neuraxial analgesia may result in unplanned cesareans. We aimed to determine the incidence of cesarean among parturients receiving neuraxial analgesia for vaginal delivery, identify associated factors, and develop a predictive model.

METHODS

In this prospective observational cohort study, we evaluated parturients receiving neuraxial analgesia for vaginal delivery and analyzed factors associated with progression to cesarean. Multiple logistic regression with a step-up procedure was performed. The dataset was split into training (70%) and testing (30%) databases, with the latter used to assess performance metrics. Bootstrap validation with 5,000 repetitions was performed.

RESULTS

We evaluated 331 parturients and 94 (28.4%) underwent cesarean. Variables differing between cesarean and vaginal delivery groups (p < 0.05) included patient age, body mass index, gestational age, cervical dilation at analgesia initiation, time under analgesia, labor conducted/monitored by nurses, and oxytocin use after analgesia initiation. Three variables remained predictive [odds ratio (95% Confidence Interval (95% CI))]: patient age: 1.0436 (1.0091 to 1.0835), p = 0.018; time under analgesia: 1.0043 (1.0008 to 1.0081), p = 0.018; and oxytocin use after analgesia initiation: 0.0921 (0.0400 to 0.1945), p < 0.001. Predictive area under the curve (95% CI) was 71.8% (60.5%‒83.1%). Arrest of descent (35.1%) and fetal distress (34.0%) were the leading indications for cesarean.

CONCLUSIONS

Among parturients receiving neuraxial labor analgesia, older patients, longer analgesia duration, and no oxytocin use after analgesia initiation increase the probability of cesarean, with moderate predictivity. Arrest of descent and fetal distress were the main causes of cesarean.