Abstract
Objective: To develop and validate a predictive model assessing
the risk of cesarean delivery in
primiparous women based on the
findings of magnetic resonance
imaging (MRI) studies.
Design: Observational study
Setting: University teaching hospital.
Population: 168 primiparous women with clinical findings
suggestive of cephalopelvic
disproportion.
Methods: All women underwent MRI measurements prior to the
onset of labor. A nomogram model to predict the risk of cesarean
delivery was proposed based on the MRI data.
The discrimination of the model was
calculated by the area under the receiver operating characteristic curve
(AUC) and calibration was assessed by calibration plots. The
decision curve analysis was applied
to evaluate the net clinical benefit.
Main Outcome Measures: Cesarean delivery.
Results: A total of 88 (58.7%) women achieved vaginal
delivery, and 62 (41.3%) required cesarean section caused by obstructed
labor. In multivariable modeling, the maternal body mass index before
delivery, induction of labor,
bilateral femoral head distance,
obstetric conjugate, fetal head circumference and fetal abdominal
circumference were significantly associated with the likelihood of
cesarean delivery. The discrimination calculated as the AUC was 0.845
(95% CI: 0.783-0.908; P< 0.001). The sensitivity and specificity of the nomogram
model were 0.918 and 0.629, respectively. The model demonstrated
satisfactory calibration. Moreover, the decision curve analysis proved
the superior net benefit of the model compared with each factor
included.
Conclusion: Our study provides a nomogram model that can
accurately identify primiparous women at risk of cesarean delivery
caused by cephalopelvic disproportion based on the MRI measurements.
Keywords : Nomograms, Pelvimetry, Cephalopelvic disproportion,
Cesarean section