Interpretation (in light of other evidence)
Antenatal estimation of the relationship between the maternal pelvis and fetal size is essential to evaluate obstetrical prognosis. Various pelvic diameters have been proposed to assess the risk of unplanned cesarean delivery. Several studies reported a significant positive correlation between obstetric conjugate and vaginal delivery15, 19. Besides, Joyce et al.20 asserted the minimum obstetric conjugate of 10 cm that required a fetus of 3400g to pass through the birth canal. Similarly, our final model incorporated the obstetric conjugate as a significant predictive factor for dystocia. Another significant variable is the interspinous diameter, which is considered as the parameter representing mid-pelvis. Our study showed that the interspinous diameter was significantly associated with the cesarean delivery group by univariable analysis, which agrees with the previous report15, 21. However, the final multivariate model eliminated the interspinous diameter as an independent factor to predict the overall cesarean risk for dystocia. Moreover, we identified the bilateral femoral head diameter as a predictive factor in both univariable and multivariate analysis, which can be a novel parameter measuring the transverse distance of the mid-pelvis. The mid-pelvis was considered as an elemental plane of the birth canal, through which the fetus’ head usually passes in sagittal orientation. Therefore, the bilateral femoral head distance, which is located in the central of the maternal pelvis in anatomical structure, can be a reliable parameter to assess the capacity of the mid-pelvis.
Because unplanned cesarean delivery is associated with a greater risk of maternal morbidity and surgical complications than scheduled cesarean delivery and vaginal delivery22, it would be clinically beneficial to predict the labor outcome in women with suspected CPD. There have been previous efforts to provide predictive models for quantifying CPD. Morgan et al. originally proposed the fetal pelvic index 6, which combined fetal head and abdominal circumferences with maternal pelvic inlet and mid pelvis circumferences to identify CPD. However, it has been criticized for the clinically useful as a poor overall prediction value23, 24. Burke et al. developed a nomogram based on five parameters (maternal age, height, BMI, fetal abdominal circumference and fetal head circumference) to assess the risk of primary cesarean delivery with limited predictive accuracy (AUC = 0.69)25. Nevertheless, previous models were either limited diagnostic accuracy25, 26 or too complex for routine clinical practice27, 28. Additionally, some prediction tools required intrapartum factors that counseling about likely outcomes cannot be made until labor occurs29. Our nomogram model based on antepartum factors may be a clinically beneficial and accurate tool for predicting cesarean delivery risk in nulliparous women.
Previous studies have demonstrated that MRI is a reliable and accurate method to evaluate labor outcomes irrespective of the experience. A randomized trial conducted by Van Loon et. al.30asserted that MRI pelvic measurements allowed the better choice of delivery modes with a significantly lower emergency cesarean section rate of breech presentation at term. Some researchers attempted to identify if MRI pelvimetry could predict the success of vaginal delivery in women with previous cesarean sections31. As the study comprised only 16 patients, we considered the results should be supported by larger ones. Sporri et. al.21 confirmed the significant association between MRI pelvimetry and CPD in 42 women, especially the parameters of the interspinous dimension and obstetric conjugate. However, the inclusion criteria were unrepresentative and MRI pelvimetry was performed postpartum. Previous attempts have shown that several MRI pelvimetric parameters are associated with CPD, such as anterior-posterior mid-pelvis diameter, interspinous diameter, and mid-pelvis capacity15, 32, 33. Another application of MRI is to estimate fetal weight; compared with ultrasound biometry, it is more accurate 34. To date, there has not been an established method for accessing the effects of multiple risk factors incorporated MRI measurements for determining the probability of primary cesarean delivery in an individual woman. Our study expands the findings of previous investigations to develop a nomogram model based on pelvic dimensions and fetal biometry measured by MRI to predict successful vaginal delivery.