Sonographic prediction of shoulder dystocia in infants of diabetic mothers

Abstract

To determine if the difference between the abdominal diameter and biparietal diameter (AD-BPD difference), as measured by ultrasound examination, predicts shoulder dystocia in borderline macrosomic infants of diabetic mothers. A retrospective study was performed of births occurring from January 1990 through June 1995. Eligibility requirements included diabetic pregnancy, ultrasound examination within 2 weeks of delivery, estimated fetal weight of 3800-4200 g, and vaginal delivery. The mean AD-BPD difference was compared in normal deliveries and those complicated by shoulder dystocia, using the Student t test and by multiple regression analysis. A receiver operating characteristic curve was generated to determine if an AD-BPD cutoff value could be used clinically to predict shoulder dystocia. Thirty-one patients, six with dystocia, were eligible for the study. The mean AD-BPD differences for those with and without shoulder dystocia were 3.1 and 2.6 cm, respectively, a statistically significant difference (P = .05). Comparing the groups with and without shoulder dystocia, no significant differences could be found in mean age, parity, weight, birth weight, or gestational age. Shoulder dystocia occurred in six of 20 patients (30%) in whom the AD-BPD difference was at least 2.6 cm but in none of 11 patients in whom it was less than 2.6 cm, also a statistically significant difference (P = .05). The AD-BPD difference was greater in borderline macrosomic fetuses of diabetic mothers who experienced shoulder dystocia than in those who had uncomplicated vaginal deliveries. Applying an AD-BPD cutoff value of 2.6 cm to this population prospectively would have provided excellent sensitivity, specificity, and predictive value in identifying those fetuses at high risk for birth injury.