Antenatal prediction of fetal macrosomia in pregnancies affected by maternal pre-gestational diabetes

Aims Higher rates of fetal macrosomia may occur in infants of women with pre-gestational diabetes compared with non-diabetic controls. Antenatal predication of fetal macrosomia remains challenging. Ultrasound over-estimated fetal weight could result in over-classification of fetuses as macrosomic with corresponding inappropriate clinical interventions. Previously we had studied a measurement - the anterior abdominal wall measurement (AAW) - to predict fetal macrosomia in fetal estimation of weight. The purpose of the study was to study whether specific third trimester ultrasound measurements with measures of glycaemic control (HbA1c) predicted macrosomia in babies born to women with pre-gestational diabetes. In particular, a new variant of this measurement (fetal anterior abdominal wall thickness (AAW), abdominal circumference (AC) ratio: AAW:AC) was investigated. Methods This was a prospective cohort study in a tertiary referral maternity hospital. Serial growth scans including measurement of AAW and AC: AAW ratio was performed at 30, 33- and 36-weeks' gestation. Birth-weight data was collected, and macrosomia was defined as >90th centile based on gestational age and gender of the baby. Serial HbA1c as measured at the first antenatal visit, 14, 20- and 36-weeks' gestation were reported for this study. Results Of the 416 pregnancies analyzed, mean maternal age was 33.3 years. One in five women were primigravida's. The mean birthweight was 3548 g (+/- 581 g), of which 142 (34%) babies were classified as macrosomic. The median gestational age at delivery was 38(3) weeks (31(4) - 40(2) weeks). There were 37 (9%) babies born preterm at <37 weeks' gestation. Mean AC measurements in fetuses that would be born with macrosomia compared with those with a non-macrosomic birth weight were 282 mm vs. 266 mm at 30 weeks, 318.3 mm vs. 297 mm at 33 weeks and 350 mm vs. 325 mm at 36 weeks' gestation (all p < .001). Mean AAW measurements in macrosomic fetuses compared with normal size fetuses were 3.7 mm vs. 3.3 mm at 30 weeks, 4.9 mm vs 4.3 mm at 33 weeks and 5.9 mm vs. 5.3 mm at 36 weeks' gestation (all p < .001). The mean AC: AAW was 0.01 for both normal and macrosomic fetuses at 30 weeks. There was no clinical or statistical difference in AC:AAW ratios between non-macrosomic and macrosomic infants. Binary logistic regression showed that AC at 36 weeks was most predictive of macrosomia (76.5%), followed by AAW at 30 weeks (68.5%). Using a combination of HbA1c booking, 14, 20, 36 weeks and AAW 30, 33, 36 weeks and AC 30, 33, 36 weeks predicted macrosomia in 80.9%. The ratio of AC: AAW did not act as a useful antenatal clinical predictor of macrosomia at birth. Conclusions Abdominal circumference at 36 weeks was the single best predictor of fetal macrosomia. A combined model of HbA1c, AC and AAW was the best antenatal predictor of macrosomia, with intriguing clinical possibilities in the possible prevention of maternal and fetal complications of macrosomia.

Automated summary

This study investigated specific ultrasound measurements in the third trimester and measures of glycaemic control to predict fetal macrosomia in babies born to women with pre-gestational diabetes. Results showed that abdominal circumference at 36 weeks was the most predictive of macrosomia, and a combined model of HbA1c, abdominal circumference, and anterior abdominal wall thickness was the best antenatal predictor of macrosomia.