Antenatal weight estimations have limited sensitivity and specificity for the detection of macrosomia. The objective of our study was to examine the screening efficacy of the subcutaneous tissue width/femur length ratio for the intrapartum detection of fetal macrosomia in a non-diabetic population at term.
Intrapartum sonographic measurements were performed in 178 well-dated gravidas at 37–41 weeks' gestation with negative glucose tolerance screens. The biparietal diameter, femur length (FL), abdominal circumference and subcutaneous tissue width of the thigh (SCT) were determined. Subsequently, predictions for macrosomia (actual birth weights above the 90th centile) were made using varying cut-off points of the examined parameters or estimated fetal weights.
Macrosomia occurred in 27 newborns (15.1%). The SCTFL ratio was independent of gestational age (r = −0.017). Maternal age, gravidity, parity, gestational age and the ratio of male-to-female infants were similar in pregnancies resulting in appropriate-for-gestational-age and macrosomic infants (NS). There was no difference in the SCT/FL ratio between these groups (p = 0.067; 99% power to detect 2 standard deviation differences). Comparison of screening efficacy by the univariate z score for the area under receiver operating characteristic (ROC) curves (θ) revealed that the abdominal circumference had the best sensitivity-specificity trade-off (θ = 0.8843; p < 0.0001 for comparison with SCT/FL ROC curve), followed by weight estimations based on the Hadlock formula (θ = 0.8773; p < 0.0005), the Shepard formula (θ = 0.8606; p < 0.0001), subcutaneous tissue thickness alone (θ = 0.6872; p < 0.01) and the SCT/FL ratio (θ = 0.6303).
We conclude that the SCT/FL ratio is a poor sonographic predictor of fetal macrosomia in the non-diabetic pregnancy and does not improve fetal weight estimations by conventional sonographic parameters. Copyright © 1999 International Society of Ultrasound in Obstetrics and Gynecology