Early and accurate prenatal diagnosis of intrauterine growth restriction is important. Commonly used biometric parameters have limited specificity and require accurate dating. Fetal abdominal wall thickness (AWT) could be a useful supplemental parameter. We performed a retrospective study of 100 third trimester ultrasound exams and compared the sensitivity and specificity of AWT to those of weight percentile (WP) in predicting adverse perinatal outcome.
There is a statistically significant difference between the AWT of patients with normal perinatal outcome and that of patients with adverse outcome (P < 0.01). When compared with WP across the entire range of the receiver operating characteristics curve, AWT [area under the curve (AUC), 0.76] has an efficacy similar to that of WP (AUC, 0.72; P = 0.30). However, AWT has superior performance over WP (AUC, 0.72 vs AUC, 0.61, respectively, P = 0.04) in the high specificity range (70%–100%) of the receiver operating characteristics curve, where the consequences of a false negative greatly outweigh those of a false positive. In our study population, with a cutoff value of 4 mm, AWT was a useful and more specific predictor of adverse perinatal outcome than WP. Abdominal wall thickness may be more useful in situations when dating is uncertain.