Detecting fetal macrosomia with abdominal circumference alone.

Objective: To determine if abdominal circumference (AC) can identify macrosomia (> or = 4,000 g) at or beyond 37 weeks.

Study Design: Prospectively, parturients at term admitted for delivery underwent sonographic mensuration of AC. A receiver-operating characteristic (ROC) curve was constructed to determine if AC can differentiate between normal (birth weight < 3,999 g) and macrosomia. A likelihood ratio was calculated. P < .05 was considered significant.

Results: The mean gestational age of the 256 subjects was 39.1 +/- 1.5 weeks, and the prevalence of macrosomia was 8.2% (21/256). Inspection of the ROC curve indicated that AC > or = 350 mm can identify macrosomic fetuses. The area under the ROC curve (0.79 +/- 0.04 for macrosomia) was significantly different than the area under the nondiagnostic line (P < .005). The likelihood ratio for AC to detect macrosomia was 2.9 (95% confidence interval, 2.1-4.0). Based on the proportion of macrosomia in our population, we would require over 1,000,000 newborns for a macrosomia analysis to obtain narrow confidence intervals around a clinically useful likelihood ratio.

Conclusion: Using the guidelines proposed by the Evidence-Based Medicine Working Group, AC is slightly useful in detecting macrosomia among term parturients.