[Routine application of results of a randomised trial: performance of ultrasound in daily routine practice for screening fetal macrosomia].

Abstract

Objective: To evaluate the performance of ultrasound at 36 weeks of gestation for screening fetal macrosomia according to the criteria of the randomized "DAME" trial.

Method: Retrospective observational study over 2 years in a type 3 maternity hospital. The primary outcome was birth weight above the 97th percentile on Audipog birth weight curves. The sensitivity, specificity, positive and negative predictive value of ultrasound at 36 SA for predicting macrosomia were calculated. For each patient, the absolute value of the difference between actual and estimated weight [EPF difference=(Birth weight - Estimated weight)/Birth weight] was calculated. Among undetected macrosomic newborns (≥97th Audipog percentile), risk factors associated with misdiagnosis were investigated (false negatives versus true positives).

Results: Ultrasound estimation of fetal weight at 36 SA to detect fetal macrosomia had a sensitivity of 43.7% and a specificity of 82.1%. With a positive predictive value for ultrasound of 36,9% and a negative predictive value of 85.8%. The mean absolute value of the difference between estimated and actual birth weight was 287g (95% CI [115-409]), with a mean relative difference of 6.8% (95% CI [3.0-11.3]). A relative difference of less than 10% between the estimated weight and the birth weight was found in 68.5% of the patients. Among macrosomic newborns (≥97th percentile), a third-trimester estimated fetal weight at or above the 90th percentile, along with a history of fetal macrosomia in a previous pregnancy, reduced the risk of false negatives in the screening ultrasound at 36 weeks of gestation.

Conclusion: Ultrasound performed at 36 weeks' gestation in a population targeted for suspected fetal macrosomia in routine practice has moderate sensitivity and positive predictive value. It would be timely for teams also applying this protocol to share their results, in order to verify whether or not our findings can be generalized to a larger scale.