The impact of routine obstetric ultrasonographic screening in a low-risk population

OBJECTIVES: Our goal was to develop a framework for evaluating the current controversy regarding routine obstetric ultrasonography in a population of low-risk pregnancies. STUDY DESIGN: A retrospective chart review was performed for all low-risk pregnancies from a single obstetric practice during 1990 to 1994, to determine the accuracy of screening ultrasonography for fetal anomalies. All patients received a routine ultrasonographic examination at 18 to 20 weeks' gestation. Neonatal records for all patients were evaluated for the presence of both major and minor anomalies. The data were analyzed with attention to the classification of anomalies (all anomalies vs major anomalies, detectable vs nondetectable). RESULTS: A total of 860 fetuses in 854 pregnancies were evaluated. Anomalies were present in 5.35% (46/860); these were major anomalies in 1.16% (10/860) and minor anomalies in 4.19% (36/860). The sensitivity, specificity, and positive and negative predictive values for the diagnosis of all anomalies were 8.7%, 99.9%, 80%, and 95.7%, respectively. However, if only major anomalies detectable by ultrasonography are included, these values become 75%, 100%, 100%, and 99.9%, respectively. There was one false-positive diagnosis not affecting outcome, a small ventriculoseptal cardiac defect. Postnatal ascertainment of anomalies was excellent, as determined by an incidence of ventriculoseptal defects of 1 in 120. CONCLUSION: Distinguishing between major and minor anomalies and between ultrasonographically detectable versus nondetectable anomalies is essential in the evaluation of the diagnostic accuracy of screening ultrasonography. Any comparisons of studies examining the effectiveness of prenatal screening for congenital anomalies with ultrasonography should use the same outcome: major anomalies identifiable by ultrasonography.