Diagnostic Accuracy of Fetal Echocardiography in Congenital Heart Disease

Background: The accuracy of fetal echocardiography (FE) is not well defined, and reporting of diagnostic discrepancies (DDs) is not standardized. The authors applied a categorization scheme developed by the American College of Cardiology Quality Metric Working Group and applied it to FE. Methods: A retrospective single-center study was conducted of prenatally diagnosed major structural congenital heart disease, defined as expected need for intervention within the first year of life. DDs between pre- and postnatal findings were identified and categorized. Minor DDs had no clinical impact, moderate DDs had impact without harm, and severe DDs resulted in adverse events. Multivariate regression analysis was used to determine factors associated with discrepancy. Results: From December 2008 to September 2017, 17,096 fetal echocardiograms were obtained, among which 222 fetuses with a median gestational age at first FE of 24 weeks were included. There were 30 DDs (13.5%), of which the majority were false negatives (56.7%). Most were minor or moderate in severity, with one severe DD. The majority were possibly preventable (90%), with the most common contributing factor being technical limitations (43.3%). The most common anatomic segment involved was the ventricular septum (23%), primarily missed septal defects. Comparing cases with DDs versus those without, those with DDs were more likely to have high anatomic complexity (16.7% vs 3.6%, P = .01), maternal comorbidities (40.0% vs 22.1%, P = .03), and a younger maternal age (median, 27 vs 30 years, P = .02). They were also more likely to have later gestation at initial FE (median, 29.5 vs 24 weeks, P = .003), to have fewer total fetal echocardiograms (median, 2 vs 3, P = .002), and to have a fellow as the initial sonographer (36.7% vs 16.7%, P = .03). There were no significant differences in maternal race/ethnicity, fetal comorbidities, and interpreting physician experience between cases with DDs and those without. On multivariate analysis, variables associated with DD included high anatomic complexity, maternal comorbidities, and fellow as initial imager. A greater number of fetal echocardiograms was associated with reduced DD. Conclusions: FE had a DD rate of 13.5%, mostly minor and moderate in severity. Factors associated with DD included high anatomic complexity, maternal comorbidities, fellow as the initial sonographer, and fewer fetal echocardiograms. Strategies to reduce DD could include a regular secondary review and repeat FE, particularly when anatomic complexity is high.