Detection of fetal cardiac anomalies: cost‐effectiveness of increased number of cardiac views

ABSTRACT Objective To compare the recommended three‐view fetal heart screening method to detect major congenital heart disease (CHD) with more elaborate screening strategies to determine the cost‐effective strategy in unselected (low‐risk) pregnancies. Methods A decision‐analytic model was designed to compare four screening strategies to identify fetuses with major CHD in a theoretical cohort of 4 000 000 births in the USA. The four strategies were: (1) three views: four‐chamber view (4CV) and views of the left (LVOT) and right (RVOT) ventricular outflow tracts; (2) five views: 4CV, LVOT, RVOT and longitudinal views of the ductal arch and aortic arch; (3) five axial views: 4CV, LVOT, RVOT, three‐vessel (3V) view and three‐vessels‐and‐trachea view; and (4) six views: 4CV, LVOT, RVOT and 3V views and longitudinal views of the ductal arch and aortic arch. Outcomes related to neonatal mortality and neurodevelopmental disability were evaluated. The analysis was performed from a healthcare‐system perspective, with a cost‐effectiveness willingness‐to‐pay threshold set at $100 000 per quality‐adjusted life year (QALY). Baseline analysis, one‐way sensitivity analysis and Monte‐Carlo simulation were performed. Results In our baseline model, screening with five axial views was the optimal strategy, detecting 3520 more CHDs, and resulting in 259 fewer children with neurodevelopmental disability, 40 fewer neonatal deaths and only slightly higher costs, compared with screening with the currently recommended three views. Screening with six views was more effective, but also cost considerably more, compared with screening with five axial views, and had an incremental cost of $490 023/QALY, which was over the willingness‐to‐pay threshold. The five‐view strategy was dominated by the other three strategies, i.e. it was more costly and less effective in comparison. The data were robust when tested with Monte‐Carlo and one‐way sensitivity analysis. Conclusion Although current guidelines recommend a minimum of three views for detecting CHD during the mid‐trimester anatomy scan, screening with five axial views is a cost‐effective strategy that may lead to improved outcome compared with three‐view screening. Copyright © 2020 ISUOG. Published by John Wiley & Sons Ltd.