Evaluating first-line genetic testing strategies for inpatients with congenital heart defects

Genetic testing strategies used to determine the etiology of congenital heart disease/defects (CHD/CHDs) vary between and within institutions, leading to potentially missed diagnostic opportunities. There has been little investigation comparing the diagnostic utility of gene panels among more comprehensive strategies used in the genetic evaluation of patients with CHD. In this descriptive study, we investigated the diagnostic yields of different genetic testing strategies in a real-world cohort of 263 patients with CHDs with genetic diagnoses. We counterfactually determined the diagnostic yield of a virtual gene panel designed for this study. We compared the diagnostic yield of the gene panel to other testing strategies, including chromosomal microarray (CMA), CMA + the gene panel, and genome sequencing. We assessed diagnostic yield differences according to clinical presentations to determine if phenotypes can inform optimal testing strategies. The virtual gene panel would have identified 51.3% of genetic disorders in this cohort, and 25.9% of genetic disorders would have remained undetected; another 22.8% may have needed additional testing to fully characterize the diagnoses. A combined approach of the virtual gene panel and CMA increased the diagnostic yield compared with panel-only testing or CMA alone (87.8% vs. 51.3% and 63.1%, respectively). The gene panel plus CMA would have increased the diagnostic yield by 24%-35% compared with CMA or panel testing alone in patients with extracardiac anomalies, 19%-41% in syndromic patients, and 0%-70% across CHD classifications. This combined approach also eliminated the potential need for follow-up testing; however, genome sequencing had a higher diagnostic yield across all clinical presentations (99.6%). CHD gene panels and CMA used individually or in combination are suboptimal first-line testing strategies, missing up to 36.5% of genetic disorders in our sample. Given the wide spectrum of phenotypes and genetic etiologies, our results support consideration of standardized genome sequencing for patients with CHDs.