Screening for Cardiac Disease with Genetic risk scoring, Advanced ECG, Echocardiography, Protein Biomarkers and Metabolomics

Abstract

Introduction Screening patients for cardiovascular disease has not been widely advocated due to cost implications and is reserved for high risk or symptomatic patients. We undertook an exploratory study to evaluate the promising low-cost methods for screening, including genetic risk scoring (GRS), advanced ECG (A-ECG), echocardiography and metabolomics. Methods 78 patients underwent advanced 5-min ECG and echocardiography, including global longitudinal strain (GLS), and echocardiographic calcium scoring (eCS). A GRS of 27 SNPs (GRS27) related to coronary disease and 3 SNPs for atrial fibrillation was used, as well as hs-troponin (Abbott, Singulex, Roche), NTproBNP (Roche) testing and targeted plasma metabolomics using GC-MS. Results were correlated with the presence of coronary artery disease (CAD) (CT coronary angiography (CTCA)), measures of left ventricular hypertrophy (LVH) (echocardiography and CTCA), and LV systolic dysfunction (LVSD) (echocardiography). Results LV dysfunction was accurately identified by using either A-ECG (AUC 0.97, 0.89 to 0.99) or NTproBNP. eCS demonstrated accurate discrimination of CAD (AUC 0.84, 95% CI 0.72 to 0.92, p < 0.0001. Troponin I (Abbott/Singulex) had the highest sensitivity and accuracy for the detection of LVH measured by either CT or echocardiography (AUC 0.85, 95% CI 0.73 to 0.92), however specificity was reduced by the presence of LV systolic dysfunction. Metabolomics and A-ECG identified underlying abnormal mechanisms related to both LVH (glycine metabolism) and LV dysfunction, (Citric Acid cycle). Metabolomics provided incidental utility by identifying metformin adherence and nutritional biomarkers. Conclusion A multi-omic approach to screening can be achieved at relatively low cost, and high accuracy, but will need to be evaluated in larger populations to prove its utility.