Machine learning improves cardiovascular risk definition for young, asymptomatic individuals

Authors: SANCHEZ-CABO, FATIMA - Instituto De Salud Carlos Iii; ROSSELLO, XAVIER - Instituto De Salud Carlos Iii; FUSTER, VALENTIN - Instituto De Salud Carlos Iii; BENITO, FERNANDO - Instituto De Salud Carlos Iii; MANZANO, JOSE PEDRO - Instituto De Salud Carlos Iii; SILLA, JUAN CARLOS - Instituto De Salud Carlos Iii; FERNANDEZ-ALVIRA, JUAN - Instituto De Salud Carlos Iii; OLIVA, BELEN - Instituto De Salud Carlos Iii; FERNANDEZ-FRIERA, LETICIA - Instituto De Salud Carlos Iii; LOPEZ-MELGAR, BEATRIZ - Instituto De Salud Carlos Iii; MENDIGUREN, JOSE - Banco De Santander; SANZ, JAVIER - Instituto De Salud Carlos Iii; ORDOVAS, JOSE - Jean Mayer Human Nutrition Research Center On Aging At Tufts University; ANDRES, VICENTE - Instituto De Salud Carlos Iii; FERNANDEZ-ORTIZ, ANTONIO - Instituto De Salud Carlos Iii; BUENO, HECTOR - Instituto De Salud Carlos Iii; IBANEZ, BORJA - Instituto De Salud Carlos Iii; GARCIA-RUIZ, JOSE MANUEL - Instituto De Salud Carlos Iii; LARA-PEZZI, ENRIQUE - Instituto De Salud Carlos Iii

Submitted to: Journal of the American College of Cardiology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 8/3/2020
Publication Date: 9/28/2020
Citation: Sanchez-Cabo, F., Rossello, X., Fuster, V., Benito, F., Manzano, J., Silla, J., Fernandez-Alvira, J.M., Oliva, B., Fernandez-Friera, L., Lopez-Melgar, B., Mendiguren, J.M., Sanz, J., Ordovas, J.M., Andres, V., Fernandez-Ortiz, A., Bueno, H., Ibanez, B., Garcia-Ruiz, J., Lara-Pezzi, E. 2020. Machine learning improves cardiovascular risk definition for young, asymptomatic individuals. Journal of the American College of Cardiology. 76(14):1674-1685. https://doi.org/10.1016/j.jacc.2020.08.017.
DOI: https://doi.org/10.1016/j.jacc.2020.08.017

Interpretive Summary: Changes in heart function occur many years before someone has a heart attack or stroke. However, the traditional heart disease prediction tools underestimate cardiovascular risk in apparently healthy individuals. The purpose of this work, conducted by investigators in Spain and at the HNRCA in Boston, was to better predict future heart attacks or strokes at the individual level using a machine-learning model. The model was based on easily measured variables to predict the presence and extent of subclinical atherosclerosis (SA) in young, asymptomatic individuals participating in the Progression of Early Subclinical Atherosclerosis (PESA) Study. The model generated with machine-learning approaches uses age, systolic blood pressure, and 10 commonly used blood/urine tests and dietary intake values to identify young, asymptomatic individuals with an increased risk of CVD based on their extension and progression of SA. These individuals are likely to benefit from imaging tests or pharmacological treatment, thus increasing our ability to prevent cardiovascular events.

Technical Abstract: Background Clinical practice guidelines recommend assessment of subclinical atherosclerosis using imaging techniques in individuals with intermediate atherosclerotic cardiovascular risk according to standard risk prediction tools. Objectives The purpose of this study was to develop a machine-learning model based on routine, quantitative, and easily measured variables to predict the presence and extent of subclinical atherosclerosis (SA) in young, asymptomatic individuals. The risk of having SA estimated by this model could be used to refine risk estimation and optimize the use of imaging for risk assessment. Methods The Elastic Net (EN) model was built to predict SA extent, defined by a combined metric of the coronary artery calcification score and 2-dimensional vascular ultrasound. The performance of the model for the prediction of SA extension and progression was compared with traditional risk scores of cardiovascular disease (CVD). An external independent cohort was used for validation. Results EN-PESA (Progression of Early Subclinical Atherosclerosis) yielded a c-statistic of 0.88 for the prediction of generalized subclinical atherosclerosis. Moreover, EN-PESA was found to be a predictor of 3-year progression independent of the baseline extension of SA. EN-PESA assigned an intermediate to high cardiovascular risk to 40.1% (n = 1,411) of the PESA individuals, a significantly larger number than atherosclerotic CVD (n = 267) and SCORE (Systematic Coronary Risk Evaluation) (n = 507) risk scores. In total, 86.8% of the individuals with an increased risk based on EN-PESA presented signs of SA at baseline or a significant progression of SA over 3 years. Conclusions The EN-PESA model uses age, systolic blood pressure, and 10 commonly used blood/urine tests and dietary intake values to identify young, asymptomatic individuals with an increased risk of CVD based on their extension and progression of SA. These individuals are likely to benefit from imaging tests or pharmacological treatment.