Development of a high density-lipoprotein (HDL) functionality score associated with predictive cardiovascular risk algorithms and subclinical atherosclerosis in Brazilian individuals

Author(s)

Maria Camila Pruper de Freitas

• Document type: Doctoral Thesis
• Press: São Paulo.
• Institution: Universidade de São Paulo (USP). Faculdade de Saúde Pública (FSP/CIR)
• Defense date: 2019-05-16

Examining board members

Nágila Raquel Teixeira Damasceno; Alessandra Carvalho Goulart; Maria Cristina de Oliveira Izar; Tatiana Natasha Toporcov

• Advisor: Nágila Raquel Teixeira Damasceno

Abstract

Introduction: current studies have not presented association between high density-lipoprotein cholesterol (HDL-C) increase, induced by drugs or genetic mutations, and coronary events reduction. HDL plays different functional cardioprotective role. Objective: to develop a HDL functionality score (HFS) and to assessment its association with predictive cardiovascular risk algorithms and subclinical atherosclerosis outcomes in Brazilian subjects. Methods: cross-sectional study based in two steps. In the first step, the HFS predictor of cardiovascular risk disease (HFS-CVR) was developed and validated on CARDIONUTRI study subsample (n=354). In second step the HFS predictor of subclinical atherosclerosis (HFS-SA) was developed and validated on ELSA-Brasil study subsample (n=4549). CARDIONUTRI study evaluated paraoxonase 1(PON1) and cholesterol ester transfer protein (CETP) activity, apolipoprotein AI (APOAI) concentration, HDL antioxidant capacity, and HDL subfractions by standard Lipoprint® method. ELSA-Brasil study evaluated the size of HDL and subfractions by Vertical Auto Profile (VAP) and Nuclear Magnetic Resonance (NMR) method, and the diagnosis of subclinical atherosclerosis by computed tomography, quantifying coronary artery calcification (CAC) and CAC score. Results: in the development of HFS-CVR, the large HDL presented greater strength of association with cardiovascular risk in the multiple final model (OR = 0.797; p <0.001). The HFS-CVR showed satisfactory performance by Framingham risk score (AUC = 0.899; p <0.001), Reynolds risk score (AUC = 0.722; p <0.001) and Adult Treatment Panel III/2013 guidelines (AUC = 0.864; p <0.001). In addition, HFS-CVR presented satisfactory reproducibility and was associated with subclinical atherosclerosis on ELSA-Brasil sample using large HDL measurements derived from the VAP method (AUC = 0.864; p <0.001 and r = 0,252; p <0,001) or the NMR method (AUC = 0.876; p <0.001 and r = 0.277; p <0,001). HFS-AS was developed from the HDL size (nm), because presented greater association with subclinical atherosclerosis in the final multiple model (OR = 0.549; p <0.001). HFS-AS demonstrated satisfactory performance (AUC = 0.769; p <0.001). Conclusion: the HFS demonstrates strong association with cardiovascular risk and subclinical atherosclerosis, independent of HDL-C, with emphasis on large HDL. Controversial results, between HDL subfractions and cardiovascular irsk seem to maintain a relation with the different methodologies used in analysis. Therefore, the validation of the methods and the inclusion of the HDL size as a cardiovascular risk marker reveal a promising future as an adjunct in the estimation of cardiovascular risk, drug management and decision making in clinical practice. (AU)

• FAPESP's process: 15/06565-5 - Development of High Density Lipoprotein (HDL)Functionality Score and Association with Cardiovascular Risk Prediction Algorithms and Clinical Outcomes in Brazilian Subjects
• Grantee: Maria Camila Pruper de Freitas
• Support Opportunities: Scholarships in Brazil - Doctorate