Full text
| |
| Author(s): |
Vytor Hugo Pereira Mendes
Total Authors: 1
|
| Document type: | Master's Dissertation |
| Press: | São Paulo. |
| Institution: | Universidade de São Paulo (USP). Faculdade de Medicina (FM/SBD) |
| Defense date: | 2016-01-11 |
| Examining board members: |
Alexandre da Costa Pereira;
Luís Henrique Wolff Gowdak
|
| Advisor: | Alexandre da Costa Pereira |
| Abstract | |
Importance of coronary artery disease (CAD) in \"modern\" society is attested by the large increase of people affected by it. Data from the World Health Organization estimated that ischemic heart disease is the second leading cause of death worldwide by 2030. Futhermore, with the advent of predictive risk tools and genome-wide association studies, several studies have demonstrated the use of genetic risk scores (GRS) for predicting CAD, but these scores shows a slight improvement in the assessment. This study built a genetic risk score for CAD, using polymorphisms previously associated with CAD and phenotypes related to disease. One thousand, Three hundred forty-nine individuals, from the Brazilian Longitudinal Study of Adults Health (in Portuguese, Estudo Longitudinal da Saúde do Adulto-BRASIL), were genotyped for our study. Using single nucleotide polymorphisms (SNPs) obtained from the online database of GWAS evaluated the association of each polymorphism chosen with risk factors for cardiovascular disease. Individuals underwent determination of calcium artery coronary score (CACS) and CAC > 100 was used as the end-point. Zero inflated models were used for the construction of GRS. Framingham Risk Score (FRS) for CHD in 10 years was calculated as described by Wilson and D\'Agostino. In addition, our GRS consists of 47 genetic variants, adjusted for sex and age, associated with CHD or other phenotypes of coronary heart disease or risk factors. GRS showed significantly better accuracy than the Framingham Risk Score to predict the risk of CAD (AUC, 0.90; 95% CI: 0.88 0.93; P < 0.01). Also other two scores were created: first, a model with only age and sex; second, only the genetic information. We compare the Age+Sex model with the GRS and FRS through ROC curve and evaluate the performance of models compared with FRS. NRI and IDI analysis were performed to evaluate the reclassification of GRS and Age + Sex model in relation to the FRS model. The score only the SNPs was used to compare their addition to a model with risk factors. Finally, we analyzed the association between the GRS, risk factors and CAC using generalized linear models. In short, we create a GRS composed of 47 SNPs associated with coronary heart disease or cardiovascular risk factors that showed good results in risk prediction for CAD. GRS improved reclassification of risk for coronary artery disease, and significantly improves the discrimination between affected and unaffected individuals (AU) | |
| FAPESP's process: | 12/16494-0 - Genetic determinants of coronary artery disease in a sample of Brazilian population |
| Grantee: | Vytor Hugo Pereira Mendes |
| Support Opportunities: | Scholarships in Brazil - Master |