Shah, T; (2007) Linking observational and genetic approaches to determine the role of C-reactive protein in coronary disease risk. Doctoral thesis , UCL (University College London).

Preview

Text Shah.Tina_thesis.pdf Download (72MB) | Preview

Abstract

Inflammation may contribute to atherogenesis. C-reactive protein (CRP) is an acute phase protein whose circulating concentrations provide an index of infection or inflammation. Prospective studies have consistently shown associations of CRP among apparently healthy individuals with risk of future hypertension, diabetes and coronary heart disease (CHD) events, leading to the proposal that CRP may play a causal role in atherogenesis, and its measurement could help predict CHD events. However, CRP is associated with a range of risk factors and biomarkers linked to atherosclerosis, so its predictive utility could be limited once these are accounted for. Its causal relevance is also uncertain because of the potential for confounding and reverse causation in observational studies. Experimental data have not resolved controversy on causation, as recent studies have indicated that apparently pro-atherogenic actions of CRP may have been mediated by preservatives or contaminants in commercial CRP preparations. Studies were therefore undertaken to evaluate the role of CRP in the prediction and pathogenesis of atherosclerotic events, by linking observational with genetic approaches. Appropriate analytical tools were used to evaluate the performance of CRP as a screening test for CHD. Single nucleotide polymorphisms (SNPs) in the CRP gene were identified using a bioinformatic approach and linkage disequilibrium was assessed to identify 'tagging SNPs" from which haplotypes could be inferred. A literature-based systematic review was undertaken to obtain precise estimates of the effect of SNPs on CRP concentration and to confirm the expected balanced distribution of potential confounders. In new studies, associations between haplotype (or genotype for a potentially functional promoter variant), and CRP concentration were tested. The effect of SNPs and haplotypes on CRP following an acute inflammatory stimulus was also assessed. The frequency of CRP SNPs and their effect on CRP were also examined in different ethnic groups. Having confirmed that SNPs and haplotypes of the CRP gene provide an unbiased proxy for CRP itself, associations with pro-atherogenic phenotypes, incident hypertension, diabetes and CHD were tested. Despite showing association with CHD, CRP performed poorly as a predictive test. Reasons for the poor performance were explored and comparisons were made with established risk factors and emerging biomarkers. SNPs and haplotypes were associated robustly with CRP, but not with other risk factors for CHD. A Mendelian randomisation approach linking genetic and non-genetic information provided no positive evidence that CRP induces a proatherogenic phenotype, nor increases risk of hypertension, diabetes or CHD. However, larger sample sizes will be required to exclude a small but potentially important causal influence of CRP on cardiovascular risk. In the absence of a selective CRP-lowering drug, this approach provides the only currently practical method for obtaining randomised evidence on the causal relevance of CRP in atherogenesis.

Download activity - last month

Export as XMLCSVJSON

Download activity - last 12 months

Export as JSONXMLCSV

Downloads by country - last 12 months

Export as CSVXMLJSON

Archive Staff Only

View Item