| Background. It has been hypothesized that variability in phenotypic characterization of coronary artery disease (CAD) contributes to the heterogeneity of genetic association studies, and that more consistency in phenotype definition might improve detection of true genetic associations. To explore this, we assessed the extent of phenotypic heterogeneity, and quantified its impact, in a large literature sample of association studies included in prior meta-analyses.;Methods and results. We searched for large (n≥15 studies) meta-analyses of genetic associations and reviewed all genetic association studies included therein. From each primary study, we extracted phenotypic definitions, demographics, design characteristics, and numerical data. For each association, we assessed the magnitude and consistency of genetic effects within and across CAD phenotypes using meta-analytic methodologies. We included 965 individual studies investigating 32 distinct variants in 22 genes. We grouped CAD phenotypes in 3 categories: acute coronary syndromes (ACS) (426 studies, 44%), angiographically-documented disease (323 studies, 34%) and "broad, not otherwise specified CAD" (216 studies, 22%). Genetic effects were commonly discordant between phenotypes in the 18 meta-analytically significant associations and the different CAD phenotypes explained between-study heterogeneity in 9 of these associations. However, no CAD phenotype was steadily associated with larger and more consistent genetic effects in the meta-analyses. Similar results were obtained for additional phenotypic characteristics of cases and controls that were examined (premature disease or age-matching and angiographic verification of controls).;Conclusions. Substantial phenotypic heterogeneity exists in genetic associations for CAD and appears to partly account for between-study heterogeneity. However, we did not detect any evidence for improved signal detection with use of specific phenotypic definitions. |
