Association of cognitive impairment with statin use in coronary artery disease across APO (ε) genotypes in AllofUS

STRUCTERED ABSTRACT BACKGROUND Coronary artery disease (CAD) and Impaired Cognitive (IC) disease share sociodemographic, genetic, and clinical factors, but the association of IC with statin use in CAD remains unclear. OBJECTIVES To determine the association between IC and statin use in CAD based on APO (ε) genotype, sex, and lipid levels. DESIGN, SETTING, AND PARTICIPANTS We performed a retrospective study of AllofUS (AoU) participants with CAD (age≥60 yrs) enrolled from 2017 to 2023. We defined CAD as having a history of angina/myocardial infarction/chronic ischemic heart disease or having percutaneous coronary intervention/CABG, and IC defined as mild cognitive impairment or all-cause dementia, using ICD/SNOMED codes. MEASURES We assessed the association between IC and statin use using logistic regression analysis, while adjusting for clinical factors, sociodemographics, and APO (ε) genotypes before and after propensity score matching. We further performed stratified analysis by sex, and APO (ε) genotypes. We finally assessed the association between IC and statin users, based magnitude on the change in lipid levels before CAD and after IC (TC-Total cholesterol, LDL – low density lipoprotein, HDL-High Density Lipoprotein). Significance was defined at p < 0.05. RESULTS The cohort included 22,089 participants with CAD and 1343 with IC. Thirty-nine percent of participants were females, 77% were European, 13% were African American, and 9% were of Admixed American ancestry. The proportion of IC was higher ( 6.8% vs 3.5%, p<0.001 ) in statin users (n=17,191) vs non-statin users (n=4,898). IC was significantly associated with statin use ( OR:1.70;1.40-2.10, p = 4.9e-7 ) after adjustment for clinical factors, sociodemographics, and APO (ε) genotypes. After propensity-score matching between IC and CAD, we observed an association between IC and statin use ( OR:1.55;1.24-1.94, p =1e-4 ). In stratified analysis, the association between IC and statin use was strongest in the APO ε3/ε3 group ( OR:2.04;1.53-2.75, p = 1e-6 ), and in females ( OR:2.20;1.60-3.06, p = 2.e-6 ) compared to males (OR:1.43;1.10-1.90, p = 0.01). We finally observed an increased magnitude of association between IC and statin users having higher HDL increase ( > 10 mg/dl: OR:1.95;1.44-2.66, p=1e-5 ) as compared to statin users with lesser HDL increase ( ≤ 10mg/dl: OR:1.61;1.22-2.15, p=8e-4 ). CONCLUSION In the AllofUS cohort, IC was significantly associated with statin use in CAD participants. We observed the strongest association in the APO ε3ε3 group, among females, and with a greater increase in HDL levels in statin users. Key Message What is already known on this topic Coronary artery disease (CAD) and Impaired Cognitive (IC) disease, i.e., mild cognitive impairment or all-cause dementia, share genetic, sociodemographic, and clinical risk factors but the association of IC with statin use in CAD remains unclear. What this study adds We observed an association between IC and statin use in CAD participants after adjusting for sociodemographics, clinical factors, and APO (ε) genotypes. The association persisted after propensity score matching for sociodemographics, clinical factors, and APO (ε) genotypes. Further, when CAD participants were stratified across APO (ε) groups and by sex, we observed strongest magnitude of association between IC and statin use in the APO ε3ε3 compared to other APO (ε) genotypes, and in females compared to males. Among CAD participants with documented baseline and latest lipids (total cholesterol (TC), low-density lipoprotein (LDL), high-density lipoprotein (HDL)), IC was associated with statin use regardless of baseline lipid levels or latest lipid levels. While we did not observe any change in association between IC and statin use, based on the magnitude of decrease in TC and LDL levels, we observed a higher magnitude of association between IC and statin use, with a greater increase in HDL levels. How this study might affect research, practice or policy Our observations highlight the association of IC and statin use in CAD and the role of APO (ε) genotype evaluation, and serial lipid level assessments for evaluating for statin associated IC in CAD.