Exploring causes of cardiovascular disease beyond the major risk factors

Abstract

Background: Despite success in reducing cardiovascular disease (CVD) rates based on well-established risk factors, residual risk of CVD exits, and the disease burden is increasing with population aging. To address this gap, I explored CVD etiology from different perspectives hopefully to generate new insights. ABO blood group is related to CVD for reasons that are incompletely understood. Similarly, one of the most widely used means of preventing and treating CVD, i.e., statins, has pleiotropic effects. CVD rates are higher in men. Statins have paradoxical effects that would be expected to increase CVD risk via increasing body mass index (BMI), which I investigated for similar reasons. Finally, coronavirus disease 2019 (COVID-19) has also drawn attention to the role of thrombotic risk, where the underlying mechanisms remain unclear.

Methods: Using the UK Biobank I conducted phenome-wide association studies for ABO blood group, using tag single nucleotide polymorphisms (SNPs) antigens (O, B, A1 and A2), and sex-specifically for statins, using SNPs from HMGCR. To obtain specificity for statins, I also assessed if any non-lipid associations found for statins were evident for other major lipid modifiers, i.e., proprotein convertase subtilisin/kexin type 9 (PCSK9) inhibitors and ezetimibe. I explored how statins increasing BMI related to IHD sex-specifically. Finally, given statins reduce thrombotic risk possibly via plasminogen activator inhibitor-1 (PAI-1), and response to infections via type 1 interferons (IFN-Is) increases thrombotic risk, I assessed if IFN-Is affected PAI-1.

Results: The O blood group antigen was inversely associated with circulatory diseases, total cholesterol, low-density lipoprotein cholesterol (LDL-c), and ovarian cancer, and positively associated with erythrocyte traits, leukocyte counts, diastolic blood pressure (DBP) and healthy body composition. The A1 antigen generally had opposite associations to O, with more marked associations for men than women for deep vein thrombosis, DBP, leukocyte traits, body composition traits, and for women than men for some erythrocyte and lipid traits. Genetically mimicking statins lowered cholesterol, increased glycated haemoglobin (HbA1c) and affected body composition as expected but also decreased serum calcium, sex hormone-binding globulin (SHBG) in women and platelet count, but increased basal metabolic rate and mean platelet volume. Associations were more evident for women than men for lipid traits, calcium and SHBG. Genetically mimicking PCSK9 inhibitors or ezetimibe reduced LDL-c but did not affect body composition. However, associations for statins with ischaemic heart disease (IHD) were largely similar after adjusting for BMI. Four of the nine genetically predicted IFNs, i.e., IFNα4, IFNα6, IFNα14 and IFNβ were positively associated with PAI-1, accounting for multiple comparisons.

Conclusion: Systematic examination revealed complex associations for both ABO blood group and statins, with some similarity for cholesterol, but differences for blood pressure, blood cell traits and body composition, and HbA1c, serum calcium, and SHBG, with some differences by sex. Statins increasing BMI did not reduce protective effects on IHD. The IFN-I response to COVID-19 may raise PAI-1, which could be a potential target for preventing infection-induced thrombotic events. Future studies on the relevance of these findings in other populations, and of the complex sex-specific associations, are recommended.