Abstract P011: ST2 As A Risk Predictor For Cardiovascular Disease, Heart Failure And Mortality In The Prospective Finrisk97 Population Cohort

Abstract

Introduction: ST2 is a receptor for the inflammatory cytokine IL33. The ST2-IL33 pathway has been associated with development of atherosclerosis and coronary artery disease in humans although the mechanism is not fully established. Increased ST2 levels have been associated with heart failure and death in patients with acute myocardial infarction and in the general population where it has also been associated with diabetes. We hypothesized that ST2 levels can identify people in the general population with elevated risk of subsequent cardiovascular disease (CVD) and add to existing risk prediction algorithms. Methods: We measured high sensitivity soluble ST2 (Presage) in 7,551 men and women (25-74 years) from the FINRISK97 prospective population cohort. Cox proportional hazards modeling using age as the time scale evaluated the ability of sST2 to predict cardiovascular endpoints including fatal and non-fatal cardiovascular disease (coronary heart disease, stroke, heart failure), incident diabetes and death over 14 years follow up. Missing data (~10%) was multiply imputed and participants with prevalent disease at baseline were excluded for corresponding endpoints of disease. Discrimination and reclassification statistics (c-index, NRI) for 10 year absolute risks were calculated with internal (10-fold) cross validation to compare the ability of ST2 to improve upon Framingham risk factors, N-terminal pro-brain natriuretic peptide (NTProBNP) and renal function (eGFR), established markers of CVD risk. Results: sST2 concentrations significantly predicted all-cause death (hazard ratio (HR) comparing the highest quartile with the lowest was 1.28 95% confidence interval (C.I.) 1.04-1.57 (1024 events), major adverse cardiac events (MACE) (HR 1.22 (95% C.I. 1.001-1.49) (1068 events) and stroke (1.44 (95% C.I. 0.99-2.1) (370 events) after adjustment for Framingham risk factors, NTProBNP and eGFR. sST2 showed suggestive but non-significant association with heart failure (HR 1.21 (95% C.I. 0.92-1.58) (577 events) after similar adjustment. Similarly, sST2 showed a trend toward elevated but non-significant risk for diabetes (HR 1.14 (95% C.I. 0.87-1.49) (377 events) after adjustment for age, area, sex, smoking, systolic blood pressure, antihypertensive medication, glucose and prevalent CVD. No improvement in the c-index was observed for models adding sST2 to Framingham and other risk factors. Clinical NRI was improved in fully adjusted models for death 9% ( p =1x10 -6 ) and stroke 4% ( p =0.028) with reclassification benefit mainly for non-cases. IDI was improved for MACE ( p =0.02) and stroke ( p =0.02). Conclusions: In a healthy general population from Finland (FINRISK97), sST2 improves prediction of death, MACE and stroke beyond Framingham covariates, NTProBNP and eGFR but does not improve prediction of heart failure.