The predictive potential of asymptomatic mild elevation of cardiac troponin I on mortality risk of stable patients with vascular disease

Due to improved analytical performance of the newest generation of troponin assays, several patients have mild elevations of this parameter. Nevertheless, they do not show any signs of acute coronary syndrome. We speculated whether non-acute cardiac troponin I (cTnI) concentrations may improve prediction of residual mortality risk in clinically stable outpatients with chronic vascular disease. We followed 830 patients (mean age 65.2years) after myocardial infarction, coronary revascularization or first ischemic stroke (pooled Czech samples of EUROASPIRE III and EUROASPIRE-stroke surveys, interviewed in 2006/2007) in a prospective cohort study. In addition to standard protocol, troponin I and brain natriuretic peptide (BNP) was estimated from frozen samples. Vital status and declared cause of death from death certificates was registered to ascertain a 5-year all-cause and cardiovascular mortality. During a median follow up of 2050days (5.6years) 168 patients died. In the multivariate Cox proportional hazard model, cTnI≥0.03ng/mL independently predicted an all-cause 5-year mortality with HRR 1.76 (95% CI: 1.09–2.83). In the Cox model, the better predictor of mortality was BNP >150ng/L [HRR 3.47 (95% CI: 2.23–5.41)]. However, the combination of BNP with cTnI did not substantially improve its sensitivity or predictive power. We cannot confirm the utility of asymptomatic mild cTnI elevation as a tool to detect residual risk of stable patients with vascular disease. On the other hand, BNP seems to be more appropriate for this purpose. •Mortality predictors in stable patients with vascular disease were studied.•Non-acute troponin I was independently associated with total mortality.•For stratification of residual risk is brain natriuretic peptide more suitable.