Introduction: Although highly sensitive troponin I (hsTnI) concentrations are valuable in the diagnosis of patients with acute myocardial infarction (MI), their utility in patients without acute MI remains unclear.
Hypothesis: Among 991 patients without acute MI referred for diagnostic coronary angiography for various indications, we hypothesized hsTnI concentrations would correlate with prevalent coronary artery disease (CAD) and predict incident cardiovascular (CV) events and mortality.
Methods: We measured hsTnI using a single molecule counting assay (Singulex, Alameda, CA) in samples obtained at the time of angiography. During mean follow up of 3.3 years, logistic and Cox proportional hazards regressions were used to relate hsTnl to prevalent CAD and incident CV events.
Results: The median hsTnI concentration in the study sample was 4.19ng/L; 38% of patients had hsTnI concentrations ≥99th percentile concentration for this assay (6ng/L). Across increasing hsTnI concentration quartiles, patients had a higher prevalence of CAD (TABLE). In multivariate models, hsTnI ≥99th percentile independently predicted obstructive CAD (odds ratio 2.50, P<0.001) as well as incident MI (hazard ratio [HR] 2.68, P<0.001), CV death (HR 2.29, P=0.001) and all-cause death (HR 1.84, P=0.004). In those with >70% coronary stenoses, hsTnI ≥99th percentile independently predicted incident MI (HR 1.78, P=0.01), CV mortality (HR 2.50, P=0.003) and composite endpoint of MI/all-cause death (HR 1.97, P<0.001). In subjects with coronary stenosis <70%, hsTnI ≥99th percentile even more strongly predicted incident MI (HR=8.36, P<0.001), CV mortality (HR=3.60, p=0.03) and composite endpoint of MI/all-cause death (HR= 3.61, P<0.001).
Conclusion: In a large prospective cohort of patients without MI undergoing diagnostic coronary angiography, hsTnI concentrations were associated with a higher prevalence of CAD and predicted incident MI, CV death and all-cause death.