Response to Letter Regarding Article “High-Sensitivity Troponin I and Incident Coronary Events, Stroke, Heart Failure Hospitalization, and Mortality in the ARIC Study”

Li and colleagues¹ noted in their letter that there were significant differences in cardiac risk factors across high-sensitivity troponin I (hs-TnI) quantiles in our study.² Specifically, higher quantiles were associated with more risk factors. This should not come as a surprise as hs-TnI reflects subclinical myocardial injury. Thus, risk factors known to contribute to clinical cardiovascular disease (CVD) are also likely to be associated with subclinical disease. However, we show that elevated hs-TnI is associated with incident clinical CVD events and mortality independent of traditional risk factors in adjusted regression models. The covariates used for adjustment were prespecified. In particular, we sought to demonstrate that elevated hs-TnI was independently associated with incident CVD events beyond variables included in the Pooled Cohort Equation (PCE) and added incremental predictive value beyond the PCE. We believe this provided clinical relevance as the PCE is currently widely adopted in the United States and recommended by the /American Heart Association/American College of Cardiology guidelines for risk stratification in primary prevention. We also included other variables in further adjustment models such as body mass index and estimated glomerular filtration rate, which are important in analyses of incident heart failure (HF) events. Lastly, many covariables such as total cholesterol, HDL cholesterol, and LDL cholesterol are interrelated, and care must be taken to choose covariables strategically in adjustment models as to avoid overadjustment.

We agree with Li et al on the importance of assessing the prognostic value of the change in hs-TnI over time. This is a topic of ongoing research. Members of this writing group previously evaluated the association between change in concentration of high-sensitivity troponin T (hs-TnT) with heart disease and death. Temporal increase in hs-TnT over a 6-year period was shown to be independently associated with increased risk for incident coronary heart disease (CHD), HF, and death in individuals without known history of CHD or HF.³ Studies such as these highlight the clinical value of cardiac troponins, which can be utilized for risk stratification but may also have potential as a tool to track disease progression and response to therapy.

As we have shown in our study, a combination of biomarkers (i.e., hs-TnI and hs-TnT) may be complementary and provide additional prognostic value beyond traditional risk factors in CVD risk assessment. We agree that in certain patient populations, the use of a combination of biomarkers may be particularly powerful when added to traditional risk factors. Saeed et al⁴ previously demonstrated the use of hs-TnT, N-terminal pro–B-type natriuretic peptide, and high-sensitivity C-reactive protein greatly augmented CVD risk prediction in older adults. However, it is also important to recognize when using a combination of biomarkers in risk prediction that simply increasing the number of biomarkers may yield diminishing returns,⁵ especially in younger populations for whom traditional risk factors used in the PCE have very good predictive value. The most effective approach will likely be selecting biomarkers that reflect distinct pathways of pathogenicity.