Background

High-sensitivity Troponin T and I (hsTnT/I) concentration has been shown to associate with coronary artery disease (CAD) severity and long-term outcome. Here we aimed to explore whether hsTn blood levels add predictive information independent of CAD severity and further confounders in patients with angiographically graded CAD.

Methods

From 2015 to 2020, 3,012 patients undergoing coronary angiography were included in the observational Hamburg INTERCATH cohort. In 2,209 consecutive patients Troponin levels were quantified using hsTnT (Roche Diagnostics Elecsys) and hsTnI (Abbott Diagnostics ARCHITECT STAT) assays. Heart transplant recipients, patients with a suspected acute coronary syndrome, previous Coronary artery bypass grafting, and without angiographically documented CAD were excluded, leaving 1,487 patients for analyses. The Gensini score was used to grade CAD severity. All-cause mortality was defined as endpoint. Cumulative incidence curves using the Kaplan-Meier method were created and stratified by hsTnT/I quartiles. Multivariable Cox analyses were computed for the association of hsTnT/I with all-cause mortality adjusting for age, gender, arterial hypertension, hyperlipoproteinemia, smoking, diabetes mellitus, body-mass index, and Gensini Score.

Results

Mean age was 69.9±10.3 years (24.6% women). Mean Gensini score was 26.8±31.7. Patients were followed-up for a median of 4.35 years. During the follow-up period 354 (23.9%) patients died. For both hsTnT as well as hsTnI, quartiles differentiated all-cause mortality across all categories (Figure 1A and B). This association remained significant after multivariable analyses adjusting for classical cardiovascular risk factors and CAD severity after 36 months of follow-up for hsTnT (Figure 1C), across all quartiles. The association proved especially strong in comparison between the 1^st and 4^th Quartile (Hazard Ratio [HR] 2.7, Interquartile range [IQR] 2.0-3.5; p<0.001). This was also observed for hsTnI (Figure 1D), albeit with a HR of 2.0 in comparison between the highest and lowest Quartile (IQR 1.6, 2.5; p<0.001).
 

Conclusion

Both hsTnT and hsTnI were independently associated with all-cause mortality in an all-comers CAD cohort even after adjustment for CAD severity and classical cardiovascular risk factors in patients presenting with chronic coronary syndromes.