Abstract
The clinical use of advanced imaging modalities for early determination of infarct size and prognosis is limited. As a specific indicator of myocardial necrosis, cardiac troponin T (cTnT) can be used as a surrogate measure for this purpose. The present study sought to investigate the use of peak and serial 6-hour fixed-time high-sensitive (hs) cTnT for estimation of infarct size, left ventricular (LV) function, and prognosis in consecutive patients with ST-segment elevation myocardial infarction. The infarct size was expressed as the 48-hour cumulative creatine kinase release. LV function at 3 months was assessed using the echocardiographic wall motion score index and LV ejection fraction using radionuclide ventriculography. Adverse outcomes, comprising all-cause death, implantable cardioverter-defibrillator implantation, or hospitalization for heart failure, were recorded at 1 year of follow-up. In 188 patients, the peak and all fixed-time values correlated significantly with the 48-hour cumulative creatine kinase release, wall motion score index, and LV ejection fraction. The hs-cTnT value at 24 hours demonstrated the greatest correlation (r = 0.86, r = 0.47, and r = −0.59, respectively; p <0.001 for all). In the multivariate regression models adjusted for the clinical parameters, almost all were independently associated with the 48-hour cumulative creatine kinase release, wall motion score index, and LV ejection fraction, with the hs-cTnT value at 24 hours having the largest effect. Moreover, all cTnT values independently predicted adverse outcomes, again, with the hs-cTnT value at 24 hours showing the largest influence (hazard ratio 3.77, 95% confidence interval 2.12 to 6.73, p <0.001). In conclusion, not only peak, but all fixed-time hs-cTnT values were associated with infarct size, LV function at 3 months, and adverse outcomes 1 year after ST-segment elevation myocardial infarction. The value 24 hours after the onset of symptoms had the closest associations with all outcomes. Therefore, serial sampling for a peak value might be redundant.