Prognostic value of day-by-day troponin-T dynamics in acute myocardial infarction: a mixed-effects model analysis

Background. . Modeling daily trajectories of high-sensitivity cardiac troponin T (hs-cTnT) in patients during the acute phase of myocardial infarction (MI) may better reflect infarct size and reperfusion success than single-point measurements. However, its association with in-hospital outcomes remains insufficiently explored. Purpose – To evaluate the relationship between daily hs-cTnT trajectories, time-to-peak, and clinical outcomes during the hospital phase in patients with acute MI. Materials and Methods. This single-center study included patients with acute myocardial infarction (AMI, n = 191), in whom high-sensitivity cardiac troponin T (hs-cTnT) levels were measured every 24 hours during hospitalization. Cases without biochemical evidence of myocardial necrosis were not included. The reperfusion strategy for the entire cohort was uniform: early percutaneous coronary intervention performed within 1 hour from hospital admission. Measurements were performed using the Roche Elecsys s100 analyzer (high-sensitivity hs-cTnT assay). Values exceeding the upper measurement limit of the test (26 ng/ml) were treated as the maximal measurable value (26 ng/ml) for statistical analysis. Trajectory analysis was performed using linear mixed-effects models with a random intercept for each patient and fixed effects for «time since individual peak (day 1)», «binary event» (in-hospital death, invasive mechanical ventilation, reduced left-ventricular ejection fraction, reduced cardiac index, multivessel disease), and the interaction «time × event». Separate models were fitted for each clinical event. Secondary analyses included grouping by day of peak (1/2/3/4; subset n = 154) and tertiles of Δhs-cTnT (difference between peak and lowest post-peak value before discharge). All analyses were conducted in R version 4.4.2. Results. . The median age of patients was 64.0 (55.0–71.0) years; most were men (83.2%), and multivessel coronary disease was present in 64.4%. During hospitalization, hs-cTnT concentrations significantly decreased over time (β = −1.24; SE 0.10; t = −12.32; p<0.001). However, the decline was markedly slower in patients who died (time × death interaction: β = +1.23; SE 0.34; t = 3.68; p<0.001). In other events – including invasive mechanical ventilation, reduced ejection fraction, decreased cardiac index, or multivessel disease – the rate of troponin decrease demonstrated patterns that should be confirmed in future studies. The troponin peak most commonly occurred on the 2nd day (59.1%). Patients whose peak occurred on the 1st or 3rd day tended to show a gradual decline in glomerular filtration rate (eGFR), possibly reflecting concomitant impairment of renal perfusion. Analysis by Δhs-cTnT tertiles showed that lower Δhs-cTnT values were associated with longer hospital stays, more advanced stages of cardiogenic shock (according to the SCAI classification), and lower final left-ventricular ejection fraction (p=0.02). Therefore, considering the interaction trends with most clinical events, the pattern of post-peak hs-cTnT dynamics proved decisive. A slower post-peak troponin decline was associated with higher in-hospital mortality, whereas a faster decline was associated with more stable left-ventricular ejection fraction and cardiac index. Conclusions. Among patients with acute myocardial infarction, a slower post-peak decline in hs-cTnT was associated with an increased risk of in-hospital mortality, while a faster decline was associated with more favorable hemodynamic profiles. Time-to-peak and Δhs-cTnT tertiles may serve as potential risk markers and warrant further validation in larger cohort studies.