| Description: |
Background Despite well-established risk factors for myocardial infarction (MI) such as hypertension, smoking, hyperlipidemia and diabetes, long-term risk stratification remains suboptimal. Hematological biomarkers have emerged as potential predictors of major adverse cardiovascular events (MACE), but their true long-term prognostic value remains unclear. Purpose This study aims to assess the long-term prognostic value of hematological biomarkers in MI patients using a large cardiovascular registry. By accounting for traditional risk factors as confounders, we aim to determine whether routine blood parameters contribute independently to risk stratification beyond established predictors. Methods This retrospective cohort study analyzed data from the Styrian Registry on Genuine Myocardial Infarction (STRONG-MI). It included all MI patients who underwent invasive coronary angiography between January 2007 and March 2016, in Austria. Multivariable Cox regression models adjusted for age, sex, smoking, BMI, hyperlipidemia, MI characteristics, previous stroke, diabetes, renal function and cumulative comorbidity burden were used to assess the association between hematological parameters and 3-point MACE over a median follow-up of 85 months (interquartile range 13–128). Results A total of 10,920 MI patients were included, with a mean age of 66.6 (standard deviation 12.7), 65% of whom were men. Non-ST-Elevation Myocardial Infarction (NSTEMI) was more prevalent than ST-Elevation Myocardial Infarction (STEMI) (62% vs. 38%). A U-shaped association was observed between Hb levels and 3-point MACE, with a 10% and 8% increased risk in the lowest and highest Hb tertiles at admission (adjusted hazard ratio [AHR] 1.10; p = 0.006; AHR 1.08, p = 0.035, respectively). Similarly, patients in the lowest tertiles of erythrocyte count, mean corpuscular hemoglobin (MCH), and mean corpuscular hemoglobin concentration (MCHC) had an increased risk of 3-point MACE (AHR 1.08, p = 0.033; AHR 1.08, p = 0.026; AHR 1.07, p = 0.037), whereas ... |