Prognostic implications of a causal endotype–based classification of myocardial infarction

Background and Aims The Fourth Universal Definition of Myocardial Infarction (UDMI) distinguishes type 1 from type 2 myocardial infarction (MI) but does not fully capture the heterogeneity of underlying mechanisms and prognosis. A causal endotype–based classification of MI has been proposed to better reflect pathophysiology and clinical risk. We evaluated the prognostic performance of this classification and its association with 1-year clinical outcomes. Methods Consecutive patients from the prospective, multicenter AMIPE registry with an adjudicated MI between 2017 and 2023 and ≥1-year follow-up were included. Patients were classified by primary etiologic cause into cardiac/coronary, cardiac/non-coronary, systemic, or indeterminate groups. The primary outcome was 1-year all-cause death; secondary outcomes included major adverse cardiovascular events (MACE), cardiovascular death, and recurrent MI. Cox and Fine–Gray competing-risk models were used, with cardiac/coronary MI as the reference. A sensitivity analysis compared atherothrombotic and non-atherothrombotic mechanisms within the cardiac/coronary endotype. Results Among 6282 patients, 5330 (84.9%) had cardiac/coronary MI, 302 (4.8%) cardiac/non-coronary MI, 503 (8.0%) systemic MI, and 147 (2.3%) indeterminate MI. 1-year all-cause mortality occurred in 9.8%, 15.9%, 25.8%, and 1.4% of patients, respectively. Compared with cardiac/coronary MI, adjusted mortality risk was higher in systemic MI (HR: 2.50; 95% CI: 2.05–3.05) and cardiac/non-coronary MI (HR: 1.46; 95% CI: 1.08–1.96), and lower in indeterminate MI (HR: 0.22; 95% CI: 0.06–0.89). Mortality differences were largely driven by non-cardiovascular death. Recurrent MI risk was lower in cardiac/non-coronary and systemic MI, while MACE rates were similar across endotypes. Atherothrombotic and non-atherothrombotic coronary mechanisms showed comparable outcomes. Conclusions A causal endotype–based classification identifies distinct MI groups with meaningful prognostic differences and competing risks, offering incremental insight beyond the current UDMI.