Long-term outcomes of time-dependent phenotypes of perioperative myocardial injury

Background: Perioperative myocardial injury (PMI) is associated with increased short-term major adverse cardio- and cerebrovascular events (MACCE) and mortality. We evaluated how different phenotypes of PMI based on timing of injury impact long-term MACCE and mortality among patients undergoing major noncardiac surgery (NCS). Methods: A prospective, observational study was carried out in 1290 patients aged ≥50 yr undergoing major noncardiac surgery. High-sensitivity cardiac troponin T (hs-cTnT) was measured before surgery and up to 3 days after surgery. Patients were classified into four groups: (1) no hs-cTnT elevation, (2) isolated preoperative hs-cTnT increases, (3) normal preoperative hs-cTnT with perioperative elevation, and (4) increased preoperative hs-cTnT with perioperative elevation. The main outcomes were MACCE and mortality up to 3 yr after operation. Results: At 3-yr follow-up, MACCE had occurred in 17.1%, 37.9%, 45.2%, and 50.7% and mortality was 21.4%, 30.3%, 30.1%, and 33.8% in groups 1-4, respectively. All PMI phenotypes were independently associated with MACCE. Patients with 'normal preoperative hs-cTnT with perioperative elevation' appeared to have the highest risk. In this phenotype, the risk was greatest at 30 days (adjusted odds ratio, 4.5; 95% confidence interval, 2.3-8.8) and persisted over 3 yr (adjusted odds ratio, 3.5; 95% confidence interval, 2.0-5.9). PMI was associated with increased mortality, but the relationship was not sustained after multivariable adjustment. Conclusions: Increased hs-cTnT based on timing identifies prognostically important subgroups. Perioperative increases, regardless of preoperative levels, are associated with a high risk of MACCE that is sustained up to 3 yr after surgery. This supports the use of time-based PMI phenotypes for prognostic enrichment studies to mitigate the risk of MACCE. Clinical trial registration: NCT03436238.