ABC | Volume 112, Nº3, March 2019

Original Article Elevated High-Sensitivity Troponin I in the Stabilized Phase after an Acute Coronary Syndrome Predicts All-Cause and Cardiovascular Mortality in a Highly Admixed Population: A 7-Year Cohort Leandro Teixeira de Castro, 1 Itamar de Souza Santos, 1 Alessandra C. Goulart, 1 Alexandre da Costa Pereira, 2 Henrique Lane Staniak, 1 Marcio Sommer Bittencourt, 1 Paulo Andrade Lotufo, 1 Isabela Martins Bensenor 1 Centro de Pesquisa Clínica e Epidemiológica da Universidade de São Paulo (USP), 1 São Paulo, SP – Brazil Instituto do Coração (InCor) - Faculdade de Medicina da Universidade de São Paulo, 2 São Paulo, SP – Brazil Mailing Address: Leandro Teixeira de Castro • Rua Oscar Freire, 2185 - Apto 54. Postal Code 05409-011, Pinheiros, São Paulo, SP – Brazil E-mail: leo920@gmail.com, leandro.castro@usp.br Manuscript received April 24, 2018, revised manuscript July 05, 2018, accepted August 02, 2018 DOI: 10.5935/abc.20180268 Abstract Background: High-sensitivity cardiac troponin I (hs-cTnI) has played an important role in the risk stratification of patients during the in-hospital phase of acute coronary syndrome (ACS), but few studies have determined its role as a long-term prognostic marker in the outpatient setting. Objective: To investigate the association between levels of hs-cTnI measured in the subacute phase after an ACS event and long-term prognosis in a highly admixed population. Methods: We measured levels of hs-cTnI in 525 patients 25 to 90 days after admission for an ACS event; these patients were then divided into tertiles according to hs-cTnI levels and followed for up to 7 years. We compared all-cause and cardiovascular mortality using Cox proportional hazards models and adopting a significance level of 5%. Results: After a median follow-up of 51 months, patients in the highest tertile had a greater hazard ratio (HR) for all-cause mortality after adjustment for age, sex, known cardiovascular risk factors, medication use, and demographic factors (HR: 3.84, 95%CI: 1.92-8.12). These findings persisted after further adjustment for estimated glomerular filtration rate<60ml/min/1.73m 2 and left ventricular ejection fraction < 0.40 (HR: 6.53, 95% CI: 2.12‑20.14). Cardiovascular mortality was significantly higher in the highest tertile after adjustment for age and sex (HR: 5.65, 95% CI: 1.94‑16.47) and both in the first (HR: 4.90, 95% CI: 1.35‑17.82) and second models of multivariate adjustment (HR: 5.89, 95% CI: 1.08-32.27). Conclusions: Elevated hs-cTnI levels measured in the stabilized phase after an ACS event are independent predictors of all‑cause and cardiovascular mortality in a highly admixed population. (Arq Bras Cardiol. 2019; 112(3):230-237) Keywords: Coronary Artery Disease / mortality; Troponin I; Prognosis; Metabolic Syndrome; Biological Variation, Population; Risk Factors. Introduction Acute coronary syndrome (ACS) is a major driver of mortality and the leading cause of years of life lost worldwide. 1 In recent decades, several therapeutic interventions have been proven beneficial in the treatment of ACS, and structured strategies for early diagnosis and appropriate treatment have been recommended by several cardiology societies. 2-5 Because of the progress made in therapeutics for ACS, a heterogeneous group of survivors from this condition has received long-term follow-up from medical services. The prognosis of patients in the stabilized phase after ACS varies widely; 6 validation of easily obtainable, low-cost prognostic markers may enhance long-term risk stratification in this population. Several studies showed cardiac troponins (cTns) to be more sensitive and specific for diagnosingmyocardial infarction, and to have greater correlation with higher mortality than the previous reference standard, creatine kinase isoenzyme MB (CK-MB). 7-11 Over the past two decades, new assays have been developed which conferred greater sensitivity to the diagnosis of myocardial infarction; these high-sensitivity cardiac troponins (hs-cTns) showed greater accuracy in discriminating patients at higher risk for death, even in those who had undetectable first-generation cTn levels. 12 More recently, the use of hs-cTnT as a prognostic marker in the subacute phase after an ACS episode has been studied in an European cohort of white patients. 13 These findings have not been replicated in more heterogeneous populations in the developing world. In this single-center observational cohort, we aimed to study the association of elevated levels of hs-cTnI with long-term all-cause and cardiovascular mortality in a highly admixed population in Brazil. Methods The Strategy of Registry of Acute Coronary Syndrome (ERICO) study design has been described in detail elsewhere. 14,15 In brief, ERICO is a prospective cohort 230