ABC | Volume 112, Nº3, March 2019

Original Article Castro et al Troponin I after acute coronary syndrome Arq Bras Cardiol. 2019; 112(3):230-237 in the present study (median second collection interval: 39 days after the index event; interquartile range: 33-50 days). The initial diagnosis was STEMI in 144 (27.4%) patients, NSTEMI in 230 (43.8%), and UA in 151 (28.8%). The mean age was 61.6 years and 60.2%were men. The main cardiovascular risk factors found in this population were hypertension (75.5%), current or previous smoking (66.9%), and sedentary lifestyle (69.9%). Cutoff values for hs-cTnI among the tertiles were: < 0.012, 0.013-0.023 and > 0.023 mcg/L. Most hs-cTnI samples collected from patients in the subacute phase after ACS were below the 99th percentile of the method (83.8%). Patients were followed for a median of 51 months; baseline characteristics of the study according to hs-cTnI tertiles are shown in Table 1. Individuals in the highest tertile were more likely to be male, have STEMI or NSTEMI diagnoses at the index event, and chronic kidney disease and sedentary lifestyle at baseline. From the data collected after patient admission, lower glomerular filtration rate (calculated according to CKD-Epi) and lower ejection fraction estimated in echocardiography were strongly correlated with persistently elevated levels of hs-cTnI (p < 0.001). Figure 1 shows the Kaplan-Meier curves for cumulative survival according to each hs-cTnI tertile during follow-up. We found significantly lower survival rates in individuals in the highest tertile (p < 0.001). Analyses evaluating cardiovascular mortality as the main outcome showed similar findings (Figure 2). Table 2 shows the results of the Cox regression analyses. Participants in the third tertile of troponin, using the first tertile as the reference, presented a hazard ratio (HR) of 4.14 (95% Confidence Interval [95% CI]: 2.19-7.86) for all‑cause mortality after adjustment for age and sex; this effect persisted in multivariate adjustment models 1 (HR: 3.84, 95% CI: 1.92- 8.12) and 2 (HR: 6.53, 95% CI: 2.12-20.14). For cardiovascular mortality, there were significant differences between the first and the third tertiles after adjustment for age and sex (HR: 5.65, 95% CI: 1.94-16.47), and both multivariate adjustment models 1 (HR: 4.90, 95% CI: 1.35-17.82) and 2 (HR: 5.89, 95% CI: 1.08-32.27). Discussion In this cohort of patients with hs-cTnI levels measured 25 to 90 days after an ACS event, participants had sociodemographic characteristics and cardiovascular comorbidities similar to that of large international registries, like the Global Registry of Acute Coronary Events (GRACE). 18 As in this registry, our cohort had a predominantly male population with a high prevalence of hypertension; other cardiovascular risk factors, such as heart failure and smoking history, were more prevalent in our study. The most frequent ACS type in our study was NSTEMI (41.5% of participants), which is also consistent with the current trend in the incidence of MI, 19 although contrasting with the smaller frequency of NSTEMI in the GRACE cohort (26% of participants). 18 Medication use on the first follow-up visit was similar to the treatment received on discharge by participants in the Brazilian Registry on Acute Coronary Syndromes (BRACE) study, which included hospitals of all regions of Brazil. 20 When we evaluated the percentage of patients receiving each therapeutic group, we found clear similarities between our study and BRACE for the use of aspirin (83.6% vs 86.0%, respectively), clopidogrel (53.0% vs 50.1%), betablockers (64.2% vs 69.8%), ACE inhibitors/angiotensin receptor blockers (68.3% vs 70.6%) and statins (76.4% vs 82.7%). These data also show that adherence to guideline-recommended therapies was still not optimal by the time of enrollment of these participants. Most patients in our study (83.8%) had hs-cTnI levels below the 99th percentile during the subacute phase after an ACS event; nevertheless, even at this range, those in the highest tertile had a greater hazard ratio for all-cause and cardiovascular mortality compared to the first tertile. Elevated levels of hs-cTnI at 25 to 90 days post-ACS remained an independent risk factor for all-cause and cardiovascular mortality after adjustment for multiple confounders. The mechanisms by which some patients present persistent elevations in cardiac troponin levels are not well established. Previous experimental studies demonstrated the incidence of chronic myocardial injury after induced mechanical coronary obstruction in rats; 21 accelerated apoptosis due to chronic myocardial dysfunction has also been shown in patients with heart failure. 22 Other speculated mechanisms include normal myocyte turnover, cellular release of proteolytic degradation products, higher myocyte cell wall permeability, and the formation of blebs in the cellular walls with the presence of these proteins. 23 The association between higher levels of cardiac troponins and worse outcomes in out- of-hospital settings has been reported by previous studies. In 2007, Eggers et al. 16 analyzed a cohort of patients with earlier-generation cardiac troponin I (cTnI) measured at 6 weeks, 3 and 6 months after an ACS event. Throughout this study, the subgroup of patients with permanently elevated levels of cTnI (≥ 0.01) had a greater probability of death during follow-up than patients with transiently elevated or negative cTnI. 16 In 2012, two studies addressed the prognostic role of high‑sensitivity cardiac troponin T (hs-cTnT) in the stabilized phase after a cardiac event. Ang et al. 13 followed 326 patients for a median of 30 months, after measurement of hs-cTnT 7 weeks post-ACS; after adjustment for age, ACS subtype, hypertension, type 2 diabetes, smoking, anemia, BNP, estimated GFR, and echocardiographic findings, hs‑cTnT remained a strong predictor of death and AMI during follow‑up. 1 Koenig et al. 24 studied 1050 patients for a median of 8.1 years after an ACS event or CABG, with hs‑cTnT levels measured approximately 43 days after the event; patients in the highest quartile were at increased risk for new cardiac events throughout the observation period. One study published in 2014 also addressed the prognostic role of hs-cTnI after an ACS episode. White et al. 25 followed 7,836 patients who had suffered an ACS event; after a median of 6 years follow-up, patients in the highest tertile were at increased risk for CAD death and MI. Compared to our study population, these studies followed patients with similar ages and estimated GFR, but with lower baseline frequency of hypertension, diabetes, and dyslipidemia. 232