ABC | Volume 114, Nº6, June 2020

Special Article Barros and Silva et al. Brazilian Registry of Acute Coronary Syndromes Arq Bras Cardiol. 2020; 114(6):995-1003 using the variance analysis ANOVA. Categorical variables were described by absolute and relative frequencies. The proportions were compared using the Chi-square test or Fisher’s exact test. Generalized Estimating Equations (GEE) models were used to assess drug therapy over time. To compare the major cardiovascular events according with the final diagnosis, Cox proportional hazards model and Kaplan-Meier curves were used. The identification of independent predictors for the composite endpoint (cerebrovascular accident (CVA), reinfarction and death) was performed using Cox proportional hazards model with the final diagnosis and the baseline factor analyzed. This analysis of predictors was initially performed in a univariate fashion and variables with a p <0.15 were included in the multivariate analysis. P-values were presented as two-sided and p<0.05 was considered statistically significant in the final analyses. Additionally, in the multivariate analysis, an interaction test was performed between the selected variables. All analyses were performed using R Statistical software, version 3.6.1. Results Between August 2010 and April 2014, 5,047 patients were recruited from this nationwide registry, 265 of whom (5.25%) had undiagnosed chest pain and were excluded from the clinical follow-upbecause they didnot fulfill the research inclusion criteria. Thus, 4,782 ACS patients were actually included in the analysis and followed in this prospective registry, in 53 hospitals from the 5 Brazilian federal regions. In a total of 410 patients (8.6%), it was not possible to obtain the final 12-month data. Baseline Characteristics The patients’ clinical profile revealed the inclusion of approximately 7 0%of patients diagnosed with AMI at admission, almost one-third had diabetes, and around 90% presented at least one risk factor, with themost frequent being systemic arterial hypertension (Table 1). Medical prescription adherence to evidence-based therapies The prescription adopted soon after admission shows that full adherence to medications currently recommended in the guidelines was of 62.1 % (Table 2). This adherence includes dual antiplatelet therapy (aspirin/P2Y12 inhibitor) combined with parenteral anticoagulants, statins and betablockers. Out of the 1,714 patients presented with AMI (STEMI), 1,412 (82.4%) individuals were treated with some modality of reperfusion of the myocardium (either fibrinolysis or primary percutaneous coronary intervention). When analyzing the prescription of reperfusion therapies for AMI, there are distinct and decreasing percentages, according to the Brazilian federal region: 87.3%, 84.5%, 72.8%, 66.7% and 65.7%, (p < 0.001), for the South, Southeast, Northeastern, Midwest, and Northern Brazilian states, respectively. As the severity of the clinical presentation of these three components of the ACS increased, there was a progressive increase in the prescription of invasive strategies, either coronary angiography (68.0%, 83.1% and 90.4%; p < 0.001), or myocardial revascularization procedures (38.2%, 54.4% and 76.4%; p < 0.001), in case of unstable angina, NSTEMI and STEMI, respectively. The preferred revascularization procedure in these patients was percutaneous coronary intervention with rates > 95% of coronary stent use in patients treated percutaneously. The percentage of percutaneous revascularization among all ACS patients ranged according with the diagnosis: unstable angina, NSTEMI and STEMI (33.6%, 47.4% and 75.1%, respectively; p <0.001). We observed that the prescription of a P2Y12 inhibitor at hospital discharge varied according with the type of ACS (66.4% for unstable angina, 77.7% for NSTEMI and 90.9% for STEMI; p<0.001), and type of coronary disease treatment the patient received (PCI (94.2%), surgical (25%) or clinical (66.2%); p <0.001). The evolution of the main therapies, from admission to discharge, at the end of 30 days and in 6 and 12 months shows a progressive reduction in the use of the therapies recommended, especially in relation to therapy with the use of P2Y12 receptor inhibitors (Figure 1). Clinical Outcomes Clinical outcomes were measured cumulatively at the end of the first 12 months of evolution (Figure 2). Among patients with UA, there was no association between the occurrence of the composite events (mortality, reinfarction or cerebrovascular accident (CVA)) at the end of the first 12 months and the performance of myocardial revascularization procedure (Table 3). In the presence of NSTEMI, a significant reduction was observed in the incidence of major cardiovascular events, including cardiovascular mortality, among those submitted or not to myocardial revascularization (mortality = 6.29 per 100 people/year versus 12.06 per 100 people/year; p < 0.001 and major cardiovascular outcomes = 13.18 per 100 people/year versus 17.96 per 100 patients/year; p = 0.038), respectively. STEMI patients had a significant reduction in mortality rates and incidence rates of major cardiovascular events when submitted to myocardial revascularization (mortality = 8.02 per 100 people/year versus 18.54 per 100 people/year; p < 0.001 and cardiovascular events = 13.11 per 100 people/year 21.69 per 100 people/year; p<0.001). In themultivariate analysis (Table 4), the following factors were associatedwith the occurrence of major cardiovascular events: age, public health care, AMI, CVA, renal failure, diabetes and quality of therapy (complete or not). There was no significant interaction between the covariables. The rate of events among SUS patients was 16.6 per 100 patients/year, whereas in the private associated network it was 9.10 per 100 patients/year (p<0.01). In the analysis per federal region, the 1-year death rate was significantly higher in the Northern region (19.8%; CI95% 12.6-27.0), followed by the Southeast (8.0%; CI95% 7.0-9.1), South (6.8%; CI95% 4.8-8.7) and Northeast regions (5.6%; CI95% 3.7-7.5). The Midwest region had the lowest number of patients with intermediate mortality rate between the Northern region and the rest of Brazil (14.2%; CI95% 2.8-25.5). When comparing the predictors of events between the North region and the 3 regions with the lowest rates of events (South, Southeast and Northeast), we observed a greater incidence of STEMI (51.0% x 35.3%; p <0.01), SUS health care (100% x 51.8%; p < 0.01) and incomplete treatment among the patients from the North region of Brazil (47.9% x 37.2%; p < 0.01). 997