ABC | Volume 115, Nº2, August 2020

Original Article Viana et al. ACS prognosis based on models and outcomes Arq Bras Cardiol. 2020; 115(2):219-225 with at least one of the following characteristics: 1) positive myocardial necrosis marker, defined by troponin T ≥ 0.01 μg / L or troponin I ≥ 0.034 μg / L, which corresponds to values​ above the 99th percentile; 7,8 2) ischemic electrocardiographic changes, consisting of T wave inversion (≥ 0.1 mV) or ST segment deviation (≥ 0, 05 mV); 3) documented coronary artery disease, defined by a history of myocardial infarction or previous angiography showing coronary obstruction ≥ 50% of the luminal diameter. In addition, for inclusion in the analysis, patients needed to have undergone a coronary angiography procedure during hospitalization. Those who disagreed with participating in the registry and / or who had previously undergone surgical myocardial revascularization procedures were excluded. The protocol is in accordance with the Declaration of Helsinki, and was released by the Institutional Research Ethics Committees. All patients signed an informed consent form. Predicative Scores (SYNTAX e GRACE) The SYNTAX score was calculated by an experienced interventional cardiologist who was blind to the clinical picture and the outcomes. This physician assessed each obstruction of the coronary tree with an obstruction percentage ≥ 50% in vessels with a diameter ≥ 1.5 mm, following the SYNTAX score tutorial 9 and considering various angiographic parameters. The GRACE score was calculated on patient admission, consisting of eight variables: five of them computed in a semi-quantitative way, that is, different weight for each age group, systolic blood pressure, heart rate, plasma creatinine and Killip class; three of them are computed in a dichotomous way, with the ST segment elevation, elevation of myocardial necrosis marker and cardiac arrest at admission. 10 Hospital Clinical Outcomes The scores were tested in relation to the prediction of two types of hospital outcomes with different connotations: (1) non-fatal recurrent coronary outcomes (infarction, re- infarction or refractory angina), which represent the complexity of the coronary instability process; (2) cardiovascular death, which represents the body’s inability to adapt to the ischemic myocardial event. Nonfatal infarction was recorded as a consistent elevation of troponin T or I, above the previously described limits, in patients whose values ​were negative in the first 24 hours. For patients with infarction on admission, a new peak of CK-MB (≥ 50% of the previous value and above the normal value) was necessary to define a re-infarction. Elevation of necrosis markers related to percutaneous procedure or revascularization surgery were not recorded as a new event. Refractory angina was defined as recurrent chest pain, with at least two episodes, despite the use of nitrate and double product control. Cardiovascular death was defined as sudden death or cardiovascular hospitalization followed by death. Data Analysis Categorical variables were expressed as a percentage. Numerical variables were expressed as mean and standard deviation or median and interquartile range in cases of significant leak to normal distribution. The normality of the variables was analyzed using the Kolmogorov-Smirnov statistical test. Numerical variables were compared with the unpaired Student’s t-test or Mann-Whitney and categorical with the chi-square test or Fisher’s exact test. Receiver Operating Characteristic (ROC) curves were constructed for the GRACE and SYNTAX score values to predict the outcomes of non-fatal recurrent events and cardiovascular death, with the areas below the curve (C-statistic) compared by the Hanley-McNeil test. Statistical significance was defined by p <0.05. SPSS Statistical Software (version 21.0, SPSS Inc., Chicago, Illinois, USA) and MedCalc Software (version 12.3.0.0, Mariakerke, Belgium) were used for data analysis. Sample Size Calculation The sample was sized to offer statistical power for the comparison of SYNTAX C-statistics versus GRACE: to obtain 80% statistical power (one-tailed alpha of 0.05) in detecting 0.05 superiority of C-statistics (for example, 0.65 versus 0.70), it would be necessary to include 192 patients in the analysis. Results Sample Description During the study, 822 patients were included in the register, wherein 370 were submitted to coronarography procedure, while 05 were excluded once they already have previous revascularization surgery. Of the 365 patients analyzed, the media of age was 64 ± 14 years old, 58% were male, 19% of whom had ST-segment elevation myocardial infarction. Coronary Disease with tri-arterial or left main coronary involvement was present in 36% of the sample. The median of SYNTAX Score was 9 (IQR = 2,5 - 20) and GRACE’s 117 (IQR = 90 -144). Analyzing the risk terciles predicted in the SYNTAX Score, 11 81% of patients had a low value (0 to 22), 10% had an intermediate value (23 to 32) and only 8.5% had high value (≥ 33). Regarding the GRACE score, 10 44% had low-risk (<109), 28% had intermediate-risk (110 to 139) and 29% had high-risk (≥ 140). The incidence of cardiovascular death during hospitalization was 4.4% (16 patients) and non-fatal ischemic events were 10.7% (39 patients). Other clinical characteristics are described in Table 1. Outcome Predictions by Score For death outcome, SYNTAX and GRACE scores presented discriminatory capacity, with similar c-statistic: 0,80 (95% CI: 0,70 – 0,92) and 0,89 (95% CI 0,81 – 0,96), respectively — p = 0,019 — Figure 1A. When the scores were divided into risk terciles, both scores showed an increase of mortality in the third tercile: respectively, 2,4%, 2,7% and 30% on SYNTAX (P < 0,001) and 0%, 0,9% and 12% on GRACE (p < 0,001) — Figure 2, panel A and B. Regarding non-fatal recurrent events, SYNTAX score presented a predictor value (C-statistic = 0,64; 95% CI 0,55 -0,73). However, the GRACE score showed no association 220