ABC | Volume 114, Nº5, May 2020

Original Article Torralba et al. HEART, TIMI, and GRACE scores for MACE prediction Arq Bras Cardiol. 2020; 114(5):795-802 in cardiovascular medicine, with a monthly average of 9,000 emergency consultations, 15% of which correspond to cardiovascular emergencies. Patients over 18 years of age who presented at the emergency department with acute chest pain between August 2017 and February 2018 were included in the study. According to the institutional protocol, patients were evaluated by the cardiologist; electrocardiography was performed, and high sensitivity troponin I (hsTnI) was measured, initially and 3 hours later if needed, using ARCHITECT STATassay (Abbot, Lake Bluff, IL, USA). Acute myocardial infarction (AMI) was diagnosed when hsTnI values were greater than 0.026 ng/mL (reference value 0.0 – 0.026 ng/ml). In this case, patients were admitted for in‑hospital care, coronary arteriography, and either percutaneous or surgical revascularization. When values were negative, but pain was considered of intermediate or high probability, the patients were admitted for further evaluation with a non-invasive stratification strategy. Patients with myocardial infarction with ST elevation and non-cardiac causes of chest pain, such as pneumonia, trauma, or psychogenic pain, were excluded from this study. The HEART, TIMI, and GRACE risk scores were calculated for this group of patients at the time of consultation. Risk Scores The methods for calculating the GRACE, TIMI, and HEART scores have been described in previous articles and are briefly summarized below. 14,16,17 The score calculations were performed using the information documented in the electronic medical record, the first electrocardiogram upon presentation, and the first laboratory values measured, including troponin measurement with the hsTnI assay. The HEART score consists of the following 5 categorical variables: the patient’s medical history, electrocardiogram, age, risk factors for coronary heart disease, and troponin. Each variable has a maximum value of 2 points adding up to a maximum score of 10, which indicates a patient with maximum risk. The GRACE score consists of the following 5 categorical variables: age, heart rate, blood pressure, creatinine, and Killip class; and the following 3 nominal variables: cardiac arrest, ST-segment deviation, and troponin elevation. Each item is assigned a value, and the sum of these values determines the risk of MACE. Finally, the TIMI score consists of the following 7 dichotomous nominal variables: age over 65 years, more than 3 risk factors for coronary artery disease, significant coronary artery stenosis, symptoms of severe angina, ST-segment deviation, use of aspirin in the last week, and elevation of troponin. The maximum TIMI score is 7, with higher scores indicating higher risks. Ethics The study was carried out in accordance with the principles laid out in the Declaration of Helsinki of the World Medical Association, the Nuremberg Code, and the World Health Organization International Ethical Guidelines for research involving humans, as well as domestic regulations related to basic health care. The study received approval from the ethics and research committee of the Fundación Cardioinfantil, and the patients included in the study provided their informed consent. Data management The data included demographic information of the patients, as well as information on clinical history, laboratory values, electrocardiographic findings, and vital signs. Laboratory values included creatinine and hsTnI, which are the assays currently applied in the institutional protocol. The attending cardiologist evaluated the 12-lead electrocardiogram according to the guidelines of the American Heart Association. 18 If necessary, the exam was submitted to a second blinded attending cardiologist for evaluation. An encrypted database was created to which only the authors of the study had access, and an algorithm was developed for the automatic calculation of the risk scores. Follow-up Follow-up was performed at 30 days, reviewing the electronic medical record and employing a telephone survey. A structured format was applied with 4 clear questions regarding the occurrence of major cardiovascular events (death, myocardial infarction, surgical revascularization, or percutaneous revascularization), to determine the presence of the primary outcome. Outcomes The diagnosis of AMI was made when troponin values rose above the 99th percentile of reference values (hsTnI > 0.026 ng/mL), and evidence of myocardial ischemia was documented on electrocardiogram. We applied the criteria described by the third universal definition of AMI, which was valid at the time the protocol was written. 19 We also defined ST-elevation myocardial infarction (STEMI), non-ST-elevation myocardial infarction (NSTEMI), and signs of ischemia according to the guidelines validated at the time of protocol design. 20 Percutaneous revascularization was defined as any intervention through a catheter in the coronary arteries, and surgical revascularization was defined as any cardiac surgery in which coronary artery grafts were made. MACE were defined as death from any cause, myocardial infarction, and surgical or percutaneous myocardial revascularization. Follow up was completed 30 days after admission to the emergency department. Statistical analysis We calculated a sample of 550 patients, to obtain 185MACE using Simel and Samsa’s method of maximum sensitivity, 21 in order to yield a power of 80% and a confidence interval of 95% with an alpha error of 5%. For each score, the best cutoff point was calculated using the Youden index, 22 considering p values of 0.05 significant . Subsequently, the C statistic, the positive and negative likelihood ratio (LR), sensitivity, and specificity were calculated. Then, the LR was calculated for each risk stratum. The difference between LR was calculated using the test for adequate binomial proportions (Chi-square test and Fisher’s exact test), considering p values of 0.05 significant . 796