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 Table 1 – Population characteristics of patients with and without cardiovascular events at 30 days Population (n = 519) MACE (n = 224) No MACE (n = 295) Average age (%) 64.31 (12.11%) 66.9 (11.69%) 62.3 (13.7%) Male sex, n (%) 291 (56.06%) 207 (59.5%) 84 (40.5%) Without cardiovascular risk factors (BMI > 30, smoker, DM2, family coronary artery disease, age < 55 years, hypertension, hypercholesterolemia) 64 (12.3%) 40 (40%) 24 (60%) 1 – 2 risk factors 348 (67%) 247 (70.97%) 101 (29.02%) 3 or more risk factors 98 (18.8%) 74 (75.5%) 24 (24.5%) Previous coronary heart disease as the only factor 84 (16.1%) 61 (72.6%) 23 (27.3%) BMI: body mass index; DM2: diabetes mellitus type 2. Table 2 – Occurrence of MACE (AMI, percutaneous revascularization, surgical revascularization, or death) according to risk groups HEART score Patients (n) MACE (n) MACE (%) Low (0 - 3) 194 6 3.1 Intermediate (4 – 6) 182 84 46.2 High (7 – 10) 143 134 93.7 TIMI score Low (1 – 2) 336 21 10.1 Intermediate (3 – 4) 130 119 55.6 High (5 – 7) 53 84 86.6 GRACE score Low (< 88) 183 65 22.2 Intermediate (89 – 118) 165 88 60.7 High (> 118) 171 71 87.7 * Total MACE = 351. † MACE per patient: 351 MACE / 224 patients = 1.56 MACE / patient. Table 3 – Operative characteristics for the HEART, TIMI, and GRACE scores HEART score ≤ 3 (CI 95%) TIMI score ≤ 2 (CI 95%) GRACE score ≤ 108 (CI 95%) SENS 99.5% (97 – 99.9) 90% (86 - 94) 70.9% (64.5 – 76.8) SPEC 36.6% (31.1 – 42.4) 63% (57.5 – 68.8) 77.2% (72 – 81.9) NPV 99% (95 - 99) 89.9% (85 – 91.9) 77.8% (72.3 – 82.3) PPV 54% (48 - 97) 65.2% (59.6 – 75.6) 70.3% (64 - 76) LR (+) 1.57 (1.4 – 1.7) 2.47 (2.1 – 2.8) 3.125 (2.4 – 3.9) LR (−) 0.012 (0.001 – 0.08) 0.147 (0.09 – 0.22) 0.375 (0.3 – 0.4) * SENS: sensitivity; SPEC: specificity; NPV: negative predictive value; PPV: positive predictive value; LR: likelihood ratio. The use of hsTnI maintained the previously demonstrated superior performance of the HEART score compared to the TIMI and GRACE scores. The HEART score allows for more accurate differentiation of patients with low risks of presenting major cardiovascular events, which will enable physicians to opt for earlier discharge and which may allow savings in hours of in-hospital stay and unnecessary diagnostic tests. This could lead to better care for patients and more efficient distribution of healthcare system resources. Author contributions Conception and design of the research: Torralba F, Navarro A, Ortiz C; Acquisition of data: Torralba F, Navarro A, Botero A, Alarcón F; Analysis and interpretation of the data: Torralba F, Castellanos JC, Botero A, Alarcón F, Isaza N, Isaza D; 799