IJCS | Volume 33, Nº4, July and August 2020

387 Table 6 - Relationship between the hematological scoring system and the TIMI RISK score in predicting mortality among patients with non-ST-segment elevation myocardial infarction Scale of points TIMI RISK Total 0 to 3 4 to 7 ≥ 26 points 8 (19.0 %) 35 (35.7%) 43 < 26 points 34 (81.0 %) 63 (64.3%) 97 Total 42 98 140 Kappa coefficient = 0.162 (50.7% overall agreement). Patients with non-STEMI and STEMI were also classified as low risk (TIMI RISK 0 to 3 and Killip I and II) and high risk (TIMI RISK 4 to 7 and Killip III and IV), respectively, which facilitated the comparison with the hematological scoring system. In our sample, 70% of the patients had STEMI and 82.1% of these at high risk of mortality (Killip III and IV) had a score ≥ 26 points in the scoring system proposed. Also, 81% of the patients with non-STEMI at low risk of mortality (TIMI RISK 0 to 3) had a score < 26 points in the scoring system. Therefore, this newhematological scoring systemcould complement these extensively used risk scores in AMI patients. As mentioned before, themain purpose of this hematological scoring system is to improve clinical surveillance during hospitalization based on these laboratory variables, which would be of help in therapeutic decision making. This hematological scoring system is dynamic, and changes in the risk profile may reflect the response to a treatment proposed. In this study, the instrument showed an 89.1% probability of identifying the outcome among those who died in this population. However, the hematological scoring system had a low positive predictive value (26.8%), probably due to the effective treatment employed. In this sample, 70%of these patients had STEMI and of these 43.9% had ≥ 26 points in the scoring system, and 30.7% of patients with non-STEMI had ≥ 26 points in the scoring system. In the present study, total mortality was 11.8% (55 patients): 43/326 (13.2%) STEMI and 12/140 (8.6%) non-STEMI. Few studies have evaluated the performance of a scoring system including laboratory variables as a prognostic marker in AMI. Yanishi et al., 16 developed a simple stratification model using white blood cell count, hemoglobin, C-reactive protein, creatinine and blood sugar levels for predicting in-hospital mortality in STEMI (ROC curve of the derivation and validation in laboratory model of 0.81 and 0.74 respectively, p < 0.01). A recent study by Ibrahim et al., 17 proposed a scoring system using clinical variables (male sex and previous percutaneous coronary intervention) and four biomarkers (midkine, adiponectin, apolipoprotein C-I, and kidney injury molecule-1) to predict with high accuracy the presence of obstructive coronary artery disease andmortality. In this study, elevated scores were predictive of ≥ 70% stenosis in all subjects (OR: 9.74; p < 0.01). At optimal cut-off, the score had 77% sensitivity, 84% specificity, and a positive predictive value of 90% for ≥ 70% stenosis. In another recent publication, Gerber et al., 18 demonstrated the importance of risk stratification for informed decision in clinical care. The present study has some limitations. First, patients were selected in a single center. Second, heparin could inhibit platelet aggregation, but not platelet size. However, we increased the sample size, and used standardized and predetermined protocol to minimize possible bias. Conclusions The proposed hematological scoring system is a surveillance tool based on laboratory data, shown to be associated with in-hospital mortality in AMI patients. This simple and low-cost tool can be used to assess inflammation and hypoxemia caused by in-hospital complications using complete blood count parameters measured by an automated method. In addition, the scoring system is easy to use and interpret by all the multidisciplinary team members and can be calculated in the laboratory. Further studies would help to confirm the usefulness and importance of this scoring system based on hematological laboratory parameters for clinical surveillance of inpatients with AMI. Acknowledgements The authors thank the PROCAPE staff for their constant support during the course this work. Author contributions Conception and design of the research: Monteiro Júnior JGM, Torres DOC Silva MCFC. Acquisition of Monteiro Júnior et al. Hematological scoring system in AMI Int J Cardiovasc Sci. 2020; 33(4):380-388 Original Article