BACKGROUND: The probability of adverse events estimate is crucial in acute coronary syndrome condition. OBJECTIVES: To develop a risk score for the brazilian population presenting non-ST-segment elevation acute coronary syndrome. METHODS: One thousand and twenty seven (1,027) patients were investigated prospectively at a cardiology center in Brazil. A multiple logistic regression model was developed to estimate death or (re)infarction risk within 30 days. Model predictive accuracy was determined by C statistic. RESULTS: Combined event occurred in 54 patients (5.3%). The score was created by the arithmetic sum of independent predictors points. Points were determined by corresponding probabilities of event occurrence. The following variables have been identified: age increase (0 to 9 points); diabetes mellitus history (2 points) or prior stroke (4 points); no previous use of angiotensin converting enzyme inhibitor (1 point); creatinine level increase (0 to 10 points); the combination of troponin I level increase and ST-segment depression (0 to 4 points). Four risk groups were defined: very low (up to 5 points); low (6 to 10 points ); intermediate (11 to 15 points ); high risk (16 to 30 points ). The C statistic was 0.78 for event probability, and 0.74 for risk score. CONCLUSION: A risk score of easy application in the emergency service was developed to predict death or (re)infarction within 30 days in a brazilian population with non-ST-segment elevation acute coronary syndrome.
Cardiovascular diseases; unstable angina; myocardial infarction; prognostic; risk factors
