Abstract

Wiviott, S. D., Morrow, D. A., Frederick, P. D., Giugliano, R. P., Gibson, C. M., McCabe, C. H., Cannon, C. P., Antman, E. M., Braunwald, E. Performance of the thrombolysis in myocardial infarction risk index in the National Registry of Myocardial Infarction-3 and -4: a simple index that predicts mortality in ST-segment elevation myocardial infarction J Am Coll Cardiol. 2004;44(4):783-9.

OBJECTIVES: We sought to evaluate a simple risk index based on age and vital signs in a community sample of patients with ST-segment elevation myocardial infarction (STEMI). BACKGROUND: A simple risk index based on age and vital signs (heart rate x [age/10](2)/systolic blood pressure) developed from patients with STEMI accurately predicts mortality in clinical trials of fibrinolysis. The application of such a tool in an unselected population is necessary to evaluate its utility in clinical practice.

METHODS: To evaluate the Thrombolysis In Myocardial Infarction (TIMI) risk index for routine practice, we tested it in the National Registry of Myocardial Infarction (NRMI)-3 and -4. The risk index was evaluated as a continuous variable in patients with STEMI from NRMI and in subgroups based on age and reperfusion status.

RESULTS: A total of 153,486 patients with STEMI were eligible. As anticipated, STEMI patients in NRMI had a higher risk index profile, as compared with those in the clinical trial (median 26.9 vs. 20, p < 0.0001). Classification of NRMI patients with STEMI into risk groups revealed a significant graded relationship with mortality (0.9% to53.2%, p(trend) < 0.0001, c statistic 0.79). The discriminatory capacity of the risk index was particularly strong in the 81,679 patients receiving reperfusion therapy (0.6% to60%, p(trend) < 0.0001, c statistic 0.81). For the 71,807 patients not receiving reperfusion therapy, a strong graded relationship remained (1.9% to 52.2%, p(trend) < 0.0001, c statistic 0.71). Among the elderly, although the distribution of scores was shifted toward higher risk, the performance remained (0% to 53.1%, p(trend)< 0.0001, c statistic 0.71).

CONCLUSIONS: A simple risk index from baseline clinical variables routinely obtained at the first patient encounter predicted mortality in a large unselected heterogeneous group of patients with STEMI.

Trial: