Little is known about the prognostic value of leukocyte count on admission for patients with chest pain. In total, 1,461 patients who presented to the emergency department with non-ST-segment elevation chest pain were studied by clinical history, electrocardiography, serial troponin I determination, and leukocyte count on admission. End points were 1-year mortality and major events (mortality or infarction). Overall patient distribution by quartiles of leukocyte count showed increased mortality (6%, 7%, 6%, and 17%, p = 0.0001) and major events (13%, 13%, 15%, and 24%, p = 0.0001) in the fourth quartile. After adjustment for other risk factors, the fourth quartile cut-off value (>10,000 cells/ml) predicted mortality (hazard ratio 2.0, 95% confidence interval 1.4 to 2.8, p = 0.0001) but not major events (p = 0.07). When analysis was performed to assess troponin status, in the subgroup with increased troponin (n = 634, 16% mortality), a leukocyte count >10,000 cells/ml was related to mortality (hazard ratio 2.2, 95% confidence interval 1.5 to 3.4, p = 0.0001). However, in the subgroup with normal troponin levels (n = 827, 4.2% mortality), there were no differences in mortality between patients with or without a leukocyte count >10,000 cells/ml (4.4% vs 4.2%, p = 0.8), with survival curves showing a tight overlap (p = 0.9). Further, in the subgroup with normal troponin levels, leukocyte count was not significantly different between patients with or without ST depression (7,969 +/- 2,171 vs 8,108 +/- 2,356 cells/ml, p = 0.6) and was not associated with mortality in patients with ST depression (p = 0.7). In conclusion, leukocyte count on admission is predictive of mortality in patients with chest pain and non-ST-segment elevation myocardial infarction. However, in the absence of myocardial necrosis, leukocyte count lacks prognostic value.