Background: The conventional 12-lead electrocardiogram showed low sensitivity for detecting patients with acute inferolateral or posterior myocardial infarct. Some of these patients were diagnosed as non-ST-elevation myocardial infarct (NSTEMI). So we evaluated the incidence, clinical characteristics, angiographic characteristics and clinical outcome of patients with NSTEMI but angiographically occluded culprit artery. Methods: We evaluated 314 patients with NSTEMI who underwent coronary angiography in our hospital from May 2003 to Dec 2008. Results: 53 (16.9%) patients showed occluded culprit artery. There was no difference in age, sex, medical history and medication. Shorter symptom duration was seen in occluded artery group (1198.1 짹 2125.7 vs 614.7 짹 782.5, p < 0.01). Killip class IV was frequent in patients with occluded culprit artery (3.8% vs 15.1%, p < 0.01). Higher peak troponin I level were shown in patients with occluded culprit artery (Troponin I, ng/ml: 23.9 짹 19.1 vs. 32.6 짹 17.1, p < 0.01). Right (RCA) and left circumflex coronary (LCX) arteries were more common in patients with occluded culprit arteries (LAD: 18.9%; LCX: 43.4%; RCA: 37.7%; Left main: 0%). There was no difference in disease extent. There were no difference in clinical outcome including target lesion revascularization, target vessel revascularization, myocardial infarct but overall mortality was significantly higher in patients with occluded culprit artery. Especially in-hospital cardiac mortality was significantly higher in occluded artery group (0.4% vs 15.1%, p < 0.01). Conclusion: The incidence of NSTEMI from occluded artery was about 17%. These patients showed higher killip class and short symptom duration. Because patients with occluded culprit artery who present NSTEMI showed higher mortality, improved early diagnostic tools for detecting and early risk stratification of these patients are needed.