Background: The release kinetic data of ischemia modified albumin (IMA) was based on the transient occlusion model using coronary balloon. The clinical setting is quite different in patients with acute chest pain because of various flow limitations from transient to prolonged occlusion. There was no data about the diagnostic efficiency of ischemia modified albumin measurements according to the time frame in patients with suspected acute coronary syndrome (ACS) in emergency department (ED). Method: From Jun. 2004 to Jul. 2005, we enrolled consecutive 302 patients (mean age; 60.9짹 14.1 years, male %: 59) presenting to the ED due to suspected ACS within 12 hours of chest pain. The diagnosis of ACS was based upon clinical findings, results of serial ECG/troponin and coronary angiography. The ideal cutoff value of IMA for ACS was calculated by Receiver Operator Characteristic (ROC) curve analysis according to the time frame of chest pain onset Results: The clinical diagnosis at discharge was 86 (28%) of acute myocardial infarction, 97 (32%) of Unstable angina, 62 (21%) of non-cardiac chest pain due to obvious cause, 51 (17%) of non-cardiac chest pain due to unknown cause. Results of ROC analysis are shown in the table. Conclusion: There was no diagnostic efficiency of IMA measurements after 5 hours of chest pain onset in patients with suspected ACS in emergency department (ED).