Objectives: We investigated whether ischemia-modified albumin (IMA), a novel biomarker of acute myocardial ischemia, is reliable for diagnosing ischemic chest pain and predicting future cardiac events in clinical practice. Methods: We enrolled 390 patients (age 61.7 ± 39.9 years) who arrived at the emergency department within 6 h of pain onset. We compared serum IMA levels of patients with ischemic chest pain (ICP) to those with nonischemic chest pain (NICP). Results: NICP was diagnosed in 162 patients and ICP in 205 patients. Median serum IMA levels did not differ between the NICP (99.0 U/ml; 95% confidence interval, 98.2–101.2 U/ml) and the ICP group (99.0 U/ml; 95% confidence interval, 99.4–102.6 U/ml; p = 0.320). During a 2-year follow-up, adverse cardiac events including cardiac death, myocardial infarction and recurrent angina occurred in 25 of 205 patients (12.2%) in the ICP group, but IMA was not predictive of cardiac events. Conclusions: IMA was neither able to discriminate acute chest pain due to myocardial ischemia from that of other origin nor to predict cardiac events. Therefore, use of serum IMA levels for triage of patients with acute chest pain does not seem effective in clinical practice.