Background: There is increasing evidence that inflammation may play a role in the pathogenesis of at least some types of AF. And, it has been demostrated that high-sensitivity C-reactive protein (hs CRP) levels are significantly increased in patients with atrial fibrillation (AF). However, it is not known cleary yet, whether hs CRP could be related to AF in patients with acute myocardial infarction (AMI), which belongs to inflammatory response. We evaluated the predictive role of hs CRP for AF in patients with AMI. Methods: We prospectively studied 192 patients with AMI from May 2005 to May 2006. We compared variables of EF, the degree of CAD, hs CRP, comobid diseases, previous medical history between patient with AF and without AF. We evaluated the following events: atrial fibrillation, rehospitalization due to cardiovascular accidents or ischemic stroke, and cardiovascular death. Results: During a mean follow up period of 8.5짹3.7 months, incidence of newly-onset atrial fibrillation was 8.8%(17/192). Patients with AF were older (p=0.015), had hs CRP>37mg/dL (p=0.001), EF<35% (p=0.016), culprit lesion at RCA or LCx (p=0.016), higher rates of previous MI (p=0.04), higher heart rate (p=0.02), and QTc>430 (p=0.017) than without AF. In multivariate analysis, variables independently associated with atrial fibrillation were: hsCRP> 37mg/dL (odd ratio=5.13, 95% CI: 1.2-20.8, p=0.022), EF<35% (odd ratio:0.18, 95% CI:0.04-0.79, p=0.039), culprit lesion located at RCA or LCx(odd ratio:0.17, 95% CI: 0.037 -0.921, p=0.024). At 12 months, rehospitalization rate (29.4% vs 8.4%, p=0.006 by log rank test) and cardiac mortality rate (5.9% vs 0.6 %, p=0.03 by log rank test) were higher in patients with atrial fibrillation than without AF. CONCLUSION: Our data show that hs-CRP could predict newly-onset AF in AMI patients with other variables.