BACKGROUND: The aim of this study was to investigate whether C-reactive protein (CRP) has prognostic impact on rhythm conversion by antiarrhythmic agents (AAD) of persistent atrial fibrillation (AF). METHOD: This study enrolled 57 consecutive patients with persistent AF lasting more than 2 weeks who had preserved left ventricular ejection fraction (LVEF) (mean EF = 51.8짹8.4%). The success of pharmacologic cardioversion was evaluated by 24 hour Holter monitoring between 4 and 8 weeks after prescribing AAD. All patients divided into 2 groups according to the presence and maintenance of sinus rhythm. The level of CRP was analyzed as continuous and categorical variables (q1: less than 0.57mg/l, q2: from 0.58 to 1.15mg/l, q3: from 1.16 to 2.54mg/l, q4: more than 2.54mg/l). With other serologic markers, we analyzed LVEF, left atrial volume index (LAVI), and LV mass in transthorasic echocardiography (TTE). RESULTS: Out of study population, successful pharmacologic cardioversion was observed in 18 patients (group I, 31.6%). In baseline characteristics, the patients in group I had lower rate of LAVI>32 mm3/m2 than those in group 2 (27.8% vs. 59.5%, p=0.025). However, serum level of CRP had no difference between 2 groups both in continuous (mean CRP level: 0.23짹0.24 in group I vs. 0.24짹0.43 in group II, p>0.05) and quartile analyses. Also, CRP level did not show the correlation with echocardiographic parameters such as LAVI (correlation coefficient=0.14, p=0.334) after adjustment for age, sex, and some parameters on TTE. Multivariate analysis revealed that LAVI>32 mm3/m2 seemed to increase the failure rate of pharmacologic cardioversion [Odd ratio = 2.33 (95% confidence interval 0.94 to 11.45), p=0.079]. CONCLUSION: In this study, left atrial size had impact on cardioversion rate, but there was a consistent absence of association between CRP and pharmacologic cardioversion of persistent AF both in univariate and multivariate analysis.