Red cell distribution width (RDW) and neutrophil/lymphocyte ratio (NLR) have been found to be associated with cardiovascular diseases. Only a few trials have investigated the correlation of these parameters with postoperative atrial fibrillation (AF). However, the correlation of these parameters in non-valvular AF is still unclear. We retrospectively analyzed consecutive AF patients from medical records and included 117 non-valvular AF patients (103 paroxysmal and 14 chronic AF). All subjects underwent physical examination and echocardiographic imaging. Complete blood counts (CBCs) were analyzed for hemoglobin, RDW, neutrophil and lymphocyte counts as well as mean corpuscular volume. Results of CBC tests within the previous year were also included and the averages were used. The demographic and echocardiographic properties of non-valvular AF group were comparable to the control group except for left atrial volumes which were increased in AF (median 33.1, IQR 26.3-41.1 cm(3) vs. median 26.4, IQR 24.2-28.9 cm(3); p = 0.01). RDW levels were significantly higher in the AF group (median 13.4 %, IQR 12.9-14.1 %) compared to the control (median 12.6 %, IQR 12.0-13.1 %; p = 0.01). NLR was not statistically different in the AF group and the controls (2.04 ± 0.94 vs. 1.93 ± 0.64, respectively; p = 0.32). Hs-CRP levels were higher in the AF group compared to the controls (median 0.84, IQR 0.30-1.43 mg/L vs. median 0.29, IQR 0.18-0.50 mg/L, respectively; p = 0.01). Multivariate logistic regression analysis revealed RDW (OR 4.18, 95 % CI 2.15-8.15; p = 0.01), hs-CRP (OR 3.76, 95 % CI 1.43-9.89; p = 0.01) and left atrial volume (OR 1.31, 95 % CI 1.06-1.21; p = 0.01) as the independent markers of non-valvular AF. Multivariate linear regression analysis revealed that hemoglobin levels (standardized β coefficient = -0.252; p = 0.01) and the presence of AF (standardized β coefficient = 0.336; p = 0.01) were the independent correlates of RDW levels. Elevated RDW levels, not NLR, may be an independent risk marker for non-valvular AF.