We investigated the heart rate variability (HRV) parameters in patients with rheumatoid arthritis (RA) and assessed their relationship with disease characteristics. Twenty-three female patients with RA [age 48+/-7 (mean+/-SD) years] free of cardiovascular diseases and 23 age- and gender-matched healthy controls were evaluated. After careful clinical examination, the following parameters were obtained after 24-h Holter recordings: average of all normal-to normal (NN) intervals over the entire 24-h ECG recording (meanNN, ms); the standard deviation for the time between NN complexes (SDNN, ms); the standard deviation of the average NN intervals for each 5-min period (SDANN, ms) and the square root of the mean-squared differences of successive NN intervals (rMSSD, ms). We also assessed quantitative parameters of the Poincare plot: the standard deviation of the points perpendicular to the line-of-identity (SD1, ms); the standard deviation along the line-of-identity (SD2, ms) and their ratio (SD12). HRV parameters excluding SD2 were significantly lower in patients with RA, than in control group (p<0.05). Significant correlations of SDNN and SDANN with swollen joints count, Ritchie articular index, disease activity score (DAS) and disease duration were found. SDNN also correlated with leucocyte count and smoking. SD1 significantly correlated only with disease duration. Relationships between SDNN and smoking, swollen joints count and DAS were confirmed using multivariate analysis. Our data indicate that in patients with RA reduced HRV is independently associated with high disease activity and smoking. HRV assessment may be useful as a part of cardiovascular risk stratification in RA patients.