Leptin is a peptide hormone that has an essential role in the regulation of body weight by inhibiting food intake and stimulating energy expenditure. The role of leptin in the modulation of the immune response and inflammation has been regarded as important. In rheumatoid arthritis (RA) patients it was reported that fasting leads to an improvement of clinical and biological measures of disease activity, which was associated with a marked decrease in serum leptin. These features suggest that leptin may also influence the inflammatory mechanisms of arthritis in humans. In this study we assessed serum leptin levels in RA and osteoarthritis (OA) patients and found a correlation between serum leptin level and other markers as well as bone mass density changes, activity of disease, disease duration and the age of the patients. The blood was collected from 30 RA and 30 OA patients who constituted the control group. Serum leptin level was determined using the DRG Leptin ELISA Kit-a solid phase enzyme-linked immunosorbent assay based on the sandwich principle. The serum level of leptin in RA patients ranged from 1.8 to 81.1 ng/ml and median value was 11.2. There was a positive correlation between body mass index (BMI) of RA patients and serum level of leptin (correlation coefficients Spearman's r = 0.81). According to correlation coefficients, serum leptin level is independent of age of RA patients, stage of disease, number of painful and swollen joints, duration of morning stiffness, disease duration as well as value of titre of the Waaler-Rose, disease activity score (DAS 28) value and presence of rheumatoid nodules. There was a negative correlation between serum leptin level and glomerular filtration rate (GFR). No correlation between the serum leptin level and T-score was found. An influence of steroid treatment on the serum leptin level was not shown. The median serum leptin level in OA patients was 9.2 ng/ml. There was a positive correlation between body mass index of OA patients and serum level of leptin (correlation coefficients Spearman's r = 0.57). No correlation was found between serum leptin level and patient's age, duration of disease and value of laboratory data. There were no correlations between serum leptin level and visual analogue pain scale (VAS) for the lower-limb afflicted patients as well as stage of disease according to Kellgren and Lawrence's score in OA patients. There was a negative correlation between serum leptin level and T-score value in OA patients (r = -0.58, P < 0.05). No statistically significant differences were found between serum leptin levels for RA and OA patients.