The aim of this study was to investigate a new bone densitometric technology based on digital radiogrammetry (DXR) with respect to its ability to measure severity-dependent variations of bone mineralization in patients with rheumatoid arthritis. One hundred six randomly selected patients suffering from verified rheumatoid arthritis underwent digitally performed plain radiographs of the non-dominant hand and measurements of dual-energy X-ray absorptiometry (DXA) regarding total femur and lumbar spine. Using DXR the radiographs were analyzed retrospectively for bone mineral density (BMD) calculation. The severity was classified using Larsen score and Steinbroker stage blinded by two radiologists. A third radiologist reviewed the incongruently scored cases. Mean values of calculated parameters changed as follows from Larsen 1 to Larsen 5: Bone mineral density (DXR-BMD) decreased from 0.55 to 0.44 g/cm2 (p=0.000), DXR-MCI decreased from 0.44 to 0.33 (p=0.001), DXA-BMD (total femur) decreased from 0.92 to 0.78 g/cm2 (p=0.090) and DXA-BMD (lumbar spine) decreased from 0.91 to 0.84 g/cm2 (p=0.595). Similar results were verified for the Steinbroker stage. The relative decrease of BMD measured by DXR between the highest and lowest score was 20% for Steinbroker stage and Larsen score (p<0.05). The relative decrease of BMD using DXA revealed not such a significant result. Similar results were verified for metacarpal index (estimated by DXR). Correlations between BMD determined by DXR and by DXA were all significant (R=0.45 for lumbar spine and R=0.59 for total femur). Consequently, less than 35% of the DXR-BMD value is explainable by corresponding DXA values. The DXR-based BMD calculation seems to be able to distinguish severity and progress of the disease in contrast to those of DXA at lumbar spine and total femur.