Some controversial issues in the current literature in relation to osteoporosis and rheumatoid arthritis are updated and discussed. Because most studies agree that osteoporosis in postmenopausal women and in men with RA is more evident at the hip and radius than at the spine, and that the most important determinants of bone loss are disability, local disease activity, and cumulative corticosteroid dose, osteoporosis is not a common systemic extra-articular manifestation of RA. In early arthritis, periarticular osteoporosis does indeed reflect disease activity because it is closely related to the acute phase reactants, but once periarticular osteoporosis is established it is no longer a marker of disease severity. The threshold dose for corticosteroid-induced osteoporotic fractures is the cumulative rather than the actual dose. Statements based on quantitative tomography concerning the acute effects (and their reversal) of corticosteroids on bone have to be interpreted with care because of important body composition changes, in particular in bone marrow fat, during corticosteroid treatment. At present there is no evidence that anti-resorbing drugs can change the progress of RA erosions, probably because erosions are the result of non-osteoclast mediated mechanisms. Stress fractures in RA are underdiagnosed and are often confused with synovitis, and therefore it is likely that they are more frequent than commonly thought, especially in the lower limbs. Methotrexate osteopathy is known in oncological practice. Whether low dose methotrexate is toxic for bone is not clear, but a number of clinical observations suggest that the occurrence of spontaneous fractures and lower extremity pain is more frequent in methotrexate treated patients than expected. Prospective studies are necessary to confirm these impressions.