The effects of cyclooxygenase 2 inhibitors on cartilage erosion and bone loss in a model of Mycobacterium tuberculosis-induced monoarticular arthritis in the rat.

Selective cyclooxygenase 2 (COX 2) inhibitors NS-398 and nimesulide were investigated for their effects on patellar cartilage and bone content in a model of Mycobacterium tuberculosis (M.tb)-induced monoarticular arthritis in the rat. The protective/destructive properties of these nonsteroidal antiinflammatory drugs (NSAIDs) were compared with piroxicam, known to accelerate cartilage breakdown and reduce bone erosion in this model in comparison to untreated arthritic controls. Male CFHB Wistar rats were injected intraarticularly with heat killed M.tb into the left stifle joint, resulting in loss of patellar cartilage glycosaminoglycans (GAG), bone erosion and inflammation. The right stifle joint received saline. Animals were dosed daily, p.o., with NS-398 (1, 10 mg/kg), nimesulide (0.5, 5 mg/kg) or piroxicam (10 mg/kg). Four days after M.tb injection, patellar GAG content, bone weight and joint swelling were measured in drug-treated animals and untreated arthritic controls. Changes in the left joint were compared to the right. The expression and distribution of COX 2 protein was determined by immunocytochemistry in synovial tissue from arthritic controls over the time course. Focal accumulations of inflammatory cells were positively immunolabelled for COX 2 in the synovium from the left stifle joint of untreated arthritic animals, 6 h after injection of M.tb. Labeling of inflammatory cell infiltrates increased and was widespread in the synovium at 24 h. By day 4 fibroblasts were positively labelled for COX 2 in addition to polymorphonuclear and mononuclear leukocytes. Piroxicam and nimesulide at the higher dose significantly exacerbated M.tb-induced cartilage GAG loss while NS-398 was without effect. Both COX 2 inhibitors did not alter M.tb-induced patellar bone loss. In contrast, piroxicam significantly reduced bone loss. All COX inhibitors significantly reduced joint swelling. In conclusion, the selective inhibition of COX 2 may result in the amelioration of synovitis with a lowered risk of NSAID-induced cartilage damage in rheumatic disease.