Do NSAIDs Adversely Affect Joint Pathology in Osteoarthritis?

Abstract

Shortly after the market introduction of NSAIDs in the UK, there were several reports that these drugs accelerate the progression of the osteoarthritis (OA) disease process. There have been suggestions that this effect may be related to dose and to duration of administration. In animal work, it has been shown that NSAIDs have little effect on normal loaded joints but that they diminish cartilage metabolism in diseased joints. Our hypothesis proposes that NSAIDs adversely affect joint pathology in OA by affecting prostaglandin-regulation of cartilage proteoglycans, enhancing interleukin-1 production and attenuating vasodilation, so reducing the OA-enhanced joint perfusion and reducing the repair process. These effects combine to increase the progression of the OA pathology in subjects receiving potent prostaglandin synthesis. To test this hypothesis studies were performed in two phases to examine the long-term effects of treatment of OA patients with NSAIDs of varying potency as prostaglandin-synthesis inhibitors and differing effects on other molecular and cellular events in inflammation. In Phase I, 105 OA patients awaiting hip arthroplasty were treated prospectively with a strong or a weak prostaglandin synthesis inhibitor, indomethacin or azapropazone respectively. In Phase II subjects received diclofenac, naproxen, piroxicam, tiaprofenic acid or the analgesics, dextropropoxyphene and paracetamol. A few patients elected for no treatment. In both phases, pain and radiological joint space were monitored up to the arthroplasty, following which the histology of the excised femoral head was determined. As judged by histopathological data, the treatment groups in both Phase I and Phase II had similar pain relief from all the drugs and were at a similar pathophysiological end-point when they came to arthroplasty. In the indomethacin group, the ‘affected’ hips lost joint space more rapidly than did the contralateral hips, a difference not seen in the azapropazone group. The patients receiving azapropazone who had higher synovial concentrations of vasodilator prostaglandin E2 took longer than the indomethacin group to reach the arthroplasty end-point. The patients in Phase II all appeared to have the same progression in OA as one another, and these appeared similar to those in the indomethacin group in Phase I.

These results show that potent prostaglandin synthesis inhibitors all have about the same propensity to accelerate joint injury in OA; weak prostaglandin inhibitors or analgesics are less likely to cause this acceleration of joint destruction.