In this chapter an attempt has been made to draw together the known biology of cartilage and some of the mechanisms thought to be responsible for its failure in arthritis. The picture is far from complete but we are now in a good position to use this information to help appraise the pros and cons of the wide range of drugs now available to treat articular disorders. For convenience, these drugs were classified as NSAIDs, corticosteroids and chondroprotective agents. The influence of each of these classes on the metabolism of cartilage was examined in the light of published laboratory and clinical studies. It has been clearly shown that not all NSAIDs are the same. While many of the older drugs provided no benefit to cartilage metabolism, and in some instances suppressed it, the more recently discovered molecules appear to be free of these undesirable effects. Tiaprofenic acid, diclofenac and piroxicam emerged as drugs with little or no harmful effects on cartilage metabolism when used at concentrations within the human therapeutic range. For all NSAIDs, their potential effects on cartilage must be weighed against their respective anti-inflammatory potency, half-life, and effects on the gastric mucosa and other tissues. Other chapters in this book have addressed these important problems. The long-acting corticosteroids, betamethasone and triamcinolone hexacetonide, also appear to offer some benefit in the management of OA; however, as in RA, their use should be restricted to short-term applications. In terms of cartilage metabolism the chondroprotective agents pentosan polysulphate, Arteparon and Rumalon have been the most extensively studied class of drugs. While the laboratory studies have provided convincing evidence of their chondroprotective efficacy, it has been difficult to prove this clinically. This dichotomy of opinion (laboratory versus clinical) stems largely from the inadequacy of the methodologies currently available for the objective clinical assessment of patient response to anti-rheumatic drug therapy. With the advent of nuclear magnetic resonance imaging techniques and monoclonal antibodies to detect specific proteoglycan breakdown fragments in synovial fluid and plasma, the prospects for a unified research approach for the evaluation of these agents may now be possible.