Rheumatic diseases are the most prevalent causes of disability in western countries, and non-steroidal anti-inflammatory drugs (NSAIDs) are still the most commonly used remedies. However, NSAIDs cause several serious adverse effects, the most important being from gastric injury to gastric ulceration and renal damage. Attempts to develop non-steroidal anti-inflammatory remedies devoid of these shortcomings-especially gastrointestinal toxicity-have followed several strategies. Non-steroidal anti-inflammatory drugs have, therefore, been associated with gastroprotective agents that counteract the damaging effects of prostaglandin synthesis suppression; however, a combination therapy introduces other problems of pharmacokinetics, toxicity, and patient's compliance. More recently, incorporation of a nitric oxide (NO)-generating moiety into the molecule of several NSAIDs was shown to greatly attenuate their ulcerogenic activity; however, several findings suggest a possible involvement of NO in the pathogenesis of arthritis and subsequent tissue destruction. A most promising approach seemed to be the preparation of novel NSAIDs, targeted at the inducible isoform of prostaglandin synthase (COX-2); they appear to be devoid of gastrointestinal toxicity, in that they spare mucosal prostaglandin synthesis. However, a number of recent studies have raised serious questions about the two central tenets that support this approach, namely that the prostaglandins that mediate inflammation and pain are produced solely via COX-2 and that the prostaglandins that are important in gastrointestinal and renal function are produced solely via COX-1. So, a growing body of evidence shows that COX-2 (not only COX-1) also plays a physiological role in several body functions and that, conversely, COX-1 (not only COX-2) may also be induced at sites of inflammation. More recent and puzzling data shows that COX-2 is induced during the resolution of an inflammatory response, and at this point it produces anti-inflammatory (PGD2 and PGF2alpha), but not proinflammatory (PGE2) prostaglandins; inhibition of COX-2 at this point thus results in persistence of the inflammation. Moreover, COX-2 selective NSAIDs have lost the cardiovascular protective effects of non-selective NSAIDs, effects which are mediated through COX-1 inhibition (in addition, COX-2 has a role in sustaining vascular prostacyclin production). The generation of other very important products of the arachidonic acid cascade (besides cyclooxygenase-produced metabolites) is inhibited neither by non-selective nor by COX-2 selective NSAIDs. The products generated by the 5-lipoxygenase pathway (leukotrienes) are particularly important in inflammation; indeed, leukotrienes increase microvascular permeability and are potent chemotactic agents. Moreover, inhibition of 5-lipoxygenase indirectly reduces the expression of TNF-alpha (a cytokine that plays a key role in inflammation). These data and considerations explain the efforts to obtain drugs able to inhibit both 5-lipoxygenase and cyclooxygenases, the so-called dual acting anti-inflammatory drugs. Such compounds retain the activity of classical NSAIDs, while avoiding their main drawbacks, in that curtailed production of gastroprotective prostaglandins is associated with a concurrent curtailed production of the gastro-damaging and bronchoconstrictive leukotrienes. Moreover, thanks to their mechanism of action, dual acting anti-inflammatory drugs could not merely alleviate symptoms of rheumatic diseases, but might also satisfy, at least in part, the criteria of a more definitive treatment. Indeed, leukotrienes are pro-inflammatory, increase microvascular permeability, are potent chemotactic agents and attract eosinophils, neutrophils and monocytes into the synovium.