It is almost 20 years since non-steroidal anti-inflammatory drugs (NSAIDs) were shown to inhibit the synthesis of prostaglandins. Although our knowledge of the molecular nature and pharmacology of prostaglandins and other eicosanoids has increased considerably during this period, their role in tissue physiology and inflammation is still incompletely understood.'2 It seems generally agreed, however, that inhibiting their synthesis has not helped in modifying the progression of inflammatory rheumatic diseases.

The eicosanoids, like other low molecular weight regulatory factors, such as platelet activating factor, the kinins, histamine, and seratonin, seem to act as local hormones and mediate quite rapid tissue responses. More recently, interest has focused on a group of protein based cell regulatory factors, or cytokines, whichusually act in a similarly localised fashion but mediate longertermeffects. 3 4Although the definition ofcytokines is somewhat loose at present, regulatory factors that fall within this category include the interleukins, tumour necrosis factor (s), the interferons, bone marrow colony stimulating factors, and a variety of other'growth factors'(fibroblast, epidermal, transforming, and platelet derived growth factors). This nomenclature is not accurately descriptive and is to some extent misleading. For example, most of the interleukins act as growth and differentiation factors not only for leucocytes but for a variety of other cells, including bone marrow, fibroblasts, and neural cells. Interferons canregulate the growth of haemopoietic and other cells, and cytokines in each category can regulate lymphocyte and inflammatory cell activity. These activities, and the ability of many of them to regulate growth and function of connective tissue cells, indicate that cytokines are likely to be directly or indirectly involved in the abnormal regulation of various tissues affected by rheumatic diseases. A few, such as interleukin 1, interleukin 6, tumour necrosis factor a, and interferon gamma, have received special attention. Interestingly, and with one notable exception (interleukin 6), many cytokines have been found to influence eicosanoid metabolism. The question therefore arises as to the importance of this activity and how it is regulated. The problem in answering this is that although there are considerable data relating to the molecular nature of these mediators and the phenomena that are regulated by them, there is little indication of how any of them act at the cellular level, let alone how they relate to each other. 134