Inflammation alters the properties of sensory receptors, in particular those with small-calibre afferent fibres, the Aδ and C groups, that include among their numbers the nociceptors. Early attempts to follow the lead of Keele & Armstrong (1964) in their classical studies of algogens acting on the cantharadin blister base, used electrophysiological methods to explore the recently accessible non-myelinated afferent units (Fjallbrant & Iggo, 1961). In these studies on healthy normal animals the algogens tested had relatively weak actions, and in some cases appeared to excite non-nociceptors as effectively as they did nociceptors. More recent work in other laboratories (Franz & Mense, 1975; Beck & Handwerker, 1976; Perl, 1976) extended the search to include, among other algogens, bradykinin (BK) and the prostanoids. Interactions of such agents with a potent non-steroidal anti-inflammatory drug (NSAID), acetylsalicyclic acid (ASA), were also relatively disappointing in the sense that they had little action in interfering with the action of administered BK or prostanoids, such as prostaglandins E1 or E2 (PGE1, PGE2) (Mense, 1982).

The development of several animal models of inflammation has now provided an opportunity to examine with more precision the nature of the inflammatory response and, in the present context, the behavioural (de Castro et al., 1981), physiological (Guilbaud, Iggo & Tegner, 1985) and pharmacological (e.g. Guilbaud & Iggo, 1985) consequences of inflammation. In my own laboratory we have been using a model based on the complete Freund's adjuvant (CFA) arthritic rat (Pearson & Wood, 1959), developed for electrophysiological study of joint afferent fibres, initially with Guilbaud and Tegner (Guilbaud et al., 1985) and subsequently modified to give a monoarthritis restricted to one joint (Grubb et al., 1991).