Aspirin (acetylsalicylic acid) is metabolically converted to salicyclic acid by the action of carboxylesterases. Although metabolic drug interactions involving aspirin are theoretically possible, there appear to have been no studies to date which have shown conclusively that aspirin hydrolysis is altered by coadministered drugs. However, a number of treatments are known to affect the rate or extent of aspirin absorption, including activated charcoal, antacids, cholestyramine and metoclopramide. Caffeine and metoprolol have been reported to increase peak salicylic acid concentration following aspirin administration, and coadministration of dipyridamole and aspirin results in higher plasma aspirin concentrations. The mechanism(s) responsible for these latter observations remains unknown. Salicylic acid is extensively bound to plasma albumin, and many of the reported drug interactions involve displacement of the coadministered drug from plasma protein. Protein binding displacement appears to be the basis of salicylic acid interactions with diclofenac, flurbiprofen, ibuprofen, isoxicam, ketoprofen, naproxen, phenytoin and tolmetin. Following displacement of these agents increased clearance of total drug occurs, and consequently the plasma concentration of total drug decreases. Although generally not measured, unbound concentration of the interacting drug should not be markedly altered. Salicylic acid also increases total plasma clearance of fenoprofen but, unlike the interactions with the other propionic acid non-steroidals, plasma protein binding displacement does not appear to be involved. Induction of fenoprofen metabolism is a possibility, although there is no firm evidence from other studies that salicylate is able to induce the metabolism of coadministered drugs. Since salicylic acid is extensively metabolised, it is not surprising that it is able to inhibit the metabolism of certain coadministered drugs and chemicals, an effect which has been reported for salicylamide, valproic acid, m-xylene, and zomepirac. The interactions with salicylamide, m-xylene and zomepirac are probably competitive in nature since mutual inhibition of salicylic acid metabolism occurs. There is an additional component of protein binding displacement in the interactions with valproic acid and zomepirac, resulting in increased unbound drug concentration. Certain coadministered drugs (or chemicals) may alter the metabolism of salicylic acid; inhibition of its metabolism has been demonstrated following treatment with benzoic acid, salicylamide, m-xylene, zomepirac and possibly cimetidine. In contrast, salicylic acid elimination is enhanced in oral contraceptive steroid users and by corticosteroid treatment. Oral contraceptive steroids induce both salicylic acid glucuronidation and salicylurate formation. Induction of metabolism has also been proposed to account for the effects of corticosteroids, but this is still to be proven. Despite the fact that a relatively small percentage of an administered dose of salicylic acid is eliminated unchanged in the urine, it is nevertheless capable of decreasing the clearance of some renally excreted drugs. Salicylic acid both displaces acetazolamide and methotrexate from plasma protein and reduces their renal clearances, resulting in an increased unbound concentration of these drugs. The renal clearance of indomethacin is also reduced by salicylic acid, as is absorption and reabsorption, but its biliary clearance is apparently increased, resulting in little overall change in plasma clearance. The renal clearance of salicylic acid itself is pH-dependent and known to be increased by antacids in a dose-dependent manner. Relatively few of the pharmacokinetic interactions reported to occur with salicylates have been shown to be of clinical importance, although the effects of activated charcoal, antacid and metoclopramide on aspirin absorption are of relevance in specific instances and toxicity has been reported as a result of coadministration of salicylate with acetazolamide, methotrexate and valproic acid. Although other interactions could potentially result in increased or diminished response, this has yet to be demonstrated in patients.