The pharmacokinetics of oral gold (auranofin) in some respects resemble, and in other respects differ from, those of existing parenteral gold compounds such as gold sodium thiomalate (GST). This may in part relate to physicochemical differences as GST is a water-soluble polymeric compound in vitro whereas auranofin is lipid-soluble and characteristically monomeric. Furthermore, intramuscularly administered gold is greater than 95% bioavailable, whereas only 20 to 30% of an orally administered dose of auranofin is absorbed. Following a standard 50mg intramuscular injection of GST, serum gold concentrations rise sharply, peaking between 4 and 8 mg/L in approximately 2 hours and declining to an average of 3 mg/L by 7 days. With repeated injections of GST stable serum concentrations of gold (3 to 5 mg/L) are eventually achieved (usually within 5 to 8 weeks) although absolute concentrations may vary widely between patients. On the other hand, long term treatment with auranofin is associated with lower and more stable serum concentrations of gold (0.5 to 0.7 mg/L), on the standard dosing regimen of 6 mg daily. Both compounds are retained within the body for prolonged periods. However, the amount of gold retained with auranofin is significantly less compared with GST (less than 5% of a tracer dose of auranofin--about 20% of the absorbed dose--is retained by 100 days whereas the retention for a single labelled dose of GST over a similar interval is greater than 50%). Excretory patterns of GST and auranofin also differ. Most of an absorbed dose of GST (greater than 70%) is excreted by the kidneys whereas only 50% of an absorbed (15% of an administered) dose of auranofin is excreted in the urine. Both compounds are avidly bound by plasma proteins and auranofin shows a particularly strong association with circulating cellular elements. In human subjects, parenterally administered gold is widely distributed among bodily tissues, showing a predilection for tissues of the reticuloendothelial system as well as the kidney and adrenal cortex. Comparable studies in humans are not available for auranofin but animal studies have shown comparatively less affinity for the liver, kidney and spleen. Valuable insight has been gained in analysing the comparative pharmacokinetics of oral and injectable gold compounds. Unfortunately, attempts to correlate pharmacokinetic findings with clinical response or pharmacodynamic changes, as a whole, remain largely unsuccessful with these agents.