Aims Glucocorticoids suppress the release of tumour necrosis factor-a (TNF-a) by macrophages in vitro and cause monocytopaenia in vivo. These actions may contribute to anti-inflammatory and immunosuppressant effects. We therefore examined relationships between prednisolone concentration, suppression of monocyte TNF-a release, monocytopaenia and suppression of total cortisol concentration in healthy volunteers treated with a single dose (1.5 mg kg −1 ) of the glucocorticoid, prednisolone. Methods Monocyte numbers, total cortisol concentration and prednisolone concentration were measured in blood samples collected over 48 h after the dose. Plasma from these samples was also tested for its capacity to suppress lipopolysaccharideinduced TNF-a release from monocytes in autologous whole blood cultures.Results At 4 h after the dose, monocyte numbers in peripheral blood had fallen to a mean of 18% of the pre-dose level whilst plasma total cortisol had fallen to 9% of the pre-dose concentration. Monocyte numbers recovered in concordance with elimination of prednisolone and there was a significant relative monocytosis at 24 h. The recovery of plasma cortisol was delayed in comparison, with cortisol remaining significantly suppressed at 24 h. Plasma samples taken at 2 h after the dose (corresponding to peak plasma prednisolone concentration) suppressed the lipopolysaccharide-stimulated production of TNF-a by autologous blood monocytes to 27% of pre-dose control. Plasma collected at intervals over the 48 h from dosing also suppressed monocyte TNF-a release in relation to the prednisolone concentration therein. Suppression was largely reversed by the glucocorticoid antagonist, mifepristone. A similar relationship between prednisolone concentration and TNF-a suppression was observed when prednisolone was added to blood samples collected from the volunteers when they were drug-free. Conclusions Blood concentrations of prednisolone achieved after a dose of 1.5 mg kg −1 are sufficient to suppress monocyte TNF-a release and cause a biphasic change in peripheral blood monocyte numbers. Suppression of TNF-a is principally a direct glucocorticoid effect, rather than a consequence of other prednisoloneinduced changes to blood composition.