Ten healthy male volunteers received 5 mg of dexamethasone sodium phosphate (DEX) i.v. and, on an other occasion, by way of nebulization. Plasma DEX and hydrocortisone (HC) concentrations as well as blood lymphocyte count (BLC) were monitored over 12 hr and at 24 to 28 hr after DEX administration. Bioavailability of DEX after inhalation was about 27% of DEX i.v. DEX-induced depletion of plasma HC could be predicted with a pharmacokinetic model. The reonset rate of HC-production was dependent of DEX dose. BLCs declined after DEX administration, reaching a minimum between 4- and 8-hr postdosing. The DEX- and HC-induced depression of BLC could be described by an integrated pharmacokinetic-pharmacodynamic competitive-interaction model that assumes that both agonists act on the same receptor. With this model the potency and efficacy of DEX and HC with respect to lymphocytopenia could be estimated simultaneously. The potency of DEX was 10 times greater than the potency of HC. The estimated efficacy suggests that HC is only a partial agonist; the maximal lymphocytopenic effect (Emax) of HC was estimated at 80% (27-99%) of the efficacy of DEX. Our results indicate that DEX should be preferred instead of HC in conditions in which the lymphocytopenic effect is the primary systemic corticosteroid treatment goal.