Most studies of dexamethasone (DXN) effects on experimental glioma have used doses 10-500 higher (on mg/kg basis) than those used for patients with brain tumour. The relevance of findings to patients with glioma are therefore uncertain. In order to evaluate the effects of clinical doses of DXN (0.22 mg kg(-1)day(-1)) on the pathophysiology of an experimental glioma we have treated rodents with established C6 gliomas for 3 days. The effects of therapy on local cerebral blood flow (LCBF), tumour blood flow (TBF), tumour capillary permeability (TCP), and inducible nit ric oxide synthase (iNOS) mRNA expression were evaluated. DXN caused a significant reduction in TCP (21 +/- 1.9 to 7.7 +/- 2.2 ml.gm(-1)min(-1)10(-3)) and iNOS mRNA production within and around tumour, but no significant change in either TBF or LCBF. The reduction in TCP was identical to that reported after higher doses of DXN and is probably mediated by glucocorticoid receptors. Further in vivo stud ies using either behavioural or neuropathological paradigms in rodents with established cerebral glioma should be treated with similar doses of DXN to optimise clinical relevance.