Rationale & objectives Patients with severe asthma have a poor therapeutic response to corticosteroid therapy, and corticosteroid responsiveness cannot be easily measured in these patients. We hypothesised that this poor response is associated with a reduced effect of corticosteroids to inhibit cytokine release from activated peripheral blood mononuclear cells (PBMCs). Methods Patients with severe asthma were defined by ATS criteria. We compared the suppression of lipopolysaccharide (LPS)-induced cytokine release (MCP-1, MIP-1, RANTES, TNF, IL-1, IL-8, IFN-,IL-6, IL-10 and GM-CSF) by dexamethasone from peripheral blood mononuclear cells (PBMCs) of severe asthma (n=16) and non-severe asthma (n=19) patients and normal volunteers (n=10). Results There was no difference in baseline spontaneous or stimulated release of these cytokines between groups. LPS-induced release of 10 cytokines was less suppressed by dexamethasone (10^-6M) in severe asthmatics compared to non-severe asthmatics, with statistical significance achieved for IL-1(p<0.03), IL-8 (p<0.03) and MIP-1 (p<0.003), and borderline significance for IL-6 (p=0.054). There was less difference between the 2 groups for dexamethasone at 10^-8M. Nuclear histone deacetylase (HDAC) and histone acetyltransferase (HAT) activities were both reduced in severe asthmatics compared to non-severe asthmatics (p<0.01). HDAC activity reduction correlated directly to the degree of steroid insensitivity of GM-CSF (r=0.57, p<0.01) and IFN- (r=0.56, p< 0.05) release. Reduction in HAT activity related to corticosteroid use rather than asthma severity. Conclusions Patients with severe asthma have diminished corticosteroid sensitivity of PBMCs when compared to non-severe asthmatics, associated with a reduction in HDAC activity that parallels the impaired corticosteroid sensitivity.