BACKGROUND: Body mass index (BMI) and mental stress seem to exert part of their cardiovascular risk by eliciting inflammation. However, the adverse effects of stress on inflammatory activity with BMI are not fully understood. We investigated whether higher BMI is associated with reduced glucocorticoid inhibition of inflammatory cytokine production following stress in men while controlling for age and blood pressure. We measured glucocorticoid inhibition of lipopolysaccharide (LPS)-stimulated release of the proinflammatory cytokine tumor necrosis factor (TNF)-alpha. METHODS: Forty-two men (age range 21-65 years; BMI range 21-34 kg/m(2)) underwent the Trier Social Stress Test (combination of mock job interview and mental arithmetic task). Whole blood samples were taken immediately before and after stress, and during recovery up to 60 min post-stress. Glucocorticoid sensitivity of LPS-stimulated TNF-alpha expression was assessed in vitro with and without coincubating increasing doses of dexamethasone. Moreover, salivary cortisol was measured during the experiment and on a normal day for assessment of baseline circadian cortisol. RESULTS: Higher BMI was associated with lower glucocorticoid sensitivity of monocyte TNF-alpha production after stress (main effect of BMI: p<0.001) and with more pronounced decreases of glucocorticoid sensitivity following stress (interaction of stress-by-BMI: p=0.002). Neither LPS-stimulated TNF-alpha release nor baseline glucocorticoid sensitivity were associated with BMI. Similarly, BMI was not associated with salivary cortisol, either in reaction to stress or in circadian cortisol secretion. CONCLUSIONS: Our data suggest that with increasing BMI, glucocorticoids are less able to inhibit TNF-alpha production following stress. This might suggest a new mechanism linking BMI with elevated risk for adverse cardiovascular outcomes following stress.