Although it is well established that coronary heart disease (CHD) patients with depression exhibit increased mortality compared with equally ill cardiac patients without depression, the mechanisms mediating this effect remain obscure. Depression is characterized by vulnerability to stress and heightened stress responsiveness, and stress can theoretically act through several biological pathways to contribute to excess mortality from CHD. Mechanisms connecting stress, depression and cardiovascular mortality have not been previously explored in detail. The purpose of this study was to assess the effects of stress and depression on myocardial perfusion and plasma cortisol concentrations in CHD patients. Patients with CHD with and without depression (n = 28) underwent single photon emission computed tomography imaging of myocardial perfusion at rest and during a stressful cognitive challenge. Severity of ischaemia was measured by summing perfusion defect scores across myocardial segments and subtracting out rest from stress scores. Plasma cortisol concentrations were measured at baseline and in response to the stressful challenge. There were no differences in stress-induced myocardial ischaemia or plasma cortisol response to stress between CHD patients with and without depression. Depressed CHD patients with a history of psychological trauma (n = 5) had an increase in stress-induced ischaemia scores [7; standard deviation (SD) = 5] compared with CHD patients with depression without a history of psychological trauma (2 SD = 2) and CHD patients without depression or psychological trauma (1; SD = 2) (F = 8.51; degree of freedom = 2,23; p = 0.007). Eighty per cent of CHD/depression trauma-exposed subjects had stress-induced ischaemia as opposed to 38 per cent of CHD/depression subjects without trauma exposure and 23 per cent of subjects with CHD without depression or trauma. Self-reported nervousness during the cognitive stressor was correlated with stress-induced ischaemia. These preliminary findings suggest that depression with a history of prior exposure to traumatic stress is associated with increased risk for stress-induced cardiovascular ischaemia. Copyright © 2008 John Wiley & Sons, Ltd.