The frequency and severity of atherosclerosis of the cardiac transplant make it an essential complication of cardiac transplantation. Coronary angiography is the usual diagnostic method but it has severe limitations. In order to evaluate other diagnostic methods coronary angiography and non-invasive techniques: echocardiography, exercise stress ECG, exercise radionuclide ejection fraction, stress Thallium scintigraphy, were performed practically simultaneously in 60 patients after cardiac transplantation. These non-invasive methods were said to be positive in the presence of, respectively, a segmental wall motion abnormality, ischaemic ST segment depression, absence of increased ejection fraction on exercise, reversible or irreversible myocardial hypofixation. Coronary angiography was considered as the reference procedure for distinction between "normal coronary circulation" (no angiographically detectable lesion) and "graft atherosclerosis" (at least one coronary stenosis irrespective of the severity and extension). None of the non-invasive methods had an adequate sensibility when compared with coronary angiography (echocardiography 0.27, exercise stress ECG 0.28, exercise radionuclide ejection fraction 0.64, myocardial scintigraphy 0.62) or negative predictive value (echocardiography 0.56, exercise stress ECG 0.58, exercise radionuclide ejection fraction 0.68, myocardial scintigraphy 0.66). This inadequacy of the non-invasive technique may be explained by the fact that they are more adapted to the diagnosis of myocardial ischaemia than that of coronary studies. In addition, the extent of the coronary lesions may have masked discordance between 2 segments by the global hypovascularisation. The results of this study indicate that the non-invasive methods studied cannot be recommended for diagnosis of atherosclerosis of cardiac transplants.