The diagnostic yield of echocardiography, radionuclide ventriculography (first pass and multiple gated) and contrast ventriculography was evaluated in 11 patients in whom a diagnosis of left ventricular pseudoaneurysm had been made during the last ten years. The diagnosis was made by two dimensional echocardiography (associated with Doppler in the last 5 patients) in 8 of 11 patients (sensitivity: 73%). The major limitation of the technique, in addition to the impossibility of an adequate recording due to a suboptimal acoustic window, is the poor definition of the neck of the pseudoaneurysm, particularly in the inferior localization. By contrast, echocardiography is the only technique which permits the direct visualization of thrombi within the pseudoaneurysm. Multiple gated radionuclide ventriculography was diagnostic in 7 of 10 patients (sensitivity: 70%). Its major limitations are the poorer spatial resolution to visualize the pseudoaneurysm neck and the thrombi within the cavity. First pass radionuclide ventriculography was diagnostic in 4 of 6 patients (sensitivity: 67%). In one of them it improved on the diagnostic yield of the multiple gated technique. Contrast ventriculography was diagnostic in 5 of 7 patients (sensitivity: 71%). The diagnosis was missed due to technical problems in one patient and to significant thrombosis within the pseudoaneurysm in another. Thus, none of the imaging studies has optimal sensitivity. Therefore, all play a complementary role in the diagnosis of ventricular pseudoaneurysm.