One-hundred and seven exercise stress tests and coronary angiograms were reviewed retrospectively, in order to evaluate the usefulness of R wave amplitude changes (ΔR) during exercise compared with ST segment depression in the screening of patients with coronary artery disease (CAD). We also attempted to correlate ΔR with the severity of CAD as expressed by coronary arteriography and left ventriculography. Thirty-six patients showed no coronary artery narrowing (0-V); the remaining 71 patients with stenosis of 70% of at least one of the major coronary arteries were divided into three groups. Sixteen patients had single vessel disease (1-V); five (31%) in this group showed abnormal left ventricular wall motion. Thirty-one patients had two-vessel disease (2-V); 22 (71%) of the 31 demonstrated abnormal left ventricular wall motion. Twenty-four patients had three-vessel disease (3-V); 20 (83%) of the 24 showed abnormal left ventricular wall motion. We considered ΔR values ≥ 0 and ST segment depression ≥ 1 mm. significant for diagnosis of CAD. The sensitivity of the ΔR method in predicting CAD was superior to the method based upon ST segment depression; however, the latter was significantly (P We found ΔR values ≥ 0 more frequently in the 2-V and 3-V groups as compared with the 1-V group. Patients of the 2-V and 3-V groups had a significantly higher incidence of abnormal left ventricular wall motion (P Even though the accuracy of the ΔR method was greater in more severe CAD, it seems to be offset by a concomitant decrease in specificity.