Currently, the success of coronary angioplasty is defined by anatomic criteria. Because of the known limitations of coronary arteriography, the translesional pressure gradient and coronary vasodilatory reserve were studied in 15 patients undergoing coronary angioplasty with the intent of defining a physiologically successful result. Coronary vasodilatory reserve was measured by a digital radiographic technique that has been previously validated against directly measured coronary sinus flow (r = 0.90, p less than 0.0001). A significant reduction in luminal stenosis from 71 +/- 12 to 34 +/- 11% (p less than 0.001) was accompanied by a reduction in translesional gradient from 47 +/- 19 to 21 +/- 12 mm Hg (p less than 0.001) and an increase in coronary vasodilatory reserve from 1.03 +/- 0.15 to 1.29 +/- 0.13 (p less than 0.001). There was a significant correlation between changes in luminal stenosis and changes in translesional gradient (r = 0.61, p less than 0.05), although a change of 20% or less in luminal diameter was accompanied by no change in pressure gradient. A more significant relation between changes in gradient and in coronary hyperemic reserve existed (r = 0.77, p less than 0.005). The relation was accurate even for small changes in gradient. Because saphenous vein bypass grafts have been shown to increase coronary vasodilatory reserve to at least 1.20, it is proposed that this physiologic criterion be used to define the success of revascularization by angioplasty. In patients in whom this value was achieved, translesional gradient was invariably 25% or less of ostial pressure and 20 mm Hg or less.(ABSTRACT TRUNCATED AT 250 WORDS)