Determination of the clinical and hemodynamic significance of coronary stenoses is often difficult and inexact. Angiography and coronary vasodilator reserve have been shown to be imperfect tools to determine the physiologic significance of coronary stenoses. Spectral flow velocity data, both proximal and distal to coronary stenoses, using an 0.018-in intracoronary Doppler-tipped angioplasty guidewire, were compared to translesional pressure gradients and angiography during cardiac catheterization. Patients were divided into 2 groups based on resting translesional gradients: Group 1 had gradients <20 mm Hg and group 2 had gradients greater-than-or-equal-to 20 mm Hg. Proximal average peak velocity, diastolic velocity integral, and total velocity integral were statistically significantly lower in Group 1. The distal average peak velocity, and diastolic and total velocity integrals were all significantly (p < 0.01) decreased in patients with gradients > 20 mm Hg (group 2). The ratio of proximal-to-distal total flow velocity integral was also higher in group 2 patients (2.3 +/- 0.9) compared with group 1 (1.1 +/- 0.2; p < 0.001). There was a strong correlation between translesional pressure gradients and the ratios of the proximal-to-distal total flow velocity integrals (r = 0.8, p < 0.001) with a weaker relationship between quantitative angiography and pressure gradients (r = 0.6, p < 0.001). Angiography was a poor predictor of translesional gradients in angiographically intermediate stenoses (range 50-70%; r = 0.2, p = NS), while the flow velocity ratios continued to have a strong correlation (r = 0.8, p <0.0001). Translesional coronary flow velocity is a new and easily applicable method for determining the hemodynamic significance of coronary stenoses that can be applied at the time of diagnostic catheterization. These data will provide a rational approach to decisions for questionable angiographic findings for coronary interventional techniques.