This study was performed to assess the accuracy of qualitative angiographic grading in persons with aortic regurgitation (AR) or mitral regurgitation (MR) and to determine the factors that may influence the reliability of such grading. In 230 patients (152 men, 78 women, aged 52 +/- 14 years) with AR or MR, forward cardiac index was measured by the Fick and indicator dilution techniques and left ventricular (LV) angiographic index by the area-length method, from which the regurgitant volume index was calculated. In 124 other patients (89 men, 35 women, aged 52 +/- 11 years) without regurgitation, there was good agreement between forward and angiographic cardiac indexes (r = 0.87, p less than 0.001). In the 83 patients with AR, the regurgitant volume indexes in those with 1+ (0.87 +/- 0.57 liters/min/m2) and 2+ (1.72 +/- 1.19 liters/min/m2) angiographic regurgitation were not significantly different from one another, but were significantly different from those with 3+ (3.0 +/- 1.42 liters/min/m2) and 4+ (4.80 +/- 2.25 liters/min/m2) regurgitation; at the same time, the regurgitant volume indexes of patients with 3+ and 4+ AR were not significantly different from one another. In the 147 patients with MR, the regurgitant volume indexes in patients with 1+ regurgitation (0.61 +/- 0.64 liters/min/m2) were significantly lower than other grades, but the regurgitant volume indexes of 2+ (1.14 +/- 0.85 liters/min/m2) vs 3+ (2.14 +/- 1.37 liters/min/m2) and of 3+ vs 4+ (4.60 +/- 2.31 liters/min/m2) were not significantly different. With AR and MR, regurgitant flow within each angiographic grade varied widely, especially in grades 3+ and 4+, and there was considerable overlap of regurgitant volume indexes between grades.(ABSTRACT TRUNCATED AT 250 WORDS)