A Doppler echocardiographic study was performed to assess whether the Monostrut model of the Björk-Shiley valve (Shiley, Inc., Irvine, Calif.) had an improved hemodynamic performance in comparison with the spherical disc model in the aortic position. Twenty retrospectively randomly selected patients were studied, 10 with each valve type. Within each valve type two sizes of valve were studied, 21 and 23 mm. The two groups were comparable with respect to age, postoperative time, fractional shortening, New York Heart Association functional class preoperatively, and body surface area. Pulsed and continuous wave Doppler measurements were recorded at rest. Continuous wave Doppler recordings were performed every 2 minutes after exercise with supine bicycle ergometry until 10 minutes after exercise. Peak and mean gradients across the aortic valve prostheses were estimated. Both groups achieved a significant and comparable rise in heart rate with exercise. The mean gradients +/- standard error of the mean at rest and 2 minutes after exercise were 19.7 +/- 1.9 mm Hg and 30.9 +/- 2.2 mm Hg, respectively in the spherical disc group compared with 14.9 +/- 1.1 mm Hg and 23.6 +/- 1.7 mm Hg in the Monostrut group (p < 0.05 and p < 0.025, respectively). Peak transvalvular gradient at rest was 30.7 +/- 2.7 mm Hg in the spherical group compared with 23.9 +/- 1.9 mm Hg in the Monostrut group (p < 0.05). We conclude that the Monostrut Björk-Shiley valve prosthesis has better hemodynamic performance than the spherical disc model in the aortic position.