The Doppler echocardiographic estimation of cardiac output at the mitral valve site is often underestimated in adults with slow heart rates because the mitral valve remains open in mid-diastole when flow is markedly reduced. Therefore we tested several approaches to this measurement in 17 adults with nonvalvular heart disease who had thermodilution catheters in the right side of the heart. Superior correlations with thermal output values were obtained by a new method that excludes mitral orifice measurements during mid-diastole when flow less than 10 cm/sec (r = 0.94) compared with the standard method (r = 0.89). Also, the new method resulted in significantly less underestimation of thermal cardiac output in patients with heart rates less than 70 beats/min (-10%) compared with the standard method (-34%). In addition, use of a constant maximal two-dimensional echocardiographic mitral orifice correction factor of 0.77 with the new method to account for variations in mitral valve orifice during the cardiac cycle, as opposed to 0.68 with the standard method, resulted in similar results as compared with determining individual correction factors from M-mode echoes. We conclude that: (1) the mitral orifice approach is accurate for measuring cardiac output in adult patients with nonvalvular heart disease; (2) a new method that excludes mid-diastolic mitral orifice measurements is superior to the standard method; and (3) use of a constant two-dimensional echocardiographic mitral valve orifice correction factor obviates the need for M-mode echoes.