The Hatle formula was derived empirically in native mitral stenosis and may not be valid for normal prosthetic valves. Bileaflet mechanical prostheses open fully at low flows and have minimal interindividual variation in orifice area. In these valves effective area and measured manufacturer's area should be similar. We studied 60 patients aged 58 +/- 12 yr at a mean of 5 months after implantation with a CarboMedics prosthesis. There was a coexistent aortic prosthesis in 21. All diastolic measurements were averaged over 5 beats and stroke volume was calculated from the integral of the subaortic velocity trace and the cross-sectional area of the left ventricular outflow tract. For the whole group, area by the Hatle formula was 3.1 +/- 0.7 cm2 and measured area was 2.8 +/- 0.4 cm2. There was no significant correlation between these values (p = 0.329). Pressure half-time was more closely correlated with peak transmitral velocity (p = 0.012), RR interval (p = 0.015), diastolic time interval (p = 0.062) and stroke volume (p = 0.074). We conclude that the Hatle formula should not be applied to normal bileaflet mitral prostheses where pressure half-time reflects nonprosthetic factors more closely than orifice area.