Recent studies have suggested that excision of the mitral valve apparatus during mitral valve replacement impairs left ventricular performance. However, functional measurements in humans have been difficult to obtain in a load-independent fashion. To investigate this concept, 12 patients (mean age, 65 +/- 8 years; mean New York Heart Association functional class, 3.3 +/- 0.7) with 4+ mitral regurgitation (n = 8) or mitral stenosis (valve area, 1.2 +/- 0.2 cm2) (n = 4) underwent prosthetic valve replacement using crystalloid cardioplegia. No patient required therapeutic inotropic support, every patient had at least the anterior mitral leaflet excised, and paced heart rate was maintained constant throughout. Left ventricular volume was measured with radionuclide angiocardiography, left ventricular pressure with a 3F micromanometer, and left ventricular wall volume with two-dimensional transesophageal echocardiography. Left ventricular preload was varied over a mean end-diastolic pressure range of 9 to 20 mm Hg and an end-diastolic volume range of 134 to 170 mL to generate four to five steady-state pressure-volume loops before and ten minutes after cardiopulmonary bypass. Left ventricular performance was estimated with the stroke work/end-diastolic volume relationship, which is insensitive to load. After bypass, no significant change (p greater than 0.1) was noted in wall volume for patients with mitral regurgitation or mitral stenosis (175 +/- 68 to 189 +/- 63 mL/m2 and 130 +/- 22 to 127 +/- 19 mL/m2, respectively).(ABSTRACT TRUNCATED AT 250 WORDS)