This study documented mitral flow velocity patterns in anesthetized patients with ischemic heart disease and left ventricular dysfunction and investigated the relationship between transmitral flow velocity measurements and the hemodynamic response to ventricular pacing. Twenty-four patients in sinus rhythm without evidence of mitral valve disease undergoing elective myocardial revascularization were studied. Anesthesia consisted of a high-dose opioid-muscle relaxant-oxygen technique. After endotracheal intubation a 3.5-MHz phased-array transesophageal echocardiographic probe was inserted and positioned to obtain a long-axis view of the left atrium and left ventricle. The Doppler sample volume was placed at the mitral annulus with minimal cursor angulation, and the biphasic velocity tracing of transmitral blood flow was recorded. A hemodynamic profile was obtained, and cardiac output was measured in triplicate by thermodilution. Ventricular pacing was then instituted and the hemodynamic and thermodilution measurements were repeated. The peak early filling (E) velocity was 0.34 +/- 0.11 m/s, and the peak atrial (A) velocity was 0.38 +/- 0.09 m/s. The mean E:A ratio was 0.91 +/- 0.3, and the median value was 0.88. In sinus rhythm the cardiac index of those patients with an E:A greater than 0.88 (group 1) was 1.97 +/- 0.32 l/min and those with an E:A less than 0.88 (group 2) was 1.76 +/- 0.50 (NS). During ventricular pacing the patients in group 1 (1.56 +/- 0.32 l/min) had significantly higher cardiac indices than those in group 2 (1.21 +/- 0.31 l/min) (P less than 0.02).(ABSTRACT TRUNCATED AT 250 WORDS)