Series Of Cross-sectional Images Research Articles

Measurement of regional myocardial blood flow with N-13 ammonia and positron-emission tomography in intact dogs

N-13 ammonia mimics certain properties of microspheres. It rapidly clears from blood into myocardium where it becomes fixed in proportion to myocardial blood flow. Used with positron emission tomography as a means for quantifying in vivo myocardial indicator concentrations, N-13 ammonia may be useful for noninvasive determination of myocardial blood flow with the arterial reference sampling technique. This possibility was examined in 27 experiments in 10 chronically instrumented dogs at control, high and low blood flows. Myocardial blood flow was calculated in vivo from the myocardial N-13 tissue activity concentrations derived from serial cross-sectional images of the heart, the 2 minute arterial input function and the withdrawal rate of arterial blood. These calculations were compared with blood flow determined by the standard microsphere technique. Blood flow determined in vivo with N-13 ammonia and positron emission tomography correlated with microsphere blood flow by y = -36.2 + 1.53x -0.0027x2 (r = 0.94 with a standard error of the estimate of 16 ml/min per 100 g). For flows from 44 to 200 ml/min per 100 g, the relation between in vivo and in vitro measured myocardial blood flow was nearly linear but reached a plateau at flows higher than 200 ml/min per 100 g. These results indicate that in dogs, blood flow in the physiologic range can be quantified in vivo with N-13 ammonia and positron emission tomography.

Accurate volume determination in the isolated ejecting canine left ventricle by two-dimensional echocardiography.

Two-dimensional echocardiography can provide serial cross-sectional images of the left ventricular cavity. We examined whether such serial images from steady-state ejecting hearts would allow three-dimensional reconstruction and accurate volume estimation without major geometric assumptions. Cross-circulated, paced dog hearts were suspended in a blood-filled tank. Serial cross-sectional images were taken at 3-mm intervals along the vertical axis. Left ventricular cavity and muscle areas of each image were planimetered with a light-pen system and summated for volume: total volume = sigma (areas x 3 mm). Direct left ventricular volume was measured through the cardiac cycle with a volumetric chamber connected to a balloon in the ejecting left ventricle. In six hearts, 67 separate direct volume measurements (range 9.5--54.7 ml) from various points in the cardiac cycle were compared with the simultaneous echo volume measurements. By least squares linear regression, echo volume = 1.01 (direct volume) - 0.44 ml; r = 0.972, SEE = 2.93 ml. Provided accurate cross-sectional localization is available, these studies suggest that extremely accurate steady-state left ventricular volume can be determined noninvasively in the ejecting heart from multiple cross-sectional images.