SUMMARY A new method for obtaining absolute left ventricular volume from gated blood pool studies was evaluated in a torso phantom and in 35 patients who also underwent single-plane contrast ventriculography.Gated 400 left anterior oblique and static anterior views were acquired. Left ventricular volume at end-diastole was given by the ratio of the attenuation-corrected end-diastolic count rate from the gated study to the count rate per milliliter from a blood sample. Attenuation correction was made by dividing the end-diastolic count rate by e ud, where u = the linear attenuation coefficient of water and d = the distance from the skin marker to the center of the left ventricle in the anterior view divided by sin 400 to yield the depth of left ventricle in the left anterior oblique view. In the phantom studies, the correlation between radionuclide and true volume was 0.99 (radionuclide = 1.03 true -3 ml); the standard error of the estimate was 8 ml. In the patient studies, the radionuclide end-diastolic volume was used to calibrate the left ventricular time-activity curve, yielding left ventricular volume throughout the cardiac cycle. The correlation between radionuclide and angiographic enddiastolic volume was 0.95 (radionuclide = 0.97 angiographic + 3 ml); the standard error of the estimate was 36 ml. The correlation between radionuclide and angiographic end-systolic volume was 0.95 (radionuclide -1.01 angiographic + I ml); the standard error of the estimate was 33 ml. This method permits direct determination of absolute left ventricular volume without assumptions about the shape of the ventricle or the necessity of using regression equations to convert volume "units" to true volume.MEASUREMENT of left ventricular volume throughout the cardiac cycle would be useful in the diagnosis and treatment of patients with heart disease. Contrast left ventriculography has generally been used to measure volume, and is the standard for evaluating newer approaches. As described by Sandler and Dodge,", 2 the angiographic method assumes that the left ventricle is either an ellipsoid or a prolate spheroid, which may not be accurate for hearts with regional dysfunction.3 Angiography is invasive, making serial studies difficult, and may itself produce hemodynamic changes that affect left ventricular performance.4"7 Radionuclide scintigrams have been analyzed in an analogous way using dimensions to obtain volumes,8"-and although the invasive aspects and hemodynamic problems of contrast angiography are avoided, the same problems with geometric assumptions remain.A nongeometric approach with radionuclide tracer dilution curves has been described, proach has yielded only an index of left ventricular volume rather than an absolute measurement in milliliters.We describe a new nongeometric method for obtaining absolute left ventricular volume from gated blood pool studies.
Materials and Methods
Theoretical ConsiderationsIf a homogeneous distribution of activity is assumed to exist throughout the blood after injection and mixing of a bl...