Ventricular Nonmixing as a Source of Error in the Estimation of Ventricular Volume by the Indicator-Dilution Technic

1. In the anesthetized dog, end-diastolic, end-systolic and stroke volumes for the left ventricle have been calculated according to the method of Holt from dilution curves recorded at the aortic root. For heart rates of less than 120 beats per minute, the catheterdensitometer systems used gave 90 to 95 per cent of the true concentration in the aortic root at the end of diastole.2. Average volumes of 2.96, 1.63 and 1.33 ml./Kg. of body weight were calculated for each of these parameters. The interindividual and intraindividual variations about these means amounted to ± 30 to 50 per cent.3. In approximately 15 per cent of the dilution curves the concentration of dye in the aortic root during early systole (phase of maximal ejection) showed a transient, markedly low value compared with that of the end of diastole. This can be explained only on the basis of the preferential ejection from the heart in early systole of the undyed, newly entering blood from the left atrium, which had mixed poorly with the residual volume of the ventricle. Qualitative inspection of cineangiograms confirmed this impression and demonstrated that a residuum of contrast medium failed to mix with the incoming blood from the left atrium.4. In addition, there is evidence for the possibility of overestiination of stroke volumes from dilution curves recorded immediately proximal to the injection chamber.5. Incomplete mixing of the left artial blood with the residual volume of the ventricle precludes accurate quantitation of the changes of volume in the ventricle by the simple formulas proposed by previous authors. This phenomenon may serve to explain the considerable variability that is found in the present study and in the literature.