In Vivo Tests

Abstract

In addition to an analysis of error sources and their influence on the accuracy of the result as will be given in Chapter XI, in vivoexpetiments are necessary. Besides the overall testing of the system, especially of the time programming by microcomputer, the effects of the geometric model fit and the interindividual variability have to be established in the real situation. During “steady state” of the circulation under controlled circumstances--especially with a constantly paced rhythm--thermodilution gives a measure for the cardiac output (from which the stroke volume can be derived by dividing through the heart rate) with a coefficient of variation of 4.3% (VAN DER WERF,1965). During this procedure, an electromagnetic flow meter can be calibrated, so that the unknown aortic lumen, haematocrit and other factors may be accounted for. In this way, stroke volume may also be established with varying heart rate. Although the blood flow into the coronary arteries escapes measurement--and this error may be estimated to be about 3% of mean flow--in determining stroke volume it may be neglected because the maximum coronary flow occurs well after the end of the ventricular systole.