A physical model of the thermodilution method: influences of the variations of experimental setup on the accuracy of flow rate estimation.

The thermodilution method has been widely used to estimate cardiac output by injecting a cold solution into circulating blood. It is uncertain if radial heat transfer from the vascular/cardiac wall to circulating injectate can cause inaccurate results with this method. In this study, we have introduced a physical experimental model of the thermodilution method without recirculation of the cold solution. To test the accuracy of the thermodilution method, the experimental setup included an aluminum tube to allow radial heat transfer. Variations of the following parameters were conducted: (i) the real flow rate, (ii) the distance between injection point of cold solution and the temperature sensor, (iii) the volume of injectate, and (iv) the temperature of injectate. By following the above variations, we have calculated different correction factors eliminating the influence of radial heat transfer on the estimation of flow rate by the thermodilution method. The results indicate that changes in both injectate temperature and volume have no influence on the estimation of flow rates. The experimental variations, which can cause greater radial heat transfer, seem to be responsible for the result of the smaller estimation of the flow rate than the real value. These variations include (i) a decreased real flow rate and (ii) increased distances between the injection point of cold fluid and the thermosensor. Such an incorrect estimation could be eliminated by using correction factors. The correction factor seems to be a function of the area of the thermodilution curve, assuming no recirculation.