Thermal dilution measurement of cardiac output in dogs using an analog computer.

Thermal dilution cardiac output determinations in dogs were compared to simultaneously performed Fick oxygen measurements. The purpose of this study was to validate in dog experiments a method for thermal dilution measurement which employs a double-thermistor catheter combined with an automatic computer as described by Olsson et al. Dilution and injectate temperature are entered directly into the calculation. The method does not employ logarithmic extrapolation, integration of the dilution signal being terminated when a preset cut-off level is reached. Errors due to recirculation, thermal capacitance of the right heart and heat exchange with the catheter's dead space require the use of an empirically derived correction factor, which in dogs was found to be significantly different from the factor used for human thermal dilution curves. With the appropriate cut-off level and correction factor a good agreement was found between the results of the thermal dilution and the Fick method. The regression equation for 47 experiments was found to be COtd = 0.95 COFick + 0.08; the correlation coefficient was 0.94.