[The effects of changes in mean aortic pressure on the reliability of cardiac output determination from the aortic pressure curve (author's transl)].

In seven postoperative patients the effects of changes in mean aortic pressure on the reliability of cardiac output determination from the aortic pressure curve by the method of Wesseling et al. [8] were tested. Reference cardiac output (CO) was measured by thermal dilution technique. The wave impedance Zo, needed for the "contour method" of Wesseling, was derived from the initial CO-determinations. Overall correlation of both methods was good with a correlation coefficient of 4 = 0.8740. The percentage error of the contour method was -4.59 +/- 25.01%. In 4 of 7 patients there was an individual differing effect of pressure changes on the impedance Zo and by this on the error of the method. 1/Zo was clearly correlated to mean aortic pressure (r = -0.4021 to r = -0.8381). To examine the reliability of contour method also in great variations of mean aortic pressure additionally 6 aortic vessels of dogs and 1 human aortic vessel were tested in a perfusion model. In all cases increase of blood pressure was followed by a systematic failure of the contour method between +77% and +1112%. Corresponding to the mean pressure the individual failure curves were of exponential character. 1/Zo and mean aortic pressure showed high negative correlation coefficients (r = -0.9056 to r = -0.9923). The pressure dependent changes of the impedance Zo were comparable to the corresponding changes of volume elasticity coefficients. Subsequently for every vessel and patient respectively an estimation function for pressure dependent impedance Zo was calculated by use of two CO-determinations at different pressure levels. The correction of Zo by these functions resulted in a considerable reduction of the pressure dependent failure.