264

Introduction: A practical approach for achieving continuous and minimally invasive cardiac output (CO) monitoring in critically ill patients is to analyze an arterial blood pressure (ABP) waveform obtained with an existing arterial line. However, current pulse contour analysis (PCA) devices may not be able to adequately predict CO reductions during blood loss. Recently, a new PCA technique was introduced that aims to enhance accuracy by analyzing the slow, inter-beat variations in the waveform wherein confounding physiologic effects are minimal. Hypothesis: The aim was to test this technique against gold standard aortic flow probe measurements in hemorrhaging animals. Methods: Six healthy swine were studied under a protocol approved by the Michigan State All-University Committee on Animal Use and Care. During an open-chest recovery surgery, an ultrasonic flow probe was implanted around the ascending aorta for reference CO. After about a week of recovery, general anesthesia and mechanical ventilation were instituted. A fluid-filled catheter was placed in a radial artery for the ABP waveform for analysis. The measurements were recorded during graded hemorrhage (to about 50% of the estimated total blood volume) and re-transfusion of the shed blood. The new technique and competing PCA devices were applied to the ABP waveforms. The CO estimates of each technique were then calibrated using the mean aortic flow probe value per subject. The agreement between the calibrated estimates and reference measurements was assessed via Bland-Altman analysis. Results: On average, CO dropped to 36% of its baseline value during the hemorrhage, while heart rate increased by 150% perhaps due to a central ischemic response. The new technique correctly predicted CO reductions in all animals and then a CO increase after the blood re-transfusion. The CO error ranged between 9 to 21% per subject, with an average of 14%. Two competing devices did not predict CO reductions during the hemorrhage, whereas the average CO error of a third device was 33%. Conclusions: The new PCA technique was able to reliably track CO changes during major hemorrhage. Future efforts will focus on the integration of the technique into a real-time device.