In our opinion, TAVI is not an appropriate setting to compare the ClearSight and PAC methods because they provide measurements at two different points of circulation, namely, downstream and upstream of the aortic valve, respectively. During aortic valve placement, surgical maneuvers alter the blood pressure waveform, and consequently, pulse contour analysis and CO computing. Furthermore, the PAC and ClearSight measure the right and left ventricle CO, respectively. The two CO measurements should be equivalent if the tricuspid and aortic valves are normal. However, in the case of tricuspid regurgitation (for the PAC) and aortic stenosis and/or regurgitation (for ClearSight), CO measurements can be profoundly affected. Among the limitations of the study, the authors did cite the issue of aortic regurgitation; however, they should have better clarified how aortic valve diseases would affect the measurements. Such differences should be expected even without a specific targeted study.

Moreover, blood pressure, which is the result of the interaction between the heart, blood volume, and circulatory bed, is expected to differ depending on whether it is measured invasively or non-invasively [2]. Conversely, Wang et al. [3] found both acceptable agreement and discrepancies between these two methods of hemodynamic monitoring during cardiac surgery. Using the same devices in a larger sample, the authors found that the mean ClearSight CO was 4.21 L/min and the mean PAC CO was 3.90 L/min, with a mean bias of 0.32 L/min, a 95% CI of 0.22–0.42 L/min, and a percentage error as low as 26.4%. We believe that Wang et al. [3] considered the limitations of comparing the two different CO measurement sites and the effect of valvular disease because they excluded patients with cardiac valvular disease of the tricuspid and aortic valves. In addition, the method used by Yahagi et al. [1] to average the ClearSight CO is problematic, since they used only two values to compute the average ClearSight CO value, whereas they used six values to compute the average PAC CO.

We also believe that their analysis of systemic vascular resistance (SVR) deserves some comment. Currently, the use of SVR (or SVRI if indexed) is a debatable issue. As reported by Magder, the principles behind SVRI computation are not applicable to human circulation: Poiseuille’s Law implies that the resistance within a pipe is determined by the ratio of the pressure gradient between the two ends over the inside flow. However, human circulation cannot be considered a pipe or electric circuit. In fact, this oversimplification does not consider the auto-regulatory mechanisms, one for each organ, which make blood flow vary according to the tissue demand, which we cannot currently “measure” at bedside [4,5]. Furthermore, these mechanisms are affected differently by anesthesia and various diseases. Thus, the standard SVRI computation method is limited and ineffective, if not misleading, for hemodynamic assessment. In addition, from a mere mathematical perspective, computing the SVRI value from the CO to explain the discrepancy between CO measurements is redundant and will not lead to any significant finding.

In conclusion, if properly employed, the non-invasive method of CO measurement is generally suitable and reliable for the intraoperative assessment and management of hemodynamics. However, the operator should be aware of the advantages and disadvantages of such devices. The surgical setting must also be considered as a critical factor that can limit the reliability of hemodynamic assessments. Finally, we emphasize that the cardiac index is preferable over the CO as it represents a more patient-related value of blood flow.