Pulmonary artery catheter: Rest in peace? Not just quite yet…

Patients with severe liver disease have a hyperdynamic circulation with an elevated cardiac output (CO), low systemic vascular resistance, and increased heart rate.1 During liver transplantation (LTx), most clinicians prefer to keep CO elevated in order to maintain tissue perfusion; this practice is based on common sense and supported indirectly by a clinical study by Parker et al.2 CO is one of the most important parameters that the anesthesiologist follows because it is a major factor in oxygen transport. CO, cardiac output; LTx, liver transplantation; PA, pulmonary artery; PAC, pulmonary artery catheter; SvO2, mixed-venous oxygen saturation; TEE, transesophageal echocardiography; USCOM, ultrasound cardiac output monitoring. Intraoperative hemodynamic management during LTx is based on the interpretation of the CO, heart rate, pulmonary artery (PA) pressures, filling pressures, and mixed-venous oxygen saturation (SvO2) and is influenced by the stage of the procedure and other factors (for example, the use of venovenous bypass).1 The pulmonary artery catheter (PAC) readily provides all this hemodynamic information, and every LTx anesthesiologist is very familiar with the PAC. Modified PACs give us continuous CO or additional information about the right ventricle (ejection fraction and end-diastolic and end-systolic volumes). The right ventricular ejection fraction catheter is used infrequently despite the better preload information it provides: filling pressures do not reflect preload very well because of compliance issues.3 Transesophageal echocardiography (TEE) gives other important hemodynamic information that is unavailable from the PAC: TEE allows the direct visualization of the volume status, overall contractility, regional wall motion, embolization, and large vessels.4 However, it does not provide the kind of hemodynamic information that is available from the PAC and is not as good at trending information (especially preload). Few LTx anesthesiologists are familiar with it, and the interpretation is highly user-dependent. Finally, TEE monitoring cannot be continued easily postoperatively. For these reasons, TEE is used fairly infrequently during LTx. The few anesthesiologists who do use TEE consider the information that it provides to be complementary to that provided by the PAC; in other words, TEE does not replace the PAC at this moment and is indicated only when there is a specific condition (for example, cardiac disease and pulmonary hypertension). In this issue of Liver Transplantation, 2 teams report that CO determined by ultrasound cardiac output monitoring (USCOM) correlates very well with thermodilution CO, the gold standard.5, 6 These reports on USCOM are important because they are the first to document that in these hyperdynamic patients there is also a good correlation. USCOM is a noninvasive technique (the probe is placed at the suprasternal notch) and therefore is expected to have a very low complication rate. Whether USCOM-derived information such as the corrected flow time and stroke volume variation will provide clinically valuable information regarding preload, as promised by some, remains to be seen. Also, USCOM estimations of PA pressures based on Doppler interrogation of the PA remain unvalidated (especially in those with pulmonary hypertension), and the ability to estimate left ventricular contractility is currently just a promise. Particularly worrisome is that 3 of 13 patients in one of these studies were excluded: in 2 patients, diagnostic-quality images could not be obtained, and this occurred after the patients were screened preoperatively to see whether the suprasternal insonation window was appropriate. Why is there such an interest in alternative monitoring devices? This is mainly the result of the large controversy regarding the efficacy and safety of the PAC in the last several years.7 However, consensus conferences have not been able to resolve this controversy: there is no high-level evidence for or against the PAC.7 Importantly, there is growing evidence of better outcome in high-risk patients in general, especially when the information is used in a systematic fashion.7 This would support the continuing use of the PAC during LTx, a high-risk procedure, even though there are no outcome studies in this patient population. Nevertheless, there is a recent trend of abandoning the PAC, even during LTx, especially in Europe. Then what are the alternative monitoring tools? Again, TEE is one alternative, but its use is limited to the operative period, and few LTx anesthesiologists have sufficient experience. Importantly, TEE provides little direct information on CO. USCOM, however, does provide direct, continuous, and accurate information on CO. So then, could USCOM become a viable alternative to the PAC? In my opinion, the answer is no. The main problem with abandoning the PAC is that hemodynamic management is based not just on CO but on other parameters as well; this other information cannot be provided by USCOM, at least not at this moment. For example, a sudden change in filling pressures or SvO2 is extremely valuable information that allows the proper detection and identification of certain intraoperative events. Could the simultaneous use of less invasive monitors, such as a combination of superior vena cava oxygen saturation (central venous SO2) with USCOM, provide a viable alternative to the PAC? Such alternative combined monitoring techniques have not been validated. It is even unclear whether central venous SO2 is a viable alternative to SvO2. More importantly, central venous catheterization is still required (it is also necessary to facilitate fluid and drug administration), and if we assume that a large number of complications reported with the PAC are due to complications resulting from central venous catheterization, then there is very little risk reduction. In conclusion, the value of the 2 studies on USCOM in this issue is limited to a validation of CO measurements with a noninvasive technique in patients undergoing LTx. It remains to be determined what the real clinical benefits are from the other measurements made by the device, such as the peak ejection velocity, minute distance, and systolic flow time. It is unclear whether USCOM can become an acceptable alternative monitoring tool when a PAC cannot be placed; those with expertise in TEE may prefer to insert a TEE probe. Finally, we may have to place more emphasis on the very early detection and correct identification of a major event, which likely has a much larger impact on overall outcome. Although it is possible that USCOM could provide some of the required information during such an event, there is likely going to be a delay by attempts to exclude technical errors. Thus, the publication of these 2 articles should not be interpreted as a message to abandon the use of the PAC during LTx.