Volume 21, Issue 5 , Pages 629-631, October 2007
Cardiac Output From the Arterial Catheter: Deceptively Simple
Article Outline
RECENTLY, EDWARDS LIFESCIENCES, LLC, Irvine, CA, introduced a system (FloTrac/Vigileo) that calculates cardiac output (CO) continuously from the arterial pressure signal. It is unique among such systems in that it does not require calibration with another cardiac output assessment method. The system consists of a transducer (similar to any other pressure transducer) connected by a wire to a small computer/display box. All that is necessary is to connect it to an arterial catheter; turn it on; plug in a few patient numbers; zero the transducer; and, voila, cardiac output numbers! The entire process takes less than 5 minutes. Simple, right? Well, yes, as long as clinicians think cardiovascular physiology, the pathophysiology of various disease states, cardiovascular support technology, and the relationship between the arterial pressure wave and its associated flow are all “simple.”
It certainly is true that this device, which has been shown to be accurate, at least in postoperative cardiac surgical patients, is a wonderful piece of engineering. Its use is now spreading rapidly throughout the world in intensive care units and operating rooms. This is because it can be set up quickly and easily, does not require central venous access or central arterial access (as do pulmonary artery catheterization and transpulmonary thermodilution techniques, respectively), and it reports potentially valuable accurate hemodynamic information. The value of early goal-directed therapy in improving outcomes in the emergency room, intensive care unit, and operating room is now well documented.1, 2, 3, 4, 5 It is widely recognized that effective early goal-directed therapy requires either a measure of tissue oxygenation (eg, central venous oxygen saturation) or a surrogate (cardiac output). Pronounced stroke volume variation and pulse pressure variation over the respiratory cycle are both well known to predict potential fluid responsiveness,6, 7, 8, 9 and the FloTrac system quantifies stroke volume variation continuously. Fluid and pharmacologic therapy algorithms based on CO and stroke volume variation are now used, improving outcome in patients who are critically ill and/or undergoing major surgery. The FloTrac system is thus a very logical, minimally invasive choice for the implementation of hemodynamic management protocols.
Two articles appear in this issue of the Journal of Cardiothoracic and Vascular Anesthesia that add considerably to the assessment of this new technology. The study by Breukers et al10 is an investigation of the agreement of the system with intermittent thermodilution CO using a pulmonary artery catheter (PAC) in postoperative cardiac surgery patients. They studied 20 patients at 3 time points: T0, within 1 hour of surgery; T1, 3 hours after surgery; and T2, the morning after surgery. They found good agreement, with the pooled bias being −0.14 L/min and the precision (1 standard deviation) being 1.00 L/min. Also, the reproducibility of the FloTrac measurements appeared superior to that of the PAC (3% v 7.3%). The authors attributed this to the fact that thermodilution measurements were made at random points in the respiratory cycle, and the respiratory cycle has more marked effects on right-sided thermodilution than on left-sided measurements. Of particular clinical importance, they found that the two technologies tracked changes in CO with good concordance and discrimination. Because CO is used to track changes resulting from fluid therapy and other hemodynamic maneuvers, this agreement is of paramount importance, perhaps more important than agreement of the actual CO values themselves. One trend they noted was that as time went on, the FloTrac CO readings increased in relation to the PAC values. At T0, the FloTrac CO reading was, on the average 0.5 L/min less than that of the PAC, whereas at T2 the FloTrac reading was 0.37 L/min greater than that of the PAC. The authors speculate that because this was associated with an increase in blood pressure, an increase in vascular tone during the postoperative period may have contributed to this phenomenon.
This means it is necessary to step back and reexamine what is known about how this device works. The basic principle is that stroke volume and arterial pulsatility are proportional to one another, and the proportionality constant, κ, is a number describing the resistance and compliance of the arterial tree: 
Pulsatility is calculated by using the standard deviation of the arterial pressure wave over a 20-second period. The κ value, calculated every minute by the latest operating system (v. 1.07), is based on patient weight, height, age, mean arterial pressure, skewness (shifting of individual waves left or right), and kurtosis (“shortness” of the individual waves). The exact calculation method and the relative weight given each factor are proprietary information. It can be imagined, however, if mean arterial pressure is not weighted as heavily as perhaps it should be under certain circumstances, the κ value might misrepresent the arterial tone. This factor as well as the fact that the device showed a slight tendency to overestimate low CO values and underestimate high values should lead the company to revisit and possibly adjust the κ value calculation algorithm in the next operating system.
Lorsomradee et al11 in their article begin to answer the following important question: “what happens to the FloTrac readings when the arterial wave pulsatility is affected by factors other than simply the stroke volume?” They found that, in the case of aortic regurgitation, the FloTrac value was consistently higher than that of the continuous cardiac output (CCO) PAC. Again, going back to how the device works and considering the primary hemodynamic aberration in aortic regurgitation is important. The pulse pressure is abnormally high in aortic regurgitation because of a low diastolic pressure. The low diastolic pressure results from backward flow through the aortic valve during diastole. The FloTrac, measuring stroke volume on the left side of the circulation by calculating pulsatility, has no way to “know” that blood is flowing backwards during diastole. It “knows” only that the pulsatility, or standard deviation of the arterial waveform over 20 seconds, is high. This is always of theoretical concern in arterial pressure–based CO assessment technologies, and Lorsomradee et al quantify the overestimation of net CO relative to the CCO PAC very nicely. Perhaps, in future aortic regurgitation patients, calculating such a difference might be useful in determining the regurgitant fraction.
In the case of intra-aortic balloon pump support, the FloTrac, in many instances, threw its hands up in the air and said “I don’t get it.” This is probably because the system has an algorithm to define when “beats” or “stroke volumes” start and end. The biphasic waves of balloon counterpulsation simply “do not compute,” and the machine interprets them as an unstable signal.
A third finding was that, upon phenylephrine injection or sternotomy, the FloTrac overestimated the CO relative to the CCO PAC. The authors do not say exactly how long after these events the comparisons were made, but the implication is that the comparisons were rather immediate. The timing of the comparisons is crucial. The FloTrac calculates a new CO value every 20 seconds, whereas it takes longer than that, depending on thermal noise, for a new value to be reported by the CCO Vigilance monitor (Edwards Lifesciences, LLC). So, when sudden hemodynamic changes occur, it should be expected that the PAC CCO will lag behind the FloTrac. Indeed, this is the case, as noted in my previous investigation (Fig 1).12
Fig 1.
CCO measurements using PAC and arterial pressure–based cardiac output (APCO) over time. The CCO changes are delayed and blunted compared with those of APCO. The 2 measurement methods agree closely during “steady-state” conditions.
Another important factor is the frequency with which the κ value is recalculated. In the original operating system (v. 1.0), which was used in the study depicted in Figure 1, the κ value was recalculated only every 10 minutes. Even with the most recent operating system, in which the κ value is recalculated every minute, a bolus of phenylephrine can result in an abrupt increase in pulsatility that is initially interpreted by the FloTrac as an increase in stroke volume. Of course, phenylephrine does not normally cause an appreciable increase in stroke volume but simply increases vascular tone. Over a period of 1 to 2 minutes, the FloTrac “learns” of the new, increased vascular tone, recalculates κ, and then reports stroke volume values more accurately. If the investigators made their comparisons before the recalculation of κ, they revealed an overestimation of CO by the FloTrac versus the CCO catheter. The same would be true of sternotomy. Sternotomy probably led to an increase in vascular tone from sympathetic nervous system stimulation. In this case, as with the administration of phenylephrine, the FloTrac likely interpreted the initial increase in pulsatility as an increase in stroke volume. The latest operating system, v. 1.07, in calculating a new κ value every minute, is a major improvement over previous ones; the period of “reactivity” on vasoconstriction administration is now much shorter. In my personal experience with version 1.07, I still notice this falsely high CO immediately after vasoconstrictor administration as well as falsely low CO immediately after administration of a pure vasodilator (nitroprusside). As expected, however, these events are now brief (correction appears to occur within 1-2 minutes).
The point of this discussion is not that one technology is “better” or “more accurate” than the other but rather that the user must understand the physiology of the circulatory system, the pathophysiology of the disease processes, and the mechanism by which each technology produces its data in order to use and compare the technologies effectively. For example, when a patient is removed from cardiopulmonary bypass support and the intermittent thermodilution curve does not return to baseline, we know not to rely on the reported cardiac output. We also realize that the situation will be corrected after the thermal changes in the chest “settle down.” This is not to say the PAC is “bad” or “wrong” but rather that it has limitations under certain circumstances. Likewise, the FloTrac is not generically “accurate” or “inaccurate.” It has certain limitations in its current form that, as far as this author is concerned, do not detract from its potential usefulness in the vast majority of situations. The proper use of any monitoring technology requires knowledge both of its strengths and its limitations. “Kudos” to both groups of authors for giving us further insight into the performance and potential pitfalls of this most interesting and attractive technology.
References
- Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med. 2001;345:1368–1377
- Randomised controlled trial assessing the impact of a nurse delivered, flow monitored protocol for optimisation of circulatory status after cardiac surgery. BMJ. 2004;329:258–261
- Early goal-directed therapy after major surgery reduces complications and duration of hospital stay (A randomised, controlled trial [ISRCTN38797445]). Crit Care. 2005;9:R687–R693
- Goal-directed intraoperative fluid administration reduces length of hospital stay after major surgery. Anesthesiology. 2002;97:820–826
- . Intraoperative intravascular volume optimisation and length of hospital stay after repair of proximal femoral fracture: Randomised controlled trial. BMJ. 1997;315:909–912
- Stroke volume and pulse pressure variation for prediction of fluid responsiveness in patients undergoing off-pump coronary artery bypass grafting. Chest. 2005;128:848–854
- Prediction of fluid responsiveness in patients during cardiac surgery. Br J Anaesth. 2004;93:782–788
- Non-invasive prediction of fluid responsiveness during major hepatic surgery. Br J Anaesth. 2006;97:808–816
- Assessment of fluid responsiveness in mechanically ventilated cardiac surgical patients. Eur J Anaesthesiol. 2005;22:658–665
- Cardiac output measured by a new arterial pressure waveform analysis method without calibration compared with thermodilution after cardiac surgery. J Cardiothorac Vasc Anesth. 2007;21:632–635
- Uncalibrated arterial pulse contour analysis versus continuous thermodilution technique: Effects of alterations in arterial waveformy. J Cardiothorac Vasc Anesth. 2007;21:636–643
- . Cardiac output determination from the arterial pressure wave: Clinical testing of a novel algorithm that does not require calibration. J Cardiothorac Vasc Anesth. 2007;21:3–7
PII: S1053-0770(07)00225-X
doi:10.1053/j.jvca.2007.07.001
© 2007 Elsevier Inc. All rights reserved.
Volume 21, Issue 5 , Pages 629-631, October 2007
