Working the Angles

Cardiac output (CO) is the most commonly used index of the heart's overall performance as a phasic, self-actuated, variable speed, hydraulic pump. Under ideal circumstances, the stroke volume (SV) generated during each cardiac cycle is precisely measured from the right or left ventricular (RV or LV, respectively) volumes obtained at end-systole and end-diastole, thereby allowing calculation of continuous CO on a beat-to-beat basis as the product of heart rate and SV. However, this method of determining CO requires real-time measurement of continuous volume of either ventricular chamber,1 which, although feasible under controlled conditions in the cardiac catheterization laboratory or radiology suite, is clearly impractical when evaluating a hemodynamically unstable patient in the operating room or intensive care unit. In lieu of such a direct determination of CO from RV or LV volume itself, many other techniques for estimating CO have been developed since Fick's original approximation based on arterial and venous oxygen tension analysis,2 each of which has its own inherent advantages and limitations.3–6 Intermittent bolus pulmonary artery (PA) thermodilution continues to be the standard method by which CO is estimated in critically ill patients,7 but a number of variations (e.g., thermal filament heating, transpulmonary thermodilution, lithium dilution) based on this technique's principles have been successfully applied in many clinical settings.8 Several minimally invasive techniques, which avoid the well-known complications of PA catheterization,5 have also been developed (e.g., arterial pulse waveform contour analysis, esophageal Doppler ultrasound, transthoracic electrical bioimpedance),4 although questions remain about the accuracy and precision with which these methods are capable of reliably quantifying CO.9 CO may also be estimated using ultrasound interrogation of blood ejected from the LV or RV with transthoracic or transesophageal echocardiography. An ultrasound beam is directed along the path of proximal aortic or PA blood flow, and the frequency shift (Δf; the difference between transmitted and received frequencies) generated by the movement of red blood cells within the vessel is used to calculate the velocity (V) of blood flow from Austrian physicist Christian Andreas Doppler's famous equation: V = c · (Δf)/2ft · (cosθ), where c is the velocity of ultrasound in blood (1540 m/s), ft is the transmitted frequency, and θ is the angle at which the ultrasound beam intersects the blood flow. When blood flow velocity is measured using this technique, a parabolic velocity envelope is generated over the duration of ventricular ejection. The product of the area inscribed by this velocity envelope (known as the velocity–timeintegral [VTI] with units of length) and the cross-sectional area of the interrogated vessel describes a cylinder of blood that has moved through the ultrasound beam during ejection, the volume of which is the SV for the cardiac cycle. This method of determining SV and CO is somewhat labor intensive, demands a certain amount of technical expertise, and may distract the clinician from more urgent aspects of patient care. Nevertheless, a reasonable correlation between Doppler-derived CO obtained from ultrasound interrogation of the PA10 or the LV outflow tract11 (a technique with which cardiac anesthesiologists are especially familiar) and conventional intermittent bolus thermodilution CO has been previously demonstrated. Two essential requirements for accurately determining CO using the Doppler method should be readily apparent from this discussion, that is, accurate measurement of the diameter of the vessel in which blood flow velocity is measured and parallel alignment of the ultrasound beam with blood flow. Small errors in diameter measurement are magnified because the cross-sectional area of a circle varies by the square of its radius. The cosθ term in the Doppler equation indicates that blood flow velocity measurement is inherently angle dependent, and hence the magnitude of velocity and the resulting area of its inscribed envelope will be reduced if θ is not equal to 0° or 180° (the value of cosθ is less than 1). The latter observation is especially problematic because the angle of interrogation (θ) often cannot be determined with the degree of accuracy required to make truly quantitative measurements of blood flow velocity for more than a few consecutive cardiac cycles. As a consequence, the fundamental angle dependency of currently available ultrasound equipment (e.g., esophageal, intravascular,12 transthoracic echocardiography, transesophageal echocardiography) imposed by the Doppler equation often restricts the ability of such devices to precisely quantify continuous SV and CO. In the current issue of Anesthesia & Analgesia, Akamatsu et al. resurrect a clever technical approach13 and use an astute application of basic trigonometry to circumvent the angle dependency of intravascular Doppler velocity measurement using a modified PA catheter.14 The authors implanted a pair of adjacent miniature ultrasonic transducers (1 mm3 in size; pulse repetition frequency of 40 kHz; range of measurable velocities of 0 to 76.5 cm/s) facing away from each other in an orthogonal orientation approximately 2.5 cm from the tip of a conventional PA catheter. The transducers were fixed at 45°angles to the PA catheter and transmitted 2 simultaneous ultrasound beams (one directed toward and the other away from PA blood flow), each of which generated a separate blood flow velocity profile. In this way, the orthogonal transducers are able to effectively “triangulate” PA blood flow velocity, such that the error of Doppler beam alignment (cosθ) with flow associated with the first transducer is corrected by an opposing but complimentary beam misalignment of the second transducer. The authors provide mathematical proof that true PA blood flow velocity may be obtained as the square root of the sum of the squares of each transducer's velocity profile independent of the angle at which blood flow interacts with the PA catheter itself. The VTI of each composite PA blood flow velocity profile was then continuously calculated. The authors did not formally measure PA cross-sectional area, but instead calibrated their device using the PA catheter's conventional thermodilution function, generating an average SV from which a “correction coefficient” (incorporating PA cross-sectional area) was derived as the ratio of SV to VTI. This “correction coefficient” was then used to calculate continuous CO in subsequent experiments conducted in midazolam–fentanyl anesthetized, acutely instrumented dogs.14 The dual Doppler transducer-equipped PA catheter was correctly positioned in the main PA by optimizing the morphology and magnitude of the composite blood flow velocity waveform. Doppler CO was simultaneously compared with CO derived using an implanted PA electromagnetic flowmeter (considered a “gold standard” in laboratory CO measurement) and thermodilution over a range of CO values (0.47 to 3.03 L/min) produced by administration of dobutamine and propranolol. Indeed, Doppler CO was very highly correlated with CO obtained using the electromagnetic flowprobe (r2 = 0.99; P < 0.001) and nearly as well correlated with standard thermodilution CO (r2 = 0.85; P < 0.001). Bland–Altman analysis also demonstrated excellent limits of agreement and minimal bias between dual transducer Doppler CO and the 2 other established techniques. The preliminary results of Akamatsu et al.14 are quite promising, but many questions remain about the possible clinical applicability of this new device. The performance of the dual-transducer Doppler PA catheter in the presence of preexisting RV or LV dysfunction, hemodynamic instability, or significant underlying cardiac disease is untested, and further laboratory and clinical research will be required to identify the accuracy and precision which with the device is capable of quantifying continuous CO under these conditions. In particular, the ability of any available CO technology, including the Doppler PA catheter described here, to measure continuous CO may be most relevant during low-CO states. It is unclear, on the basis of the current data, how well the current device performs in such circumstances, although the dual-transducer Doppler PA catheter's correlation with electromagnetic flowprobe-derived CO appeared to be reasonable when CO was <1 L/min (a substantially reduced value in dogs). The current device may allow real-time, dynamic assessment of changes in CO in comparison with the time-averaged, modestly delayed continuous CO display of commercially available thermal filament PA technology, but whether this difference translates into a clinically meaningful benefit in hemodynamically unstable patients has not been defined. The requirement for thermodilution-derived calibration of the dual transducer Doppler PA catheter is rather cumbersome; instead, integration of M-mode echocardiography technology to measure beat-to-beat proximal PA diameter would be preferable because of the pressure dependence of PA dimension. The dual-transducer technology is incorporated into a conventional PA catheter, and therefore, patients in whom this new device is inserted will most likely be exposed to a similar set of potential adverse consequences in comparison with other PA catheters that are currently in clinical use.8 It also appears highly unlikely that this new device will affect the long-standing and as yet unresolved debate about the relative clinical efficacy of PA catheters in critically ill patients.15–17 Nevertheless, PA catheterization continues to be recommended for the hemodynamic monitoring of patients undergoing high-risk cardiac, thoracic, or major vascular procedures.18,19 The new dual-transducer Doppler continuous CO device described by Akamatsu et al.14 may theoretically eliminate many of the problems associated with the intermittent bolus thermodilution CO technique,8 but whether this new technology will successfully supplant its time-tested predecessor remains to be established. DISCLOSURES Name: Paul S. Pagel, MD, PhD. Contribution: This author wrote the manuscript. Attestation: This author approved the submission of the final manuscript. Name: Judith A. Hudetz, PhD. Contribution: This author wrote the manuscript. Attestation: This author approved the submission of the final manuscript.