Improved Right Heart Function With a Compliant Inflow Artificial Lung in Series With the Pulmonary Circulation Scott D. Lick, MD, Joseph B. Zwischenberger, MD, Dongfang Wang, MD, Donald J. Deyo, DVM, Scott K. Alpard, BA, and Sean D. Chambers, PhD Departments of Surgery and Anesthesiology, The University of Texas Medical Branch, Galveston, Texas, and MC3 Corporation, Ann Arbor, Michigan
Background. We previously reported a 50% incidence of immediate right heart failure using a rigidly housed, noncompliant inflow artificial lung in series with the pulmonary circulation in a healthy ovine survival model. Three device modifications resulted: (1) an inflow cannula compliance chamber, (2) an inlet blood flow separator, and (3) modification of the artificial lung outlet geometry, all to reduce resistance and mimic the compliance of the pulmonary vascular bed. Methods. In 7 sheep, arterial grafts were anastomosed end-to-side to the proximal and distal main pulmonary artery, with the paracorporeal artificial lung interposed. A pulmonary artery snare between anastomoses diverted full pulmonary blood flow through the artificial lung for up to 72 hours.
Results. Six of 7 sheep exhibited good cardiac function throughout the test period: mean central venous pressure was 6.8 mm Hg (range, 4 to 11 mm Hg), mean cardiac output, 4.17 ⴞ 0.12 L/min (range, 2.4 to 6.3 L/min); before and after device mean pulmonary arterial pressure, 21.8 and 18.5 mm Hg, and left atrial pressure, 10.8 mm Hg. Conclusions. This modified artificial lung prototype with an inflow compliance chamber, blood flow separator, and modified outlet geometry has greatly improved cardiac function and initial survival in our healthy ovine model.
A
We have implanted prototype ALs in a healthy ovine survival model in series with the pulmonary circulation (pulmonary artery-to-pulmonary artery) for up to 7 days of survival [4]. Immediate right heart failure in this initial series, though, was 50%. The resultant AL modifications for this study include an inflow separator and a modification of the outlet geometry, both designed to reduce flow resistance, and an inflow compliance chamber, designed to mimic the compliance of the pulmonary bed. The modified prototype AL achieved total gas exchange with greatly improved cardiac function and survival at 48 to 72 hours in healthy sheep.
lthough mechanical ventilation and extracorporeal membrane oxygenation have been used successfully as bridges to transplantation for periods of time up to weeks [1–3], each has limitations. Mechanical ventilation allows only partial support, limited by the ventilatory and gas exchange capabilities of the (remaining) lung parenchyma. Extracorporeal membrane oxygenation is labor-intensive, expensive, and nonambulatory. Our objective is to develop a paracorporeal artificial lung (AL) to be used as a clinical bridge to transplantation or recovery. For successful application, the patient must be hemodynamically stable, and the AL should provide total gas exchange (O2 and CO2). The device should allow ambulation, and to decrease posttransplantation infection risk, be completely removable at the time of transplantation. To simplify the device, and to avoid the blood trauma of a pump, we prefer the AL be powered by the patient’s right heart. Finally, the device should be tolerated in series with the pulmonary circulation, to allow for complete gas exchange, and hence must be of ultralow impedence to avoid right heart failure.
Presented at the Poster Session of the Thirty-seventh Annual Meeting of The Society of Thoracic Surgeons, New Orleans, LA, Jan 29 –31, 2001. Address reprint requests to Dr Lick, Department of Cardiothoracic Surgery, University of Texas Medical Branch, 301 University Blvd, Route 0528, Galveston, TX 77555-0528; e-mail:
[email protected].
© 2001 by The Society of Thoracic Surgeons Published by Elsevier Science Inc
(Ann Thorac Surg 2001;72:899 –904) © 2001 by The Society of Thoracic Surgeons
Material and Methods Artificial Lung Design The AL is designed and manufactured by MC3 Corporation (Ann Arbor, MI) to have minimal resistance and impedance to right heart flow. Whereas conventional cardiopulmonary bypass oxygenators have blood enter the outside of a rigid housing, flow through the fiber bundle, and then exit across the housing, the AL prototype has blood enter through a center channel into the middle of a cylindrical fiber bundle, flow radially around evenly spaced, parallel-wound fibers, and then flow out a tangentially placed outlet (Fig 1). All blood flow must cross gas-exchange fibers to reach the AL outlet. The 0003-4975/01/$20.00 PII S0003-4975(01)02842-9
900
LICK ET AL COMPLIANT ARTIFICIAL LUNG
Ann Thorac Surg 2001;72:899 –904
Fig 1. Initial design of the MC3 artificial lung.
mean pressure drop with a conventional cardiopulmonary bypass oxygenator is approximately 25 mm Hg; bench testing of the initial AL design showed only an 8-mm Hg drop at 5 L/min flow. In the first-generation AL prototype (used in our first series of sheep, which had 50% right heart failure), flow visualization studies showed most of the inflow going directly to the distal end of the AL, and a relative lack of flow to the proximal fibers near the blood inflow port. To distribute the inflow in a progressive pattern, an inflow separator was inserted into the center of the fiber bundle. The geometry of the outlet was also altered, thus enlarging its effective diameter along the fiber bundle axis. These design changes dropped the bench-tested transdevice pressure gradient from 8 mm Hg to 6 mm Hg at 5 L/min of steady flow. Because of the dynamic, pulsatile nature of flow and pressure, resistance to continuous flow alone does not determine right heart work as it ignores the capacitance and compliance in the circulation. As such, right heart work is most accurately quantified by calculating the pulmonary input impedance. Impedance is determined by performing a Fourier series analysis of the pressure and flow profiles, thus yielding the magnitude and phase of each profile at the respective frequency, or harmonic, component. Impedance is then defined as the ratio of the pressure magnitude to the flow magnitude at each harmonic [5]. Mathematical modeling of an AL attached to the pulmonary circulation shows that adding compliance to the system is the most effective method in lowering the pulmonary inflow impedance [6]. The closer the compliance element to the right heart outflow tract, the more effectively it dampens resultant impedance harmonics. Thus, the MC3 device has an inflow cannula compliance chamber (Fig 2). The compliance chamber consisted of a solvent-casted, polyurethane bulging segment encased in an open-face housing, so that pressure would not be generated external to the chamber. The chamber was passive, accepting the stroke volume by filling the bulg-
ing section without significant elastic expansion. The compliant chamber was designed to significantly lower the impedance of the AL, thus decreasing the work of the right ventricle on attachment of the AL to the pulmonary artery (see Appendix). Bench-top experiments, using a mock circuit and pulsatile pump that generates a flow pulse similar to the right ventricle, demonstrated that the compliance chamber decreased the impedance modulus at harmonics 1 through 7 by an average factor of 15 times (Fig 3).
Implantation All animals received humane care according to the “Guide for the Care and Use of Laboratory Animals” (1996) prepared by the U.S. Department of Health and Human Services and published by the National Institutes of Health. The study was approved by the Institutional Animal Care and Use Committee of The University of Texas Medical Branch, Galveston, TX, with strict adherence to the Institutional Animal Care and Use Committee guidelines regarding humane use of animals. Our management of the sheep parallels our standards of patient care. The principal investigators make daily rounds on
Fig 2. Modified artificial lung with inflow compliance chamber.
Ann Thorac Surg 2001;72:899 –904
LICK ET AL COMPLIANT ARTIFICIAL LUNG
901
Fig 3. Input impedance of the MC3 artificial lung (IAL) compared with that of one with a compliant chamber placed at the inflow to the artificial lung. The impedance was measured in an in vitro, pulsatile circuit that yielded a flow and pressure profile similar to the pulmonary artery. The results are for an average flow rate of 5 L/min and a pressure pulse of 22/9 mm Hg.
the sheep to outline and supervise management. Our management team also consists of a full-time staff veterinary anesthesiologist who oversees all anesthesia, sedation, and animal management issues. Medical students volunteer for 24-hour cageside care 7 days per week. Institutional Animal Care and Use Committee personnel, with no conflict of interest, make daily rounds to check compliance with the animal management protocol. Our ovine implantation technique [7] and perioperative anesthetic care [8] have been described in detail. All sheep were anticoagulated with continuous intravenous heparin, titrated to keep activated clotting time between 230 and 300 seconds. Seven healthy adult ewes were used for the experiment. Briefly, through a left fourth interspace thoracotomy, 16-mm or 18-mm polyethylene terephthalate fiber arterial grafts bonded to 5/8-inch silicone tubing are anastomosed end of graft–to–side of proximal and distal pulmonary artery. The cannulas are connected to the MC3 prototype AL. The long ovine main pulmonary artery, averaging 5.5 cm [9], allows room for a snare between anastomoses. The snare is occluded, diverting full blood flow through the AL. A left atrial pressure catheter is placed, and the wound is closed over a pleural and pericardial drain. A pulmonary artery catheter (Edwards Critical Care, Irvine, CA) placed preoperatively, remains unmanipulated throughout the operative and postoperative period. A custom-made Doppler flow probe (Transonic Inc, Ithaca, NY) is placed on the AL outflow graft. The animals are extubated, returned to their cages, allowed to awaken, and then transferred to the ovine intensive care unit where they are allowed free access to food and water until sacrifice. The inflow compliance chamber of this series of devices added approximately 6 inches to total length when compared with the first-generation prototype AL. To keep the longer device from contacting the back of the animal’s cage, we shortened the overall length by exiting the cannulas through a 4-cm second thoracotomy inci-
sion with sixth rib resection, rather than tunneling subcostally, as is commonly done with ventricular-assist device cannulas.
Results The artificial lungs averaged 220 mL/min O2 and 166 mL/min CO2 gas exchange (O2 and CO2 exchange were calculated by the Fick equation from simultaneous blood withdrawal through 25-gauge needles temporarily inserted into device inflow and outflow cannulas to measure change in O2 saturation). Six of the 7 sheep exhibited good cardiac function: mean central venous pressure was 6.8 mm Hg; mean cardiac output was 4.17 L/min; before and after device mean pulmonary arterial pressures were 21.8 and 18.5 mm Hg; and mean left atrial pressure was 10.8 mm Hg. Artificial lung resistance averaged (21.8 ⫺ 18.5 mm Hg)/4.17 L/min ⫽ 0.79 Wood units. By comparison, the noncompliant AL group averaged 2.8 L/min cardiac output, with AL resistance 7 mm Hg/2.8 L/min ⫽ 2.5 Wood units. Two sheep exsanguinated through the interstices of their knitted arterial grafts 3 and 5 hours postoperatively. We changed to pretreated, woven grafts for the remainder of the animals and had no further short-term graft bleeding. Two sheep were electively sacrificed at 48 hours because of a falling hematocrit caused by diffuse, slow bleeding into the chest; two others were electively sacrificed at 72 hours (planned termination of the study). The seventh sheep died approximately 1 hour after implantation with low cardiac output and low filling pressures; no cause of death could be determined at autopsy. We consider this an anesthetic-related death. Autopsy studies of all 7 sheep showed no evidence of pulmonary emboli. All 48-hour and 72-hour survivors had bilateral lung atelectasis and between 500 mL and 1,000 mL of bloody fluid in the left pleural space, likely from diffuse oozing exacerbated by heparin. No clots
902
LICK ET AL COMPLIANT ARTIFICIAL LUNG
were found in the ALs. Two ALs failed to exchange gas after 24 hours, but blood flow and cardiac and pulmonary pressures remained unchanged. These were exchanged for functioning ALs without difficulty.
Comment This modified AL prototype with an inflow compliance chamber, blood flow separator, and modified outlet geometry has greatly improved cardiac function and initial survival in our healthy ovine model. In our earlier series [7], initial right heart failure was 50% (4 of 8 sheep), as manifest by high central venous pressure, visible right heart distension, and low cardiac output. In the first series, we attempted to overcome right heart failure in two ways. In 2 sheep we used inotropic agents (dopamine, epinephrine, and calcium gluconate), but the animals died with high central venous pressure and low cardiac output within hours. In the other 2 sheep we partially released the pulmonary artery snare (AL blood flow, 1 L/min, with the remaining right heart output passing through the main pulmonary artery), but the AL clotted within hours, despite intravenous heparin infusion titrated to keep the activated clotting time more than 250 seconds. This latter group of 2 sheep successfully underwent device change-out on the next elective operating day (second and third postimplantation day) using balloon catheter thrombectomy of the cannulas, and both animals immediately tolerated full pulmonary artery snare (full blood flow through the AL). Right heart work is highest when an AL is placed in series with the pulmonary circulation [6]. To decrease right heart work, AL design must mimic the high compliance and low resistance of the natural pulmonary circulation. The pulmonary circulation has only onetenth the capacity of the systemic circulation yet must accommodate the same ejected volume [10]. The pulmonary vessels, which are much shorter than the systemic vessels, accommodate this volume because of thinner walls and greater compliance. Our second-generation AL prototype incorporates three simultaneous changes to decrease right heart work: an inflow separating cone, which improves dispersion of flow, rendering it more uniform across the fiber bundle; an increase in the outlet effective diameter along the fiber bundle axis; and an inflow cannula compliance chamber, which preserves right ventricular pulsatile work, spreading it over time during diastole. The resultant AL modifications therefore reduce resistance and more closely mimic the compliance of the native pulmonary circulation. An AL may be placed in parallel (pulmonary artery– to–left atrium), or with a combined outflow to both the left atrium and pulmonary artery (hybrid configuration), or in a pure pulmonary artery–to–pulmonary artery configuration. The in-parallel configuration diverts blood away from the native lungs and toward the left atrium; if a snare is placed on the pulmonary artery distal to the anastomosis, most or all blood is diverted through the AL. If fully snared, this guarantees full gas exchange through the AL, but deprives the pulmonary vascular
Ann Thorac Surg 2001;72:899 –904
bed of blood flow. The lungs are the dominant site of processing vasoactive compounds (activation of the vasoconstrictor angiotensin; deactivation of the vasodilators prostaglandin E and bradykinin). Vasodilators are also deactivated, but on the whole the vasoconstricting effect predominates, leading to a drop in blood pressure if blood flow through the lungs is reduced [11]. Moreover, complete exclusion of pulmonary blood flow might predispose to pulmonary bed thrombosis, as has been demonstrated in cardiopulmonary support with extracorporeal membrane oxygenation in a failed transplanted heart with minimal cardiac ejection across the pulmonary bed [12]. If the pulmonary artery is unsnared (competitive flow), some pulmonary blood flow is preserved, and right heart work is decreased by the addition of an alternate resistance bed in competition with the native lungs, so that flow through each (lung or AL) is determined by relative impedances. A pulmonary artery–to–left atrium native lung competitive configuration with a prototype similar to the device used in this study (without a compliance chamber) has been used successfully for up to 1 week [13]. Blood flow through the native lungs varied from 10% to 90%, and averaged 47%. The competitive flow configuration would clearly have an advantage in patients with pulmonary hypertension: it would allow a decrease in right heart work. However, the potential drawbacks of this competitive pulmonary artery–to–left atrium configuration are (1) right-toleft microemboli or macroemboli, especially during device change-out; (2) partial gas exchange; and (3) massive rightto-left shunting if the AL must be removed from the circuit and replaced with a loop (eg, when the patient is intolerant of anticoagulation). The pulmonary artery–to–left atrium configuration also has less bundle flow pulsatility than the pulmonary artery–to–pulmonary artery configuration. The pulsatility of blood flow through the fiber bundle is of concern for two reasons. First, oxygen transfer is more efficient when blood passes through the fiber bundles in steady rather than pulsatile flow [14]. Second, pulsatility creates periods of high and low shear stress; the low-stress periods predispose to thrombosis [15]. However, a compliance chamber helps to lower bundle flow pulsatility. Despite these inherent disadvantages of additive resistance beds and increased bundle flow pulsatility, we continue to pursue the in-series configuration because we predict it will have the most long-term clinical utility. Experience with long-term ventricular-assist device support shows that critically ill patients needing mechanical assistance are often intolerant of any anticoagulation for periods of time that can last up to weeks. With an AL in the pulmonary artery–to–pulmonary artery configuration, the AL could be removed and replaced with a loop connected to each cannula, the anticoagulation stopped, and the native lungs used to support the patient until anticoagulation can be safely restarted. A standby flow loop could not be used in any form of the pulmonary artery–to–left atrium configuration, as it would create massive right-to-left cardiac shunting and hypoxia. Although antithrombotic surface coating will likely be used
Ann Thorac Surg 2001;72:899 –904
in a production model of an AL, anticoagulation will still be necessary when an AL is used long term. An AL has inherent problems of less complete washout, more surface area, and a less favorable shear profile [16] compared with a ventricular-assist device. We are cognizant that the pulmonary artery–to– pulmonary artery configuration will not be applicable to patients with fixed pulmonary hypertension. Pulmonary vascular resistance, though, often is not fixed, but active and variable. Oxygenated blood is a known pulmonary vasodilator, and has been shown effective in an ovine model of acute lung injury [17]. We noticed that initial washout from the AL often led to an acute rise in pulmonary pressures, which reversed within a few minutes after starting oxygen flow through the AL. We have not tried pharmaceutical selective pulmonary vasodilators (nitric oxide, prostaglandin E) in conjunction with an AL. The MC3 AL has improved through multiple modifications leading to the current prototype. The casings were made by a rapid prototyping technique using polyester resins inside a computer-generated silicone rubber mold. This allowed multiple design iterations and prototype fabrication with ease. However, the necessary biomaterials are not especially durable, and the tolerances are imprecise relative to production membrane oxygenators. In essence, these are hand-built prototypes. Hence, there were multiple device failures requiring change-out in the series. The next step will be creation of a clinical product, using longer-term and more durable (but costly) techniques, and with an anticipated reduced failure rate. Similarly, we have not studied the blood–surface interactions at this early development stage because the final biomaterials will likely have considerably different characteristics (heparin-bonded or otherwise coated casing and potting; silicone impermeable fibers) and be more blood compatible. At this phase of development, we have focused on feasibility of design, physiologic performance, and initial survival. Because this modified prototype has greatly improved right heart function, we are optimistic about the clinical future of the AL in series with the pulmonary circulation. We next plan to study the device in our smoke and burn lung injury sheep (lethal dose 50 and lethal dose 100) models of adult respiratory distress syndrome [18] in the pulmonary artery–to–pulmonary artery configuration. Ultimately, we see the AL used as a bridge to transplant or recovery for patients with relatively normal (or minimally elevated and reversible) pulmonary resistance.
LICK ET AL COMPLIANT ARTIFICIAL LUNG
4.
5. 6. 7. 8.
9.
10. 11.
12.
13.
14. 15. 16.
17. 18.
19. 20. 21. 22.
References 1. Meyers BF, Lynch JP, Battafarano RJ, et al. Lung transplantation is warranted for stable, ventilator-dependent recipients. Ann Thorac Surg 2000;70:1675– 8. 2. Demertzis S, Haverich A, Ziemer G, et al. Successful lung transplantation for posttraumatic adult respiratory distress syndrome after extracorporeal membrane oxygenation support. J Heart Lung Transplant 1992;11:1005–7. 3. Jurmann MJ, Schaefers HJ, Demertzis S, Haverich A, Wahlers T, Borst HG. Emergency lung transplantation after ex-
903
tracorporeal membrane oxygenation. ASAIO J 1993;39: M448 –52. Lick SD, Alpard SK, Montoya P, Deyo DJ, Witt SA, Zwischenberger JB. A prototype low-resistance total artificial lung in series with the pulmonary circulation. ASAIO J Abstracts 2000;46:181. McDonald DA. The relation of pulsatile pressure to flow in arteries. J Physiol (Lond) 1955;127:533–52. Boschetti F, Perlman CE, Cook KE, Mockros LF. Hemodynamic effects of attachment modes and device design of a thoracic artificial lung. ASAIO J 2000;46:42– 8. Lick SD, Zwischenberger JB, Alpard SK, Witt SA, Deyo DJ, Merz SI. Development of an ambulatory artificial lung in an ovine survival model. ASAIO J 2001; in press. Witt SA, Alpard SK, Lick SD, Deyo DJ, Jayroe JB, Zwischenberger JB. Total artificial lung perioperative management: a 7-day survival study in sheep. Crit Care Med 1999;27(Suppl): A22. Harper DD, Alpard SK, Deyo DJ, Lick SD, Traber DL, Zwischenberger JB. Anatomic study of the pulmonary artery as a conduit for an artificial lung. ASAIO J Abstracts 2000; 46:184. Smith J, Kampine J, eds. Circulatory physiology. Baltimore: Williams & Wilkins, 1984. Takewa Y, Tatsumi E, Taenaka Y, et al. Hemodynamic and humoral conditions in stepwise reduction of pulmonary blood flow during venoarterial bypass in awake goats. ASAIO J 1997;43:M494 –9. Adamson RM, Dembitsky WP, Daily PO, Moreno-Cabral R, Copeland J, Smith R. Immediate cardiac allograft failure. ECMO versus total artificial heart support. ASAIO J 1996;42: 314– 6. Lynch WR, Haft JW, Montoya JP, et al. Partial respiratory support with an artificial lung perfused by the right ventricle: chronic studies in an active animal model. ASAIO J Abstracts 2000;46:202. Boschetti F, Cook KE, Perlman CE, Mockros LF. Does blood flow pulsatility affect oxygen transfer in artifical lungs? ASAIO J Abstracts 2000;46:194. Cook KE, Maximer JB, Hubbard JE, Mavroudis C, Mockros LF. Effect of shear stress on coagulation and inflammation in implantable artificial lungs. ASAIO J Abstracts 2000;46:194. Cook KE, Mockros LF. Biocompatibility of artificial lungs. In: Vaslef SN, Anderson RW, eds. The artificial lung. Austin, TX: Landes Bioscience Publishers, Tissue Engineering Series. 2001, in press. Lazar EI, Weinstein S, Stark CJ. Lung injury does not increase vascular resistance if pulmonary blood is fully saturated. Surg Forum 1993;44:649–51. Alpard SK, Zwischenberger JB, Tao W, Deyo DJ, Traber DL, Bidani A. New clinically relevant sheep model of severe respiratory failure secondary to combined smoke inhalation/ cutaneous flame burn injury. Crit Care Med 2000;28:1469–76. McDonald DA. The relation of pulsatile pressure to flow in arteries. J Physiol (London) 1955;127:533–52. Bergel DH, Milnor WR. Pulmonary vascular impedance in the dog. Circ Res 1965;16:401–15. Nichols WW, O’Rourke MF. McDonald’s blood flow in arteries: theoretical, experimental and clinical principles. London: Edward Arnold, 1998. Montoya JP, Merz SI, Bartlett RH. Significant safety advantages gained with an improved pressure-regulated blood pump. J Extracorporeal Technology 1996;28:71– 8.
Appendix The use of impedance to characterize the vascular circulation was first proposed by McDonald [19]. He argued that because of the dynamic, pulsatile nature of flow and pressure in the cardiovascular system, static resistance did not fully characterize the load that the heart must
904
LICK ET AL COMPLIANT ARTIFICIAL LUNG
pump against, as it ignored the capacitance and inertance in the circulation. The capacitance and inertance can be accounted for by incorporating the relationship between the time-varying components of the pressure and flow profiles. The time-varying components are quantified by converting the pressure and flow profiles from the time domain to the frequency domain with a Fourier series expansion of the respective profiles, as follows:
冘 N
P共t兲 ⫽ P ⫹
Pn sin 共nt ⫹ n兲
n⫽1
冘 N
˙ ⫹ ˙ 共t兲 ⫽ Q Q
˙ n sin 共nt ⫹ n兲 Q
n⫽1
˙ (t) are the time-domain pressure and Where, P(t) and Q ˙ are the mean pressure flow profiles, respectively; P and Q and flow magnitudes, respectively; n is an integer corresponding to the harmonic number; N is the total number ˙ n are the pressure and of harmonics in the series; Pn and Q flow magnitudes, respectively, of the nth harmonic; n is the frequency at the nth harmonic; t is time; and n and n are phase angles of the nth harmonic. The impedance (Zn) is then defined as the ratio of the pressure magnitude to the flow magnitude at each harmonic component, as follows:
Zn ⫽
Pn Qn
Ann Thorac Surg 2001;72:899 –904
The calculation of impedance is based on the assumption that the system under study is linear, and this has been shown to be true in the pulmonary circulation, as well as the systemic circulation [20, 21]. To measure the input impedance of the paracorporeal AL and modified prototype AL with compliant chamber, a mock circuit was constructed at MC3, Inc. to provide the flow and pressure profile that is normally generated by the right ventricle. The circuit consisted of a pulsatile pump [22], two rigid reservoirs, and a centrifugal pump. The test fluid was water at room temperature. The pulsatile pump perfused the test device (ie, prototype AL or modified prototype AL with the compliant chamber). The two reservoirs were used to vary the mean and diastolic pressures, and the centrifugal pump maintained steady water levels between the two reservoirs. The pressure wave at the inlet to the device, which was equivalent to the pulmonary artery pressure profile, was measured with a dynamic pressure transducer (Omega Engineering Inc, Stamford, CT), and the flow wave was measured with an ultrasound transducer (Transonic Systems Inc, Ithaca, NY). The pressure and flow signals were acquired digitally with a data acquisition system (National Instruments, Austin, TX). The digitally acquired signals were processed using a fast Fourier transform algorithm (MathWorks, Natick, MA) to derive the corresponding frequency components of the signals, and the impedance was calculated at each frequency as shown above.
