During the past 2 decades, mechanical circulatory support has evolved into a standard therapy for patients with end-stage heart failure1–3 as a bridge to cardiac transplantation,4 a destination therapy,4 or a bridge to myocardial recovery.5,6 In 2002, the milestone Randomized Evaluation of Mechanical Assistance for the Treatment of Congestive Heart Failure (REMATCH) trial demonstrated clinical success with pulsatile mechanical circulatory support as a long-term therapy for patients with end-stage heart failure.2 Subsequently, the US Food and Drug Administration approved use of the pulsatile HeartMate XVE as a destination therapy in patients ineligible for cardiac transplantation.7
In more recent years, pulsatile left ventricular assist devices (LVADs) that mimic native cardiac function with alternating systoles and diastoles have been replaced by devices which continuously (and nonphysiologically) unload blood volume from within the failing left ventricle into the aorta. Indeed, in 2008, the Food and Drug Administration approved the use of the HeartMate II continuous flow LVAD in patients for bridge-to-transplant.8 Such next-generation devices are smaller, more reliable, more energy efficient, less thrombogenic, and less surgically traumatic to implant, but continuous flow devices do not generate normal pulsatile hemodynamics.9 As a result, a new population of “pulseless” patients undergoing long-term LVAD support has emerged. In this patient population, concerns have been raised about the effects of chronic, nonpulsatile blood flow on cardiovascular architecture, end-organ perfusion, and long-term patient outcomes.10
Previously, with in silico computer modeling,11in vitro mock circulatory systems,12 and in vivo in cows,13 pigs,13,14 and human patients,13,15 we have demonstrated marked differences in vascular input impedance, pulsatility, ventricular pressure-volume relationships, and coronary blood flow during continuous versus pulsatile blood flow. To further characterize the influence of pulsatility on ventricular unloading, this preliminary study is a first step in understanding the chronic consequences of diminished pulsatility. The major goal of this feasibility study was to examine acute hemodynamic responses to continuous versus pulsatile mechanical circulatory support in calves with induced cardiac dysfunction. Standard indices of cardiovascular performance were measured during each support mode. The results of this study will assist in guiding future investigation into the effects of chronic continuous and pulsatile mechanical circulatory support.
Materials and Methods
All animals received humane care and were handled in accordance with National Institutes of Health and University of Louisville Animal Care Committee guidelines. Experimental procedures followed animal study protocols approved by the University of Louisville Institutional Animal Care and Usage Committee.
Male mixed-breed calves (n = 14, 80–110 kg) were used. As described previously, chronic heart failure was induced with oral Monensin (n = 5),16 serial intravenous doxorubicin (n = 2),17 or serial selective left main coronary artery microembolization with 90-μm polystyrene microspheres (n = 2).18 In additional animals, acute heart failure was induced with intravenous esmolol, phenylephrine, and hypervolemia (n = 5).19 An appropriate moderate degree of heart failure was defined by a 20% reduction in cardiac output (CO) and/or an ejection fraction <40%.
After the induction of heart failure, a nonsurvival hemodynamic study with the implantation of a continuous flow LVAD (Thoratec HeartMate II, n = 6; Heartware HVAD n = 2) and/or pulsatile flow LVAD (Thoratec PVAD n = 6, IVAD n = 2) was performed. Animals were preanesthetized with atropine (30 mg), anesthetized with isoflurane (3%–5%) and room air, and anticoagulated with heparin (200–250 units/kg). Fluid-filled arterial and venous catheters were placed in the right carotid artery and jugular vein for blood sampling. A left thoracotomy was performed. Ribs 4 and 5 were resected. A single-tip, high-fidelity micromanometer catheter (Millar Instruments, Houston, TX) was placed in the left atrium, and a dual pressure-volume conductance catheter (Millar Instruments, Houston, TX) was advanced from the aorta into the left ventricle for simultaneous measurement of left ventricular and aortic blood pressures. Transit-time ultrasonic flow probes (Transonics, Ithaca, NY) were placed around the pulmonary artery and LVAD outflow graft to measure CO and LVAD output, respectively.
The LVAD was implanted. The left ventricular apex was cored and cannulated. The device and outflow graft were deaired and anastomosed to the descending aorta.
In each animal, blood pressure and flow waveforms were recorded during heart failure baseline, maximum support mode (LVAD operated at maximum support, aortic valve closed during continuous support, or opening intermittently/incompletely during pulsatile support), and during moderate support mode (LVAD flow [LVADF] titrated to ∼50% of maximum support mode). Heart failure baseline and support modes for each device were maintained for 10 minutes each to achieve steady-state conditions before collection of 30-second data sets.
After completion of the experimental protocol, animals were euthanized by increasing the anesthetic depth and a bolus intravenous injection of supersaturated KCl.
Instrumentation, Data Reduction, and Statistics
All transducers were pre and postcalibrated against known physical standards to ensure measurement accuracy. Calibration curves for the volume conductance catheter were constructed using static and dynamic tests pre- and postexperiment. Data were collected at 400 Hz, signal conditioned, and A/D converted for digital analysis using our GLP compliant data acquisition system.20
To determine hemodynamic status during each support mode, pressure and flow waveforms were used to derive pulse pressure (PP), surplus hemodynamic energy (SHE), energy equivalent pressure (EEP), and (EEP/mean aortic pressure [MAP] − 1) × 100% as indices of stored pulsatile energy.10 Mean LVADF, heart rate (HR), CO, stroke volume (SV), left ventricular end-diastolic volume (LVEDV), left ventricular end-systolic volume (LVESV), mean left atrial pressure, left ventricular end-diastolic pressure (LVEDP), left ventricular peak systolic pressure (LVPSP), aortic systolic blood pressure, aortic diastolic blood pressure, MAP, and ±dP/dt were derived as standard indices of cardiovascular performance. Rate-pressure product, a standard index of cardiac metabolic demand,21 was calculated as the product of HR and LVPSP. Hemodynamic parameters were calculated on a beat-to-beat basis for each 30-second data set with the Hemodynamic Evaluation and Assessment Research Tool program22 developed in Matlab (Version 6.5, MathWorks, Natick, MA). All analyzed beats in each data set (∼30–50 beats/30-second data set) were averaged to obtain a single representative mean value for each calculated parameter. Pressure-volume loops were constructed by plotting left ventricular pressure against left ventricular volume. Each loop represented one complete cardiac cycle.
One-way repeated measures analysis of variance with Tukey posttest was performed to compare baseline, moderate, and maximum support mode within either the continuous flow or pulsatile flow groups. Unpaired Student's t-tests were performed to compare across baselines and continuous versus pulsatile support modes (GraphPad Prism version 4.00, San Diego, CA). All analyses were two tailed, and a p < 0.05 (95% confidence interval) was considered statistically significant.
Figure 1 demonstrates morphological differences in the hemodynamic waveforms during each support mode. Continuous and pulsatile LVADs each reduced ventricular volumes by different mechanisms. As continuous unloading increased, the pressure-volume relationship collapsed. Although left ventricular volumes were reduced significantly, the range of left ventricular pressures decreased to values below aortic diastolic pressure. As a result, the aortic valve remained chronically closed. In comparison, as pulsatile unloading increased, the decrease in left ventricular volumes was comparable with continuous unloading, but the normal range of left ventricular pressures was preserved.
Figure 2 illustrates hemodynamic energy profiles during each support mode. During moderate and maximum continuous unloading, PP, SHE, and (EEP/MAP − 1) × 100% were significantly decreased compared with baseline and compared with pulsatile unloading. PP but not SHE or (EEP/MAP − 1) × 100% were significantly lower during maximum continuous support versus moderate continuous support. In comparison, during moderate and maximum pulsatile unloading, PP and SHE were unchanged. However, during maximum pulsatile unloading, (EEP/MAP − 1) × 100% was significantly higher than baseline.
Table 1 presents hemodynamic parameters during heart failure baseline, moderate and maximum continuous support, and moderate and maximum pulsatile support. Heart failure baselines for continuous and pulsatile flow groups were not statistically different. Moderate and maximum LVADFs were significantly different. As expected, continuous unloading of the failing left ventricle altered all hemodynamic parameters studied except HR and MAP, whereas pulsatile unloading preserved normal physiologic values.
Pulsatile and continuous unloading reduced LVEDVs to a similar degree. However, during continuous but not pulsatile support, LVESV and SV were significantly reduced. As such, continuous unloading diminished the variation between end-systolic and end-diastolic volumes, whereas pulsatile unloading preserved a wider range.
Importantly, as determined by rate-pressure product, maximum continuous unloading dramatically decreased cardiac metabolic demands to values that are not normally observed in calves, whereas pulsatile unloading decreased this parameter more modestly.
We have initiated large animal feasibility studies to characterize the differences between varying levels of continuous and pulsatile blood flow with an LVAD. As a preliminary step toward these goals, this study demonstrated that important differences exist between acute continuous versus pulsatile mechanical unloading of the failing left ventricle. Specifically, these data revealed three important findings: 1) continuous unloading diminished hemodynamic energy cycling that is typical of pulsatile hemodynamics; 2) although continuous and pulsatile LVADs similarly reduced LVEDVs, during increasing levels of continuous unloading, the variation between end-systolic and end-diastolic volumes and pressures decreased to nonphysiologically low values and resulted in chronic closure of the aortic valve. In comparison, pulsatile unloading decreased left ventricular volumes but preserved a normal range of physiologic pressures; and 3) continuous unloading dramatically reduced cardiac metabolic demands to values that are not normally observed in mammals, whereas pulsatile unloading did so to a lesser degree. The acute bovine model does not provide information regarding the chronic effects of continuous versus pulsatile unloading of the failing left ventricle. As such, these results are intended to guide the development of further in vivo studies to examine myocardial and arterial remodeling during chronic mechanical circulatory support.
Our findings suggest mechanisms by which diminished pulsatility may induce histological changes within the cardiovascular system. When cells in the heart and vasculature are exposed to chronically deranged mechanical forces such as during hypertension, pathological remodeling may occur.23 For example, chronic hypertension is well documented to induce concentric myocardial hypertrophy, arterial medial thickening, and alterations of the collagen/elastin ratio. During isolated systolic hypertension, PPs are known to exceed 80–100 mm Hg and affect cardiac and arterial tissues. Even the acute blood pressure elevations that are observed in malignant hypertension may induce acute pathological vascular remodeling that include hyperplastic arteriolosclerosis and fibrinoid necrosis. During continuous unloading of the cardiovascular system with an LVAD, hemodynamics are similarly deranged but in the opposite direction of hypertension. Our data demonstrated that continuous volume unloading reduced the peak left ventricular systolic pressure from a baseline mean of 91 mm Hg to 56 mm Hg. Considering that the baseline diastolic arterial blood pressure was 62 mm Hg, in this situation, the aortic valve did not consistently open. In normal and heart failure patients, a LVPSP of 56 mm Hg is not compatible with life nor is a PP of 7 mm Hg as was produced by continuous unloading.
Similarly, continuous unloading decreased ±dP/dt and rate × pressure product to values that are not encountered in bovids. Moderate reductions in metabolic demands while maintaining a partial cardiac workload are likely important, especially if the goal of treatment is myocardial recovery. However, the optimal reduction in cardiac metabolism during LVAD support has not been established, and excessive myocardial unloading in which the heart performs too little workload may induce myocardial atrophy.24,25 Therefore, it is our hypothesis that narrow hemodynamic energy profiles and diminished myocardial and arterial stretch in continuous flow LVAD patients may induce architectural changes in cardiovascular tissues and may have long-term consequences.
Continuous versus Pulsatile Blood Flow: Clinical Implications for Long-Term Therapy
Chronic, nonpulsatile blood flow may alter the structure and physiological reactivity of the arterial vasculature. As the PP decreases, vascular resistance increases. Indeed, we have reported acute increases in vascular impedance clinically in response to continuous flow mechanical circulatory support.15 Physiologic studies in canines have demonstrated that systemic vascular resistance is inversely related to PP and that systemic vascular resistance during nonpulsatile circulation is as much as 34% greater than during pulsatile circulation.26 Histologic studies in goats undergoing prolonged, nonpulsatile cardiopulmonary bypass demonstrated atrophic changes in the tunica media of the aorta that included decreased total smooth muscle volume, reduced cell size, changes in the constituent volume ratio of collagen and elastin, and decreased number of myofilaments and chromatin in smooth muscle cells.27 Additional studies in the same model demonstrated decreased arterial vasoconstrictive response to norepinephrine28 and phenylephrine29 and suggested that nonpulsatile blood flow affects vascular reactivity. The net effect of nonpulsatile vascular remodeling and loss of vascular reactivity is stiff and unresponsive arteries. Such changes to arterial viscoelastic performance characteristics may account for novel pathologies observed in patients with prolonged nonpulsatile blood flow.
For example, recent reports have documented an increased incidence in ischemic and hemorrhagic stroke,30 acquired von Willebrand disease,31,32 diastolic hypertension,33 and gastrointestinal arteriovenous malformations and bleeding34 in patients who received chronic continuous flow mechanical circulatory support. These cardiovascular pathologies likely share a common etiology related to the absence of pulsatile stretch of arteries and/or the influence of altered shear stresses on the endothelium. Although superior clinical efficacy has been confirmed with next generation continuous flow devices versus first generation pulsatile devices,3 better clinical outcomes are necessary before prolonged LVAD support is more widely accepted. To this end, it may be worthwhile to introduce pulsatility with continuous flow devices, or better yet, to provide moderate ventricular assistance earlier in the progression of the disease and maintain native pulsatility. In support of this idea, our data demonstrated that moderate continuous unloading preserves more normal cardiac and arterial hemodynamics.
Our results also have important implications for device selection, operation, and weaning. Mounting evidence suggests that complete “volume-unloading” of the failing ventricle by an LVAD can promote reverse myocardial remodeling in select patients.6 In fact, myocardial recovery has most frequently been reported in patients who received pulsatile ventricular assistance.5,6 In these instances, functional recovery has been accompanied by favorable changes at the histological and molecular levels. However, a strategy of volume unloading by a continuous flow LVAD to promote myocardial recovery has not been fully characterized and may have limitations. For example, continuous unloading provides hemodynamic improvement by restoring CO and end-organ blood flow, but as the level of flow increases, variation in end-systolic and end-diastolic left ventricular volumes and pressures diminishes and eliminates the native workload of the heart. It is unclear how variations in SV affect myocardial remodeling. However, it is known that normal ventricular geometry depends on a consistent physiologic pressure-volume relationship. As we have demonstrated, pulsatile unloading normalizes this relationship by decreasing volume without affecting pressure. Furthermore, greater reductions of left ventricular volume are possible with a pulsatile device because suction events are less likely. These findings suggest a mechanism by which low native-heart volume output, loss of mechanical stretch, and less-aggressive volume unloading during continuous LVAD therapy may cause the heart to become a stiff and nonfunctional chamber. As a result, cardiac disuse atrophy24,25 may later prevent weaning of the device.
Of additional concern, this study demonstrated that continuous volume unloading of the heart decreased peak-systolic left ventricular pressures to the point where the aortic valve remained chronically closed. In this situation, fused valve leaflets, acquired aortic stenosis, or total occlusive thrombosis of prosthetic aortic valves has occurred.35 In these situations, maximum continuous unloading of the failing left ventricle may cause adverse consequences and limit myocardial recovery.
Four different heart failure models were used to create a state of diminished cardiac function. Although variability existed between the models, we did not observe statistically significant differences between measured hemodynamic parameters at heart failure baseline within the continuous versus pulsatile groups.
The hemodynamic effects of synchronous versus asynchronous pulsatile unloading of the failing left ventricle were not examined in this study. Although synchronous pulsatile unloading may be an effective strategy to unload the failing ventricle, it is not clinically used in ambulatory LVAD patients due to the difficulties associated with arrhythmia and chronic electrocardiographic acquisition in heart failure patients.
A meaningful analysis of the pressure-volume relationship was not possible. Pressure-volume analyses during LVAD support that include stroke work and end-systolic elastance are invalid and may be misleading.36 Although an LVAD unloads blood volume from within the heart, the heart continues to generate contractile force. Even if the aortic valve remains closed, systolic ejection of blood occurs through the LVAD and may account for the small PP that is observed during continuous flow mechanical circulatory support. As a result, a parameter such as stroke work does not differentiate between the quantity of work performed by the heart versus the LVAD.
Inherent error may exist during hemodynamic measurement. Flow probes and pressure-volume transducers may be inaccurate up to 10%.37
Continuous flow LVADs altered the physiologic profile of myocardial and vascular hemodynamic energy utilization, whereas pulsatile LVADs preserved normal physiologic values. As a result, the hemodynamic profile was deranged during continuous but not pulsatile unloading of the failing left ventricle. As continuous unloading increased, the pressure-volume relationship collapsed and the aortic valve remained closed. In contrast, as pulsatile unloading increased, the left ventricular pressure-volume relationship normalized by decreasing left ventricular volumes while maintaining normal left ventricular pressures.
Supported by American Heart Association Scientist Development Grant no. 0730319N and Whitaker Foundation Grant no. RG-01-0310.
The authors thank Tim Horrell, Mickey Ising, Cary Woolard, Laura, Karen Lott, Menaka Nadar, and the University of Louisville Health Sciences veterinary staff for their assistance.
1. Goldstein DJ, Oz MC, Rose EA: Implantable left ventricular assist devices. N Engl J Med
339: 1522–1533, 1998.
2. Rose EA, Gelijns AC, Moskowitz AJ, et al
: Long-term mechanical left ventricular assistance for end-stage heart failure. N Engl J Med
345: 1435–1443, 2001.
3. Slaughter MS, Rogers JG, Milano CA, et al
: Advanced heart failure treated with continuous-flow left ventricular assist device. N Engl J Med
361: 2241–2251, 2009.
4. Birks EJ, Yacoub MH, Banner NR, et al
: The role of bridge to transplantation: Should LVAD patients be transplanted? Curr Opin Cardiol
19: 148–153, 2004.
5. Frazier OH, Myers TJ: Left ventricular assist system as a bridge to myocardial recovery. Ann Thorac Surg
68: 734–741, 1999.
6. Birks EJ, Tansley PD, Hardy J, et al
: Left ventricular assist device and drug therapy for the reversal of heart failure. N Engl J Med
355: 1873–1884, 2006.
7. Reitz BA: Mechanical devices and US Food and Drug Administration (FDA) approval. Semin Thorac Cardiovasc Surg Pediatr Card Surg Annu
8. Thoratec: Thoratec Announces Filing of PMA Seeking Destination Therapy Approval for HeartMate II. [Press release]. 2009.
9. Takatani S: Progress of rotary blood pumps. Artif Organs
30: 317–321, 2006.
10. Ji B, Undar A: Comparison of perfusion modes on microcirculation during acute and chronic cardiac support: Is there a difference? Perfusion
22: 115–119, 2007.
11. Giridharan GA, Ewert DL, Pantalos GM, et al
: Left ventricular and myocardial perfusion responses to volume unloading and after load reduction in a computer simulation. ASAIO J
50: 512–518, 2004.
12. Koenig SC, Pantalos GM, Gillars KJ, et al
: Hemodynamic and pressure-volume responses to continuous and pulsatile ventricular assist in an adult mock circulation. ASAIO J
50: 15–24, 2004.
13. Koenig SC, Giridharan GA, Ewart DL, et al
: Human, bovine and porcine systematic vascular input impedances are not equivalent: Implications for device testing and xenotransplantation in heart failure. J Heart Lung Transplant
27: 1340–1347, 2008.
14. Pantalos GM, Giridharan G, Colyer J, et al
: Effect of continuous and pulsatile flow left ventricular assist on pulsatility in a pediatric animal model of left ventricular dysfunction: Pilot observations. ASAIO J
53: 385–391, 2007.
15. Travis AR, Giridharan GA, Pantalos GM, et al
: Vascular pulsatility in patients with a pulsatile- or continuous-flow ventricular assist device. J Thorac Cardiovasc Surg
133: 517–524, 2007.
16. Litwak KN, McMahan A, Lott KA, et al
: Monensin toxicosis in the domestic bovine calf: A large animal model of cardiac dysfunction. Contemp Top Lab Anim Sci
44: 45–49, 2005.
17. Tomlinson CW, McGrath GM, McNeill JH: Adriamycin cardiomyopathy: Pathological and membrane functional changes in a canine model with mild impairment of left ventricular function. Can J Cardiol
2: 368–374, 1986.
18. Sabbah HN, Stein PD, Kono T, et al
: A canine model of chronic heart failure produced by multiple sequential coronary microembolizations. Am J Physiol
260: H1379–H1384, 1991.
19. Bartoli CR, Wilson GC, Giridharan GA, et al
: A novel subcutaneous counterpulsation device: Acute hemodynamic responses to pharmacologically-induced hypertension, hypotension, and heart failure. Artif Organs
2010 (in press).
20. Koenig SC, Woolard C, Drew G, et al
: Integrated data acquisition system for medical device testing and physiology research in compliance with good laboratory practices. Biomed Instrum Technol
38: 229–240, 2004.
21. Rooke GA, Feigl EO: Work as a correlate of canine left ventricular oxygen consumption, and the problem of catecholamine oxygen wasting. Circ Res
50: 273–286, 1982.
22. Schroeder MJ, Perreault B, Ewert DL, Koenig SC: HEART: An automated beat-to-beat cardiovascular analysis package using Matlab. Comput Biol Med
34: 371–388, 2004.
23. Haga JH, Li YS, Chien S: Molecular basis of the effects of mechanical stretch on vascular smooth muscle cells. J Biomech
40: 947–960, 2007.
24. Kinoshita M, Takano H, Taenaka Y, et al
: Cardiac disuse atrophy during LVAD pumping. ASAIO Trans
34: 208–212, 1988.
25. Kinoshita M, Takano H, Takaichi S, et al
: Influence of prolonged ventricular assistance on myocardial histopathology in intact heart. Ann Thorac Surg
61: 640–645, 1996.
26. Angell James JE, Daly Mde B: Effects of graded pulsatile pressure on the reflex vasomotor responses elicited by changes of mean pressure in the perfused carotid sinus-aortic arch regions of the dog. J Physiol
214: 51–64, 1971.
27. Nishimura T, Tatsumi E, Takaichi S, et al
: Prolonged nonpulsatile left heart bypass with reduced systemic pulse pressure causes morphological changes in the aortic wall. Artif Organs
22: 405–410, 1998.
28. Nishimura T, Tatsumi E, Nishinaka T, et al
: Diminished vasoconstrictive function caused by long-term nonpulsatile left heart bypass. Artif Organs
23: 722–726, 1999.
29. Nishinaka T, Tatsumi E, Nishimura T, et al
: Change in vasoconstrictive function during prolonged nonpulsatile left heart bypass. Artif Organs
25: 371–375, 2001.
30. Lahpor J, Khaghani A, Hetzer R, et al
: European results with a continuous-flow ventricular assist device for advanced heart-failure patients. Eur J Cardiothorac Surg
37: 357–361, 2010.
31. Klovaite J, Gustafsson F, Mortensen SA, et al
: Severely impaired von Willebrand factor-dependent platelet aggregation in patients with a continuous-flow left ventricular assist device (HeartMate II). J Am Coll Cardiol
53: 2162–2167, 2009.
32. Geisen U, Heilmann C, Beyersdorf F, et al
: Non-surgical bleeding in patients with ventricular assist devices could be explained by acquired von Willebrand disease. Eur J Cardiothorac Surg
33: 679–684, 2008.
33. Haft J, Armstrong W, Dyke DB, et al
: Hemodynamic and exercise performance with pulsatile and continuous-flow left ventricular assist devices. Circulation
116(suppl 11): I8–I15, 2007.
34. Crow S, John R, Boyle A, et al
: Gastrointestinal bleeding rates in recipients of nonpulsatile and pulsatile left ventricular assist devices. J Thorac Cardiovasc Surg
137: 208–215, 2009.
35. Rose AG, Park SJ: Pathology in patients with ventricular assist devices: A study of 21 autopsies, 24 ventricular apical core biopsies and 24 explanted hearts. Cardiovasc Pathol
14: 19–23, 2005.
36. Vandenberghe S, Segers P, Steendijk P, et al
: Modeling ventricular function during cardiac assist: Does time-varying elastance work? ASAIO J
52: 4–8, 2006.
37. Slaughter MS, Bartoli CR, Sobieski MA, et al
: Intraoperative evaluation of the HeartMate II flow estimator. J Heart Lung Transpl
28: 39–43, 2009.