| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL ARTICLE |
Clinic of Cardiac Surgery, University of Schleswig-Holstein, Luebeck, Germany
Prof. Dr. med. Norbert W Guldner, MD Tel: +49 451 500 2887 Fax: +49 451 500 6035 Email: guldner{at}uni-luebeck.de, Ratzeburger Allee 160, D- 233538 Luebeck, Germany.
ABSTRACT
Experimental valveless muscular blood pumps (biomechanical hearts) in goats can pump more than 1 L·min–1, but due to a high pendulum volume, no significant flow contribution to the circulation is gained. Thus valved and valveless biomechanical hearts were compared for efficacy. Heart failure was induced in 5 adult Bore goats by repeated intracoronary embolization. A valved and balloon-equipped pumping chamber was integrated into the descending aorta, simulating standard biomechanical circulatory support. The valveless biomechanical heart supported a failing heart with a baseline cardiac output of 2,670 ± 710 mL·min–1 by contributing additional flow of 113 ± 37 mL·min–1. The biomechanical heart model incorporating an outlet valve offered an additional 304 ± 126 mL·min–1, and the use of 2 valves significantly enhanced pulmonary blood flow by 1,235 ± 526 mL·min–1. The use of 2 valves in biomechanical hearts seems to be essential to achieve adequate circulatory support. Double-valved biomechanical hearts driven by an appropriate skeletal muscle ventricle may contribute to the therapy of heart failure.
Key Words: Heart-Assist Devices • Heart Failure • Heart Valves • Hemodynamics • Skeletal Muscle Ventricle
INTRODUCTION
More than 95% of patients with endstage heart failure have no definitive treatment options. This is due to the severe shortage of donor hearts and the technical and economic limitations of cardiac assist devices and artificial hearts.1 An additional cardiac output of 1–2 L·min–1 would give most patients with endstage heart failure a better quality of life and longer survival. The latissimus dorsi muscle has several advantages as a muscular blood pump: it is nearly always available, there is no concern about foreign tissue rejection or need for immunosuppression, there is less risk of infection, and this procedure might be less costly than heart transplantation or treatment with a fully implantable cardiac assist device or artificial heart.2 Dynamic cardiomyoplasty has been applied clinically in >2,000 patients.3,4 It has not been convincing in many cases, possibly due to 3 reasons: the electrically transformed type I fibre muscle tissue might be too weak to squeeze the enlarged heart during systole; this transformed muscle might contract too slowly following the more rapid heart systole; and the now transformed and thus oxygen-demanding muscle might be over-stimulated, resulting in severe muscle damage.
Skeletal muscle ventricles (SMV) with markedly smaller diameters than those used in dynamic cardiomyoplasty, and a longer time of contraction during diastole, might overcome these limitations.5,6 Furthermore, partially transformed muscles might be more appropriate for support of the failing heart.7,8 For more than 20 years, SMV have been applied in dogs after modelling to the ventricular shape, allowing a vascular delay, and electrical preconditioning of >6–8 weeks to overcome muscle fatigue.7 They are integrated into the circulation in a second operation, and a small SMV can pump for >4 years without relevant muscle damage.6 However, a 2-stage procedure is clinically unacceptable because of the high operative risk in such severely ill patients, and the additional risk of bleeding. The results with small SMV in dogs may not be extrapolated to patients, because for a ventricular volume of 100 mL, much greater power is needed to overcome intracavitary pressure. In large animal models, clinically relevant stroke volumes of 30–60 mL were evaluated over 6–24 months.9–13 Muscle performance was assessed with various stimulation patterns, pressure loads, and drugs.9,11 These studies resulted in the biomechanical heart, providing a pumping chamber with appropriate stroke volumes.12 This can be integrated into the circulation in a single operative procedure, with pharmacological stimulation by the β2 agonist, clenbuterol. The pumping chamber stabilizes the ventricular pump cavity and improves flow characteristics to minimize thromboembolic complications, as well as avoiding muscle damage by over-stretch-induced ischemia and preventing ventricular chamber rupture. Biomechanical hearts of a clinically relevant size, but without valves, have been tested previously.12 This study was undertaken to compare biomechanical hearts without valves and those with 1 or 2 valves, using a standardized stroke volume.
MATERIALS AND METHODS
Experiments were carried out in 9 adult Boer goats aged 3.3 ± 1.2 years, weighing 79 ± 6.6 kg. The study was authorized and supervised by a representative of the Society for the Prevention of Cruelty to Animals of the District Administration (V 742-722-41.122-6 [138-5/03]). General anesthesia was induced with xylazine hydrochloride 100 mg (Rompun 2%; BayerVital GmbH, Germany) and atropine sulphate 1 mg (Atropinsulfat; Fresenius, Germany) before orotracheal intubation, and maintained with 10–30 mL·h–1 of propofol 2% (Disoprivan 2%; GlaxoWellcome, Germany). Blood pressure was monitored via an ear artery. A left dorsolateral thoracotomy was performed, the descending aorta was divided, and each end was connected with a ring-armored PTFE prostheses of 16 mm in diameter (Impra Medica GmbH, Munich, Germany). Both prostheses led into a 16-mm silicone tube, splitting in a Y-configuration to one valved tube and one valveless tube. Both tubes reunited before entering the pumping chamber (Figure 1
). By clamping appropriate tubes, configurations without, and with 1 or 2 valves were used to evaluate pump efficacy. A test setting with one outlet valve is illustrated in Figure 1. Muscular pump function was simulated by an intraventricular balloon with a standard volume of 50 mL, driven by a clinically used intraaortic balloon pump (Datascope, Fairfield, NJ, USA). Synchronization of the balloon pump with diastole in 1 : 2 mode was based on the arterial pressure curve at 20, 150, and 300 ms after the pressure dip caused by aortic valve closure. The most effective hemodynamic setting was used for analysis.
|
Blood flow was measured using ultrasound cuffs (Transonic Systems Inc, Ithaca, NY, USA) around the pulmonary artery and the ascending and descending aorta (Figure 1
). A conductance catheter (Leycom; Cardio Dynamics, Zoetermeer, Netherlands) was inserted through the left carotid artery and led into the LV.15 Catheters were inserted into the left coronary artery for embolization and measuring intracoronary flow velocity. A 0.014-inch Doppler-tipped guidewire (FloWire; Cardio-Metrics EndoSonics Corp, CA, USA) was used for intracoronary flow velocity measurements.
Data are presented as mean ± standard deviation. A Wilcoxon test was performed for comparison between groups of paired samples (n = 5). A 95% confidence limit was chosen as the criterion of statistical significance (p 
0.05). All statistical calculations were performed with Winstat for Exel 2005.1 (R. Fitch Software Germany).
RESULTS
Baseline values were measured under steady-state conditions with cardiac output of 2,670 ± 710 mL·min–1, mean aortic pressure of 52 ± 7 mm Hg, LVEDP of 23 ± 6 mm Hg, coronary flow velocity of 295 ± 31 mm·s–1, and dP/dt of 1,685 ± 213 mm Hg/s. Our acute heart failure model had a high complication rate with ventricular fibrillation occurring after 2–4 intracoronary embolizations at intervals of 10 min, thus data of only 5 goats could be evaluated. The 5 valveless models induced additional pulmonary blood flow of 113 ± 37 mL·min–1. After 5 min of flow equilibration, mean aortic pressure increased by 2 ± 1.5 mm Hg, and coronary flow velocity by 25 ± 6 mm·s–1. The simultaneously evaluated LVEDP decreased from 23 to 21 mm Hg, while dP/dt rose by 47 ± 10 mm Hg/s. With a single distal outlet valve in the 5 models, additional pulmonary blood flow was 304 ± 126 mL·min–1. Mean aortic pressure increased by 20 ± 7 mm Hg, and intracoronary flow velocity by 29 ± 9 mm·s–1, while LVEDP decreased by 4 ± 2 mm Hg, and LV contractility rose by 92 ± 32 mm Hg/s. When 2 valves were deployed in the 5 models, additional pulmonary blood flow was 1,235 ± 526 mL·min–1 (p < 0.05) with cardiac output of 2,670 ± 710 mL·min–1. Corresponding mean aortic pressure increased by 19 ± 9 mm Hg (p < 0.05) and intracoronary flow velocity by 59 ± 18 mm·s–1 (p < 0.05). LVEDP decreased from 23 ± 6 to 9 ± 5 mm Hg (p < 0.05) and contractility increased from 1,685 ± 213 to 2,155 ± 239 mm Hg/s (
dP/dt 470 ± 192 mm Hg/s; p < 0.05). Thus the configuration with 2 valves caused the highest additional flow, mean aortic pressure, coronary flow velocity, and LV contractility, with markedly increased stroke work as demonstrated by the enlarged area surrounded by the pressure-volume loop (Figures 2, 3, and 4).
|
|
|
These results demonstrate impressive circulatory benefits from the use of 2 valves in a model of biomechanical support for the failing heart. An optimally working SMV with a stroke volume of 50 mL was mimicked by the balloon pump. Evaluations of chronically working skeletal muscles were not the subject of this study. Heart failure models have been investigated by several groups.16 Acute models should be distinguished from chronic settings.17,18 Our acute model had a high rate of ventricular fibrillation although we pretreated the myocardium with a high dose of beta blockers to minimize rhythm disorders. Internal defibrillation with electric shocks was mostly unsuccessful. Four of the 9 goats were lost to ventricular fibrillation with circulation breakdown.
Cardiac output was measured with an ultrasonic flow probe around the pulmonary artery; a flow probe around the ascending aorta would not have included coronary flow. The mean measuring duration of 5 min for each valve setting was regarded as valid. Mean aortic pressure was registered after at least 10 min of pumping. Immediately at the start of pumping, there was a pressure dip in the proximal aorta of roughly 1 min. This pressure decrease was 5–15 mm Hg, and it was not included in our evaluations. This decrease must be analyzed separately because there is a direct correlation between mean aortic pressure and flow within the coronary arteries. Intracoronary flow velocity measured in our test setting corresponds to the coronary flow necessary for oxygen supply to the myocardium. Hemodynamic parameters (LVEDP, dP/dt) were measured by the pressure sensor of the conductance catheter. We aimed for LVEDP >20 mm Hg during intracoronary embolization. We could also demonstrate unloading of the failing LV by the pumping model. The pressure-volume loops indicated decreased LVEDP as well as stroke work. These data of double-valved biomechanical hearts may contribute to the development of therapy for acute heart failure by mechanical assist devices.
A hemodynamically effective single-valved device was applied clinically in 1966 within the ascending aorta (using the native aortic valve).19 The cause of death 20 h after completing the operation was not fully understood. Placing this pump in the ascending aorta risks cerebral thromboembolic complications, and circulatory regulative disorders are not excluded. In our double-valved model in the descending aorta, a retrograde thromboembolism to the brain from the double-valved pumping chamber seems very unlikely. To improve valveless biomechanical hearts developed in chronic animal studies, knowledge of the hemodynamic efficacy of none, 1, and 2 valves is required.12 We showed that the use of 2 valves and the aortic-aortic topographical setting were appropriate for efficient blood pumping. Blood flow could be increased by more than 1 L·min–1 in goats with a latissimus dorsi muscle mass of approximately 300 g (human latissimus dorsi weights 600 g). The decreased mean aortic pressure at the start of pumping should be investigated further to avoid any complications of coronary perfusion.
Long-term pumping with a reduced stimulation frequency 24 h per day has proved feasible with SMV in dogs and goats.6,16,20,21 Nevertheless, controlled stimulation of the muscle seems to be mandatory. This has been shown in basic experimental work as well as in clinical application, using a new concept in dynamic cardiomyoplasty with less stimulation energy.7,8,20 Muscular blood pumps with controlled stimulation might be practical only if the blood-contacting surface prevents thromboembolic complications. This surface might be achieved by a titanium coating, as described elsewhere.21 A titanium coating might be followed by autologous tissue engineering with progenitor cells of the circulating blood. This biotechnology has already been demonstrated in goats.22 These findings may make muscular blood pumps clinically more feasible.
Presented at the 4th International Cardiac Bio-Assist Association Congress, Singapore, March 12–13, 2008.
REFERENCES
Asian Cardiovasc Thorac Ann 2009;
17:592-597
© 2009 by SAGE Publications
DOI: 10.1177/0218492309349066
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
