(Circulation. 2005;112:438-448.)
© 2005 American Heart Association, Inc.
New Drugs and Technologies |
From the Division of Cardiology, Columbia University Medical School, New York (D.M., D.B.), and IMPULSE Dynamics, Orangeburg (D.B.), NY.
Correspondence to Daniel Burkhoff, MD, PhD, Columbia University, Division of Cardiology, Black Building 812, 650 W 168th St, New York, NY 10032.
| Abstract |
|---|
|
|
|---|
Key Words: heart failure transplantation disease management
| Introduction |
|---|
|
|
|---|
Many of the major advances achieved over the past 2 decades came with the recognition that the progression of heart failure is due to neurohormonal activation and abnormal mechanical stresses on the myocardium (increased preload and afterload; Figure 1). This vicious cycle of progressive pump enlargement and dysfunction is collectively referred to as ventricular remodeling. The structural and functional aspects of remodeling can be illustrated on a ventricular pressure-volume diagram (Figure 2A). With sustained neurohormonal activation and its consequences, the pressure-volume relations gradually shift toward larger volumes so that ejection fraction is reduced with marked myocardial hypertrophy. The goal of many heart failure treatments is to prevent, slow, or reverse this process (Figure 2B). Randomized studies have shown that pharmacological blockade of the key neurohormonal pathways interrupts the vicious cycle, retards progression, and improves survival (Figure 1, blue). However, results of several recent trials suggest that attempts to block additional neurohormonal pathways may be detrimental.6 This finding has led to the concept that we may have reached the limit to which neurohormonal and cytokine mechanisms can be blocked in heart failure patients.7
|
|
Accordingly, there have been intensive efforts to develop and test device-based therapies for patients with both acute and chronic heart failure. For example, cardiac resynchronization therapy8 has been approved by the US Food and Drug Administration (FDA) for NYHA class III and IV chronic heart failure (CHF) patients with abnormal electric conduction, and recent clinical trials support broader use of implantable cardiac defibrillators in patients with ventricular dysfunction.9,10 Such devices have significantly affected the treatment of CHF patients. Other experimental forms of therapy such as myogenesis (eg, stem cell, myoblasts)11 and other electrical therapies such as cardiac contractility modulating nonexcitatory electrical impulses12 and less invasive defibrillators are under active investigation. These classes of devices are the topic of a recent review in this series. The focus of the present review is on mechanical devices to treat various stages of heart failure that are currently available or are under development. Not included for the sake of brevity, but still noteworthy, are classes of devices under development to aid in comprehensive disease management programs and chronic hemodynamic monitoring systems aimed at prevention and early detection of heart failure exacerbations to reduce hospitalizations and increase longevity.
| Surgical Approaches to Treatment of CHF |
|---|
|
|
|---|
In ischemic cardiomyopathy, myocardium affected by infarctions can heal in different ways, generating either an akinetic scar or a dyskinetic scar (ie, aneurysm). Ventricular reduction surgery in this setting therefore removes noncontractile material. The effects of these ventricular reduction surgeries on ventricular pressure-volume relations has not been studied, and clinical studies of these procedures are generally confined to reports of nonrandomized registries. Removal of dyskinetic scar (aneurysmorophy) is clinically accepted and reported to be associated with satisfactory outcomes.16 The effects of removing akinetic scar (often referred to as the Dor procedure17 or surgical anterior ventricular restoration [SAVR]18) are less clear. The same theoretical modeling that correctly predicted detrimental effects of the Batista operation predicts that aneurysmorophy significantly improves pump function (which matches most of the data in the literature) and predicts that pump function will not be significantly improved by SAVR (Figure 3).19 The SAVR operation is now being studied prospectively as part of the NIH-sponsored Surgical Treatment of Ischemic Heart Failure (STICH) trial (http://www.stichtrial.org/disclaimer/index.cfm). Without firm scientific evidence that proves the benefits of such surgery, many physicians are reluctant to refer patients.
|
One limitation of these surgeries that could potentially confound the study is the lack of a uniform method of performing the procedure. Thus, the amount of tissue removed and the ultimate size and geometry of the reconstructed chamber are more or less at the discretion of the surgeon, with no firm guidelines. Two devices under development provide a model (CHF Technologies and Chase Medical) that is inserted into the chamber to provide a guide for both resizing and reshaping the left ventricle. These devices are being studied independently outside the STICH trial.
Another method proposed to decrease wall stress and to induce reverse remodeling is by passive ventricular restraint devices (ACORN Cardiovascular Inc; Paracor Medical, Inc; Myocor Inc). This concept evolved from an earlier investigational approach called cardiomyoplasty20 that tested the concept that the latisimus dorsi muscle could be freed, wrapped around the heart, and stimulated to contract in synchrony with the native heart, providing a boost to ventricular performance. However, it was discovered that after a relatively short time, continual stimulation of the skeletal muscle caused fatigue so that the graft no longer contracted with each stimulation.21,22 Rather, the skeletal muscle ultimately became a passive wrap around the heart. Although the overall clinical trial failed to show benefit, some patients appeared to benefit, and there was a reduction in heart size with some improvement in function. This led to the concept that the same effect could be readily achieved with a passive material appropriately tailored and fit snug to the heart.23 Studies in animals in which CHF was induced by repeated coronary artery microembolization have indeed shown reverse remodeling, including a leftward shift back toward normal of the end-systolic pressure-volume relationship (Figure 4), improved pump function and reversal of the fetal gene program associated with heart failure, which is interpreted as being a beneficial consequence of reduced wall stress.23 Importantly, despite the fact that the heart is wrapped to slightly decrease volume acutely, there is no induction of a restrictive physiology. A recently completed multicenter, randomized study of this technology with
300 patients has suggested that this approach improves NYHA functional class and reduces the requirement for subsequent major cardiac procedures such as cardiac transplantation and ventricular assist device insertion. Another device intended to perform similarly is constructed of nitinol, which may more readily permit placement of the device via port access.
|
Still another form of ventricular reshaping involves use of transventricular struts to bring opposite sides of the dilated left ventricle toward each other to reshape the ventricular circumference from 1 large circle to 2 circles, each with a smaller radius of curvature (in a kind of figure 8 configuration; Myocor, Inc).24 This device is currently under investigation for its potential use in treating heart failure and as a means of less invasively reducing the severity of mitral regurgitation.
For all these forms of surgical heart failure treatments, mitral valve repair or replacement is an important yet potentially confounding factor during clinical trials. Although mitral repair is typically effective at eliminating or reducing the regurgitant volume, it remains controversial as to whether and when to perform such surgery as a sole treatment in patients with low ejection fraction.25 Accordingly, numerous catheter-based devices are being developed to perform mitral valve repair percutaneously to reduce risk and allow it to be performed as a sole therapy to treat symptoms of heart failure.26,27 It is proposed that availability of such devices will allow treatment of more patients, including those at high risk, with reduced procedural morbidity and mortality.
Devices to Treat Acutely Decompensated Heart Failure
As patients deteriorate into more severe decompensation and cardiac output and blood pressure decline, patients become increasingly resistant to high doses of diuretics, and fluid overload ensues. Reduced renal perfusion resulting from relative hypotension and renal artery constriction are contributing factors. Because diuretics reduce glomerular filtration rate, their use, although necessary to treat symptoms, may contribute to further diuretic refractoriness. Accordingly, devices aimed at improving renal perfusion, directly dealing with fluid overload, and/or attempting to improve systemic perfusion have been developed.
One such strategy is local renal artery vasodilator drug delivery.28 Local renal vasodilator therapy can potentially improve renal blood flow and glomerular filtration rate while acting synergistically with diuretics to effect a diuresis with fewer effects on arterial blood pressure than the same drugs delivered systemically. Devices are currently under development for such local drug delivery that would allow testing of this strategy (Flowmedica, Inc).
A direct means of dealing with fluid overload in patients who have inadequate response to diuretic therapy is aquapheresis with a specialized system that provides efficient salt and water removal by a new form of ultrafiltration (Aquadex System 100, CHF Solutions).29 This system uses peripheral or central venous access and requires a low volume of extracorporeal blood to prime the system (33 cm3) compared with standard hemofiltration devices. This system removes isotonic salt and water at rates between 100 and 500 cm3/h without clinically significant changes in serum electrolytes, heart rate, or blood pressure. Because of the relative simplicity of operation, this system can be administered in inpatient settings and outpatient care centers. Four clinical trials of this system have indicated the ease, safety, and efficacy of this strategy to remove fluid. The UNLOAD trial, a prospective, randomized, multicenter, 200-patient trial, is currently underway and is testing the short-term efficacy with respect to fluid and symptoms and 30- to 90-day clinical impact of aquapheresis compared with usual diuretic strategies.
Another novel investigational approach for treating patients with acute heart failure exacerbations who are dependent on or resistant to intravenous inotropic agents involves continuous aortic flow augmentation.30 The Cancion system (Orqis Medical) consists of an extracorporeal, magnetically levitated centrifugal pump that withdraws blood from the femoral artery and returns it to the descending thoracic aorta in a continuous, nonpulsatile manner at rates between 1.1 and 1.5 L/min. Studies in experimental animals with CHF and results of preliminary clinical studies in patients with heart failure exacerbations demonstrate that this continuous flow loop results in increased cardiac output, improved renal function (decreased creatinine) and diuresis, and reduced pulmonary capillary wedge pressure. The suggested mechanisms of action include direct ventricular afterload reduction, stimulation of nitric oxide production (resulting from increased aortic shear stress) that leads to vasodilation, and increased renal perfusion. Randomized, controlled studies are currently underway.
Catheter-Based Pumps for Acute Cardiogenic Shock
As a result of either the progression to end-stage heart failure in patients with longstanding cardiomyopathy or severe ventricular dysfunction in the setting of acute myocardial infarction, cardiogenic shock (inadequate blood pressure and cardiac output to meet the metabolic needs of the body) occurs in many patients surviving to hospital admission. Medical treatment with intravenous inotropes, followed by use of the intra-aortic balloon pumping (IABP), is considered the first line of treatment for such patients.31 However, these strategies are frequently not effective. Several devices are under development for treating such patients.
Approved for short-term (6 hours) hemodynamic support is the TandemHeart percutaneous ventricular assist device (CardiacAssist). This device consists of an extracorporeal centrifugal (continuous flow) pump that withdraws blood from the left atrium via a cannula introduced transseptally via the femoral vein. Blood is then pumped (at up to
3.5 L/min) by a centrifugal pump through cannulas in 1 or both femoral arteries.32 Preliminary results showed that this device can improve hemodynamics in patients with cardiogenic shock. Studies also showed a reduction in serum lactate levels, an indication of improved tissue oxygenation and thus reversal of the cardiogenic shock state. In a more recent study, patients with cardiogenic shock resulting from acute myocardial infarction failing medical treatment were randomized to treatment with either an IABP or the TandemHeart assist device.33 The IABP had almost no effect on hemodynamics, but the TandemHeart device significantly increased cardiac output and blood pressure while significantly decreasing pulmonary capillary wedge pressure. However, mortality was comparable in both groups.
Other devices are being developed to provide acute hemodynamic support. They include 2 axial flow pumps (Impella CardioSystem AG,34 A-Med System, based on a prior system called the Hemopump35). These systems have helical propellers at the tip of catheters that are positioned retrograde across the aortic valve. Motors rotate the propellers, which draw blood from the left ventricle through the distal end of the catheter and discharge it into the aorta. These catheters can be inserted percutaneously via a femoral artery or surgically into the proximal aorta. Like the TandemHeart, these devices unload the left ventricle, thus reducing oxygen demand while improving systemic perfusion (Figure 5). Studies are underway to assess the safety and efficacy of these devices in a variety of settings in which IABPs are now used, including cardiogenic shock.
|
Devices for Chronic Decompensated CHF
Because of the limited availability of donor organs and the urgency for cardiac support in the setting of severe hemodynamic decompensation, ventricular assist devices capable of completely supporting the circulation are taking on an increasingly important role in heart failure therapy.
Several FDA-approved devices provide short-term circulatory support for postcardiotomy failure or bridge to transplantation. Those who might benefit from device placement for acute cardiac failure include patients unable to be weaned from cardiopulmonary bypass, those experiencing acute myocardial infarction with cardiogenic shock, patients with myocarditis, and patients who deteriorate while awaiting transplantation. In the last group, assist devices can be especially helpful before transplantation by providing a period for recovery of heart failurerelated end-organ impairment, thus reducing the risk of transplantation.36,37
Following the results of the Randomized Evaluation of Mechanical Assistance for the Treatment of Congestive Heart Failure (REMATCH) trial, use of left ventricular assist devices (LVADs) as destination therapy is now approved for some patients ineligible for transplantation.38 Use of the HeartMate vented electric device was shown to improve survival over a 2-year period in this population. Although this landmark study ushered in the era of destination therapy, adverse effects related to infection, stroke, bleeding, device failure, and other factors all call for technological advances to improve outcomes.
Tables 1 and 2
outline LVAD patient selection criteria for a bridge to transplantation and destination therapy, respectively. The main hemodynamic criteria include systolic blood pressure <80 mm Hg, pulmonary capillary wedge pressure >20 mm Hg, and a cardiac index <2.0 L · min1 · m2 despite maximal inotropic and/or IABP support.39 In practice, however, the clinical judgment of experienced surgeons and cardiologists influences the timing of implantation on the basis of estimated waiting time for transplantation and other comorbidities that complicate device insertion.
|
|
Many devices are available to provide temporary support, either as a bridge to another treatment or until recovery of heart function. Selecting the appropriate device is dependent on a number of considerations, including the anticipated duration of patient support, the need for right-sided support, and the patients size (Table 3).
|
Extracorporeal Devices
Centrifugal pumps are commonly used to propel blood for standard cardiopulmonary bypass,40 and systems based on these pumps are available for short-term support of critically ill patients. Left- and/or right-sided support can be obtained with these systems. If necessary, an oxygenator may be spliced into the circuit to allow extracorporeal membrane oxygenation. Cardiovascular support with extracorporeal membrane oxygenation in adults, however, has yielded poor results.
The Abiomed BVS 5000 consists of 2 extracorporeal pneumatic, pulsatile pumps that can be used for univentricular or biventricular postcardiotomy support but also is used as a bridge to transplantation. The self-regulating device can achieve flows of up to 6 L/min with adequate preload and is typically used for periods up to 7 days. Because it is easy to insert and operate, the Abiomed BVS has been particularly important in community hospitals where the device can be placed and the patient subsequently transferred to a specialized center. Data from the voluntary worldwide Abiomed BVS 5000 database revealed 1513 implants (most for biventricular support), of which 63% were for postcardiotomy failure.41
The Thoratec VAD is a polyurethane-lined pneumatically driven pusher plate device that can also be used for left- and/or right-sided support. The device has a stroke volume of 65 mL and can achieve flows up to 7 L/min. The main advantages of this device include its versatility for biventricular support and the ability to use it in small patients. Data from the Thoratec database indicate a bridge-to-transplantation success rate of 60% in 729 patients, with support durations as long as 515 days.42 It has also been reported that among 183 patients with postcardiotomy failure treated with the Thoratec VAD, 69 (38%) were ultimately weaned from the device.
Intracorporeal Devices
The HeartMate LVAD is approved in 2 forms: the implantable pneumatic (IP) and the vented electric. Both use pusher plate technology with a polyurethane diaphragm to propel blood in a pulsatile fashion at rates up to 10 L/min and provide only left ventricular support. Each device has a sintered titanium housing that promotes rapid adherence of blood elements and development of a pseudointima.43 Because of this innovation and the use of xenograft valves, anticoagulation can be avoided with a low thromboembolic rate (2.7%).44 The other advantage of this device is the fact that patients can freely ambulate and can be discharged from the hospital. Except for swimming, patients may resume almost all other activities. The challenges associated with the device include the requirement that the patient have a body surface area
1.5 mol/L2, and because the driveline is percutaneous, there is a relatively high risk of infectious complications, with as many as 48% of patients experiencing some type of infection.45 Data from the worldwide registry as of October 1999 accounted for 1934 implants from 160 different centers used to bridge to transplantation.46 Of patients surviving the first 30 days of support, 84% survived to either transplantation or explantation of the device. Patients who ultimately receive transplants do as well as non-LVAD United Network for Organ Sharing (UNOS) status 2 patients and better than non-LVAD UNOS status 1 patients.
The Novacor left ventricular assist system is another device that is FDA approved as a bridge to transplantation.46 This system is an electric dual-pusher-plate system with a polyurethane lining and 2 porcine valves with flows up to 10 L/min. Like the HeartMate, it can be used only for left-sided support and in patients weighing >60 kg. The device has a percutaneous lead that serves as both a vent and an electrical connection. The blood contacting surface of the system requires anticoagulation, which needs to be started soon after implantation if bleeding is relatively well controlled. Patients are able to ambulate and participate in many activities. Despite anticoagulation, the incidence of thrombotic stroke remains relatively high. Like all other devices with a percutaneous line, there is a significant risk for infection. The advantages of this system are the excellent mechanical reliability (few device failures) and successful outpatient experience. Results from the US multicenter trial of 129 patients satisfying all criteria for inclusion demonstrated a transplantation rate of 77%, despite a 30% systemic infection rate and a 26% embolic stroke rate. The global experience has demonstrated support times as long as 4 years. This device is currently under evaluation for destination therapy.
The Arrow Lionheart-2000 is an experimental LVAD that can pump up to 8 L/min. Unlike other LVADs, however, this device is totally implanted. The battery, controller, and a gas compliance chamber are implanted with no lines crossing the skin. The pump is powered via continuous transcutaneous energy transfer that maintains charge on the battery. The battery can operate the device for
1 hour when not being charged, but only 20 to 30 minutes daily is recommended. The device was designed for destination therapy. A 260-patient study in Europe, the CUBS trial, demonstrated reliability of the device over an average of
1 year, with survival and adverse events similar to those reported in the REMATCH trial. Additional testing is underway.
Total Artificial Hearts
Currently, 2 devices are being developed as total artificial hearts. The AbioCor Artificial Heart is a total artificial heart that is implanted in the orthotopic position. It is driven by an internal motor using hydraulically coupled chambers so that, while the left side is ejecting, the right side is filling. The device is completely implantable and has an internal battery and a transcutaneous energy transfer system that permits recharging the battery or running the device with an external coil. Several patients have received the device, and reports of initial experiences have appeared.47
The Cardiowest Total Artificial heart was recently approved by the FDA as a bridge to transplantation in patients in need of biventricular support.48 This is a biventricular, pneumatic, pulsatile pump that totally replaces the native ventricles and all 4 cardiac valves. Like the AbioCor, it is implanted in the orthotopic position. Eighty-one patients treated with this device had a rate of survival to transplantation of 79%. This device is currently powered by a large console that precludes hospital discharge and thus widespread applicability. Development of a smaller console that will allow greater patient mobility is underway.
New Devices
The newest generation of devices includes axial flow and centrifugal pumps. Many devices include magnetically levitated rotating propellers or plates that eliminate a failure mode related to wear on bearings. These pumps are smaller than the devices discussed above but generally have flow rates >5 L/min. Flow is nonpulsatile with all these devices. Potential advantages of these devices include their durability, small size, simpler mechanics, quiet function, and lack of valves (Table 3).49
The smaller size of these pumps is expected to facilitate placement and explantation, with some pumps being placed by lateral thoracotomy incision without the need for cardiopulmonary bypass. The small size should also permit use in smaller patients. Because they have fewer moving parts and points of friction, it is expected that, once perfected, they will have improved longevity. Unfortunately, if device failure occurs, there are few options other than replacement.
Ventricular Assist Devices as a Bridge to Recovery
Early in the course of LVAD use as a bridge to transplantation, it became evident that some patients exhibit substantial recovery of ventricular function.50,51 This prompted explantation of some devices in lieu of transplantation, so-called bridge to recovery therapy.5254 However, outcomes after these early experiences were poor, with many patients progressing rapidly back to heart failure or dying of heart failurerelated complications.50,52 Therefore, LVADs are not generally used with the intention of bridging patients to recovery except in relatively rare, potentially reversible situations (eg, postcardiotomy failure and fulminant myocarditis). However, many insights have been obtained into the mechanisms of remodeling through studies of hearts supported by LVADs.5566
LVADs provide profound unloading of the left ventricle, although the right ventricle remains hemodynamically loaded (Figure 6A).55 LVADs normalize the circulating neurohormal and cytokine levels66 which would be expected to affect right and left ventricles equally. The effects of LVAD support on the left ventricle include (but are not limited to) near normalization of the passive left ventricular pressure-volume relations (Figure 6B and 6C),55,60 regression of left ventricular free wall mass and cellular hypertrophy (Figure 6D), normalization of the force-frequency relationship (Figure 7A), normalization of ß-adrenergic responsiveness (Figure 7B), normalized phosphorylation of the calcium release channel (ryanodine receptor),62 normalized expression of key genes encoding for proteins involved in calcium cycling,60 and normalized tissue levels of norepinephrine (unpublished). These observations are collectively referred to as reverse remodeling.55 However, in the right ventricle (Figure 6C and 7
B),60 only normalization of ß-adrenergic responsiveness, calcium release channel phosphorylation, and tissue norepinephrine levels are observed.67 These observations indicate that some aspects of reverse remodeling are regulated primarily by hemodynamic stress, whereas other aspects are regulated by neurohormonal milieu.60
|
|
Although most characteristics improve toward normal during LVAD support, this trend is not ubiquitous. Interstitial fibrosis and total, cross-linked, and type I and III collagen contents are all increased beyond the already abnormal levels found in the heart failure state.61,68 The implications and mechanisms of these changes are under investigation.
| Summary and Conclusions |
|---|
|
|
|---|
Because this field is moving so quickly, it is likely that by the time this review appears in press, new information about several of the devices discussed will already be available, some of the approaches may already have been abandoned, and new approaches will have appeared. It is therefore hoped that the enduring message of this review relates to fundamental concepts learned over the recent decades about why and what types of device-based approaches are likely to play an increasing role in the management of patients with heart failure in the future.
| Acknowledgments |
|---|
Dr Burkhoff is employed by Impulse Dynamics, which is investigating an electrical-based treatment for CHF that is not related to the subject of this article. Dr Burkhoff also has an ownership interest in Abiomed and has served as a consultant to Acorn and Cardiac Assist.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
C. Butter, S. Rastogi, H.-H. Minden, J. Meyhofer, D. Burkhoff, and H. N. Sabbah Cardiac contractility modulation electrical signals improve myocardial gene expression in patients with heart failure. J. Am. Coll. Cardiol., May 6, 2008; 51(18): 1784 - 1789. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. S. Groth, B. A. Whitson, J. D'Cunha, R. S. Andrade, and M. A. Maddaus Diaphragmatic hernias after sequential left ventricular assist device explantation and orthotopic heart transplant: Early results of laparoscopic repair with polytetrafluoroethylene J. Thorac. Cardiovasc. Surg., January 1, 2008; 135(1): 38 - 43. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Aggarwal, F. Cheema, M. C. Oz, and Y. Naka Long-Term Mechanical Circulatory Support Card. Surg. Adult, January 1, 2008; 3(2008): 1609 - 1628. [Full Text] |
||||
![]() |
K. L. Baughman and J. A. Jarcho Bridge to Life -- Cardiac Mechanical Support N. Engl. J. Med., August 30, 2007; 357(9): 846 - 849. [Full Text] [PDF] |
||||
![]() |
|