Since 2000, two major developments have ushered in the modern era of left ventricular assist devices (LVADs): the 2001 Randomized Evaluation of Mechanical Assistance for the Treatment of Congestive Heart Failure (REMATCH) trial, which resulted in the Food and Drug Administration (FDA) approval of LVADs for long-term destination therapy and the emergence of smaller continuous-flow devices starting in 2005. It is now indisputable that implantable LVADs are an important therapeutic modality for patients with end-stage heart failure and will remain so for the foreseeable future.

Approximately 5 million Americans suffer from heart failure with over 550,000 new cases diagnosed each year and an estimated 287,000 deaths resulting from heart failure in the United States every year. In addition to the loss of life, heart failure poses a significant financial burden with estimated annual direct costs of $35 billion dollars per year in the United States alone. Despite advances in the management of heart failure, there may be as many as 100,000 people who have been treated with guidelines-based therapy but have remained relatively unresponsive in New York Heart Association (NYHA) Class IIIb or IV. In these end-stage heart failure patients with recurrent hospitalizations, few options exist. Orthotopic heart transplantation, the gold standard of therapy for end-stage heart failure, has been shown to definitively improve outcomes with median survival of 13 years but this is restricted to a select relatively young population. In addition, a critical shortage of suitable grafts limits the number of heart transplants performed in the United States to about 2,000 per year, only 2% to 4% of patients who need definitive therapy. With a growing population of end-stage heart failure patients and a static donor pool, additional treatment modalities must be pursued.

To date, the most promising new treatment to emerge for end-stage heart failure has been mechanical circulatory support with ventricular assist devices (VADs). The development of VADs began in earnest in 1964 with the National Institutes of Health establishment of the Artificial Heart Program, whose stated goal was putting a man-made heart into a human being by the end of the decade. Unlike the moon landing, this lofty goal was not met by the end of the decade; nevertheless, progress proceeded slowly and clinical use of VADs on a more routine basis began in the mid-1980s. The majority of implantable VADs are designed for the support of the left ventricle (LVADs) and our discussion will focus on these devices. Implantable LVADs function by removing oxygenated blood from the apex of the left ventricle, passing it through a mechanical pump, and returning pressurized blood to the ascending aorta. Despite the predominance of implantable LVADs, other temporary nonimplantable configurations do exist and the use of VADs designed for right ventricular assist devices (RVADs) failure, or biventricular failure is also well established.

Currently, implantable LVADs can be classified into two main categories: volume-displacement (pulsatile) and continuous-flow (nonpulsatile) pumps. Implantable LVADs are typically used for either bridge to heart transplantation (BTT) or long-term destination therapy for those patients not eligible for transplant. Initial clinical success was achieved with the so-called first-generation devices, or pulsatile pumps, which include the Heart-Mate XVE and its predecessors the Heart-Mate IP1000 and HeartMate VE (Thoratec Corp., Pleasanton, CA), the Thoratec PVAD (Paracorporeal Ventricular Assist Device) and IVAD (Implantable Ventricular Assist Device; Thoratec Corp.), and the Novacor LVAS (World Heart Corp., Oakland, CA). These first-generation LVADs came into clinical use in the mid-1980s and all of the above are FDA-approved for the BTT indication. Of these first-generation devices, only the HeartMate XVE is FDA-approved for destination therapy. In the landmark 2001 REMATCH trial, patients with irreversible heart failure who were ineligible for transplantation were randomized to either maximal medical therapy or HeartMate XVE implantation. LVAD implantation doubled the 1-year survival rate of these patients from 25% to 52%. The FDA subsequently approved the HeartMate XVE for permanent destination therapy, ushering in the modern era of LVAD therapy.

In recent years, the pulsatile first-generation LVADs have been replaced in clinical practice by second-generation, continuous flow LVADs. These devices, which include the HeartMate II (Thoratec Corp.), Jarvik 2000 FlowMaker (Jarvik Heart, Inc., New York), and MicroMed–DeBakey (MicroMed Cardiovascular, Inc., Houston, TX), have an internal rotor within the blood flow path that is suspended by contact, bloodimmersed bearings. These continuous-flow, rotary pumps were introduced to overcome many of the shortcomings of the first-generation devices, and they are simpler in design with only a single moving part: the internal rotor. Advantages over first-generation devices include smaller size requiring less extensive surgical dissection for implantation, the absence of valves that are a primary site of wear, higher efficiency with less energy requirement, and a smaller percutaneous lead. With these improvements, the second-generation devices have demonstrated improved reliability with device support of more than 6 years reported. The most extensive clinical experience is with the HeartMate II.

Currently, the HeartMate II is the only second-generation device approved for both BTT and destination therapy in the United States. In a recently published randomized clinical trial comparing the HeartMate II versus the first-generation HeartMate XVE for destination therapy, the HeartMate II was shown to be significantly better than the HeartMate XVE in achieving the primary end point of survival free from device failure or disabling stroke at 2 years. Moreover, patients with HeartMate
II support also had significantly superior actuarial survival rates at 1 (68% vs. 55%) and 2 years (58% vs. 24%). More recent evidence shows that the actuarial 1-year survival rate for those patients receiving a HeartMate II is now almost 80%, a staggering improvement from the 52% seen in the REMATCH trial barely 10 years ago, and a number which approaches the accepted survival rate for the gold-standard of heart transplantation.

Despite recent developments in device technology and patient care, there is no substitute for proper patient selection in the achievement of optimal outcomes. Whenever possible, LVAD implantation should be performed under elective and not emergent circumstances. Appropriate patient selection and proper timing of implantation are likely more important than any other aspect of a successful LVAD program. Each patient’s comorbidity profile, ability to survive surgery, social support, and expected functional level after implantation should be carefully assessed prior to surgery. Optimal timing of LVAD implantation should occur prior to the development of irreversible end-organ damage but after correction of reversible comorbidities. Risk assessment of medical and surgical issues for both BTT and destination therapy are virtually identical and close collaboration with heart-failure cardiologists is essential for the identification of those patients who will benefit most from LVAD implantation.

Numerous factors are known to influence outcomes and most risk factors are additive. Several evidence-based predictive models, composite risk scores, and defined patient profiles are available, which are useful in selecting appropriate patients for LVAD implantation. For example, the Seattle Heart Failure model is a useful tool to identify those heart-failure patients who will derive maximum benefit from LVAD. Using easily obtained clinical parameters, a heart failure patient’s probability of 1- and 2-year survival with and without LVAD can be calculated using this validated model. In addition, the Lietz and Miller 90-day predictive model identifies nine weighted risk factors affecting mortality. This model has proven useful but must be applied with care as it was developed by analysis of 222 patients undergoing HeartMate XVE implantation for BTT and is not validated for continuous-flow LVADs or destination therapy. The Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) has defined seven patient profiles and has correlated survival after LVAD with these patient profiles (Table 55.1). For example, those in profile 3, stable but inotrope dependent, have the best survival while patients in profile 1, cardiogenic shock, have high mortality and may be too sick for implantable LVAD placement. We find the combination of the INTERMACS profile and Seattle Heart Failure model most useful and believe that these tools support the trend in practice toward LVAD implantation earlier in the progression of heart failure.

One key factor for optimal patient outcome after LVAD implantation is RV function. RV failure is a leading cause of morbidity and mortality due to inability of the RV to pump sufficient blood through the pulmonary circuit. The LVAD unloads the LV allowing a decrease in pulmonary artery pressures and a corresponding reduction in RV afterload. On the other hand, LVADs increase systemic vascular return to the RV and reduced LV pressure can cause displacement of the interventricular septum into the left side, thereby reducing efficient RV function. In our practice, postimplant RV failure can be anticipated preoperatively and can be avoided through a combination of careful patient selection, prompt pharmacologic treatment of at-risk patients, and careful protection of the RV intraoperatively.

Other key factors that should be specifically addressed as part of a comprehensive preoperative assessment include nutritional status, renal function, hepatic function, gastrointestinal (GI) bleeding, pulmonary function, coagulation status, infection, neurologic function, and perhaps most importantly, psychosocial and psychiatric function. One of our key exclusion criteria for LVAD is the lack of a well-established social support system. Patients will require extensive assistance after leaving the hospital and must demonstrate the presence of a supportive network of family and friends who will be available to assist
with care. Given the complexity of management concerns, comorbidity profiles, and social issues associated with LVAD implantation, the importance of a multi-disciplinary team approach to patient care cannot be overemphasized. All social and medical risk factors are assessed and addressed by our team of LVAD professionals including the surgeon, heart failure cardiologist, LVAD coordinator, nurse practitioners, social workers, pharmacists, dieticians, and many others. Table 55.2 lists our minimum goals with regard to standard metabolic markers associated with several patient risk factors. Nutritional status, as an example, demonstrates the importance of the multidisciplinary approach to optimizing patient outcomes. Malnutrition is common in patients with heart failure and if not addressed prior to implantation, it leads to increased infection risk, decreased functional recovery, and generally poor outcomes. For patients with a prealbumin <15, enteral feeding should be initiated preoperatively and should continue through the postoperative period until the patient is taking adequate oral nutrition. Caloric intake, nutritional supplementation, and the metabolic response to changes in nutritional management must all be monitored carefully throughout the preoperative, postoperative, and outpatient course of care. This is best achieved with the input of multiple specialists. Similar proactive approaches must be taken toward the other mentioned patient comorbidities and directed therapy should begin prior to LVAD implantation and continue throughout the postoperative stay and outpatient course of therapy.

Implantation of a long-term LVAD should not be considered in patients with irreversible major end-organ dysfunction, severe hemodynamic instability, unknown neurologic status, major coagulopathy, irreversible respiratory failure requiring mechanical ventilation, bacteremia, or right heart failure. In patients who present in cardiogenic shock and accompanied end-organ dysfunction, we have found success with temporary mechanical support in the form of the TandemHeart^® Percutaneous Ventricular Assist Device (pVAD) system (Cardiac Assist, Inc., Pittsburgh, PA). This device, which can be inserted under fluoroscopic guidance in the catheterization laboratory, employs a transeptally placed inflow cannula to pump oxygenated blood from the left atrium, through an external centripetal blood pump, and back into the outflow cannula which is secured in the femoral artery. Flow rates of up to 4.0 L/min can be achieved and we have successfully used this pVAD system in multiple patients who presented in cardiogenic shock, in whom the reversibility of their condition was unknown. In our experience, patients who demonstrate significant reversal of end-organ failure within several days after initiation of pVAD support are excellent candidates for subsequent implantable LVAD therapy. To date, we have successfully placed implantable HeartMate II LVADs in four patients who were bridged temporarily with a Tandem Heart pVAD.

The physiologic goal of LVAD implantation remains the same regardless of the device implanted: mechanical unloading of the failing left ventricle along with an increase in aortic perfusion. The decompression of the LV and increase in systemic perfusion in turn lead to improved RV efficiency— via an increase in venous return to the RV and a reduction in pulmonary pressures by unloading the LV and decreasing pulmonary pressure afterload. All of these changes reduce cardiac work, improve myocardial perfusion, and allow for reversal of the neurohormonal activation and catabolic state present in patients with progressing heart failure. The combination of improved perfusion along with suspension of the heart failure milieu allows for recovery of nonreversible end-organ damage and, in rare cases, significant myocyte recovery allowing for subsequent device removal.

While the physiologic changes that accompany LVAD implantation have been proven to provide relief from heart failure for properly selected patients, the re-routing of blood flow and concomitant flow and pressure changes in the cardiac chambers and great vessels must be carefully considered in each patient. After LVAD placement, the majority of systemic blood flow bypasses the aortic valve, and the structural integrity and valvular function of the heart must be carefully evaluated in light of these anticipated changes. Echocardiography both prior to LVAD implantation and at the time of surgery is essential for the evaluation of cardiac function and identification of anatomic abnormalities that may warrant correction at the time of surgery. All atrial and ventricular septal defects should be repaired during LVAD implantation and a dedicated bubble study is an important part of the echocardiographic evaluation. Even the smallest patent foramen ovale (PFO) must be closed during the LVAD procedure as any PFO may become hemodynamically significant after LVAD placement. For this reason, it is imperative to perform careful echocardiographic evaluation for the presence of any atrial or ventricular septal defects both before and after initiation of LVAD support.