The field of heart transplantation has evolved tremendously since Alexis Carrel first explanted a canine heart and anastomosed it to the carotid artery and jugular vein of a recipient dog in 1905.1 In 1960, Norman Shumway and Richard Lower at Stanford described a technique for orthotopic canine heart transplantation and demonstrated adequate physiologic function of the denervated heart.2 Their work paved the way for the first successful human-to-human heart transplant by Dr Christiaan Barnard in South Africa in December 1967,3 followed closely by Dr Shumway’s group at Stanford in January 1968.4 By 1969, over 100 heart transplants had been performed worldwide.
However, enthusiasm for heart transplantation soon waned because early mortality due to opportunistic infections and acute rejection remained unacceptably high, with one-year survival of only 22%.5 Most centers subsequently abandoned heart transplantation, with the exception of Stanford Hospital, where research directed at improving heart transplant outcomes continued. The development of cyclosporine, a potent and effective immunosuppressive drug, represented a crucial breakthrough in the early 1980s, and its introduction was associated with an increase in the one-year survival rate to the 80% range. This, combined with improved infection prophylaxis and accuracy of rejection diagnosis, led to increasing heart transplant volumes. By the late 1980s, several hundred heart transplants were performed annually.
Heart transplant volumes continued to increase until the mid-1990s, when over 4,000 transplants were performed annually worldwide; subsequently, the number of transplants slowly declined, largely due to a critical organ donor shortage.6 By the end of 2005, 48% of heart transplant candidates had spent more than two years on the waiting list, compared with 17% in 1993.7
Patient Selection for Heart Transplantation
Heart transplantation is indicated in patients with end-stage heart failure of any etiology, including ischemic, dilated, valvular, and congenital cardiomyopathies. The decision to list a patient for transplantation is complex and can be guided by published consensus criteria.8,9 A useful test for estimating the potential benefit gained by heart transplantation is a maximal cardiopulmonary exercise test (CPX), performed to determine the anaerobic threshold and peak oxygen consumption (VO2) of patients while on optimal medical therapy.
A peak VO2 of <50% predicted, ≤12mg/kg/minute when on beta-blocker therapy, or ≤14mg/kg/minute in patients intolerant to beta-blockers is associated with survival benefit after transplantation.10 Heart failure prognostic models such as the Heart Failure Survival Score can also guide decision making.11
A major cause of early post-transplant morbidity is right heart failure,12 mainly resulting from elevated pulmonary vascular resistance (PVR) in the transplant recipient. Currently, a PVR greater than 4–5 Wood units is considered a relative contraindication to transplantation.13 The risk of post-operative right heart failure persists even if acute vasodilator testing, such as with nitric oxide or nitroprusside, demonstrates ‘reversibility’ of pulmonary hypertension.14
Other relative contraindications to heart transplantation include advanced age (generally >70 years of age), a history of cancer, and obesity. Cancer may not be a contraindication if cure or remission has been achieved and there is a low likelihood of recurrence. With regard to obesity, a body mass index (BMI) ≤30kg/m2 is recommended, given the higher incidence of infections, poor wound healing, and pulmonary complications in obese patients. In practice, increasingly high BMI cut-offs are being set, owing to the rapidly increasing prevalence of obesity in the US. At the other end of the spectrum, malnourished and cachectic patients are also considered to be poor candidates.
Other medical co-morbidities that are considered to be relative contraindications to transplantation include uncontrolled diabetes with end-organ damage, irreversible renal dysfunction (often defined as a glomerular filtration rate [GFR} <40ml/minute), and peripheral arterial disease, particularly if symptomatic. Coexistent heart and renal failure has increasingly become an indication for combined heart–kidney transplantation,15 but consensus criteria for appropriate patient selection do not yet exist. From a pulmonary standpoint, a forced expiratory volume in one second (FEV1) < one litre or recent (< four weeks pre-operatively) pulmonary infarction are considered relative contraindications to transplantation. Other relative contraindications include hepatic disease that shortens life expectancy and pre-transplant osteoporosis, owing to the need for post-operative steroid therapy.
Finally, active substance use (alcohol, tobacco, or illicit drugs) is a contraindication to transplantation, as is under-treated mental illness, medical non-compliance, lack of adequate care-giver support, and lack of financial resources. In practice, the decision regarding listing for heart transplantation is complex, requires in-depth medical, social, and psychological assessment, and rests on a consensus from a multi-disciplinary team.
The Heart Transplant Waiting List and Organ Allocation
In the US, patients awaiting heart transplantation are listed at individual transplant centers and are assigned a status code (1A, 1B, 2, or 7) based on their severity of illness. Patients who are designated Status 1A are critically ill and are given highest priority. These patients must be in the intensive care unit and must have a pulmonary artery catheter in place. The 1A status also includes patients who require mechanical circulatory support, need mechanical ventilation, require multiple inotropes (or a single high-dose inotrope) to maintain acceptable hemodynamics, or who have a life expectancy of less than seven days. Status 1B patients are those who have had a ventricular assist device for >30 days or who require continuous low-dose inotropic infusion. Patients who do not meet status 1A or 1B criteria are defined as status 2, and those who are inactive on the waiting list (owing to acute intercurrent illness or other circumstances that temporarily preclude transplantation) are listed as status 7. Support for the current status code system comes from the United Network for Organ Sharing (UNOS), which reported 40.2% one-year survival for status 1A patients who did not undergo heart transplantation, compared with 81.4% one-year survival for status 2 patients.7 For both groups, one-year survival was significantly improved in patients who received heart transplants.
Once status codes are assigned, organs are first allocated to transplant centers closest to the donor, in order to limit the graft cold ischemic time to no more than four to six hours. If no local match is found, the donor organ is offered to centers further away, with Status 1A recipients having priority over less critically ill patients within each donation service area. According to UNOS data, the median waiting time in 2004 was 50 days for patients listed as status 1A, 78 days for status 1B, and 309 days for status 2. Within one year of listing, 84% of status 1A patients are transplanted, compared with 82.5% of status 1B and 53.7% of status 2 patients.16
Current Heart Transplant Outcomes
The International Society for Heart and Lung Transplantation (ISHLT) has recorded data on close to 90,000 heart transplants worldwide since 1983, and reports from their registry are published annually.6
Heart Transplant Volumes and Indications
The number of heart transplants reported annually to the registry peaked in the mid-1990s, at 4,429 in 1995, and has remained relatively stable over the decade, with 3,000–3,500 annually. However, approximately 2,000 heart transplant surgeries are not reported to the registry yearly,17 resulting in a worldwide transplant volume that likely exceeds 5,000 per year. Reporting to the registry is mandatory in the US, but not in other countries.
The main indication for heart transplantation in the past five years has been non-ischemic cardiomyopathy, which accounted for 53% of recipients, followed by ischemic cardiomyopathy (38%). Retransplantation (3%), adult congenital heart disease (3%), valvular heart disease (3%), and other diagnoses such as intractable angina or arrhythmias account for the remaining transplants.6
The majority of transplant recipients (77%) are male, with a mean age of 54┬ü}12 years. An increasing number of elderly patients are being transplanted: 10.5% of recipients between 2002 and 2009 were ≥65 years of age, and 1% were ≥70 years of age. The number of recipients with comorbidities continues to grow with 22% of patients having pre-transplant diabetes mellitus, 41% with hypertension, and 42% with previous cardiac surgery. Recipient demographics also reflect an acutely ill patient population: 46.3% of recipients are hospitalized at the time of transplant, and 44.5% are supported by intravenous inotropes. Finally, in the past decade, the number of patients with mechanical circulatory support devices ‘bridging them to transplant’ has increased dramatically—20% of recipients have a left ventricular assist device (LVAD) and 3% have a right ventricular assist device (RVAD) at the time of transplantation.6
The median survival (time at which 50% of recipients remain alive) after heart transplantation is 10 years for all patients transplanted since 1982, with a median survival of 13 years for those who survived the first year after transplantation (see Figure 1). Currently, mortality risk is highest in the first six months post-transplant, even though short-term survival has improved significantly since the early 1980s, largely owing to refinements in surgical technique, improved peri-operative care, and prevention of acute rejection and opportunistic infections. Unfortunately, in the past 20 years, long-term survival has not improved significantly, and has mainly been limited by the development of cardiac allograft vasculopathy (a form of chronic rejection), malignancies, and complications of immunosuppressive medications. Currently, one-year survival is approaching 90%, five-year survival is approximately 75%, and 10-year survival is 50%.6 Among survivors, functional status is excellent, with approximately 90% of heart transplant recipients reporting no limitations to performing their daily activities (see Figure 2).
The most significant risk factor for one-year mortality after heart transplantation is pre-transplant mechanical circulatory support: RR for mortality is 2.73 for extracorporeal membrane oxygenation (ECMO), 1.94 for total artificial heart, 1.33 for continuous-flow LVADs, and 1.24 for pulsatile-flow LVADs. However, the ISHLT data show improvement in patients bridged to transplantation in the most recent era. Furthermore, the excess risk appears to be limited to the first six months post-transplant. Other risk factors for one-year mortality include congenital heart disease (RR 2.27) and ischemic cardiomyopathy (1.16). Finally, comorbidities that increase short-term mortality include need for hemodialysis, mechanical ventilation, prior need for blood transfusions, recent infection, increasing donor and recipient age, and prolonged allograft ischemic time.
The risk factors for five-year mortality, per the ISHLT registry report,6 are similar to those for one-year mortality, and also include recipient history of pregnancy, female allograft allocation to a male recipient, and recipient history of stroke. When excluding patients who died within the first year post-transplant, major risk factors for five-year mortality included acute rejection, need for surgical interventions post-transplant, diabetes, and renal dysfunction.
Predictors of long-term survival remain poorly understood, mainly due to lack of comprehensive clinical data collected on patients in earlier eras. Nevertheless, younger donor and recipient age, non-ischemic causes of heart failure, shorter allograft ischemic time, and PRA <10% (panel of reactive antibodies—an estimate of percentage of potential donors against whom the recipient has pre-formed antihuman leukocyte antigen antibodies) appear to be favorable prognostic variables.6,18
The leading causes of death vary by time post-transplant. Within the first year, primary graft failure and infection represent the main threats to survival. Acute rejection accounts for at most 10% of deaths. After the first year infectious complications decrease, likely due to reductions in maintenance immunosuppression, and deaths due to malignancies increase to approximately 20–25% of all deaths from three years to more than ten years post-transplant. Graft failure is the leading cause of death throughout the post-transplant period, but the etiology of graft failure is often poorly understood or inadequately described. In the intermediate- to long-term period post-transplant, graft failure is likely to be related to chronic graft injury from processes such as antibody-mediated rejection or cardiac allograft vasculopathy (CAV). Cases of confirmed CAV are responsible for 10–15% of deaths annually, starting at approximately three years post-transplant.6
Acute Allograft Rejection
In the current era, approximately 30% of heart transplant recipients experience acute rejection within the first year post-transplant, and female sex and younger recipient age appear to be significant risk factors.6 With improvements in medical therapies, acute rejection is an uncommon cause of mortality after transplantation, but remains a significant risk factor for hospitalization, graft dysfunction, and development of CAV.6 Routine screening for acute rejection was made possible through development of the transvenous endomyocardial biopsy by Philip Caves at Stanford University in 1973.19 Currently, most heart transplant recipients undergo periodic routine screening for acute rejection, with up to 12 to 15 endomyocardial biopsies performed in the first year alone.
As our understanding of the pathophysiology of acute rejection improves, we now recognize two major subtypes: acute cellular rejection (ACR) and antibody-mediated rejection (AMR). In cellular rejection, an inflammatory response occurs in which effector T-cells, macrophages, and plasma cells infiltrate the graft. Cellular rejection is classified according to extent of cellular infiltration and myocyte damage (0R-no rejection, 1R-mild rejection, 2R-moderate rejection, 3R-severe rejection)20 and grades 2R or higher are considered to be clinically significant.
Antibody-mediated rejection occurs when alloantibodies against donor human leukocyte antigens target the capillary endothelium of the transplanted heart.21–23 Although the presence of AMR is increasingly recognized, no firm consensus has been reached on criteria for its diagnosis. Currently, AMR is suspected when histology demonstrates capillary injury, endothelial cell swelling, and the presence of intravascular macrophages. Positive immunofluorescence (C3d, C4d) or immunoperoxidase staining (CD68) further supports this diagnosis.20 However, there is great controversy regarding the significance of these biopsy findings in the absence of clinical symptoms or signs of graft dysfunction.
Recent attention has focused on non-invasive means to detect acute rejection, in order to spare the patient the risks and discomfort associated with the endomyocardial biopsy, and to reduce health care costs. These monitoring strategies include: (1) monitoring of graft function via imaging, particularly echocardiography; (2) measurement of B-type natriuretic peptide (BNP) levels; (3) genomic markers of rejection; and (4) direct immune function assays.
Many studies over the past two decades have examined the relationship between echocardiographic parameters and acute rejection. Abnormal diastolic parameters, including shortening of the isovolumic relaxation time and the mitral valve pressure half-time have been shown to be sensitive (although not specific) markers of cellular rejection, even in the absence of measureable changes in systolic function.24 More recently, alterations in pulsed-wave tissue Doppler imaging has been shown to have >90% positive and negative predictive value for the diagnosis of acute cellular rejection.25 Along with echocardiographic changes, a rise in BNP levels may also suggest allograft dysfunction,26 although absolute BNP cut-offs are unlikely to be helpful in this patient population with a high incidence of pulmonary disease, renal dysfunction, and other comorbidities.
There has been considerable effort to identify genomic markers of acute rejection that reflect changes in the recipient’s immune response through the use of gene expression profiling (GEP). Multiple genetic pathways are activated during acute cellular rejection, including those involved in T-cell activation and trafficking, natural killer cell activation, and alloimmune recognition.27 Thus, the expression profile of certain genes in peripheral blood mononuclear cells (PBMCs), assayed from patient blood samples, has been demonstrated to differ between ‘quiescent’ patients and those with moderate-severe cellular rejection (≥2R) episodes.28,29
The AlloMap® molecular expression test (XDx, Inc.) is the first Food and Drug Administration (FDA)-approved test based on this research. This test has a low positive predictive value; however, its use in conjunction with clinical observations and echocardiograms has been shown to safely reduce the number of biopsies performed, without increasing the risk of serious cardiovascular events.30 Finally, direct immune function assays that monitor T-cell function (Immuknow® assay, Cylex, Inc.) are being used in some centers,31,32 although data regarding their utility is still lacking.
Cardiac Allograft Vasculopathy
Cardiac allograft vasculopathy (CAV) is a form of chronic rejection in the transplanted heart, and is the leading cause of death beyond the first year after heart transplantation.33,34 CAV is a rapidly progressive obliterative vascular disease involving the allograft coronary arteries.35 The lesions of CAV are characterized by diffuse fibrointimal proliferation composed of vascular smooth muscle cells and intercellular matrix, which characteristically spares the internal elastic lamina, thereby distinguishing it from native atherosclerosis.36 The diagnosis of CAV, which occurs in up to 45% of recipients by three years after transplantation,37 is an ominous sign with an associated mortality rate of >40% within the subsequent two years.38
Traditionally, CAV was diagnosed via coronary angiography, looking for diffuse coronary artery luminal narrowing or tapering, or less commonly focal stenosis. However, coronary angiography has since been shown to underestimate the presence of CAV because even severe intimal thickening may be compensated by vessel enlargement and remodeling.39 Owing to this limitation, intravascular ultrasound (IVUS) has gained favor for detecting CAV because of its ability to provide detailed two- and three-dimensional visualization of component layers of the diseased vessel wall40 (see Figure 3). The utility of this screening tool has been validated by studies demonstrating that an increase in coronary artery maximal intimal thickness of ≥0.5mm by one year after heart transplantation predicts long-term mortality.41,42
The etiology of CAV involves both immune and non-immune mechanisms. From an immunologic standpoint, CAV appears to be mediated by persistent recognition of donor antigens by alloreactive recipient T-cells, leading to persistent T-cell activation and proliferation. Perivascular inflammation develops in the donor heart, with recruitment of mononuclear cells and persistently high cytokine levels. This results in the recruitment and proliferation of smooth muscle cells in the intima of the coronary arteries, leading to progressive narrowing of the vessel lumen.43,44 Non-immunologic factors implicated in the pathogenesis of CAV include organ damage during cold preservation, hyperlipidemia, insulin resistance,45 and cytomegalovirus (CMV) infection.46
Several observational and clinical trials have suggested a role for statins,47 calcium channel blockers,48 angiotensin-coverting enzyme (ACE) inhibitors,49 and proliferation signal inhibitors (sirolimus and everolimus)50,51 in preventing the progression of CAV. However, despite the use of these agents, CAV continues to develop in the majority of patients after heart transplantation, leading to unacceptably high morbidity and mortality.52 Coronary revascularization (either via percutaneous coronary intervention or bypass surgery) plays a limited role owing to the diffuse nature of the disease. Retransplantation, finally, may be an option for carefully selected patients with CAV, although survival after retransplantation remains inferior to primary heart transplantation.53
Infectious complications remain an important cause of morbidity and mortality, particularly during the first year post-transplant, when immunosuppression is profound.6,54 In general, infections within the first month post-transplant are mainly nosocomial, those within the first year are often associated with activation of latent infections or opportunistic infections, and after one year infections are mostly community acquired.55 The pattern of infectious complications after transplantation has changed with early weaning of corticosteroids and the widespread adoption of effective antimicrobial prophylaxis. For instance, the introduction of trimethoprim-sulfamethoxazole has significantly reduced the incidence of pneumocystis pneumonia, and infections due to listeria, nocardia, and toxoplasma. Use of valganciclovir has reduced CMV infection and reactivation, and treatment with azole antifungals has reduced the incidence of candidiasis and aspergillosis. Gram-positive bacteria, mainly Staphylococcus species, are now emerging as an important cause of post-transplant infections.54
Current Immunosuppressive Strategies
Immunosuppression after heart transplantation can be divided into three major phases: (1) peri-operative (induction) immunosuppression; (2) maintenance immunosuppression; (3) treatment of acute rejection. Induction immunosuppression is used to provide rapid and effective protection against acute rejection. In patients with pre-existing renal dysfunction or acute kidney injury, induction may also be used to delay introduction of calcineurin inhibitors. Agents used for induction may be divided into two main categories: (1) depleting antibodies (typically antithymocyte globulin) and (2) non-depleting antibodies (such as the interleukin-2 [IL-2] receptor antagonist basiliximab). In 2009, 54% of heart transplant recipients received induction immunosuppression, 27% with IL-2 receptor antagonists and 23% with antilymphocyte antibodies, even though convincing evidence of a survival benefit or reduction in acute rejection has not been convincingly demonstrated.56,57
Most centers use triple therapy for maintenance immunosuppression (see Figure 4), consisting of a corticosteroid (mainly prednisone), a calcineurin inhibitor (cyclosporine or tacrolimus), and an antiproliferative agent (usually mycophenolate mofetil). Prednisone is used in high doses early post-transplant, and is gradually withdrawn or discontinued altogether within the first year. Tacrolimus has emerged as the calcineurin inhibitor of choice at most centers, mainly due to its greater efficacy in preventing acute rejection.58,59 In the current era, 69% of patients are treated with tacrolimus and 29% with cyclosporine by one year post-transplant.6 Finally, mycophenolate mofetil has largely replaced azathioprine as the anti-proliferative agent of choice, largely due to its greater efficacy in preventing acute rejection and in reducing mortality after heart transplantation.60 Recently, target of rapamycin (mTOR) inhibitors/proliferation signal inhibitors have been increasingly used because they have no effect on calcineurin activity and are therefore free of the adverse effects of calcineurin inhibitors, such as nephrotoxicity. The mTOR inhibitors (sirolimus and everolimus) have also been shown to reduce the progression of CAV.50,61 However, these drugs have not been uniformly adopted owing to their significant side-effect profile, which includes impaired wound healing, edema, and oral ulcers.
Treatment of acute rejection depends on the histologic type (cellular versus antibody-mediated) and severity (histologic grade and presence of hemodynamic compromise). In general, high-dose corticosteroids (intravenous methylprednisolone or oral prednisone) are given for cellular rejections graded as ≥2R, or any rejection event associated with hemodynamic compromise. Lymphocyte-depleting agents such as anti-thymocyte globulin are added in patients with high-grade (e.g. 3R) cellular rejection or those in cardiogenic shock. Treatment protocols for antibody-mediated rejection are often center-specific, continue to evolve, and have not been rigorously studied in clinical trials. AMR associated with graft dysfunction is usually treated with high-dose corticosteroids and plasmapheresis, to remove circulating donor-specific antibodies, followed by intravenous immunoglobulin and occasionally rituximab (a B-cell depleting monoclonal antibody).
Mechanical Circulatory Support
Mechanical circulatory support devices (MCSDs) have been developed to augment or supplant cardiac function in patients with intractable heart failure—a growing population worldwide. MCSDs were originally implanted for a limited duration in order to support transplant candidates who otherwise might not have survived until a suitable donor heart became available. This designation, termed ‘bridge-to-transplant’ (BTT), is still the most common reason for MCSD utilization, accounting for 43.7% of reported implants in the U.S. since March 2006.62 The BTT indication was subsequently expanded to include patients who are not actively listed but may become candidates for heart transplantation following initiation of MCSD therapy, so-called ‘bridge-to-decision’ (BTD). Less frequently, MCSDs are used to support the failing heart temporarily until it recovers [‘bridge to recovery’ (BTR)]. Finally, MCSDs are being increasingly used for permanent ‘destination therapy’ (DT) in patients who are not transplant candidates. The main adverse events associated with MCSD therapy are postoperative bleeding, stroke, right heart failure, and percutaneous lead infection.63
The ‘first generation’ MCSDs, introduced in the 1980s, were pulsatile, positive displacement pumps (Novacor®, HeartMate® XVE, Thoratec® IVADTM/PVADTM). Due to their large size, they often could not be implanted in patients with small body surface area. Furthermore, owing to the complexity of the pump function with multiple moving parts, the durability of the devices was limited. These original assist devices have now been largely supplanted by ‘second-generation’ MCSDs (Jarvik 2000, MicroMed DeBakey®, HeartMate II®), introduced from 1998 to 2000.
These devices have axial flow rotary pumps that result in continuous blood flow. The axial flow pumps are smaller than pulsatile devices and are more durable, due to fewer moving parts. The HeartMate II is the only second-generation MCSD that has FDA approval for BTT and DT. ‘Third-generation’ MCSDs (HeartWare HVADTM, Berlin Heart® Incor, DuraHeart®), currently in clinical investigation, consist of suspended rotary pumps without contact bearings. The rotor is suspended using either hydrodynamic or electromagnetic forces, thus removing the need for support bearings and reducing the number of moving parts to one. This technology increases the durability of the device and decreases device size. In fact, some newer devices are small enough to fit entirely within the pericardium.
Outcomes following implantation of MCSDs continue to improve (see Figure 5). In 2001, the landmark Randomized evaluation of mechanical assistance for the treatment of congestive heart failure (REMATCH) trial demonstrated that pulsatile-flow MCSDs dramatically reduced mortality by 48% in 129 patients with severe heart failure who were not eligible for transplantation.64 In 2007, a multicenter observational study of continuous-flow MCSDs that enrolled 133 patients demonstrated 75% six-month survival, 68% 12-month survival, and significant improvements in functional status and quality of life.63 More recently, a randomized trial comparing pulsatile flow to continuous flow MCSDs in patients ineligible for transplantation demonstrated that continuous-flow devices significantly improved the probability of survival free from stroke and device failure at two years (46 versus 11%).65
It is clear that MCSD therapy has become increasingly available, effective, and safe for selected patients with end-stage heart failure, whether for bridge-to-recovery, bridge-to-transplant, or destination therapy. Lessons to-date suggest that the optimal time for referral is before the development of major complications from heart failure, such as irreversible end-organ damage.66 Identifying the optimal timing for implantation in an individual patient’s course often requires special expertise.67
Over the past four decades, heart transplantation has reduced suffering and prolonged the lives of tens of thousands of patients with end-stage heart disease. The field has evolved considerably, with improvements in surgical techniques and post-operative care, and with the introduction of potent immunosuppressive medications and effective drugs to prevent and treat infections. Major challenges continue to exist, including the organ donor shortage, diagnosis and treatment of antibody-mediated rejection, prevention and treatment of cardiac allograft vasculopathy, and the development of more-specific immunosuppressive drugs with fewer long-term side effects. Active research efforts aimed at addressing these problems will continue to move the field forward.