INTRODUCTION — Cardiac transplantation is the definitive therapy for eligible patients with end-stage heart failure. The major limitations to survival in the early post-transplant period (first year) are nonspecific graft failure, multiorgan failure, acute rejection, and infection [1]. Beyond the first year, cardiac allograft vasculopathy (CAV, also called transplant coronary artery disease or cardiac transplant vasculopathy) is among the top three causes of death [1]. Approximately 30 percent of patients have angiographic coronary artery disease at five years and 50 percent at 10 years, with the incidence increasing progressively with time [1]. (See "Heart transplantation in adults: Prognosis", section on 'Causes of death' and "Heart transplantation: Clinical manifestations, diagnosis, and prognosis of cardiac allograft vasculopathy".)
The pathogenesis of CAV will be reviewed here. The diagnosis, prevention, and treatment of this disease are discussed separately. (See "Heart Transplantation: Prevention and treatment of cardiac allograft vasculopathy" and "Heart transplantation: Clinical manifestations, diagnosis, and prognosis of cardiac allograft vasculopathy".)
PATHOLOGY — Cardiac allograft vasculopathy (CAV) is a panarterial disease confined to the allograft and is characterized by diffuse concentric longitudinal intimal hyperplasia in the epicardial coronary arteries (figure 1) [2-4] and concentric medial disease in the microvasculature [5-7]. In contrast, traditional atherosclerosis is focal, noncircumferential, and most often observed proximally in the epicardial vessels.
Transplant recipients also frequently develop proximal coronary artery disease. However, these lesions more closely resemble traditional atherosclerosis pathologically and probably evolve from pre-existing disease in the donor heart that is accelerated by the plethora of cardiac risk factors after transplantation [4].
Serial intravascular (intracoronary) ultrasound testing has shown that most of the intimal thickening occurs during the first year after transplantation [8]. Lumen loss is also due to arterial remodeling, with early expansion and late constriction of the external elastic membrane area. (See "Heart transplantation: Clinical manifestations, diagnosis, and prognosis of cardiac allograft vasculopathy", section on 'Epidemiology'.)
Coronary angioscopy has demonstrated heterogeneous intimal lesions with varied clinical correlates (picture 1) [9].
CAV is also associated with intracoronary mural and occlusive thrombi, which can cause acute myocardial infarction [10,11]. Thrombi typically develop more than two months after transplantation [10]. Early, thin mural thrombi primarily contain platelets, while later organized thrombi, which are often occlusive, primarily consist of fibrin [11]. The important role of platelets is further supported by a strong association between platelet activation and the development and progression of CAV [12].
PATHOGENESIS — Cardiac allograft vasculopathy (CAV) appears to be multifactorial in origin with both immunologic and nonimmunologic factors implicated. Among the factors associated with vasculopathy are cellular and antibody-mediated rejection, donor-specific anti-HLA antibodies, cytomegalovirus infection, and hypercholesterolemia.
Immunologic events appear to be most important, since CAV develops in the donor's but not the recipient's arteries. It has been suggested that CAV reflects an accelerated "normal" healing process following allograft-induced immunologic injury [13].
Consistent with the importance of inflammation is the observation that elevations in serum C-reactive protein, an acute phase reactant, predict a greater likelihood of CAV and allograft failure [14-16]. (See "Heart transplantation in adults: Prognosis".)
Immunologic factors — The development of CAV involves both the cellular and humoral arms of the immune system. Immunologic mechanisms that result in the initiation and maintenance of cellular rejection correlate with the development of vascular injury. These factors include histocompatibility antigen HLA mismatching, T cell activation, endothelial cell activation, and altered cytokine expression. Increasing evidence suggests that donor specific antibodies and antibody mediated rejection are also major contributors to CAV [17]. The basic aspects of transplantation immunobiology are reviewed in detail separately. (See "Transplantation immunobiology".)
HLA mismatch — Several studies have shown a greater incidence of HLA mismatching in recipients who develop CAV than in those who do not [2,3]. HLA-DR mismatching has generally been the most important of the HLA subtypes [18], but a positive correlation with HLA-A mismatching has also been observed [19,20]. Development of de novo donor specific anti HLA Class I or II antibodies after transplant have been associated with CAV [21].
Acute cellular rejection — The number of episodes of moderate to severe cellular rejection appears to correlate with the development of CAV [2,22-25]. In addition a summary of overall rejection, the total rejection score, has also been associated with CAV [26]. This observation is consistent with animal data that suggest that the immunogenicity of the graft is probably the most important stimulus to the development of vasculopathy [27]. However, some studies have not supported a relationship between rejection ≥2R (3A) and vasculopathy [28]. (See "Heart transplantation in adults: Diagnosis of allograft rejection".)
Antibodies and antibody-mediated rejection — Humoral activation, resulting in the production of anti-HLA and antiendothelial antibodies [29-33], enhances the development of CAV, and is associated with acute antibody-mediated rejection (also called humoral or vascular rejection) [29,32,34]. (See "Heart transplantation in adults: Diagnosis of allograft rejection", section on 'Acute antibody-mediated (humoral) rejection'.)
Episodes of antibody-mediated rejection are associated with the development of CAV and poorer long-term outcomes [33-35].
Anti-HLA and antiendothelial antibodies also appear to increase the risk of CAV independent of antibody-mediated rejection. In one study, for example, cardiac transplant recipients who developed anti-HLA antibodies had a lower four-year survival rate than those who did not develop such antibodies (90 versus 38 percent) [30]. CAV was responsible for many of the late (more than one year post-transplant) deaths in these patients. At two years post-transplant in another report, a higher incidence of CAV was found in patients who continued to make anti-donor HLA antibodies than in those who did not (18 versus 3 percent) [31].
A higher rate of accelerated coronary artery disease has also been associated with the presence of antiendothelial cell antibodies (72 versus 5 percent in those without such antibodies) [33].
T cell activation — A complex interplay between activated T lymphocytes and endothelial cells may explain the sequential and temporal changes observed in the infiltrating T cell subtypes found in the coronary arteries of transplanted hearts (figure 2) [36-41]. T cells, HLA-DR+ endothelial cells, and HLA-DR+ macrophages are frequently seen in vasculopathic coronary artery lesions [41].
Endothelial cells, in particular, show significantly increased expression of MHC class I alloantigens [36]. These antigens are thought to be recognized by CD8+ cells, resulting in the secretion of CD8+ derived cytokines that further activate coronary endothelial cells. The activated endothelial cells, in turn, express increased levels of MHC class II antigens, which subsequently activate CD4+ T cells. As a result, CD8+ cells are predominant in the early vasculopathic lesion, while CD4+ cells are found in increasing numbers in later pathologic stages.
Endothelial cell activation — Activation of the endothelial cell provides the proper milieu for the recruitment of proinflammatory cells from the vasculature, the initiation of the immune response, and the subsequent development of CAV. Cytokines released from T cells, macrophages, and other cells (see 'Cytokines' below) dramatically increase the expression of intercellular cell adhesion molecule-1 (ICAM-1), vascular cell adhesion molecule-1 (VCAM-1), and P-selectin on the surface of allograft endothelial cells at the time of acute cellular rejection [42-44].
The prognostic significance of endothelial activation was suggested in a study in which serial endomyocardial biopsy specimens were obtained in 121 donor hearts just prior to and at three months following transplantation [45]. The patients with evidence of endothelial activation during the first three months following transplant were at significantly greater risk for developing vasculopathy (64 versus 28 percent in patients without activation).
Serum levels of soluble ICAM-1 (sICAM-1) also may predict the risk of CAV. Elevated levels correlate with the biopsy finding of ICAM-1 on arterial and arteriolar endothelium and are associated with a greater risk of vasculopathy and graft failure [46].
Thus, evidence of either endothelial activation on endomyocardial biopsy or elevated serum sICAM-I concentration can predict the development of CAV.
In a cohort of 76 heart transplant recipients biopsied at 12 months post-transplantation, antibody-mediated rejection (AMR) positive biopsies showed significantly greater endothelial localization of VEGF than time-matched AMR negative biopsies [47]. The presence of diffuse endothelial expression of VEGF was associated with a 2.5-fold risk of developing CAV while biopsy evidence of AMR was associated with a fivefold risk.
Endothelial dysfunction (assessed via Doppler flow wire arterial responsiveness to acetylcholine, adenosine, and nifedipine) was found to be an independent predictor of cardiovascular events and death [48].
Cytokines — Cytokine expression by infiltrating lymphocytes and macrophages has been detected in transplant vasculopathic lesions. Among the cytokines involved are interleukin (IL)-1, IL-6, tumor necrosis factor (TNF)-alpha, platelet derived growth factor (PDGF), insulin-like growth factor-I (IGF-1), macrophage chemoattractant protein-1 (MCP-1), fibroblastic growth factor, vascular endothelial growth factor, transforming growth factor-alpha (TGF-alpha), and TGF-beta-1 [49-56]. These cytokines have trophic and proliferative effects on coronary artery smooth muscle cells.
Interferon-gamma is produced by certain T cell subsets and may play a central role in CAV by activating macrophages and enhancing the production of MHC products, other components of the antigen presentation pathway, and the adhesion molecules ICAM-1 and VCAM-1 [57]. Interferon-gamma deficient mice appear to be protected against transplant arterial disease but not graft rejection [57].
An increasing number of genetic polymorphisms have been associated with the development of vasculopathy. These include the genes for TNF-alpha and IL-6 [58,59].
Nonimmunologic factors — Nonimmunologic factors contribute to the development of CAV. These factors may act indirectly through immunologic pathways.
Age and sex — Age and sex in both the donor and recipient are predictors of CAV:
●Among donor characteristics, older age and male sex are associated with higher risk [25,60,61]. In a study of 489 one-year heart transplant survivors who underwent 1435 coronary angiograms, the relative risk for CAV was 1.26 for every 10 years of donor age [25].
●Among recipient characteristics, younger age is associated with higher risk [18,25,62].
Hyperlipidemia — Elevations in serum total cholesterol, low-density lipoprotein (LDL)-cholesterol, oxidized LDL-cholesterol, and triglycerides are common after heart transplantation. These changes are probably related to immunosuppressive therapy with corticosteroids and cyclosporine and to obesity [63]. In comparison, lipoprotein(a) levels decrease during the first six months after transplantation and are not associated with vasculopathy [64,65]. (See "Heart transplantation: Hyperlipidemia after transplantation".)
The lipid abnormalities appear to correlate with the severity of the vasculopathy [66-68]. Support for a direct role of hypercholesterolemia comes from a study in mice in which hypercholesterolemia increased the rates of neointima formation and vascular occlusion by a mechanism that depends upon smooth muscle cell accumulation [69]. Further support comes from the observation in heart transplant recipients that lipid-lowering with statins is associated with a significant reduction in the incidence and severity of CAV [70]. (See "Heart Transplantation: Prevention and treatment of cardiac allograft vasculopathy", section on 'Statins'.)
Cytomegalovirus infection — Cytomegalovirus infection has been associated with a higher incidence of CAV [71,72]. One or more of the following mechanisms may contribute:
●A direct endothelial assault, which results in the enhancement of vascular adhesiveness, activation of the coagulation cascade, and elaboration of cytokines.
●Adverse vascular remodeling with greater net lumen loss [73].
●Stimulation of cellular immune responses in the vasculature [63], perhaps via induced expression of MHC antigens on the endothelial cells [74].
Glycemic control and insulin resistance — Both impaired glycemic control and insulin resistance may play a role in the pathogenesis of CAV. As an example, an association has been noted between an elevated HbA1c concentration, a marker of chronic glucose intolerance, and the incidence and severity of CAV [75].
The metabolic syndrome, also called the insulin resistance syndrome or syndrome X, is characterized by abdominal obesity, hypertension, diabetes, and an atherogenic lipid profile (hypertriglyceridemia and low high density lipoprotein [HDL] cholesterol) and is seen frequently after heart transplantation. In a series of 66 patients, high serum insulin or glucose concentrations were associated with a significantly lower likelihood of freedom from CAV (57 versus 82 percent in patients with normal values) and with reduced survival [76]. (See "Metabolic syndrome (insulin resistance syndrome or syndrome X)".)
In 98 consecutive heart transplant recipients, high C-reactive protein (CRP) and insulin resistance (defined as triglyceride to HDL ratio >3.0), were significantly associated with CAV identified by angiography. The combination of increased CRP and insulin resistance was synergistic and associated with a fourfold increased risk in developing CAV [77].
Coronary disease history — Coronary artery disease in either the donor [60] or recipient [18,22] appears to contribute to CAV. In one report, CAV at three years after transplantation was more common in hearts from donors with angiographic coronary disease and from older donors (figure 3) [76]. Although older donor age is one of the strongest predictors for CAV, donor age did not affect recipient survival or freedom from ischemic events at a mean follow-up of 3.8 years.
Fibrinolysis — Animal models suggest that the plasminogen system may contribute to the development of CAV by mediating elastin degradation, macrophage infiltration, media remodeling, medial smooth muscle migration, and the formation of a neointima; as an example, plasminogen deficient mice have a reduced risk of vasculopathy [78].
In contrast, deficient fibrinolysis may be a contributing factor to CAV in humans. Grafts with persistent depletion of tissue-type plasminogen activator (tPA) and expression of its inhibitor, plasminogen activator inhibitor-1 (PAI-1) are much more likely to develop CAV (78 versus 24 percent) and patients with such grafts are much more likely to receive a second transplant or die (30 versus 2.5 percent) [79].
The cause of t-PA depletion is unknown, but may have a genetic basis. It is possible that genotype-specific overexpression of PAI-1 leads to local depletion of t-PA. Consistent with this hypothesis is the observation that PAI-1 levels and impairment of fibrinolysis are linked to the presence of PAI-1 gene polymorphisms in the recipient, ie, 1/1, 1/2, or 2/2 genotypes [80]. In one series of 48 transplant recipients, a donor 2/2 PAI-1 genotype was associated with a significant risk of CAV [81]. The actuarial freedom from any coronary artery disease at 12 and 24 months for the 1/1, 1/2, and 2/2 genotypes was 100 and 100 percent, 92 and 92 percent, and 75 and 45 percent, respectively. There was no association with donor t-PA genotype.
Grafts with a persistent loss of vascular antithrombin also have a higher incidence of vasculopathy. In one report, the vascular disease was more severe and progressed more rapidly in such grafts compared with those that lost and later recovered antithrombin [82].
Endothelial dysfunction — The development of vasculopathy may correlate with the presence of endothelial dysfunction and cardiac death, as established by paradoxical coronary artery constriction in response to acetylcholine or to the cold pressor test [83-86]. (See "Coronary endothelial dysfunction: Clinical aspects".)
The following observations illustrate this relationship:
●In a report of 20 patients, endothelial dysfunction in the early post-transplant period was associated with a significant increase in angiographic vasculopathy at one year (58 versus 13 percent in those without endothelial dysfunction) [84].
●In a series of 45 patients followed with annual angiography, IVUS, and Doppler assessment of endothelial function, the development of endothelial dysfunction was significantly associated with both CAV and a higher event rate [86].
Inducible nitric oxide synthase is upregulated in grafts with CAV and may play a protective role. Consistent with this hypothesis are the observations that CAV is exacerbated in animals deficient in inducible nitric oxide synthase and reduced in animals given gene therapy with endothelial nitric oxide synthase [87,88]. A similar protective effect can be achieved in animals by lowering the levels of asymmetric dimethylarginine, an endogenous inhibitor of nitric oxide synthase [89].
On the other hand, an excessive nitric oxide response may be deleterious, promoting CAV by a process that may involve reaction with the free radical superoxide to form peroxynitrite, a strong oxidant that damages cellular proteins [90].
There are a number of different responses to vessel injury, some of which are cytoprotective and/or inflammatory. Using a global proteomic approach examining protein expression in endomyocardial biopsies from patients with and without angiographic vasculopathy, an increase in heat shock protein 27 (HSP27) was found in patients without vasculopathy [91]. This observation suggests that normal vascular expression of HSP27 may protect against vessel injury, possibly by reducing apoptosis.
ACE gene polymorphism — Angiotensin converting enzyme (ACE) polymorphism may be associated with the development of CAV [92,93]. In a series of 80 heart transplant recipients, the DD genotype in the donors but not the recipients was associated with the development CAV [92]. This observation suggests the importance of tissue rather than circulating ACE. (See "Pathophysiology of heart failure: Neurohumoral adaptations", section on 'ACE gene polymorphism'.)
Endothelin — The endothelial cells release endothelin-1, a potent direct vasoconstrictor that also may stimulate the proliferation of vascular smooth muscle cells and fibroblasts [94]. Intense endothelin-1 immunoreactivity has been demonstrated in arteries with CAV, but not in normal coronary arteries [95].
The potential pathogenetic importance of endothelin-1 was shown in an animal model in which there was local upregulation of endothelin-1 in the thickened neointima and media of the coronary arteries [96]. The administration of bosentan, a nonselective endothelin receptor antagonist, suppressed the development of graft atherosclerosis. (See "Pathophysiology of heart failure: Neurohumoral adaptations", section on 'Endothelin'.)
Vascular remodeling — Adverse vascular remodeling can occur. When an atheromatous plaque develops within an artery, the artery can respond with compensatory dilatation, resulting in preserved net lumen area, or with shrinkage and a further net reduction in lumen area [97-99]. Intravascular ultrasound (IVUS) has allowed an increased understanding of this process. In one report, 78 percent of patients had overcompensation or partial compensatory dilatation, but 22 percent had no compensatory dilatation or shrinkage in the coronary segment [97].
IVUS has also shown that 95 percent of heart transplant recipients had an increase in intimal area over time. Among these patients, only 37 percent had some compensatory dilation, suggesting that abnormal vascular remodeling contributes to the decrease in luminal area [97,98].
Early left ventricular dysfunction — In a review of 117 patients, multivariate analysis showed that mean lumen diameter loss at one year was inversely related to early left ventricular function as defined by fractional shortening on echocardiography performed within the first week after transplantation [61]. It was speculated that myocardial injury associated with brain death in the donor, myocardial preservation, and ischemia-reperfusion injury might lead to injury to coronary endothelium as well as the myocardium.
Etiology of brain death in donor — Explosive brain death (eg, gunshot wound, head trauma, or intracerebral hemorrhage, as opposed to ischemic stroke) in the donor has been associated with an increased risk of late CAV and reduced survival in the recipient in some [100,101] but not all studies [61]. Systemic activation of matrix metalloproteinase (MMP)-2 and MMP-2 is associated with intracerebral hemorrhage and may contribute to progression of hemorrhagic stroke [101]. Although activation occurs before the heart is removed from such donors, endomyocardial biopsies obtained in the recipient at one week after transplantation show a marked increase in expression of both MMP-2 and MMP-9 [101].
Other — A number of other factors may importantly contribute to an increased risk of CAV [18,62]. These include:
●Recipient obesity [62].
●Pretransplantation diagnosis of ischemic compared with nonischemic cardiomyopathy (see 'Coronary disease history' above).
●Longer graft ischemia time (the time between heart explant from the donor and implantation in the recipient) and the development of fibrosis [102].
●Increases in elastase activity [103], thrombospondin-1 (a matrix glycoprotein that inhibits angiogenesis and facilitates the smooth muscle proliferation that is characteristic of CAV) [104], and the expression of tissue factor and the vitronectin receptor [105].
●Hepatitis C virus seropositivity in the donor [106].
●Possibly hyperhomocysteinemia [107,108]. (See "Heart Transplantation: Prevention and treatment of cardiac allograft vasculopathy", section on 'Homocysteine'.)
●Angiogenesis within the intima [109].
●Higher von Willebrand factor levels [110].
SUMMARY
●Cardiac allograft vasculopathy (CAV) appears to be multifactorial in origin with both immunologic and nonimmunologic factors implicated. Among the factors associated with vasculopathy are cellular and antibody-mediated rejection, donor-specific anti-HLA antibodies, cytomegalovirus infection, and hypercholesterolemia. (See 'Pathogenesis' above.)
●Beyond the first year after cardiac transplantation, CAV, also called transplant coronary artery disease or cardiac transplant vasculopathy) is among the top three causes of death. (See 'Introduction' above.)
●CAV is a panarterial disease confined to the allograft and is characterized by diffuse concentric longitudinal intimal hyperplasia in the epicardial coronary arteries (figure 1) and concentric medial disease in the microvasculature. In contrast, traditional atherosclerosis is focal, noncircumferential, and most often observed proximally in the epicardial vessels. (See 'Pathology' above.)
●CAV appears to be multifactorial in origin with both immunologic and nonimmunologic factors implicated. Among the factors associated with vasculopathy are cellular and antibody mediated rejection, donor specific anti-HLA antibodies, cytomegalovirus infection, and hypercholesterolemia. Immunologic events appear to be most important, since CAV develops in the donor's but not the recipient's arteries. (See 'Pathogenesis' above.)