Introduction

Many synonyms, ranging from chronic rejection to transplant atherosclerosis, have been used over the years to identify the progressively stenosing vascular process in transplanted organs that is now most frequently designated as cardiac allograft vasculopathy (CAV; see Table 1) [1–11].

Table 1 Coronary cardiac allograft vasculopathy: synonymous
Full size table

The International Society for Heart Lung Transplantation (ISHLT) has recently published a working formulation of a standardized nomenclature for cardiac allograft vasculopathy-2010, recognizing the necessity of clarifying the abundant confusion in terminology and interpretations of various clinical and imaging technologies. This has resulted in a standard language based on best scientific evidence and clinical practice [12–15]. CAV can be defined as a rapidly progressive form of atherosclerosis characterized by an accelerated progressive and diffuse intimal hyperplastic lesions of the vascular tree that leads to insidious vessel narrowing and eventually to allograft ischemia. Clinical presentation is heterogeneous, varying from silent myocardial ischemia to congestive heart failure and sudden death [16, 17]. There are many aspects that are peculiar to CAV compared to classical native atherosclerotic lesions (ATS). CAV is considered a diffuse concentric process, affecting homogeneously the entire coronary vascular tree, with a rapid onset even within months, a mainly immune-mediated process and at histological evaluation characterized by intimal hyperplasia composed by SMC, intact internal elastic lamina, and various mononuclear cell infiltrations and preserved tunica media (see Table 2) [18, 19]. However, this represents a simplistic approach to the CAV lesion evaluation, since the assessment of the vascular tree of re-transplanted hearts or autopsy post-transplanted heart has revealed a series of highly heterogeneous intimal lesions in terms of component makeup, ranging from early diffuse intimal thickening to complicated atheromas [9, 20–22]. Even in these CAV hearts, we can identify non-culprit and culprit lesions (i.e., the lesions that are responsible for the onset of clinical symptoms). Another important feature which can be noticed not only in the heart but also in other solid organ transplants as kidney [23, 24] is the involvement of intramyocardial arteries and arterioles, the so-called microvasculopathy [25], as well as the venous and post-capillary vascular bed in the process of CAV [26]. The latter can be characterized by venulitis with mononuclear inflammatory cells infiltration of the intimal layer or transmural wall or by intimal hyperplastic cell proliferation with cross-sectional luminal stenosis. This is generally considered trivial in terms of hemodynamic graft performance although systematic studies on this subject are lacking. In the definition of terms, “accelerated” atherosclerosis is usually associated to the term CAV, since the rapidly progressive growth of lesions determines many biological and clinical manifestations of CAV. We all know that ATS is a long-life remodeling process affecting the wall of large- and medium-sized arteries, which starts in childhood and manifests in the advanced adult age, and with males being more precociously involved (about 10/15 years earlier than females). On the other hand, CAV develops much more rapidly, affects males and females in a similar degree and proportion, and not only adults but also pediatric patients become symptomatic early after transplantation even within the first year. This accounts for more than 30 % of involvement at 3 years post-transplant (Table 2). The attention has been focused on risk factors both for ATS and CAV in a clear attempt to adopt preventive strategies to reduce morbidity and mortality related to these diseases. However, ATS still represents the main cause of cardiovascular death in the general population, and CAV still represents the second most frequent cause of death among transplanted patients in the long-term follow-up [27]. This is true also when we disaggregate the mortality data considering different era in the follow-up [13].

Table 2 Key points
Full size table

The heterogeneity of pathological features

The heterogeneity in clinical symptoms from silent ischemia to severe congestive heart failure and even sudden death is paralleled by heterogeneity in the pathological features of CAV lesions. This varies from what is considered the classical CAV fibrocellular lesion characterized by concentric diffuse intimal thickening of epicardial and intramyocardial coronary arteries with preserved tunica media to classical ATS-like lesions characterized by atheromas with focal, eccentric plaques, and a lipid pool (fibrolipid plaques). Inflammatory patterns may range from transmural vasculitis involving epicardial and intramyocardial vessels to a more selective superficial involvement with endothelialitis (Fig. 1) [18–22]. Fibrocellular lesions are present in about two-third of CAV cases in the first years after transplantation, but their incidence drops markedly on the long term when fibrolipid plaques are more frequently detected (Figs. 2, 3, and 4) frequently associated with atrophy or attenuation of the media in a similar pattern as in native ATS [16]. Challenging questions are related to the presence of these fibrolipid plaques in CAV: Are the fibrolipid plaques in CAV pre-existent plaques in the coronary arteries of the donor, or do they arise the novo, superimposed on the classical fibrocellular CAV lesions? And even more: is the rich SMC donor intimal thickening, which is insignificant from a clinical point of view, a predisposing substrate (a “fertile soil”) for CAV development? Many of these questions remain to be elucidated, which could explain why CAV still is the Achilles’ heel of heart transplantation outcomes [16].

Fig. 1
figure 1

Histopathologic characteristic of CAV and ATS. Similarities and differences in phenotypes and biological mechanisms of cardiac allograft vasculopathy (CAV) and atherosclerosis (ATS). This figure reports only the main characteristics of both types of lesions

Full size image
Fig. 2
figure 2

Coronary allograft vasculopathy: pathological features. View of concentric diffuse intimal thickening of epicardial coronary artery tree in a gross pathology; b concentric lesion, with focal attenuation of tunica media; and c eccentric fibrocellular lesion; d, e concentric diffuse intimal thickening of the small intramyocardial arteries, with focal attenuation of tunica media in (d)

Full size image
Fig. 3
figure 3

Classic atherosclerosis lesion. a Azan-mallory histo-staining of a classic ATS lesion, showing focal eccentric plaque with thin fibrous cap covering a lipid pool, with atrophy of tunica media; b detailed vision of black inset in (a), showing numerous cholesterol clefts in the lipid pool, with atrophy of tunica media

Full size image
Fig. 4
figure 4

Coronary allograft vasculopathy: pathological features a epicardial artery showing vasculitis. Power view in the black inset; b H&E showing both the artery (A) and vein (V) affected by vasculitis; c, d endothelialitis of small coronary arteries. All sections H&E stained

Full size image

Destabilizing factors in plaque tissue of CAV and ATS

In recent years, the attention has been directed to evaluate the microenvironment in atherosclerotic plaque in order to specifically address the issue of inflammation and neoangiogenesis and its role in ATS development and progression and to identify potential substrate of plaque instability [28–31]. Kockx et al. and Virmani et al. showed that intraplaque hemorrhages (IPH) can be seen as a possible plaque destabilizing factor, since hemorrhages were more frequently identified in the vulnerable thin-cap fibroatheromas and in advanced lesions [28, 32, 33]. In addition, IPH occurred in areas of extensive neovascularization, where dilated, very-thin-walled, and leaky microvessels are located, as can be visualized by diffuse perivascular staining for von Willebrand factor [28, 34].

We recently had the opportunity to evaluate the coronary plaques of cardiac transplanted patients who died because of CAV and to compare them with the coronary plaques in their native hearts which were removed at the time of transplantation (i.e., in the same patients) [34]. The results of our study are reported below.

Inflammatory patterns in CAV and ATS

Immunohistochemical studies have characterized the inflammatory profile of lesions and showed that CAV fibrocellular lesions display a pattern of inflammation similar to the inflamed myocardium during an acute cellular rejection. In this type of inflammatory response, CD3+ T cells dominate the infiltrate, accounting for about 70–75 % of the inflammatory cells, CD20+ B cells for about 10 %, and CD68+ macrophages for about 15 % [35]. By contrast, macrophages are the dominant cell type in ATS plaques accounting for more than 55 % of the plaque cells, CD3+ cells as the second important cell type (40–45 %), but almost no B-lymphocytes [36]. Our attempts to characterize the inflammatory cells subtypes in CAV showed similarly that fibrocellular plaques of CAV presented overall low numbers of macrophages and more lymphocytes compared to ATS, but that fibrolipid plaques in CAV and in ATS had the same inflammatory patterns (Table 2).

Plaque microvessels and intraplaque hemorrhages in CAV

Overall, we investigated 24 hearts of 12 patients, evaluating 70 plaques of which 35 lesions were taken from native hearts and 35 from the transplanted heart at autopsy [34]. Microvessels, which are an important feature of the advanced fibro lipid plaques in ATS, were detected in our study in 74.3 % of the 35 CAV lesions and showed perivascular leakage of von Willebrand factor as sign of endothelial damage. In these CAV lesions, presence of microvessels was strongly associated with IPH and plaque inflammatory cells.

These intraplaque hemorrhages were even more frequently encountered in CAV lesions than in ATS, (60 % of plaques versus 20 % of plaques respectively, p = 0.0016). Not only fresh hemorrhages but also older (lytic or organized) hemorrhages could be detected frequently in the same plaque (40 % of plaques were positive both for fresh and old hemorrhages) (Fig. 5). This implies that, similar to the situation in advanced native plaques, plaque hemorrhage appears to be an ongoing phenomenon and hence may contribute significantly to lesion progression and/or degree of lumen stenosis. Interestingly, these hemorrhages were not only found in fibrolipid plaques, as one would expect, but also in the “typical” fibrocellular lesions of CAV (Fig. 3) [34].

Fig. 5
figure 5

Intraplaque hemohrrage (IPH) in CAV and ATS. a The graph reports the distribution of incidence of IPH in CAV and ATS plaques and divided by different types of lesions as: early, fibrocellular, and fibrolipidic lesions. Note as in both CAV and ATS early lesions, no IPH was identified. CAV shows more incidence of IPH in both fibrocellular and fibrolipidic lesion compared to ATS (p < 0.01); b,c,d immunohistology (Gycophorin A, panels b and c) and iron stain (panel d) of plaque hemorrhage

Full size image

The remodeling process of CAV is time dependent, and one might expect a dramatic reduction in the numbers of fibrocellular plaques over the years concomitant with their transition into fibroatheromas, which is due to the exposure to atherosclerotic risk factors such as hypercholesterolemia or immunologic risk factors. However, in the long term, we observed that about half of the plaques still were fibrocellular in nature. Moreover, one would expect that in post-transplant patients, the plaques would have been more inflamed, but instead, we detected lack of inflammation in some of CAV plaques several years after transplantation. These differences are probably also risk factors related (Table 2).

Adventitia

Since the adventitia of arteries is supposed to be actively involved in the process of tissue remodeling of atherosclerosis affected arteries, we have recently further explored patterns of inflammation, microvessels, and leakage in the adventitia and perivascular tissues of CAV versus ATS. Lesions containing plaque hemorrhages had more adventitial inflammation in CAV than in ATS, but the abundant infiltrations diminished on the long term (post-transplant intervals of several years).

CAV lesions showed more adventitial inflammation and particularly more fibrosis compared to ATS. Inflammation followed a pattern of diffusely distributed inflammatory cells which was present in nearly all lesions early after transplantation, but in the late observations (years after transplantation), some plaques were completely devoid of adventitial inflammation (Table 2).

Distinct perivascular fibrosis appears to be an early complication of CAV-affected arteries, which may contribute importantly to remodeling phenomena, particularly in terms of shrinkage. The presence of adventitial fibrosis and microvessels seemed not to be influenced by time [37, 38].

Inflammation in ATS and CAV and in vivo imaging

ATS is recognized as an inflammatory disease of the arteries; although, the inflammatory component in the plaques is usually present multifocally instead of diffusely [39–41]. Inflammation is recognized to be a destabilizing factor in the natural history of the plaque and eventually may lead to plaque disruption complicated by thrombosis [41]. CAV also has a strong inflammatory component but usually in relation to the concomitant presence of acute cellular rejection in the myocardium. The role of ACR in CAV development and progression is still controversial. Some authors reported a strong association with CAV progression and worse outcomes [17], while others had been unable to obtain similar results [42]. High rejection score was an independent predictor for CAV onset suggesting an immune basis for CAV onset and nonimmune modulation for progression [17]. Moreover, so-called endothelialitis has now been recognized as a common feature of the microvascular inflammatory burden associated to antibody-mediated rejection (AMR). Symptomatic antibody-mediated rejection has been recognized as a risk factor for development of CAV and worse outcome [43–45]. Recently, Wu et al. demonstrated that also asymptomatic AMR is associated with development of CAV at 5 years follow-up [46]. Evidence of strong correlation between complement activation (C4d and C3d) on EMBs and development of CAV exists [47, 48], supporting the concept of donor specific antibodies damage to endothelium and activation of m-TOR proliferative pathway [49–51].

The production of antibodies to donor HLA antigens before or after cardiac transplantation is a major risk factor for the development of AMR. The pathological effect of DSA binding to the transplanted organ is likely to involve signaling pathways. Engagement of HLA class I molecules by anti-HLA antibodies increased the activation of ERK1-2, S6K, and S6RP through target of rapamycin (mTOR) complexes 1 and 2 (mTORC1 and mTORC2) [52]. Recent multicenter trials have evaluated the outcomes of immunosuppressive therapies with the mTOR-derived antiproliferative signal inhibitors such as everolimus [53]. These studies were performed using in vivo imaging techniques evaluation of CAV by means of virtual histology (VH). These techniques have previously been used also for identification of ATS plaque characteristics in an attempt to identify plaques at risk of complications such as plaque rupture. They have been used also in CAV assessment, among patients treated with everolimus and low-dose calcineurin inhibitor (CNI) compared to standard CNI therapy [54]. Among these techniques, VH represents a novel technological tool that utilizes backscatter radiofrequency data obtained during IVUS (intra-vascular ultrasound) pullback to construct a tissue map providing a qualitative assessment of plaque and intimal wall composition. It has been demonstrated that VH possesses 94–97 % ex vivo and 87–97 % in vivo accuracy for characterization of four basic tissue components (fibrous, fibrofatty, calcified, and necrotic core components) among patients with ischemic heart disease [55]. Moreover, another technique—namely, contrast ultrasound imaging and dynamic contrast-enhanced MR (DCE-MR)—was applied by Dong et al. group to study also the vasa vasorum in the adventitia of the plaques [56]. DCR-MR allows the assessment of neovascular architecture and functional characteristics, such as fractional plasma volume and permeability, also providing high spatial resolution that allows localization of the measurements, including the adventitial boundary and the source of the vasa vasorum [56].

Thanks to application of virtual histology in CAV assessment of patients treated with everolimus and low-dose calcineurin inhibitor (CNI), the investigators were able to identify four morphologic components of plaques—namely, fibrous tissue, fibro fatty tissue, calcifications, and necrotic cores, in which the last two have been considered as surrogates of inflammation into the plaque. Moreover, the results showed that the progression of CAV in the treated group was characterized by a statistically significant increase in the calcified and necrotic components. These gave new insights in the detailed morphology of CAV lesions; although, we question the interpretation given by the authors to consider calcified and necrotic components as surrogate markers for inflammation in plaques [57–59]. Nowadays, imaging techniques are still unable to detect reliably sites of inflammation in plaques (Table 2) [60].

Conclusions

Cardiac allograft vasculopathy can be considered a particular type of atherosclerosis with many similarities but also significant differences compared to ATS, which makes it difficult for pathologists to systematically classify the lesions and to elucidate their origins. With donation, the patients receive a heart, which already has its own history of coronary atherosclerosis and in which all the phenotypes of ATS lesions are potentially already present. The immunological/inflammatory- and microvessel-derived hemorrhagic responses over time after transplantation cause coronary arterial pathology that is characterized by the onset of de novo lesions but also remodeling of already existing lesions (Fig. 5). Detailed knowledge of the entire spectrum of plaque types in CAV is not only important for understanding the pathogenesis of disease but also clinically for the interpretation of in vivo imaging techniques.