http://www.jhltonline.org ORIGINAL CLINICAL SCIENCE Endothelitis in cardiac allograft biopsy specimens: Possible relationship to antibody-mediated rejection Fabio Tavora, MD, a Raghava Munivenkatappa, MD, b John Papadimitriou, MD, b Cinthia Drachenberg, MD, b Charles Sailey, MD, b Mandeep Mehra, MD, b and Allen Burke, MD b From a Messejana Heart and Lung Hospital, Fortaleza Brazil; the b University of Maryland Medical Center, Baltimore, Maryland. KEYWORDS: heart transplantation; antibody mediated rejection; endothelitis; vasculitis; complement BACKGROUND: The incidence of endothelitis in cardiac transplants, and the relationship to clinical symptoms and humoral rejection, is unclear. Recently, the finding of intravascular macrophages has been found to represent antibody-mediated rejection. This study investigated the role of intravascular T lymphocytes in antibody-mediated rejection. METHODS: A total of 819 sequential biopsy specimens from 93 cardiac allograft recipients were prospectively studied. Rejection was graded according to International Society for Heart and Lung Transplantation (ISHLT) criteria and inflammatory infiltrates characterized by immunohistochemical staining for CD3, CD4, CD8, CD68, and CD20. Endothelitis was defined as lymphocyte and macrophage infiltrates within arteriolar, capillary, or venular walls, with endothelial swelling, in contrast to perivascular inflammation of cellular rejection. Complement C4d was identified in capillary walls by immunofluorescent staining and immunohistochemical staining on paraffin sections. RESULTS: Endothelitis was identified in 27 specimens (3%) from 14 patients (15%). ISHLT rejection grades were 0 in 6 specimens, 1R in 20 (1A in 8; 1B in 12), and 2R (3A) in 1. In all cases, there were admixtures of macrophages and T lymphocytes. Inflammation was most prominent in venules. C4d was localized in 12 of the 27 specimens (44%). C4d was localized in 31 of 796 specimens without endothelitis (p 0.001). The endothelial infiltrates were CD3, CD4, CD8, and CD68. Twelve of 14 patients had 0 panel reactive antibodies (PRA), 9 were above 10%, and 8 were above 25%; 5 patients were treated for clinical antibody-mediated rejection, and 4 had possible cardiac allograft vasculopathy by ultrasound imaging (mean follow-up, 40 ). CONCLUSION: Endothelitis is present in more than 10% of heart transplant recipients and is associated with complement deposition on biopsy samples. Approximately 33% of patients have clinical evidence of humoral rejection. The eventual risk for developing graft vascular disease remains undetermined. J Heart Lung Transplant 2011;xx:xxx 2011 International Society for Heart and Lung Transplantation. All rights reserved. Antibody-mediated rejection (AMR) results from complement-mediated endothelial damage in the allograft. Initial AMR was considered to be essentially devoid of inflammation characteristic of cellular rejection. However, the intraluminal macrophages and neutrophils are currently recognized histologic features. Nevertheless, the diagnostic criteria for antibody AMR varies from center to center, and Reprint requests: Allen Burke, MD, University of Maryland Medical Center, 22 S Greene St, Baltimore, MD 200910. Telephone: 1-410-328-1346. E-mail address: allen.burke@gmail.com 1053-2498/$ -see front matter 2011 International Society for Heart and Lung Transplantation. All rights reserved. doi:10.1016/j.healun.2010.10.009 the incidence differs according to patient population and methods of diagnosis, which depend on clinical, serologic, and immunohistochemical findings. Despite the revision of the International Society for Heart and Lung Transplantation (ISHLT) panel, there is no consensus on histologic features and immunohistochemical findings of complement deposition that define AMR. Endothelitis is defined as intimal inflammation in muscular arteries (arteritis), arterioles, and capillaries (capillaritis), and is diagnosed in heart transplant biopsy specimens when there is arteriolar and capillary inflammation.
2 The Journal of Heart and Lung Transplantation, Vol xx, No x, Month 2011 Table 1 Clinical Characteristics of Patients With Endothelitis Patient Age at transplant (years) Sex Indication LVAD Immunosuppression 1 20 F NICM Yes Tac, MMF, Pred 2 22 F Post-partum Yes Tac, MMF, Pred 3 22 F Post-partum No Tac, MMF, Pred 4 27 M NICM No Tac, MMF, Pred 5 47 M NICM Yes Tac, MMF, Pred 6 50 F NICM No Tac, MMF, Pred 7 51 M NICM Yes Tac, MMF, Pred 8 60 M Ischemic Yes Cyc, Pred, MMF 9 60 M Ischemic No Tac, MMF, Pred 10 62 M Ischemic Yes Tac, MMF, Pred 11 63 M Ischemic No Tac, Sir, Pred 12 64 M Ischemic Yes Tac, MMF, Pred 13 65 M NICM (valvular, rheumatic) No Cyc, Pred, MMF 14 67 M Ischemic No Cyc, Pred, MMF F, female; LVAD, left ventricular assist device; M, male; MMF, mycophenolate mofetil; NICM, non-ischemic cardiomyopathy; Pred, prednisone; Sir, sirolimus; Tac, tacrolimus. Autopsy studies long ago described necrotizing vasculitis associated with intimal thickening and complement deposition. 1 Anecdotal evidence suggests that vasculitis involves the entire coronary tree, from epicardial to small intramural branches. 2 Also termed intimitis, especially in renal transplants, 3 endothelitis has been described in epicardial arteries in autopsies of heart allografts as a possible precursor lesion to cardiac allograft vasculopathy (CAV). Intimitis or endothelitis in the cardiac transplants has been associated with cytomegalovirus infection 4 and the development of CAV based on autopsy studies. 5,6 Figure 1 Endothelitis is shown in endomyocardial biopsy specimens of a small arteriole. (A) Hematoxylin-eosin stained section of an arteriole with chronic inflammation within the lumen and wall of the vessel. (B) CD4 immunohistochemical staining shows T lymphocytes infiltrating the wall of the arteriole. (C) CD8 immunohistochemical staining shows T lymphocytes infiltrating the wall of the arteriole. Original magnification 400.
Tavora et al. Endothelitis and Antibody Mediated Rejection 3 Although perivascular inflammation is a hallmark of low-grade cellular rejection, intimal inflammation and endothelial reaction is not mentioned as a component of lowgrade cellular rejection. 7 Vasculitis is considered only a component of severe rejection with myocyte necrosis. The coexistence of cellular and AMR has, however, been well documented. In the revised ISHLT criteria for biopsy specimen diagnosis, intravascular macrophages have been identified as a typical feature of AMR, as well as vascular neutrophils. 7 Vasculitis, including lymphocytic inflammation, is not considered in these revised criteria as a separate diagnosis or one associated with AMR. The purpose of the current study was to provide a workable definition of endothelitis (intimitis) on heart biopsy specimens in patients without high-grade rejection, and determine the incidence and clinical significance, especially related to AMR, and histologic evidence of complement deposition. Material and methods All heart allograft biopsy specimens at the University of Maryland Medical Center were prospectively entered into a database during a 4-year period from 2004 to 2008. Cases with 2 biopsy fragments with myocardium were retained for study. Institutional Investigational Review Board approval was obtained for the research. Recorded data at the time of biopsy were grade of rejection, presence of endothelitis, Quilty effect, and prior biopsy site. Before 2006, cellular rejection was graded according to the 1990 working formulation of the ISHLT, and after 2006, by the older grading system as well as the 2005 revision. 7 Quilty was defined as the presence of discrete aggregate of endothelial T lymphocytes, with or without central B cells. Tangentially sectioned subendocardial infiltrates were diagnosed as cellular rejection vs Quilty by CD20 infiltrates, appearance on different levels, and consensus agreement by 2 of the study pathologists (A.P.B., C.I.D.). Immunohistochemical stains were done by Ventana automated immunostainer. C4d antibody was commercially obtained from Thermo Scientific, Waltham, Massachusetts. A dilution of 1:200 was used, with staining on an automated Ventana Immunostainer. C4d positivity was defined as 50% of capillaries with distinct endothelial staining. Each specimen was stained at 3 levels for hematoxylineosin, with additional levels for immunostains. Endothelial markers CD31, CD34, and factor VIII-related antigen were assessed by immunohistochemistry on those cases assessed as endothelitis or possible endothelitis. In addition, CD3, CD4, and CD8 immunohistochemical stains were performed. Endothelitis was defined in distinction to perivascular cellular inflammation by intravascular or capillary inflammation with endothelial cell reaction, characterized by nuclear enlargement. Endothelial markers were used to identify inflammatory cells within the lumen and wall of the vessel. The distinction of interstitial cellular rejection (old 1A, or 1R) was based on intravascular inflammation or sub-endothelial infiltration with endothelial cell swelling. Clinical information obtained in patients diagnosed with endotheliitis included duration of transplant, graft function, presence of panel reactive antibodies, and presence of donor-specific antibodies. Follow-up data included development of subsequent rejection or AMR, CAV, graft failure, or death. Duration of transplant was recorded in all biopsy specimens as well as indication for biopsy (surveillance vs suspicion of rejection). Results There were 839 biopsies in 93 patients in the study period that met the inclusion criteria. The 93 patients comprised 71 men (age, 57 13 years) and 22 women (age, 42 16 years; p 0.0001 vs men) at time of transplant. The mean numbers of biopsies during the study period per patient was 8 4 for men, and 11 4 for women (p 0.02 vs. men). The mean duration of transplant at the last biopsy was 26 32 for men and 18 11 for women. Of the 93 patients, 3 of 22 women and 22 of 71 men showed no evidence of rejection on any biopsy specimen (p 0.1, men v. women), 14 of 22 women and 39 of 71 men showed evidence of Quilty on 1 specimens (p 0.5, men vs women), and 14 (4 women, 10 men) showed evidence of endothelitis on 1 biopsy, for a total of 27 biopsies. Their ages at time of transplant ranged from 20 to 67 years (Table 1). The indication for transplant and presence of prior LVAD are presented in Table 1. Most patients were on an immunosuppressive regimen of tacrolimus, mycophenolate mofetil (MMF), and prednisone (Table 1). Twelve of the 14 had evidence of alloantibodies (Table 1), 10 10% and 9 25%. Nine of the 14 had donor-specific antibodies, including 5 with major histocompatibility complex class 2 (Table 1). The histologic features of endothelitis were primarily intravascular macrophages in 3, with an admixture of T cells and macrophages in the remaining 24. The lymphocytes expressed CD3 by immunohistochemical staining, with a predominance of CD4-positive T cells in each of the 24 cases (Figures 1 and 2). CD31, CD34, and factor VIII immunohistochemical stains demonstrated intraluminal inflammatory cells and occasionally inflammatory cells that were surrounded by endothelial cell cytoplasm (Figure 3). There was a sub-population of CD8- positive cells in 21 of the 24 cases with intravascular T cells (Figures 2 and 4). The vessels involved were small arterioles (Figure 1), venules (Figures 2 5), and capillaries (Figure 5). The media of vessels was often not intact due to infiltration of the vessel cross-section by inflammation, rendering the distinction between small arteriole and capillary difficult in some cases. In most specimens there was a mixture of inflammatory cells, including macrophages and lymphocytes, within the vascular lumens in a given specimen, although 1 patient had primarily lymphocytic endothelitis in 1 biopsy specimen and macrophages in a subsequent biopsy (Figure 5). In general, the larger vessels were more likely to have a predominant lymphocytic infiltration, with capillaries more predominantly macrophages, although even capillaries had a significant proportion of lymphocytes in most cases (Figure 4). C4d staining was present in 10 of the patients with specimens with endothelitis and in 12 of the specimens with endothelitis (Table 2). There was no significant correlation with serologic results of panel reactive antibodies or donorspecific antibodies (Table 2). Even in the biopsy specimens with predominant intravascular macrophages, there was a lymphocytic endothelitic component (Figure 5) as well as perivascular inflammation
4 The Journal of Heart and Lung Transplantation, Vol xx, No x, Month 2011 Figure 2 Endothelitis is shown in endomyocardial biopsy specimens of venules. (A) Hematoxylin-eosin stained section shows marked expansion of the vessel by chronic inflammation obscuring the lumen and wall of the vessel. Occasional red blood cells are noted in the lumen. (B) CD4 immunohistochemical staining shows T lymphocytes infiltrating the vessel wall. (C) CD68 immunohistochemical staining shows a minor subpopulation of macrophages in the infiltrate. CD8 immunohistochemical staining shows T lymphocytes infiltrating the vessel walls. Original magnification 400. Figure 3 Endothelitis is shown in venules. (A) Immunohistochemical staining for CD34 of the vessel in the specimen shown in Figure 2 demonstrates endothelial cells in the intima surrounded by chronic inflammation but within the lumen as well as surrounding the vessel. Original magnification 200. (B) Immunohistochemical staining for CD31 of vessel shown in Figure 2 demonstrates endothelial cells in the intima surrounded by chronic inflammation, but within the lumen as well as surrounding the vessel. Original magnification 200. (C) Immunohistochemical staining for factor VIII-related antigen demonstrates endothelial cells in the intima with chronic inflammatory cells (not staining) on the intimal luminal surface. Original magnification 400.
Tavora et al. Endothelitis and Antibody Mediated Rejection 5 Figure 4 Lymphocytic vasculitis, with C4d deposition, is shown in capillaries and venules. (A, original magnification 200, B, original magnification 400 ) Hematoxylin and eosin staining demonstrates lymphocytes not only in a perivascular distribution, as seen in mild rejection, but also within venular walls. (C) CD8 immunohistochemical stain has identified T lymphocytes within the vessel. Original magnification 400. (D) C4d immunohistochemical staining is shown for endothelial complement. Original magnification 200. (Figures 6 and 7). This perivascular infiltrate was considered a component of cellular rejection if extensive. Patients with endothelitis on biopsy specimens were a mean 7.2 9.1 years (range 2 weeks 5 years) after transplant vs 11.8 26.2 years for others (p 0.3 Wilcoxon rank sum test for non-parametric data). The number of specimens showing endothelitis in patients with 1 positive biopsy specimen ranged from 1 in 16 (6%) to 3 in 6 specimens (50%). Thirteen of the 14 patients with endothelitis on any specimen demonstrated rejection on any biopsy, and of the individual 23 biopsy specimens in these patients, 20 of 23 showed cellular rejection. Nine of these were assigned ISHLT grade 1R, 1B; and the remaining 14 showed grade 1R, 1A. Eight of 12 patients with endothelitis on any biopsy specimen demonstrated C4d positivity in an endothelial pattern; in all but 1 patient, C4d was seen only in specimens with endothelitis. Of the 27 biopsy specimens with endothelitis, 12 (44%) showed endothelial staining for C4d. On follow-up (40 9 ), 4 patients had possible CAV, one with symptomatic CAV (Table 3). One patient died of infection and sepsis 2.5 years after transplant. The 3 patients with biopsy specimens showing primarily intravascular macrophages (Patients 1, 10, and 14, Table 2) all showed clinical evidence of AMR, 1 progressing to CAV as evidenced by intravascular ultrasound, and 1 with possible CAV. In 2 of the patients, there were other specimens with predominantly lymphocytic endothelitis. Compared with patients without endothelitis on any biopsy specimen, the duration of transplant at the last biopsy was less long and the incidence of rejection was higher, at borderline significance (Table 4). Discussion The current study shows that the rate of endothelitis is 15% for patients and 3% for heart biopsy specimens. The higher rate by patient indicates that recurrence is typical, because in the 12 patients, there was a range of 6% to 50% of the proportion of biopsy specimens with endothelitis. There was a high association with cellular rejection, especially low grade (particular
6 The Journal of Heart and Lung Transplantation, Vol xx, No x, Month 2011 Figure 5 (A) Capillaries and small arterioles are expanded and filled with CD68-positive macrophages with (B) diffuse endothelial staining for C4d. The patient had a history of antibody-mediated rejection and evidence of lymphocytic vasculitis on biopsy specimens 2 years prior. (C) The routinely stained section shows mild interstitial inflammation, with (D) a population of T lymphocytes around capillaries (anti-cd3 avidin-biotin technique). Original magnification 400. Table 2 Immunohistochemical and Serologic Data for Patients With Endothelitis PRA, % DSA Patient C4d IHC endothelitis (other biopsies) C4d MHC 1 MHC 2 Class 1/class 2 1 1/2 endothelitis (0/12 other) 0 0 No 2 1/1 endothelitis (0/15 other) 36 3 No 3 0 40 29 Yes (both) 4 0 26 0 Yes (1) 5 2/5 endothelitis (0/10 other) 9 40 Yes (2) 6 2/3 endothelitis (2/7 other) 41 2 Yes (1) 7 1/1 endothelitis (0/14 other) 86 11 Yes (both) 8 1/1 endothelitis (0/15 other) 6 0 Yes (2) 9 1/2 endothelitis (0/8 other) 17 0 Yes (both) 10 0 28 17 Yes (1) 11 1/3 endothelitis (0/3 other) 5 0 No 12 0 33 0 No 13 1/1 endothelitis (0/9 other) 0 6 No 14 1/3 endothelitis (0/3 other) 38 97 Yes (2) IHC, immunohistochemistry; MHC, major histocompatibility complex; PRA, panel reactive antibody; DSA, donor-specific antibodies.
Tavora et al. Endothelitis and Antibody Mediated Rejection 7 Figure 6 Perivascular inflammation associated with endothelitis in a patient with clinical antibody-mediated rejection. (A) There is extensive perivascular inflammation with macrophages (CD68 stain). Original magnification 200. (B) A higher magnification of a different area shows intra-arteriolar macrophages (arrows). Original magnification, 400. (C) CD3 immunohistochemical stain shows diffuse interstitial T lymphocytes associated with capillaries in a pattern of 1B rejection. Original magnification, 200. (D) A venule from the same specimen shows diffuse infiltration of the vessel wall by macrophages (CD68; immunohistochemistry, avidin-biotin technique). Original magnification, 400. Figure 7 Perivascular inflammation associated with endothelitis. (A) A low magnification image stained for CD3 T lymphocytes shows a pattern similar to 1B rejection. Original magnification, 100. (B) Scattered vessels demonstrated intense infiltration with T lymphocytes. Original magnification 200. (C) Immunohistochemical staining for CD68 shows clear intravascular macrophages. Original magnification 200.
8 The Journal of Heart and Lung Transplantation, Vol xx, No x, Month 2011 Table 3 Follow-up Data for Patients With Endothelitis Patient Endothelitis biopsies/total biopsies Time post-tx endothelitis biopsies Range of cellular rejection (biopsies/ total) History of AMR or CAV IVUS Follow-up. 1 2/14 primarily macrophages 2 week 1 month ISHLT 1 and 2, 8/14 Possible CAV Possible mild vessel narrowing Lost to follow-up at 13 2 1/16 2 All ISHLT 1, 8/16 No Mild distal vessel narrowing Died 2 years 6 3 1/15 2 weeks 14 Multiple episodes AMR ISHLT 1 and 2, 6/15, same as mild Normal IVUS; significant biventricular dysfunction 4 2/5 2 weeks ISHLT 1 and 2, 3/5 No IVUS, normal dobutamine stress echo Rejection 3B, endothelitis, AMR 4 years 4 OK 24 5 5/15 3 9 All ISHLT 1, 8/15 No None OK 4 years 1 month 6 3/10 2 weeks 3 ISHLT 1, 1/10, AMR with RV No OK 7 different biopsy dysfunction 7 1/15 2 ISHLT 1, 1/14 (different biopsy) AMR, post- transplant plasmapheresis Minimal intimal hyperplasia; normal ejection fraction OK 3 years 8 recent normal biopsies 8 1/16 10 ISHLT 1 (endothelitis) and 2, 2/16 9 2/10 2 ISHLT 1 and 2, 8/10 Episodes of cellular rejection (noncompliance) No No OK 2 years 6 No OK 6 10 1/3 primarily 1 month ISHLT 1 (same B-cell cross match; No OK at 14 macrophages biopsy) severe non-hla AMR 11 3/6 10 21 All ISHLT 1, 4/6 No Normal IVUS Doing well, 7 years 1 month 12 1/3 2 0 No No Persistent endothelitis, 13 13 1/10 12 All ISHLT 1, 6/10 No None OK 7 years 5 14 3/6 primarily macrophages in 1 biopsy 24 60 ISHLT 1, 2/6 Multiple episodes AMR; CAV with severe LV dysfunction CAV AMR clinically 11 years 3, with endothelitis AMR, antibody-mediated rejection; CAV, cardiac allograft vasculopathy; HLA, human leukocyte antigen; ISHLT, International Society for Heart and Lung Transplantation; IVUS, intravascular ultrasound; LV, left ventricular; RV, right ventricular. previous grade 1B). The inflammation within vessels that defined endothelitis was often associated with perivascular inflammation, explaining at least in part the association with cellular rejection. Furthermore, old 1B rejection, as defined by diffuse increased inflammation without discrete infiltrates, may be an overlap condition with endothelitis. The relationship between cellular rejection, endothelitis, and AMR is not clear. The current study defined endothelitis as separate from perivascular infiltrates of mild rejection by involvement of the vessel wall, namely, subintimal inflammation, luminal inflammation, and endothelial reaction. None of the patients had severe rejection, which is associated with vascular necrosis. Of interest, the current study showed that the nature of the inflammatory infiltrates was most typically lymphocytic, as opposed to macrophage cells, which have been proposed as a criterion, when intraluminal, for AMR. 7 The current study showed an association between C4d capillary deposition and endothelitis, because there was a significantly increased rate of C4d-positive reaction in the endothelitis specimens compared with others, and there was a clear association, among patients with endothelitis, between episodes of vasculitis and complement deposition. However, almost 33% of biopsy specimens with endothelitis showed no complement deposition. It is unknown if endothelitis with complement represents a non-immune mechanism of endothelial inflammation, whether other forms of complement, such as C3d are present, or if there is a temporal difference between
Tavora et al. Endothelitis and Antibody Mediated Rejection 9 Table 4 Comparison of Clinical Features of Patients With and Without Endothelitis on any Biopsy Specimen No Endothelitis endothelitis Feature a (n 14) (n 79) p-value Age at last biopsy, 49 18 54 14 0.3 years Sex, No. 0.6 Male 10 61 Female 4 18 Biopsies, No. 10.0 4.8 8.7 4.5 0.3 Duration of transplant, 16 12 25 32 0.06 Incidence on any biopsy, No. Rejection 13/14 55/79.04 Quilty 8/14 45/79 0.9 C4d staining 10/14 7/79.0001 a Continuous data are shown as the mean standard deviation. inflammation and complement deposition in some patients. Using histologic findings for the diagnosis of AMR has some limitations. The current criteria suggest that intravascular macrophages, endothelial swelling, neutrophils, hemorrhage, and edema are evidence for AMR, as well as C4d deposition. 7 In the current study, endothelitis was characterized primarily by macrophage and lymphocytic infiltrates and endothelial swelling. The latter finding is less specific and reproducible than the finding of intimal inflammatory cells, with the use of endothelial markers. The known problems with standardization of complement deposition include the antibody used, fixation, immunohistochemical technique, and use of C4d vs C3d. In the current study, no case showed neutrophils, hemorrhage, or edema, features that are often associated with severe cellular rejection. 8,9 It is unclear why neutrophils were not seen in any case of endothelialitis in the current study, which is a common feature of capillaritis in the kidney. 10 In renal biopsies, a quantitative scoring method is used for C4d deposition as well as capillaritis. 10 The method is based on the proportion of peritubular capillaries with inflammation, the overall number of luminal inflammatory cells, and the proportion of capillaries staining positive ( 50% generally considered positive). The current study documented a strong correlation between capillaritis and C4d deposition, as has been shown in the kidney. A scoring pattern similar to that in the kidney may eventually be applicable to the heart. It is unclear if both lymphocytic endothelitis and intravascular macrophages are different stages of AMR or separate processes. Because of overlap of both findings in the same biopsy specimens as well as in different specimens of the same patients, the current study suggests that they are both manifestations of AMR. Interestingly, intracapillary macrophages were seen in patients with clinical AMR, indicating the possibility that this represents a later phase in endothelitis. Similar to allograft rejection in kidneys, it may be useful to grade AMR in heart biopsy specimens by the degree of vascular inflammation, beginning with scattered intimal lymphocytes, progressing to diffuse lymphocytic endothelitis, and culminating in intimitis with intravascular macrophages. Because endothelitis often has perivascular inflammation, concomitant mild rejection (grade 1R) is commonly seen in association, as the current study shows. The diagnosis of endothelitis rests on intravascular and subintimal lymphocytes and macrophages with an endothelial reaction, features absent in cellular rejection. Diffuse inflammation within capillary vessels, without significant perivascular inflammation, was present in some of the cases presented in the current study and has morphologic overlap with previously designated low-grade cellular rejection 1B. Clinical evidence of AMR in the current study was present in 5 of 14 patients, and serologic evidence (PRA 10%) in 9 of 14 patients. These data suggest that endothelitis is associated with the development of graft dysfunction through an antibody-mediated mechanism. The relationship between endothelitis and epicardial coronary disease is less clear. However, vasculitis as a component of cardiac allograft rejection has been described for more than 10 years. Berglin et al 11 described 5 of 151 patients with lymphocytic vasculitis on heart biopsy specimens, which was associated with severe graft dysfunction and response to plasmapheresis. The current study supports the concept that lymphocytic infiltration of small vessels in heart biopsy specimens may be a component of humoral rejection or AMR. 11 The presence at autopsy of vasculitis has been associated with cyclosporine and to the subsequent development of graft vasculopathy. 5,12 In the current study, arterioles, venules, and capillaries were involved, generally in the same biopsy specimen. In addition, an overlap was found between primarily macrophage and lymphocytic infiltration of vessels in the same specimen as well as temporally among biopsy specimens in a given patient. In conclusion, intimal inflammation in cardiac allograft biopsy specimens comprises not only macrophages but T lymphocytes as well and may be suggestive of AMR and may occur at a relatively frequent rate. The relationship between the future development of CAV needs further study. Disclosure statement None of the authors has a financial relationship with a commercial entity that has an interest in the subject of the presented manuscript or other conflicts of interest to disclose. References 1. Foerster A. Vascular rejection in cardiac transplantation. A morphological study of 25 human cardiac allografts. APMIS 1992;100:367-76.
10 The Journal of Heart and Lung Transplantation, Vol xx, No x, Month 2011 2. Cuda JD, Baldwin WM 3rd, Steenbergen C, Judge DP, Dropulic LK, Halushka MK. Extensive cardiac allograft vasculitis and concurrent fat necrosis 6 years after orthotopic heart transplantation. J Heart Lung Transplant 2007;26:1212-6. 3. Emovon OE, King JA, Smith SR, et al. Clinical significance of eosinophils in suspicious or borderline renal allograft biopsies. Clin Nephrol 2003;59:367-72. 4. Koskinen P, Lemstrom K, Bruggeman C, Lautenschlager I, Hayry P. Acute cytomegalovirus infection induces a subendothelial inflammation (endothelialitis) in the allograft vascular wall. A possible linkage with enhanced allograft arteriosclerosis. Am J Pathol 1994;144:41-50. 5. Paavonen T, Mennander A, Lautenschlager I, Hayry P. Endothelialitis in accelerated allograft arteriosclerosis in human cardiac transplant recipients. Transplant Proc 1992;24:342-3. 6. Paavonen T, Mennander A, Lautenschlager I, Mattila S, Hayry P. Endothelialitis and accelerated arteriosclerosis in human heart transplant coronaries. J Heart Lung Transplant 1993;12:117-22. 7. Stewart S, Winters GL, Fishbein MC, et al. Revision of the 1990 working formulation for the standardization of nomenclature in the diagnosis of heart rejection. J Heart Lung Transplant 2005;24:1710-20. 8. Berry GJ, Brunt EM, Chamberlain D, et al. A working formulation for the standardization of nomenclature in the diagnosis of heart and lung rejection: Lung Rejection Study Group. The International Society for Heart Transplantation. J Heart Transplant 1990;9:593-601. 9. Billingham ME, Cary NR, Hammond ME, et al. A working formulation for the standardization of nomenclature in the diagnosis of heart and lung rejection: Heart Rejection Study Group. The International Society for Heart Transplantation. J Heart Transplant 1990;9:587-93. 10. Solez K, Colvin RB, Racusen LC, et al. Banff 07 classification of renal allograft pathology: updates and future directions. Am J Transplant 2008;8:753-60. 11. Berglin E, Kjellstrom C, Mantovani V, Stelin G, Svalander C, Wiklund L. Plasmapheresis as a rescue therapy to resolve cardiac rejection with vasculitis and severe heart failure. A report of five cases. Transpl Int 1995;8:382-7. 12. Mennander A, Paavonen T, Hayry P. Cyclosporine-induced endothelialitis and accelerated arteriosclerosis in chronic allograft rejection. Transplant Proc 1992;24:341.