- Email: [email protected]
Graft vascular disease in the great vessels and vasa vasorum
Graft Vascular Disease in the Great Vessels and Vasa Vasorum MIRI FUJITA, MD, MARY E. RUSSELL, MD, MARILYN A. MASEK, REED A. ROWAN, PHD, KAZUO NAGASHIMA, MD, AND MARGARET E. BILLINGHAM, MD To determine whether the great vessels were subject to transplantation-associated arteriosclerosis
(graft vascular disease) we studied
sections of aorta and pulmonary
arteries from 19 cardiac explant
cases. Sections of the same vessels from five autopsied hearts from patients who had received liver and/or kidney transplants with immunotherapy
as well as from eight recipient hearts served as con-
trols. In addition,
eight donor hearts (three aortae and eight pul-
monary arteries) that were not used for heart transplantation examined. Intimal proliferation the pulmonary
were
was noted in the aorta but affected
arteries only slightly, whereas the vasa vasorum of
both vessels were involved and were occluded or stenosed by thickened intima containing
recanalized
capillaries.
On morphometric
study the ratio of intima to media in the aorta was significantly higher
in the heart transplant
group
than in the control
group
(0.14 k 0.14 [n = 191 u 0.02 f 0.02 [n = 161; P < .Ol). This intimal thickening was not correlated with the use of cyclosporine other clinicopathologic
factors. The cytomegalovirus
or with
infection rate
was higher in the heart transplant group (85%) than in the control group
(40%). HUMPATHOL 24~1067-1072. Copyright 0 1993 by W.B.
Saunders Company
Transplantation-associated coronary artery disease (TX-CAD) is the leading cause of death in long-term cardiac allograft recipients.’ Included in the etiologies of TX-CAD are cyclosporine cytotoxicity to the endothelium,’ viral infection (mostly with cytomegalovirus),3 and immune-mediated endothelial damage, initially based on experience with human renal transplantation4 and more recently on immunohistochemical study and endothelial cell cultures in human heart transplantation.5 There has been no agreement on the etiology of TX-CAD. One of the problems in previous studies has been that the coronary arteries used for pathologic analysis have been severely damaged, showing the end stage of TX-CAD. The aorta is the most common site of arteriosclerosis, providing a good example for the study of early events of TX-CAD. However, there have been few reports as to whether the aorta was involved in transplantation-associated arteriosclerosis. Moreover, the aortic involvement with some immune-mediated diseases in which the coronary arteries, except for their ostia, are From the Department of Pathology, Stanford University School of Medicine, Stanford, CA; the Department of Cardiology, Harvard School of Public Health, Boston, MA; and the Department of Pathology, Hokkaido University School of Medicine, Sapporo, Japan. Accepted for publication April 26, 1993. Key words: transplantation-associated arteriosclerosis, aorta, pulmonary artery, morphometric analysis. Address correspondence and reprint requests to Miri Fujita, MD, Department of Pathology, Hokkaido University School of Medicine, North 15, West 7, Kita-ku, Sapporo 060, Japan. Copyright 0 1993 by W.B. Saunders Company 0046-8177/93/2410-0004$5.00/O
1067
spared has suggested that the aorta is antigenically different from the coronary artery. It is known that there is a difference in major histocompatibility class II antigen expression between renal arteriolar endothelium and glomerular capillary endothelium.‘j Thus, examination of the aorta in patients with cardiac transplantation could provide important findings for analysis of the mechanism of graft arteriosclerosis as well as for ordinary arteriosclerosis. For the present study we selected hearts and aortae that were obtained during surgery for cardiac re-transplantation because it is important to obtain fresh surgical material for precise morphologic analysis. Our observation of the intima showed that the aortic intima of the cardiac transplant patients was markedly thickened compared with that of the media and that this intimal thickening was more prominent than that seen in ordinary arteriosclerosis. Therefore, we applied morphometric measurement to obtain quantitative data on the intimal thickening and analyzed the results statistically in comparison with various clinical and therapeutic factors.
MATERIALS
AND METHODS
Nineteen cardiac allografts with great vessels were studied in 19 patients. All the allografts were obtained for re-transplantation at Stanford University between October 1978 and September 1991 because of severe TX-CAD. The patients’ ages at re-transplantation ranged from 19 to 55 years (mean age, 43 + 11 years [&SD]). Patients aged 19 to 54 years (mean age, 41 + 14 years, n = 9) were treated with cyclosporine (Cys+) and patients aged 32 to 55 years (mean age, 45 f 8 years, n = 10) were not treated with cyclosporine (Cys-). The ages of the donors ranged from 16 to 37 years (mean age, 25 ? 6 years); the donors for the nine Cysf patients were aged 16 to 35 years (mean age, 25 + 6 years) and the donors for the 10 Cys- patients were aged 17 to 37 years (mean age, 24 * 7 years). The length of graft survival ranged from 5.5 to 143 months (mean, 53 f 39 months): 5.5 to 91 months in the nine Cys+ patients (mean, 37 + 27 months) and 17 to 37 months in the 10 Cys- patients (mean, 68 ? 44 months). The reasons for the first heart transplantations were cardiomyopathy in eight patients, CAD in eight patients, and Adriamycin (Adria Laboratories, Columbus, OH)-induced cardiotoxicity, rheumatic pancarditis, and infective endocarditis after aortic valve replacement with xenograft in one patient each. Autopsy cases served as controls since the intimal thickness was not influenced by postmortem hours. There were five autopsied hearts from liver and/or kidney transplant patients whose ages ranged from 41 to 60 years (mean age, 49 f 8 years), eight hearts (removed for transplantation because of cardiomyopathy in six cases, pulmonary hypertension in one case, and CAD in one case) from patients whose ages ranged
HUMAN PATHOLOGY
Volume 24, No. 10 (October 1993)
from 17 to 58 years (mean age, 37 f 15 years), and eight donor hearts (the aorta as well as the aortic valves had already been excised for valve transplantation in five cases, so that three aortae and eight pulmonary arteries were obtained) from adult patients whose ages were not available. We did not use CAD cases that had undergone transplantation for controls because most of them were used for other studies. To evaluate the degree of coronary arteriosclerosis we used a three-tiered grading system consisting of severe (+++), moderate (++), and mild (+) levels: “severe” denotes that more than 80% of the luminal stenosis was found in all three coronary arteries, “moderate” denotes that the lumen was stenosed from 40% to 80% in all coronary arteries, and “mild” denotes that the lumen was narrowed from 20% to 40% in all coronary arteries. However, when 80% and 40% occlusions were found in the arteries we used the term “moderate to severe” (++ - +++) and when 80% and 30% occlusions were found we used the term “mild to severe” (+ - +++) (Table 1). The aortae and pulmonary arteries of the allografts were excised cross-sectionally at the position between the valve and the suture line of the allografts; those of the controls were excised at the same site. To avoid the elect of local trauma associated with surgery we extracted the specimens more than 1 cm from the suture line; in addition, if we encountered cases of severe adventitial fibrosis we omitted them from the evaluation. In routine processing, paraffin-embedded sections were stained with hematoxylin-eosin, van Gieson’s method for elastic fibers, and Masson’s trichrome stain for collagen fibers. Intimal thickening of the aortae and pulmonary arteries was graded from mild (+) to severe (+++). “Mild” is defined as between approximately one fifth and one third of the whole vessel wall, “moderate” as between one third and one half of the wall, and “severe” as more than half of the wall. Two pathologists (M.F. and M.E.B.) performed the grading independently. The vasa vasorum of both vessels also were examined for the presence of intimal thickening. For determination of the ratio of intima to media of the great vessels morphometry’ was performed by one of the authors (R.A.R.), who had no prior information. Briefly, using an ocular grid
lntimal Proliferation
TABLE 1.
that had 100 counting points of line intersections and an ohjective lens of X2.5 magnification of an ordinary light microscope. the number of grid points lying over the intima and the media was counted independently by movement of a glass slide until the entire vessel was covered. The ratio of intima to media was the quotient of the total number of counting points over the intima divided by that over the media. We reviewed clinical and pathology records, including slides with specimens graded from mild (+) to severe (+++) for coronary atherosclerotic lesions, data on cytomegalovirus infection (serology, culture, and histology), and drugs given for immunotherapy. We used statistical analysis by t-test and one-way analysis of variance in the morphometric study to determine whether the ratio of intima to media of the great vessels differed between the heart re-transplant group and the control group, and whether the ratio correlated with clinical parameters, including the use of cyclosporine, age of donor or recipient, length of survival of grafts, presence of CAD in the original hearts, association with acute rejection in the myocardium, and status following another transplantation. The factor of gender, however, could not be analyzed because there were only two female patients (cases no. 13 and 18) and only one female donor (case no. 13) (Table 1).
RESULTS Clinical and histopathologic data are listed in Table 1. Ten patients had no CAD in their original hearts, whereas nine patients had moderate to severe CAD. Transplantation-associated CAD was severe in 15 cases, moderate in one case, and severe to moderate in three cases. These histologic features have been described elsewhere by Billingham.’ Case no. 7 was unusual in that the transplanted heart showed right ventricular hypertrophy, suggesting the presence of car pulmonare. Among 13 patients for whom data on cytomegalovirus infection were available, 11 (85%) were infected, of whom seven of nine were in the Cys+ group and four
of the Great Vessels in Transplant
Coronary Artery Disease Intima/ Intimal
Patient Case
Age (yr,/
No.
Sex
Thickening
Media
Donor RtXS””
Age
for First
(YG SW
Duration
CAD
of Graft
Original
(I”“)
Heart
Transplant
in
Aorta CAD
in
use of
CMV
Graft
Infection
VZKd Awta
Cyclosporine
PA Vasa
Vasorum _
PA _
1
19/M
28/M
ADR
~
ff-ttt
+
+
t+
2
53/M
24/M
CAD
13
++
+++
+
+
3
52/M
16/M
Rheumatic
14
-
+++
+ _
+
t
+
_ _
5.5
pancarditis >
+
+
+
+
+
t
+
+
+
+
++
++
++
+
+
+
++
+ (+vein) t
t _
t
t
- (net)
t
32/M
CAD
20
+t+
+++
43/M
18/M
CAD
28
+++
++-+++Re
6
27/M
20/M
CM
31
-
t+t
7
47/M
25/M
Post-AVR
34
-
+++
8
51/M
28/M
CAD
38
+-+++
++t
AR
?
9
54/M
32/M
CM
38
++
t+
AR
t _
+++Ath
t
t++Ath
IF.
AR
+
_
45/M
5
10
37/M
19/M
CAD
40
+++
t++
Re
+ >
11
48/M
23/M
CM
53
-
t+t
AR
t
Aorta
_
>
4
Re
Vasorum
(KVH)
PA
0.23
0
0.05
0
0.02
0
0.12
0
0.05
0.02
t
0.07
0
+
0 27
n.25
t
0.03
0.04
(twin)
0.12
0.01
O.?O
0
(vast)
0.61
0
l?
55/M
37/M
CM
54
-
++t
?
_
++
t
_
+
0.23
0
I3
4X/F
35/F
CM
55
++-ttt
t
+ (+vein)
-
+
0.22
0
40/M
17/M
CAD
56
+++
+
+
0.05
0
I9
53/M
29/M
CM
71
_
++Ath
+
_
0.13
0
16
24/M
23/M
CM
91
? + _
+ _
t+Ath
14
+t+
+-+t
+ (+vein)
0.13
0
52/M
lR/M
CAD
+
+
0.09
0
-
ft+
t
t
+ _
_
+
17
+ _
_
0.04
0.01
t++
t
+
t
_
+
tft
18
32/F
18/M
CM
92 139
19
42/M
26/M
CAD
143
Abbreviations: association addition
CMV.
cytomegatovirus;
with acute rejection
to the artery,
ADR,
in the myocardium;
the vein also was affected;
_
t++
+++
+++
+++ AR
Adriamycine-induced AVR, RVH,
aortic
+
cardiotoxicity;
valve replacement;
right ventricular
hypertrophy;
_ _
Re. followed IE, infective net.
by third
endocarditis;
ischernic
necrosis;
+
heart
transplantation
CM,
cardiomyopathy;
vast. vasculitis:
_
due Ath,
0.04 0
to renewed
TX-CAD;
with atheroma;
PA, pulmonary
artery
twin.
AR, in
GRAFT
VASCULAR
DISEASE IN THE GREAT
of four were in the Cys- group. This shows a higher incidence than that detected in the control group. The rate of cytomegalovirus infection was 40% (two of five cases) in the subgroup of liver/kidney transplant recipients and 40% (two of five cases) in the subgroup of heart transplant recipients. Acute cellular rejection in the myocardium was accompanied by TX-CAD in five allograft recipients (26%; three of five in the Cys+ group and two of five in the Cys- group), with international grades Ia (focal mild) to IIIa (multifocal moderate).8 Three heart transplant cases (16%; three of three in the Cys- group) underwent re-transplantation. Aortic intimal thickening was recognized in all retransplant cases (loo%), ranging in grade from mild to severe (Fig 1, top). It was composed of usually intact endothelial cells, mononuclear cell infiltrates beneath the endothelium (often referred to as “endothelialitis”), and foamy macrophages. Moreover, the thickening was characterized by the presence of an increased number of “modified” smooth muscle cells, which have rather round nuclei and polygonal or asteroid cytoplasms that are arranged haphazardly.g In addition, there was loose matrix substance consisting of proteoglycan and immature collagen (Fig 1, bottom). During 40 months of graft survival, four patients (21%, two of four in the Cys+ group and two of four in the Cys- group) had atheromatous changes in the intima extending to the media of the aorta (Fig 2). The arterial vasa vasorum of the aorta were involved in 16 of the 19 cases (84%; seven of 16 in the CysS group and nine of 16 in the Cys- group), including three in which the venous vasa vasorum also were involved (Fig 3). On the other hand, the pulmonary arteries showed much less intimal change with mild to moderate thickening in three of the 19 cases (16%; two of three in the Cys+ group and one of three in the Cys- group). The intima showing a moderate grade of thickening consisted of mostly myofibroelastic proliferation and no atheromatous changes were observed (Fig 4, top). The vasa vasorum of the pulmonary artery were affected in 12 of the 19 cases (63%; seven of 12 in the Cysf group and five of 12 in the Cys- group). Occluded vasa vasorum with recanalization were occasionally encountered (Fig 4, bottom). Most recanalized capillaries were found in intima with relatively preserved internal elastic lamina. In case no. 11 vasculitis was found in the vasa vasorum and the corresponding pulmonary arterial wall showed extensive necrosis (Fig 5). In the control subgroup of liver/kidney transplants the case with the longest length of graft survival (14 months) and one case of systemic lupus erythematosus showed intimal thickening of the aorta and pulmonary artery, respectively. The histology, however, was different from that of cardiac allografts with regard to cellularity, types of matrix, and the ratio of intima to media. In allografts we found more increased cellularity, less mature collagen, and an increased ratio of intima to media compared with the values in the two cases of the control subgroup. There was no difference in the histology of the intima between the control subgroup of liver/kidney transplant cases and the nontransplant control subgroup (recipients and donors). 1069
VESSELS
(Fujita et al)
FIGURE 1. Aortic intimal thickening in case no. 13. (Top) Moderate grade of intimal thickening and accumulation of foamy macrophages were seen adjacent to the media in the lower half. (Elastic van Gieson stain; original magnification x100.) (Bottom) lntimal thickening with superficial mononuclear cell infiltrates showing the features of endothelialitis. Foamy macrophages and modified smooth muscle cells also were seen in the loose matrix. (Masson’s trichrome stain; original magniftcation X200.)
The results of the morphometric studies were as follows: the ratio (mean + SD) of intima to media in the aorta was 0.14 f 0.14 (n = 19) in the heart transplant group and 0.02 & 0.02 (n = 16) in the control group (P < .Ol), 0.01 + 0.01 (n = 5) in the control subgroup of liver/kidney transplant cases, and 0.02 +- 0.02 (n = 11) in the nontransplant control subgroup. There was a significant difference between the heart transplant and liver/kidney transplant groups (P < .Ol) and the nontransplant subgroup (P < .Ol). There was no significant difference between the liver/kidney transplant control subgroup and the nontransplant control subgroup. The highest ratio of 0.03 was found in the liver/kidney transplant case with the longest length of graft survival.
HUMAN PATHOLOGY
Volume 24, No. IO (October
1993)
allografts explanted because of severe TX-CAD. The intimal thickening consisted of irregularly arranged smooth muscle cells, which are referred to as of synthetic (or secretory) phenotype. They are usually termed “modified” smooth muscle cells in electron microscopic studies and are found in the thickened intima and injured media.“~” These are different from ordinary contractile smooth muscle cells in the normal vascular media, which have elongated nuclei and spindle-shaped cytoplasms arranged in a unidirectional fashion. In our present cases these cells had the light microscopic appearance of “modified” smooth muscle cells. Endothelialitis, often described in the transplanted coronary arteries, was also found in the aortic intima, suggesting an immune-mediated inflammation.’ Kennedy et al” described nine autopsy cases examined up to 10 months after heart transplantation, of which two showed aortic intimal thickening and seven had necrotizing and/or obliterative vasa vasorum of the aorta. Our study was confined to re-transplant cases due to severe TX-CAD and included longer lengths of graft survival so that a higher incidence of aortic intimal thickening was recognized than that reported by Kennedy et al.” The pulmonary artery was affected much less than the aorta, except in one case with right ventricular hypertrophy showing moderate intimal proliferation, This
FIGURE 2. Aorta showing the highest increased ratio of intima to media with atheromatous change in the intima extending to the media (case no. 1 I). (Elastic van Gieson stain; original magnification x64.)
Although the intima to media ratio in the Cys+ heart transplant subgroup (0.19 + 0.18, n = 9) was higher than that in the Cys- heart transplant subgroup (0.10 f 0.07, n = lo), the difference between them was not significant; only one ratio (case no. 11) was remarkably high (Fig 2), whereas the others were similar. The factors of presence of original CAD, accompanying acute rejection in the myocardium, status following a re-transplantation, length of graft survival, and patients’ and donors’ ages resulted in no significant difference in the ratio or no correlation with the ratio of intima to media of the aorta in the heart transplant group. The ratio of intima to media in the pulmonary artery was 0.02 f 0.06 (0.03 + 0.08, n = 9 in the Cysf group and 0.01 f 0.01, n = 10 in Cys- group) in heart transplant cases, 0.01 f 0.02 (n = 5) in the control subgroup of liver/kidney transplant cases, and 0.01 + 0.01 (n = 16) in the nontransplant control subgroup. There was no significant difference among the three groups and no significant correlation between the ratio and the various factors listed above, including the use of cyclosporine. DISCUSSION In this study we found that the same histology of intimal proliferation was observed in all aortae of cardiac 1070
FIGURE 3. Vasa vasorum of the aorta seen in case no. 9. There are three arteries with marked intimal proliferation (arrows) and one vein with lesser intimal change seen at the left lower corner. (Elastic van Gieson stain; original magnification x64.)
GRAFT
VASCULAR
DISEASE IN THE GREAT VESSELS
(Fujita et al)
suggests that pressure increases intimal proliferation in allografts, as is the case in the saphenous vein of aortocoronary bypass grafts.i3 A similar observation has been made that an anomalous coronary artery deriving from the pulmonary artery showed no atherosclerotic changes compared with the finding that coronary arteries originating from the aorta developed severe atherosclerosis.14 A mild to moderate degree of intimal thickening of the pulmonary artery was reported in the cases of combined heart-lung transplantation, but the severity of the grade of the lesion was not described.15 In addition to the pressure effects, the difference between the aorta and pulmonary artery is in the oxygen content of the blood flowing through them. The intimal proliferation of the pulmonary artery was experimentally induced by hyperoxemic normotensive perfusion’“; however, there was little information about the oxygen content obtained in our series. The vasa vasorum in both the aorta and pulmonary artery also were affected by graft vascular disease. The vasa vasorum of the root of the aorta are thought to arise mostly from the proximal coronary arteries and partly from surrounding recipient tissue.‘* The vasa vasorum of the pulmonary artery are thought to arise mainly from bronchial arteries” and partly from the proximal coronary arteries or their branches. An example is the conus artery or conus branch, which is easily recognized at least up to the level
FIGURE 4. Pulmonary artery showing intimal thickening and affected vasa vasorum in case no. 7. This heart revealed remarkable right ventricular hypertrophy suggesting pulmonary hypertension. (Top) Moderate grade of thickening consisted of mostly myoftbroelastic proliferation in the intima, and no atheromatous changes were observed. (Elastic van Gieson stain; original magnification x40.) (Bottom) A vas vasorum of pulmonary artery showing marked intimal thickening that contained hyalinously thickened capillaries (arrows). Note the almost intact internal elastic lamina. (Elastic van Gieson stain; original magnification X300.)
FIGURE 5. Pulmonary artery showed ischemic necrosis of the entire wall due to vasculitis of vasa vasorum in case no. 11. (Elastic van Gieson stain; original magnification x64.)
1071
HUMAN PATHOLOGY
Volume 24, No. 10 (October
of the pulmonary valve. Occluded vasa vasorum with recanalized capillaries in intima may suggest that endothelialitis has occurred in these vessels. In addition, it appears that the vasa vasorum of both vessels were affected similarly in graft vascular disease because they had the same systemic pressure. A somewhat lower incidence of affected vasa vasorum of the pulmonary artery could be due to the dominant supply from the bronchial arteries rather than from the coronary arterial supply. In one case we found that aseptic vasculitis of the vasa vasorum of the pulmonary artery and the vascular wall drained by this vessel was necrotic, which may suggest the presence of immunemediated inflammatory processes on the vasa vasorum. Application of the point counting method to evaluate the intimal thickening was shown to correlate well with the semiquantitative measurements of grading; hence, we were able to compare the predisposing factors statistically. The use of cyclosporine, which is thought to be toxic to endothelial cells, showed no significant correlation with graft vascular disease. This could be because not only cyclosporine but also other immunosuppressive drugs were used in our cases, since it was reported that in rat aortic allografts combined immunotherapy did not induce the intimal thickening compared with cyclosporine alone.” Other factors also showed no correlation with the intima to media ratio. Thus, there were no predictive factors of graft vascular disease and TX-CAD. The cytomegalovirus infection rate of 85% was higher than the rate of 40% seen in our control group and the previously reported rate of 45% seen in heart and heart-lung transplant cases.ig This suggests a correlation between TX-CAD or graft vascular disease of the aorta and cytomegalovirus infection. Hruban et al detected cytomegalovirus nucleic acid in lesions of TxCAD, but not in a control group of autopsy cases.” Although various risk factors for TX-CAD were evaluated previously on a large scale at Stanford University Medical Center, there were no correlations between the development of TX-CAD and recipient age or the number of rejection episodes during the first year after transplantation; most of rejection episodes tended to occur during this period. Moreover, there also were no relationships between TX-CAD and the mean daily dose of prednisone, HLA mismatch, plasma cholesterol levels, fasting blood sugar level, or cause of heart transplantation, which included CAD, cardiomyopathy, and valvular heart diseases. Only the donor age and plasma triglyceride level in the recipients seemed to be related to TX-CAD.’ A number of investigators have studied many other risk factors, including obesity, anti-donor HLA antibodies, cytotoxic B-cell antibodies, and cytomegalovirus infection.21~22 One of the most frequent possible contributors to TX-CAD seemed to be a combination of lipid abnormalities and immune reaction.” In conclusion, the basic process of TX-CAD, possibly caused by cytomegalovirus infection and/or immune-mediated inflammation, could occur in the aorta and vasa vasorum of the great vessels in association with factors such as pressure and/or blood oxygen content. 1072
1993)
However, to understand the precise mechanism of endothelialitis and intimal thickening it is necessary to investigate more cases and to correlate the histologic findings with the total number of acute rejection episodes, the number of severe rejection episodes, HLA mismatches, and posttransplant hyperlipidemia. REFERENCES I. Billingham ME: Cardiac transplant atherosclerosis. Transplant 19:19-25, 1987 2. Sedmak DD. Orosz CG: The role of vascular endothelial cells in transplantation, Arch Pathol Lab Med 115:260-265, 1991 3. Grattan MT, Moreno-Cabral CE, Starnes VA, et al: Cytomegalovirus infection is associated with cardiac allograft rejection and atherosclerosis. JAMA 261:3561-3566, 1989 4. Porter KA, Thomson WB, Owen K, et al: Obliterative vascular changes in four human kidney homotransplants. Br Med J 2:639-645, 1963 5. Solomon RN, Hughes CCW, Schoen FJ, et al: Human coronary transplantation-associated arteriosclerosis. Evidence for a chronic immune reaction to activated graft endothelial cells. Am J Pathol 138: 791-798, 1991 6. Mullet--Ruchholtz W: The immunopathology of transplantation, in Sale GE (ed): The Pathology of Organ Transplantation. Boston, MA, Butterworths, 1990, pp l-33 7. van Diest PJ, Baak JPA: Morphometry, in Bibbo M (ed): Comprehensive Cytopathology. Philadelphia, PA, Saunders, 1991, pp 946964 8. Billingham ME, Cary NRB, Hammond ME. et al: A working formulation for the standardization of nomenclature in the diagnosis of heart and lung rejection: Heart rejection study group. J Heart Transplant 9:587-593, 1990 9. Gotlieb Al, Havenith MG: Atherosclerosis: Lesions and pathogenesis, in Silver MD (ed): Cardiovascular Pathology, vol 1 (ed 2). New York. NY. Churchill Livingstone. 1991. DD 225-265 10. Campbell GR, Campbell YH, Manderso;l’JA, et al: Arterial smooth muscle. A multifunctional mesenchymal cell. Arch Path01 Lab Med 112:977-986, 1988 1 1. Kockx MM, Cambier BA, Bortier HE, et al: The modulation of smooth muscle cell phenotype is an early event in human aortocoronary saphenous vein grafts. Virchows Arch A Pathol Anat Histopathol 420:155-l 62, 1992 12. Kennedy LJ. Jr, Bieber CP, Mitchinson MJ: Aortic histopathology in human cardiac allografts. J Thorac Cardiovasc Surg 62:4250,197l 13. Virmani R, Atkinson JB, Forman MB: Aortocoronaty bypass grafts and extracardiac conduits, in Silver MD (ed): Cardiovascular Pathology, vol 2 (ed 2). New York, NY, Churchill Livingstone. 199 1, pp 1607-1647 14. Baroldi C: Diseases of extramural coronary arteries, in Silver MD (ed): Cardiovascular Pathology, vol 1 (ed 2). New York, NY, Churchill Livingstone. 199 1, pp 487-563 15. Yousem SA, Tazelaar HD: Combined heart-lung transplantation, in Sale GE (ed): The Pathology of Organ Transplantation. Boston, MA, Butterworths, 1990, pp 153-l ‘78 16. Cohn LH, Kosek J, Angel1 WW: Pulmonary arteriosclerosis produced by hyperoxemic normotensive perfusion. Circulation 42: 114, 1970 (suppl 3) (abstr) 17. Tobin CE: Some observations concerning the puhnonic vasa vasorum. Surg Gynecol Obstet 111:297-303, 1960 18. Mennander A, Tiisala S, Paavonen T. et al: Chronic rejection of rat aortic allograft. II. Administration of cyclosporin induces accelerated allograft arteriosclerosis. Transplant Int 4: 173-l 79, 1991 19. Atkinson JB, Virmani R: Pathology of heart and combined heart-lung transplantation, in Virmani R, Atkinson JB, Fenoglio JJ (eds): Cardiovascular Pathology. Philadelphia, PA. Saunders. 199 1, pp 31 o-333 20. Hruban RH, Wu TC, Beschomer WE, et al: Cytomegdlovirus nucleic acids in allografted hearts. HUM PATHOL 21:981-983, 1990 2 1, Winters CL: The pathology of heart allograft rejection. Arch Pathol Lab Med 115:266-272, 1991 22. Hosenpud JD, Shipley GD, Wagner CR: Cardiac allograft vasculopathy: Current concepts, recent developments, and future directions. J Heart Lung Transplant 11:9-23, 1992 Proc
(graft vascular disease) we studied
sections of aorta and pulmonary
arteries from 19 cardiac explant
cases. Sections of the same vessels from five autopsied hearts from patients who had received liver and/or kidney transplants with immunotherapy
as well as from eight recipient hearts served as con-
trols. In addition,
eight donor hearts (three aortae and eight pul-
monary arteries) that were not used for heart transplantation examined. Intimal proliferation the pulmonary
were
was noted in the aorta but affected
arteries only slightly, whereas the vasa vasorum of
both vessels were involved and were occluded or stenosed by thickened intima containing
recanalized
capillaries.
On morphometric
study the ratio of intima to media in the aorta was significantly higher
in the heart transplant
group
than in the control
group
(0.14 k 0.14 [n = 191 u 0.02 f 0.02 [n = 161; P < .Ol). This intimal thickening was not correlated with the use of cyclosporine other clinicopathologic
factors. The cytomegalovirus
or with
infection rate
was higher in the heart transplant group (85%) than in the control group
(40%). HUMPATHOL 24~1067-1072. Copyright 0 1993 by W.B.
Saunders Company
Transplantation-associated coronary artery disease (TX-CAD) is the leading cause of death in long-term cardiac allograft recipients.’ Included in the etiologies of TX-CAD are cyclosporine cytotoxicity to the endothelium,’ viral infection (mostly with cytomegalovirus),3 and immune-mediated endothelial damage, initially based on experience with human renal transplantation4 and more recently on immunohistochemical study and endothelial cell cultures in human heart transplantation.5 There has been no agreement on the etiology of TX-CAD. One of the problems in previous studies has been that the coronary arteries used for pathologic analysis have been severely damaged, showing the end stage of TX-CAD. The aorta is the most common site of arteriosclerosis, providing a good example for the study of early events of TX-CAD. However, there have been few reports as to whether the aorta was involved in transplantation-associated arteriosclerosis. Moreover, the aortic involvement with some immune-mediated diseases in which the coronary arteries, except for their ostia, are From the Department of Pathology, Stanford University School of Medicine, Stanford, CA; the Department of Cardiology, Harvard School of Public Health, Boston, MA; and the Department of Pathology, Hokkaido University School of Medicine, Sapporo, Japan. Accepted for publication April 26, 1993. Key words: transplantation-associated arteriosclerosis, aorta, pulmonary artery, morphometric analysis. Address correspondence and reprint requests to Miri Fujita, MD, Department of Pathology, Hokkaido University School of Medicine, North 15, West 7, Kita-ku, Sapporo 060, Japan. Copyright 0 1993 by W.B. Saunders Company 0046-8177/93/2410-0004$5.00/O
1067
spared has suggested that the aorta is antigenically different from the coronary artery. It is known that there is a difference in major histocompatibility class II antigen expression between renal arteriolar endothelium and glomerular capillary endothelium.‘j Thus, examination of the aorta in patients with cardiac transplantation could provide important findings for analysis of the mechanism of graft arteriosclerosis as well as for ordinary arteriosclerosis. For the present study we selected hearts and aortae that were obtained during surgery for cardiac re-transplantation because it is important to obtain fresh surgical material for precise morphologic analysis. Our observation of the intima showed that the aortic intima of the cardiac transplant patients was markedly thickened compared with that of the media and that this intimal thickening was more prominent than that seen in ordinary arteriosclerosis. Therefore, we applied morphometric measurement to obtain quantitative data on the intimal thickening and analyzed the results statistically in comparison with various clinical and therapeutic factors.
MATERIALS
AND METHODS
Nineteen cardiac allografts with great vessels were studied in 19 patients. All the allografts were obtained for re-transplantation at Stanford University between October 1978 and September 1991 because of severe TX-CAD. The patients’ ages at re-transplantation ranged from 19 to 55 years (mean age, 43 + 11 years [&SD]). Patients aged 19 to 54 years (mean age, 41 + 14 years, n = 9) were treated with cyclosporine (Cys+) and patients aged 32 to 55 years (mean age, 45 f 8 years, n = 10) were not treated with cyclosporine (Cys-). The ages of the donors ranged from 16 to 37 years (mean age, 25 ? 6 years); the donors for the nine Cysf patients were aged 16 to 35 years (mean age, 25 + 6 years) and the donors for the 10 Cys- patients were aged 17 to 37 years (mean age, 24 * 7 years). The length of graft survival ranged from 5.5 to 143 months (mean, 53 f 39 months): 5.5 to 91 months in the nine Cys+ patients (mean, 37 + 27 months) and 17 to 37 months in the 10 Cys- patients (mean, 68 ? 44 months). The reasons for the first heart transplantations were cardiomyopathy in eight patients, CAD in eight patients, and Adriamycin (Adria Laboratories, Columbus, OH)-induced cardiotoxicity, rheumatic pancarditis, and infective endocarditis after aortic valve replacement with xenograft in one patient each. Autopsy cases served as controls since the intimal thickness was not influenced by postmortem hours. There were five autopsied hearts from liver and/or kidney transplant patients whose ages ranged from 41 to 60 years (mean age, 49 f 8 years), eight hearts (removed for transplantation because of cardiomyopathy in six cases, pulmonary hypertension in one case, and CAD in one case) from patients whose ages ranged
HUMAN PATHOLOGY
Volume 24, No. 10 (October 1993)
from 17 to 58 years (mean age, 37 f 15 years), and eight donor hearts (the aorta as well as the aortic valves had already been excised for valve transplantation in five cases, so that three aortae and eight pulmonary arteries were obtained) from adult patients whose ages were not available. We did not use CAD cases that had undergone transplantation for controls because most of them were used for other studies. To evaluate the degree of coronary arteriosclerosis we used a three-tiered grading system consisting of severe (+++), moderate (++), and mild (+) levels: “severe” denotes that more than 80% of the luminal stenosis was found in all three coronary arteries, “moderate” denotes that the lumen was stenosed from 40% to 80% in all coronary arteries, and “mild” denotes that the lumen was narrowed from 20% to 40% in all coronary arteries. However, when 80% and 40% occlusions were found in the arteries we used the term “moderate to severe” (++ - +++) and when 80% and 30% occlusions were found we used the term “mild to severe” (+ - +++) (Table 1). The aortae and pulmonary arteries of the allografts were excised cross-sectionally at the position between the valve and the suture line of the allografts; those of the controls were excised at the same site. To avoid the elect of local trauma associated with surgery we extracted the specimens more than 1 cm from the suture line; in addition, if we encountered cases of severe adventitial fibrosis we omitted them from the evaluation. In routine processing, paraffin-embedded sections were stained with hematoxylin-eosin, van Gieson’s method for elastic fibers, and Masson’s trichrome stain for collagen fibers. Intimal thickening of the aortae and pulmonary arteries was graded from mild (+) to severe (+++). “Mild” is defined as between approximately one fifth and one third of the whole vessel wall, “moderate” as between one third and one half of the wall, and “severe” as more than half of the wall. Two pathologists (M.F. and M.E.B.) performed the grading independently. The vasa vasorum of both vessels also were examined for the presence of intimal thickening. For determination of the ratio of intima to media of the great vessels morphometry’ was performed by one of the authors (R.A.R.), who had no prior information. Briefly, using an ocular grid
lntimal Proliferation
TABLE 1.
that had 100 counting points of line intersections and an ohjective lens of X2.5 magnification of an ordinary light microscope. the number of grid points lying over the intima and the media was counted independently by movement of a glass slide until the entire vessel was covered. The ratio of intima to media was the quotient of the total number of counting points over the intima divided by that over the media. We reviewed clinical and pathology records, including slides with specimens graded from mild (+) to severe (+++) for coronary atherosclerotic lesions, data on cytomegalovirus infection (serology, culture, and histology), and drugs given for immunotherapy. We used statistical analysis by t-test and one-way analysis of variance in the morphometric study to determine whether the ratio of intima to media of the great vessels differed between the heart re-transplant group and the control group, and whether the ratio correlated with clinical parameters, including the use of cyclosporine, age of donor or recipient, length of survival of grafts, presence of CAD in the original hearts, association with acute rejection in the myocardium, and status following another transplantation. The factor of gender, however, could not be analyzed because there were only two female patients (cases no. 13 and 18) and only one female donor (case no. 13) (Table 1).
RESULTS Clinical and histopathologic data are listed in Table 1. Ten patients had no CAD in their original hearts, whereas nine patients had moderate to severe CAD. Transplantation-associated CAD was severe in 15 cases, moderate in one case, and severe to moderate in three cases. These histologic features have been described elsewhere by Billingham.’ Case no. 7 was unusual in that the transplanted heart showed right ventricular hypertrophy, suggesting the presence of car pulmonare. Among 13 patients for whom data on cytomegalovirus infection were available, 11 (85%) were infected, of whom seven of nine were in the Cys+ group and four
of the Great Vessels in Transplant
Coronary Artery Disease Intima/ Intimal
Patient Case
Age (yr,/
No.
Sex
Thickening
Media
Donor RtXS””
Age
for First
(YG SW
Duration
CAD
of Graft
Original
(I”“)
Heart
Transplant
in
Aorta CAD
in
use of
CMV
Graft
Infection
VZKd Awta
Cyclosporine
PA Vasa
Vasorum _
PA _
1
19/M
28/M
ADR
~
ff-ttt
+
+
t+
2
53/M
24/M
CAD
13
++
+++
+
+
3
52/M
16/M
Rheumatic
14
-
+++
+ _
+
t
+
_ _
5.5
pancarditis >
+
+
+
+
+
t
+
+
+
+
++
++
++
+
+
+
++
+ (+vein) t
t _
t
t
- (net)
t
32/M
CAD
20
+t+
+++
43/M
18/M
CAD
28
+++
++-+++Re
6
27/M
20/M
CM
31
-
t+t
7
47/M
25/M
Post-AVR
34
-
+++
8
51/M
28/M
CAD
38
+-+++
++t
AR
?
9
54/M
32/M
CM
38
++
t+
AR
t _
+++Ath
t
t++Ath
IF.
AR
+
_
45/M
5
10
37/M
19/M
CAD
40
+++
t++
Re
+ >
11
48/M
23/M
CM
53
-
t+t
AR
t
Aorta
_
>
4
Re
Vasorum
(KVH)
PA
0.23
0
0.05
0
0.02
0
0.12
0
0.05
0.02
t
0.07
0
+
0 27
n.25
t
0.03
0.04
(twin)
0.12
0.01
O.?O
0
(vast)
0.61
0
l?
55/M
37/M
CM
54
-
++t
?
_
++
t
_
+
0.23
0
I3
4X/F
35/F
CM
55
++-ttt
t
+ (+vein)
-
+
0.22
0
40/M
17/M
CAD
56
+++
+
+
0.05
0
I9
53/M
29/M
CM
71
_
++Ath
+
_
0.13
0
16
24/M
23/M
CM
91
? + _
+ _
t+Ath
14
+t+
+-+t
+ (+vein)
0.13
0
52/M
lR/M
CAD
+
+
0.09
0
-
ft+
t
t
+ _
_
+
17
+ _
_
0.04
0.01
t++
t
+
t
_
+
tft
18
32/F
18/M
CM
92 139
19
42/M
26/M
CAD
143
Abbreviations: association addition
CMV.
cytomegatovirus;
with acute rejection
to the artery,
ADR,
in the myocardium;
the vein also was affected;
_
t++
+++
+++
+++ AR
Adriamycine-induced AVR, RVH,
aortic
+
cardiotoxicity;
valve replacement;
right ventricular
hypertrophy;
_ _
Re. followed IE, infective net.
by third
endocarditis;
ischernic
necrosis;
+
heart
transplantation
CM,
cardiomyopathy;
vast. vasculitis:
_
due Ath,
0.04 0
to renewed
TX-CAD;
with atheroma;
PA, pulmonary
artery
twin.
AR, in
GRAFT
VASCULAR
DISEASE IN THE GREAT
of four were in the Cys- group. This shows a higher incidence than that detected in the control group. The rate of cytomegalovirus infection was 40% (two of five cases) in the subgroup of liver/kidney transplant recipients and 40% (two of five cases) in the subgroup of heart transplant recipients. Acute cellular rejection in the myocardium was accompanied by TX-CAD in five allograft recipients (26%; three of five in the Cys+ group and two of five in the Cys- group), with international grades Ia (focal mild) to IIIa (multifocal moderate).8 Three heart transplant cases (16%; three of three in the Cys- group) underwent re-transplantation. Aortic intimal thickening was recognized in all retransplant cases (loo%), ranging in grade from mild to severe (Fig 1, top). It was composed of usually intact endothelial cells, mononuclear cell infiltrates beneath the endothelium (often referred to as “endothelialitis”), and foamy macrophages. Moreover, the thickening was characterized by the presence of an increased number of “modified” smooth muscle cells, which have rather round nuclei and polygonal or asteroid cytoplasms that are arranged haphazardly.g In addition, there was loose matrix substance consisting of proteoglycan and immature collagen (Fig 1, bottom). During 40 months of graft survival, four patients (21%, two of four in the Cys+ group and two of four in the Cys- group) had atheromatous changes in the intima extending to the media of the aorta (Fig 2). The arterial vasa vasorum of the aorta were involved in 16 of the 19 cases (84%; seven of 16 in the CysS group and nine of 16 in the Cys- group), including three in which the venous vasa vasorum also were involved (Fig 3). On the other hand, the pulmonary arteries showed much less intimal change with mild to moderate thickening in three of the 19 cases (16%; two of three in the Cys+ group and one of three in the Cys- group). The intima showing a moderate grade of thickening consisted of mostly myofibroelastic proliferation and no atheromatous changes were observed (Fig 4, top). The vasa vasorum of the pulmonary artery were affected in 12 of the 19 cases (63%; seven of 12 in the Cysf group and five of 12 in the Cys- group). Occluded vasa vasorum with recanalization were occasionally encountered (Fig 4, bottom). Most recanalized capillaries were found in intima with relatively preserved internal elastic lamina. In case no. 11 vasculitis was found in the vasa vasorum and the corresponding pulmonary arterial wall showed extensive necrosis (Fig 5). In the control subgroup of liver/kidney transplants the case with the longest length of graft survival (14 months) and one case of systemic lupus erythematosus showed intimal thickening of the aorta and pulmonary artery, respectively. The histology, however, was different from that of cardiac allografts with regard to cellularity, types of matrix, and the ratio of intima to media. In allografts we found more increased cellularity, less mature collagen, and an increased ratio of intima to media compared with the values in the two cases of the control subgroup. There was no difference in the histology of the intima between the control subgroup of liver/kidney transplant cases and the nontransplant control subgroup (recipients and donors). 1069
VESSELS
(Fujita et al)
FIGURE 1. Aortic intimal thickening in case no. 13. (Top) Moderate grade of intimal thickening and accumulation of foamy macrophages were seen adjacent to the media in the lower half. (Elastic van Gieson stain; original magnification x100.) (Bottom) lntimal thickening with superficial mononuclear cell infiltrates showing the features of endothelialitis. Foamy macrophages and modified smooth muscle cells also were seen in the loose matrix. (Masson’s trichrome stain; original magniftcation X200.)
The results of the morphometric studies were as follows: the ratio (mean + SD) of intima to media in the aorta was 0.14 f 0.14 (n = 19) in the heart transplant group and 0.02 & 0.02 (n = 16) in the control group (P < .Ol), 0.01 + 0.01 (n = 5) in the control subgroup of liver/kidney transplant cases, and 0.02 +- 0.02 (n = 11) in the nontransplant control subgroup. There was a significant difference between the heart transplant and liver/kidney transplant groups (P < .Ol) and the nontransplant subgroup (P < .Ol). There was no significant difference between the liver/kidney transplant control subgroup and the nontransplant control subgroup. The highest ratio of 0.03 was found in the liver/kidney transplant case with the longest length of graft survival.
HUMAN PATHOLOGY
Volume 24, No. IO (October
1993)
allografts explanted because of severe TX-CAD. The intimal thickening consisted of irregularly arranged smooth muscle cells, which are referred to as of synthetic (or secretory) phenotype. They are usually termed “modified” smooth muscle cells in electron microscopic studies and are found in the thickened intima and injured media.“~” These are different from ordinary contractile smooth muscle cells in the normal vascular media, which have elongated nuclei and spindle-shaped cytoplasms arranged in a unidirectional fashion. In our present cases these cells had the light microscopic appearance of “modified” smooth muscle cells. Endothelialitis, often described in the transplanted coronary arteries, was also found in the aortic intima, suggesting an immune-mediated inflammation.’ Kennedy et al” described nine autopsy cases examined up to 10 months after heart transplantation, of which two showed aortic intimal thickening and seven had necrotizing and/or obliterative vasa vasorum of the aorta. Our study was confined to re-transplant cases due to severe TX-CAD and included longer lengths of graft survival so that a higher incidence of aortic intimal thickening was recognized than that reported by Kennedy et al.” The pulmonary artery was affected much less than the aorta, except in one case with right ventricular hypertrophy showing moderate intimal proliferation, This
FIGURE 2. Aorta showing the highest increased ratio of intima to media with atheromatous change in the intima extending to the media (case no. 1 I). (Elastic van Gieson stain; original magnification x64.)
Although the intima to media ratio in the Cys+ heart transplant subgroup (0.19 + 0.18, n = 9) was higher than that in the Cys- heart transplant subgroup (0.10 f 0.07, n = lo), the difference between them was not significant; only one ratio (case no. 11) was remarkably high (Fig 2), whereas the others were similar. The factors of presence of original CAD, accompanying acute rejection in the myocardium, status following a re-transplantation, length of graft survival, and patients’ and donors’ ages resulted in no significant difference in the ratio or no correlation with the ratio of intima to media of the aorta in the heart transplant group. The ratio of intima to media in the pulmonary artery was 0.02 f 0.06 (0.03 + 0.08, n = 9 in the Cysf group and 0.01 f 0.01, n = 10 in Cys- group) in heart transplant cases, 0.01 f 0.02 (n = 5) in the control subgroup of liver/kidney transplant cases, and 0.01 + 0.01 (n = 16) in the nontransplant control subgroup. There was no significant difference among the three groups and no significant correlation between the ratio and the various factors listed above, including the use of cyclosporine. DISCUSSION In this study we found that the same histology of intimal proliferation was observed in all aortae of cardiac 1070
FIGURE 3. Vasa vasorum of the aorta seen in case no. 9. There are three arteries with marked intimal proliferation (arrows) and one vein with lesser intimal change seen at the left lower corner. (Elastic van Gieson stain; original magnification x64.)
GRAFT
VASCULAR
DISEASE IN THE GREAT VESSELS
(Fujita et al)
suggests that pressure increases intimal proliferation in allografts, as is the case in the saphenous vein of aortocoronary bypass grafts.i3 A similar observation has been made that an anomalous coronary artery deriving from the pulmonary artery showed no atherosclerotic changes compared with the finding that coronary arteries originating from the aorta developed severe atherosclerosis.14 A mild to moderate degree of intimal thickening of the pulmonary artery was reported in the cases of combined heart-lung transplantation, but the severity of the grade of the lesion was not described.15 In addition to the pressure effects, the difference between the aorta and pulmonary artery is in the oxygen content of the blood flowing through them. The intimal proliferation of the pulmonary artery was experimentally induced by hyperoxemic normotensive perfusion’“; however, there was little information about the oxygen content obtained in our series. The vasa vasorum in both the aorta and pulmonary artery also were affected by graft vascular disease. The vasa vasorum of the root of the aorta are thought to arise mostly from the proximal coronary arteries and partly from surrounding recipient tissue.‘* The vasa vasorum of the pulmonary artery are thought to arise mainly from bronchial arteries” and partly from the proximal coronary arteries or their branches. An example is the conus artery or conus branch, which is easily recognized at least up to the level
FIGURE 4. Pulmonary artery showing intimal thickening and affected vasa vasorum in case no. 7. This heart revealed remarkable right ventricular hypertrophy suggesting pulmonary hypertension. (Top) Moderate grade of thickening consisted of mostly myoftbroelastic proliferation in the intima, and no atheromatous changes were observed. (Elastic van Gieson stain; original magnification x40.) (Bottom) A vas vasorum of pulmonary artery showing marked intimal thickening that contained hyalinously thickened capillaries (arrows). Note the almost intact internal elastic lamina. (Elastic van Gieson stain; original magnification X300.)
FIGURE 5. Pulmonary artery showed ischemic necrosis of the entire wall due to vasculitis of vasa vasorum in case no. 11. (Elastic van Gieson stain; original magnification x64.)
1071
HUMAN PATHOLOGY
Volume 24, No. 10 (October
of the pulmonary valve. Occluded vasa vasorum with recanalized capillaries in intima may suggest that endothelialitis has occurred in these vessels. In addition, it appears that the vasa vasorum of both vessels were affected similarly in graft vascular disease because they had the same systemic pressure. A somewhat lower incidence of affected vasa vasorum of the pulmonary artery could be due to the dominant supply from the bronchial arteries rather than from the coronary arterial supply. In one case we found that aseptic vasculitis of the vasa vasorum of the pulmonary artery and the vascular wall drained by this vessel was necrotic, which may suggest the presence of immunemediated inflammatory processes on the vasa vasorum. Application of the point counting method to evaluate the intimal thickening was shown to correlate well with the semiquantitative measurements of grading; hence, we were able to compare the predisposing factors statistically. The use of cyclosporine, which is thought to be toxic to endothelial cells, showed no significant correlation with graft vascular disease. This could be because not only cyclosporine but also other immunosuppressive drugs were used in our cases, since it was reported that in rat aortic allografts combined immunotherapy did not induce the intimal thickening compared with cyclosporine alone.” Other factors also showed no correlation with the intima to media ratio. Thus, there were no predictive factors of graft vascular disease and TX-CAD. The cytomegalovirus infection rate of 85% was higher than the rate of 40% seen in our control group and the previously reported rate of 45% seen in heart and heart-lung transplant cases.ig This suggests a correlation between TX-CAD or graft vascular disease of the aorta and cytomegalovirus infection. Hruban et al detected cytomegalovirus nucleic acid in lesions of TxCAD, but not in a control group of autopsy cases.” Although various risk factors for TX-CAD were evaluated previously on a large scale at Stanford University Medical Center, there were no correlations between the development of TX-CAD and recipient age or the number of rejection episodes during the first year after transplantation; most of rejection episodes tended to occur during this period. Moreover, there also were no relationships between TX-CAD and the mean daily dose of prednisone, HLA mismatch, plasma cholesterol levels, fasting blood sugar level, or cause of heart transplantation, which included CAD, cardiomyopathy, and valvular heart diseases. Only the donor age and plasma triglyceride level in the recipients seemed to be related to TX-CAD.’ A number of investigators have studied many other risk factors, including obesity, anti-donor HLA antibodies, cytotoxic B-cell antibodies, and cytomegalovirus infection.21~22 One of the most frequent possible contributors to TX-CAD seemed to be a combination of lipid abnormalities and immune reaction.” In conclusion, the basic process of TX-CAD, possibly caused by cytomegalovirus infection and/or immune-mediated inflammation, could occur in the aorta and vasa vasorum of the great vessels in association with factors such as pressure and/or blood oxygen content. 1072
1993)
However, to understand the precise mechanism of endothelialitis and intimal thickening it is necessary to investigate more cases and to correlate the histologic findings with the total number of acute rejection episodes, the number of severe rejection episodes, HLA mismatches, and posttransplant hyperlipidemia. REFERENCES I. Billingham ME: Cardiac transplant atherosclerosis. Transplant 19:19-25, 1987 2. Sedmak DD. Orosz CG: The role of vascular endothelial cells in transplantation, Arch Pathol Lab Med 115:260-265, 1991 3. Grattan MT, Moreno-Cabral CE, Starnes VA, et al: Cytomegalovirus infection is associated with cardiac allograft rejection and atherosclerosis. JAMA 261:3561-3566, 1989 4. Porter KA, Thomson WB, Owen K, et al: Obliterative vascular changes in four human kidney homotransplants. Br Med J 2:639-645, 1963 5. Solomon RN, Hughes CCW, Schoen FJ, et al: Human coronary transplantation-associated arteriosclerosis. Evidence for a chronic immune reaction to activated graft endothelial cells. Am J Pathol 138: 791-798, 1991 6. Mullet--Ruchholtz W: The immunopathology of transplantation, in Sale GE (ed): The Pathology of Organ Transplantation. Boston, MA, Butterworths, 1990, pp l-33 7. van Diest PJ, Baak JPA: Morphometry, in Bibbo M (ed): Comprehensive Cytopathology. Philadelphia, PA, Saunders, 1991, pp 946964 8. Billingham ME, Cary NRB, Hammond ME. et al: A working formulation for the standardization of nomenclature in the diagnosis of heart and lung rejection: Heart rejection study group. J Heart Transplant 9:587-593, 1990 9. Gotlieb Al, Havenith MG: Atherosclerosis: Lesions and pathogenesis, in Silver MD (ed): Cardiovascular Pathology, vol 1 (ed 2). New York. NY. Churchill Livingstone. 1991. DD 225-265 10. Campbell GR, Campbell YH, Manderso;l’JA, et al: Arterial smooth muscle. A multifunctional mesenchymal cell. Arch Path01 Lab Med 112:977-986, 1988 1 1. Kockx MM, Cambier BA, Bortier HE, et al: The modulation of smooth muscle cell phenotype is an early event in human aortocoronary saphenous vein grafts. Virchows Arch A Pathol Anat Histopathol 420:155-l 62, 1992 12. Kennedy LJ. Jr, Bieber CP, Mitchinson MJ: Aortic histopathology in human cardiac allografts. J Thorac Cardiovasc Surg 62:4250,197l 13. Virmani R, Atkinson JB, Forman MB: Aortocoronaty bypass grafts and extracardiac conduits, in Silver MD (ed): Cardiovascular Pathology, vol 2 (ed 2). New York, NY, Churchill Livingstone. 199 1, pp 1607-1647 14. Baroldi C: Diseases of extramural coronary arteries, in Silver MD (ed): Cardiovascular Pathology, vol 1 (ed 2). New York, NY, Churchill Livingstone. 199 1, pp 487-563 15. Yousem SA, Tazelaar HD: Combined heart-lung transplantation, in Sale GE (ed): The Pathology of Organ Transplantation. Boston, MA, Butterworths, 1990, pp 153-l ‘78 16. Cohn LH, Kosek J, Angel1 WW: Pulmonary arteriosclerosis produced by hyperoxemic normotensive perfusion. Circulation 42: 114, 1970 (suppl 3) (abstr) 17. Tobin CE: Some observations concerning the puhnonic vasa vasorum. Surg Gynecol Obstet 111:297-303, 1960 18. Mennander A, Tiisala S, Paavonen T. et al: Chronic rejection of rat aortic allograft. II. Administration of cyclosporin induces accelerated allograft arteriosclerosis. Transplant Int 4: 173-l 79, 1991 19. Atkinson JB, Virmani R: Pathology of heart and combined heart-lung transplantation, in Virmani R, Atkinson JB, Fenoglio JJ (eds): Cardiovascular Pathology. Philadelphia, PA. Saunders. 199 1, pp 31 o-333 20. Hruban RH, Wu TC, Beschomer WE, et al: Cytomegdlovirus nucleic acids in allografted hearts. HUM PATHOL 21:981-983, 1990 2 1, Winters CL: The pathology of heart allograft rejection. Arch Pathol Lab Med 115:266-272, 1991 22. Hosenpud JD, Shipley GD, Wagner CR: Cardiac allograft vasculopathy: Current concepts, recent developments, and future directions. J Heart Lung Transplant 11:9-23, 1992 Proc