The Conduction System in Human Cardiac Allografts

Path. Res. Pract. 188, 783-790 (1992)

The Conduction System in Human Cardiac Allografts A Histological and Immunopathological Study A. Foerster Department of Pathology, National Hospital of Norway, Os/o, Norway

SUMMARY Twenty-five hearts were lost in the first 103 orthotopic human cardiac allografts in 100 recipients. Twenty-two patients died. The first nine recipients were treated with Cyclosporin A and high dose Prednisolone and the subsequent recipients were immunosuppressed with a tripie regimen consisting of Azathioprine in addition to Cyclosporin A and low dose Prednisolone. Twenty grafts were examined at autopsy and three after retransplantation. Fourteen hearts with acute rejection and vascular pathology were investigated regarding the histological and immunopathological alterations due to cellular and vascular rejection. The recipient sinoatrial node was available for examination in 10, the donor node in 13 and the atrioventricular conductive tissue in all 14 grafts. The conduction system was involved in 8 out of 11 allografts attacked by acute cellular rejection. In one allograft in which the patient developed A V block, the cellular infiltrates were almost exclusively observed in the atrioventricular tissue. Most of the cells were T lymphocytes, while macrophages and B cells were present to alesser extent. There was little evidence of permanent structural damage to the atrioventricular tissue after recurrent mild or moderate episodes of acute cellular rejection. In the sinus nodes, however, risk of traumatic damage to the recipient and donor sinus node was about half and a quarter respectively. Acute and chronic vascular rejection seemed to be equally distributed in the sinus node, atrioventricular arteries and in the intramural branches of the coronaries. No morphological differences were established between the two types of immunosuppressive regimes.

Introduction There has been limited interest in the investigation of the conductive system in human cardiac allografts. Since the introduction of the endomyocardial biopsy technique in 19734, many publications have dealt with the histopathological monitoring of the effectiveness of immunosuppressive treatment. Bieber et al. l published a study of the conduction system in dogs in 1969 and in 20 human heart transplants in 19702 . In 1986 Stovin and Hewitt ll examined 11 orthotopically transplanted human hearts at retransplantation or autopsy. The study was a time-related examination of the hisropathological changes of the conductive tissue in patients dying at various intervals posttransplant. The aims of the present study were: 1) to © 1992 by Gustav Fischer Verlag, Stuttgart

investigate the different types of allograft rejection in the cardiac conductive system by immunohistochemical techniques, 2) to study the morphological changes in the rejection process, 3) to analyse the relation between morphology and function of the conductive system.

Material and Methods From November 1983 to January 1990, 103 human cardiac transplantations were performed on 100 patients. Follow-up details are available up to March 1990. From 25 grafts lost 14 were entering the study. The basic clinical and transplant data are shown in Table 1. 0344-0338/92/0188-0783$3.50/0

784 . A. Foerster Table 1. Clinical and pathological profile of 14 cardiac recipients TX. No.

Recipient Diag-

Graft survival (days)

age

sex

40 39 39 39 52 52 16 51 55 56 56 47 54 34

F F M M M M M M M M M M M

Alive

Donor age

sex

19 39 19 39 36 37 17

M F M M M M M M M M F M M M

Terminal clinical stage

Cause of deathlgraftloss

Blaod group danar

reCl-

A

A

B

B B A A A A

nOSIS

2 4 5 7 8

9 35 46 49 59 61 67 88 94

CM CM CM RE]. CHD CHD ADRlA CHD VD CM CM CHD CHD CM

F

70 1167 9 1

no no

na

no no no na no no yes na no no yes

8

56

106

189 267 9 11 118

153 10

21

26 19 19 28 42 33

AV-block multiorgan failure respiratory failure graft failure right bundle branch block right bundle branch block AV-block (borderline) aeute arrhythmia AV-block graft failure ventricular arrhythmia aeute arrhythmia

Taxo

Aid, Rej. ACF Rej. Rej. Rej. Rej. graftscleros. Rej.'· Rej. Vase. rej. Vase. rej. Rej.'·

0 0

A

0

A

0 0

A

0

A

0

A

pient

0

0 0 0

A

0 0

HLA· Cellular Vase. mismatch rejecrion rejection episodes AlB DR mild moclJsev

2/2 2/1 2/2 1/2 1/2 1/2 1/1 1/2 1/2 1/1 1/2 0/2 2Il

1/1

1 1

0

2 2 2 1 1 1 2 2 1 2

0 0

1 0 0

2

1

0 0 0 0

0

1 0 0 0

2

0

1 1 0 1 2 2 1

1

1

1 0 0 0

1 1

1

1 1 1 1

*Graft failurc due to ABO-mismatch. CM = dilated cardiomyopathy, CHD = coronary hcart discasc, ADRIA = adriamycin induced cardiotoxicity, VD = valvular heart disease, Rej = acute rejection, Vasc rej = vascular rejection, TOXO = toxoplasmosis, ACF = acute cardiac failurc. Thc first nine patients were treated with Cyclosporin A (CyA) and high-dose prednisolone. Preoperatively 15 mglkg of Cy A was administered orally. The postoperative dose was adjusted to maintain a plasma concentration of 200-300 nglml. Methylprcdnisolone (1.0 g i.v.) was given intraoperatively, followed by Prednisolone at 1 mglkgld initially and then gradually rcduced to 0.2 mglkgld after seven weeks.

The last 91 patients werc trcated with CyA, prednisolone and azathioprine. Prednisolone administration was initiated at a lower dose, 0.2 mglkgld and gradually lowered to 0.1 mglkgld. Methylprednisolonc was givcn intraopcrativcly and also on thc first postoperative day (125 mg X 3 i.v.). Additionally, this group was treated with azathioprine 4 mglkg preoperatively, followed by 2 mglkgld posttransplant. The CyA whole-blood concentra-

Table 2. Histopathological findings in thc conduction tissue from 14 human cardiac allografts TX. No.

Rec.

Donor

Rec.

Donor

SInUS

SInUS

SInUS

SInUS

node

node

nodal artery

nodal artery

2 4 5

F+

F+ CR++

8

F++

F+

CVR++

CR+

AVR++

r-

9 35

AS+ AS+

AS+ AS+

46 49

F+++

59"61

F+

67

F+

CR+

88

F+

CR+

94':·

AVR+ CVR+++ AS+++ AVR+ CVR+ AVR+ CVR++

Atrio-ventr. Atrioconductive ventr. tissue artery

F(+) CR++ necrotic CR+ F(+) CR+++ CR++ CR+ F+

CR++ CR++ CR++

AS+ PAP+ AVR+

AVR+ AVR++ CVR++ AVR++ CVR++ AVR+ CVR++ AVR+ AVR++ CVR++ AVR+++ CVR++ AVR+ CVR++ AVR+

Coronary arteries

Last EMB

epicard.

days before death

intram.

CVR++ CVR+++ 11 CVR++ PAP++ 3 AVR+ AVR+ no biopsy AS+ performed AS++ AS+ 6 AVR+ CR++ CVR++ CVR++ AVR+ AVR+

2 1

Myocard

result CR(+) CR+++ CR(+) CR+++ CR+ CR(+) AVR CR(+) CR(+)

CVR++

CVR++

1

AVR+ AVR++

AVR++ AVR++

2 0

AVR CR+ AVR CR+

AVR+ AVR+ 3 CVR++ CVR+++ AVR+ AVR+ 10 CVR++ CVR+++ AVR+ 2

CR+

" Retransplantation - -F = fibrosis, CR = cellular rejection, CVR = chronic vascular rejection, AS acute vascular rcjection, EMB = endomyocardial biopsy, - = normal.

CR+ CR+

= arteriosclerosis, AVR =

The Conduction System in Human Cardiac Allografts . 785 Table 3. Antisera for immunohistochemical analysis of the conductive tissue in human cardiac allografts Designation

Type of re agent

Chief reactivity

Source

Working dilution

Anti-human CD 3

Polyclonal rabbit antiserum (IgG fraction)

T cells

1 : 200

Dacoparts

Antihuman CD 45 R

Monoclonal mouse antibody (supernatant)

B cells

1:

80

Dacoparts

MTl

Monoclonal mouse antibody (IgG, fraction)

T cells

I:

80

Eurodiagnostics

MBl

Monoclonal mouse antibody (IgG, fraction)

B cells

1:

40

Eurodiagnostics

MAC 387

Monoclonal mouse antibody (ascites)

Macrofagcs

1: 2400

tion, which was monitored by the specific monoclonal antibody RlA test, was kept at 200-300 ng/ml during the first month and then gradually reduced to 100-200 ng/ml. The immunosuppressive regime was monitored by frequent endomyocardial biopsies (EMB) from the right ventricle, which were obtained by percutaneous internal jugular venous approachs. Additional specimcns were taken whenever rejection was clinically suspected and also in the follow-up of treated events of rejection. Acute rejection episodes were treated by 1.5 mglkg prednisolone orally or methylprednisolone 1.0 g i.v. for three days. In addition, rabbit antithymocyte globulin was given in doses of 2.5 mg/kgld i.v. for three days when rejection was severe or ongomg. Twelve grafts were studied at autopsy and two after retransplantation. Blocks for histological examination were made after fixation in 4% phosphate-buffcrcd formaldehyde solution, using a modified technique described by Davies 7 , and then processed for embedding in paraffin and serially sectioned at 6 f.lm. Every fifth, sixth and sevcnth section was mounted and stained with haema-

Dacoparts (courtesy D. Joncs)

toxylin-azophloxin-saffron, by the van Gieson technique and with methyl-green pyronine (MGP). It was possible to examine the recipient sinus node (RSN) in 10 grafts, the donor node (DSN) in 13 and the atrioventricular node (A VN), the bundle of His and the proximal parts of the bundle branches in all 14 grafts (Table 2). Immunohistochemical examination was performed in 14 cases. Subsequent sections were trypsinised for 5-10 min bcfore staining. Sampies stained for B cells were not trypsinised. The cells infiltrating the conductive tissue, myocardium and the vessel walls were analysed with different mono- and polyclonal antisera against T and B cells as weil as macrophages (Table 3). Incubation with antisera was carried out ovcrnight at room temperature. Scctions were then washed ami stained according to thc AP AAP immunoalkaline phosphatase procedurc 6,9. Thc amount of infiltrating cells was arbitrarily graded from 0 to +++ (Tablc 4). Dcpositions of immunoglobulins, complement and fibrin in the vessel walls and endothelial lining were studied in a Leitz Orthoplan microscope equipped with a Ploem.-type vertical

Table 4. Immunohistochemical findings in human allograft rejection TX. No.

T cells

B cells

Macrophages

IgM

5

Cell++ Vasc+

+

+ +

+

8

Ccll++ Vasc+

+

+ ++

9

Cell++ Vasc+

+

++ +

35

Cell+ Vasc++

+ +

46

Cell+ Vasc++

59

Cell+ Vasc++

61 67

IgG

IgA

C3

+

+

+

+

+ ++

+

+

+

+

+ +

+

+

+

+

+ ++

+

CellVasc++

+

+

Ccll+ Vasc++

+

88

Cell+ Vasc++

94

CcllVasc+

Cell

+

= cellular rejection, Va sc = vascular rejcction.

Fibrin

+

+

+

+

+

+

+ ++

+

+

+

+

+ +

+

+

+

+

+

+

786 . A. Foerster illuminator which allowed selective observation of green and red fluorescence. Sampies were trypsinised for 30 min and incubation was performed at room temperature overnight. Sections stained with monoclonal rhodamine-conj ugated rabbit anti-human immunoglobulin IgA (TRITC), fluorescein-conjugated rabbit anti-human IgM and IgG (FlTC)" were investigated. Additional specimens were lIsed tO examine purified complement factor (C 3) and fibrin by a direct immunofllloresce nce method. The results of the immllnohistochemical analyses are shown in Table 4.

Results Acute Cellular Rejection In patients who died from acute cellular rejection, neither the grade of infiltration of the conductive tissue nor the type of lymphoid cells differed significantly from the myocardial involvement elsewhere in the interventricular septum (T able 2). Cellular infiltrates were ne ver seen in the RSN. Lymphocytes were observed in the DSN and always in combination with rejection infiltrates in other parts of the conductive tissue in four out of 13 patients, mainly within the AVN but also in the myocardium. The AVN was involved in acute cellular rejection in 8 out of 14 recipients. The myocardium also showed various grades of lymphocytic infiltration. In three cases acute cellular rejection was found only in the myocardium. One patient (No. 9"}showed a unique cause for rejection. He suffered from ischemic heart disease (IHD) and received a transplant from a 37-year-old male. Early postoperatively he experienced one episode of mild and two of moderate acute cellular rejection which responded to enhanced immunosuppressive treatment. Fifty-seven days posttransplant he developed progressing right bundle branch block and died suddenly the day after. Multiple blocks of myocardial üssue revealed only small clusters of inactive lymphocytes which did not stain for MGP. The caudal part of the atrial septum, however, as weil as the AVN and the right bundle branch were heavily infiltrated with MGP-positive large lymphocytes (Fig. 1). Immunohistochemical studies disclosed mainly T-Iymphocytes and various numbers of macrophages and Blymphocytes. The infiltrations were arbitrarily graded from 0 to +++, as shown in Table 4. The infiltrates affected predominantly the transitional fibres within the cranial part of the AVN. Lymphocytes and macrophages were almost absent in the penetrating bundle of His. Extensive myocytolysis of the Purkinje fibres was present with interstitial edema and some focal hemorrhagic areas. Vascular Rejection The vascular damage caused by the rejection process produced a spectrum ofhistopathological findings varying from acute vascular rejection with fibrinoid necrosis to chronic arteriosclerotic lesions. x·

From DACO-Immunoglobu lin NS.

Fig. 1. Atrioventricular node (between the arrows) infiltrated by lymphocytes. Transitional fibres lie between arrows and th e annulus fibrosis below. Haematoxylin-azophloxin-saffron. Original magnification 312.5 x.

The donor sinus nodal artery (DSNA) showed a mild to moderate lymphocytic infiltration, mainly subendothelially and to a lesser extent within the media in four cases (Table 2). Evidence of chronic vascular damage, such as subintimal fibrosis and foamy cells, was mild in one, moderate in two and severe with stenotic luminal obstruction in one case. Acute and chronic vascular rejection were observed in combination in three cases. The atrioventricular artery (AVA) was involved in all but two patients. In case No. 67 a massive lymphocytic infiltrate was found within vessel walls, mostly in the subendothelial space but also in the tunica media, with sectorial and segmental necrosis of the smooth muscle fibres. Two patients (Nos. 35 and 46) revealed similar changes and, in the cases of Nos. 35 and 67, dense cellular infiltrates were present also within the AVN. Acute and chronic vascular rejection were combined in five patients, particularly within the intramural branches of the coronary arteries. Chronic vascular damage was more pronounced in the AVA than in the epicardial coronary arteries. One patient (No. 4) showed an unusual form of chronic vascular

The Conduction System in Human Cardiac A1lografts . 787

rejeetion in which the intramural arteries and the AVA both revealed a proliferative arteriopathy consisting of a concentric and symmetrie increase of smooth muscle cells combined with an intimal fibrocellular proliferation. This patient also had mild ordinary arteriosclerotic changes in the recipient sinus nodaLartery (RSNA) and in the DSNA. The epicardial arteries were quite severely stenosed by asymmetrie intimal fibrocellular proliferation and foamy cells. This patient died of multi-organ failure due to AIDS. Patient No. 2 showed at autopsy, 68 days posttransplant, severe damage to both the intramural and epieardial coronary arteries, whieh revealed chronic vascular rejection with asymmetrie luminal narrowing combined wirh scattered cellular infiltrates in the myocardium. This patient died of multi-organ failure after treatment for toxoplasmosis. Patient No. 35 was admitted to the intensive ca re unit beeause of acute cardiac failure two days before death, after 106 days of uneomplicated follow-up. Posttransplant EMB did not show any Iymphocytic infiltrates and there had been no signs of any ischemie myocardial damage.

Two subsequent biopsies which were performed shortly before death revealed some focal necrosis wirh a pleomorphic leucocytic infiltrate. At autopsy, however, aeute as weB as chronic vascular rejection were found segmentally in aB branches of the coronary arteries. Lymphocytic infiltrates wirhin the vessel walls and fibrinoid necrosis of the media were pronounced in the AVA (Figs. 2 and 3), while an obliterative fibrocellular proliferation was marked throughout the coronary arteries. The myocardium showed small clusters of T-eells and disseminated ischemie muscular necrosis. The T-cells were not found only in the myocardium bur infiltrated rhe AVN quire extensively. IgM and IgG rogerher wirh complement (C 3 ) were deposited along the endothelial lining of the vessel walls (Fig. 4). A communication errar resulted in transplantation of a blood group ABO-incompatible donorheart in two cases (Nos. 59 and 94, Table 1). The blood graup mismatches were detected du ring operation. Graft function was initialIy exeellent in both reeipients. Patient No. 59 was successfully retransplanted on day 9 after rapidly failing graft

Fig. 2. Atrioventricular artery showing intimal fibrocellular proliferation, foamy cells and Iymphocytic cells within the vessel wall. Haemotoxylin-azophloxin-saffron. Original magnification 312.5x.

Fig. 3. Branch of the atrioventricular artery with extensive, circumscribed fibrinoid necrosis of the vessel wall infiltrated by mononuclear cells. Haemotoxylin-asophloxin-saffron. Original magnification 312.5 x.

788 . A. Focrstcr

of mild Iymphocytic infiltrates. The AVN revealed a focal area of atrophy in the upper, periannular region. This patient had a postoperative transient sinus node dysfunetion and died of acute arrhythmia with total AV-block after experiencing only one early episode of moderate cellular rejection. Arteriosclerosis and FibrosislNecrosis

Two patients (Nos. 2 and 4) showed mild arteriosclerotic changes without foamy cells in the RSNA. Both recipients were transplanted because of dilated cardiomyopathy without signs of IHD. One of them (No. 4) also showed similar damage to the DSNA and mild fibrosis of the AVN. Five out of 10 grafts revealed mild to moderate fibrosis in the RSN, one of them (No. 8) also in the DSN. The AVN was necrotic in one case only (No. 7), probably because of acute cardiac failure. The contractile myocardium in this ca se was, however, histologically well-preserved with only scattered smalllymphocytic foei. The electrocardiogram showed no significant abnormality.

Fig.4. Immunoglobulin (IgM) deposits subendothelially. Fibrocellular stenosing proliferation and lymphocytic infiltrates with an atrioventricular macrophagcs within the proximal part artery. FITC-conjugated rabbit antihuman IgM. Original magnification 125 X.

ot

funetion in spite of repeated plasmapheresis. The second patient was retransplanted after ten days, before graft dcterioration. Both patients survived. The subsequent histopathological examination of these patients revealed only patchy Iymphocytic infiltrates in the myocardium and no damagc of the conductive tissue. However, mild lymphocytic infiltrates were visible within the AVAas weil as in the epicardial branches, but were in one ca se (No. 59) more pronoullccd in the intramural eoronary arteries, mainly subcndothelially. Immunohistochemical studies showed definite linear deposits of IgG and IgM (Fig. 5) in the endotheliallining together with complement (C3), in line with the histological findings in patient No. 59. The stainings were almost negative in patient No. 94. One patient (No. 49), who died suddenly 267 days posttransplant due to chronic vascular rejection, showed a multifocal and ischemic interstitial fibrosis due to narrowing oE the larger epi<:ardial and intramural arteries. The DSNA and AVA were severely stenosed proximally by a fibrocellular, subintimal proliferation with adventitial foci

Fig. 5. Immunoglobulin (IgM) linear deposits in the subendothelial lining of small intramural arteries and within myocytes. FITC-conjugated rabbit antihuman IgM. Original magnification 500x.

The Conduction System in Human Cardiac Allografts . 789

Discussion The present study shows that the conduction system is involved in acute as weil as in chronic rejection. lt is of interest to attempt to assess to what extent this tissue participates in the immunologic pracess. Stovin and Hewitt l1 eoncluded that the involvement in acute rejection is almost always discrete within the conduction system itself, with very little spread from it or into it fram the adjacent myocardium. They also concluded that the conductive tissue was neither selectively involved nor spared during the rejeetion process. These authors had the impression that it was not attacked as intensely as the myocardium by an acute eellular rejeetion infiltrate. They claimed that the SNA and AV-tissue was more frequently involved in acute cellular and chronic vaseular rejection than the arteries in the adjaeent myocardium. lt should, however, be emphasised that the small number of reeipients in the current and former 2 , 11 investigations does not allow significant eonclusions. Twelve out of 18 patients died in Biebers study2, mainly because of neerotising arteritis due to inadequate immunosuppression. That study revealed myocardial damage predominantly localiscd to the subendocardial region and papillary muscles, favouring the hypothesis that ischemia seeondary to vascular oeclusion or narrowing was the primary mechanism. The authors argued that early and severe involvement of the eonduetion system eould be related to decreased vascularisation~ of the AVN and conduction bundle, as weil as the subendothelial bundle branches. These pathologic changes resulted in a variety of electrophysiologic and funetional disturbances. Bieber et al.2 concluded that the particular vulnerability of the conduetion system to immunologie damage could be an organspeeific feature of cardiac rejeetion which manifested itself in a variety of arrhythmias potentially reversible by appropriate immunosuppression. They were not able to substantiate any relations hip between the number of histoeompatibility mismatehes and the severity of rejeetion. üur results differ in several ways from those of Stovin and Hewitt ll . Firstly, it was not possible to find eellular infiltrates in the eonduetion system in about one third of our eases. In 57% of the cases aetivated lymphoeytes were deteeted in the atrioventricular eonduetion (AV-) system, but not in the sinus node (SN). The SN was not infiltrated unless the AV-system was compromised as weil. üne patient (No. 9) showed a speeifie and nearly seleetive involvement of the AV-system without any substantial signs of rejeetion in the eontraetile myoeardium. This unique ease suggests that negative EMB in eombination with arrhythmias or bundle bloek may call for inereased immunosuppression, implantation of a pacemaker, or both. The vascular involvement in our series appears to be similar to the findings reported by Stovin and Hewitt ll . lt was more frequent in arteries supplying the eonduetive tissue than in the larger epicardial branehes. Again, however, the series are·· too small to permit definite eonclusions, and it must be emphasised that a clear

histopathologieal evaluation to a great extent depends on the number of sections of the conduction system which are examined. Surprisingly, Stovin and Hewitt ll observed lymphoeytie aggregates in the RSN. In spite of numerous sections through reeipient and donor tissue, we were not able to detect rejection infiltrates of activated T-eells in this node. lt is difficult to explain the possibility of donor lymphoeyte reaetivity against "self" unless an autoimmune reaetion has taken plaee. The possibility of damage to the DSN and RSN by surgieal proeedures eaused by interruption of the arteries at the atrial suture line is obvious. In half of the eases the RSN was normal, while in the remaining it was more or less fibrotie, probably beeause of damage to its vaseular supply. In only three DSN was it possible to find morphologie ehanges sueh as atrophy, dystrophie ealcification and fibrosis. In one ease only (No. 49), the DSN was eompletely destroyed and fibrotic. It was diffieult to assess whether the DSN fibrosis had been secondary to surgieal trauma or to rejection. In our experience the atherosclerosing process is more pronounced in the epicardial and the larger intramural branches of the coronary arteries, while concentrie proliferative arteriopathy is more common in the intramural arteries and arterioles. The distribution and degree of these two main arte rial changes are at varianee with the findings of Pueei et al. 10. They investigated long-term eardiac allografts and found that all their six long-surviving grafts showed occlusive coronary disease resembling ordinary atherosclerosis, while four two-year survivors revealed the more characteristic concentric eoronary disease. The conduetion system was not examined. Immunohistochemieal analysis showed that the cellular infiltrates in active rejeetion were dominated by T-cells. Based on studies on frozen sections, it was shown that the prevalent lymphoeytes in ongoing or lethai rejection were CD 8 positive (eytotoxic) T-cells. The formalin-induced cross-bindings between proteins are difficult to destabilise with proteolytic enzymes, and the variation in fixation time may be one explanation for the unequal distribution of cells. Linear deposits of immunoglobulin along the endothelial lining may represent a humoral mechanism due to immunologie allograft damage to the vascular endothelium 3 . Small foci of aetivated lymphoeytes within the conduetion tissue may probably disappear as they do elsewhere in the myocardium. Pronouneed forms of allograft rejection, including myoeyte necrosis or myocytolysis, are in most eases also convertible by inereased immunosuppression. However, as EMB and autopsy studies have revealed, myocytie damage results in focal scars and dystrophie calcifications. There is reason to believe that moderate or severe forms of cardiac allograft rejection within the conduction tissue may lead to cardiac arrhythmias, whieh in term may produce sudden death. Eight out of 22 patients died suddenly with electrophysiological disturbances most probably due to damage to the conduction system. However, there are indieations that most patients with rejection infiltrates in the eonduction system recover without lasting

790 . A. Foerster

symptoms. Some patients show no significant electrocardiographic disturbance in spite of extensive lymphocytic infiltrates in the conduction tissue. This suggests that the conduction system has a certain reserve capacity. A previous study of risk factors leading to mortality in humanheart transplantation 8 showed a significant correlation between histocompatibility mismatching in the HLA-DR locus and lethai rejection. The presence of two HLA-DR mismatches between donor and recipient was found to be an independent risk factor for lethai rejection (relative risk 3.3) with a significant correlation by univariate analysis (p < 0.02). This study also shows a significant relationship between histocompatibility mismatch and acute cellular rejection episodes. This also seems to be the case in the cardiac conductive tissue system.

Acknowledgements The author would particularly like to thank the Laboratory for Immunohistochemistry and Immunopathology (LIIP AT) at the University of Oslo, for providing the anti sera and carrying out the immunohistochemical stainings. I also thank Svein Simonsen, MD, at the Medical Department B for his contribution of the clinical findings.

Received April 4, 1991 . Accepted July 4, 1991

References 1 Bieber CP, Stinson EB, Shumway NE (1969) Pathology of the conduction system in cardiac rejection. Circulation 39: 567-575 2 Bieber CP, Stinson EB, Shumway NE, Payne R, Kosek J (1970) Cardiac transplantation in man. VII. Cardiac allograft pathology. Circulation 41: 753-771 3 Billingham ME (1987) Cardiac transplant atherosclerosis. Trans plant Proc 4 Supp 5: 19-25 4 Caves PK, Stinson EB, Billingham ME, Rider AK, Shumway NE (1973) Diagnosis oE human cardiac rejection by serial cardiac biopsy. J Thorac Cardiovasc Surg 66: 461-466 5 Caves PK, Stinson EB, Graham AF, Billingham ME, Grehl TM, Shumway NE (1973) Percutaneous transvenous endomyocardial biopsy. JAMA 225: 288-291 6 CordeIl JL, Falini B, Erber WN, Ghosh AK, Abdulaziz Z, MacDonald S, Pulford K, Stein H, Mason D (1984) Immunoenzymatic labelling of monoclonal antibodies using immune complexes of alkaline phosphatase and monoclonal anti-alkaline phosphatase (APAAP-complexes). J Histochem Cytochem 32: 219-229 7 Davis MJ. Pathology of conducting tissue of the heart (1971) Butterworths, 29-37, London 8 Foerster A, Abdelnoor M, Ff0ysaker T, Geiran 0, Lindberg H, 0vrum E, Pfeffer P, Thorsby E (1988) Human heart transplantation. Rejection risk factors. APMIS Suppl 2: 160-173 9 Masson DY, CordeIl J, Brown M, Pa liesen G, Ki::er ER, Rothbard J, Crumpton M, Gatter KC (1989) Detection of T cells in paraffin wax-embedded tissue using antibodies against a peptide sequence from the CD 3 antigen. J Clin Pathol 42: 1194-1200 10 Pucci AM, Forbes CRD, Billingham ME (1990) Pathologic features in long-term cardiac allografts. J Heart Transplantation 9: 339-345 11 Stovin PGI, Hewitt S (1986) Conduction tissue in the transplanted human heart. J Pathol149: 183-189

Key words: Human cardiac transplantation - Conductive tissue - Transplantate rejection Arnold Foerster, Dept. of Pathology, Rikshospitalet, National Hospital of Norway, University of Oslo, Pilestredet 32,0027 Oslo 1, Norway