POSSIBLE MECHANISM OF ACTION OF TRANSFUSION EFFECT IN RENAL TRANSPLANTATION

468

excision of a lesion rather than its simple incision could hardly be justified on treatment grounds. Recently, however, a middle-aged man, admitted routinely for haemorrhoidectomy, described the sudden development, three days before admission, ofatender hard lump at the anus separate from his permanently prolapsed pile. It was a typical "perianal haematoma". A rare opportunity for excising an entire lesion (in the course of haemorrhoidectomy) had arrived. Microscopical examination showed a clot (there was no sign of lamination to suggest thrombosis) distending a vein under the appendage-less anoderm (fig. 2).

Nothing

about

the

gross

appearance

of

"perianal

haematoma" suggests haemorrhage, and it is difficult

to

understand how either the name or the aetiological attribution originated. Without knowledge of the existence there of the venous saccules, though, perhaps no other explanation would readily suggest itself. One reason why histological elucidation is not easily

Preliminary Communication POSSIBLE MECHANISM OF ACTION OF TRANSFUSION EFFECT IN RENAL TRANSPLANTATION A. M. MACLEOD D. A. POWER R. J. MASON K. N. STEWART W. G. SHEWAN N. EDWARD G. R. D. CATTO

Department of Medicine, University of Aberdeen, and Transfusion Service, Aberdeen Royal Infirmary, Aberdeen

Blood

The mechanism by which blood transfusions given before renal transplantation improves allograft survival was studied in 31 transplant recipients. The presence of non-cytotoxic, Fc receptor blocking antibodies to donor and leukaemic B lymphocytes in pre-transplant sera correlated with both improved graft survival (p<0 ·03 and <0·1, respectively) and the number of blood transfusions given (p<0 ·05 and <0 ·03, respectively). Moreover, 6 out of 10 previously untransfused prospective transplant recipients developed these potentially protective antibodies during a course of elective blood transfusions. These results indicate that such non-cytotoxic, Fc receptor blocking antibodies in pretransplant recipient sera (a) are associated with improved allograft survival, (b) correlate with the number of blood transfusions given, and (c) can develop in response to blood transfusion.

Summary

obtained has already been given. In addition, only a small minority of lesions come under the surgeon’s knife. Most patients either do not seek treatment or are managed by their own doctor, and in many of the few who are referred to hospital resolution is well on its way by the time of out-patient attendance. The lesion is neither "perianal" (but subanodermal; the perianal zone encompasses the hairy skin beyond) nor a "haematoma". It is neither truly external (being marginally intra-anal), nor thrombosed (being a clot), nor a "pile" (a word which should be confined to the state of disruption with partial or complete descent of an anal cushion). I suggest the term "clotted venous saccule" is more apt. REFERENCES 1. Thomson H. Haemorrhoids and all that. Practitioner 1982; 226: 619-28. 2. Thomson H. The anal cushions - a fresh concept in diagnosis. Postgrad Med 55: 403-05.

J 1979,

antibodies; if these are cytotoxic to lymphocytes from donors

they can be detected by a positive crossmatch test. It has been suggested that the beneficial effect of transfusion on allograft survival is due to the ability of these complement fixing antibodies to exclude unsuitable donors by the crossmatch test. However, donor selection is an inadequate explanation for the transfusion phenomenon.The other principal theory is that transfusions exert an immunosuppressive effect of suppressor T lymphocytes’,’ or through the production 8 antibodies. enhancing Our investigation of the development of alloantibodies following blood transfusion in prospective transplant patients showed that the presence, in pre-transplant serum, of non-cytotoxic antibodies which block Fc receptor sites on B lymphocytes was associated with improved allograft survival. All patients in our previous study had received pretransplant blood transfusions. In this paper we provide evidence that these apparently beneficial antibodies can be induced in potential transplant recipients by elective blood transfusions. The Grampian area has a stable population and operates an integrated dialysis and transplant unit on the same site as the Blood Transfusion Service. It was, therefore, possible to study a group of patients with reliable blood transfusion histories and to correlate the development of non-cytotoxic, Fc receptor blocking antibodies both with allograft survival and with the number of blood transfusions given. Furthermore, 10 previously untransfused prospective transplant recipients on the chronic dialysis programme were given a course of transfusions during which they were assessed for the development of Fc receptor blocking antibodies.

PATIENTS AND METHODS

Patients

INTRODUCTION

BLOOD transfusions

given

before renal

transplantation

improve allograft survival.’-’ Since the appreciation of this fact many centres, which previously limited the number of blood transfusions given to potential transplant recipients, now transfuse these patients more liberally. The beneficial effect of transfusion is directly related to the total number of units of blood given,’ but the mechanism of action remains unclear. Blood transfusions induce complement fixing

31 recipients of cadaver donor renal transplants were studied. Recipient selection was based on blood group compatibility, a negative standard crossmatch test, and the best available HLA-A and B antigen match. Allograft survival was assessed at 1 year. Graft

failure was considered to have occurred when there was a return to maintenance dialysis therapy, graft nephrectomy, or death of a patient with or without a functioning graft. The number of pretransplant blood transfusions was obtained from records kept by the Blood Transfusion Service. None of the patients had been

469 TABLE II-ANTI-DONOR EAI AND ANTI-CLL PANEL EAI IN RELATION TO

transfused outside the Grampian region. 10 previously untransfused patients, 8 on maintenance haemodialysis and 2 on chronic ambulatory peritoneal dialysis, were given electively at least 3 blood transfusions over 3 months before being placed on the waiting list for renal transplantation. These patients were given further transfusions when therapeutically necessary.

NUMBER OF PRE-TRANSPLANT BLOOD TRANSFUSIONS I

I

I

Sera

samples (4 ml) were obtained from the 31 transplant recipients in the 12 h preceding transplantation. Serum samples (4 ml) were also taken from the 10 previously untransfused dialysis patients, before each transfusion (given at intervals of approximately 1 month) and after the last transfusion. All sera were stored at -20°C until required for assay and then thawed, heat inactivated, and ultracentrifuged at 100 000 g for 90 min. Serum

RESULTS

EAI and Renal Allograft Survival

Graft survival correlated with the presence of EAI in preserum (table 1). 91% of those with pre-transplant anti-donor EAI, and only 45% of those who did not have these antibodies, had grafts which survived 1 year (p<0 03). Similarly, 88% of those with detectable pre-transplant EAI against the CLL panel had grafts which survived 1 year, compared with only 33% among those without such activity

transplant

Target Lymphocytes The target lymphocytes used in this study were (a) specific donor B lymphocytes prepared from donor splenocytes by nylon wool column separation 10 and cryopreserved in liquid nitrogen; and (b) B lymphocytes from a panel of 6 patients with chronic lymphatic leukaemia (CLL). Surface membrane immunoglobulin studies showed that at least 90% of the leukaemic lymphocytes were B

(p<0-01). EAI and Number

of Pre- Transplant Blood Transfusions EAI activity was detected in pre-transplant sera significantly more often in patients who had received more than 5 blood transfusions (table 11). 48% of those patients who

lymphocytes. EA Rosette Inhibition Assay EAI

was

performed by

a

(EAI) modification of

a

microrosette ‘

technique." Briefly, erythrocyte antibody complexes (EAs) were made by incubating, without complement, washed ox erythrocytes with rabbit anti-ox erythrocyte antibodies. B lymphocytes were incubated with test serum, or with ultracentrifuged fetal-calf serum (FCS) as control, in the wells ofamicrotitre plate for 45 min at 4°C.

TABLE III-DEVELOPMENT OF ANTI-CLL PANEL EAI DURING A COURSE OF BLOOD TRANSFUSIONS

TABLE I-ANTI-DONOR EAI AND ANTI-CLL PANEL EAI IN RELATION TO RENAL ALLOGRAFT SURVIVAL

Results shown as number of CLL panel members in panel = 6).

against which EAI directed

(total no.

had been transfused with over 5 units of blood had pretransplant anti-donor EAI, compared with none of those who had received 5 units of blood or less (p<0’05). Of the 16 had 1 who anti-CLL had received 5 patients panel EAI, only units of blood or less, compared with 65% of those who had received more than 5 units of blood (p<0-03). The cells were then washed and incubated overnight at 40C with 2% ox EAs. The cells were resuspended and between 200 and 400 lymphocytes were counted, under simultaneous ultraviolet light and phase-contrast illumination, in a Neubaur counting chamber with an acridine-orange coated coverslip. The percentage of rosette forming cells was determined. A rosette was defined as three or more erythrocytes/lymphocyte. Control cell suspensions contained 40-60% rosetting cells. The minimum value for positive inhibition was taken as 40% of the FCS control. The percentage EAI was calculated using the formula99 % EAI = 100 f 1-

EA

% EA rosettes in test

serum

% rosettes in control FCS

]

Inhibiting Antibody

As before9the EA rosette blocking factor was shown column chromatography to be an IgG antibody.

by

DEAE

Development of EAI During Course of Transfusions The development of EA inhibiting antibodies during the course of transfusions is shown in table III. Only 1 of the 10 patients had detectable EAI against the CLL panel before transfusion. However, EA inhibiting antibodies developed in a further 50% of these patients during the transfusion programme. The EAI was directed against lymphocytes from different numbers of panel members, the numbers varying from 2 (patients 6 and 9) to 6 (patient 8). The EAI was thus selective but in no case was it directed against a panel member’s lymphocytes according to his HLA-DR tissue type. In 2 patients (6 and 9) activity was noted after the first transfusion, in 2 others (8 and 10) after the second, and in 1 (5) after the third. Patient 7, a multiparous woman, showed EAI against 2 panel cells before transfusion and against a further 2 during the transfusion programme.

Statistics DISCUSSION

Statistical analysis was performed by the use of Fisher’s exact test for fourfold tables.

The results

presented

in this paper indicate that

non-

470

cytotoxic Fc receptor blocking antibodies detected by EAI are associated with improved allograft survival, are rarely present unless the patient has been transfused with more than five units of blood, and can develop during a course of blood transfusions. These data thus suggest a method by which blood transfusions improve renal allograft survival. The immunological responses to blood transfusion have been studied by several investigators. Two studies6,7 have reported the development of suppressor T lymphocytes after blood transfusion, while others have shown the development of lymphocytotoxic antibodies. 5, 12 Such lymphocytotoxic antibodies when directed against donor peripheral blood lymphocytes in the crossmatch test are associated with early graft rejection. However, the effect on graft survival when the antibodies are directed only against donor B lymphocytes is uncertain;

some

studies

have

shown

them

to

be

beneficial,13,14 some detrimental,15 and some to have no significant influence on graft outcome.16,9 In this study we report for the first time the development of non-cytotoxic antibodies, which are detectable by EAI and which occur after transfusion and correlate with good transplant outcome. In recent years the EAI assay has been investigated in both animal models and in clinical studies. In the inbred rat renal transplant model, the presence ofEA inhibiting antibodies in recipient pre-transplant sera directed to donor B lymphocytes correlated with improved graft survival. 17 These antibodies were stimulated by the administration of donor strain lymphocytes before transplantation (active enhancement). The control, non-immunised group did not exhibit EAI and survival of their grafts was not enhanced. Results from clinical investigations are, however, conflicting. One studyl8 showed that pre-transplant EAI was associated with poor allograft outcome, whereas anotherl9 indicated a slight but not significant improvement in graft survival. Neither study provided details of patients’ transfusion histories. The present study not only shows a significant beneficial effect of pre-transplant EAI on graft outcome but also that such antibodies are present more frequently in patients given more than 5 blood transfusions. Indeed in the group given 5 blood transfusions or less no patient showed anti-donor EAI and only 1 showed EAI against the panel of leukaemic B lymphocytes. Of the 10 patients who were given a course of transfusions only 1 had these Fc receptor blocking antibodies before transfusion. She was a multiparous woman and we have found, as part of one of our current studies, that these antibodies may be present in the sera of previously untransfused women who have had one or more pregnancies. Pregnancy may therefore also stimulate the development of these antibodies. The mechanism by which antibodies block the Fc receptor sites on B lymphocytes is unclear. Immune complexes which can block Fc receptor sites non-specifically were removed by ultracentrifugation of the serum for 90 min at 100 000 g. Moreover the sera of patients given a course of transfusions reacted, in all but 1 case, with lymphocytes from only certain panel members, which makes non-specific receptor blockade unlikely. Morito et al.10 have shown that certain anti-HLA tissue typing sera inhibit rosette formation in a similarly selective manner. These results and the data presented in this paper indicate that an allelic alloantigen system other than the already defined HLA A, B, and DR antigen systems is expressed on human B lymphocytes. The EA inhibition assay may therefore detect antibodies to these alloantigens. Two studies9°ZO have shown that EAI was not directed against lymphocytes according to their tissue type and thus the Fc

are not directed against the classic HLA-A, B, or DR antigen systems. The EA rosette test has been used in an effort to determine the antigenic nature of molecules associated with Fc receptors. Initially, the EA rosette inhibition seen when lymphocytes were incubated with alloantibodies was thought to represent a specific association of MHC-coded molecules with the Fc receptor.l’,zi This has been questioned by the subsequent finding that heteroantibodies which recognise membrane structures, such as 02 microglobulin, can also cause nonspecific EA inhibition.22 Our study shows the selective inhibition of Fc binding and suggests that the EA inhibition assay may be a valuable means of detecting non-cytotoxic antibodies to alloantigens which cannot be detected by standard lymphocytotoxicity tests. The antibodies may represent an immunosuppressive humoral response to blood transfusion. How they prolong allograft survival remains to be determined.

receptor blocking antibodies

We thank the Scottish Hospitals Endowments Research Trust for financial assistance; Prof. M. MacLeod, Mr J. Engeset, and Mr D. Hamer-Hodges for permission to study patients under their care; Dr H. B. M. Lewis for allowing part of the work to be done in his department; and Mrs Marie Robertson, Mrs Shona Kidd, Mr Derry Campbell for technical.assistance. Correspondence should be addressed to A. A. MacL.

REFERENCES 1. Festenstein

H, Sachs JA, Paris AMI, Pegrum GD, Moorhead JF. Influence of HLA

and blood transfusions on outcome of 502 London Transplant Group renal-graft recipients. Lancet 1976; i: 157-61. 2. Hunsicker LG, Oei LS, Freeman RM, Thompson JS, Corry RJ Effect of blood transfusions on cadaver renal allograft survival. Transpl Proc 1979; 11: 156-59 3. van Es AA, Balner H. Effect of pre-transplant transfusions on kidney allograft survival Transpl Proc 1979; 11: 127-37. 4. Opelz G, Terasaki PI. Dominant effect of transfusions on kidney graft survival Transplantation 1980; 29: 153-58. 5. Opelz G, Graver B, Terasaki PI. Induction of high kidney graft survival rate by multiple transfusion Lancet 1981; i: 1223-25. 6. Marquet RL, Heystek GA, Borleffs JC. Suppressor cells in rats with actively induced unresponsiveness to allogeneic heart grafts. Transpl Proc 1981; 13: 589-91 T 7. Smith MD, Williams JD, Coles GA, Salaman JR. The effect of blood transfusions on suppressor cells in renal dialysis patients. Transpl Proc 1981; 13: 181-83. 8. Solheim BG. The role of pre-transplant blood transfusions. Transpl Proc 1979; 11:

matching

138-44. 9. MacLeod

AM, Mason RJ, Stewart KN, et al. The association of Fc receptor blocking antibodies and human renal transplant survival. Transplantation (in press). 10. Stewart KN. Separation of T and B lymphocytes by nylon fibre columns. Med Lab Sci 1981; 38: 123-25. JP, Carpenter CB, d’Apice AJ F, Strom TB. The role of non-classical Fc

11. Soullilou

receptor associated 143: 405-21.

Ag B antigens (Ia) in rat allograft enhancement. J Exp Med 1976,

Soullilou JP, Bignon JD, Peyrat MA, Guimbretiere J, Guenel J. Systematic transfusion in haemodialyzed patients awaiting grafts. Transplantation 1980; 30: 285-89 13. d’Apice AJF, Tait BD. Improved survival and function of renal transplants with positive B-cell crossmatches. Transplantation 1979; 27: 324-28. 14. Mohanakumar T, Giedlin M, Duval C, et al. B-lymphocyte specific antibodies in human renal allografts. Transplant Proc 1979; 11: 397-99. 15. Albrechtsen D, Arnesen E, Solheim BG, Thorsby E. Significance of HLA-DR matching and of B-cell crossmatch tests in vitro and in cadaver renal transplantation. Transpl Proc 1979; 11: 743-47. 16. Ting A, Morris PJ. Pre and post-transplant B-cell antibodies in renal transplantation Transpl Proc 1979; 9: 393-96. 17. Suthanthiran M, Catto GRD, Kaldany A, et al. Differential antibody responses to Ag-B (A region) and Ia (B region) antigens during enhancement of rat renal allografts Transplantation 1979; 28: 4-9. 18. Suthanthiran M, Gailiunas P, St Louis G, Fagan G, Carpenter CB, Garovoy MR Presensitisation to donor B cell (’Ia’) antigens is associated with early allograft failure Transplant Proc 1977; 9: 1807-09. 19. Soullilou JP, Peyrat MA, Guenel J. Studies of the antibodies against HLA, Ia-like, Fc and/or C3 receptors present in pre-transplant sera: Anti B cell antibodies not associated with accelerated graft loss. Transplant Proc 1978; 10: 475-77. 20. Morito T, Tanimoto K, Hariuchi Y, Takeo J. Fc-rosette inhibition by pregnant women’s sera and by rabbit anti-&bgr;2-microglobulin. Int Archs Allergy Appl Immunol 1978; 56: 247-55. 21. Catto GRD, Carpenter CB, Strom TB, Williams RM. Passive enhancement of rat renal allografts by antibodies to a non-SD (Ag-B) locus analogous to Ia and demonstration of linkage to the MHC. Transplant Proc 1977; 9: 957. 22. Sarmay G, Sanderson A, Ivanyi J. Modulation of Fc receptors on human peripheral blood lymphocytes by antisera against &bgr;2 micro-globulin, Ia or immunoglobulin Immunology 1979; 36: 339-44. 12.