- Email: [email protected]
INDUCTION OF HIGH KIDNEY GRAFT SURVIVAL RATE BY MULTIPLE TRANSFUSION
Saturday 6 June
INDUCTION OF HIGH KIDNEY GRAFT SURVIVAL RATE BY MULTIPLE TRANSFUSION BEVERLY GRAVER GERHARD OPELZ PAUL I. TERASAKI
Department of Surgery, UCLA School of Medicine, Los Angeles,
California,
U.S.A.
that adopted a policy of liberal blood transfusion 174 recipients preoperative of cadaver donor transplants were followed up. Those who received no pretransplant transfusions had a 23% 1-year graft survival rate whereas those who had >10 transfusions had an 87% survival rate (p<0.000001). 3% of those transfused reactive antibodies and >70% had acquired highly cytotoxic no antibodies, so transfusions do not exert their beneficial effect by excluding strong immune responders. Since transplant survival rates improved with the number of transfusions, they probably produce their beneficial effect by inducing a state of unresponsiveness. Multiple transfusion seems a simple and effective way of improving kidney transplant survival rates. Moreover, the immunosuppressive effect is specific in that it does not alter the immune response to infectious agents.
Summary
In 33
centres
Introduction THE paradoxical beneficial effect of pretransplant blood transfusion on the outcome of kidney graftshas now been confirmed by more than fifty independent studies. Perhaps the most important question to resolve is whether transfusion produces better survival rates by inducing a state of immune
mresponsiveness or by selecting out highly immune :esponsive patients. 1,2 The basis for the latter alternative is ’1M blood transfusion causes the production of cytotoxic antibodies in responsive patients and therefore donors with tigens against which the patients responded are eliminated 3 the crossmatch test. We provide evidence that transfusion unresponsiveness in patients. Patients were liberally transfused preoperatively and carefully monitored for the velopment of cytotoxic antibodies to determine how many clients were rendered unsuitable for transplantation by
aduces
"nsfusion. We also monitored the effect of numbers of ansfusions on graft survival.
1981
Patients and Methods The Prospective Transfusion Study began in February, 1979. Two previous attempts at testing the effectiveness of deliberate pretransplant transfusion failed because potential participants could not agree on which transfusion protocol to follow. Therefore a strict transfusion protocol was not part of the present study; instead, emphasis was placed on close monitoring of all transfusions, and the only stipulation was that participating centres agreed to transfuse
liberally. Patients qualified for this study either if they had received no transfusions at all, or if they had received less than 3 units of blood, all in the year before entry and properly recorded. A computerised system was developed at the UCLA Tissue Typing Laboratory for data collection, storage, communication with participating centres, and analysis. Every month records were updated and returned to the centres for verification-they include date, amount and type of blood product transfused, and results of periodic screenings for lymphocytotoxic antibodies. Only first transplants from cadaver donors were included in this analysis. They were done between February, 1979, and October, 1980, at 33 transplant centres in five countries (Austria, Canada, France, Sweden, and the United States). Minimum clinical followup was 3 months. Graft survival rates were computed by actuarial methods.Statistical significance was calculated by weighted regression4 or Student’s t test as indicated. Technical or nonimmunological failures were not excluded, and patient death was counted as graft failure.
Results
331 patients qualified for the study, and 174 have received transplants so far. With up to 20 transfusions given, 3% of patients produced antibodies that reacted against >90% of the random panel, 807o formed antibodies against 50% - 90% of the panel, 1707o formed antibodies against 1070 - 500/0 of the panel, and 72% did not have detectable lymphocytotoxins. A detailed report of antibody studies in these patients is being 5 published elsewhere. Despite their enrolment in the study, 22 patients received transplants early, before being transfused. 133 patients received only packed cells, 5 patients received packed cells and whole blood transfusions, and 14 patients received frozen blood (3 frozen blood exclusively). Graft survival in relation to units of blood transfused is shown in the accompanying 8232
1224 TABLE II-EFFECT OF TRANSFUSIONS GIVEN DURING TR41SPL.:".
SURGERY
Results
are
given as means + SE.
Discussion kidney graft survival rates according to the number of transfusions received before transplantation. Numbers of transfusions are indicated at end of each curve and numbers of patients are given in parentheses. Numbers of graft survivals for each group at 6 mo are as indicated. Number of patients at risk at 6 mo are indicated. p (by weighted regression analysis’) was <0-0001 at 3, 6, and 12 months indicating that the improvement in graft outcome was dependent on an increased number of pre transplant transfusions.
Actuarial
There was a striking dose-related increase in graft survival from 23±9% at 1 year in non-transfused patients to. 87±6% in patients with >10 transfusions (p<0-000001 by Student’s t test). Only 3 patients received > 15 units of blood. When only non-frozen blood was considered (because frozen blood was found to be less effective in previous studies6,7) there was some further improvement in the correlation: the 29 patients with >10 transfusions of nonfrozen blood had an 89±6% graft survival rate at 1 year, compared with 66±7% among 45 patients with 5-10 transfusions, 59±7% among 75 patients with 1-4 transfusions, and 20:t80/0 among 25 patients without transfusions (p regression <0-0001). Of the 174 transplant recipients, 140 showed no cytotoxic antibody reactivity in their sera at any time before transplantation. Their graft survival rates correlated strongly with the number of transfusions (table 1). 21 patients had
figure.
TABLE I-TRANSFUSION EFFECT IN PATIENTS WITHOUT CYTOTOXIC ANTIBODIES
I
p
I
(regression) <0-0001.
cytotoxic antibodies against 10-50% of the panel and 13 patients had antibodies against >50070 of the panel; their respective graft survival rates were 50±1207o and 62±14°7o at 1 year. The numbers of patients in these subsets were too small for a meaningful analysis by number of transfusions. The number of transfusions during surgery did not show a significant association with graft survival (table II). The distribution of recipient’s sex, age, original disease leading to renal failure, duration of pretransplant dialysis, interval between last transfusion and transplantation, and HLA match did not differ among the different transfusion categories, and none of these factors influenced our findings
significantly.
The results of our prospective trial confirm those of man; previous retrospective studies indicating that transfusions improve the transplant survival rate.’ In fact, in the current study, an even greater effect of pretransplant transfusion is noted: with no transfusions the 1-year survival rate was 23%; compared with 87% with >10 transfusions. Apart from confirmation of retrospective studies, this prospective trial is important in answering certain questions that could not be clearly resolved from retrospective data.
First, it is now clear that transfusion does not exert its effect by selection but by induction of unresponsiveness. Out of a group of 331 patients, only 3% had highly reactive cytotoxic antibodies and more than 7007o remained entirely free of antibodies. The graft survival rate in patients with cytotoxic antibodies indicates that they had not been compromised by transfusion. The remaining patients who did not have antibodies showed a striking improvement in their survival rates with increasing numbers of transfusions, which suggests a dose-dependent induction effect. A larger proportion of patients with cytotoxic antibodies is necessary for improved graft outcome in transfused patients to be attributed to selection. Considered in terms of induction of unresponsiveness, multiple stimuli seem to be better than one. It thus follows that transfusion at the time of surgery would not be effective The results from this trial confirmed the data obtained in the North American cooperative study8 and conflict with those from two centres.9,10 The prospective nature of this trial reduces errors in transfusion record-keeping. In retrospective studies the numbers of transfusions are probably underestimated since records are often not complete. Thus, accurate graft survival rates for patients with no transfusions are likely to be those provided by our study (23% 1-year survival), the higher rates of up to 40% in retrospective studies8,11,12 probably being an overestimation of the true rate. However, the number oi patients we had who received no transfusion was small, soou’ survival rate may be spurious. The non-transfused patienn could also have been given their transplants before transfusion because of an urgent need, in which case rheB would form a high-risk group, but none of them had bee: listed on the waiting list as "urgent". This prospective study also puts paid to the argument rh, patients with certain types of diseases are transfused mo often, and that the high survival rate of transfused patier.;may be related to the requirement for transfusion rather they to the transfusion itself.l3 Our patients had a varied : underlying renal diseases and were transfused whether or r.,’: there were "medical" reasons. In an earlier study we noted that the centre at which :h: transplant is done is one of the strongest influences c:
1225
transplantation success rates.14 It is reassuring that the transfusion effect is seen so clearly despite the number (33) of taking part in this study. Our results were not biased bv centre-e.g., the 22 non-transfused patients came from 11 transfusions different centres and the 31 patients with >10 centres
from 10 centres. The transfusion effect
seems to
be strong
enough to override the centre effect. Since transfusion is so far the strongest identified influence on cadaver transplant survival, it is important to consider transfusion data when evaluating new drugs (e.g.,
cyclosporin). Perhaps the most important feature of transfusion is its specificity in that it does not alter the immune response to infectious agents. It is also inexpensive and easily available; and the increase in haematocrit levels produced by transfusion gives patients an improved quality of life while they wait for a transplant. The findings of this trial also provide assurance that mass-sensitisation of recipients 15 does not occur after transfusion-only 307o become highly sensitised. If transfusion induces a state ofunresponsiveness, how best is the induction process achieved? Increasing the number of transfusions improves the transplant success rate and frozen blood seems to be less effective than packed cells;8 but we do not know whether fresh blood is better than banked blood, which component of the blood is the best inducer, and what is the best volume to be given. We hope to continue our collaborative studies to obtain the answers rapidly. Meanwhile, we suggest that a policy of conservative transfusion, or of transfusion only at the time of transplantation, is unwarranted. We thank the following dialysis and transplant physicians and their staffs for providing the transfusion and transplant data on which this analysis was based:
G.
Beathard, Austin, Texas; T. Nebout, Creteil, France; R. Bull and R. Easterling, East Lansing, Michigan; J. Dossetor, Edmonton, Canada; R. Schweizer, Hartford, Connecticut; D. Nelson, Houston, Texas; M. .Uargreiter, Innsbruck, Austria; W. Flanigan, Little Rock, Arkansas; R. . Mson, Lackland, Texas; R. Johnson, B. Lockwood, and C. Stiller, London, Canada; A. Abukurah, T. Berne, A. Dauer, A. Dumke, A. Erlbaum, D. Gentile, 0. Haig, P. Kinion, D. Lee, L. Lewin, D. Martin, R. Mendez, B. tlurray, P. Parsa, C. Reddy, H. Rodiles, M. Rosenblatt, B. Salick, B. Self, L. Soberon, R. Soderblom and K. Titlow, Los Angeles, California; S. Acchiardo, L Bntt, G. Howell, and R. Shane, Memphis, Tennessee; P. Barre, M. Davidman, R. Guttmann, and D. Hollomby, Montreal, Canada; R. Steenberg, Omaha, Nebraska; B. van der Werf, Phoenix, Arizona; S. Chatterjee and D. Johnson, Sacramento, California; C. Capp, G. Herron, B. Levin, and P. Weber, San Francisco, California; R. Baltzan, Saskatoon, Canada; P. llontambault, Sherbrooke, Canada; I. Fehrman, Stockholm, Sweden; M. Sloan, Tampa, Florida; B. Harwell and R. Medlock, Tulsa, Oklahoma; C. Reeve, Vancouver, Canada; J. Light, Washington, D.C.; and J. Jeffrey,
TRIAL OF HEPARIN VERSUS ATENOLOL IN PREVENTION OF MYOCARDIAL INFARCTION IN INTERMEDIATE CORONARY SYNDROME ANNE M. TELFORD
CHARLES WILSON
Cardiac Unit, Waveney Hospital, Ballymena, Co. Antrim, Northern Ireland
Summary
A randomised, trolled study of
double-blind, placebo conmorbidity and mortality was carried out using heparin, atenolol, and a combination of both drugs, in 214 patients with the intermediate coronary syndrome. During the trial period, transmural myocardial infarction developed in 9 (17%) out of 54 patients on placebo, 8 (13%) out of 60 on atenolol, 1 (2%) out of 51 on heparin, and 2 (4%) out of 49 on heparin and atenolol combined (p = 0.024). The improved prognosis in the heparin-treated
patients was maintained at follow-up. All five deaths occurred among patients who did not receive heparin. These results show that intravenous heparin therapy was of benefit in preventing myocardial infarction in patients with the intermediate coronary syndrome. Introduction DURING the days or weeks preceding myocardial infarction 27-67°70 of patients have one or more episodes of ischaemic chest pain without significant infarction.’’"The electrocardiogram (ECG) usually shows ST depression with or without T wave changes, often associated with small increases in serum cardiac enzyme levels. Patients presenting with this syndrome have been studied with varying results.6,12-19 However, when neither anticoagulants nor
beta-adrenergic blocking drugs were used, myocardial infarction rates of 22% and 49070, and mortality rates of 16% and 24% within three months have been reported.6,14 Thus, these patients constitute a high risk group who should be identified as having an intermediate coronary syndrome, so that therapeutic measures may be directed towards the prevention of myocardial infarction and possibly death. Several uncontrolled or inadequately controlled studies have suggested that anticoagulation may improve the prognosisbnz-14 but anticoagulants have been recommended less frequently since the introduction of beta-adrenergic blocking drugs, the benefits of which are well accepted in stable angina. A few reports20,21have suggested a beneficial effect in
Winnipeg, Canada. Requests for reprints should be addressed to G. 0., Department of Surgery, !-’CLA School of Medicine, Los Angeles, California 90024, U.S.A. REFERENCES
Opelz G, Mickey MR, Sengar DPS, Terasaki PI. Effect of blood transfusions on subsequent kidney transplants Transplant Proc 1973; 5: 253-59. Opelz G, Mickey MR, Terasaki PI. Identification of unresponsive kidney transplant recipients Lancet 1972; i 868-70.
Dixon WJ (ed) Biomedical Computer Programs. Berkeley, California, University of California Press, 1974. 4 Dunn OJ, Clark VA (eds) Applied Statistics Analysis of variance and regression. New
York: Wiley, 1974 5 Opeiz G, Graver B, Mickey MR, Terasaki PI. Lymphocytotoxic antibody responses to transfusions in potential kidney transplant recipients. Transplantation (in press). Opelz G, Terasaki PI. Poor kidney transplant survival in recipients with frozen blood transfusions or no transfusions Lancet 1974; ii: 696-98. Es AA, Balner H. Effect of pretransplant transfusions on kidney allograft survival.
:an
Transplant Proc 1979, 11: 127-37. Opelz G, Terasaki PI. Dominant effect Transplantation 1980, 29: 153-58.
of transfusions
on
kidney graft
survival
9. Williams KA, Ting A, French ME, Oliver D, Morris PJ. Peroperative bloodtransfusions improve cadaveric renal-allograft survival in non-transfused recipients. Lancet 1980; i. 1104-06. 10. Hunsicker LG, Oei LS, Freeman RM, Thompson JS, Corry RJ. Effect of blood transfusions on cadaver renal allograft survival. Transplant Proc 1979; 11: 156-59. 11. Festenstein H, Sachs JA, Paris AMI, Pegrum GD, Moorhead JF Influence of HLA matching and blood-transfusion on outcome of 502 London transplant group renalgraft recipients. Lancet 1976; i: 157-61. 12. Solheim BG, Flatmark A, Jervell J, Arnesen E. Influence of blood transfusions on kidney transplant and uremic patient survival Scand J Urol Nephrol 1977; 42 (suppl): 65-69. 13. Guttmann RD. Interrelationship of time on dialysis-dependent uremia and pretransplant blood transfusions. Nephron 1978; 22: 196-99. 14. Opelz G, Mickey MR, Terasaki PI. Comparison of kidney transplant survival among transplant centers Transplantation 1975; 19: 226-29. 15. Solheim BG. The role of pretransplant blood transfusions Transplant Proc 1979; 11: 138-44 16. Persijn GG, van HooffJP, Kalff MW, Lansbergen Q, van Rood JJ. Effect of blood transfusions and HLA matching on renal transplantation in The Netherlands. Transplant Proc 1977, 9: 503-05. 17. Opelz G, Terasaki PI. International histocompatibility workshop study on renal transplantation. Terasaki PI, ed. In: Histocompatibility testing 1980, Los Angeles, California. UCLA Tissue Typing Laboratory, 592-624.
INDUCTION OF HIGH KIDNEY GRAFT SURVIVAL RATE BY MULTIPLE TRANSFUSION BEVERLY GRAVER GERHARD OPELZ PAUL I. TERASAKI
Department of Surgery, UCLA School of Medicine, Los Angeles,
California,
U.S.A.
that adopted a policy of liberal blood transfusion 174 recipients preoperative of cadaver donor transplants were followed up. Those who received no pretransplant transfusions had a 23% 1-year graft survival rate whereas those who had >10 transfusions had an 87% survival rate (p<0.000001). 3% of those transfused reactive antibodies and >70% had acquired highly cytotoxic no antibodies, so transfusions do not exert their beneficial effect by excluding strong immune responders. Since transplant survival rates improved with the number of transfusions, they probably produce their beneficial effect by inducing a state of unresponsiveness. Multiple transfusion seems a simple and effective way of improving kidney transplant survival rates. Moreover, the immunosuppressive effect is specific in that it does not alter the immune response to infectious agents.
Summary
In 33
centres
Introduction THE paradoxical beneficial effect of pretransplant blood transfusion on the outcome of kidney graftshas now been confirmed by more than fifty independent studies. Perhaps the most important question to resolve is whether transfusion produces better survival rates by inducing a state of immune
mresponsiveness or by selecting out highly immune :esponsive patients. 1,2 The basis for the latter alternative is ’1M blood transfusion causes the production of cytotoxic antibodies in responsive patients and therefore donors with tigens against which the patients responded are eliminated 3 the crossmatch test. We provide evidence that transfusion unresponsiveness in patients. Patients were liberally transfused preoperatively and carefully monitored for the velopment of cytotoxic antibodies to determine how many clients were rendered unsuitable for transplantation by
aduces
"nsfusion. We also monitored the effect of numbers of ansfusions on graft survival.
1981
Patients and Methods The Prospective Transfusion Study began in February, 1979. Two previous attempts at testing the effectiveness of deliberate pretransplant transfusion failed because potential participants could not agree on which transfusion protocol to follow. Therefore a strict transfusion protocol was not part of the present study; instead, emphasis was placed on close monitoring of all transfusions, and the only stipulation was that participating centres agreed to transfuse
liberally. Patients qualified for this study either if they had received no transfusions at all, or if they had received less than 3 units of blood, all in the year before entry and properly recorded. A computerised system was developed at the UCLA Tissue Typing Laboratory for data collection, storage, communication with participating centres, and analysis. Every month records were updated and returned to the centres for verification-they include date, amount and type of blood product transfused, and results of periodic screenings for lymphocytotoxic antibodies. Only first transplants from cadaver donors were included in this analysis. They were done between February, 1979, and October, 1980, at 33 transplant centres in five countries (Austria, Canada, France, Sweden, and the United States). Minimum clinical followup was 3 months. Graft survival rates were computed by actuarial methods.Statistical significance was calculated by weighted regression4 or Student’s t test as indicated. Technical or nonimmunological failures were not excluded, and patient death was counted as graft failure.
Results
331 patients qualified for the study, and 174 have received transplants so far. With up to 20 transfusions given, 3% of patients produced antibodies that reacted against >90% of the random panel, 807o formed antibodies against 50% - 90% of the panel, 1707o formed antibodies against 1070 - 500/0 of the panel, and 72% did not have detectable lymphocytotoxins. A detailed report of antibody studies in these patients is being 5 published elsewhere. Despite their enrolment in the study, 22 patients received transplants early, before being transfused. 133 patients received only packed cells, 5 patients received packed cells and whole blood transfusions, and 14 patients received frozen blood (3 frozen blood exclusively). Graft survival in relation to units of blood transfused is shown in the accompanying 8232
1224 TABLE II-EFFECT OF TRANSFUSIONS GIVEN DURING TR41SPL.:".
SURGERY
Results
are
given as means + SE.
Discussion kidney graft survival rates according to the number of transfusions received before transplantation. Numbers of transfusions are indicated at end of each curve and numbers of patients are given in parentheses. Numbers of graft survivals for each group at 6 mo are as indicated. Number of patients at risk at 6 mo are indicated. p (by weighted regression analysis’) was <0-0001 at 3, 6, and 12 months indicating that the improvement in graft outcome was dependent on an increased number of pre transplant transfusions.
Actuarial
There was a striking dose-related increase in graft survival from 23±9% at 1 year in non-transfused patients to. 87±6% in patients with >10 transfusions (p<0-000001 by Student’s t test). Only 3 patients received > 15 units of blood. When only non-frozen blood was considered (because frozen blood was found to be less effective in previous studies6,7) there was some further improvement in the correlation: the 29 patients with >10 transfusions of nonfrozen blood had an 89±6% graft survival rate at 1 year, compared with 66±7% among 45 patients with 5-10 transfusions, 59±7% among 75 patients with 1-4 transfusions, and 20:t80/0 among 25 patients without transfusions (p regression <0-0001). Of the 174 transplant recipients, 140 showed no cytotoxic antibody reactivity in their sera at any time before transplantation. Their graft survival rates correlated strongly with the number of transfusions (table 1). 21 patients had
figure.
TABLE I-TRANSFUSION EFFECT IN PATIENTS WITHOUT CYTOTOXIC ANTIBODIES
I
p
I
(regression) <0-0001.
cytotoxic antibodies against 10-50% of the panel and 13 patients had antibodies against >50070 of the panel; their respective graft survival rates were 50±1207o and 62±14°7o at 1 year. The numbers of patients in these subsets were too small for a meaningful analysis by number of transfusions. The number of transfusions during surgery did not show a significant association with graft survival (table II). The distribution of recipient’s sex, age, original disease leading to renal failure, duration of pretransplant dialysis, interval between last transfusion and transplantation, and HLA match did not differ among the different transfusion categories, and none of these factors influenced our findings
significantly.
The results of our prospective trial confirm those of man; previous retrospective studies indicating that transfusions improve the transplant survival rate.’ In fact, in the current study, an even greater effect of pretransplant transfusion is noted: with no transfusions the 1-year survival rate was 23%; compared with 87% with >10 transfusions. Apart from confirmation of retrospective studies, this prospective trial is important in answering certain questions that could not be clearly resolved from retrospective data.
First, it is now clear that transfusion does not exert its effect by selection but by induction of unresponsiveness. Out of a group of 331 patients, only 3% had highly reactive cytotoxic antibodies and more than 7007o remained entirely free of antibodies. The graft survival rate in patients with cytotoxic antibodies indicates that they had not been compromised by transfusion. The remaining patients who did not have antibodies showed a striking improvement in their survival rates with increasing numbers of transfusions, which suggests a dose-dependent induction effect. A larger proportion of patients with cytotoxic antibodies is necessary for improved graft outcome in transfused patients to be attributed to selection. Considered in terms of induction of unresponsiveness, multiple stimuli seem to be better than one. It thus follows that transfusion at the time of surgery would not be effective The results from this trial confirmed the data obtained in the North American cooperative study8 and conflict with those from two centres.9,10 The prospective nature of this trial reduces errors in transfusion record-keeping. In retrospective studies the numbers of transfusions are probably underestimated since records are often not complete. Thus, accurate graft survival rates for patients with no transfusions are likely to be those provided by our study (23% 1-year survival), the higher rates of up to 40% in retrospective studies8,11,12 probably being an overestimation of the true rate. However, the number oi patients we had who received no transfusion was small, soou’ survival rate may be spurious. The non-transfused patienn could also have been given their transplants before transfusion because of an urgent need, in which case rheB would form a high-risk group, but none of them had bee: listed on the waiting list as "urgent". This prospective study also puts paid to the argument rh, patients with certain types of diseases are transfused mo often, and that the high survival rate of transfused patier.;may be related to the requirement for transfusion rather they to the transfusion itself.l3 Our patients had a varied : underlying renal diseases and were transfused whether or r.,’: there were "medical" reasons. In an earlier study we noted that the centre at which :h: transplant is done is one of the strongest influences c:
1225
transplantation success rates.14 It is reassuring that the transfusion effect is seen so clearly despite the number (33) of taking part in this study. Our results were not biased bv centre-e.g., the 22 non-transfused patients came from 11 transfusions different centres and the 31 patients with >10 centres
from 10 centres. The transfusion effect
seems to
be strong
enough to override the centre effect. Since transfusion is so far the strongest identified influence on cadaver transplant survival, it is important to consider transfusion data when evaluating new drugs (e.g.,
cyclosporin). Perhaps the most important feature of transfusion is its specificity in that it does not alter the immune response to infectious agents. It is also inexpensive and easily available; and the increase in haematocrit levels produced by transfusion gives patients an improved quality of life while they wait for a transplant. The findings of this trial also provide assurance that mass-sensitisation of recipients 15 does not occur after transfusion-only 307o become highly sensitised. If transfusion induces a state ofunresponsiveness, how best is the induction process achieved? Increasing the number of transfusions improves the transplant success rate and frozen blood seems to be less effective than packed cells;8 but we do not know whether fresh blood is better than banked blood, which component of the blood is the best inducer, and what is the best volume to be given. We hope to continue our collaborative studies to obtain the answers rapidly. Meanwhile, we suggest that a policy of conservative transfusion, or of transfusion only at the time of transplantation, is unwarranted. We thank the following dialysis and transplant physicians and their staffs for providing the transfusion and transplant data on which this analysis was based:
G.
Beathard, Austin, Texas; T. Nebout, Creteil, France; R. Bull and R. Easterling, East Lansing, Michigan; J. Dossetor, Edmonton, Canada; R. Schweizer, Hartford, Connecticut; D. Nelson, Houston, Texas; M. .Uargreiter, Innsbruck, Austria; W. Flanigan, Little Rock, Arkansas; R. . Mson, Lackland, Texas; R. Johnson, B. Lockwood, and C. Stiller, London, Canada; A. Abukurah, T. Berne, A. Dauer, A. Dumke, A. Erlbaum, D. Gentile, 0. Haig, P. Kinion, D. Lee, L. Lewin, D. Martin, R. Mendez, B. tlurray, P. Parsa, C. Reddy, H. Rodiles, M. Rosenblatt, B. Salick, B. Self, L. Soberon, R. Soderblom and K. Titlow, Los Angeles, California; S. Acchiardo, L Bntt, G. Howell, and R. Shane, Memphis, Tennessee; P. Barre, M. Davidman, R. Guttmann, and D. Hollomby, Montreal, Canada; R. Steenberg, Omaha, Nebraska; B. van der Werf, Phoenix, Arizona; S. Chatterjee and D. Johnson, Sacramento, California; C. Capp, G. Herron, B. Levin, and P. Weber, San Francisco, California; R. Baltzan, Saskatoon, Canada; P. llontambault, Sherbrooke, Canada; I. Fehrman, Stockholm, Sweden; M. Sloan, Tampa, Florida; B. Harwell and R. Medlock, Tulsa, Oklahoma; C. Reeve, Vancouver, Canada; J. Light, Washington, D.C.; and J. Jeffrey,
TRIAL OF HEPARIN VERSUS ATENOLOL IN PREVENTION OF MYOCARDIAL INFARCTION IN INTERMEDIATE CORONARY SYNDROME ANNE M. TELFORD
CHARLES WILSON
Cardiac Unit, Waveney Hospital, Ballymena, Co. Antrim, Northern Ireland
Summary
A randomised, trolled study of
double-blind, placebo conmorbidity and mortality was carried out using heparin, atenolol, and a combination of both drugs, in 214 patients with the intermediate coronary syndrome. During the trial period, transmural myocardial infarction developed in 9 (17%) out of 54 patients on placebo, 8 (13%) out of 60 on atenolol, 1 (2%) out of 51 on heparin, and 2 (4%) out of 49 on heparin and atenolol combined (p = 0.024). The improved prognosis in the heparin-treated
patients was maintained at follow-up. All five deaths occurred among patients who did not receive heparin. These results show that intravenous heparin therapy was of benefit in preventing myocardial infarction in patients with the intermediate coronary syndrome. Introduction DURING the days or weeks preceding myocardial infarction 27-67°70 of patients have one or more episodes of ischaemic chest pain without significant infarction.’’"The electrocardiogram (ECG) usually shows ST depression with or without T wave changes, often associated with small increases in serum cardiac enzyme levels. Patients presenting with this syndrome have been studied with varying results.6,12-19 However, when neither anticoagulants nor
beta-adrenergic blocking drugs were used, myocardial infarction rates of 22% and 49070, and mortality rates of 16% and 24% within three months have been reported.6,14 Thus, these patients constitute a high risk group who should be identified as having an intermediate coronary syndrome, so that therapeutic measures may be directed towards the prevention of myocardial infarction and possibly death. Several uncontrolled or inadequately controlled studies have suggested that anticoagulation may improve the prognosisbnz-14 but anticoagulants have been recommended less frequently since the introduction of beta-adrenergic blocking drugs, the benefits of which are well accepted in stable angina. A few reports20,21have suggested a beneficial effect in
Winnipeg, Canada. Requests for reprints should be addressed to G. 0., Department of Surgery, !-’CLA School of Medicine, Los Angeles, California 90024, U.S.A. REFERENCES
Opelz G, Mickey MR, Sengar DPS, Terasaki PI. Effect of blood transfusions on subsequent kidney transplants Transplant Proc 1973; 5: 253-59. Opelz G, Mickey MR, Terasaki PI. Identification of unresponsive kidney transplant recipients Lancet 1972; i 868-70.
Dixon WJ (ed) Biomedical Computer Programs. Berkeley, California, University of California Press, 1974. 4 Dunn OJ, Clark VA (eds) Applied Statistics Analysis of variance and regression. New
York: Wiley, 1974 5 Opeiz G, Graver B, Mickey MR, Terasaki PI. Lymphocytotoxic antibody responses to transfusions in potential kidney transplant recipients. Transplantation (in press). Opelz G, Terasaki PI. Poor kidney transplant survival in recipients with frozen blood transfusions or no transfusions Lancet 1974; ii: 696-98. Es AA, Balner H. Effect of pretransplant transfusions on kidney allograft survival.
:an
Transplant Proc 1979, 11: 127-37. Opelz G, Terasaki PI. Dominant effect Transplantation 1980, 29: 153-58.
of transfusions
on
kidney graft
survival
9. Williams KA, Ting A, French ME, Oliver D, Morris PJ. Peroperative bloodtransfusions improve cadaveric renal-allograft survival in non-transfused recipients. Lancet 1980; i. 1104-06. 10. Hunsicker LG, Oei LS, Freeman RM, Thompson JS, Corry RJ. Effect of blood transfusions on cadaver renal allograft survival. Transplant Proc 1979; 11: 156-59. 11. Festenstein H, Sachs JA, Paris AMI, Pegrum GD, Moorhead JF Influence of HLA matching and blood-transfusion on outcome of 502 London transplant group renalgraft recipients. Lancet 1976; i: 157-61. 12. Solheim BG, Flatmark A, Jervell J, Arnesen E. Influence of blood transfusions on kidney transplant and uremic patient survival Scand J Urol Nephrol 1977; 42 (suppl): 65-69. 13. Guttmann RD. Interrelationship of time on dialysis-dependent uremia and pretransplant blood transfusions. Nephron 1978; 22: 196-99. 14. Opelz G, Mickey MR, Terasaki PI. Comparison of kidney transplant survival among transplant centers Transplantation 1975; 19: 226-29. 15. Solheim BG. The role of pretransplant blood transfusions Transplant Proc 1979; 11: 138-44 16. Persijn GG, van HooffJP, Kalff MW, Lansbergen Q, van Rood JJ. Effect of blood transfusions and HLA matching on renal transplantation in The Netherlands. Transplant Proc 1977, 9: 503-05. 17. Opelz G, Terasaki PI. International histocompatibility workshop study on renal transplantation. Terasaki PI, ed. In: Histocompatibility testing 1980, Los Angeles, California. UCLA Tissue Typing Laboratory, 592-624.