Anaphylactic transfusion reactions follow the adminis - Europe PMC

1 downloads 0 Views 1MB Size Report
t6mes qui ont accompagn& ces violentes reactions a . .tabli pour 10 patients. .... hypothyroidism, and patients 6, 9 and 10 hypog.mma- globulinemia. In spite of ...
CAROL LASCHINGER, M Sc FRANCES A. SHEPHERD, MD, FRCP [C] DEREK H. NAYLOR, PH D During a 6-year period (1977 to 1982) blood samples from 152 Canadian patients were referred to the national reference laboratory of the Canadian Red Cross Society because the referring hospitals had not been able to determine the cause of the patients' severe nonhemolytic transfusion reactions. Twenty-one patients were found to be IgA deficient, and 12 of them had strong class-specific anti-IgA antibodies, which were presumed to have been responsible for the reactions. The spectrum of symptoms that accompanied these violent reactions was documented for 10 of the patients. As a probable minimum, the incidence of anti-IgA-mediated reactions averaged 1.3 per million units of blood or blood products transfused during this period. Au cours d'une periode de 6 ans (de 1977 . 1982) des pr&l.vements de sang provenant de 152 patients canadiens ont . envoy.s pour analyse au laboratoire central de la Soci.t. canadienne de Ia Croix-Rouge parce que les h6pitaux de provenance n'avaient pu determiner la cause des reactions non h.molytiques graves qu'avaient subies les patients suite . une transfusion. Une carence en IgA a .t& retrouv&e chez 21 patients, et 12 d'entre eux avaient des anticorps anti-IgA sp&cifiques de classe forts que l'on croyait responsables des r&actions. Le spectre des sympt6mes qui ont accompagn& ces violentes reactions a . .tabli pour 10 patients. L'incidence moyenne des r.actions . mediation par les anticorps anti-IgA de 1,3 par million d'unit.s de sang ou de d&iv&s sanguins transfusees au cours de cette periode represente probablement une incidence minimum.

Anaphylactic transfusion reactions follow the administration of incompatible blood or blood products and are characterized by the rapid development of fever, tachycardia, hypotension and bronchospasm.1'2 These reactions may occur within minutes of the start of a transfusion, when only a very small volume of the offending blood product has been infused. They can be life threatening but account for a very small proportion of all reactions associated with transfusion therapy. . One of the causes of an anaphylactic reaction is the combination of a patient's anti-immunoglobulin A (antiIgA) antibodies with the IgA usually present in transfused blood. Class-specific anti-IgA antibodies, which From the national reference laboratory, Canadian Red Cross Society Blood Transfusion Service, Toronto Reprint requests to: Dr. Derek H. Naylor, Canadian Red Cross Society Blood Transfusion Service, 95 Wellesley St. E, Toronto, Ont. M4Y 1H6

occur in some individuals who have either a selective deficiency of IgA or hypogammaglobulinemia, appear to be one of the causes of severe anaphylactic reactions.56 The relative rarity of anti-IgA-mediated transfusion reactions has made it difficult to determine the precise spectrum of symptoms. As well, most estimates of incidence have been derived from populations considerably smaller than 1 million. The centralized nature of the Canadian Red Cross Society Blood Transfusion Service, however, provided us with a unique opportunity to assess both symptoms and incidence in a national population. We report our observations and the results of laboratory studies for the 6-year period 1977 through 1982. Methods and materials

Between 1977 and 1982 inclusive, blood samples from 152 patients were referred to us at the national reference laboratory of the Canadian Red Cross Society because the hospital or regional laboratory had been unable to identify the cause of the patients' transfusion reactions. Most of these patients had experienced immediate nonhemolytic transfusion reactions; a few had had adverse reactions to injections of plasma fractions. We had access to the medical records of the patients whose serum we identified as having class-specific anti-IgA antibodies. IgA testing

We tested for IgA deficiency by double diffusion, reacting rabbit anti.IgAhu antiserum (Canadian Hoechst, Montreal) with the patient's undiluted serum. This procedure was capable of detecting concentrations of IgA as low as 0.035 gIL. Passive hemagglutination was used to detect anti-IgA antibodies and to determine their specificities.7 After absorption with uncoupled cells test serum was reacted at a 1:8 dilution with group 0 erythrocytes coupled to monoclonal IgA proteins: 4 IgAl, 1 IgA2m(1), 1 IgA2m(2) and normal IgA. Uncoupled cells and cells coupled to purified monoclonal 1gM, normal IgG or albumin served as controls. The purified proteins used to sensitize the erythrocytes were stored in aliquots at - 200C and were thawed immediately before use. Class-specific anti-IgA would cause the agglutination of all cells that had been coupled with any form of IgA. Anti-IgA of limited specificity was considered to be present when the test serum caused only the agglutination of cells coated with normal IgA or with one or two of the monoclonal IgA proteins, provided this agglutination could be specifically inhibited by incubation with the uncoupled protein. The immunoglobulin class of the anti-IgA was determined with dithiothreitol according to the method of Olson and coworkers.' CAN MED ASSOC J, VOL. 130, JANUARY 15, 1984

141

Table I-Clinical and laboratory observations on 10 patients* who had experienced anaphylactic transfusion reactions mediated by anti-IgA antibodies Anti-IgA antibody titre in relation to reactio4

(and interval) Patient no.

Blood product Signs and symptoms Immunoglobulin (and amount) of reactiont status

1

Packed erythrocytes (30 mL)

2

Platelet concentrate (30 mL)

3

Whole blood (5-10 mL) Whole blood (5 minutes' worth) Packed erythrocytes (100 mL) Plasma (1 unit)

4 5 6 7 8 9

10

Packed erythrocytes (100 mL) Packed erythrocytes (20 mL) Plasma (100 mL) Platelet concentrate (5-10 mL)

Blood pressure (BP) b.pulse t, diaphoresis, shaking chills, vomiting, cyanosis BP ., pulse I, vertigo, flushing, nausea, shaking chills Erythema, dizziness, choking BP b diaphoresis, tachycardia

Subsequent

Immediately Before after Subsequently

HLA antibodies

Selectively IgA deficient

NT

1:512

Selectively IgA deficient

NT

1:256

1:2000(6 wk) Strong, polyspecific

Selectively IgA deficient Selectively IgA deficient

NT

1:200

NT

1:400

1:1000 (2 wk) Weak 1:2000 (2 yr) anti-HLA-B12 1:1000(2 yr) None

Selectively IgA deficient

NT

1:1000

Erythema, dyspnea, Hypogammadisorientation globulinemic BP ., pulse f, Selectively IgA diaphoresis deficient

NT

Urticaria, Selectively IgA vomiting, deficient headache BP ., pulse t, Hypogammaflushing, globulinemic erythema, headache Swollen tongue, Hypogammaperiorbital edema, globulinemic dyspnea

Bronchospasm

NT

tolerance of blood products

Strong anti-HLA-A2

NT

-

1:64

1:1000(2 yr)

-

NT

1:64

NT

NT

NT

1:100

1:20

1:8

1:240

1:32

1:320 (5 wk) None

None

Frozen washed cells and plasma lacking IgA Frozen washed cells

1:500 (1 mo) Strong, 1:500 (5 mo) polyspecific (6 mo) None

*The two other patients found to have anti-IgA antibodies were excluded because of inadequate documentation. t = increased; = decreased. 4:NT = not tested.

Washed cells, washed random-donor platelets, platelets lacking IgA

reaction could not be identified in 9 of the 21 IgA-deficient patients. Five out of 12 samples with class-specific anti-IgA antibodies were assessed for complement fixation. In the serum of patient 2 complement was fixed with an end-point dilution of 1:4; the test results in the other four patients were uninformative because the patients' serum was anticomplementary at low dilutions. At intervals of weeks or years after the reaction the serum levels of anti-IgA antibodies were generally found to have increased (Table I). Blood samples from before the time of the transfusion reaction were available in only two cases and showed that the serum levels had been higher than at the time of the reaction. Limited-specificity anti-IgA antibodies of low titre (less than 1:20) were found in 6 of the 131 patients whose serum levels of IgA were normal but in none of the IgA-deficient patients. Of the 105 patients so tested, 42 had cytotoxic HLA antibodies, as did 4 of 8 patients who possessed classspecific anti-IgA antibodies (Table I). Each year between 1977 and 1982 an average of 1 526 277 units of blood or blood products were transfused in Canada, with about 6130 immediate nonhemolytic transfusion reactions being reported to us (4112 per million units transfused). As the only facility capable of identifying anti-IgA antibodies, we determined this to be the cause of the reaction in 12 patients (1.3 anti-IgA-mediated reactions per million units). Discussion The rarity of anti-IgA-mediated transfusion reactions has made it difficult to define the full spectrum of characteristic symptoms. Our access to the records of 12 patients provided us with a unique opportunity. The severity and explosive development of the symptoms experienced by the 10 patients for whom documentation was complete were typical of the reactions described by others."2'4'5 Certain types of symptoms, such as a drop in blood pressure and some form of respiratory

I A

distress, were particularly frequent, but no single symptom appeared in all 10 patients. The next most common manifestations (about equally frequent) were disorientation, nausea or vomiting, tachycardia or increased pulse rate, and flushing of the skin. Diaphoresis and shaking chills were observed less often. Fever and urticaria were not prominent manifestations - a finding that was in contrast to the observations in the other reports. HLA antibodies are frequently found in the serum of patients who have immediate nonhemolytic transfusion reactions," and they are capable of producing some of the symptoms we have discussed.'2 Our referral service found such antibodies in 42 (40%) of the 105 samples tested. Three of the 10 patients whose serum was found to have class-specific anti-IgA antibodies also had strong HLA antibodies. However, the severity, wide spectrum and rapid development of the symptoms seen in these three patients after only a small volume of blood had been infused indicate that HLA antibodies were probably not responsible.'2 Other possible causes of anaphylactic transfusion reactions include the presence of erythrocyte antibodies in the recipient's blood'2 and bacterial contamination of the blood being transfused. Neither of these causes was implicated in any of the 10 patients with class-specific anti-IgA antibodies. Limited-specificity anti-IgA antibodies are of uncertain clinical significance.'3 Their presence in six blood samples was, therefore, not considered to be the probable cause of the reaction. IgA deficiency has been reported to occur more frequently in patients with autoimmune disorders,'4 and 6 of the 10 patients described here had such disorders. Patient 2 had pernicious anemia, patient 4 rheumatoid arthritis, patient 5 chronic persistent hepatitis, patient 6 hypothyroidism, and patients 6, 9 and 10 hypog.mmaglobulinemia. In spite of the exacting analyses carried out, 98 of the 152 transfusion reactions we investigated remain unexplained. In almost half of these 98 cases, though, tests for HLA antibodies could not be carried out.

I

A £

Controlling inventory costs

pm

*

0 0

0

9** * a

*

: :

Overhead costs in medical offices can often be reduced by implementing controls on ordering supplies. A very simple system, but one which works well, utilizes a recipe box and cards. On each card, write the name of the article being purchased and then establish column headings from left to right as follows: date ordered, quantity ordered, name of supplier, date received and unit cost. On the bottom right corner, write the minimum re-order quantity. Using this system, you will be able to tell how long your supplies normally last, spot wastage or loss, determine re-order and delivery dates and monitor price changes. CAN MED ASSOC I, VOL. 130, JANUARY 15, 1984

143

sion reactions associated with anti-IgA antibody. N Engi J Med Frequency of anti-IgA -mediated reactions 1969; 280: 188-193 The average annual incidence of immediate transfu2. LEIKOLA J, KOISTINEN J, LEHTINEN M, VIROLAINEN M: IgAinduced anaphylactic transfusion reactions: a report of four cases. sion reactions reported to us from across Canada was Blood 1973; 42: 111-119 about 6130 per million units transfused. Other estimates 3. BJERRUM OJ, JERSILD C: Class-specific anti-IgA associated with have been 5000, 16 800 and 38 100 such reactions per severe anaphylactic transfusion reactions in a patient with pernimillion units.'5-'7 cious anaemia. Vox Sang 1971; 21: 411-424 The average annual incidence of anti-IgA-mediated 4. PINEDA AA, TA5WELL HF: Transfusion reactions associated with reactions was 1.3 per million units transfused. This anti-IgA antibodies: report of four cases and a review of the literature. Transfusion 1975; 15: 10-15 figure is much lower than the estimates of 20 to 50 5. VYAS GN, PERKINS HA, FUDENBERG HH: Anaphylactoid transreactions per million units reported by Vyas and Perfusion reactions associated with anti-IgA. Lancet 1968; 2: 312kins'8 and 40 per million units reported by Decary and 315 associates'9 in a limited study of Toronto hospitals. It is 6. VYAS GN, PERKINS HA, YANG YM, BASANTANI GK: Healthy unlikely that every immediate nonhemolytic transfusion blood donors with selective absence of immunoglobulin A: prevention of anaphylactic transfusion reactions caused by antibodies to reaction of unknown cause was brought to our attention, IgA. J Lab Gun Med 1975; 85: 838-842 even though our centralized service for the identification 7. VYAS GN, HOLMDAHL H, PERKINS HA, FUDENBERG HH: of anti-IgA antibodies is the only one available in Serologic specificity of human anti-IgA and its significance in Canada. Therefore, our figures must represent an transfusion. Blood 1969; 34: 573-581 underestimate of the number of anti-IgA-mediated 8. OLSON PR, WEIBLEN BJ, O'LEARY JJ, MOSCOWITZ AJ, MCCULtransfusion reactions occurring each year. We still feel, LOUGH J: A single technique for the inactivation of 1gM antibodies using dithiothreitol. Vox Sang 1976; 30: 149-159 though, that our results are more reliable than those 9. COLOMBANI J, D'AMORA J, GABB BW, SMITH G, SVEJGAARD A: previously available from studies of small populations. A micro-technique of platelet complement fixation with a note on The true incidence of anti-IgA-mediated reactions the micro-technique of lymphocyte complement fixation. In RAY would likely fall between our overall estimate and that IF IR, HARE BD, PEDERSON P, KAYHOE D (eds): Manual of Tissue Typing Techniques (DHEW publ no [NIH] 75-545), US of Decary and associates. Dept of Health, Education, and Welfare, National Institute of We were not able to derive any estimate for the Allergy and Infectious Diseases, transplantation and immunology national incidence of transfusion reactions mediated by branch, Bethesda, Md, 1974: 86 HLA antibodies since many hospital laboratories and 10. RAY JG, HARE DB, PEDERSON PD, MULLALLY DI (eds): NIAID research units carry out such analyses and do not report Manual of Tissue Typing Techniques (DHEW publ no [NIH] the results to us. 76-545), US Dept of Health, Education, and Welfare, National Institute of Allergy and Infectious Diseases, research resources Class-specific anti-IgA antibodies were present in branch, Bethesda, Md, 1976: 22 57% of our IgA-deficient samples. Comparable frequen- 11. PERKINS HA, PAYNE R, FERGUSON I, WooD M: Nonhemolytic cies o? antibody (44%) have been reported by others in febrile transfusion reactions. Quantitative effects of blood compoIgA-deficient hospital populations.7'20 Limited-specificity nents with emphasis on isoantigenic incompatibility of leukocytes. anti-IgA antibodies were present in 6 (4%) of the 152 Vox Sang 1966; 11:578-600 patients we tested. This figure compares well with the 12. MILLER WV, HOLLAND PV, SUGARBAKER E, STROBER W, WALDMANN TA: Anaphylactic reactions to IgA: a difficult 6.3% reported by Koistinen and Leikola2' for patients transfusion problem. Am J Clin Pathol 1970; 54: 618-621 who had experienced immediate nonhemolytic transfu- 13. RIVAT L, RIVAT C, DAvEAU M, ROPARTZ C: Comparative sion reactions. frequencies of anti-IgA antibodies among patients with anaphyBetween 1977 and 1982, 218 348 healthy blood lactic transfusion reactions and among normal blood donors. Clin Immunol Immunopathol 1977; 7: 340-348 donors were screened for IgA deficiency in order to provide blood products to patients with anti-IgA anti- 14. KOIsTINEN J, SARNA 5: Immunological abnormalities in the sera of IgA-deficient blood donors. Vox Sang 1975; 29: 203-213 bodies; 429 individuals were found to be IgA deficient. 15. KASPRISIN DO, YOGORE MG, SALMASSI 5, BOLF EC: Blood The frequency of IgA deficiency, then, was 1/509, components and transfusion reactions. Plasma Ther Transfus compared with the figures ranging from 1/396 to 1/650 Technol 1981; 2 (1): 25-29 reported by others.6'22 Of the 429 IgA-deficient donors 16. AHRONS 5, KISSMEYER-NIELSEN F: Serological investigations of 1,358 transfusion reactions in 74,000 transfusions. Dan Med Bull 71 (17%) had class-specific anti-IgA antibodies, com1968; 15: 259-262 pared with reported frequencies of 9.6% to 24.0%.6423 SV, SCHMIDT PJ, MCGINNISS MH, WORKMAN WG: Only 5 (1%) of the 429 had limited-specificity anti-IgA 17. KEVY Febrile, nonhemolytic transfusion reactions and the limited role of antibodies. Holt and coworkers23 had arrived at a figure leukoagglutinins in their etiology. Transfusion 1962; 2: 7-16 of 7%. 18. VYAS GN, PERKINS HA: Anti-IgA in blood donors (C). Transfusion 1976; 16: 289-290 Conclusion 19. DECARY F and 13 others: An investigation of non-hemolytic transfusion reactions. Vox Sang (in press) The symptoms of anaphylactic reactions must be NADORP JHS, Voss M, BUYS WC, VAN MUNSTER JJ, VAN recognized promptly so as to have the transfusion 20. TONGEREN JHM, AALBERSE RC, VAN LOGHEM E: The signifihalted. In Canada, samples of the patient's blood should cance of the presence of anti-IgA antibodies in individuals with an be referred to our central reference service for analysis IgA deficiency. Eur J Gun Invest 1973; 3: 3 17-323 once the more common causes of such reactions are 21. KOISTINEN I, LEIKOLA J: 'Weak anti-IgA antibodies with limited specificity and nonhemolytic transfusion reactions. Vox Sang ruled out. Patients whose serum contains anti-IgA 1977; 32: 77-8 1 antibodies should thereafter receive blood or blood 22. KOISTINEN J: Selective IgA deficiency in blood donors. Vox Sang products lacking IgA. 1975; 29: 192-202 23. HOLT PD, TANDY NP, ANSTEE DJ: The screening of blood donors References 1. SCHMIDT AP, TASWELL HF, GLEICH GJ: Anaphylactic transfu144

CAN MED ASSOC J, YOL. 130, JANUARY 15, 1984

for IgA deficiency: a study of the donor population of south-west England. J Clin Pathol 1977; 30: 1007-1010