British Jourrial of Haernatology, 1990, 75. 408-413

Transfusion of ABO-mismatched platelets leads to early platelet refractoriness ROBERTCARR,J A N E L. HUTTON,*JILL A. J E N K I N S , ~GEOFFREYF. LUCAS$ AND NIGELW . AMPHLETT$ University Department of Haematology, Royal Liverpool Hospital, *Department of Statistics and Computational Mathematics. University of Liverpool, tMersey and North Wales Regional Transfusion Centre, Liverpool, and $Blood Group Reference Laboratory, Southmead Hospital, Bristol Received 2 6 October 1989; accepted for publication 5 March 2 990

Summary.Forty-three consecutive patients previously unexposed to platelets and undergoing treatment for acute leukaemia or autografting for relapsed Hodgkin’s lymphoma were randomized to receive transfused platelets of either their own ABO group (OG) or of a major mismatched group (MMG). The 26 evaluable patients were equally distributed between the two study groups. Nine of 1 3 (69%) MMG patients became refractory with a median onset at transfusion 7 (15 d). compared with only one of 13 (8%) OG patients (P=O.OOl). Refractoriness was associated with the formation of high titre isoagglutinins, anti-HLA and platelet specific antibodies. In one patient refractoriness appeared to be due to high titre isoagglutinins alone. Six other patients developed an increase in isoaggluti-

nin titre sufficient to adversely affect platelet increments. Patients receiving ABO-mismatched platelets had a higher incidence of anti-HLA antibodies ( 5 v. 1)and platelet specific antibodies (4 v. 1). ABO-mismatched platelets transfused prior to the onset of refractoriness resulted in increments similar to those achieved by ABO-matched platelets. The study demonstrates that ABO-mismatched platelets are as effective as matched platelets in patients with low titre isoagglutinins requiring only few transfusions. However, the greater incidence of early refractoriness induced in MMG patients indicates that ABO-mismatched platelets should not be given to patients with marrow failure requiring long-term support.

The ever-increasing demand for platelets and ways to improve their delivery has been a major concern of blood transfusion centres during the past decade. Limited supplies of individual blood groups frequently necessitate transfusion of platelets across ABO barriers. Platelets carry ABH antigens on their surface: some of which are intrinsic to the platelet membrane, the remainder being adsorbed from the surrounding plasma (Dunstan et al, 1985). Studies with healthy volunteers (Aster, 1965) and alloimmunized patients receiving HLA-matched platelets (Duquesnoy et al, 1979; Tosato et al, 1978) have shown that the transfusion of ABO-mismatched platelets results in a platelet increment approximately 20% less than expected for an ABO-matched transfusion. There is no published data on the effect of repeated transfusions of ABO-mismatched platelets into non-immunized, thrombocytopenic patients. In 1984,The Mersey &North Wales Regional Transfusion Centre adopted a policy whereby platelets were routinely

issued irrespective of the ABO group of the recipient. Removing the need to keep a stock of platelet units of every ABO group constantly available reduced the wastage of timeexpired units from 19% during the 4 years prior to 1984 to 5.5% during 1985 and subsequent years. However, under this policy 50% of platelet units given to group 0 patients, for example, were ABO incompatible. A prospective randomized study was established to examine the efficacy and the consequences of giving repeated transfusions of ABO-mismatched platelets to non-immunized, thrombocytopenic patients. METHODS Study design. Between July 1987 and February 1989, 43 consecutive patients admitted for remission induction chemotherapy for newly diagnosed acute leukaemia, or for high dose chemotherapy with autologous marrow rescue (autograft) as treatment for relapsed Hodgkin’s lymphoma were recruited into the study. Patients were recruited only if they had not previously received platelet concentrates. Prior pregnancy and blood transfusion was recorded (Table I).

Correspondence:Dr R. Carr. The University Department of Haematology, Duncan Building,Royal Liverpool Hospital, Prescot Street. P.O. Box 147, Liverpool L69 3BX.

408

Transfusion of ABO-Mismatched Platelets Eligible patients were randomized (random number generation, sealed envelope) to receive platelets of their own ABO group (OG) or of a mismatched ABO group (MMG): Thus group 0 patients received exclusively group A platelets ( n = 7): group A patients, group B platelets (ti = 5): group B patients, group A platelets ( n= 1).Patients were studied until marrow recovery from remission induction chemotherapy or marrow autografting, or until refractoriness to transfused platelets developed. Platelet concentrates. Each platelet transfusion was equivalent to 5 units made up of either five single platelet concentrates, or as a ‘triple’ pheresis pack plus two single concentrates. The number of platelets, red cells and leucocytes in each dose was counted. Platelets for autograft patients were irradiated with 1 500 cGy immediately prior to transfusion using a ‘Gammacell 1000’ blood products irradiator, a dose shown not to affect in vivo survival or function (Leitman & Holland, 1985). Platelets were given through a standard blood giving set without filtration to remove leucocytes. Indications for platelet transfusion and assessment of response. Prophylactic platelets were given whenever the platelet count was < 2 0 x 1OY/l.The platelet increment was measured at l h and 20 h post transfusion and expressed as a corrected increment (CI) (Daly et al, 1980): CI = (post- pretransfusion platelet count) x body surface area (mL) no. ofplatelets transfused ( x 10”) Definition of refractoriness. A patient was designated refractory when three consecutive transfusions resulted in a 1 h CI 50 x l O Y / l (n= 1). Comparison of study groups (Table I) The 26 evaluable patients were divided equally between the two treatment groups. Seven OG patients had leukaemia (one ALL, six AML) and eight MMG patients had leukaemia Table I. Comparison of the two study groups

No. of patients Sex: male/female Previous pregnancy Previous blood transfusion No. platelet transfusions per patient: median (range)

No. donors per transfusion: mean f SD)

Cell dose per transfusion (means (SD)) Platelets x 1,eucocytes x 1 Oy Red cells x loy

Own group

Mismatched group

13 3/10

13 1112

8 4 9

(4-30)

L

2

7 (5-1 9)

4.0 10.3)

2.92 (0.7) 0 . 3 9 (0.2) 8.90 (3.9)

2.97 (0.6) 0.42 (0.3) 8.78 (3.5)

4 10

Robert Carr et a1

(one ALL, one MDS, six AML). The randomization led to an excess, in the OG patients, of females, a history of pregnancy or blood transfusion during the preceding year: nine of this group therefore had potentially been sensitized for antibody formation. Four of the MMG patients had the potential for prior sensitization by pregnancy or transfusion during the previous year. The time span of marrow failure requiring platelet support was similar in the two groups. The number of units of blood administered during the study period and the numbers of leucocytes and red cells in the platelet preparations was equivalent between the two groups.

Incidence of refractoriness and antibodg formation (Table 11) Nine (69%) MMG patients became refractory to transfused platelets compared with only one (8%) OG patient (P=O.O01). The median onset of refractoriness occurred at Table 11. Distribution of patient ABO-groups, platelet refractoriness and new antibody formation

Study group Diagnosis Own group Leukaemia

Autograft

New antibodies Patients

5 €3 8 A 12 0 18 0 25 0 28 A 36 A

Anti-HLAT

14 0 16 0 30 0 32 0 42 A

Platelet$

R

13 A 20 0 21 B 35 0 39 A 43 0

Mismatch group Leukaemia 10 A 15 B 17 0 22 0 24 A 33 A 37 0 38 A Autograft

Anti-A/B*

CLTg

n

128 CLT CLT

R R

R

256 512 >2048

R

2 (B7)

n n

PSIFT CLT 2 56

R =Refractoriness developed during study.

* Isoagglutinin titre, where

> 3 doubling dilution

increase over

baseline occurred. t Anti-HLA antibodies: grade and (specificity).NS = No specificity identified. $ Platelet specific antibodies detected by PSIFT or CLT. 4 Transient antibody.

transfusion 7 (life table analysis, Fig 1). occurring at a median of 15 d (range 6-20 d) after the first transfusion. Refractoriness was associated with increases in isoagglutinin titre and the formation of anti-HLA and platelet specific antibodies. Isoagglutinins. The repeated administration of ABO-mismatched platelets induced a significant rise ( 2 3 doubling dilutions) in the corresponding anti-A/B titre in seven of the 1 3 patients. Four were group A and three were group 0. In almost all cases the rise in titre to > 128 occurred suddenly and was coincident with the onset of poor platelet increments. The ability to form high-titre isoagglutinins appeared to be impaired in the highly immunosuppressed autograft patients (Holohan et a!, 1981).Only patient 42, who had received less immunosuppressive conditioning chemotherapy, developed a rise in titre but this patient was unable to fulfil the criteria for refractoriness, as the rise coincided with marrow recovery. No OG patient had a variation in titre of more than one doubling dilution. HLA antibodies. Lymphocytotoxic antibodies of broad specificity developed in four (50%)of the eight MMG leukaemia patients and appeared simultaneously with the increase in isoagglutinin titre, at between 10 and 19 (median 13) days after the first transfusion. Only two other patients developed HLA antibodies: patient 16 in the MMG autograft group and patient 2 5 who was the only OG patient affected. Both these individuals were multiparous females who developed antibodies of identifiable specificity, suggesting an anamnestic response from pregnancy-related immunization. Platelet specific antibodies. Six patients had anti-platelet antibodies detected by PSIFT and/or CLT on recruitment into the study (four OG, two MMG). These patients were not excluded from analysis as the antibodies did not appear to influence initial platelet increments or increase in reactivity following platelet transfusion. Four MMG patients and one OG patient developed platelet specific antibodies, which did not react in the LCT or LIFT assays, during the course of the study. In three of these (patients 2 1 , 30, 3 2 ) the antibodies were associated with poor platelet increments. Patient 30 developed plateletspecific IgM antibodies detectable by PSIFT which coincided with the onset of refractoriness. In patient 2 1, transient antibodies detected by CLT coincided with a period of poor platelet increment although refractoriness, as defined, did not occur. Patient 32 developed CLT reactive antibodies some time after the onset of otherwise unexplained refractoriness. Two additional patients ( 1 7 and 24) developed positive CLT reactions. In these two cases positive reactions were coincident with episodes of fever and it is not known whether the samples contained endotoxins which may have caused monocyte activation to give a false positive CLT result.

Clinical conseyiiences of mismatched platelets No transfusion reactions attributable to the incompatible red cells contaminating the mismatched platelet units occurred. Bleeding requiring transfusion support was observed only in two of the mismatch group patients at the onset of refractoriness.

Transfusion of ABO-Mismatched Platelets 100

I1

I

I

I l l

41 1

I

80 %

60

40

20

MMG

5

10

15

20

25

30

number of platelet transfusions

Fig 1.Estimated survival curves of refractoriness by number of platelet transfusions. Ordinate: probability (%) of not becoming refractory. Log rank statistic= 1 0 . 3 (P=0.0014).

Eflect of ABO-mismatch on platelet increments The underlying haematological condition, i.e. leukaemia or autografting for Hodgkin's lymphoma. had no significant influence on platelet increments prior to refractoriness (P=0.9 and P=0.8 at 1 and 20 h respectively), therefore patients in both diagnostic groups were combined for further analyses. When only the transfusions administered before the onset of refractoriness were analysed, there was no significant difference between the study groups. ABO-mismatched platelets achieved increments similar to those resulting from ABOmatched platelet transfusions (P=O.9 at 1 and 2 0 h) (Fig 2 ) . Sequential analysis of platelet increments indicated that the transition from full responsiveness to refractoriness was abrupt, and that there was no gradual decline in increments with ABO-mismatched platelets prior to the onset of refractoriness. DISCUSSION The transfusion of platelets across ABO barriers is accepted practice when compatible platelet units are unavailable (NIH Consensus Conference, 1987). When HLA-matched, but ABO-mismatchedplatelets are transfused into alloimmunized patients, increments are reported to be 20% less than achieved by ABO-matched platelets (Duquesnoy et nl, 1979: Tosato ef a]. 1978). Absolute refractoriness due to high titre anti-A and anti-B has been reported in two patients receiving HLA-matched platelets (Brand et al, 1986). There is little data on the use of ABO-mismatched platelets in non-immunized patients. Van Eys et a2 (1978) found that ABO-mismatched platelets given intermittently to paediatric

oncology patients were as effective as matched transfusions. A recent study (Lee & Schiffer, 1989) of leukaemic patients

given a maximum of two ABO-mismatched transfusions showed that the second mismatched platelet dose gave increments less than matched platelets, especially in patients with raised isoagglutinin titres. There is no data on the consequences for non-immunized patients of repeated infusions of ABO-mismatched platelets throughout the period of leukaemia or chemotherapy-induced marrow failure. The present study of 293 platelet transfusions given to 26 non-immunized patients has shown that, in the short term, ABO-mismatched random donor platelets achieve increments at 1 h and 20 h survival post-transfusion no different from those achieved by matched platelets. However, it is important to note that only one patient receiving mismatched platelets had an isoagglutinin titre of greater than I 6 prior to the onset of refractoriness. Aster (1965), using normal volunteers, demonstrated an inverse relationship between isoagglutinin titre and in vivo recovery of incompatible platelets. with poor recoveries when the corresponding titre was greater than 40. Repeated doses of ABO-mismatched platelets led to a high incidence (69%) of refractoriness associated with not only high-titre isoagglutinins, but also anti-HLA and platelet specific antibodies. The onset of refractoriness at a median of 2 weeks (range 1-3 weeks) in the patient group receiving ABO-mismatched platelets, was earlier than would be expected from an analysis of the literature by Slichter (1986) who found that the median time to platelet alloimmunization was 4 weeks with a range of 3-26 weeks. The induction of high-titre isoagglutinins by ABO mismatched platelets played an important role in the high

Robert Carr et a1

412

l h

237

l2

37-38

'38

"C

1 20 h

e37 Own Group

37-38

>38

"c

Mismatch Group

Fig 2. Unadjusted mean corrected platelet increments, by study group and patient temperature. Increments at 1 and 20 h for transfusions administered before the development of high titre isoagglutinins, anti-HLA,or platelet specific antibodies. The trend to lower increments with increasing pyrexia did not reach statistical

significance.

incidence of refractoriness. In all six mismatch group patients with leukaemia who became refractory, the onset was associated with an abrupt increase in isoagglutinins to a titre sufficient to severely impair increments (Aster, 1965), irrespective of other antibody formation. In one case (no. 38) refractoriness was mediated exclusively by high-titre anti-B. The sudden increases in titre were not heralded by a gradual decline in platelet responsiveness and could not have been predicted from titres measured as little as 24 h before. In addition to high-titre isoagglutinins. the study group receiving mismatched platelets had a greater incidence of lymphocytotoxic, anti-HLA antibody formation, although this incidence did not differ from the 28-55% reported in the literature (Dutcher et al. 1981; Gmur et al, 1983; McGrath et al, 1988; Murphy et nl, 1987). The observation that patients receiving mismatched platelets developed HLA antibodies

earlier than expected can only be considered preliminary in view of the small number of cases. However, the close temporal association between HLA antibodies and high-titre isoagglutinins raises the possibility that stimulation by incompatible ABH antigens may encourage early development of HLA antibodies. The association is not without precedent: it has previously been noted following hydatidiform molar pregnancies (Lawler & Fisher, 1987) and renal transplantation across ABO barriers (Brynger et al, 1984).in which apparently HLA-mediated rejection episodes are associated with increases in isoagglutinin titre. It has also previously been observed that anti-A/B titres and anti-HLA antibodies tend to vary in parallel when leukaemia patients have received ABO-mismatched platelet units (Schiffer et nl. 1976). Platelet-specific antibodies were similarly more frequent in the mismatch group but this also did not differ from the reportedincidence (McGrathet al, 1988;Murphyet al, 1987). In four of the patients the antibodies were detected only by the platelet chemiluminescence test (Lucas et nl, 1987). In two of these patients the antibodies were associated with reduced platelet increments in the absence of other identifiable cause, which suggests that the CLT may be a useful additional technique for identifying clinically significant platelet antibodies. It is possible that the increased tendency to develop early refractoriness stems from the excess of males in the study group receiving mismatched platelets, although preferential immunoresponsiveness by males undergoing chemotherapy has not previously been reported. On the other hand, pregnancy and prior transfusion, factors known to predispose to alloimmunization, were more prevalent (nine versus four patients) amongst subjects receiving own group platelets. The routine issue of platelets irrespective of ABO group reduces wastage and so leads to economic saving. However, the high incidence of sudden onset refractoriness observed in patients regularly receiving mismatched platelets, in contrast to the significantly lower incidence among patients restricted to matched platelets, casts doubt on the advisability of such a policy. Refractoriness due to anti-A/B antibodies, once identified, may be circumvented by switching to ABOmatched transfusions, although this was ineffective in the five patients who had simultaneously developed other antibodies. As a consequence of these findings, Regional policy has reverted to using matched platelets whenever possible. The results of this study suggest that mismatched platelets should not be given to patients with marrow failure requiring long-term platelet support. On the other hand, the good increments achieved when mismatched platelets are first given to non-immunized patients with low isoagglutinin titres suggests that such platelets may be given to most patients requiring very short-term support, without having to increase the number of units administered. ACKNOWLEDGMENTS Assistance with data collection by the medical, technical and nursing staff of the Department of Haematology. Royal Liverpool Hospital, and with the provision of appropriate

Transfusion of ABO-Mismatched Platelets platelet units by the staff of the Mersey & North Wales Kegional Transfusion Centre is gratefully acknowledged. Our thanks also to Chris West for computer graphics.

REFERENCES Aster, R.H. (1965) Effect of anticoagulant and ABO incompatibility on recovery of transfused human platelets. Blood, 26, 732-743. Bishop, J.F., McGrath. K.. Wolf, M.M.. Matthews. J.P., De Luise. T., Holdsworth. R., Yuen. K., Veale. M.. Whiteside, M.G.. Cooper, I.A. & Szer. J. ( 1 988) Clinical factors influencing the efficacy of pooled platelet transfusions. Blood. 71, 383-387. Borne, A.E.G.Kr. von dem. Vergheught. F.W.A., Oosterhof, F.. Reisz, E. von, Brutel de la Riviere. A. & Engelfriet, C.P. (1978) A simple immunofluorescence test for the detection of platelet antibodies. British Jourrial of Haernatologg. 39, 195-207. Brand, A,, Sintnicolaas. K.. Claas. F.H.J. &Eernisse. J.G. (1986)ABH antibodies causing platelet transfusion refractoriness. Transfusion. 26,463-466. Brynger. H.. Rydberg, L., Samuelsson. G. & Sandberg. L. (1984) Experience with 14 renal transplants with kidneys from blood group A (subgroup AJ) to 0 recipients. Transplantation Proceedings. 16, 1175-1 176. Daly, P.A.. Schiffer. C.A.. Aisner, J . & Wiernik. P.H. (1980) Platelet transfusion therapy: one-hour posttransfusion increments are valuable in predicting the need for HLA-matched preparations. Journal ojthe American Medical Association. 243, 43 5-438. Decary. F.. Vermeulen. A. & Engelfriet. C.P. (1975) A look at HLA antisera in the indirect immunofluorescence test (IIFT).Histocornpatibility Testing. pp. 380-390. Munksgaard. Copenhagen. Dunstan. R.A.. Simpson, M.B., Knowles, R.W.&Rosse, W.F. (1985) The origin of ABH antigens on human platelets. Blood, 65, 61 5619. Duquesnoy. R.J.. Anderson. A.J., Tomasulo. P.A. & Aster, R.H. (1979) ABO compatibility and platelet transfusions of alloimmunised thrombocytopenic patients. Blood, 54, 595-599. Dutcher, J.P., Schiffer. C.A.. Aisner, J. & Wiernik, P.H. (1981) Longterm follow up of patients with leukaemia receiving platelet transfusions: identification of a large group of patients who do not become alloimmunised. Blood, 58, 1007-101 1. Eys, J. van, Thomas, D. & Olivos, B. (1978) Platelet use in pediatric oncology: areview of 393 transfusions. Transfusion, 18,169-173. Gmur. J., Felten. A. von, Osterwalder, B., Honegger. H.. Hormann,

41 3

A., Sauter. C.. Deubelbeiss.K., Berchtold, W.. Metaxas. M.. Scali. G. & Frick. P.G. (1983). Delayed alloimmunisation using random

single donor platelet transfusions: a prospective study in thrombocytopenic patients with acute leukemia. Blood. 62, 473-479. Holohan. T.V.. Terasaki, P.I. & Deisseroth. A.B. (1981) Suppression of transfusion-related alloimmunisation in intensively treated cancer patients. Blood, 58. 122-128. Lawler. S.A. & Fisher, R.A. 11987) Immunogenicity of hydatidiform mole. Placenta, 8, 195-199. Lee, E.J. & Shiffer. C.A. (1989) ABO compatibility can influence the results of platelet transfusion: results of a randomised trial. Transfusion. 29, 384-389. Leitman. S.F. & Holland, P.V. (1985) Irradiation of blood products: indications and guidelines. Transfusion. 2 5, 29 3-300. Lucas. G.F., Hadley. A.G. & Hoburn. A.M. (1987) Anti-platelet opsonic activity in alloimmune and autoimmune thrombocytopenia. Clinical arid Laboratory Haematology, 9, 59-66. McCullagh. P. & Nelder, J.A. (1989) Generahed Linear Models. Chapman & Hall, London. McGrath, K.. Wolf, M.. Bishop, J., Veale. M.. Ayberk. H.. Szer. I.. Copper, I. & Whiteside. M. (1988) Transient platelet and HLA antibody formation in multitransfused patients with malignancy. British journal of Haeniatologg.68, 345-350. Metcalfe, P.. Minchinton. R.M., Murphy. M.F. & Waters, A.H. (1985) tise of chloroquine-treated granulocytes and platelets in the diagnosis of immune cytopenias. Vox Sanguinis. 49, 340-345. Murphy, M.F., Metcalfe, P., Ord,J., Lister,T.A.& Waters,A.H. (1987) Disappearance of HLA and platelet-specific antibodies in acute leukaemia patients alloimmunized by multiple transfusions. British ]ourrial of Haematologg. 67, 255-260. National Institutes of Health Consensus Conference (198 7) Platelet transfusion therapy. Transfusion Medicine Reviews, 1, 19 5-200. Schiffer. C.A.. Lichtenfeld, J.L.. Wiernik. P.H.. Mardiney. M.R. & Joseph. J.M. (1976) Antibody response in patients with acute nonlymphocytic leukemia. Cancer. 37, 21 77-2182. Slichter, S.J. (1986) Prevention of platelet allo-immunisation. Transfusion Medirine: Recent Technological Advarices. pp. 8 3-1 16. Alan R. Liss, New York. Park, M.S., Ozturk. G. & Imaki, Y. (1978) Terasaki. P.I., Bernoco, D., Microdroplet testing for HLA-A, -B, -C. and -D antigens. American Journal of Cliriical Pathology, 69, 103-107. Tosato, G.. Applebaum, F.R. &Deisseroth,A.B. (1978)HLA-matched platelet transfusion therapy of severe aplastic anemia. Blood, 52, 846-854.

Transfusion of ABO-mismatched platelets leads to early platelet refractoriness.

Forty-three consecutive patients previously unexposed to platelets and undergoing treatment for acute leukaemia or autografting for relapsed Hodgkin's...
571KB Sizes 0 Downloads 0 Views