The Evaluation and Management of Platelet Refractorinessand Alloimmunization

Eduardo Delafior-Weiss and Paul D. Mintz

T HE ARREST of hemorrhage associated withan increase in the patient's platelet count after

blood transfusion was first reported by Duke in1910.' It was not until the 1960s that clinicalstudies showed that therapeutic platelet transfu-sions decreased fatal hemorrhagic complications in

thrombocytopenic patients2 and that prophylacticplatelet transfusions reduced episodes of bleeding.3More recent reports have suggested that a thresholdof I O,OOO/~ for prophylactic platelet transfusion is

equivalent to 20,OOO/~ in the absence of addi-tional complicating clinical factors such as platelet

dysfunction, associated coagulopathy, or primarydefects of vascular integrity.4.5 The hemostaticbenefit of platelet transfusion is presumably medi-ated through an increase in the posttransfusion

platelet count above a target level, and clai"!s thatplatelet transfusion may be beneficial even in theabsence of such an increase are not commonly

accepted.

and is calculated by the following fonnul.

(platelet increment per JJL)X (weight in kg)X (blood volume as 75 mUkg)X 100 (for percent)PPR; -

(platelet count of transfusedproduct per ~)X (volume of product in mL:

The posttransfusion platelet count is dependent onthe pretransfusion platelet count, the dose of plate-lets transfused, and other factors that affect therecovery, both technical and clinical. In individualswithout splenomegaly, one third of radiolabeledautologous platelets are reversibly pooled in thespleen at any given time, whereas two thirds remainin the circulation,S so adjustments to the recoveryneed to be made. Recovery of autologous plateletsis on the orderof66 ~ 8% and approaches 100% inasplenic individuals.9 Recovery is modestly re-duced in thrombocytopenic patients with countsbelow 50,000/~ to approximately 50% :!: 22%.10

Another measurement of platelet transfusionrecovery in the clinical setting is the posttransfu-sion corrected count increment (CCI). The CCI iscalculat~d as follows7:

DEFINITION OF REFRACTORINESS

Platelet refractoriness is empirically defined as alack of response in posttransfusion platelet incre-ments after 2 or more consecutive transfusions ofan adequate dose of allogeneic platelets. Refractori-ness is due to the shortened survival of the trans-fused platelets in the recipient's circulation. Histori-cally, it has been described in 20% to 60% ofpatients who have received multiple platelet trans-fusions.6 The causes of platelet refractoriness canbe grouped into immune and nonimmune causes.

The percent platelet recovery (PPR) after platelettransfusions is the percentage of transfused plate-lets circulating at the time the count is determined

(Platelet increment per ilL)X (body surface area in square meters)

CCI =Number of platelets transfused (X 1011)

From the CCI, the formula for the calculation ofan effective dose (ED) of platelets is derived fromthe following formula:

(desired PLT count increment per~)X (body surface area in square meters)

ED(XIOJI) =ex{x:Cted corrected count increment

The actual units of the CCI are (square meters)/(~ X 1011), although as an index it is typicallywritten without units. The CCI between 10 min-utesll and I hour posttransfusion12 is used typicallyto determine the adequacy of response to platelettransfusions. The expected posttransfusion CCIbetween 10 minutes and I hour is greater than

From the Departments of Pathology and Internal Medicine.

University of Virginia Health System, Charlottesville. VA.Address reprint requests to Paul D. Mintz. MD. Directol;

Clinical Laboratories and Blood Bank. University of VirginiaHealth System. Box 800286. Charlottesville. VA 22908.

Copyright c 2000 by WB. Saunders Company0887- 7963/0011402-0007$3. 00/0

180Transfusion Medicine Reviews, Vol 14, No 2 (April), 2000: pp 180-196

QV'r

PLATELET REFRACTORINESS AND ALLOIMMUNIZATION 181

transfusion or posttransfusion counts when thesewere not available. Iii This study underscores the

importance of obtaining platelet counts immedi-ately before transfusion and I{) minutes to I hour

posttransfusion.Although platelet retractoriness in individual

patients can be attributed to single identifiableclinical factors, including fever, neutropenia, dis-seminated intravascular coagulation (DIC), andsepsis/infection, in many cases it is multifactorial,and no single clinical factor is an accurate predictorof posttransfusion platelet increments across pa-tients.17

NONIMMUNE CAUSES OF REFRACTORINESS

Platelet Quality

When the posttransfusion platelet count does notincrease as expected, one possibility is that aninadequate dose of platelets was administered as aresult of platelet loss due to washing or filtration.Although most studies have found that leukoreduc-tion filters result in the loss of less than 10% ofplatelets during filtration, some studies have re-ported loss of as much as 24 ::!: 15% of apheresisplatelets and 20 ::!: 9% of pooled platelets.18 Suchlosses may be independently associated with de-creased posttransfusion CCI.19 The platelet doseshould be adjusted to the patient's blood volume,and some patients require proportionally moreplatelets to achieve adequate increments. Otherfactors that may detrimentally atTect the quality ofthe platelets during storage include pH lower than6.2, incorrect storage temperature, and mode ofagitation. Gamma-irradiation of platelets with 25Gy does not affect platelet function or posttransfu-sion survival.2o However, in the Trial to ReduceAlloimmunization to Platelets (TRAP) study, ultra-violet B (UVB) irradiation at dosages used toprevent alloirnmunization (1.5 J/cm2) was indepen-dently associated with lower posttransfusion CCIS.19

Platelet Age

There are conflicting reports regarding the ben-efit of fresh platelets.21.22 In one study, patients withinfection, DIC. or splenomegaly were refractory toolder platelets but not to fresh platelets}3 Thepossible advantage of fresher platelets in improv-ing the response to platelet transfusions in patientswith non immune refractoriness has been attributedto activation of platelets during storage.24 In one

7.500, representing recovery of at least 20% to30%. Values below these have been associated withaccelerated platelet destruction.13 Although occa-sionally low CCI or percent recoveries occur for avariety of reasons other than refractoriness, poorresponses to at least 2 sequential transfusions of anadequate dose of platelets is considered evidence ofplatelet refractoriness. Repeatedly low CCls be-tween 10 minutes and I hour posttransfusion aregenerally attributed to an immune cause,12 althoughin some alloimmunized patients good incrementscan still be achieved with incompatible platelets.14In some patients with other reasons for accelerateddestruction, the CCI at these times may be low. ACCI measured 18 to 24 hours after the transfusionof platelets is usually approximately 60% of theI-hour CCI and should be greater than 4,500,13equivalent to a recovery greater than 15%. Areduced CCI at 18 to 24 hours after a nonnal CCI atI hour has been interpreted as representing evi-dence of increased consumption owing to nonim-mune clinical causes,I5 although in s~me patientsthis can represent immune destruction.

The maximal lifespan of rein fused autologousradiolabeled platelets is approximately 9.5 := 0.6days, with 85% removed by senescence.1o A fixednumber of platelets are lost at random, presumablymaintaining the vascular integrity. Overall survivalof platelets is decreased proportionally to thedegree of thrombocytopenia. When the plateletcount is greater than I OO,OOO/~L, platelet life spanis independent of the count, but life span ismodestly reduced in patients with counts between50,000 and lOO,OOO/~L to 7.0 + 1.5 days, andmarkedly reduced when platelet counts fall below50.000/~ to 5.1 + 1.9 days.lo A fixed number ofapproximately 7,100 platelets/lJlid is consumed.Because this represents only 18% of the nonnalrate of platelet turnover (41,200 platelet/lJlid), thefixed platelet consumption does not have a signifi-cant effect on overall platelet survival when countsare over lOO,OOO/~. However, as the platelet countdecreases below 1 OO,OOO/~, a larger proportion ofplatelets are devoted to daily endothelial supportand overall platelet lifespan decreases. to

The importance of precise timing of samples forcorrect estimation of the CCI has been shown bydel Rosario and Kao,I6 in a study in which theinvestigators used the average reduction rate of 800platelets/1JL/h to make approximations to the pre-

DELAFLOR-WEISS AND MINTZ

study, the proportion of activated platelets in S-day-old platelet concentrates was 15% to 20%. andthese were cleared more rapidly from the circula-tion after transfusion than were fresher platelets.25After an initial phase of splenic sequestration andlater release to the circulation, some transfusedplatelets may recover their normal function andlose evidence of activation within 24 hours aftertransfusion.26 In several studies, recoveries were10% to 59% higher with fresher platelet products.usually in those patients with fever. O[C, or spleno-megaly.22.27 In the TRAP study,19 transfusion ofplatelets stored for less than 48 hours resulted ingreater posttransfusion increments. The criteria fordiagnosing platelet refractoriness in this studyincluded failure to respond to fresh or ABO-matched platelets as an initial therapeutic maneu-ver in refractory patients. Slichter et aJ2K havereported the preliminary results of a small prospec-tive randomized trial comparing fresh and storedapheresis and pooled platelet concentrates. A de-crease in the viability of stored platelets becamestatistically significant only after 5 days of storagefor both products. As a practical matter. the freshestplatelets available for transfusion at most institu-tions may be I to 2 days old because of transmis-sible disease testing and shipment.

sion reactions were also associated with lowerI-hour CCIs}3

When alloantibodies cannot be demonstrated bythe laboratory and fewer than 3 weeks have elapsedsince starting chronic transfusion therapy in pa-tients without a history of previous exposure toHLA antigens by transfusion. solid organ trans-plant. or previous gestations. nonimmune clinicalrefractoriness is the likely cause. Although plateletrefractoriness in individual patients can be attrib-uted to single identifiable clinical factors, in manycases it is multifactorial. and no single clinicalfactor is a predictor of posttransfusion plateletincrements across patients. L 7

Although fever has been implicated in decreasedplatelet recoveries. it is unclear whether the sepsis.DIC. or the antibiotic treatment associated withfever rather than the fever per se are thecause}.6.15.34.35 There is no evidence that decreasingfever with antipyretics will improve recoveries.Treating the underlying cause of fever and controlof DIC may be of value at improving the plateletcounts.

Clinical Condition.\'

Platelet refractoriness most commonly resultsfrom clinical conditions associated with acceler-ated platelet consumption. These may coexist withalloantibody-mediated refractoriness.1S.17 Spleno-megaly, DIC, amphotericin B therapy, bone mar-row transplantation, veno-occlusive disease of theliver, and graft-versus-host disease have been re-ported to reduce platelet recovery.6.IS.29-32 In pa-tients undergoing hematopoietic stem cell transplan-tation, poor posttransfusion CCls at 16 hours weresignificantly correlated with high total bilirubin.total body irradiation. high serum tacrolimus. andhigh serum cyclosporin by multivariate analysis.32A multivariate analysis of clinical factors associ-ated with decreased posttransfusion increments inthe TRAP study showed fever, infection, bleeding,splenomegaly, gamma-irradiation. concurrent am-photericin B therapy, as well as a history ofprevious pregnancy in women, to be independentlyassociated with decreased CCls.19 Those transfu-sions associated with moderate to severe transfu-

IMMUNE CAUSES OF REFRACTORINESS

ABO Incompatibility

There is a highly variable ABH expression.greater than 20-fold. on the surfaces of platelets ofdifferent persons.36 and only 7% of non-group aplatelets express high levels of A or B antigens asdetermined by enzyme-linked immunosorbent as-say and immunoblotting studies. Expression corre-lates with the levels of glycosyltransferase in theserum of the donor. In contrast to red cells, whichexpress type 2 ABH chains. platelets express equalamounts of type 2 and type I chains.37

Severnl studies have addressed the significanceof ABa group compatibility in the response toplatelet tmnsfusion.3841 In recipients with hightiters of anti-A or anti-B isoagglutinins (eg. >64).ABa-incompatible platelet transfusions may con-sistently fail to produce adequate responses,42 butincompatible transfusions have been reported to beas effective as ABa-identical platelets in patientswith lower titer isoagglutinins.43 In one report.CCIs were higher in patients who received ABaidentical pooled platelets, intermediate in thosewho received ABa plasma incompatible platelets.and lowest in recipients of ABa major incompat-ible concentrates.41 Aster38 reported that the meanrecovery of radiolabeled AI platelets transfused

PLATEl IRINESS AI JIMMUNIZA"

into 0 recipients was only 19% and that of AI Bplatelets transfused to 0 recipients only 8%, ascompared with recovery of 63% observed whenABO-compatible platelets were transfused. In thesame study, transfusion of incompatible group Bplatelets resulted in recoveries similar to those ofcompatible transfusions. ABO antigen density onthe platelet ~urt.ace may influence the result oftran~fusions, because group Az platelets expressapproximately 40-fold less A antigen than Alplatelets and may provide good recoveries in groupa patients even in recipients with high titers ofanti-A isoagglutinins.44 Transfusion of ABO-incompatible donor plasma also may result inreduced platelet recoveries. ABO incompatibledonor pla~ma in platelet products result in poorresponses presumably from the fonnation of circu-lating immune complexes (CIC) with the soluble Aand B substance in the recipient's plasma withresultant increased platelet clearance.40 Based onthe available clinical evidence. a trial of plateletconcentrates ABa compatible with recipient plasmato refractory patients with isoagglutinin titers above64 is reasonable and may result in adequateCCIS.J9.41 This practice defines the role of ABOincompatibility in particular patients and may be allthat is necessary.

are a very weak stimulus for the fonnation ofantibodies to HLA class I antigens. because alloan-tigen recognition usually requires expression ofboth class I and class II HLA antigens, found on thesurface of donor leukocytes. Residual donor leuko-cytes in red blood cells or platelet concentratesprovide a source of antigen-presenting cells (APC).52Leukoreduction to less than I to 5 X I Q6 pertransfusion minimizes the APCs in the transfusedproduct and decreases the total antigen load, shownin various studies to be effective in preventingalloimmunization and refractoriness to platelet trans-fusion.50.53.54 Ultraviolet B irradiation abolishesAPC stimulation in mixed lymphocyte cultures andmay cause specific unresponsiveness to HLA anti-gens.55 Leukoreduction and UV-B irradiation wereequivalent in preventing alloimmunization, asshown in the TRAP study. 50 The rate of alloimmu-

nization was not statistically different betweenrecipients of leukoreduced platelets, 18%, andrecipients of UV-B irradiated platelets, 21 %, asopposed to 45% in patients receiving untreatedblood products. Refractoriness was lower in theintervention arms: 3% and 5% in recipients ofleukoreduced and UV-B irradiated products, respec-tively, compared with the controls (13%). In aseparate prospective randomized study of patientsselected for low-risk histories for alloiminunization(not previously exposed to HLA antigens), a reduc-tion in alloimmunization from 42% to 7% wasobserved with leukoreduced blood components.54

Primary alloimmunization to HLA antigens oc-curs at a median of 3 to 4 weeks (range, 2 to 56weeks) after the first transfusion in those whoreceive multiple transfusions. 55 Patients with previ-

ous exposure to HLA antigens are at high risk todevelop alloimmunization and platelet refractori-ness and may do so earlier than patients withoutsuch history.46.56 In the absence of previous transfu-sion, alloimmunization develops in 4% of womenafter one pregnancy and 26% after four pregnan-cies.51 In the TRAP study, leukoreduction or UVBirradiation reduced the risk of HLA-aIloimmuniza-tion even in patients with previous exposure toHLA alloantigens.50 In contrast, Sintnicolaas et a}46found that leukoreduction did not prevent second-ary HLA-alloimmunization (43% v 44%) or plate-let refractoriness (41 % v 29%, P = .52) in patients

with a history of pregnancy in a randomizedprospective study. When prefonned alloantibodiesare present before the current course of transfusion.

HLA Incompatibility

HLA antibodies are usually detected by lympho-cytotoxicity testing. The percentage of panel cellsto which the patient has developed HLA antibodiesis called the panel reactive antibody (PRA) level.4sPRA may be used to screen patients receivingchronic transfusion therapy, and increases in PRAlevels occurring over time may be used to predictalloimmune refractoriness.4s.46 Reactivity with over70% of the panel lymphocytes is the usual cutoffused in the clinical setting to diagnose alloimmuni-zation.

Fewer than 50% of refractory patients haveevidence of platelet or lymphocytotoxic (LCT)HLA antibodies.47.48 Most studies have shown aclear relationship between the presence of HLALCT antibodies and concomitant platelet refractori-ness!.12.3S.-I9 although in I series only 30% of allalloimmunized patients were refractory to platelettransfusion.so

Platelets express 73% of the whole blood load ofclass I HLA-A and HLA-B antigen,SI but they donot express class II HLA antigens. Platelets alone

184 DELAFLOR-WEISS AND MINTZ

to the recipient but possessing nonidentical anti-gens that are considered to be cross-reactive andnot immunogenic to the recipient; (3) identificationof antibody specificities and selection of donorswho lack the corresponding antigen; (4) findingselectively mismatched donors for those antigensthat are known not to be strongly expressed indonor platelets or for which expression is variable.

For HLA-alloimmunized patients matched toprivate antigens. prospective donors are classified

..

based on the number of antigens identical to therecipient (Table I). Platelet count increments withA and BU matches are equivalent and superior to Cand 0 matches in alloimmunized refractory pa-tients.60.63 One study63 found that CCIs with BXmatches were lower than with A and BU matchesbut higher than those with C and 0 matches.Overall. only 80% of alloimmunized clinicallystable patients benefit from fully HLA-matchedproducts.6.64 Failure rates with platelet transfusionsselected using CREGs has been reported to be ashigh as 41 %.63 Some transfusions within CREGsare more successful than others. Bidirectional trans-fusion within donor-recipient cross-reactive pairsAI/All. A2/A28. B5/18. and B7/B22 resulted ingood platelet increments. whereas Al/A3. A3/AII.85/BI5. B5/B17. 88/B14. BI2/B21. and B7/B27did poorly.65

Reasons for the failure of HLA-matched transfu-sions include undiagnosed clinical factors. alloim-munization to platelet-specific or minor histocom-

Table 1. Grades of HLA Matching Between

Recipient and Donor

Grade

B1U

82U

B1X

B2UX

All 4 antigens in the donor are identical to the

recipientOnly 3 donor antigens are identified. All are

present in the recipient.

Only 2 donor antigens identified. Both are pre-sent in the recipient.

Three donor antigens are identical to therecipient, fourth is a member of a cross-reac-tive group.

Two of the 3 detectable donor antigens are pre.sent in the recipient. The third is a member

of a cross-reactive group.Two donor antigens are identical to the

recipient, 2 are cross-reactive.Three donor antigens are identical to the

recipient, 1 antigen is mismatched.Two donor antigens are identical to the

recipient. 2 mismatched.

they tend to persist, whereas de novo antibodies aremore likely to be transient,58 Patients receivinghigh-dose steroids may have their HLA-alloimmu-nization delayed to a mean of 7 to 8 weeks (range.2-15 weeks),4'1.5'1

Selection of HLA-matched platelets. Selectionof platelets for transfusion based on HLA matching

improves posttransfusion platelet recovery in pa-tients with platelet refractoriness due to alloimmu-nization to HLA antigens.47.60 Daly et al12 reportedsignificantly improved CCIs by transfusing HLA-matched products to patients who had LCT antibod-ies and poor I-hour CCIs to pooled platelets. Peterset al61 studied the kinetics of HLA-matched 1111n-labeled allogeneic platelets in 5 patients withalloimmune refractoriness and found that recoveryat 1 hour postinjection and median platelet lite spanimproved with HLA matching.

HLA antibodies may be classified in 2 groups:those that recognize only an epitope in a particularHLA allele (private epitopes), such as HLA A2 !}rHLA B 12. and those that recognize more than Igene product. In the latter group, there are antibod-ies that may recognize structurally similar epitopesin different gene products or identical epitopespresent in different alleles (public epitopes). Differ-ent epitopes that are sufficiently similar to berecognized by a specific HLA alloantibody are saidto belong to a cross-reactive group.

Most patients who develop alloimmunization arebroadly immunized to class I HLA antigens. andonly a minority of refractory patients have clear-cutspecificities to private class I antigens.62 This broadreactivity is not caused by multiple alloantibodiesbut instead is usually attributable to 1 or 2 antibod-ies directed against public specificities present in a

relatively large percentage of the population. Forexample, Bw4 and Bw6 are public antigens en-coded by a diallelic system. These antigens areassociated with 2 different sets of HLA-B antigens.Special techniques such as binding assays usinglabeled antiglobulin reagents may be required todetect these antibodies.

HLA matching requires knowledge of the patientand prospective platelet donor HLA phenotypes.The principles of selecting compatible plateletproducts are based on (I) Selecting a donor with anHLA phenotype identical to the recipient (or not

expressing any antigens other than those present inthe recipient), or selecting crossmatch-compatibleplatelets; (2) Selecting a donor with partial identity

B2X

D

Adapted from McFarland.'45

PLATELET REFRACTORINESS AND ALLOIMMUNIZATION 185

demonstrated in two thirds of patients who losetheir HLA alloantibodies and may explain thisphenomenon.71 Patients who demonstrate alloim-mune refractoriness deserve a trial of unmatchedplatelets at intervals after the diagnosis of refracto-riness. Serial determination of LCT or plateletantibody may guide this therapy.

Selection of platelet.\" ba.\"ed on platelet cross-match. Ordering HLA-matched platelets does notguarantee an adequate match. In one study, 43% ofsuch products were BX or C matches}2 Therationale for platelet cross-matching is to select acompatible platelet concentrate irrespective of thenature of alloimmunization (ie, to HLA or platelet-specific antigens). Platelets compatible with therecipient's serum in vitro are selected from singledonor platelet units available at the time of testingthe inventory. Because ABC major incompatibilityoften leads to an incompatible cross-match, onlyABC-compatible platelet units should be cross-matched.

Platelet cross-matching has the advantage ofallowing selection of platelets from a larger pool ofdonors who otherwise would have not been se-lected because either they are poor matches (BX, C,or D) or their HLA type is unknown.73 Some bloodcenters include the HLA type of the donor, ifknown, on all single-donor platelets. This can guidethe hospital transfusion service in selecting prod-ucts to cross-match. Even in highly alloimmunizedpatients, 5% of unselected screened donors may becompatible by cross-match.

In most studies, cross-matching has been foundto be equivalent to HLA matching in terms ofpredicting adequate responses to platelets transfu-sion. The sensitivity and specificity of cross-matching to predict good responses in patients havebeen reported to be greater than 80% in clinicallystable patients}4.7S Although selection of cross-matched platelets usually results in adequate recov-eries, in one study this strategy was inferior to HLAgrades A and BU matching}3-7S Friedberg et al76observed equivalent results with cross-match andHLA match, although HLA matching benefitedonly the subgroup of patients with good HLAmatches (grades A or BU). None of 31 cross-matchincompatible platelets provided an adequate incre-ment. An incompatible cross-match result predictspoor responses, but a compatible cross-match doesnot exclude a poor response.6 As with red cellantibodies, the sensitivity of the cross-match is

patibility antigens. ABO group mismatch. and useof older products. In addition to the clinical factorsmentioned. development of intragroup antibodiesto cross-reactive HLA antigens may contribute tofailure of the BX matches.

Platelet .\"election based on specific HLA antigenmismatch. Characterization of the patient's anti-body may be useful in certain circumstances tochoose donors by selective mismatch in patientswho are less broadly immunized. For example.patients who develop antibodies against a singleHLA class I antigen (eg, HLA A2) may receiveeffective platelet transfusions specifically selectedto be negative for the HLA A2 antigen.66.67 Pooledplatelets. some of which would be expected not toexpress the antigen. also may be effective. Inpractice. this may be tried if it is difficult to findgood HLA matches for a particular recipient andpartially matched transfusions are unsuccessful.

Trans.fusin,~ HLA antigen-positive platelets inthe presence of HLA antibody. There are. 50,000to 120.000 HLA molecules per platelet.51 Expres-sion of some of the class I HLA antigens on theplatelet surface may vary by a factor of 35-fold forHLA-B 12.68 8-fold for HLA-B8.69 32-fold forBw4. and 20-fold for Bw6!0 In many patients withantibody to HLA B 12. transfusion of HLA-B 12antigen-positive platelets may result in adequateresponses. Expression of B 12 is enhanced whenA 11 is also present. but decreased when A2. A3. orAw28 are present.68 Expression of Bw4 is strongerwhen associated with B5 or B27. and weaker withB13. Expression of Bw6 is stronger with B7. Bl5or B35, but weaker with B8 or B 14.70 Platelets withlow antigen expression may not be efficientlydestroyed by the antibody. In one study, 69% ofimmunized patients responded to HLA-B 12 antigen-positive platelets.48 In selected refractory patients,it may be possible to use this information to selectamong incompatible platelets for transfusion, ifcompatible products are not available.

Return to unselected platelets. LCT HLA anti-bodies may be transient. They have been reportedto disappear in 40% of patients after I week toseveral months. despite continued platelet transfu-sions,6.49.58.59 and previously refractory patientshave regained good responses to unmatched plate-lets for periods from 2 weeks to 36 months.49 Otherstudies have described similar findings.58.59 Devel-opment of anti-idiotypic antibodies reactive withthe variable regions of HLA antibodies has been

186 DELAFLOR-WEISS AND MINTZ

and platelet refractoriness is evolving. Thirteenhuman platelet antigen (HPA) systems have beendescribed (Table 2). Cases in which poor plateletrecoveries are seen after transfusion of HLA-matched and ABO-identical platelets prompt con-sideration of the presence of platelet-specific allo-immunization.58.84 In multitransfused patients,alloimmunization to platelet-specific antigens ismost commonly directed to antigens with pheno-

typic frequencies below 30%,85.86 whereas alloanti-bodies directed to the high-i.nciqence antigens arerare.56.87

Although Murphy and Waters88 estimated thefrequency of platelet-specific antibodies to be 20%to 25% in multi transfused patients, in the recentTRAP study,5O only 8% of the patients developedantibodies to platelet-specific antigens. Simulta-neous alloimmunization to HLA and platelet-specific antigens is not rare. KickIer et al89 identi-fied antibodies to GPllb/IIIa in 9 of 293multi transfused thrombocytopenic patients, and all9 were also alloimmunized to HLA antigens. In onereport, 90% of patients alloimmunized to plateletantigens also had HLA antibodies.90 Between 20%and 25% of patients broadly alloimmunized toHLA antigens also may be alloimmunized to plate-

let-specific antigens. The contribution of platelet-specific alloimmunization to platelet refractorinesshas not been entirely defined. Some studies havefound no clear correlation between platelet-specificantibodies and poor responses to platelet transfu-sion.91.92 Leukoreduction does not appear to affectthe rate of alloimmunization to platelet-specific

antigens.soFor a patient with a documented platelet-specific

alloantibody who is refractory to HLA-matchedplatelets, the transfusion of platelets lacking the

corresponding antigen may be beneficial. Keko-mill et al93 reported 67 successful transfusions to 6patients who were simultaneously alloimrnunizedto HLA and platelet-specific antigens, with plate-lets matched to both antigen groups. Donor selec-tion based on platelet-specific antigen genotyping

by polymerase chain reaction using allele-specificprimers to HPA sequences has been reported.94Empiric transfusion of platelets of known HLA and

platelet-specific antigen types may detennine whichalloantibodies are causing refractoriness in an indi-vidual patient.

insufficient to detect all alloantibodies. The abilityof the cross-match to predict posttransfusion re-sponses may be lower (60% to 80%) in unselectedrefractory thrombocytopenic patients than in thosewho are clinically "stable" (80% to 100%), that is,without associated clinical factors./) A theoreticallimitation of using cross-matched rather than HLA-matched platelets may be the exposure of thepatient to a greater number of mismatched HLAantigens to which alloimmunization is possible.

Several methods of cross-matching have beendescribed and found similarly useful to predict theCCI, including solid-phase red cell adherence(SPRCA),77 flow cytometry,78 immunofluores-cence, LCT cross-matches, radiolabeled antiglobu-lin test, complement fixation, platelet serotoninrelease, platelet factor 3 release, and platelet aggre-gation.79 Rachel et al8o found an overall correlationof 97% between SPRCA cross-matching and trans-fusion outcome. The commercially availableSPRCA assay (Capture-P@, IMMUCOR, Norcross,GA) was described by Bock et al.s! SPRCA testscannot differentiate between alloimmunization toHLA or platelet-specific antigens. Chloroquine oracid stripping of the platelets before SPRCA or

platelet cross-match, differential platelet/lympho-cyte reactivity, monoclonal antibody-specific immo-bilization of platelet antigens assay,82 or differentialcoating of HLA and human platelet antigen (HPA)antigen by enzyme immunoassay83 may be of valuein differentiating HLA from HPA antibodies.

We believe that either platelet cross-matching orordering HLA-matched platelets is an equally ap-propriate strategy to treat the alloimmunized, refrac-tory patient. The decision regarding which to useshould be made initially by each hospital in consul-tation with its blood supplier and should be deter-mined largely by local factors (eg, Does the hospi-tal perform platelet cross-matching? Does the bloodcenter have a large HLA-ty~d donor base?) Pa-tients not responding to cross-match-<:ompatibleplatelets should be given HLA-matched products incase the cross-match is not sufficiently sensitive todetect alloantibodies present. Patients for whomgood HLA matches are not available may benefitfrom the determination of cross-match compatibil-ity with the best HLA-matched products that can befound.

Selection of platelets based on platelet-specificantigen matching. The understanding of the roleof platelet-specific antigens in alloimmunization

PLATELET REFRACTORINESS AND ALLOIMMUNIZATION 187

Table 2. Human Platelet Alloantigens

HPA-2a (KOb)

HPA-2b (Ko., Sibs)Ib Thr145

Met145

HPA-3 HPA-3a IBak., Lek.)

HPA-3b tBakb)

lIe843Ser843

lib

HPA-4 HPA-4a (Pen', Yuk"'HPA-4b (Pen", Yuk')

ilia Arg143Gln143

HPA-5a (B,o, Zav.)

HPA-5b (Br., Zav., Hc.)Lys505

Glu505

la

HPA-6 HPA-6a (Cab)

HPA-6bW (Cae, rue)

ilia Arg489Gln489

HPA-7a (MOb)

HPA-7bW (Mo.)

Ilia Pro407Ala407

HPA-8 HPA-8aHPA-8bW (Sf')

Ilia Arg636Cys636

HPA-9 Val837

Met837

Arg62

Gln62

Arg633

His633

Gly15

Glu15

Thr799

Met799

Not known

Not known

Not known

libHPA-9aHPA-9bW (Maxo)HPA-l0aHPA-10bW (LaO)HPA-l1a

HPA-11bW(GroO)HPA-12aHPA-12bW (IVO)HPA.13a

HPA.13bW(SitO)HPA. (CeO'HPA. (Vao)HPA. (PeO)

Ilia

HPA.11 ilia

HPA-12 Ib

HPA-13 la

HPA-tHPA-tHPA-t

iliailiaIb

NOTE, These also are known as "platelet-specific antigens," 145.'46

Abbreviations: HPA, human platelet alloantigen; NA, not available.

'Phenotypic frequencies for the Caucasian population only.tNumber not yet assigned,

without endogenous platelets available for testing,it may be difficult to prove the existence ofautoantibodies, especially at a time when mostcirculating platelets are previously transfused allo-geneic platelets. When strongly suspected clini-cally, or a previous diagnosis of autoimmunethrombocytopenia already has been made, a trial ofhigh-dose steroids, splenectomy, or intravenousimmunoglobulin (IVIG) may be justified.99

Many drugs have been implicated in the develop-ment of immune or nonimmune thrombocytopenia.Some of these are likely to be used in patients with

Autoantibodies, Drug-Induced Antibodies andPassively Acquired Antibodies

The presence of platelet autoantibodies mayimpair the response to transfused allogeneic plate-lets; however. decreased platelet survival occurs inonly a minority of patients in whom they aredetected. Platelet autoantibodies have been re-ported in acute leukemia. lymphoproliferative dis-orders. solid tumors. sepsis. cytomegalovirus infec-tions. and hematopoietic stem cell transplantationand its complications.9I,92.9s-98

In patients with severe thrombocytopenia and

DELAFLOR.WEISS AND MINTZ

collection autologous platelets cryopreserved for25 patients receiving high-dose chemotherapy forbreast cancer. Mean platelet recovery was 63%(range. 44% to 81 %) after freezing and thawing.Twenty-three patients did not require allogeneicplatelet transfusions. and 21 patients had I-hourCCls greater than 7.500. Platelet size was largerthan fresh controls. and there was reduced adeno-sine diphosphate-induced aggregation. but otheragonist responses were adequate. For patients whowill go through prolonged periods of thrombocyto-penia. several collections may be necessary. Cryo-preservation may be used for those alloimmunizedpatients with few or no compatible donors. but itsimplementation is difficult. and this practice is notwidespread.

hypoproliferative thrombocytopenia requiring mul-tiple transfusions (eg, cephalosporins, penicillin,methycillin, TMP-SMX, pentamidine, vancomy-cin, and amphotericin 8).6 Diagnosis dcpends oncliciting a temporal relationship of thrombocytope-nia and platelet refractoriness with the use of thedrug and demonstration of drug-related antibod-ies.1lXJ Evaluation for drug-induced antibodies maybe warranted in the subset of patients with temporalass()Ciation to the exposure of an offending drugand the onset of refractorine~s. 8ecau~e of thecomplexity of the clinical situation, it may not beeasy to identify which factors are responsible forrefractoriness, and the offending agent should bewithdrawn when the diagnosis is strongly sus-pected, if possible.

The presence of circulating immune complexesand clearance of platelets by the reticuloendothelialsystem may induce thrombocytopenia. 101 As noted,

ABO-incompatible plasma in transfused plateletsmay contribute to formation of immune com-plexes.40 The presence of IgG antibodies to. plasmaproteins was found to increase progressively withplatelet transfusions and to correlate with platelet-bound IgG and refractoriness.,o2 The degree towhich this phenomenon poses a problem in mostpatients is unknown.

Antifibrinolytics

Bleeding was controlled with epsilon-aminoca-proic acid (EACA) in 90% (36 of 40) of patientswith autoimmune thrombocytopenia or bone mar-row failure and alloimmune refractoriness to plate-let transfusions.'os-1o1 In one randomized studycomparing platelet transfusions with and withoutEACA in 18 thrombocytopenic leukemic patients,there was no significant difference between groupsin "major bleeding" or number of platelet transfu-sions required. However, the authors reported "over-all less bleeding" in the treated group.IOS Tranex-amic acid was found to reduce hemorrhage in adouble-blind trial in thrombocytopenic patients,without an increase in thromboembolic complica-tions.l09 Antifibrinolytic therapy is worth trying forbleeding thrombocytopenic patients without com-patible platelets available, or for those who fail torespond to compatible products.

MANAGEMENT WHEN COMPATIBLEPLATELETS ARE UNAVAILABLE

OR INEFFECTIVE

Alternative strategies for managing thrombocyto-penic refractory patients have been reported mainlyin small uncontrolled studies or case reports, andinterpretation of the benefits of these modalities isdifficult. Positive results may be more likely to bereported and published than negative results. Noclear-cut recommendations can be made regardingthese therapies (Table 3).

Intravenous Immunoglobulin

Although IVIG therapy may be efficacious forsome cases of autoimmune thrombocytopenia,llothere are conflicting data on the treatment ofplatelet transfusion refractoriness with IVIG. Mostpublished studies have shown some benefit,lll-119but in 3 studies no advantage was noted. 120-122 Of a

total of 66 patients reported, 37 (56%) had improve-ment in platelet counts posttransfusion, reductionin bleeding, or decreased platelet transfusion re-quirements. It is possible that successful studies aremore likely to be reported. KickIer et all18 reporteda randomized placebo-controlled clinical trial ofalloirnrnunized refractory patients treated with IVIG

Cryopreservation of AutoLogous PLateLets

Platelets collected during recovery of chemother-apy may be frozen in the cryoprotectant dimethylsulfoxide until needed. Approximately 50% of theplatelets are recovered after processing and transfu-sion. The platelets have normal function and sur-vival. Funke et allO3 reported 78 transfusions ofautologous cryopreserved platelets to 8 alloimmu-nized patients. The in vitro platelet recovery was85 :t 4%, with a posttransfusion CCI of 11,000.More recently, Torretta et al104 used single-

PLATELET REFRACTORINESS AND ALLOIMMUNtZATION 189

Table 3. Treatment of Platelet Refractoriness

8

25823

85% Recovery63% Recovery

17

913No difference in "major bleed-

ing"; "overall less bleeding"

5 9 loading, 6 g/d thereafter

16-24 g/d8-24 g/d100 mg/kg loading dose12-24 g/d thereafter6 g/d109*

17 EACA

9EACA14 EACA9 Platelets + EACA

9 Platelets6 Tranexamic acid (6 Place-

bo-no response)5

IVIG111

112113114

33319

22313

0.4 g/kg/d x5 d0.4 g/kg/d x5 d0.9 gikg0.4 g/kg/d x5 d followed by 0.8

g/kg/d x 5 d

2 g/kg/d x2-5 d0.4 g/kg/d x5 d0.4-08 g/kg/d x5 d0.4 g/kg/d x5 dPt. 1: 0.4 gikg/d x5 dPt. 2: 1 gikg/d x2 d0.4 gikg/d x5 d

0.4 g/kg/d x5 d0.4-0.6 gikg/d x5 d

32652

116117

118*119

32

107 (5 Placebo-no response)

2

5

7

11

75

00038 (51%)

11 (61%) 10 LId x2-3 d18

120

121122Total

Plasma exchange124

Staphylococcal A protein126 6

127 3 0

0.5-2 L plasma processed. 1-4

treatments2 L plasma processed. 4 treat-

ments

13 6(46%)TotalMassive platelet transfusion

130 2 20 Units to adsorb the anti-body. Improved incrementsobserved thereafter.

211

111

12.5 mg/kg/d300 mg/d3.3 mg/kg twice daily

34 Disappearance of alloantibody

Cyclosporin A131132133

Antithymocyte globulin135

HLA stripped platelets136 1 Patient

2 Normal subjects

1 73% Recovery in normal sub-jects; results in patient com-

parable to HLA-matchedGI bleeding stoppedGI bleeding stopped

21

11

137138

Vinblastine-loaded platelets139 Effect lasted a few days;

bleeding stopped

1 (11 courses)

-Prospective controlled study.

DELAFLOR.WEISS AND MINTZ190

at 4(X) mg/kg/d for 5 days. Five of 7 patientsreceiving IVIG had significant increases in CCI at Ito 6 hours compared with pretreatmcnt values, butonly one had improved platelet survival at 24hours. Improvement in I-hour CCI was seen in 4patients receiving HLA-mismatched platelets. Re-sponse~ ~een with IVIG are usually rapid, rangingfrom I to 9 days, and last for 6 to 8 weeks. Onereport documented a response lasting over I year ina patient receiving chronic platelet support.117 Pro-posed mechanisms include macrophage Fc receptorblockade, platelet coating, and displacement ofpreviously bound antibodies from the platelet sur-face.123 All studies on the treatment of refractori-ness are confounded by the spontaneous loss ofantibody. Physicians must be mindful that a courseof high-dose therapy costs several thousand dollars.

Other Options

Plasma exchange. Therapeutic plasmapheresiswas reported in only 1 case series of 18 ~lloimmu-nized patients.124 A response was seen in 8 of 13patients with detectable LCT antibodies at the timeof pheresis. The authors concluded that the bestresponses are seen with 10-liter plasma exchangesperfonned daily for 3 days. Plasmapheresis alsomay be considered in patients with autoantibodiesor drug-induced thrombocytopenia as a means ofremoving IgG or immune complexes. This therapyis employed only rarely owing to its limitedreported success.

Anti-D immune globulin. Heddle et al125 re-ported a randomized trial of weekly intravenousanti-D (20 ~g/kg) or placebo given to 43 Rh-positive acute leukemia patients at diagnosis, toprevent the development of refractoriness and toimprove posttransfusion platelet increments. Nodifferences were observed between groups, exceptfor an increase in red cell transfusions in thetreatment group.

Immunoadsorption: Staphylococcal Protein AColumns. There are 2 small case series on the useof extracorporeal immunoadsorption in the treat-ment of platelet refractoriness. In a study byChristie et al,126 6 of 10 patients refractory toplatelet transfusions responded to treatment withstaphylococcal protein A columns (PROSORBA,IMRE Corp., Seattle, WA) after failing other immu-nosuppressive therapy. Responding patients hadposttransfusion CCIs at 10 minutes to 2 hours of48,000 compared with 16,000 pretreatment. Eight

of these patients had demonstrable platelet antibod-ies. Patients recei ved from I to 14 treatments, andthe total volume of treated plasma ranged from 500to 2,0<)0 mL per treatment at a rate of 10 to 20mUmin with apheresis equipment (CS-3000, Bax-ter, Deerfield, IL) or by an off-line procedure. Inanother clinical study, Lopez-Plaza et al127 foundno effect in 3 patients. It has been hypothesized thatalthough staphylococcal protein A binds IgG (ex-cept IgG3), the key effect is not immunoadsorptionbut immunomodulation by induction of anti-idiotypic antibodies.128.129 The usefulness of thisexpensive approach needs to be confirmed.

Massive platelet transfusion. Transfusion oflarge doses of platelets to immunoadsorb alloanti-body or block the phagocytic system has beendescribed in 2 patients. 130 After 20 pooled platelet

units were transfused, the subsequent doses re-sulted in improved responses and produced aclinically evident effect on hemostasis. Alterna-tively, continuous platelet transfusion (eg, I doseevery 4 to 6 hours) has been tried with low successin patients refractory to compatible platelets or forwhom such products are unavailable.

Immunosuppressive therapy. After 2 to 3 weeksof treatment with 20 mg/kg/d cyclosporin A, refrac-tory dogs had improvements in recovery and sur-vival. Response correlated wi~h drug levels.13!Improvement has been reported in 4 patients.131-133Yamamoto et al132 reported reduction in LCTantibodies in a 26-year-old patient with aplasticanemia treated with 300 mg cyclosporin A daily for77 days. PRA decreased from 12 to 3 of 20 cells.Slichter et al134 observed reversal of alloimmuniza-tion with antithymocyte globulin (ATG) and procar-bazine in an animal model.134 Transient improve-ment was shown in one clinical report.135

HIA Stripped Platelets. Shanwell et al136 treatedone HLA-aIloimmunized refractory patient withplatelets stripped of HLA antigens with citric acidtreatment at pH 2.8. Good posttransfusion incre-ments were documented on 2 occasions. Twoadditional patients have been treated success-

fuIly.m.138Vinblastine Loaded Platelets. Wong et al139

described the use of vinblastine-loaded platelets(VLP) in a patient with aplastic anemia, thrombocy-topenia, and bleeding who developed refractorinessassociated with LCT antibody. The infusion ofVLP24 hours before an infusion of pooled plateletsresulted in good increments. The effect was main-

PLATELET REFRAC AND Al IIMMUNIZATION

penia in breast cancer patients after high-dosechemotherapy and may decrease the need forplatelet transfusions. \42 rHuIL-11 (Neumega, Genet-

ics Institute, Cambridge, MA) has been approvedfor the prevention of severe thrombocytopenia andreduction in the need for platelet transfusions afterchemotherapy in patients with solid tumors receiv-ing nonmyeloablative chemotherapy.14\ IL-II is oflimited value in patients receiving myeloablativechemotherapy.141.142 Combinations of IL-6 andgranulocyte-colony-stimulating factor have beenobserved to enhance platelet recovery minimallyafter intensive chemotherapy, although they are notcurrently used for this indication. 142

tained only for a few days. t\ total of II courses ofVLP were infused, until the bleeding symptomsimproved. The authors based their study on thesuccessful results obtained by Ahn et all-lO inpatients with idiopathic thrombocytopenia, andhypothesized that VLP produce reticuloendothelialsystem blockade.

Thrombopoietin and platelet growth .factors.Hematopoietic growth factors, particularly throm-bopoietin (TPO) and interleukin-li (IL-ll), mayaccelerate endogenous platelet recovery, decreas-ing dependency on platelet transfusions. 141 TPO

increases the size and number of megakaryocytes,stimulates expression of platelet-specific markers(eg, CD4l and CD6l), and sensitizes platelets tothe aggregating effects of agonists.'41-'44 Plateletsremove TPO from the circulation, and it has beensuggested that platelet transfusions may blunt therecovery of megakaryocytes. TPO generates impres-sive thrombocytosis in mice and humans andmodifies thrombocytopenia after chemorac}iother-apy in animals.141.142 TPO has been available forclinical trials as recombinant human TPO (rHuTPO)or polyethyleneglycol-conjugated recombinant hu-man megakaryocyte growth and development fac-tor (PEG-rHuMGDF) and has been shown to returnplatelet counts to normal significantly faster, withhigher nadir counts.143.144 In patients receiving

nonmyeloablative chemotherapy (eg, carboplatinfor gynecologic malignancies), platelet counts startto increase after 3 to 4 days of TPO, with a peakeffect at 12 to 14 days, with up to a 50% decrease inthe need for platelet transfusion.141.144 However,neither minimal to no effect in platelet nadirs ortime to recovery nor a reduction in the need fortransfusion of platelets has been observed afterhematopoietic stem cell transplants or myeloabla-tive chemotherapy for acute leukemia.141 PEG-rHuMDGF has been shown to increase plateletpher-esis yield 3-fold in stimulated donors.141 However,all clinical trials with PEG-rHuMGDF were stoppedin 1998 because of development of antibodies tothe recombinant molecule that cross-react withendogenous TPO. These antibodies caused throm-bocytopenia in almost 10% of normal treatedvolunteer plateletpheresis donors. This problem hasnot been associated with rHuTPO.141.142

IL-ll works in concert with IL-3 to increase thesize, number, and ploidy of megakaryocyte colo-nies and to increase platelet counts.142 It has beenreported to prevent severe cumulative thrombocyto-

APPROACH TO THE REFRACTORY PATIENT

When refractoriness to platelet transfusion isdiagnosed, a systematic approach to elucidating itsorigin may help direct clinical management. Bloodproduct variables, clinical factors, and the possibil-ity of alloimmunization must all be considered.Physicians should be mindful that clinical factorsare more common causes of refractoriness thanalloimmunization and that in some patients thenature of the platelet concentrates themselves canbe significant as well. Furthermore, many clinicaland product variables (eg, fever, bone marrowtransplantation, ABO compatibility, days of stor-age) affect platelet recovery in some patients butnot in others.

If a patient develops refractoriness to platelettransfusions, a subsequent transfusion of an ad-equate dose of ABO-identical platelets that are asfresh as possible (preferably less than 2 days old)should be tried. In addition, verification of thetiming of the pretransfusion and posttransfusionplatelet counts is important. If a long time haselapsed between either count and the transfusion,the diagnosis of refractoriness may not be accurate.Treatment of clinical factors such as infection,DIC, veno-occlusive disease, and adjustment ofdrugs may be necessary to improve platelet countsand responsiveness to platelet transfusion. If refrac-toriness has occurred within 3 weeks of the firsttransfusion in a patient not previously exposed toforeign HLA antigens, an alloimmune cause is less

likely.Patients should be typed for class I HLA antigens

at the time of diagnosis. If a transfusion of anadequate dose of ABO-identical platelets stored for2 days or less does not result in an acceptable

192 DELAFlOR-WEISS AND MINTZ

should be provided. if possible. For individualpatients alloimmunized to both HLA and platelet-~pecific antigens. responsiveness to individual trans-fusions of known HLA and platelet antigen typesmay help determine which alloantibody or alloanti-bodies are principally responsible for refractori-ness.

Patients should be monitored every few weeksfor the persistence of alloantibody. because manypatients lose detectable alloimmunization even withcontinuing transfusions. Such patients may nolonger require HLA-matched or cross-match-compatible platelets. although failure to detectantibodies in vitro does not necessarily mean thatthere will be no alloantibody-mediated refractori-ness in vivo, owing to imperfect test sensitivity.

Patients who are refractory despite the use of theavailable strategies described above may requireadditional treatment or preventive strategies. al-though the efficacy of these modalities is unproven.A course of anti fibrinolytic therapy to arrest hemor-rhage is relatively safe and inexpensive and may behighly efficacious. IVIG (0.4 g/kg/d X 5 days).although expensive and of unproven efficacy. mayresult in good platelet recovery in some patients.Other options described above are rarely usedowing to their very limited reported success.

The transfusion of leukocyte-reduced cellularcomponents to prevent alloimmunization hasemerged as a cornerstone of addressing the prob-lem of platelet transfusion refractoriness. Althoughthis strategy is of proven value. some patients willcontinue to develop alloimrnune-mediated refracto-riness to platelet transfusion.5O The evaluation andmanagement of these patients poses a complicatedclinical challenge. The conditions that can contrib-ute to refractoriness have variable significance indifferent patients, and the available therapeuticoptions have variable success in different patients.Each refractory patient has a unique constellationof factors that contribute to her or his own insuffi-cient response to platelet transfusions.

increase in the platelet count. then testing for LCTantibodies is indicated. If LCT antibodies aredetected. then HLA-matched or cross-match-compatible platelets should be transfused. Some ofthese patients also may need ABO-identical orfresh platelets. Some patients may have rising titersof anti-A or anti-B. if they received ABO-incompatible platelet transfusions earlier in theirtreatment. and may thereafter require ABO-identical platelets. When only low-grade HLAmatches are available (eg. BX. C. or D). cross-match-<:ompatible platelets. if available. may pro-vide better responses. If only platelets that arepartially HLA-matched can be collected. it mayalso prove helpful to cross-match these products.Physicians should remember that HLA class Iantigens cause refractoriness in only a minority ofpatients who are alloimmunized to them.

If there is difficulty finding well-matched plate-lets for a particular patient. the use of plateletsexpressing HLA CREGs may benefit some pa-tients. Also. mismatching for antigens net wellexpressed by platelets (eg. HLA-B 12) may proveuseful. Patients not responsive to partially matchedplatelets may be evaluated for alloimmunization topublic epitopes (eg. Bw4 or Bw6). If tound.patients should be treated with platelets selected forcompatibility for the alloantibody detected.

When it is difficult to find well-matched plateletsand patients are not responding to transfusions ofplatelets expressing HLA CREGs. determination ofthe specificity of the HLA alloantibodies may helpguide the selection of products for transfusion insome patients. This strategy has proved particularlyuseful for patients alloimmunized to HLA-A2.

In patients who continue to be refractory toHLA-matched or cross-match-compatible platelettransfusions. other causes should be suspected. Ifplatelet transfusions well matched for HLA anti-gens do not result in expected in posttransfusionplatelet count increments. testing for platelet-specific antibodies should be considered. If suchantibodies are present. antigen-negative platelets

REFERENCESI. Duke WW: The relation of blood platelets to hemolThagic transfusion on hemolThage in patients with acute leukemia: An

disease: Description of a method for detemlining the bleeding autopsy study. Cancer 19:1937-1942.1966time and coagulation time and report of three cases of hemor- 3. Higby DI. Cohen E, Holland IF. et al: The prophylacticrhagic disease relieved by transfusion. lAMA 55:1185-1192, treatment of thrombocytopenic leukemic patients with platelets:1910 A double blind study. Transfusion 14:440-446. 1974

2. Han T, Stutzman L. Cohen E. et a1: Effect of platelet 4. Heckman KD, Weiner aI, Davis CS. et a1: Randomized

PLATELET REFRACTORINESS AND AllOIMMUNIZATION 93

stu~y of prophylactic platelet transfusion threshol~ ~uringin~u(;tion thcrapy for a~ult acute leukemia: 1(),OOO/~1 versus

20.(xx)/~I.JClinOncoI15:1143-1149,19975. Rebulla P, Finazzi G, Marangoni F. et al: A multicenter

ran~omize~ trial of the threshold fur prophylactic platelettransfusions in adults with acute myeloid leukemia. N Engl J

Me~337:1870-1875,1997n. Slichter SJ: Mechanisms and Management of Platelet

Refractoriness, in Nan(;e SJ (ed): Transfusion Medicine in the1990's. Arlington, VA, American Association of Blood Banks,

199(),pp95-1797. Herman J: Platelet transfusion therapy, in Mintz PD (ed):

Transfusion Therapy: Clinical Principles and Practice. Bethesda.MD, AABB Press, 1999, pp 65- 79

8. Aster RH, Jand! JH: Platelet sequestration in man. I.

Methods. J Clin Invest 43:843-845, 1964

9. Harker LA: The role of the spleen in thrombokinetics. JClin Lab Med 77:247-253,1971

10. Hanson SR, Slichter SJ: Platelet kinetics in patients withbone marrow hypoplasia: Evidence for a fixed platelet require-ment. Blood 66: 1105-1109,1985

II. O'Connell B, Lee EJ, Schiffer CA: The value of 10-

minute posttransfusion platelet counts. Transfusion 28:66-67,1988

12. Daly PA, Schiffer CA, Aisner J, et al: Platelet transfusion

therapy: One-hour posttransfusion increments are v:rtuable inpredicting the need for HLA-matched preparations. JAMA

243:435-438,198013. Bishop JF, Matthews JP, Yuen K, et ..1: The definition of

rcfractoriness to platelet transfusions. Transfus Med 2:35-41,1992

14. S(;hiffer CA, O'Connell, Lee EJ: Platelet transfusiontherapy for alloimmunized patients: Selective mismatching forHLA B 12, an antigen with variable expression on platelets.

Blood74:1172-1176,198915. Bishop JF, Matthews JP. McGrath K, et al: Factors

innuencing 20-hour increments after platelet transfusion. Trans-fusion 31:392-396,1991

16. del Rosario MLU, Kao KJ: Determination of the rate ofreduction in platelet counts in recipients of hematopoietic stemand progenitor cell transplant: Clinical implications for platelettransfusion therapy. Transfusion 37: 1163-1168, 1997

17. Friedberg RC, Mintz PD: Causes of refractoriness to

platelet transfusion. CUff Opin Hematol 2:493-498, 199518. Kao KJ, Mickel M, Braine HG, et al: White cell reduction

in platelet concentrates and packed red cells by filtration: Amulticenter clinical trial. Transfusion 35:13-19,1995

19. Enright H, Gemsheimer T, Woodson R, et al: Factors

inftuencing the response to platelet transfusion. Blood 90:

lI75A,I997(abstr)20. Moroff G, George VM, Siegl AM, et al: The inftuence of

irradiation on stored platelets. Transfusion 26:453-456, 198621. Shanwell A, Larsson S, Aschan J, et al: A randomized

trial comparing the use of fresh and stored platelets in thetreatment of bone marrow transplant recipients. Eur J Haematol

49:77-81, 199222. Peter-Salonen K, Bucher U, Nydegger UE: Comparison

of posttransfusion recoveries achieved with either fresh or

stored platelet concentrates. Blut 54:207-212, 1987

23. Norol F, Kuentz M, Cordonnier C, et al: Inftuence of

clinical status on the efficiency of stored platelet transfu!;ion. Br1 HaematoI86:125-129. 1994

24. Funheer R. Moddennan PW, Veldman H. et al: Detectionof platelet activation with monoclonal antibodies and flow

I:ytometry: Changes during platelet storage. Transfusion 30:20-

25,1990

25. Rinder HM, Murphy M. Mitchell lG. et al: Progressiveplatelet activation with !;torage: Evidence for shortened survivalof activated platelets after trdnsfusion. Transfusion 31 :409-414,

1991

26. Michelson AD. Barnard MR. Hechtman HB. et al: In vivotracking of platelets: Circulating degranulated platelets rapidlyloose surface P-selectin but continue to circulate and function.

Blood 84:320A. 1994 (abstr)

27. Murphy S, Gardner PH: Platelet preservation: Effect ofstorage temperature on maintenance of platelet viability-Deleterious effect of refrigerated storage. N Engl 1 Med

280:1094-1098,196928. Slichter SJ, Grabowski M. Townsend-McCall. et al:

Prospective randomized transfusion trial to directly comparefresh and stored apheresis platelets (AP) and pooled randomdonor platelet concentrates (PC) in thrombocytopenic patients.

Blood 92:2771 A. 1998 (abstr)

29. McFarland lG, Anderson AF, Slichter Sl: Factors influ-encing the transfusion re!;ponse to HLA-.~clected apheresisdonor platelets in patient!; refractory to random platelet concen-trates. Br 1 HaematoI73:380-386. 1989

30. Bishop lF, McGrath K. Wolf MM. et al: Clinical factors

influencing the efficacy of pooled platelet transfusions. Blood

71:383-387.198831. KJingemann HG, Self S, Banaji M, el al: Refractorine!;s

to random-donor platelet transfusions in patients with aplasticanemia: A multivariate analysis of data from 264 cases. Br 1

HaematoI66:115-12I,198732. Ishida A, Handa M, Wakui M. et al: Clinical factors

influencing posttransfusion platelet increment in patients under-going hematopoietic progenitor cell transplantation: A prospec-

tive analysis. Transfusion 38:839-847, 1998

33. Enright H. GernsheimerT. McCullough J, et al: Moderateto severe reactions to platelet transfusion: Experience of theTRAP multicenter trial. Blood 90: 1172A. 1997 (abstr)

34. Doughty HA, Murphy MF, Metcalfe P. et al: Relative

importance of immune and nonimmune causes of plateletrefractoriness. Vox Sang 66:200-205. 1994

35. Alcorta I, Pereira A. Ordinas A: Clinical and laboratoryfactors associated with platelet transfusion refractoriness: Acase-control study. Br 1 HaematoI93:220-224, 1996

36. Ogasawara K, Ueki J, Takenaka M. et al: Study on theexpression of ABH antigens on platelets. Blood 82:993-999,

1993

37. Dunstan RA, Simpson MB, Knowles RW. et al: Origin of

ABH antigens on human platelets. Blood 65:615-619, 1985

38. Aster RH: Effect of anticoagulant and ABO incompatibil-ity on recovery of transfused human platelets. Blood 26:732-

743,196539. Lee EJ, Schiffer CA: ABO compatibility can influence

the results of platelet transfusion: Results of a randomized trial.

Transfusion 29:384-389, 1989

40. Heal 1, Masel D, Rowe 1. et al: Circulating immune

194 DELAFLOA-WEISS AND MINTZ

HLA and platelet-specific antibodies in acute leukemia patientsalloimmunized by multiple transfusions. Br 1 Haematol 67:255-260, 1987

59. Pamphilon DH, Farrell DH, Donaldson C, et al: Develop-ment of Iymphocytotoxic and platelet-reactive antibodies: A

prospective randomized study in patients with acute leukemia.

VoxSang57:177-181,198960 Duquesnoy RJ: Donor selection in platelet transfusion

therapy of alloimmunized thrombocytopenic patients, in Green-walt Tl, lamieson GA (eds): The Blood Platelet in TransfusionTherapy. New York, NY, Alan R. Liss, 1978, pp 229-2~3

61. Peters AM, Porter IB, Saverymuttu SH, et al: Thekinetics of unmatched and HLA-matched I I I In-labelled homolo-

gous platelets in recipients with chronic marrow hypoplasia andanti platelet immunity. Br 1 Haematol60: 117-127, 1985

62. MacPherson BR, Hammond PB, Maniscalco CA: Alloim-

munization to public HLA antigens in multi-transfused plateletrecipients. Ann Clin Lab Sci 16:38-44, 1986

63. KickJerTS, Braine HG, Ness PM: The predictive value of

platelet crossmatching for alloimmunized patients. Transfusion25:385-389, 1985

64. Duquesnoy Rl. Filip 01, Rodey GE, et al: Successfultransfusion of platelets "mismatched" for HLA antigens to

alloimmunized thrombocytopenic patients. Am 1 Hematol 2:219-226, 1977

65. Dahlke MB, Weiss KL: Platelet transfusion from donorsmismatched for cross-reactive HLA antigens. Transfusion 24:

299-302, 1984

66. Hussein MA, Lee EJ, Fletcher R, Schiffer CA: The effectof Iymphocytotoxic antibody reactivity on the results of single

antigen mismatched platelet transfusions to alloimmunizedpatients. Blood 87:3959-3962,1996

67. Duquesnoy Rl. Filip 01, Aster RH: Influence of HLA-A2on the effectiveness of platelet transfusions in alloimmunized

thrombocytopenic patients. Blood 50:407-412, 1977

68. Liebert M, Aster RH: Expression of HLA-B 12 on

platelets, on lymphocytes and in serum: A quantitative study.Tissue Antigens 9: 199-208, 1977

69. Aster RH, Szatkowski N, Liebert M, et al: Expression of

HLA-BI2, HLA-B8, Bw4, and w6 on platelets. Transplant Proc

9:1695-1696,1977

70. Szatkowski NS, Aster RH: HLA antigens of platelets, IV:Influence of "private" HLA-B locus specificities on the expres-sion of Bw4 and Bw6 on human platelets. Tissue Antigens

15:361-368,198071. Atlas E, Freedman 1, Blanchette V, et al: Down regulation

of the anti-HLA alloimmune response by variable regionreactive (anti-idiotypic) antibodies in leukemic patients trans-fused with platelet concentrates. Blood 81 :538-542, 1993

72. Friedberg RC, Donnelly SF, Boyd lC, et al: Clinical andblood bank factors in the management of platelet refractorinessand alloimmunization. Blood 81:3428-3434,1993

73. O'Connell BA, Lee El, Rothko K, et al: Selection of

histocompatible apheresis platelet donors by cross-matchingrandom donor platelet concentrates. Blood 79:527-531,1992

74. MoroffG, Garratty G, Hal 1M, et al: Selection of plateletsfor refractory patients by HLA matching and prospective

crossmatching. Transfusion 32:633-640, 1992

75. Gelb AB, Leavitt AD: Crossmatch-compatible platelets

improve corrected count increments in patients who are refrac-

complexes involving the ABO system after platelet transfusion.Br J HaematoI85:566-572. 1993

41. Heal JM. Blumberg N. Masel D. et al: An evaluation of

crossmatching. HLA. and ABO matching for platelet transfu-,ions to refractory patients. Blood 70:23-30. 19R7

42. Brand r\. Sintnicolaas K. Claas FHJ. et al: ABH antibod-ies cau~ing platelet tran~fusion refractoriness. Tran~fu~ion 26:

463-466. 1986

43. Carr R. Hutton J. Jenkins J. et al: Tran~fu.,ion ofABO-mismatched platelets Icads to early platelet refractoriness.Br J HaematoI75:40R-413. 1990

44. Skogen B. Rossebo Hansen B. Husebekk A. et al:Minimal expression of blood group A antigen on thrombocytesfrom A2 individuals. Transfusion 28:456-459. 1988

45. Hogge DE. McConnell M. Jacobson C. et al: Platelet

refractoriness and alloimmunization in pediatric oncology andbone marrow transplant patients. Transfusion 35:645-652. 1995

46. Sintnicolaas K. vaan Marwijk M. van Kooij HC. et al:

Leukocyte depiction of random single-donor platelet transfu-sions does not prcvent secondary human Icukocyte antigen-

alloimmunization and refractoriness: A randomized prospectivestudy. Blood 85:824-828.1995

47. Yankee RA. Graff KS. Dowling R. et al: Selection of

unrelated compatible platelet donors by lymphocyte HLAmatching. N Engl J Med 288:760-764. 1973

48. Schiffer CA. O'Connell B. Lee EJ: Platelet tra."sfusiontherapy for alloimmunized patients: Selective mismatching forHLA B 12. an antigen with variable expression on platelets.Blood 74: 1172-1176. 1989

49. Lee EJ. Schiffer CA: Serial measurement of Iymphocyto-toxic antibody and response to non matched platelet transfusionsin alloimmunized patients. Blood 70: 1727-1729. 19R7

50. The Trial to Reduce Alloimmunization to Platelets Study

Group: Leukocyte reduction and ultraviolet B irradiation ofplatelets to prevent alloimmunization and refractoriness toplatelet transfusions. N Engl J Med 337: 1861-1869. 1997

51. Kao KJ. Cook DJ. Scomik JC: Quantitative analysis of

platelet surf.ace HLA by W6/32n anti-HLA monoclonal anti-body. Blood 68:627-632.1986

52. Tew JG. Thorbecke GH. Steinman RM: Dendritic cells inthe immune-response: Characteristics and recommendcd nomen-clature. J Reticuloendothelial Soc 31 :371-380. 1982

53. Oksanen K. Kekomaki R. Ruutu T. et at: Prevention ofalloimmunization in patients with acute leukemia by use ofwhite cell-reduced blood components: A randomized trial.Transfusion 31:588-594.1991

54. van Marwijk Kooy M. van Prooijen HC. Moes M. et al:Use of leukocyte-depleted platelet concentrates for the preven-tion of refractoriness and primary HLA alloimmunization: A

prospective. randomized trial. Blood 77:201-205.199155. Kripke ML: Immunologic unresponsiveness induced by

ultraviolet radiation. Immunol Rev 80:87-102. 1984

56. Novotny VMJ. van Doom R. Witvliet MD. et al:Occurrence of allogeneic HLA and non-HLA antibodies aftertransfusion of prestorage filtered platelets and red blood cells: A

prospective study. Blood 85:1736-1741.199557. Taanig E. Skibsted L: The frequency of platelet alloanti-

bodies in pregnant women and the occurrence and managementof neonatal alloimmune thrombocytopenic purpura. Obstet

Gynecol Surv 45:525-525. 1990

58. Murphy MF. Metcalfe P. Ord J. et al: Disappearance of

PLATELET REFRAC )RINESS AND ALLOIMMUNIZATION

tory to randomly o;clccted platclets. Transfu.o;ion 37:624-630,1997

7(;. Friedherg RC, Donnelly SF, Mintz PO: Independent roles

for platclct crossmatching and HLA in the selection of plateletsfor alloimmuni7.ed patients. Transfusion 34:215-220,1994

77. Rachel JM, Sinor LT, Tawfik OW, et al: A solid phase red

ccll adherence test for platelet cross-matching. Med Sci Lab

42:194-195. 19li5

78. Sintnicolaas K. Lowenberg B: A flow cytometric platelet

immunofluorescence crossmatch for predicting successful HLA-matched platelet transfusions. Br J Haematol 92:1005-1010,1996

79. Wu KK, Hoak JC, Koepke JA, et al: Selection of

compatible platelet donors: A prospective evaluation of three

cross-matching techniques. Transfusion 17:638-643, 1977

80. Rachel JM, Summers TC, Sinor LT, et al: Use of a solidphase red blood cell adherence method for pretransfusionplatelet compatibility testing. Am J Clin Pathol 9():63-68, 1988

Ii I. Bock M, Heim MU, Schleich I, et al: Platelet crossmatch-

ing with Capture-P: Clinical relevance. Infusiontherapie 4:183-

185,198982. Kiefel V. Santoso S, Weisheit M, et al: Monoclonal

antibody-specific immobilization of platelet antigens (MAIPA):A new tool for the identification of platelet-reactive antibodies.

Blood70:1722-1726,1987 .li3. Moore SB, DeGoey SR: Serum platelet antibody testing:

Evaluation of ~olid phase enzyme immunoassay and compari-son with indirect immunofluorescence. Coagulation TransfusMed 109:190-195, 1998

84. Brubaker DB, Romine M: Relationship of HLA and

platelet-reactive antibodies in alloimmunized patients refractoryto platelet therapy. Am J Hematol 26:341-352, 1987

85. Schnaidt M, Northoff H, Wernet D: Frequency andspecificity of platelet-specific alloantibodies in HLA immunizedhaematologic oncologic patients. Transfus Med 6: 111-114, 1996

86. Wernet D. SchnaiJt M, Mayer G, et al: Serologicalscreening, using three different test systems of platelet trans-fused patients with hematologic oncologic disorders. Vox Sang

65:108-113.1993

87. Freedman J, Hornstein A: Simple method fordifferentiat-ing between HLA and platelet-specific antibodies by flowcytometry. Am J HematoI38:314-320, 1991

88. Murphy MF, Waters AH: Immunologicat aspects ofplatelet transfusions. Br J Haematol60:409-414, 1985

89. Kickier TS. Kennedy SD, Braine HG: Alloimmunizationto platelet-specific antigens in glycoproteins IIb-lIIa and Ib/IXin multiply transfused thrombocytopenic patients. Transfusion

30:622-625,199090. Kurz M, Grinix H, Hocker p, et al: Specificities of

anti-platelets antibodies in multitransfused patients with hemato-oncological disorders. Br J HaematoI95:564-569, 1996

91. Godeau B, Fromont P, SerorT, et al: Platelet alloimmuni-zation after multiple transfusions: A prospective study of 50patients. Br J Haematol 31 :395-400, 1992

92. Meenaghan M, Judson p, Yousaf K, et al: Antibodies toplatelet glycoprotein V in poly transfused patients with hemato-

logical disease. Vox Sang 64: 167 -170, 1993

93. Kekomaki S, Volin L, Koistinen p, et al: Successfultreatment of platelet transfusion refractoriness: The use ofplatelet transfusions matched for both human leukocyte antigens

(HLA) and human platelet alloantigens (HPA) in alloimmunizedpatients with leukemia. Eur 1 Haematol 60: 112-118. 1998

94. Skogen B. Bellissimo DB, Hessner Ml, et al: Rapiddetennination of platelet alloantigen genotypes by polymerasechain reaction using allele-specific primers. Transfusion 34:955-

960. 199495. Van tier Lelie 1, van der Plas-Van Dallen CM, von dem

Borne AEGKr: Platelet autoantibodies in septicaemia. Br 1HematoI58:755-758, 1984

96. Hebart H, Einsele H, Klein R, et al: CMV infection after

allogeneic bone marrow transplantation is associated with theoccurrence of various autoantibodies and monoclonal gammapa-thies. Br 1 HematoI95:138-144, 1996

97. Sivakumaran M, Hutchinson RM, Pringle H. et al:Thrombocytopenia following autologous bone marrow transplan-tation: Evidence for autoimmune aetiology and B-cell clonalinvolvement. Bone Marrow Transplant 15:531-536, 1995

98. Bierling P. Cordennier C, Fremont P. et al: Acquiredautoimmune thrombocytopenia after allogeneic bone marrowtransplantation. Br 1 Haematol 59:643-646, 1985

99. Bussel IB, Pham LC, Aledort L, et al: Maintenancetreatment of adults with chronic refractoty immune thrombocy-topenic purpura using repeated intravenous infusions of gamma-

globulin.Blood72:121-127,1988100. McFarland IG: Laboratory investigation of drug-

induced immune thrombocytopenias. Transfus Med Rev 7:275-

287,1997101. Kutti 1, Zaroulis CG, Safai-Kutti S, et al: Evidence that

circulating immune complexes remove transfused platelets fromthe circulation. Am 1 Hematolll:255-259, 1981

102. Heal 1M. Cowles 1, Masel D, et al: Antibodies to plasmaproteins: An association with platelet transfusion refractoriness.

Br 1 Haematol 80:83-90. 1992103. Funke I. Wiesneth M. Koerner K. et al: Autologous

platelet transfusion in alloimmunized patients with acute leuke-

mia. Ann Hematol71: 169-173,1995104. Torretta L. Perotti C. Pedrazzoli P. et al: Autologous

platelet collection and storage to support thrombocytopenia inpatients undergoing high-dose chemotherapy and circulatingprogenitor cell transplantation for high risk breast cancer. Vox

Sang 75:224-229. 1998105. Bartholomew lR. Salgia R, Bell WR: Control of bleed-

ing in patients with immune and non immune thrombocytopeniawith aminocaproic acid. Arch Intern Med 149:1959-1961. 1989

106. Garewa! HS, Durie BGM: Anti-fibrinolytic therapy withaminocaproic acid for the control of bleeding in thrombocytope-

nic patients. Scand 1 Haematol 35:497-500. 1985107. Gardner FH. Helmer RE: Aminocaproic acid: Use in

control of hemolThage in patients with amegakaryocytic throm-bocytopenia. lAMA 243:35-37, 1980

108. Gallardo RL, Gardner FH: Antifibrinolytic therapy forbleeding control during remission induction for acute leukemia.

Blood 62:202A. 1983 (abstr)109. Avvisati G, ten Cate lW, Buller HR, et al: Traxenamic

acid control of haemorrhage in acute promyelocytic leukaemia.Lancet2:122-124,1989

110. Fehr 1. Hoffman V, Kappeler U: Transient reversal of

thrombocytopenia in idiopathic thrombocytopenic purpura byhigh-dose intravenous gammaglobulin. N Engl 1 Med 306: 1254-1258.1982

III. Kekomaki R. Elfenbein G. Gardner R. et al: Improved

196 DELAFLOR-WEISS AND MINTZ

rc~ponse of patients refractory to random-donor platelet tran~fu-,ions by intravenous gamma globulin. Am J Med 76: 199-2031984

tion to remove platelet alloantibodies. Transfusion 33: 192-194.1993

129. Franchini E: Treatment of refractoriness to platelettransfusion by protein A column therapy: The role of the IgGsubclasses. Transfusion 33:961-962.1993

130. Nagasawa T. Kim BK. Baldini MG: Temporary suppres-sion of circulating antiplatelet alloantibc)dies by the massiveinfusion of fresh. stored. or lyophilized platelets. Transfusion

18:429-435. 1978131. Slichter SJ. Deeg HJ. Storb R: Prevention and reversal

of platelet alloimmunization with cyclo.sporin A. Blood 58:186A. 1981 (abstr)

132. Yamamoto M. Ideguchi H. Nishimura J. et al: Treatmentof platelet-alloimmunization with cyclosporin A in a patientwith aplastic anemia. Am J Hematol 33:220-221. 1990

133. Tilly H. Azagury M. Bastit D. et al: Cyclosporin fortreatment of life-threatening alloimmunization. Am J Hematol

34:75-76.1990134. Slichter SJ. Weiden PL. Kane Pl. et al: Approaches to

preventing or reversing platelet alloimmunization using animalmodels. Transfusion 28: 103-108. 1988

13.1. Sabbe UM. Class FHJ. Haak HL. et al: Anti-thymocyteglobulin (ATG) can eliminate platelet refractoriness. Blut 42:331-335. 1981

136. Shanwell A. Sallander S. Olson I. et al: An alloimmu-nized. thrombocytopenic patient successfully transfused withacid-treated. random-donor platelets. Br J Haematol 79:462-

465.1991137. Novotny VM. Huizinga TW. van Doom R. et al: HLA

class I-eluted platelets as an alternative to HLA-matchedplatelets. Transfusion 36:438-444. 1996

138. Castro E. Mancunill J. Barea L. et al: Acid elution ofplatelets HLA-ciass I antigens in the treatment of a refractorypatient. Br J HaematoIIOO:245-246. 1998

139. Wong P. Hiruma K. Endoh N. et al: Vinblastine-loadedplatelet transfusion in an alloimmunized patient. Br J Haematol

65:380-381.1987140. Ahn YS. Byrnes JJ. Harrington WJ. et al: The treatment

of idiopathic thrombocytopenia with vinblastine-loaded plate-lets. N EngIJMed298:1101-1107. 1978

141. Kuter DJ. Cebon J. Harker LA. et al: Platelet growthfactors: Potential impact on transfusion medicine. Transfusion

39:321-332.1999142. Whitsett CF: The role of hematopoietic growth factors

in transfusion medicine. in Mintz PD (ed): Transfusion Therapy:

Clinical Principles and Practice. Bethesda. MD. AABB Press.1999. pp 399-427

143. Kaushansky K: Thrombopoietin. N Engl J Med 339:746-

754.1998144. Price TH. Adamson JW: Hematopoietic growth factors

in transfusion medicine. in Ganser A. Hoelzer D (eds): Cyto-kines in the Treatment of Hematopoietic Failure. New York. NY.Marcel Dekker. 1999. pp 327-357

145. McFarland J: Platelet immunology and alloimmuniza-tion. in Rossi EC. Simon TL. Moss GS. et aI (eds): Baltimore.MD. Williams & Wilkins. 1996. pp 231-244

146. Santoso S. Kiefel V: Human platelet-specific alloanti-gens: Update. Vox San!! 74:249-253. 1998 (SUDDI2)

112. Bierling P. Cordonnier C, Rodet M. et al: High do~eintravenous gammaglobulin and platelettr;lnsfusions in leukae-mic HLA-immuni7.ed patients. Scand J Haematol 33:215-220.19114

113. Junghans RP. Ahn YS: High-do~e intravenous gamma

globulin to suppres~ alloimmune de~truction of Jonor platelets.Am J Med 76:204-208. 1984

t 14. Zeigler ZR. ShaJduck RK. Ro~enfeld CS. et al: Intrave-

nous gamma globulin decreases platelet associated IgG andimproves transfusion responses in platelet refractory _tales. AmJ Hematol38 15-23. 1991

115. Atrah HI. Sheehan T. Gribben J. et al: Improvement of

post platelet transfusion increments following intravenous immu-noglobulin therapy for leukaemic HLA-immunized patients.Scand J Haematol 36: 160-164, 1986

116 Becton OL. Kinney TR. Schaffee S. et al: High-Jose

intravenous immunoglobulin for severe platelet alloimmuniza-lion. Pediatrics 74:1 120-1123. 1984

117. Zeigler ZR. Shadduck RK. Rosenfeld CS. ct al: High-Jose intravenous gamma globulin improves responses to single-donor platelets in patients refractory to platelet transfusion.Blood 70:1433-1436. 1987

118. KicklerT. Braine HG. Piantadosi S. etal: A raodomized~placebo-controlled trial of intravenous gammaglobulin in alloim-munized thrombocytopenic patients. Blood 75:313-316. 1990

119. Kurtzberg J. Friedman HS. Kinney TR. et al: Treatmentof platelet alloimmunization with intravenous immunoglobulin:Two case reports and review of the literature. Am J Med

83:30-33. 1987120. Knupp C. Chamberlain JK. Raab SO: High-Jose intrave-

nous gamma globulin in alloimmunized platelet transfusionrecipients. Blood 65:776. 1985

121. Lee EJ. Norris O. Schiffer CA: Intravenous immuneglobu~i!, for patients alloimmunized to random donor platelet

transfusion. Transfusion 27:245-247. 1987122. Schiffer CA. Hogge DE. Aisner J. et al: High-dose

intravenous gammaglobulin in alloimmunized platelet transfu-sion recipients. Blood 64:937-940,1984

123. Kekomaki R. Myllyla G: Effect of nonnal serum on thebinding of specific antibodies and platelet-unrelated immunecomplexes to human platelets. Scand J Immunol 9:527-535.1979

124. Bensinger WI. Buckner CD. Clift RA. et ;)1: Plasma

exchange for platelet alloimmunization. Transplantation 41 :602-

605.1986125. Heddle NM. KJama L. Kelton JG. et al: The use of

anti-D to improve post-transfusion platelet response: A random-ized trial. Br 1 Haematol 89: 163-168. 1995

126. Christie 01. Howe RB. Lennon SS. et al: Treatment ofrefractoriness to platelet transfusion by protein A column

therapy. Transfusion 33:234-242. 1993127. Lopez-Plaza I. Miller K. Leitman SF: Ineffectiveness of

protein A adsorption in the treatment of platelet refractoriness. 1Clin Apheresis 7:33. 1992 (abstr)

128. Rock G: The application of protein A immunoadsorp-