Refractoriness to platelet transfusion

INTRODUCTION — This topic discusses the causes, evaluation, and management of refractoriness to platelet transfusion, as defined below.

Separate topics discuss:

●General principles of platelet transfusion – (See "Platelet transfusion: Indications, ordering, and associated risks".)

●Evaluation of thrombocytopenia

•Child – (See "Approach to the child with unexplained thrombocytopenia".)

•Adult – (See "Diagnostic approach to thrombocytopenia in adults".)

●Drug-induced thrombocytopenia – (See "Drug-induced immune thrombocytopenia".)

DEFINITION AND PREVALENCE — Platelet refractoriness refers to a suboptimal response to platelet transfusions (smaller than expected platelet count increment after transfusion) on more than one occasion [1]. In routine clinical practice, failure to observe a post-transfusion platelet increment of at least 10,000/microL is considered suspicious for refractoriness [2].

To improve the consistency of reporting in research studies, it is standard to adjust the absolute platelet increment to account for the number of platelets transfused and the size of the transfusion recipient. The corrected count increment (CCI) is the most used such metric. (See 'CCI' below.)

●The 1997 TRAP trial, which evaluated the role of leukoreduction (post-storage) in preventing platelet refractoriness (see 'Prevention' below), defined platelet refractoriness as a CCI <5000/microL after two sequential transfusions of ABO-compatible platelets [3,4].

●A 2018 guideline from American Society of Clinical Oncology (ASCO) on platelet transfusion in patients with cancer endorsed the definition of platelet refractoriness as a CCI <5000/microL following at least two ABO-compatible platelet transfusions using platelets stored <72 hours, but the guideline emphasized that evidence is weak, inter-individual variation occurs, and this criterion should not take the place of physician judgment [5].

The incidence of platelet refractoriness depends on the population, with a higher likelihood in individuals who are seriously ill, heavily transfused, or both. In the PLADO trial, which evaluated platelet dosing strategies in individuals with thrombocytopenia due to chemotherapy or hematopoietic stem cell transplant (HSCT), 102 of 734 patients who received at least two platelet transfusions (14 percent) developed platelet refractoriness by the TRAP definition [6]. Alloimmunization occurred in 40 of 816 evaluable participants (5 percent). Observational studies have reported platelet refractoriness in 13 to 44 percent of participants [2,6-10].

FACTORS ASSOCIATED WITH PLATELET REFRACTORINESS — Refractoriness to platelet transfusion is often multifactorial.

The causes are generally divided into non-immune and immune (table 1) [2]. Most cases are non-immune [6].

Contribution of both non-immune causes and alloimmunization is especially likely in individuals who have been pregnant, and in heavily transfused individuals, such as people with hematologic malignancies or after HSCT. These individuals may have infection or chemotherapy-induced thrombocytopenia causing reduced platelet lifespan plus anti-human leukocyte antigen (anti-HLA) alloantibodies causing alloimmune refractoriness.

Non-immune causes — Non-immune causes account for most episodes of platelet refractoriness. These include conditions associated with rapid platelet consumption or sequestration from the circulating blood volume. Transfused platelets can be consumed or sequestered so rapidly that the platelet count one hour after transfusion fails to increase substantially, but usually with non-immune causes the platelet count increment is higher at 10 minutes to 1 hour and lower at 24 hours. (See 'Determining the cause' below.)

Common examples include [11]:

●Sepsis, infection, or fever – Fever and infections have been associated with reduced platelet increments [2]. It is unclear whether fever is a marker of severe infection rather than an independent risk factor for platelet refractoriness [12,13]. Platelet interactions with bacteria, neutrophils, and activated endothelial cells may also contribute to reduced platelet survival [11]. Sepsis may be accompanied by other platelet consumptive processes such as disseminated intravascular coagulation (DIC) or hemophagocytosis [14-16]. (See "Evaluation and management of disseminated intravascular coagulation (DIC) in adults".)

●Splenomegaly – Platelets pool in the spleen, and splenomegaly is well established as a cause of reduced platelet count increments following platelet transfusion. Under normal conditions, approximately one-third of an individual's platelets are sequestered in the spleen, where they are in equilibrium with the circulating platelet pool. In cases of extreme splenomegaly, splenic sequestration can be increased to 90 percent (figure 1). It can be expected that a large fraction of the transfused platelets will be sequestered in an enlarged spleen [17]. (See "Splenomegaly and other splenic disorders in adults", section on 'Hypersplenism'.)

In a study of platelet responses in individuals with bone marrow suppression (mostly leukemia), prior splenectomy correlated with a greater platelet count increment, further supporting the idea that an enlarged spleen would lower the increment [12].

●DIC – DIC results in an increased rate of platelet consumption and refractoriness to platelet transfusion. This is a particular problem in cases of acute promyelocytic leukemia where DIC may result in an increased risk of fatal hemorrhage [18]. (See "Evaluation and management of disseminated intravascular coagulation (DIC) in adults" and "Clinical manifestations, pathologic features, and diagnosis of acute promyelocytic leukemia in adults", section on 'Coagulopathy and APL'.)

●Bleeding – Bleeding is considered an independent risk factor for reduced platelet increment [2]. Presumably the mechanism involves platelet loss and consumption.

●Medications – Numerous medications can cause drug-induced thrombocytopenia, by a variety of immune and non-immune mechanisms. Heparin and amphotericin are two notable examples. Details and lists of implicated drugs are presented separately. (See "Drug-induced immune thrombocytopenia", section on 'List of drugs'.)

●HSCT – Additional non-immune causes in individuals who have undergone HSCT may include hepatic sinusoidal obstruction syndrome (hepatic veno-occlusive disease [VOD]) and graft-versus-host disease (GVHD), although not all studies have documented these associations [2,8,12,19,20].

Alloimmunization — Alloimmunization (development of antibodies directed at blood cell antigens) accounts for a minority of cases of platelet refractoriness. Exposure to blood cell antigens can occur through prior transfusions, pregnancy, or hematopoietic stem cell (HSC) transplantation (HSCT).

●Anti-HLA – The most common types of epitopes are from the human leukocyte antigen (HLA) system, which are present on platelets, white blood cells (WBCs), and various immune tissues. Antibodies targeting HLA-A and HLA-B epitopes are typically implicated; anti-HLA-C antibodies have been reported as a rare cause of alloimmune refractoriness, but anti-HLA-C are typically not considered a cause of platelet refractoriness [21].

Pregnancy is the most important risk factor for HLA alloimmunization. In a 2009 study of >8000 blood donors, circulating anti-HLA antibodies were present in approximately 24 percent of previously pregnant female blood donors who had anti-HLA [3]. The rate of anti-HLA increased according to the number of pregnancies, from 1.7 percent with one pregnancy to 32 percent with four or more pregnancies [22]. In contrast, rates of anti-HLA antibodies in previously transfused and nontransfused male donors were 1.7 and 1.0 percent, respectively [3].

HLA alloimmunization can also be stimulated by transfusion; the alloimmunization risk is substantially higher if non-leukoreduced units are transfused. Alloantibodies to foreign ABO antigens develop universally during early life as an immune response to the gastrointestinal microbiome. (See "Red blood cell antigens and antibodies", section on 'ABO antibodies'.)

HLA alloimmunization does not always cause clinical platelet refractoriness. In the 1997 TRAP trial, which evaluated the role of pre-storage leukoreduction in preventing platelet refractoriness, alloantibodies developed in 17 to 45 percent of participants depending on treatment assignment, but only 10 percent of patients developed refractoriness [3].

●Anti-HPA – Less commonly seen epitopes in platelet immune refractoriness are human platelet-specific antigens (HPA). Some platelet antigens are universally expressed and do not lead to alloantibody formation, but others are polymorphic; exposure to foreign HPA can lead to alloantibody formation. The most relevant HPA are GPIa, GPIb, GPIIb, GPIIIA, and CD109.12 [4]. In the TRAP trial, antibodies to HPA epitopes developed 6 to 11 percent of the patients [3]. As would be expected, the prevalence of anti-HPA was not altered by pre-storage leukoreduction.

Although platelets express only HLA Class I antigens, HLA Class II antigens present on WBCs may be essential for the development of alloimmunization to HLA Class I antigens. Foreign HLA Class II antigens can be introduced through prior transfusion or during pregnancy [3,7].

MONITORING AND EVALUATION

When to suspect — Refractoriness to platelet transfusion may be suspected in any individual in whom there is failure to have the expected platelet count increase following platelet transfusion. However, before making the diagnosis, it is important to determine the post-transfusion platelet count increment for at least two transfusions. (See 'Post-transfusion platelet count' below.)

Obtaining an immediate post-transfusion platelet count is not required for individuals in whom refractoriness is not suspected.

Post-transfusion platelet count — The immediate post-transfusion platelet count increment is used clinically to establish the diagnosis of platelet refractoriness and to help determine the cause (algorithm 1). Generally, this refers to a platelet count obtained 10 minutes to 1 hour after the transfusion is completed.

A repeat count is obtained after 24 hours, and the pattern is used to determine the cause of refractoriness. The rationale for obtaining the immediate post-transfusion platelet count and a count at 24 hours is that in non-immune mechanisms, the platelet count initially rises, but platelets are consumed. In contrast, in alloimmune refractoriness, the platelet count never rises (figure 2). (See 'Determining the cause' below.)

CCI — The corrected count increment (CCI) is primarily used for research studies to allow comparable assessments of platelet count increment.

Determining the cause — The cause of platelet refractoriness determines management. (See 'Management' below.)

The figure illustrates typical patterns with non-immune and alloimmune refractoriness (figure 2). The reliability of these patterns for distinguishing between non-immune and alloimmune refractoriness have been questioned [1].

●In non-immune refractoriness, the immediate post-transfusion platelet increment tends to be reasonably high, with a return to the baseline platelet count within 24 hours. This is also referred to as normal platelet recovery with reduced platelet survival.

●In alloimmune refractoriness, the immediate post-transfusion platelet generally does not increase (or increases only slightly).

•Anti-HLA antibodies – The second step in determining refractoriness is to assay for anti-human leukocyte antigen (anti-HLA) alloantibodies. The calculated panel reactive antibody (cPRA) provides a score from 0 to 100 percent (the lower the score, the less reactivity). In one study, the likelihood of identifying crossmatch compatible platelets was greatest in individuals with a cPRA <70 percent [23]. If anti-HLA antibodies are identified in a patient with platelet refractoriness, they can be avoided in subsequent transfusions. (See 'Positive PRA (anti-HLA alloantibodies)' below.)

Panel reactive antibody (PRA) testing can be omitted if a non-immune cause is considered likely, such as in a patient with a large platelet count increment immediately after transfusion who is known to have one or more potential causes of non-immune platelet refractoriness (fever/sepsis, splenomegaly, disseminated intravascular coagulation [DIC]).

Methods to detect the presence and specificity of anti-HLA antibodies include lymphocytotoxicity, immunoassays such as enzyme-linked immunosorbent assays (ELISAs), and flow cytometric immunofluorescence testing, depending on the laboratory performing testing [24].

•Anti-HPA antibodies – If the screen for anti-HLA antibodies is negative, testing for antibodies to the human platelet antigen system (anti-HPA antibodies) can be performed; use is institution-dependent. In institutions where this is not done, platelet crossmatching can be used.

Not all detectable alloantibodies are clinically significant; HLA antibodies that are capable of activating complement may have a greater likelihood of causing platelet refractoriness. Higher levels of antibody may also be associated with greater risk of refractoriness [25].

PREVENTION — The major preventive intervention is to minimize risk factors, especially alloimmunization. Indications for platelet transfusion should be reviewed and unnecessary transfusions avoided. (See "Indications and hemoglobin thresholds for RBC transfusion in adults" and "Platelet transfusion: Indications, ordering, and associated risks".)

Platelets express human leukocyte antigen (HLA) class I and ABO antigens on their surface, as well as platelet specific antigens; they do not express HLA class II or Rh antigens. A 2001 guideline from the American Society of Clinical Oncology (ASCO), updated in 2018, endorsed pre-storage leukoreduction for all cellular blood products in individuals with acute myeloid leukemia (AML) receiving chemotherapy, as well as in individuals with other hematologic malignancies [5,26].

Major preventive approaches are to use leukoreduced products when possible. Medical conditions that decrease platelet survival should be treated.

●Leukoreduction – Leukoreduction decreases exposure to HLA Class I antigens on white blood cells. Many countries practice universal pre-storage leukoreduction; in the United States, most cellular products undergo pre-storage leukoreduction. Clinicians caring for patients who may require platelet transfusions should familiarize themselves with local practices and ensure that pre-storage leukoreduction is used for individuals who may require multiple platelet transfusions. Details and other indications are discussed separately. (See "Practical aspects of red blood cell transfusion in adults: Storage, processing, modifications, and infusion", section on 'Pre-storage leukoreduction'.)

The benefit of leukoreduction in reducing alloimmunization was demonstrated in the 1997 TRAP trial, which randomly assigned 530 patients with no detectable preexisting alloantibodies who were receiving induction chemotherapy for AML to receive platelet transfusions prepared in one of four ways, and documented the following incidences of platelet refractoriness during the eight weeks of the trial [3]:

•Non-leukoreduced pooled platelets (from whole blood donations) – 16 percent (95% CI, 10 to 23 percent)

•Ultraviolet B (UVB)-irradiated pooled platelets – 10 percent (6 to 16 percent)

•Leukoreduced pooled platelets – 7 percent (4 to 13 percent)

•Leukoreduced apheresis platelets – 8 percent (4 to 14 percent)

Rates of refractoriness were higher in females who had ever been pregnant; rates were similar between males and never-pregnant females [3,6]. Rates of functional anti-HLA antibodies (measured in a lymphocyte killing assay) were approximately two-fold higher than rates of refractoriness in all groups.

●ABO matching – It has been suggested that ABO matching can reduce the risk of HLA alloimmunization, although data are very limited [27]. ABO matching (with ABO identical or ABO compatible platelets) prevents platelet clearance by pre-formed antibodies in the recipient directed against ABO antigens. Many institutions give priority to patients at highest risk for platelet refractoriness when determining which patients should receive ABO compatible platelets (prioritizing hematopoietic stem cell transplant recipients). In many institutions, low inventory does not allow universal ABO matching. Observational studies have generally found better platelet count increments with transfusion of ABO compatible platelets compared with ABO incompatible platelets [2,28-34]. Rh matching is not a consideration, as platelets do not express Rh antigens.

●Treatment of other conditions – It is prudent to treat any underlying conditions that contribute to decreased platelet survival by non-immune mechanisms. (See 'Management of non-immune refractoriness' below.)

MANAGEMENT — Management depends on the underlying cause of platelet refractoriness (algorithm 1). (See 'Factors associated with platelet refractoriness' above.)

For individuals with alloimmunization, avoiding the implicated platelet antigens is most helpful. For those with non-immune mechanisms, the underlying cause of platelet consumption should be addressed.

For the most part, immunosuppressive therapies have not proven particularly helpful in treating platelet refractoriness. (See 'Limited role of immunosuppressive therapies' below.)

Management of alloimmune refractoriness

Positive PRA (anti-HLA alloantibodies) — The panel reactive antibody (PRA) or calculated PRA (cPRA) is the standard test for anti-human leukocyte antigen (anti-HLA) antibodies. PRA testing was previously performed using a lymphocyte cytotoxicity assay; most modern laboratories screen for HLA antibodies with a flow cytometric method using HLA antigen-coated beads.

Patients with clinical refractoriness who have anti-HLA antibodies are transfused with either HLA-matched platelets, HLA antigen-exclusion platelet units, or crossmatched platelet units (algorithm 1). Efficacy is considered comparable, and the decision of which to use is based on availability, inventory, cost, and other factors [4]. HLA-matched platelets are not indicated when the PRA is low (eg, <20 percent).

●HLA-matched platelets – Once a patient's HLA class I (A and B) type has been determined, platelet units with identical or similar HLA class I antigens can be selected for transfusion. Many apheresis platelet donors who donate regularly are HLA typed by the blood donation center, and information on their HLA type is readily available.

The degree of similarity is referred to as the HLA "match grade" [4]. The better the match, the more likely that the patient will have an adequate increment to transfused platelets. Platelet units with a grade A match (all four HLA A and B antigens identical between donor and recipient) are optimal. However, grade A match units are challenging to find, and it is common to use grade B match units (one antigen out of four is mismatched). Transfusing HLA-matched platelets below grade B match (C or D matches) often results in a poor platelet increment.

●HLA antigen-exclusion platelets – Antigen-negative platelets are not fully HLA-matched, but they lack the antigen(s) thought to be implicated in platelet refractoriness.

●Crossmatch-compatible platelets – Crossmatch-compatible platelets have been used as a more convenient and less expensive approach than HLA-matched platelets, although in some institutions they may be more expensive. Crossmatch compatible platelets are identified by crossmatching apheresis platelet units with the patient's serum [34-36]. Crossmatching can be performed in a few hours, as opposed to the days it may take to identify HLA-matched platelets [37]. Crossmatching can also benefit patients with uncommon HLA types for whom it would be very difficult to find an HLA-matched donor. Additionally, it avoids exclusion of HLA-mismatched but otherwise compatible donors thereby increasing the number of potentially compatible units. These serologic techniques are typically available through community blood centers with commercially available kits [38].

If HLA-compatible platelets are not available (or while they are being tested and procured), ABO-compatible platelets can be used as a means of improving the platelet count increment. However, this is unlikely to be effective for individuals with HLA alloimmune refractoriness. (See "Platelet transfusion: Indications, ordering, and associated risks", section on 'ABO, Rh, and HLA matching'.)

There are no randomized trials comparing these three approaches; for the most part, crossmatch-compatible platelets, HLA-matched platelets, and antigen-exclusion platelets are thought to provide equivalent efficacy for identifying a compatible platelet unit that can result in a reasonable platelet increment [1,4]. A small retrospective study (354 platelet transfusions in 32 patients) reported corrected count intervals (CCI) >5000/microL with 25 percent with crossmatch compatible units and 30 percent with HLA-matched units, both of which were better than random donor units (CCI >5000/microL with only 12 percent of units) [39].

In a study of 50 pediatric patients with beta thalassemia major and platelet transfusion refractoriness following hematopoietic stem cell transplantation, transfusion outcome depended upon both the degree of HLA matching and ABO compatibility, is as follows [40]:

●Matched at HLA and ABO compatible: 76 percent successful platelet transfusions

●Matched at HLA and ABO incompatible or mismatched at HLA and ABO compatible: 67 percent successful

●Mismatched at HLA and ABO incompatible: 46 percent successful

Several other observational studies have demonstrated that individuals with HLA alloimmune platelet refractoriness generally have a good platelet count response with HLA-matched donor platelets [34,41,42].

It is mandatory that when HLA-matched (or other types of HLA-compatible) platelet transfusions are used, the product must be irradiated before it is transfused. This is because individuals receiving HLA-matched blood products are at risk for transfusion-associated graft-versus-host disease (TA-GVHD) caused by lymphocytes in the product that recognize the transfusion recipient's tissues (bone marrow, skin, gastrointestinal tract) as foreign and attack them, leading to pancytopenia that is typically fatal. All HLA-matched platelets must be irradiated prior to transfusion to prevent TA-GVHD. (See "Transfusion-associated graft-versus-host disease", section on 'Partial HLA matching'.)

Alternative computerized matching techniques are now emerging and may further improve the ability of blood banks to provide apheresis platelets for HLA-alloimmunized patients. As an example, the HLAMatchmaker is a software algorithm that predicts HLA compatibility including acceptable mismatched options. This method takes into account matching for highly immunogenic HLA epitopes, HLA matching on a deoxyribonucleic acid (DNA) rather than a serological level, and as well as HLA-C antigens [36,43-47]. A randomized trial comparing HLA-matched platelets identified using HLAMatchmaker versus standard HLA matching found generally similar outcomes between the two [42]. The degree of HLA match correlated with platelet count increment.

An immediate post-transfusion count should be obtained to determine the response to these units. It is especially important to monitor response when using antigen-exclusion platelets (because the antibody profile may change) and also relevant with other methods. (See 'Post-transfusion platelet count' above.)

Positive for anti-HPA — Antibodies against human platelet antigen (HPA) are an uncommon cause of platelet refractoriness. Some centers test for anti-HLA and anti-HPA simultaneously and others do the testing sequentially (reserving anti-HPA testing for individuals who do not demonstrate anti-HLA) [1]. Sequential testing is often more cost-effective and more commonly done (algorithm 1). If the institution does not perform anti-HPA testing, platelet crossmatching can be done instead.

If anti-HPA antibodies are identified, the relevant HPA antigens can be avoided in future transfusions (eg, using HPA-typed donors or crossmatching). Typically this is restricted to HPA-1a; for other antigens, crossmatching may be needed.

Limited role of immunosuppressive therapies — Immunosuppressive therapies have been used for individuals with alloimmune platelet refractoriness with limited success. We suggest not routinely using intravenous immune globulin (IVIG), splenectomy, or other immunosuppressive approaches, consistent with other experts, although immunosuppressive therapies (IVIG, off-label rituximab or bortezomib) may be used in extenuating circumstances.

•IVIG - IVIG, with or without therapeutic apheresis, has been proposed as a strategy for immune modulation, but the results have generally been unimpressive [48-50]. In a randomized trial from 1990 involving 12 patients with platelet refractoriness, those assigned to receive IVIG (400 mg/kg daily for five days) had improved platelet recovery at 1 to 6 hours post-transfusion, while those assigned to placebo did not have any improvement (corrected count interval [CCI] 8413/microL with IVIG versus 1050/microL with placebo; p <0.007) [50]. Neither group had improved platelet counts at 24 hours.

•Rituximab or bortezomib – Rituximab and bortezomib have been described as promising in case reports, although further investigation is needed [51-54]. We have not used these agents in platelet refractoriness.

An important role of T cells in alloimmune refractoriness may explain the lack of efficacy of B cell directed therapies [55].

Management of non-immune refractoriness — Non-immune refractoriness is managed by treating the underlying cause of increased platelet consumption (algorithm 1). (See 'Non-immune causes' above.)

Examples include:

●Treatment of infections

●Treatment of underlying causes of DIC

●Discontinuation of any medications that could cause drug-induced thrombocytopenia

●Addressing sources of bleeding

It has not been determined whether treatment of fever with an antipyretic agent can improve platelet count increment independent of treating infections.

TREATMENT OF BLEEDING — Major bleeding (or minor bleeding expected to progress to major bleeding) requires a coordinated approach, especially in individuals with platelet refractoriness. Urgent involvement of the transfusion medicine service and individuals with expertise in hemostasis is advised.

The following may be appropriate:

●Platelet transfusions – Even if the post-transfusion platelet count increment is low, continued platelet transfusions may be helpful, as transfused platelets may contribute to hemostasis. For individuals with alloimmune refractoriness, the most compatible platelets available can be used. If there is no platelet count increase and no reduction of bleeding, a clinician may determine that further transfusions would not provide any benefit and only confer risks associated with transfusion, and they may decide to discontinue transfusions. (See 'Management of alloimmune refractoriness' above.)

●Other hemostatic therapies

•Antifibrinolytic therapy – Tranexamic acid or epsilon aminocaproic acid are commonly used for bleeding, especially bleeding in mucosal sites (nose, oropharynx, urogenital tract), which have increased fibrinolytic activity [56]. However, evidence is lacking to support the use of antifibrinolytic therapy in patients with thrombocytopenia; preliminary results of a randomized trial suggest that tranexamic acid does not reduce the risk of clinically important bleeding [57]. Antifibrinolytic agents are contraindicated in patients with disseminated intravascular coagulation (DIC). (See "Evaluation and management of disseminated intravascular coagulation (DIC) in adults", section on 'Avoid antifibrinolytic agents and PCCs'.)

•rFVIIa – Recombinant activated factor VII (rFVIIa) is sometimes used off-label in situations with extreme bleeding when other therapies are ineffective. Case reports have described use of rFVIIa in individuals with alloimmune platelet refractoriness [58]. Thrombosis is a risk, even in individuals with thrombocytopenia, although the exact level of risk is challenging to determine. This therapy would only be used in a rare, life-threatening bleeding episode in which platelet transfusions are ineffective. Dosing is discussed separately. (See "Recombinant factor VIIa: Administration and adverse effects", section on 'Off-label uses' and "Recombinant factor VIIa: Administration and adverse effects", section on 'General approach to administration'.)

●Treat the cause of thrombocytopenia – Whenever possible, treatment for the underlying cause of thrombocytopenia should be used, even if it may take time to become effective. For immune thrombocytopenia (ITP), intravenous immune globulin (IVIG) is generally the fastest means of increasing the platelet count. (See "Initial treatment of immune thrombocytopenia (ITP) in adults", section on 'Therapies to raise the platelet count'.)

●TPO-RAs – Thrombopoietin receptor agonists (TPO-RAs) can raise the platelet count, but their effect takes several days, which will not treat bleeding urgently. (See "Clinical applications of thrombopoietic growth factors".)

Immunosuppressive therapies are rarely helpful and do not act rapidly, but they may be used in exceptional circumstances. (See 'Limited role of immunosuppressive therapies' above.)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Transfusion and patient blood management".)

SUMMARY AND RECOMMENDATIONS

●Definition – Platelet refractoriness is defined as a platelet count increment less than expected after at least two platelet transfusions. Quantitative definitions may be used but should not substitute for clinical judgment. The prevalence varies depending on the patient population. (See 'Definition and prevalence' above.)

●Associated conditions – Conditions associated with platelet refractoriness are broadly divided into non-immune causes (fever/sepsis, splenomegaly, disseminated intravascular coagulation [DIC]) and alloimmune causes (alloantibodies to platelet antigens); non-immune factors are more common (table 1). Most alloantibodies are directed against human leukocyte antigen (HLA) antigens, typically HLA-A and HLA-B; less commonly, alloantibodies against human platelet antigen (HPA) epitopes may be responsible. Non-immune and alloimmune causes often coexist. (See 'Factors associated with platelet refractoriness' above.)

●Evaluation – Platelet refractoriness may be suspected if platelet transfusion does not result in the expected increase in platelet count (algorithm 1). It is important to determine the immediate post-transfusion platelet count (obtained 10 minutes to 1 hour after the transfusion is finished) and the count at 24 hours, for at least two transfusions. A platelet count increase of <10,000/microL(or no increase) on at least two occasions indicates refractoriness. An increase in the immediate post-transfusion count and a fall to baseline by 24 hours is consistent with a non-immune mechanism, while absence of an increase immediately post-transfusion is consistent with an alloimmune mechanism (figure 2). The corrected count increment (CCI) is generally reserved for research.

●Prevention – Other than avoiding unnecessary transfusions, pre-storage leukoreduction is the primary means of reducing alloimmunization and platelet refractoriness. Conditions that contribute to platelet consumption should be treated. (See 'Prevention' above.)

●Management recommendations – Management is summarized in the flowchart (algorithm 1).

•Alloimmune refractoriness – For patients with a component of alloimmune refractoriness (eg, positive panel reactive antibody [PRA]), we suggest using platelets that are HLA-matched, HLA antigen-negative, or crossmatch-compatible (Grade 2C). The choice among these options may be based on cost, availability, inventory, or other factors. For alloimmune refractoriness without anti-HLA antibodies, we check for anti-HPA antibodies and provide antigen-negative platelets. Immunosuppressive therapies do not appear helpful in most cases of alloimmune refractoriness and are generally reserved for extreme cases. (See 'Management of alloimmune refractoriness' above.)

•Importance of platelet irradiation – When HLA-matched, crossmatched, or antigen avoidance platelet transfusions are used, it is mandatory that the product be irradiated before it is transfused, to prevent transfusion-associated graft-versus-host disease (TA-GVHD), a fatal complication associated with partially HLA-matched blood components. (See "Transfusion-associated graft-versus-host disease", section on 'Prevention'.)

•Non-immune refractoriness – For individuals with a component of non-immune refractoriness, the underlying cause should be addressed when possible. (See 'Management of non-immune refractoriness' above.)

•Patients with bleeding – Individuals with major bleeding (or concern for progression to major bleeding) and thrombocytopenia should receive platelet transfusions with the most compatible platelets available. Other therapies may include an antifibrinolytic agent, other hemostatic therapy, and therapies to improve the platelet count in the longer-term (thrombopoietin receptor agonist [TPO-RA], immunosuppressive therapy). (See 'Treatment of bleeding' above.)

ACKNOWLEDGMENTS — The UpToDate editorial staff acknowledges James D Burner, MD; Ann Secord, MD, CDR, MC, USN Ret; Ana Ortega Lopez, MD; Alyssa Ziman, MD; and Dennis Goldfinger, MD (deceased), all of whom contributed to earlier versions of this topic review.