INTRODUCTION — Hemostasis depends on an adequate number of functional platelets, together with an intact coagulation (clotting factor) system. This topic covers the logistics of platelet use and the indications for platelet transfusion in adults. The approach to the bleeding patient, refractoriness to platelet transfusion, and platelet transfusion in neonates are discussed separately.

●Evaluation of bleeding – (See "Approach to the adult with a suspected bleeding disorder".)

●Refractoriness to platelet transfusion – (See "Refractoriness to platelet transfusion therapy".)

●Neonates – (See "Neonatal thrombocytopenia: Clinical manifestations, evaluation, and management", section on 'Platelet transfusion'.)

COLLECTION — There are two ways that platelets can be collected: by isolating and pooling platelets from units of donated whole blood or by collecting platelets via apheresis directly from a donor.

●Pooled platelets – A single unit of platelets can be isolated from every unit of donated blood by centrifuging the blood within the closed collection system to separate the platelets from the red blood cells (RBCs). The number of platelets per unit varies according to the platelet count of the donor; a yield of 7 x 10¹⁰ platelets is typical [1]. Since this number is inadequate to raise the platelet count in an adult recipient, four to six units are pooled to allow transfusion of 3 to 4 x 10¹¹ platelets per transfusion [2]. These units are called whole blood-derived platelets, platelet concentrates, or random donor pooled platelets.

Advantages of pooled platelets include lower cost and ease of collection and processing (a separate donation procedure and pheresis equipment are not required). The major disadvantage is recipient exposure to multiple donors in a single transfusion and logistic issues related to bacterial testing.

●Apheresis (single donor) platelets – Platelets can also be collected from volunteer donors in a one- to two-hour apheresis procedure. Platelets are selectively removed along with some white blood cells (WBCs) and plasma, and most RBCs and plasma are returned to the donor. A typical apheresis platelet unit provides the equivalent of six or more units of platelets from whole blood (ie, 3 to 6 x 10¹¹ platelets) [2]. In larger donors with high platelet counts, up to three units can be collected in one session. These are called apheresis or single donor platelets.

Advantages of single donor platelets are exposure of the recipient to a single donor rather than multiple donors, and the ability to match donor and recipient characteristics such as HLA type, cytomegalovirus (CMV) status, and blood type for certain recipients. (See 'Ordering platelets' below.)

The effects of platelet apheresis on the donor are covered elsewhere. (See "Blood donor screening: Overview of recipient and donor protections", section on 'Complications of apheresis'.)

Both pooled and apheresis platelets contain some WBCs that were collected along with the platelets. These WBCs can cause febrile nonhemolytic transfusion reactions (FNHTR), alloimmunization, or transfusion-associated graft-versus-host disease (ta-GVHD) in some patients.

Platelet products also contain plasma, which can be implicated in adverse reactions including transfusion-related acute lung injury (TRALI) and anaphylaxis. (See 'Complications' below.)

Several strategies are used to prevent the complications associated with the presence of WBC and plasma in platelet products. (See 'Ordering platelets' below.)

Platelets concentrates also contain a small number of RBCs that express Rh antigens on their surface (platelets do not express Rh antigens). The small numbers of RBCs in apheresis platelets negates the issue of Rh alloimmunization in most patients. However, blood banks avoid giving platelets from Rh⁺ donors to Rh^- females of childbearing potential because of the potential risk of Rh alloimmunization and subsequent hemolytic disease of the newborn. (See "RhD alloimmunization in pregnancy: Overview".)

STORAGE

Room temperature storage — Platelets are routinely stored at room temperature, because cold induces clustering of von Willebrand factor receptors on the platelet surface and morphological changes of the platelets, leading to enhanced clearance by hepatic macrophages and reduced platelet survival in the recipient [3-6].

All cells are more metabolically active at room temperature, so platelets are stored in bags that allow oxygen and carbon dioxide gas exchange. Citrate is included to prevent clotting and maintain proper pH, and dextrose is added as an energy source [2].

The shelf life of platelets stored at room temperature is generally only five days because of the bacterial infection risk that increases in relationship to the storage duration. This short shelf life contributes to the greater sensitivity of platelet inventory to shortages. However, facilities can store platelets for up to seven days if they use a container cleared or approved by the US Food and Drug Administration (FDA) for seven-day storage and if the individual platelet units are subsequently tested for bacteria using a bacterial detection device cleared by the FDA and labeled for use as a "safety measure." The use of platelet storage containers must be performed consistent with the instructions for use of the device.

Cold storage (investigational) — Storage of platelets in the cold (refrigerator temperature; 2° to 6°C) is investigational.

Cold-stored platelets could potentially reduce the risk of bacterial growth. However, there are concerns about poorer recovery and survival of cold-stored platelets [3,7,8].

Small preliminary studies suggest that platelets stored in the cold have similar efficacy and safety as platelets stored at room temperature [9,10].

Strategies for reducing pathogens — A disadvantage of room temperature storage is the increased growth of bacteria compared with blood products stored in the refrigerator or freezer. (See 'Complications' below.)

Strategies for reducing exposure to pathogens in the platelet product include:

●Donor screening for bloodborne pathogens (see "Blood donor screening: Laboratory testing", section on 'Infectious disease screening and surveillance' and "Blood donor screening: Overview of recipient and donor protections", section on 'Protection of the recipient')

●Proper skin sterilization techniques during collection, and discarding the first 15 to 30 mL of blood collected, which is most likely to be contaminated by skin bacteria

●Performing tests to screen for bacterial contamination, such as automated culture-based assays, and rapid point-of-issue tests (see "Transfusion-transmitted bacterial infection", section on 'Detection of contamination')

●Using blood products that have been subjected to pathogen inactivation or reduction treatment (see "Pathogen inactivation of blood products", section on 'Platelets')

The US Food and Drug Administration (FDA) issued a guidance document on September 30, 2019, to further reduce the risk of bacterial contamination of platelet products [11]. By October 1, 2021, facilities must select and implement additional safety measures beyond a primary bacterial culture:

●Platelets collected by apheresis may be pathogen reduced, or they must undergo either secondary culture or secondary rapid testing after the primary culture. Alternatively, large volume delayed sampling of platelets collected by apheresis can be cultured no sooner than 36 hours after collection.

●Pre-storage pools of whole blood-derived platelets have similar testing requirements for secondary cultures, rapid testing after the primary culture, or large volume delayed sampling cultured no sooner than 36 hours.

●Single units of whole blood-derived platelets, which are constrained by volume, require either rapid testing or primary cultures.

The guidance also outlines conditions for extending platelet storage up to seven days [11].

Cryopreservation — Cryopreserved platelets have the potential to dramatically increase shelf-life (to years rather than days) and in turn to greatly improve inventory and availability as well as to reduce the risk of transfusion-transmitted bacterial infection. Cryopreserved platelets are not clinically available in the United States, but they have been used in military settings and are in use or development in some countries in Europe, Asia, and Australia [12].

The largest trial of cryopreserved platelets (the cryopreserved versus liquid platelet [CLIP-I] trial) randomly assigned 121 adults undergoing cardiac surgery to receive up to three units of cryopreserved or room temperature-stored (liquid) platelets if they needed a platelet transfusion [13]. The cryopreserved platelets were prepared from apheresis units from group O donors and frozen in DMSO as a preservative; after thawing they were reconstituted with AB or group-specific plasma. The control platelets (referred to as liquid storage) could be prepared from red blood cell units or by apheresis (see 'Collection' above). All measures of blood loss, need for red blood cell transfusion, and frequency of adverse events were similar between groups, and there were no thrombotic events associated with cryopreserved platelets. Compared with the control group, the group that assigned to cryopreserved platelets had a lower platelet count increment (median platelet count on the first postoperative day, 150,000 in controls versus 112,000/microL for cryopreserved platelets); the cryopreserved platelet group also received more platelet transfusions. A reduced platelet count increment with cryopreserved platelets versus controls was also seen in a trial in a small number of hematology-oncology patients who were transfused with platelets in the setting of bleeding and thrombocytopenia [14].

This trial suggests that cryopreserved platelets are likely to be safe and effective in a surgical setting. Cryopreserved platelets could potentially be a useful way to augment blood inventories and improve product availability in settings such as combat, in remote locations, or to alleviate shortages in urban areas with high usage. However, additional trials are needed in other patient populations with other clinical indications. Considerable regulatory hurdles must also be addressed [15].

Another potential concern with cryopreserved platelets is the delay in obtaining the unit due to thawing and reconstitution with plasma [15]. This delay is approximately 10 minutes if thawed plasma is available for reconstitution and 30 to 40 minutes if plasma also has to be thawed [13].

INDICATIONS FOR PLATELET TRANSFUSION — Platelets can be transfused therapeutically (ie, to treat active bleeding or in preparation for an invasive procedure that would cause bleeding), or prophylactically (ie, to prevent spontaneous bleeding).

Actively bleeding patient — Actively bleeding patients with thrombocytopenia should be transfused with platelets immediately to keep platelet counts above 50,000/microL in most bleeding situations including disseminated intravascular coagulation (DIC), and above 100,000/microL if there is central nervous system bleeding [16]. (See "Evaluation and management of disseminated intravascular coagulation (DIC) in adults", section on 'Treatment' and "Spontaneous intracerebral hemorrhage: Acute treatment and prognosis", section on 'Triage'.)

Other factors contributing to bleeding should also be addressed. These include:

●Surgical or anatomic defect

●Fever

●Infection or inflammation

●Coagulopathy

●Acquired or inherited platelet function defect

The dose and frequency of platelet transfusions will depend on the platelet count and the severity of bleeding. (See 'Dose' below.)

Preparation for an invasive procedure — Platelets are transfused in preparation for an invasive procedure if the thrombocytopenia is severe, if there is insufficient time to use other therapies to raise the platelet count when indicated (eg, if there is insufficient time to administer intravenous immune globulin or glucocorticoids in an individual with immune thrombocytopenia [ITP]), and if the risks of bleeding are deemed high.

Most of the data used to determine bleeding risk come from retrospective studies of patients who are afebrile and have thrombocytopenia but not coagulopathy [17,18]. Typical platelet count thresholds that are used for some common procedures are as follows:

●Neurosurgery or ocular surgery – 100,000/microL

●Most other major surgery – 50,000/microL

●Endoscopic procedures – 50,000/microL for therapeutic procedures; 20,000/microL for low risk diagnostic procedures (see "Gastrointestinal endoscopy in patients with disorders of hemostasis")

●Bronchoscopy with bronchoalveolar lavage (BAL) – 20,000 to 30,000/microL [19]

●Central line placement – 20,000/microL [20]

●Lumbar puncture – 10,000 to 20,000/microL in patients with hematologic malignancies and 40,000 to 50,000 in patients without hematologic malignancies; lower thresholds may be used in patients with immune thrombocytopenia (ITP) [21-23]

●Epidural anesthesia – 80,000/microL [23]

●Bone marrow aspiration/biopsy – 20,000/microL

Prevention of spontaneous bleeding — We use prophylactic platelet transfusion to prevent spontaneous bleeding in most afebrile patients with platelet counts below 10,000/microL due to bone marrow suppression.

We use higher thresholds (ie, 20,000 to 30,000/microL) in patients who are febrile or septic. Patients with acute promyelocytic leukemia (APL) have a coexisting coagulopathy, and for them we use a platelet transfusion threshold of 30,000 to 50,000/microL. (See 'Leukemia, chemotherapy, and HCT' below.)

The threshold for prophylactic transfusion can also vary depending on the patient and on the clinical scenario. (See 'Specific clinical scenarios' below.)

There are no ideal tests for predicting who will bleed spontaneously [24]. Studies of patients with thrombocytopenia suggest that patients can bleed even with platelet counts greater than 50,000/microL [25]. However, bleeding is much more likely at platelet counts less than 5000/microL. Among individuals with platelet counts between 5000/microL and 50,000/microL, clinical findings can be helpful in decision-making regarding platelet transfusion.

●The platelet count at which a patient bled previously can be a good predictor of future bleeding.

●Petechial bleeding and ecchymoses are generally not thought to be predictive of serious bleeding, whereas mucosal bleeding and epistaxis (so-called "wet" bleeding) are thought to be predictive.

●Coexisting inflammation, infection, and fever also increase bleeding risk.

●The underlying condition responsible for a patient's thrombocytopenia also may help in estimating the bleeding risk. As an example, some patients with ITP often tolerate very low platelet counts without bleeding, while patients with some acute leukemias that are associated with coagulopathy (eg, acute promyelocytic leukemia) can have bleeding at higher platelet counts (eg, 30,000 to 50,000/microL). (See 'Specific clinical scenarios' below.)

●Compared with adults, children with bone marrow suppression may be more likely to experience bleeding at the same degree of thrombocytopenia. In a secondary subgroup analysis of the PLADO trial, in which patients were randomly assigned to different platelet doses, children had more days of bleeding, more severe bleeding, and required more platelet transfusions than adults with similar platelet counts [26]. However, these findings do not suggest a different threshold for platelet transfusion in children, as the increased risk of bleeding was distributed across a wide range of platelet counts.

●Tests for platelet-dependent hemostasis (ie, bleeding time, thromboelastography (TEG), and other point of care tests) are generally not used to predict bleeding in thrombocytopenic patients. (See "Platelet function testing", section on 'The in vivo bleeding time' and "Platelet function testing", section on 'Instruments that simulate platelet function in vitro'.)

Therapeutic versus prophylactic transfusion — By convention, most authors use the term "therapeutic transfusion" to refer both to transfusion of platelets to treat active bleeding and transfusion of platelets in preparation for an invasive procedure that could cause bleeding. The term "prophylactic transfusion" is used to refer to platelet transfusion given to prevent spontaneous bleeding.

The findings from clinical trials support the use of prophylactic transfusion for patients with hematologic malignancies and hematopoietic cell transplant (HCT), as discussed below. Although reserving platelet transfusion for active bleeding may be safe for some adults undergoing autologous HCT, such a strategy requires intensive monitoring and the ability to perform immediate imaging for suspected central nervous system (CNS) or ocular bleeding. We do not recommend reserving platelet transfusion for active bleeding in patients with HCT outside of a clinical trial or a specific protocol in a highly specialized center with the ability to support this level of vigilance. (See 'Leukemia, chemotherapy, and HCT' below.)

Patients with platelet consumption disorders (eg, immune thrombocytopenia [ITP], disseminated intravascular coagulation) and platelet function disorders are typically transfused only for bleeding or, in some cases, invasive procedures. Platelets should not be withheld in bleeding patients with these conditions due to fear of "fueling the fire" of thrombus formation. (See 'Immune thrombocytopenia (ITP)' below and 'TTP or HIT' below and 'Platelet function defects' below.)

Given the need to balance the risk of spontaneous bleeding with the potential complications of unnecessary platelet transfusion, the decision of whether to transfuse platelets based upon a clinical event (ie, for active bleeding or invasive procedures) or at a particular threshold (ie, to prevent spontaneous bleeding) is challenging. Standard practice has evolved to transfusion of platelets at a threshold platelet count of 10,000 to 20,000/microL for most patients with severe hypoproliferative thrombocytopenia due to hematologic malignancies, cytotoxic chemotherapy, and HCT [27]. However, the risks and benefits of reserving platelet transfusion for active bleeding episodes in these patients continue to be evaluated [17,28-31].

Two randomized trials have evaluated outcomes with platelet transfusion for bleeding versus routine prophylactic transfusions:

●In a 2012 trial from Germany, 400 patients with acute myeloid leukemia (AML; patients with APL were excluded) and patients undergoing autologous HCT for hematologic malignancies were randomly assigned to receive platelet transfusions when morning platelet counts were ≤10,000/microL or only for active bleeding [32]. Patients transfused only for active bleeding received fewer platelet transfusions during the 14-day period after induction or consolidation chemotherapy (1.63 versus 2.44 per patient, a 33.5 percent reduction). However, among patients with AML who were transfused only for active bleeding, there were more episodes of major bleeding (six cerebral, four retinal, and one vaginal) and there were two fatal intracranial hemorrhages compared with four retinal hemorrhages among patients transfused for a platelet count ≤10,000/microL. Patients undergoing HCT also experienced more bleeding episodes when transfused only for active bleeding, but most of these were minor.

●In the 2013 TOPPS trial (Trial of Prophylactic Platelets), 600 patients with hematologic malignancies receiving chemotherapy, autologous, or allogeneic HCT were randomly assigned to receive platelet transfusion for a platelet count ≤10,000/microL or only for active bleeding [33,34]. Compared with those who received prophylactic transfusions, patients transfused only for active bleeding received fewer platelet transfusions during the 30-day period after randomization, but had a higher incidence of major bleeding (50 versus 43 percent) and a shorter time to first bleed (1.2 versus 1.7 days) [35]. There were no differences in the duration of hospitalization, and no deaths due to bleeding. In a predefined subgroup analysis, patients undergoing autologous (but not allogeneic) HCT had similar rates of major bleeding whether they were transfused for a platelet count ≤10,000/microL or only for active bleeding (45 and 47 percent). However, this may reflect a shorter duration of severe thrombocytopenia in this population and more aggressive, early treatment of suspected bleeding in the context of a clinical trial, rather than an inherently different rate of bleeding.

SPECIFIC CLINICAL SCENARIOS — There are several common clinical scenarios that raise the questions of whether to transfuse patients prophylactically to prevent bleeding, and, if prophylactic transfusion is used, of what platelet count is the best threshold for transfusion.

Leukemia, chemotherapy, and HCT — Patients with leukemia, or those being treated with cytotoxic chemotherapy or undergoing hematopoietic cell transplant (HCT) have a suppressed bone marrow that often cannot produce adequate platelets. We use prophylactic transfusion in these settings, assuming the patient is hospitalized, afebrile, and without active infection. We generally use a threshold platelet count of 10,000/microL (ie, transfuse for a platelet count <10,000/microL). An exception is acute promyelocytic leukemia (APL), for which the threshold is higher (ie, transfuse for a platelet count of <30,000 to 50,000/microL) due to a higher bleeding risk. If fever, sepsis, or coagulopathy is present, or if the patient is not hospitalized and/or cannot be closely monitored, higher thresholds may be needed.

This approach is in line with the 2017 American Society for Clinical Oncology (ASCO) guidelines (table 1) and a 2015 practice guideline from the AABB [36,37]. It is supported by randomized trials comparing prophylactic (ie, threshold-based) and therapeutic platelet transfusion, in which patients who did not receive prophylactic transfusion had more severe bleeding [32,35,38]. While the 2017 ASCO guideline suggests that some individuals undergoing autologous HCT may omit prophylactic platelet transfusions if they are being treated in a highly specialized center, we do not believe there is sufficient evidence to support this approach in most practice settings and clinical scenarios, and we continue to use prophylactic transfusions in our practice unless the patient is enrolled in a clinical trial or is following a specific institutional protocol, as discussed above. (See 'Therapeutic versus prophylactic transfusion' above.)

●Acute myeloid leukemia (AML) – Patients with AML can have suppressed bone marrow from AML, chemotherapy, or HCT. We use standard dose prophylactic transfusion of these patients at a threshold platelet count of 10,000/microL, and transfusion for any bleeding greater than petechial bleeding. (See 'Dose' below.)

●Acute promyelocytic leukemia (APL) – Patients with APL differ from other patients with AML because they often have an associated coagulopathy that puts them at high risk for disseminated intravascular coagulation and bleeding. We prophylactically transfuse these patients at a platelet count of 30,000 to 50,000/microL, and treat any sign of bleeding, especially central nervous system bleeding, with immediate platelet transfusion. (See "Clinical manifestations, pathologic features, and diagnosis of acute promyelocytic leukemia in adults", section on 'Coagulopathy and APL' and "Initial treatment of acute promyelocytic leukemia in adults", section on 'Control of coagulopathy'.)

●Acute lymphoblastic leukemia (ALL) – Patients with ALL have thrombocytopenia from bone marrow suppression. In addition, these patients are often treated with L-asparaginase, which causes severe hypofibrinogenemia. However, the risk of life-threatening bleeding is low. As an example, in over 2500 children with ALL, only two intracranial hemorrhages occurred, and they were associated with hyperleukocytosis in one case and intracerebral fungal infection in the other [21]. We transfuse adults with ALL at a threshold platelet count of 10,000/microL. The use of platelet transfusion in children with ALL is discussed separately. (See "Overview of the treatment of acute lymphoblastic leukemia/lymphoma in children and adolescents", section on 'Bleeding'.)

●Chemotherapy for solid tumors – Cancer chemotherapy often makes patients thrombocytopenic from bone marrow suppression. Randomized trials of platelet transfusion threshold in this population have not been performed. Observational studies support a prophylactic platelet transfusion threshold of 10,000/microL [38]. A threshold of 20,000/microL may be appropriate for patients with necrotic tumors.

●Hematopoietic cell transplant (HCT) – Chemotherapy and radiation therapy administered as part of the conditioning regimen for HCT can be highly bone marrow suppressive, depending on the doses used. We use standard dose prophylactic transfusion of these patients at a threshold platelet count of 10,000/microL, and therapeutic transfusion for any bleeding greater than petechial bleeding. As noted above, patients in the TOPPS trial who were undergoing autologous HCT had similar rates of major bleeding whether they were transfused for a platelet count ≤10,000/microL or only for active bleeding; however, there are several caveats that limit the use of this approach in the vast majority of patients undergoing HCT. (See 'Therapeutic versus prophylactic transfusion' above and "Hematopoietic support after hematopoietic cell transplantation", section on 'Platelet transfusion'.)

●Aplastic anemia – Patients with aplastic anemia do not have a malignancy, but they may have severe thrombocytopenia, and they may be candidates for HCT. Issues related to platelet transfusion in these patients are discussed separately. (See "Treatment of aplastic anemia in adults" and "Treatment of acquired aplastic anemia in children and adolescents" and "Inherited aplastic anemia in children and adolescents" and "Dyskeratosis congenita and other telomere biology disorders" and "Management and prognosis of Fanconi anemia".)

Immune thrombocytopenia (ITP) — Individuals with immune thrombocytopenia produce antiplatelet antibodies that destroy circulating platelets and megakaryocytes in the bone marrow. Circulating platelets in patients with ITP tend to be highly functional, and platelet counts tend to be well above 30,000/microL. Bleeding is rare even in patients with severe thrombocytopenia (ie, platelet count <30,000/microL). (See "Immune thrombocytopenia (ITP) in adults: Clinical manifestations and diagnosis", section on 'Pathogenesis'.)

Our general approach to platelet transfusion in patients with ITP is to transfuse for bleeding rather than at a specific platelet count. (See "Initial treatment of immune thrombocytopenia (ITP) in adults", section on 'Overview of decision-making'.)

TTP or HIT — Thrombotic thrombocytopenic purpura (TTP) and heparin-induced thrombocytopenia (HIT) are disorders in which platelet consumption causes thrombocytopenia and an increased risk of bleeding; but the underlying platelet activation in these conditions also increases the risk of thrombosis.

Platelet transfusions can be helpful or even life-saving in patients with these conditions who are bleeding and/or have anticipated bleeding due to a required invasive procedure (eg, placement of a central venous catheter), and platelet transfusion should not be withheld from a bleeding patient due to concerns that platelet transfusion will exacerbate thrombotic risk. However, platelet transfusions may cause a slightly increased risk of thrombosis in patients with these conditions; thus, we do not use prophylactic platelet transfusions routinely in patients with TTP or HIT in the absence of bleeding or a required invasive procedure.

Support for this approach comes from a large retrospective review of hospitalized patients with TTP and HIT, in which platelet transfusion was associated with a very slight increased risk of arterial thrombosis but not venous thromboembolism [39]. In contrast, the review found that patients with immune thrombocytopenia (ITP) had no increased risk of arterial or venous thrombosis with platelet transfusion. Of note, this was a retrospective study in which sicker patients were more likely to have received platelets, and the temporal relationships between platelet transfusions and thromboses were not assessed.

●TTP – Of 10,624 patients with TTP in the large review mentioned above, approximately 10 percent received a platelet transfusion [39]. Arterial thrombosis occurred in 1.8 percent of patients who received platelets, versus 0.4 percent of patients who did not (absolute increase, 1.4 percent; adjusted odds ratio [OR], 5.8; 95% CI, 1.3-26.6). The rate of venous thrombosis was not different in those who received platelets and those who did not (adjusted OR 1.1; 95% CI 0.5-2.2).

In contrast, a systematic review of patients with TTP who received platelet transfusions, which included retrospective data for 358 patients and prospective data for 54 patients, did not find clear evidence that platelet transfusions were associated with adverse outcomes [40].

●HIT – Of 6332 patients with HIT in the large review mentioned above, approximately 7 percent received a platelet transfusion [39]. Arterial thrombosis occurred in 6.9 percent of patients who received platelets, versus 3.1 percent of patients who did not (absolute increase, 3.8 percent; adjusted OR, 3.4; 95% CI, 1.2-9.5). The rate of venous thrombosis was not different in those who received platelets and those who did not (adjusted OR 0.8; 95% CI 0.4-1.7).

In a series of four patients with HIT who received platelet transfusions, two of three with active bleeding had cessation of bleeding following platelet transfusion, and no thromboses occurred; a literature review was not able to identify any complications clearly attributable to platelet transfusion [41].

Management of TTP and HIT is discussed in detail separately. (See "Immune TTP: Initial treatment" and "Management of heparin-induced thrombocytopenia".)

Liver disease and DIC — Patients with liver disease and disseminated intravascular coagulation (DIC) have a complex mixture of procoagulant and anticoagulant defects along with thrombocytopenia, and therefore they are at risk for thrombosis and bleeding. There is no evidence to support the administration of platelets in these patients if they are not bleeding. However, platelet transfusion is justified in patients who have serious bleeding, are at high risk for bleeding (eg, after surgery), or require invasive procedures. (See "Evaluation and management of disseminated intravascular coagulation (DIC) in adults", section on 'Prevention/treatment of bleeding' and "Hemostatic abnormalities in patients with liver disease", section on 'Bleeding'.)

Platelet function defects — Platelet function defects can be inherited or acquired, and may be associated with thrombocytopenia or a normal platelet count. Platelet transfusion in these settings is typically reserved for bleeding.

●Inherited diseases – Platelet function is impaired in Wiskott-Aldrich syndrome, Glanzmann thrombasthenia, and Bernard-Soulier syndrome. Bleeding in patients with these conditions is treated with platelet transfusion, along with other hemostatic agents discussed below. (See "Congenital and acquired disorders of platelet function", section on 'Inherited disorders of platelet function' and 'Alternatives to platelet transfusion' below.)

●Acquired conditions – Uremia, diabetes mellitus, myeloproliferative disorders, and other medical conditions can impair platelet function. Bleeding risk can be reduced by treating the underlying condition. Platelet transfusion is typically reserved for major bleeding in these conditions. (See "Congenital and acquired disorders of platelet function", section on 'Acquired platelet functional disorders'.)

Patients who are febrile or septic can have impaired platelet function. We transfuse these patients for bleeding. We also use a higher threshold for when fever or sepsis coexist with thrombocytopenia (eg, in patients with leukemia). (See 'Leukemia, chemotherapy, and HCT' above.)

Antiplatelet agents — Aspirin, nonsteroidal anti-inflammatory drugs (NSAIDs), dipyridamole, ADP receptor (P2Y₁₂) inhibitors (eg, clopidogrel, ticlopidine), and GPIIb/IIIa antagonists (eg, abciximab, eptifibatide) are used to prevent thrombosis by interfering with normal platelet function. The antiplatelet effects of NSAIDs and aspirin are relatively weak compared with the other antiplatelet agents, but the inhibitory effects of aspirin are irreversible during the lifespan of the platelets. (See "Platelet biology", section on 'Drugs with antiplatelet actions'.)

Typically, the approach to treating mild bleeding in a patient taking an antiplatelet agent is to discontinue the drug, assuming a favorable risk-benefit ratio. For more severe bleeding or urgent surgical procedures, high quality evidence regarding the benefit of platelet transfusion is lacking, and some evidence suggests that platelet transfusion may be deleterious. These cases can be complex, however, and we favor an individualized approach based on the complete clinical picture.

Evidence suggesting platelet transfusion is not effective in some sites of severe bleeding comes from the following:

●The 2016 PATCH trial (Platelet Transfusion in Cerebral Hemorrhage) randomly assigned 190 patients with intracerebral hemorrhage (ICH) in the setting of aspirin or another antiplatelet agent to receive platelet transfusion or standard care without platelet transfusions [42]. Compared with controls, patients who received platelet transfusions had a higher incidence of a composite outcome of death or shift toward a worse score on the modified Rankin Scale for functional independence. When analyzed separately, the increase in mortality did not reach statistical significance. Serious adverse events were greater with platelet transfusion (42 versus 29 percent); enlargement of the ICH was similar in both groups at approximately 15 percent. The authors did not identify a clear mechanism for the inferior outcomes with platelet transfusion but offered several hypotheses, including the possibility of concomitant ischemia, possible proinflammatory effects of platelets, or characteristics of the hemorrhage such as location or etiology. A subsequent reanalysis of the trial by the original authors suggested that the arms of the trial were not balanced at baseline (patients in the platelet transfusion arm had a larger ICH volume and more peri-hemorrhage edema), which might account for some portion of the difference in outcomes [43].

The management of ICH is discussed in detail separately. (See "Spontaneous intracerebral hemorrhage: Acute treatment and prognosis".)

●A 2017 study retrospectively reviewed outcomes in 204 individuals with gastrointestinal bleeding associated with an antiplatelet agent who received platelet transfusion as part of their management with 204 matched controls who did not receive platelet transfusions [44]. None of the people in the study had thrombocytopenia. In a multivariate analysis, platelet transfusion was associated with a higher risk of death (adjusted odds ratio [OR] 5.6; 95% CI 1.5-27). It is very possible that the individuals who received a platelet transfusion had increased baseline risk factors for mortality compared with controls.

The role of platelet transfusion in the setting of urgent surgical procedures (eg, coronary artery bypass grafting, neurosurgical interventions, and others) is also not well defined. Some clinicians give prophylactic platelet transfusions to patients taking antiplatelet drugs who require major surgery, while other clinicians use platelet transfusion only to treat excessive surgical bleeding [45,46]. There is no compelling evidence to support platelet transfusions in such scenarios, especially when platelet counts are within normal range; at the same time, there are limitations of the studies cited above (associating platelet transfusions with no clear benefits and possibly harm), making it challenging to interpret and extrapolate the data. The AABB has noted the low quality of the evidence and does not recommend for or against platelet transfusion for patients receiving antiplatelet therapy who have traumatic or spontaneous intracranial hemorrhage [36]. This decision continues to be individualized according the specific patient factors and the judgment of the treating clinician.

Other medications may impair platelet function. As an example, the Bruton tyrosine kinase (BTK) inhibitor ibrutinib inhibits platelet aggregation by interfering with activation signals. The role of platelet transfusion in patients with ibrutinib-associated bleeding despite a sufficient platelet count is unknown, and decisions are individualized according to the platelet count and the severity and site of bleeding. This association is discussed in more detail separately. (See "Treatment of relapsed or refractory chronic lymphocytic leukemia", section on 'Ibrutinib'.)

Massive blood loss — Patients with massive blood loss from surgery or trauma are transfused with red blood cells (RBC), resulting in partial replacement of the blood volume with a product lacking platelets and clotting factors. In this setting, we transfuse RBC, fresh frozen plasma (FFP), and random donor platelet units in a 1:1:1 ratio. As an example, a patient transfused with six units of RBC would also receive six units of pooled platelets or one apheresis unit (both of which provide approximately 5 x 10¹¹ platelets) and six units of FFP. (See "Initial management of moderate to severe hemorrhage in the adult trauma patient".)

Cardiopulmonary bypass — Patients who undergo prolonged cardiopulmonary bypass can have thrombocytopenia and impaired platelet function. The use of platelet transfusion in the cardiopulmonary bypass setting is discussed separately. (See "Congenital and acquired disorders of platelet function", section on 'Cardiopulmonary bypass' and "Early noncardiac complications of coronary artery bypass graft surgery", section on 'Bleeding'.)

Neonates — This subject is discussed separately. (See "Neonatal thrombocytopenia: Clinical manifestations, evaluation, and management", section on 'Platelet transfusion'.)

ORDERING PLATELETS — When ordering platelets, one should consider platelet dose; whether to use single donor versus random donor platelets; whether to include leukoreduction, irradiation, or platelets in platelet additive solution (PAS); and whether cytomegalovirus (CMV)-negative platelets are required. For individuals receiving multiple platelet transfusions, the transfusion service or blood bank may provide platelets that are matched ABO and Rh type and/or HLA type depending on the clinical setting (eg, ABO and RhD-matched platelets for individuals of childbearing potential, HLA-matched platelets for individuals with HLA alloimmunization, and decreased platelet count increment [refractoriness to platelet transfusion]).

Dose — A standard dose of platelets for prophylactic therapy in adults is approximately one random donor unit per 10 kg of body weight, which translates to four to six units of pooled platelets or one apheresis unit, both providing approximately 3 to 4 x 10¹¹ platelets [2,36]. A standard pediatric dose is 5 to 10 mL/kg. For prophylactic transfusion there is generally no reason to transfuse platelets more often than once a day. This platelet dosing is expected to raise the platelet count by approximately 30,000/microL within 10 minutes of the infusion. (See 'Platelet count increment' below.)

Clinical trials comparing standard platelet dosing with other doses have been limited to patients with hypoproliferative thrombocytopenia due to bone marrow suppression (eg, leukemia, hematopoietic cell transplant, or chemotherapy). Two large studies that evaluated the use of higher or lower platelet doses in these groups have conflicting results, as illustrated below.

●In the PLAtelet DOse (PLADO) trial, 1272 patients with thrombocytopenia due to chemotherapy or hematopoietic cell transplant (HCT) were randomly assigned to receive standard-dose (2.2 x 10¹¹ platelets per m²), half-dose (1.1 x 10¹¹ per m²), or double-dose (4.4 x 10¹¹ per m²) platelet transfusions [25]. The primary endpoint of prolonged mucosal or deep bleeding was similar among all groups (67 to 69 percent). The half-dose group received a higher median number of platelet transfusions during the 30-day study period (five in the half-dose group versus three in the other groups) but received fewer platelets overall (9.25 × 10¹¹ versus 11.25 × 10¹¹ and 19.63 × 10¹¹ in the standard- and double-dose groups, respectively).

●In the Strategies for Transfusion of Platelets (SToP) trial, patients with hypoproliferative thrombocytopenia were randomly assigned to receive platelet transfusion at standard dose (3 to 6 x 10¹¹ platelets, equivalent to approximately 5 to 10 units) or half dose (1.5 to 3 x 10¹¹ platelets, equivalent to approximately 3 to 5 units) when platelet counts fell below a trigger value (most participating institutions used 10,000/microL) [47]. The trial was halted prematurely (at 119 patients) because of life-threatening bleeding or bleeding requiring transfusion in the low dose arm (3 of 58 patients versus none of 61 in the standard dose arm).

Based on review of these and other trials as well as a study examining prophylactic transfusion of low-dose (half of the standard dose), standard dose, or high-dose (approximately two times the standard dose) platelet transfusions in hospitalized patients with therapy-induced hypoproliferative thrombocytopenia, the AABB recommends transfusions of up to a single standard-dose apheresis unit of platelets (or the equivalent) and states that "greater doses are not more effective, and lower doses are equally effective" [25,36,47-49]. However, given the logistical issues associated with more frequent platelet transfusions when low-dose units are used, and at times conflicting results from available studies, most centers continue to use standard-dose transfusion until further data become available.

In contrast to prophylactic transfusion, patients who are being transfused therapeutically (ie, for active bleeding or in preparation for an invasive procedure), or who have a rapidly dropping platelet count, may require higher dose or more frequent platelet transfusions.

Infusion rate — For an average-sized adult, six units of pooled platelets or one apheresis unit of platelets are transfused over approximately 20 to 30 minutes. Patients at risk for transfusion-associated circulatory overload (TACO) can be transfused at a slower rate as long as the transfusion is completed within four hours of issuance from the blood bank. (See "Transfusion-associated circulatory overload (TACO)", section on 'Prevention'.)

Pooled versus apheresis platelets — The platelet count increment and hemostatic effects of pooled and apheresis platelets are comparable [38].

Apheresis platelets have the advantages of limiting the recipient exposure to a single donor, which potentially reduces the possibility of infection and alloimmunization; some centers use apheresis platelets exclusively. Many believe it is logistically easier to perform bacterial testing on apheresis platelets compared with pooled platelets. (See 'Complications' below.)

Use of apheresis platelets also permits transfusion of platelets from specific donors selected based on HLA matching or platelet cross-matching, cytomegalovirus (CMV) status, and ABO group. Patients with confirmed immune mediated platelet refractoriness due to anti-HLA antibodies should receive HLA-matched platelets or platelets negative for the corresponding antigen(s) or cross-match compatible platelets; in other cases, either pooled and apheresis platelets can be used [50].

Leukoreduction — Leukoreduction removes most of the contaminating white blood cells (WBC) from the platelet transfusion [51]. Prestorage leukoreduction is standard practice in nearly all centers in the United States:

●Reduction of HLA alloimmunization

●Reduction of CMV transmission

●Reduction of transfusion-associated immunomodulation

●Reduction of lung injury during and after cardiopulmonary bypass

●Reduction of febrile nonhemolytic transfusion reactions (FNHTR)

Leukoreduction is done by passing platelets through a filter that blocks passage of most white blood cells. Apheresis platelets can be leukoreduced during collection, and pooled platelets can be leukoreduced shortly after collection or at bedside before transfusion.

Leukoreduction can reduce the risks of several potential complications of contaminating WBC, but it is not adequate to prevent transfusion-associated graft-versus-host disease (ta-GVHD), because some WBC can pass through the leukoreduction filter. Therefore, irradiation must be used to prevent ta-GVHD. (See "Transfusion-associated graft-versus-host disease", section on 'Prevention' and "Leukoreduction to prevent complications of blood transfusion", section on 'Transfusion-associated graft-versus-host disease'.)

Leukoreduction at the bedside (ie, post storage) is not optimal for reducing FNHTR because bedside leukoreduction does not remove cytokines released from WBC during storage. (See "Leukoreduction to prevent complications of blood transfusion", section on 'Febrile nonhemolytic transfusion reactions'.)

The only drawback of leukoreduction is the cost.

Irradiation — Platelet irradiation is used to prevent ta-GVHD, in which contaminating WBCs attack host tissues and cause serious, even fatal, outcomes in both immunosuppressed and some immunocompetent individuals. Irradiation damages the nuclei of donor lymphocytes in the transfusion, so that they cannot proliferate and mount an immune response against the recipient. Platelets are anucleate, so their functions are unaffected by irradiation, although there may be a slight effect on platelet survival due to membrane damage [52]. Platelets are irradiated by exposing the bag to 25 Gy from a Cesium source.

Irradiation is not a substitute for leukoreduction, because lymphocytes inactivated by irradiation still express human leukocyte antigens (HLA) on their surfaces and can elicit an anti-HLA antibody response from the host. Irradiation is also inadequate to kill pathogens such as bacteria and viruses. Irradiation is used for the following indications [53-55]:

●Immunosuppression or imminent immunosuppression from hematopoietic cell transplant, solid organ transplant, and cytotoxic chemotherapy.

●Congenital immunodeficiency (eg, DiGeorge syndrome, Wiskott-Aldrich syndrome, Leiner's disease, 5' nucleotidase deficiency).

●Fludarabine therapy.

●Hodgkin lymphoma and other hematologic malignancies.

●Neonatal exchange transfusion.

●Premature, low birth weight neonates. (See "Neonatal thrombocytopenia: Clinical manifestations, evaluation, and management", section on 'Platelet transfusion'.)

●Intrauterine transfusion.

●A subset of donor-recipient pairs who may be closely, but not completely, HLA matched (ie, relatives and genetically homogeneous populations) [56]. The rationale for this is discussed separately. (See "Transfusion-associated graft-versus-host disease", section on 'Partial HLA matching'.)

Irradiation may not be necessary if platelets have been subjected to pathogen reduction protocols that also prevent lymphocyte proliferation (eg, photochemical treatments) [56]. These protocols, which have the added advantage of not requiring a radioactive source, are discussed separately. (See "Pathogen inactivation of blood products", section on 'Methods that damage nucleic acids'.)

CMV — Some cytomegalovirus (CMV) seronegative transfusion recipients (eg, immunosuppressed patients) are at greater risk of adverse outcomes from receiving CMV-contaminated blood products than the general population. The AABB (formerly the American Association of Blood Banks) considers transfusion of platelets from CMV-negative donors to be equivalent to leukoreduction in reducing this risk. (See "Leukoreduction to prevent complications of blood transfusion", section on 'Viral diseases'.)

ABO, Rh, and HLA matching — Platelets express ABO antigens and HLA class I antigens on their surface. They do not express Rh or HLA class II antigens.

ABO and HLA compatible platelets appear to cause a greater platelet count increment in the recipient, and they can be used to improve responses in patients who have become refractory to platelet transfusion due to alloimmunization [25]. (See "Refractoriness to platelet transfusion therapy", section on 'Management of the alloimmunized patient'.)

ABO compatibility is most important in the "forward" direction (ie, donor platelets only express ABO antigens on their surface that are also present on the recipient’s red blood cells [RBCs] and platelets).

In many cases, ABO matched platelets may not be available, and inventory constraints may result in ABO mismatched platelets being provided when type-specific platelet products are not available. Clinically significant hemolytic transfusion reactions secondary to transfusion of ABO-incompatible platelet products (eg, group O platelets given to group A patient) are uncommon, but they do occur [57].

To limit such hemolytic reactions, some transfusion services monitor and limit the volume of ABO incompatible plasma given to a patient via platelet transfusions, or volume-reduce or wash the ABO incompatible platelet products to reduce the plasma content. Some also screen for platelet products with high anti-A or anti-B titers and give products with high titers only to group O patients. However, the critical threshold has not been determined for either the volume of incompatible plasma or the anti-A/B titers. (See "Red blood cell antigens and antibodies", section on 'Blood component transfusion' and "Hemolytic transfusion reactions".)

The possibility of alloimmunization to red blood cell (RBC) antigens causing hemolytic disease of the fetus and newborn (HDFN) in a pregnant woman raises another important issue related to Rh matching of platelets [58]. Although platelets do not express Rh antigens, platelet products contain small numbers of RBCs, which could be Rh incompatible with the recipient. Thus, when an RhD-negative woman of childbearing age receives a platelet transfusion, platelets from an RhD-negative donor are used, in order to prevent alloimmunization and HDFN. The Royal College of Obstetricians and Gynaecologists (RCOG) advises administration of anti-D immune globulin (also called Rho[D] immune globulin) in this setting [59]. Even if this is not done and platelets from an RhD-positive donor are used, the risk of alloimmunization remains low. This was illustrated in a retrospective analysis of 1014 RhD-negative patients who received 6043 platelet transfusions from Rh⁺ donors (89 percent from pooled platelets); in this series no patients who received only apheresis platelets developed anti-RhD antibodies, and 12 of 315 (3.8 percent) who received pooled platelets developed anti-RhD antibodies [60]. However, in a series of 59 RhD-negative patients transfused with platelets from an RhD-positive donor for a non-hematologic condition such as pneumonia or surgery (typical dose, one to three units, given without anti-D immune globulin), alloantibodies to RhD were detected in eight (13.5 percent) [61].

To further reduce the risk of alloimmunization if only RhD-positive platelets are available, anti-D immune globulin can be coadministered with platelet transfusions. Each dose of anti-D immune globulin is considered sufficient to prevent alloimmunization for up to 15 mL of RhD-positive RBCs, and most units of platelets do not contain more than 0.5 mL of RBCs. Thus, a single dose of anti-D immune globulin is likely to be sufficient even if several units of platelets are transfused. If necessary, this can be repeated once every eight weeks (a similar interval to that used to prevent HDFN). (See "RhD alloimmunization in pregnancy: Overview" and "RhD alloimmunization: Prevention in pregnant and postpartum patients".)

WBCs present in HLA matched platelet products can cause transfusion-associated graft-versus-host disease (ta-GVHD), so all HLA-matched platelets must be irradiated. (See "Transfusion-associated graft-versus-host disease", section on 'Partial HLA matching'.)

Platelet additive solutions — After collection, platelets can be resuspended in one of several platelet additive solutions (PAS), as a substitute for a portion of the associated plasma. PAS consist of salts, buffers, and sometimes glucose [62]. Use of PAS platelets decreases but does not eliminate donor plasma exposure, and PAS may provide a less labor-intensive option for reducing allergic transfusion reactions than platelet washing or volume-reduction.

Clinical experience with PAS platelets is limited, and decisions regarding when to use them may depend on the incremental costs and expected benefits. Local availability of PAS platelets may vary, and institution-specific guidelines regarding their use should be followed. One strategy is to use PAS apheresis platelets, which contain less plasma, for patients without a coagulopathy and/or patients who have had minor allergic transfusion reactions [63].

A reduction in allergic transfusion reactions with PAS platelets has been demonstrated in two randomized trials using PAS available in Europe [64,65].

●In one trial of patients receiving multiple platelet transfusions (324 transfusions in 21 patients), platelets resuspended in 65 percent PAS with 35 percent plasma were associated with fewer allergic reactions than platelets in 100 percent plasma (5 versus 12 percent) [65]. The platelet count increases following transfusion were slightly lower with PAS platelets than non-PAS platelets (corrected count interval [CCI] at 20 hours 10,000 versus 12,000/microL). (See "Refractoriness to platelet transfusion therapy", section on 'Measuring response to platelet transfusion'.)

●In a trial that randomly assigned 168 patients to PAS versus non-PAS platelets, mild transfusion reactions were seen less commonly with PAS compared with non-PAS platelets (2 versus 6 percent) [64]. CCI was slightly lower with PAS compared with non-PAS platelets in this trial as well (CCI at 24 hours 7000 versus 8000/microL).

Neither trial showed a difference in bleeding complications with PAS versus non-PAS platelets.

A large retrospective study (5078 patients) compared outcomes with apheresis platelets resuspended a PAS solution approved by the US Food and Drug Administration (InterSol, 65 percent, with 35 percent plasma) versus 100 percent plasma [63]. The incidence of allergic transfusion reactions was reduced with PAS apheresis platelets (PAS AP) compared with non-PAS AP (1.01 versus 1.85 percent; relative risk [RR] 0.54; 95% CI 0.30-0.99). The incidence of febrile non-hemolytic transfusion reactions did not differ. Among individuals for whom paired PAS AP and non-PAS AP transfusions could be compared, there was no difference in the CCI at 12 to 24 hours, although PAS AP were associated with a slight reduction in CCI at four hours compared with non-PAS AP.

The use of PAS platelets may not be possible when HLA matched or CMV-negative products are needed. PAS platelets can be irradiated.

Other special modifications — Patients with known IgA deficiency who have a history of anaphylactic transfusion reactions or demonstrate anti-IgA antibodies can be transfused with platelets that have been washed to remove IgA-containing plasma or obtained from IgA deficient donors. (See "Selective IgA deficiency: Management and prognosis", section on 'Safe administration of blood products'.)

In addition, volume-reduced platelets can be used when exposure to ABO incompatible plasma needs to be limited, or for transfusion of volume-sensitive patients.

As noted above and separately, platelets can also be treated with a pathogen-inactivation method. (See 'Storage' above and "Pathogen inactivation of blood products", section on 'Platelets'.)

COMPLICATIONS — Platelet transfusion carries several risks including infection, transfusion reactions, alloimmunization, and posttransfusion purpura.

Complication rate with apheresis versus whole blood-derived platelets — The relative frequency of complications with apheresis versus whole blood-derived, pooled platelets has not been studied in large randomized trials.

●A 2008 systematic review and meta-analysis that evaluated several small randomized trials (mostly with fewer than 100 patients) found a greater incidence of reactions with whole blood-derived platelets; however, this was no longer significant after controlling for the use of leukoreduction [66].

●A 2016 study involving almost 800,000 platelet transfusions in France found that apheresis platelets were associated with a greater frequency of adverse reactions (approximately 6 per 1000 for apheresis platelets versus 2 per 1000 for whole blood-derived platelets) [67]. In this study, all platelets were leukoreduced (during collection for apheresis, and before storage for whole blood-derived). However, comparison may be difficult due to the different size of apheresis versus pooled platelet units and the challenges of calculating the incidence per unit when multiple units are administered.

Additional data are needed before a clear conclusion on relative risk of complications can be made.

Infection — Donor screening procedures and pathogen inactivation do not completely eliminate the risk of bacterial and other bloodborne infections, and infection by bacterially contaminated platelets represents a serious hazard of platelet transfusion that may be fatal [68-71]. This is because platelets are stored at room temperature, where bacteria can proliferate rapidly. The incidence of bacterial contamination is higher for platelets (approximately 1 in 2000) than it is for red blood cells (approximately 1 in 30,000) [72,73]. A 2020 meta-analysis that included 22 studies (over 5 million platelet units) found a bacterial contamination rate of 1:1961, with lower rates for apheresis units and platelet rich plasma collection versus buffy coat collection and a gradual decline in contamination rate year over year [74].

Evaluation and management of septic transfusion reactions are discussed separately. (See "Transfusion-transmitted bacterial infection".)

Measures that appear to reduce the rate of bacterial contamination include enhancements to the skin preparation technique, Gram staining or culturing the product, and using pathogen-inactivation technologies.

●A change to the skin preparation technique in 2012 in the United States was associated with a decrease from 4.2 instances of bacterial contaminants per year to 0.8 per year [75,76].  

●Culture of the product 24 to 36 hours after collection to identify and remove contaminated units was associated with a decrease from 492 to 82 septic transfusion reactions per million units [75]. Enhanced primary culture methods are under development.

As noted above, guidance from the US Food and Drug Administration (FDA) issued in late 2019 will require additional procedures to reduce infectious risks. (See 'Strategies for reducing pathogens' above.)

Alloimmunization — Platelets express Class I human leukocyte antigen (HLA) antigens, which can be recognized by the recipient's immune system as foreign. Production of anti-HLA antibodies can adversely affect the response to future platelet transfusions. The incidence of alloimmunization depends on the number of transfusions a patient has received. (See "Refractoriness to platelet transfusion therapy", section on 'Alloimmunization' and 'Platelet count increment' below.)

Platelet products also contain small volumes of red blood cells (RBCs), and alloimmunization to RBC antigens can occur as a result. This is especially of concern in RhD-negative women of childbearing potential, who are at risk for hemolytic disease of the fetus and newborn (HDFN) if they have an RhD-positive pregnancy. As noted above, this is one of the settings in which it may be appropriate to use matched platelets and/or to administer anti-D immune globulin (also called Rho[D] immune globulin). (See 'ABO, Rh, and HLA matching' above.)

Transfusion reactions — Transfusion of any blood product, including platelets, can lead to the following transfusion reactions:

●Transfusion-related acute lung injury (TRALI) – Transfusion-related acute lung injury (TRALI) is a form of acute lung injury that causes respiratory distress following transfusion. The true incidence of TRALI from platelet transfusion is unknown; it has decreased since institution of TRALI mitigation strategies in the late 2000s [77]. (See "Transfusion-related acute lung injury (TRALI)", section on 'Epidemiology'.)

●Transfusion-associated circulatory overload (TACO) – Platelet transfusion introduces approximately 200 mL of intravascular volume per transfusion. The incidence of TACO is in the range of one to three per 100,000 transfusions and is higher in patients predisposed to volume overload (eg, with comorbidities such as congestive heart failure, renal failure, respiratory failure, and positive fluid balance). (See "Transfusion-associated circulatory overload (TACO)".)

●Allergic and anaphylactic reactions – Allergic reactions to platelet transfusion are relatively common. They are usually due to IgE directed against proteins in the donor plasma. Common symptoms include urticaria and pruritus in mild cases, and wheezing, shortness of breath and hypotension in more severe cases. (See "Immunologic transfusion reactions", section on 'Urticarial (allergic) reactions'.)

Patients with a history of allergic transfusion reactions who require additional platelet transfusions may benefit from platelets in additive solution (PAS), which contain less plasma than non-PAS platelets. Those who continue to have allergic reactions with PAS platelets may receive concentrated or washed platelets. (See 'Platelet additive solutions' above and 'Other special modifications' above.)

Anaphylactic reactions (ie, severe allergic reactions) are a very rare complication of platelet transfusion. These are associated with rapid onset of shock, angioedema, and respiratory distress. Many cases occur due to the production of anti IgA antibodies in recipients who are IgA deficient. (See "Immunologic transfusion reactions", section on 'Anaphylactic transfusion reactions'.)

●Febrile non-hemolytic transfusion reactions (FNHTR) – These reactions are mediated by various inflammatory mediators and leukocytes and may manifest as fevers, chills, and rigors. (See "Immunologic transfusion reactions", section on 'Febrile nonhemolytic reactions' and "Leukoreduction to prevent complications of blood transfusion", section on 'Febrile nonhemolytic transfusion reactions'.)

●Ta-GVHD – Transfusion-associated graft-versus-host disease (ta-GVHD) can occur with any type of transfusion that contains lymphocytes, given the correct immunologic setting. Its incidence continues to drop due to irradiation of blood products for at-risk patients, such as patients with hematopoietic cell transplantation, immunodeficiency, or other types of immunosuppression.

A second and potentially less obvious situation that can lead to ta-GVHD in individuals who are completely immunocompetent is partial HLA matching (ie, a donor-recipient pair who are closely, but not completely, HLA matched, as can occur in relatives and genetically homogeneous populations) [56]. In this case, the HLA antigens on the donor lymphocytes are seen by the recipient lymphocytes as self, so recipient lymphocytes do not attack the donor lymphocytes; however, recipient cells also express unique HLA antigens that the donor lymphocytes recognize as foreign. This can result in donor lymphocytes destroying the recipient's tissues (eg, bone marrow, skin and liver), which can be fatal. (See 'Irradiation' above and "Transfusion-associated graft-versus-host disease".)

Posttransfusion purpura — Posttransfusion purpura (PTP) is a rare transfusion reaction to any platelet-containing product, in which thrombocytopenia develops 5 to 10 days following transfusion. This can occur in the <2 percent of individuals who lack the platelet antigen PIA1, now known as human platelet antigen 1a (HPA-1a), and have become previously sensitized to the antigen (eg, during pregnancy or prior transfusion). Transfused platelets are removed by an antibody-mediated mechanism; the patient's own HPA-1a-negative platelets are also destroyed by an incompletely understood process. Treatment is with intravenous immune globulin (IVIG), with or without a glucocorticoid, and HPA-1a-negative products should be used whenever possible if platelet transfusion is indicated. (See "Immunologic transfusion reactions", section on 'Post-transfusion purpura'.)

PLATELET COUNT INCREMENT — Following a platelet transfusion, the platelet count should rise, with a peak at 10 minutes to one hour and a gradual decline over 72 hours. A general rule of thumb is that transfusion of six units of pooled platelets or one apheresis unit should increase the platelet count by approximately 30,000/microL in an adult of average size.

Platelet count increment is typically measured within 24 hours in patients given prophylactic platelet transfusions. For patients undergoing invasive procedures, it is prudent to check that the desired platelet count was achieved before performing the procedure, which can be done within 10 minutes of the transfusion. For actively bleeding patients, cessation of bleeding is a more important clinical endpoint than the posttransfusion platelet count.

The length of time platelets have been stored has a modest effect on their survival in the recipient. As an example, compared with platelets stored for two or three days, platelets stored for five days produce a smaller increment in platelet count. This information is not usually factored into assessment of a patient's response to platelet transfusion.

Many patients who receive platelet transfusions reproducibly show a less-than-expected increase in platelet count. The definition of platelet refractoriness and its management are discussed separately. (See "Refractoriness to platelet transfusion therapy", section on 'Diagnostic approach' and "Refractoriness to platelet transfusion therapy", section on 'Management of the alloimmunized patient'.)

ALTERNATIVES TO PLATELET TRANSFUSION — There are no substitutes for platelet transfusion to rapidly increase the platelet count in a bleeding patient. Reversal of thrombocytopenia due to autoimmune platelet destruction, platelet consumption, or bone marrow suppression can take days to weeks, depending on the underlying cause.

Patients with ongoing bleeding not responsive to platelet transfusion and other interventions can also be given procoagulant bypass agents, such as prothrombin complex concentrates or recombinant factor VIIa [50]. (See "Platelet dysfunction in uremia", section on 'Desmopressin (DDAVP)' and "Medical management of the dialysis patient undergoing surgery", section on 'Bleeding diathesis'.)

In some settings, fibrinolytic inhibitors such as tranexamic acid have been effective. (See "Coagulopathy in trauma patients", section on 'Pharmaceutical hemostatic agents' and "Managing an episode of acute uterine bleeding", section on 'Tranexamic acid'.)

Stimulation of bone marrow megakaryocytes with thrombopoietin receptor agonists can take up to seven days (ie, the time it takes for new platelets to form). This might be appropriate for selected indications for preventing bleeding. (See "Clinical applications of thrombopoietic growth factors", section on 'Use of TPO receptor agonists'.)

Investigational approaches such as the use of human leukocyte antigen (HLA)-deleted platelets generated from induced pluripotent stem cells (iPSCs) (see "Genetics: Glossary of terms", section on 'Induced pluripotent stem cell (iPSC)') or the use of platelet substitutes (eg, synthetic or acellular biological materials that could replace the primary hemostatic function of platelets) have not reached clinical trials [78-81].

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: Immune thrombocytopenia (ITP) and other platelet disorders" and "Society guideline links: Transfusion and patient blood management".)

SUMMARY AND RECOMMENDATIONS

●Means of collection and content of a unit – A unit of platelets isolated from a unit of donated blood contains approximately 7 x 10¹⁰ platelets, and four to six of these units are typically pooled for transfusion. Single donor (apheresis) platelets contain approximately 3 to 6 x 10¹¹ platelets (ie, the equivalent of six or more units) per unit. (See 'Collection' above.)

●Storage – Platelets are stored at room temperature; consequently their shelf life is only approximately five days under most circumstances. Pathogen-reduced platelets are clinically available, and cryopreservation is an area of active study. (See 'Storage' above.)

●Indications

•Bleeding – Platelet transfusion can be lifesaving in bleeding patients with thrombocytopenia or reduced platelet function. Platelets should be transfused in any patient who is bleeding with a platelet count <50,000/microL (100,000/microL for central nervous system or ocular bleeding), or in any patient with an acquired or inherited platelet function defect regardless of platelet count. Platelet transfusion may also be indicated in thrombocytopenic patients undergoing invasive procedures, depending on the procedure and the platelet count. (See 'Actively bleeding patient' above and 'Preparation for an invasive procedure' above.)

Other conditions that impair hemostasis (eg, coagulopathy, fever infection, anatomic defects) should be corrected in thrombocytopenic patients when possible to reduce active bleeding and to lessen the risk of spontaneous bleeding. (See 'Actively bleeding patient' above.)

•Prophylaxis – We use prophylactic platelet transfusion to prevent spontaneous bleeding in most hospitalized afebrile patients with platelet counts below 10,000/microL due to bone marrow suppression. Patients with acute promyelocytic leukemia (APL) have a coexisting coagulopathy, and we use a platelet transfusion threshold of 30,000 to 50,000/microL in these patients. We use higher thresholds in patients who are febrile or septic. (See 'Prevention of spontaneous bleeding' above and 'Therapeutic versus prophylactic transfusion' above and 'Specific clinical scenarios' above.)

●Platelet consumption disorders – Patients with platelet consumption disorders, including immune thrombocytopenia (ITP), thrombotic thrombocytopenic purpura (TTP), heparin-induced thrombocytopenia (HIT), disseminated intravascular coagulation (DIC), liver disease, as well as those with platelet function disorders, are typically transfused only for bleeding or, in some cases, invasive procedures. Platelets should not be withheld in bleeding patients with these conditions due to fear of "fueling the fire" of thrombosis. (See 'Specific clinical scenarios' above.)

●Potential complications – Platelet transfusion has risks, including sepsis/infection, transfusion-related acute lung injury (TRALI), transfusion-associated circulatory overload (TACO), alloimmunization, allergic and anaphylactic transfusion reactions, febrile non-hemolytic transfusion reactions (FNHTR), transfusion-associated graft-versus-host disease (ta-GVHD), and posttransfusion purpura (PTP). The US Food and Drug Administration (FDA) issued guidance in late 2019 to reduce the risk of transfusion-transmitted infections from platelet products. (See 'Complications' above and 'Strategies for reducing pathogens' above.)

●Approach to refractoriness – Refractoriness to platelet transfusion is discussed separately. (See "Refractoriness to platelet transfusion therapy".)

●Dosing and special modifications – When ordering platelets, one should consider platelet dose; whether to use single donor versus random donor platelets; whether to include leukoreduction, irradiation, or platelets in platelet additive solution (PAS); whether cytomegalovirus (CMV)-negative platelets are required; and whether to match HLA, ABO, and Rh type. (See 'Ordering platelets' above.)

●Alternatives to platelet transfusion – There are limited alternatives to platelet transfusion for the acute treatment of thrombocytopenia-associated bleeding. Longer term alternatives include discontinuation of drugs that affect platelet function, treatment of underlying conditions, and other strategies to increase platelet production. (See 'Actively bleeding patient' above and 'Alternatives to platelet transfusion' above and 'Platelet function defects' above.)

ACKNOWLEDGMENT — We are saddened by the death of Dennis Goldfinger, MD, who passed away in September 2021. The UpToDate editorial staff acknowledges Dr. Goldfinger's past work for many years as an author for this topic.