Hematologic Supportive Care for Children with Cancer

Jennifer Andrews

Susan A. Galel

Wendy Wong

Bertil Glader

Transfusion therapy is a key component of supportive care of children with cancer and hematologic diseases and recipients of hematopoietic stem cell transplants. In this chapter, we provide a review of the indications for transfusion, along with a discussion of the pathophysiology, risks, and benefits of transfusion therapy. Where possible, we provide guidelines for best practices. In addition, we review the current status of cytokine therapy for children with malignancies.

ANEMIA

Definition and Prevalence

Anemia is defined as a deficiency of red blood cells (RBCs) or hemoglobin leading to a reduction in the oxygen-carrying capacity of blood. The incidence of anemia in children with cancer is not well defined in the United States, but a survey of European providers showed that over 80% of children being treated for cancer were anemic. The prevalence was the highest in children with leukemia (over 97%) and lymphoma (over 93%), but also was common in children with bone cancers, neuroblastoma, and brain tumors.¹

Pathophysiology and Etiology

Cancer-related anemia is multifactorial, but is primarily attributed to direct tumor infiltration of the bone marrow at diagnosis or relapse.² Additionally, suppression of erythropoiesis by chemotherapy and/or radiation contributes to anemia, with its severity influenced by a patient’s specific treatment regimen and intensity. Inflammatory cytokines also inhibit erythropoiesis in the bone marrow and depress erythropoietin response to anemia as in other chronic diseases.³ Less frequent causes of anemia include occult bleeding, such as intratumoral hemorrhage, and viral suppression of the bone marrow as can occur with parvovirus B19 infection in children with immunodeficiency or on chemotherapy.

Signs and Symptoms of Anemia

Cancer-related anemia is typically normocytic and normochromic with a decreased reticulocyte count.³ The signs and symptoms of anemia vary based on the rate of marrow replacement by the malignancy, the child’s compensatory change in blood volume, and the compensatory mechanisms of the cardiovascular system. Symptoms of anemia can include poor feeding, loss of appetite, headaches, dizziness, fatigue, vertigo, tinnitus, dyspnea, irritability, faintness, inactivity, and loss of concentration. Cardiac enlargement and evidence of congestive heart failure may be present with severe anemia. Tachycardia, prominent arterial pulses, bruits, tachypnea, dyspnea, and postural hypotension can occur in patients with modest-to-severe anemia. Flow murmurs reflect an increase in cardiac output, stroke volume, and heart rate associated with decreased peripheral resistance and decreased blood viscosity. Gallop rhythm may be present in a hemodynamically compromised state.

Red Cell Transfusion Guidelines

The standard therapy for chemotherapy-induced anemia has been RBC transfusions. Several guidelines have been published regarding the indications for RBC transfusions in children.⁴^,⁵ Most are based on expert opinion rather than evidence from clinical trials.⁶ Restrictive transfusion strategies using a hemoglobin “trigger” have been shown to be safe,⁵ although the need for an RBC transfusion should always be based on the patient’s clinical condition, ability to compensate for anemia, cardiopulmonary risk factors, and expected duration of the anemia.⁷

Healthy children generally tolerate relatively severe anemia (Hgb 6 to 7 g/dL or occasionally even lower).⁷ Even critically ill, hemodynamically stable children have been shown, in a randomized, prospective, multicenter trial, to have no worse outcomes when transfused at a Hgb less than 7 g/dL compared with a Hgb less than 9.5 g/dL.⁵ Although no randomized control trial evidence exists for a transfusion trigger for pediatric oncology patients, RBC transfusions should be considered for symptomatic anemia, prior to surgical procedures with anticipated blood loss or prior to radiation therapy. In asymptomatic patients, most oncologists transfuse when hemoglobin drops within a range of 6 to 7 g/dL. However, there are other important considerations aside from the hemoglobin concentration depending on whether the patient is expected to have a lower hemoglobin concentration or is on the path to recovery.

Red Cell Component Therapy

The patient and family must be informed of the risks, benefits, and alternatives to blood transfusion. According to the American Association of Blood Banks (AABB) Standards, informed consent must include at a minimum the following: (1) a description of the risks, benefits, and treatment alternatives (including nontreatment); (2) the opportunity to ask questions; and (3) the right to accept or refuse transfusion.⁸ Informed consent is required by local, state, and national health and safety codes (such as the Paul Gann Blood Safety Code in the state of California).⁹ The right to refuse transfusion is only applicable to adults or emancipated minors.

After consent is obtained, pretransfusion testing is performed. This should include ABO group, Rh type, and screening for any unexpected antibodies to red cell antigens known to cause hemolytic transfusion reactions or hemolytic disease of the fetus and newborn.¹⁰ Cytomegalovirus (CMV) serologies are recommended for high-risk populations, such as potential bone marrow recipients. RBC components require that a crossmatch be performed, for which the patient’s sample must be no more than 3 days old.¹¹

Packed RBC products are prepared from whole blood after centrifugation. RBCs can be stored in various anticoagulantpreservative solutions, including acid-citrate-dextrose (ACD), citrate-phosphate-dextrose (CPD), citrate-phosphate-dextrose-adenine (CPDA-1), or additive solutions (AS-1, AS-3, and AS-5). Additive solutions are designed to preserve RBCs and minimize hemolysis to less than 1% of the RBCs during prolonged storage. The storage life of RBCs varies, based on the type of anticoagulant-preservative solution, from 21 days (CPD) to 42 days (AS). The hematocrit of additive solution units ranges from 55% to 65%, whereas the
hematocrit of CPDA units is between 70% and 80%. These differences should be considered when calculating the amount of RBCs required to correct anemia.¹²

Leukoreduction

It is well known that leukoreduction reduces the frequency of febrile nonhemolytic transfusion reactions and CMV transmission.¹³^,¹⁴^,¹⁵^,¹⁶^,¹⁷^,¹⁸^,¹⁹^,²⁰ There is still controversy regarding the best strategies to reduce transfusion-transmitted CMV (see CMV discussion later in this chapter).

Leukoreduction of RBCs occurs via filtration soon after collection. Current guidelines state that leukoreduced RBCs must contain less than 5 × 10⁶ leukocytes per unit.¹² Universal leukoreduction of blood products is standard in Canada and much of Europe. However, in the United States, it remains controversial as a standard of care. Approximately 85% of cellular blood products are currently leukoreduced at US children’s hospitals.²¹

Irradiation

Irradiation of RBCs (and platelets) is recommended for immunocompromised patients and is effective in preventing transfusion-associated graft-versus-host disease (TA-GVHD) by halting donor T-cell replication and engraftment in the host (see TA-GVHD discussed later). Irradiation of cellular blood products is performed with either gamma rays from cesium-137 or cobalt-60 sources, or from X-rays.¹² T lymphocytes are rendered inactive with a minimum radiation dose of 2,500 cGy.

Directed Donation

Concerns about blood safety may influence a child’s family to seek out alternatives to transfusions from volunteer community blood donors. A directed blood donor is usually a relative or friend who donates blood specifically for a patient in advance of a scheduled procedure or transfusion. Directed donor units have similar rates of infectious disease marker positivity compared to volunteer community donor units when first-time blood donor status is taken into account.²² There is a concern that family members may feel pressure to donate a directed blood unit and may not reveal to their family or the donor center a critical piece of health information that should preclude them from donating. Additionally, several potential risks must be considered when a biologic parent wishes to serve as a directed donor to a child who may later need a hematopoietic stem cell transplant from a family member, including RBC and HLA (human leukocyte antigen) alloimmunization and transfusion-related acute lung injury (TRALI). These risks must be adequately explained to families wishing to proceed with directed donation, in addition to discussion of cost, timing of donation, and special needs of the recipient (for example, CMV-seronegative unit). Cellular blood components from blood relatives are irradiated to prevent TA-GVHD.

Dosage of Packed RBCs

The dose of packed RBCs is 10 to 15 mL per kg of child’s actual body weight infused over 3 to 4 hours. Care should be taken to maximize the use of each unit ordered. The expected increase in hemoglobin is dependent on the hematocrit concentration of the blood unit. Hemoglobin should increase by 2 g/dL for each 10 mL per kg of packed RBCs transfused of units stored in additive solutions (AS).²³ If CPD-1 RBCs are used, the expected hemoglobin increase is 3 g/dL for each 10 mL per kg of packed RBCs infused. This difference in expected hemoglobin increase reflects the packed RBC volumes in the different preparations.

Special Considerations for Transfusion of RBCs

Severe Anemia

The optimal rate of infusion of RBCs has not been well studied in children, but should be determined by the patient’s degree of anemia, acuity of onset of anemia, ongoing blood loss, and cardiovascular status. Traditional medical teaching suggests that children with a gradual onset of severe anemia (Hgb less than 5 mg/dL) should be transfused at a slow rate (1 mL/kg/hr) to prevent transfusion-associated circulatory overload (TACO) and resulting cardiac failure. However, a recent randomized clinical trial showed that in 43 children with severe anemia, but without evidence of cardiac failure, none developed signs of circulatory overload with RBC transfusion at rates of 1 mL/kg/hr or 3 mL/kg/hr.²⁴ Children with evidence of cardiac failure should be transfused slowly and diuretics given as necessary.²⁵

Hyperleukocytosis

Hyperleukocytosis, defined as a white blood cell (WBC) count of 100,000 per mm³ or greater, occurs in patients with acute lymphocytic leukemia (ALL), acute myeloid leukemia (AML), and chronic myelogenous leukemia (CML). Hyperleukocytosis can be life-threatening by inducing leukostasis, tumor lysis syndrome, and/or disseminated intravascular coagulopathy (DIC).²⁶ Leukostasis occurs when blasts accumulate and obstruct the vascular lumen, inducing hypoxia, especially in the central nervous system (CNS) and lungs, and this is particularly problematic in AML because of the stickiness of the blast cells.²⁷ Symptoms include dyspnea, hypoxia, and respiratory failure, as well as confusion, seizures, and cerebrovascular accident. This is a clinical diagnosis. Also, leukostasis can occur in children with a WBC count less than 100,000 per mm³, particularly if the blast count is rapidly increasing or in AML.

There have been no prospective randomized clinical trials to assess leukapheresis in adults or children with leukostasis.²⁸ Treatment remains controversial, though many centers advocate for cytoreduction with leukapheresis and/or chemotherapy with induction and/or hydroxyurea.²⁶ Early mortality in AML is not affected by treatment of leukostasis.²⁸ Transfusion of RBC may be hazardous in patients with hyperleukocytosis since RBCs further increase blood viscosity. If needed, RBC transfusion should be performed cautiously during or following leukapheresis.²⁷

Jehovah’s Witnesses

Jehovah’s Witnesses, based on their interpretation of the biblical scripture, will not accept whole blood, RBCs, platelets, plasma, or leukocyte transfusions. Some, however, agree to transfusion with blood derivatives such as albumin, immunoglobulin, and recombinant factor concentrates.²⁹ It is critical to ask patients and their families their preferences for blood products during informed consent.

If a family refuses blood products for their child based on religious beliefs, one should consider therapies that they may find acceptable, including decreased phlebotomy and nonblood adjunctive therapies such as iron, erythropoietin, and antifibrinolytic drugs. Of course, if transfusion for a minor is medically necessary, legal intervention should be sought when a child is at risk of death.³⁰ It is important to note that many Jehovah’s Witnesses families find this process acceptable.

Infants

Guidelines for transfusion in neonates and infants remain controversial. Because of conflicting long-term neurologic outcomes in premature infants in the PINT (Premature Infants in Need of Transfusion) trial versus a randomized control trial performed at the University of Iowa, the TOP (Transfusion of Prematures) trial is currently underway to assess liberal versus restrictive RBC transfusion strategies in infants.³¹^,³² In general, infants should be transfused RBCs to maintain hemodynamic stability. The specific transfusion thresholds can vary widely in healthy mature infants versus premature infants dependent on cardiopulmonary support.⁴

Radiation Therapy

Tissue hypoxia is thought to have a role in limiting the effectiveness of radiation therapy in causing tumor cell death. However, it
is not well understood what role underlying anemia plays in tissue hypoxia. Despite an incomplete understanding of the relationship between anemia and tissue hypoxia, RBC transfusions commonly are used to correct anemia during radiotherapy.³³ No pediatric studies have addressed the benefit or risk of RBC transfusion during radiotherapy. Two randomized trials in adult patients undergoing radiation therapy for squamous cell carcinoma of the head and neck have shown no benefit on outcome.³⁴^,³⁵ In addition, two retrospective studies in women with cervical cancer being treated with radiotherapy have shown no benefit from RBC transfusion, and, in fact, potential harm.³⁶^,³⁷ Patients with anemia may just have more aggressive disease than those who maintain a normal hemoglobin throughout their treatment.³³ Further research is necessary to assess the effect of RBC transfusion on radiation therapy.

End-of-Life Management of Anemia

Patients receiving palliative care may develop anemia related to their disease, marrow invasion, blood loss, or therapy. Fatigue is the most common symptom in children who have advanced cancer in the last month of life, and the majority of children and their families report significant suffering because of fatigue.³⁸ The cause of fatigue is multifactorial, and can include physiologic, psychological, and situational factors, including anemia and disrupted sleep.³⁹ No randomized control trials have been performed to assess the effectiveness of RBC transfusion on alleviation of fatigue in patients with advanced cancer.⁴⁰ However, studies in adults have shown that correction of anemia with either RBC transfusion or erythropoietin reduces fatigue and enhances the quality of life.³⁸ In children, erythropoietin improves anemia but does not improve the quality of life.⁴¹ Further studies are warranted in pediatric palliative care to address the best treatment of fatigue and anemia.

THROMBOCYTOPENIA

Platelet transfusions are common in the supportive care of pediatric patients with malignancy undergoing therapies which lead to hypoproliferative thrombocytopenia. Platelet transfusions in these conditions may be considered either prophylactic or therapeutic, depending on whether the goal is to prevent bleeding or stop it. Contemporary clinical practice has adopted the prophylactic approach as standard of care. The decision to transfuse platelets prophylactically, however, must be based on an assessment of risk versus benefit in an individual patient given controversial evidence for prophylactic platelet transfusions.

Practice Guidelines for Prophylactic Platelet Transfusions

In the 1960s and 1970s the preparation of platelet concentrates was optimized and platelet transfusions subsequently became standard practice to treat hemorrhage in patients with hypoproliferative thrombocytopenia.⁴² The criteria and thresholds for prophylactic platelet transfusions have continued to change over the last several decades as a consequence of observational and clinical trial data. For example, the early evaluation of prophylactic platelet transfusion at the National Cancer Institute (NCI) demonstrated less bleeding days in children with acute leukemia treated with repeated platelet transfusions versus those children who were not treated.⁴³ Prior to the routine transfusion of platelets, death from hemorrhage occurred in more than half of patients with acute leukemia treated at the NCI between 1954 and 1963.⁴²

Early efforts were made to establish a threshold platelet count for prophylactic transfusions based on the platelet count and its relationship to spontaneous hemorrhage. An important nonrandomized study showed gross hemorrhage (hematuria, hematemesis, melena) was more frequent in patients with platelet counts of less than 5,000 per mm³ than in patients with a platelet count between 5,000 and 100,000 per mm^3.⁴³ However, in this widely cited study there was very little difference in the frequency of hemorrhage across this wide range of thrombocytopenia. It also must be highlighted that patients in this era were frequently treated with aspirin, and quality control measures for platelet components were much less stringent. Despite these limitations, it became standard practice to transfuse platelets at a trigger of 20,000 per mm³.⁴⁴

The widespread use of a prophylactic platelet trigger of 20,000 per mm³ decreased the mortality from hemorrhage in patients with hematologic malignancies to less than 1%.⁴² As a result, the demand for platelet components increased, but this practice has its limitations. Platelet transfusions are expensive and the supply is limited. Platelet transfusions also can be associated with several adverse effects, including alloimmunization, transfusion reactions including TRALI, and bacterial sepsis (see later text). Thus, given the need to balance the benefits of transfusion with costs, safety and demand of a relatively strained resource, recent studies have again addressed the issue of defining the optimal clinical practice of prophylactic platelet transfusion.

Over the past several years there have been several large, randomized controlled trials, mainly in adult patients, designed to compare prophylactic platelet transfusion thresholds.⁴⁵^,⁴⁶^,⁴⁷^,⁴⁸^,⁴⁹ Heckman et al.⁴⁶ performed a prospective randomized trial in 78 adults undergoing induction therapy for acute leukemia. Patients with acute promyelocytic leukemia (APL), a known bleeding diathesis, coagulopathy, or platelet refractoriness were excluded from the study. The investigators demonstrated a reduction in the utilization of platelet transfusions for patients using a lower threshold of 10,000 per mm³ versus 20,000 per mm³ and found no statistical difference between the two groups regarding remission rate, hospital days, death during induction, RBC transfusion requirements, febrile days, or platelet refractoriness. No patient in either group died from hemorrhage.

In a large Italian multicenter randomized trial, Rebulla et al.⁴⁵ evaluated 255 adults undergoing remission induction for AML, excluding those with French-American-British M3 subtype (APL). Patients were randomized to a threshold of 20,000 or 10,000 per mm³ when in stable condition and a threshold of less than 20,000 per mm³ in the presence of fresh hemorrhage, fever of more than 38°C, or anticipation of invasive procedures. The threshold of 10,000 per mm³ was associated with a significant reduction in platelet requirements, and there were no significant differences in the number of patients experiencing severe hemorrhage, number of RBC products transfused, or induction deaths.

Other studies addressed the optimal platelet trigger in recipients of hematopoietic stem cell transplantation (HSCT).⁴⁷ Diedrich et al.⁴⁸ prospectively compared a threshold of 20,000 or 10,000 per mm³ in 159 patients (age range, 8 to 70 years) undergoing HSCT. They found no difference in the number of platelet units transfused or the incidence or severity of bleeding events. No deaths were attributed to bleeding. The authors concluded that a platelet threshold of 10,000 per mm³ was safe in HSCT recipients.

Taken together, the above studies now suggest that a platelet threshold of 10,000 per mm³ is both safe and effective for prophylaxis in oncology patients.

Therapeutic Platelet Transfusions

Therapeutic platelet transfusions are given for ongoing hemorrhage, the presence of clinical factors predisposing to bleeding, or before a surgical or invasive procedure (Table 39.1). Therapeutic platelet transfusions are prescribed to patients with thrombocytopenia or functional platelet impairment, or both, who have significant bleeding. Primary hemostasis is impaired when the platelet count drops below the normal range (150,000 to 400,000 per mm³). Easy bruisability, excessive menstrual blood loss, and epistaxis are reported more frequently when the platelet count drops below 75,000 per mm³. These symptoms become even more clinically significant when the platelet count drops below 20,000 per mm³. Petechiae, spontaneous hemorrhage, and
mucosal bleeding are exacerbated at platelet counts of 10,000 to 20,000 per mm³. However, as observed in the studies that utilized a prophylactic platelet transfusion threshold of 10,000 per mm³, it is unusual for a life-threatening hemorrhage to occur with platelet counts of 10,000 to 20,000 per mm³ unless aggravating clinical factors are present.⁴⁵^,⁵⁰^,⁵¹^,⁵²^,⁵³^,⁵⁴ Unfortunately, it is impossible to predict bleeding sites, and the first hemorrhage may be into a vital organ, resulting in severe morbidity or mortality.

TABLE 39.1 Platelet Transfusion Guidelines

Clinical Scenario		Transfuse When Platelet Count Less Than (thousand per mm³)
Well, stable		10
Procedures
	For diagnostic lumbar puncture and to avoid introduction of blasts	100
	Lumbar puncture (use lower number for clinically stable, sedated patients and experienced practitioner)	10-20
	Bone marrow aspirate	Not needed
	Surgery: minor (central venous catheter)	40-50
	Surgery: major (CNS, ophthalmologic, solid tumor biopsy/resection with anticipated blood loss)	50-100
Signs/symptoms
	Bleeding: minor (mucosal, epistaxis)	20
	Bleeding: major (hemoptysis, hemorrhagic cystitis, GI, CNS, tumor necrosis)	100
	APL (induction)	50
	CNS metastases (e.g., choriocarcinoma, melanoma, testicular carcinoma)	50
Coexisting laboratory abnormalities
	Hyperleukocytosis	20
	Coagulation abnormalities	50
	DIC	50
These recommendations are based on published guidelines and the standard of practice at Lucile Packard Children’s Hospital.⁶⁵^,⁷⁰
APL, acute promyelocytic leukemia; BT, bleeding time; CNS, central nervous system; DIC, disseminated intravascular coagulation; FDP, fibrin degradation product; GI, gastrointestinal.

The decision to transfuse platelets depends on many clinical factors besides the platelet count. These include the cause of thrombocytopenia; time of expected resolution; rapidity of platelet count drop; the functional ability of the platelets; and the clinical condition of the patient including the presence of fever, infection, coagulopathy, or bleeding. As the platelet count decreases, an increasing number of available (transfused) platelets are required to meet the need for hemostasis.⁵⁵ A minimal number of platelets may be required to maintain the integrity of the microvasculature.⁵⁵^,⁵⁶ This may explain why patients with chronic severe thrombocytopenia are more susceptible to spontaneous bleeding.

With improved supportive care and better understanding of the clinical situations predisposing to life-threatening hemorrhage, a logical next step may be to evaluate whether platelet transfusions should be administered only in the actively bleeding patient or in those with antecedent risk factors. In deciding when to transfuse, it is important to consider not only the absolute platelet count, but also the number of functional platelets. Posttransfusion platelet increments may not accurately predict function, and use of the platelet function assays such as thromboelastogram (TEG) or rotational thromboelastometry (ROTEM) may prove to be useful in the uncontrolled bleeding patient not adequately treated with platelet transfusion.

TABLE 39.2 Clinical Factors Increasing the Risk of Bleeding in Thrombocytopenic Children

Precipitous drop in platelet count
History of brisk or life-threatening hemorrhage (requiring resuscitative efforts or immediate PRBC transfusion)
Solid tumors: location (GI, bladder, pelvis), necrosis, radiation, recent biopsy
Comorbidities: fever, sepsis/infection, DIC, liver disease
Medications: amphotericin, NSAIDs, aspirin, fish oil
Phase of therapy: acute leukemia induction
Coagulopathy, disease-related (APL, NBL, tumor lysis, hyperleukocytosis)
Coagulopathy, acquired (liver disease DIC)
Splenomegaly
Alloimmunization

APL, acute promyelocytic leukemia; DIC, disseminated intravascular coagulation; GI, gastrointestinal; NBL, neuroblastoma; PRBC, packed red blood cells; NSAIDs, nonsteroidal anti-inflammatory drugs.

Clinical factors such as history of prior bleeding, site of bleeding, presence of fever or infection, degree of anemia, coexisting coagulopathy, rate of platelet count decrement, platelet consumptive states or medications, or hyperleukocytosis may predispose to bleeding at lesser degrees of thrombocytopenia than in the clinically stable patient (Table 39.2).⁴⁶^,⁵⁷^,⁵⁸^,⁵⁹^,⁶⁰^,⁶¹^,⁶²^,⁶³ An additional factor to consider is access to emergent medical care. Hospitalized patients tend to be closely observed, have frequent laboratory studies, have access to rapid intervention, and are less likely to be involved in trauma.⁵⁰ Outpatients, especially those living at a great distance from health care providers, may be more at risk due to less frequent monitoring, and transfusion support may be less readily accessible.⁶⁴

No randomized trials exist to answer the questions of what is the risk of bleeding in patients with severe, chronic, stable thrombocytopenia, and what is the ideal clinical treatment of such patients. Patients with myelodysplasia, aplastic anemia, and platelet refractoriness often have sustained and severe thrombocytopenia. Clinical experience suggests these patients may have no or minimal bleeding for long periods despite the low platelet counts.⁶⁵^,⁶⁶ Gmür et al.⁵⁹ reported an increased incidence of severe hemorrhage (one lethal) in four patients with platelet alloimmunization who experienced delays in transfusion due to difficulty obtaining HLA-matched products. Some patients may indeed be at risk for a severe or fatal hemorrhage, but it is not clear whether transfusing platelets when there is no incremental change will provide any benefit.

Researchers have studied a therapeutic strategy rather than a prophylactic one. A randomized clinical control trial by Wandt et al.⁴⁹ in 391 patients aged 16 to 80 years compared a therapeutic versus prophylactic platelet transfusion strategy in patients with AML and in patients undergoing autologous stem cell transplantation (SCT). The authors concluded that a therapeutic strategy was safe in patients undergoing autologous SCT, since there was a 34.1% decrease in the mean number of platelet transfusions without increased risk of major hemorrhage between the groups. However, in patients with AML, the risk of nonfatal grade 4 (mostly CNS) bleeding was increased in those patients in the therapeutic arm; and therefore they recommended continued prophylactic use of platelet transfusions.

Stanworth et al.⁶⁷ performed a randomized, open-label, noninferiority trial in 600 patients aged 16 years or older receiving
chemotherapy or undergoing SCT comparing therapeutic versus prophylactic platelet transfusion. Patients in the therapeutic arm had more days with bleeding and a shorter time to first bleed than patients in the prophylactic arm. There were similar rates of bleeding in the two groups undergoing autologous SCT. The authors recommended continued use of a prophylactic platelet transfusion strategy, though noted that a significant number of patients had bleeding despite prophylaxis.

It is now reasonable to assess the safety and feasibility of a therapeutic strategy in the current generation of pediatric cancer patients. Giving platelets to patients for specific indications is likely to reduce the use of platelet transfusions. Given the lack of clinical studies elucidating a clear platelet transfusion policy, it is not surprising that tremendous variability exists in platelet transfusion and dosing strategies among pediatric oncologists.⁶⁸ Clear criteria for both prophylactic and therapeutic indications remains an area of tremendous interest and it is recognized that more definitive data are needed.⁶⁹

Recommendations Regarding Surgical or Invasive Procedures in Thrombocytopenic Patients

Cancer patients frequently require invasive diagnostic or therapeutic procedures such as placement of central venous access devices, tumor biopsy or resection, bone marrow aspirates and biopsies, lumbar punctures, and occasionally bronchial-alveolar lavage (BAL), paranasal sinus aspirations, or endoscopic biopsies. No randomized studies have been performed to determine accurate, safe platelet thresholds for major or minor procedures in children with malignancy. On the basis of an accumulated clinical experience, consensus panels have concluded that a platelet count of 50,000 per mm³ is sufficient for major surgery and 20,000 per mm³ is safe for the performance of minor procedures, assuming the absence of associated coagulation abnormalities.⁶⁵^,⁷⁰ However, clinical practice varies and some oncologists use higher platelet counts, also taking into consideration the extent and site of surgery (Table 39.1). Bone marrow aspiration and biopsy can be performed in patients with severe thrombocytopenia without platelet support, providing that adequate surface pressure is applied.⁷¹^,⁷²^,⁷³

Conventional wisdom among oncologists has been that the most experienced practitioner available should perform the procedure. Supervision by an experienced practitioner is advisable for those in training. Bleeding complications from procedures are often related to procedural problems and lack of experience rather than patient factors, such as the platelet count.⁷¹

Lumbar Puncture

Lumbar puncture is perhaps the most common procedure performed on children with leukemia, for diagnostic purposes and to instill intrathecal chemotherapy. Most pediatric oncologists administer prophylactic platelet transfusions to thrombocytopenic patients before lumbar puncture. However, this practice is controversial because a safe platelet count has not been determined for this procedure. Thrombocytopenia is not a proven risk factor for permanent neurologic injury.⁷²

A retrospective review of children with newly diagnosed leukemia who underwent a diagnostic lumbar puncture did not report serious complications, regardless of platelet count.⁷² The reviewers evaluated 5,223 procedures performed during remission induction or consolidation for ALL; 941 had platelet counts less than 50,000 per mm³ and 199 had platelet counts of 20,000 per mm³ or less. They concluded that prophylactic platelet transfusion is not necessary in children with platelet counts higher than 10,000 per mm³. Only 29 spinal taps were performed in children with a platelet count less than 10,000 per mm³, precluding a statistically significant conclusion on its safety. Interestingly, they noted a 10.5% rate of traumatic lumbar punctures (defined as 500 or more RBCs per high-powered microscopic field).

A large retrospective study (by the same investigators as above) evaluated 5,609 lumbar punctures and looked at several modifiable and unmodifiable risk factors of traumatic (10 RBCs per µL) or bloody (500 RBCs per µL) spinal taps.⁷¹ They found that African American race, young age (<1 year), prior traumatic tap within 2 weeks, or prior lumbar puncture performed while the platelet count was 50,000 per mm³ or less were factors that increased the risk for a traumatic tap. Modifiable risk factors included lack of general anesthesia, platelet count of 100,000 per mm³ or less, an interval of less than 15 days between lumbar punctures, or a less experienced practitioner. Although the authors had previously determined that 10,000 per mm³ was a safe platelet level for routine lumbar puncture with instillation of chemotherapy, they now recommend, on the basis of this study, that the initial diagnostic lumbar puncture be performed with a platelet count of 100,000 per mm³.⁷¹ This is to ensure that the interpretation of the cerebrospinal fluid is not obscured by excessive RBCs. In addition, the authors reviewed data suggesting that bloody spinal taps may be of prognostic significance with respect to possible introduction of circulating leukemic blasts into the cerebrospinal fluid.⁷⁴

Several studies have shown an association between increased incidence of CNS leukemia relapse and traumatic lumbar puncture when circulating blasts are present at the time of diagnosis.⁷⁴^,⁷⁵ Based on studies from St. Jude Children’s Research Hospital, many leukemia practitioners now desire a platelet count of greater than 100,000 per mm³ prior to a child’s first diagnostic tap despite the safety data showing no risk of bleeding at platelet counts of above 10,000 per mm³.⁷² After the diagnostic lumbar puncture, there is much variability in practice surrounding adequate platelet count prior to instillation of intrathecal chemotherapy.

Placement of a Central Venous Access Device

A number of studies support the safe practice of central venous catheter insertion in a stable patient with a platelet count of 50,000 per mm³.⁷⁶^,⁷⁷^,⁷⁸^,⁷⁹ Several investigators have performed such procedures, including implantable infusion ports and tunneled Silastic catheters, at platelet levels of 30,000 to 50,000 per mm³ with acceptable side effects.⁸⁰^,⁸¹^,⁸² Central venous catheters should be placed by experienced and skilled physicians in patients with severe thrombocytopenia.⁷⁷ For severely thrombocytopenic patients, perioperative platelet infusion at the time of central line placement is appropriate to decrease bleeding risk even after platelets fall to pretransfusion levels hours after surgery.⁸²

Platelet Component Preparation

Platelets can be prepared by either separation of platelet concentrates from whole blood or by apheresis from a single donor. Most US blood centers provide apheresis platelet units.⁸³ Either product can effectively address thrombocytopenia. However, apheresis platelets have less risk of infectious disease transmission and bacterial contamination.⁸⁴

After centrifugation of whole blood, each whole-blood-derived unit of platelets must contain at least 5.5 × 10¹⁰ platelets. Each unit of whole-blood-derived platelets contains about 50 to 70 mL of plasma, and this can be pooled if necessary for a larger platelet dose. Apheresis platelets are collected from a single donor and must yield a minimum of 3 × 10¹¹ platelets per US Food and Drug Administration (FDA) mandate. One apheresis unit is equivalent to approximately 4 to 6 whole-blood-derived platelet units. The volume of a unit of apheresis platelets is approximately 250 mL, but these units can be divided into split units or aliquots if necessary to yield a smaller dose in an infant or small child.⁷

Apheresis platelet units are leukoreduced at the time of collection by modern apheresis machinery.⁸⁵ Platelets derived from whole blood are often pooled and leukoreduced by filtration prior to storage. Both platelet products can be stored for up to 5 days
after collection at 20°C to 24°C with continuous agitation.⁸⁵ In the United States, platelet components must be tested for bacterial contamination and found to be negative 24 to 48 hours after collection, prior to their transfusion.⁸⁶

Dosage for Platelet Transfusion

A calculated platelet dose of 10 mL per kg of body weight of either whole-blood-derived or apheresis platelets should result in a platelet count increment of 50,000 to 100,000 per mm³.²³ Whole-blood-derived platelets are ordered by 1 unit per 10 kg of body weight for children weighing more than 10 kg. For infants weighing less than 10 kg, one whole-blood-derived platelet unit is an appropriate dose.⁷ Apheresis platelet units are ordered by single units for children weighing 30 kg or more, or have a total body surface area more than 0.5 m². One-half of an apheresis unit is appropriate for a child weighing between 10 and 30 kg, or with a total body surface area of less than 0.5 m². For infants weighing less than 10 kg, syringe aliquots of 10 mL per kg should be ordered if available.

The 1-hour corrected count increment (CCI) is a very accurate determination of the response to platelet transfusion.⁸⁷ The CCI is calculated as follows:

CCI = Platelet count increment (platelet/mL) × 10¹¹ × Body surface area (m²)/Number of platelets transfused × 10¹¹

For example, a child with ALL undergoing induction chemotherapy with a body surface area (BSA) of 0.7 m² has a morning platelet count of 7, 000. After an infusion with one apheresis unit of platelets (assume this contains 3.0 × 10¹¹ platelets), a 1-hour post-platelet infusion count is 77,000.

CCI = (77,000 – 7,000) × 10¹¹ × 0.7 m²/3.0 × 1011 = 10,333 per m². The expected CCI is about 15,000 per m² of body surface per unit of platelets. One widely accepted definition of platelet refractoriness is two 1-hour CCI values on consecutive days of less than 5,000 per m².⁸⁷ The optimal dose of platelets was most recently addressed in a prospective, multicenter randomized trial termed the PLADO.⁸⁸ Almost 200 children were enrolled in this study of approximately 1,100 hospitalized patients who were thrombocytopenic as a result of chemotherapy for hematologic malignancy or HSCT. There was no clinically significant difference in bleeding measured between the patient groups randomized to receive 1.1 × 10¹¹, 2.2 × 10¹¹ or 4.4 × 10¹¹ platelets per m² per prophylactic transfusion for platelet counts <10,000/µL.⁸⁸

Infusion Volumes and Rates

Platelets should be infused through a dedicated line over 30 to 60 minutes, depending on the volume. The patient should be monitored closely during the transfusion, and especially within the initial 15 minutes of infusion. Some institutions require infusion at a slower rate during the initial 15 minutes of the transfusion (such as 1 mL per kg per hour), and then increase the rate as tolerated. However, rapid infusion rates (as high as 10 mL per kg per hour) have not been shown to adversely affect the quality of transfused platelets and may provide direct benefit to the patient.⁸⁹ Mild reactions, including hives, may be treated by administering an antihistamine such as diphenhydramine and then resuming the infusion at a slower rate. Severe reactions, including anaphylaxis, require immediate cessation of the transfusion and supportive care.

Posttransfusion Platelet Count

Traditionally, platelet counts have been measured at 60 minutes posttransfusion to assess response. However, the platelet count can be monitored as soon as 10 minutes after transfusion because of rapid equilibration.⁹⁰ This measurement is suitable for decisions regarding surgical procedures and assessing refractoriness.

Only gold members can continue reading. Log In or Register to continue