Handbook of Cancer Chemotherapy (Lippincott Williams & Wilkins Handbook Series), 8th Ed.

28. Transfusion Therapy, Bleeding, and Clotting

Mary R. Smith and NurJehan Quraishy

Disorders of the hemostatic mechanisms are common in patients with malignancy. Active cancer accounts for 20% of new venous thromboembolism (VTE) events. Patients who present with unprovoked VTE have a 10% risk of developing cancer within the next 2 years. Abnormalities associated with thromboembolic events cause significantly more morbidity and mortality than disorders leading to hemorrhage.


A. Pathophysiology

The thromboembolic risk associated with neoplasia reflects an imbalance between platelet number, platelet function, levels of coagulation factors, and generation of thromboplastins compared with the levels of inhibitors of hemostasis and fibrinolytic activity. Thrombosis may be minor and localized or widespread and associated with multiple-organ damage. There may also be hemorrhage of varying degrees of severity in association with the thromboembolic events.

1. Factors that may affect the risk of thromboembolism vary widely from patient to patient and include the following:

bull Specific type of tumor: adenocarcinomas (ovary, pancreas, colon, stomach, lung, and kidney)

bull Nutritional status of the patient

bull Type of chemotherapy

bull Response to chemotherapy (e.g., tumor lysis syndrome)

bull Liver and renal function

bull Patient immobility and venous stasis

bull Surgery (twice the risk for VTE and three times the risk for pulmonary embolism [PE] compared to patients without cancer)

bull Arterial and venous catheters.

2. Factors that can initiate thrombus formation are common to many cancers:

bull Circulating tumor cells adhere to the vascular endothelium and form a nidus for clot formation.

bull Tumors may penetrate the vessel, destroying the endothelium and promoting clot formation.

bull Neovascularization associated with many tumors may stimulate clotting.

bull Arterial thrombosis associated with tumors may result from vasospasm.

bull A systemic hypercoagulable state develops (e.g., decreased protein C).

bull External compression of vessels by tumor masses impedes blood flow and leads to stasis and clot development.

3. Platelet abnormalities associated with an increased risk of thromboembolism include thrombocytosis and increased platelet adhesion and aggregation. Tumors may produce substances that cause increased platelet aggregation with subsequent release of platelet factor 3 and ensuing acceleration of coagulation.

B. Clinical syndromes

A variety of noteworthy clinical syndromes are associated with the “hypercoagulable state” of malignancy and of its treatment.

1. Disseminated intravascular coagulation (DIC) is a syndrome with many signs, symptoms, and laboratory abnormalities (Table 28.1). As many as 90% of patients with metastatic neoplasms have some laboratory manifestation of DIC, but only a small fraction of these patients suffer morbidity from the coagulation process or subsequent depletion of coagulation factors and consequent bleeding due to DIC. The initiating factor for DIC is apparent in some situations but unknown in others.


Among the common initiators of DIC are the following:

bull Thromboplastic substances in granules from promyelocytes of acute promyelocytic leukemia (DIC may worsen with therapy). There is a significant concomitant fibrinolysis in many patients.

bull Sialic acid from mucin produced by adenocarcinomas of the lung or gastrointestinal tract

bull Trypsin released from pancreatic cancer

bull Impaired fibrinolysis associated with hepatocellular carcinoma

bull DIC in any patient may be fostered by sepsis or other causes of the systemic inflammatory response syndrome (SIRS).

2. Lupus anticoagulant in neoplastic disease. The lupus anticoagulant is an antiphospholipid antibody (immunoglobulin G or M). Antiphospholipid antibodies are reported to be associated with a number of malignant disorders including hairy cell leukemia, lymphoma, Waldenström macroglobulinemia, and epithelial neoplasms. The lupus anticoagulant leads to a prolonged activated partial thromboplastin time (aPTT) but is paradoxically associated with an increased risk of thrombosis.

3. Trousseau syndrome (tumor-associated thrombophlebitis). Suspect the possibility of neoplasia in the following circumstances:

bull An unexplained thromboembolic event occurs after the age of 40 years.

bull Thromboses occur in unusual sites.

bull The thromboses affect superficial as well as deep veins.

bull The thromboses are migratory.

bull The thromboses tend not to respond to the “usual” anticoagulant therapies.

bull An unexplained thrombosis occurs more than once.

4. Thrombotic events that occur after surgery for tumors of the lung, ovary, pancreas, or stomach.

5. Nonbacterial thrombotic endocarditis may be found in association with carcinoma of the lung. These thrombi are formed from accumulations of platelets and fibrin. The mitral valve is the most frequent site of origin of these thrombi, which frequently embolize.

6. Thrombotic thrombocytopenic purpura (TTP) is a poorly understood syndrome characterized by thrombocytopenia, microangio-pathic hemolytic anemia, fever, fluctuating neurologic signs and symptoms, and acute renal failure. TTP and the hemolytic-uremic syndrome (thrombocytopenia, hemolysis, and acute renal failure) have been associated with untreated malignancies as well as with a number of drugs used for treating malignant disease. The agent most often reported is mitomycin, but other drugs including bleomycin, cisplatin, cyclophosphamide, gemcitabine, and vinca alkaloids may also be associated with these syndromes. TTP may be difficult to diagnose in this setting because the chemotherapy suppresses platelet production, some agents may impair renal function, and many of the features of DIC are similar to those of TTP. Careful review of the peripheral blood smear is required to identify the changes in red blood cells (RBCs) that are associated with a microangiopathic hemolytic process.

There is growing evidence that damage to the endothelium is seen in association with TTP. For many patients with TTP, von Willebrand-cleaving protease (ADAMTS13) levels are very low or absent, leading to the accumulation of unusual large multimers of von Willebrand factor (vWF) and subsequent platelet clumping. The von Willebrand-cleaving proteolytic activity is thought to be inhibited by an anti-vWF–cleaving protease immunoglobulin G antibodies.

The prognosis of patients with TTP is poor, and its therapy has been varied. Plasmapheresis and transfusion with fresh frozen plasma (FFP) appear to be the best modalities of therapy. Plasmapheresis is most frequently used as it not only replaces the von Willebrand-cleaving protease missing or decreased in patients with TTP but also removes the anti-vWF–cleaving protease antibody.

Complications from platelet transfusions are not as common in TTP associated with malignancy and bone marrow transplantation as in other cases of TTP; thus, platelet transfusion can be used especially if there is a threat of bleeding.

7. Thromboembolism associated with chemotherapy

a. The use of central arterial or venous catheters has markedly facilitated the delivery of chemotherapy, but all such catheters are associated with a 5% increase in the risk of vascular thrombosis. This risk level is lower than was previously suspected. The empiric use of low doses of warfarin (1 mg/day) or low-dose low-molecular-weight heparin (LMWH) has been evaluated in recent randomized trials, and both drugs failed to show any reduction in symptomatic catheter-associated thrombosis. The risk of catheter-associated thrombosis appears to be higher in children. In a recent large randomized trial, there was an increased risk of bleeding in patients treated with low-dose warfarin.

b. Many chemotherapy agents cause significant chemical phlebitis. The most common offending agents are mechlorethamine (nitrogen mustard), anthracyclines, nitrosoureas, mitomycin, fluorouracil, dacarbazine, and epipodophyllotoxins.

c. L-asparaginase inhibits the synthesis of proteins, including coagulation factors. This inhibition may cause either hemorrhage or thrombosis. Patients with pre-existing hemostatic disorders are at particular risk for complications when using L-asparaginase. L-asparaginase also decreases antithrombin-III (AT-III) activity.

d. Tamoxifen, when used as an adjuvant, has been associated with a two- to sixfold increased risk of thromboembolic events. This effect may be magnified when tamoxifen is combined with chemotherapeutic agents. When tamoxifen is used for primary prevention, the risk of deep vein thrombosis and PE is 1.6% and 3.0%, respectively. Other selective estrogen receptor modulators, like raloxifene, are also associated with an increase in the risk of thromboembolic events. Aromatase inhibitors have a lower risk of thromboembolism than tamoxifen and are preferred for postmenopausal women, particularly if there are additional risk factors for thrombosis.

e. Estrogens may increase the risk of thromboembolism. This is likely due, at least in part, to a decrease in protein S and an increase in coagulation factors.

f. Superior vena cava syndrome is nearly always associated with thrombosis in the thoracic venous system cephalad to the site of obstruction and may lead to upper-extremity thrombosis.

g. Antiangiogenic or targeted therapy may be associated with a significant increase in the risk of VTE events. Thalidomide and lenali-domide in combination with corticosteroids or chemotherapy increases the risk of symptomatic VTE in patients with multiple myeloma. Prophylaxis with low doses of anticoagulation therapy has not been formally evaluated in this group of patients.

C. Principles of therapy for thrombosis associated with neoplasia

1. Discrete vascular thrombosis

a. General guidelines. Therapy should be directed at controlling the neoplasm. As an anticoagulant, heparin is superior to warfarin in these patients. Warfarin and antiplatelet drugs have been used with varying degrees of success in some patients with thromboembolism associated with tumors. The use of heparin, warfarin, and antiplatelet agents alone or in combination may be associated with normalization of hemostatic parameters. Despite this, patients with malignant disease are often resistant to anticoagulant therapy and may continue to have thrombotic events even while receiving what appears to be adequate treatment. Great care must be exercised in the use of both heparin and warfarin in patients with malignant disease because hemorrhage into areas of necrotic tumor can be hazardous. The use of anticoagulant therapy is generally contraindicated in patients with central nervous system metastases. Bulky disease is a relative contraindication, especially if central necrosis of the tumor is suspected and particularly if the lesion is in the mediastinum or pleural spaces.

The decision to treat thromboembolism occurring in a patient with malignancy may be difficult. One must carefully weigh the risks of therapy against expected benefits. The patient's life expectancy, concurrent therapy, and type of malignancy also influence the decision.

b. Heparin. Low doses of heparin (5000 U given via the subcutaneous [SC] route every 12 hours) can be used to protect patients with malignant disease from thromboembolism during perioperative periods.

Heparin may be used as the initial or long-term therapy for thromboembolic events in patients with malignant disease. Heparin may be administered either by the intravenous (IV) or SC routes. Generally, the IV route is preferred for initial therapy so that the anticoagulant effect begins at once and adjustment of doses can be easily achieved. An initial dose of 5000 U (70 U/kg) of heparin is given as an IV bolus followed by 1000 to 1200 U (15 U/kg)/h as a continuous infusion. One should check the aPTT 1 hour after the heparin bolus to ensure that the patient is heparinizable (i.e., not AT-III deficient), 6 hours after beginning therapy, and 6 hours after any change in the dose of heparin. Some patients with malignant disease may appear to be refractory to heparin; in all likelihood, this reflects low levels of AT-III, owing to poor production or increased consumption, both of which may occur in patients with malignant disease. (Infusion therapy with L-asparaginase has been associated with reduced levels of AT-III.) As long as the AT-III activity is above 50% of normal, it is usually possible to achieve the desired anticoagulant effect if adequate doses of heparin are given. If AT-III activity is less than 50% of normal, AT-III may be replaced using AT-III concentrates or FFP.

Heparin may be administered by the SC route for both the acute and the chronic management of thromboembolism associated with malignancy. Using the SC route may be less desirable when treating acute events because the onset of anticoagulant effect is somewhat slower (2 to 3 hours), and adjusting the therapeutic effect may be more difficult. SC heparin can be considered for chronic therapy provided that the patient can manage the twice-daily injection and weekly monitoring of the aPTT. In a patient who has been receiving IV heparin, half the total dose of IV heparin received in the previous 24 hours should be given SC twice a day (e.g., 1000 U/h by IV infusion equals 12,000 U SC twice a day). For the patient being started on SC heparin, the initial dose is 7500 to 10,000 U SC twice a day. The aPTT should be checked 6 hours after the third dose of heparin. Otherwise, the aPTT should be checked 6 hours after a SC dose of heparin. The goal for the aPTT should be similar to that of IV heparin, namely, 1.5 to 2 times the patient's baseline aPTT.

LMWHs should now be considered as the first line of therapy or prevention of VTE in patients with malignant disease. LMWH is preferred to unfractionated heparin as it can be given as an outpatient more easily and has a lower risk of heparin-associated thrombocytopenia. LMWH may have specific antineoplastic effects separate from its effect to reduce VTE. There are a number of ongoing trials designed to evaluate this antineoplastic effect. Monitoring of LMWH is indicated if the patient suffers from liver or kidney dysfunction or if the patient is significantly malnourished or debilitated. One must use anti-Xa levels as the aPTT is not indicative of the anticoagulant effect of LMWH.

c. Warfarin. In the past, warfarin was the therapy of choice for the chronic management of thromboembolic events associated with malignant disease. The use of warfarin in this setting is of concern because patients with malignant disease are frequently taking multiple medications that can alter the patient's response to warfarin. An additional concern about the use of warfarin in patients with malignancy is the development of purpura fulminans. This complication may be due to lower-than-normal protein C levels in patients who had DIC before initiation of warfarin therapy. Warfarin should not be used as the primary drug to manage VTE events or future prevention of events in patients with malignant disease.

d. The use of platelet-inhibiting drugs such as aspirin, other nonsteroidal anti-inflammatory agents, and dipyridamole has met with varying degrees of success in the prevention of repeated thromboembolic events in patients with malignant disease. Care must be taken with the use of such drugs, especially in thrombocytopenic patients, because the risk of bleeding associated with thrombocytopenia is increased.

e. Fibrinolytic therapy. Systemic malignancy is a relative contraindication to fibrinolytic therapy.

f. Vascular interruption devices such as inferior vena cava filters should only be used in patients who cannot tolerate anticoagulant therapy or who develop emboli while on adequate anticoagulant therapy.

g. Anticoagulation therapy should be continued as long as the patient is receiving anticancer therapy or has evidence of active cancer.

2. Disseminated intravascular coagulation. Therapy for DIC includes the following:

bull Urgently correct shock (if present).

bull Treat the underlying disease process. When it is not possible to treat the underlying disease process, it is unlikely that the complicating DIC can be successfully managed.

bull Replace depleted blood components (e.g., platelets, cryopre-cipitated antihemophilic factor [AHF] for fibrinogen and factor VIII, FFP for other factors) if clinically significant bleeding is present.

bull Consider the use of heparin only in the following situations:

bull In patients with acute promyelocytic leukemia (see Chapter 18).

bull When there is clear evidence of ongoing end-organ damage due to microvascular thrombosis.

bull If venous thrombosis occurs.

These latter two complications ofDIC are most likely to occur as a component of SIRS, and the treatment of the underlying cause of SIRS is necessary in addition to treatment with heparin. There is no evidence that chronic warfarin therapy is of value for treating the chronic DIC seen in some patients with neoplasia if thromboses are absent. Warfarin may predispose to the development of purpura fulminans in the presence of chronic DIC due to acquired protein C deficiency.


A. Tumor invasion

It is well recognized that bleeding may be a warning sign of cancer. Bloody sputum may indicate carcinoma of the lung, blood in the urine may be a sign of carcinoma of the bladder or kidney, blood in the stool may be due to carcinoma of the alimentary tract, and postmenopausal vaginal bleeding may be caused by endometrial carcinoma. In each of these instances, bleeding can be directly related to the invasive properties of cancer and disruption of normal tissue integrity.

B. Hemostatic abnormalities

Often, bleeding in patients with cancer is not due to the direct effects of the neoplasm but rather to indirect effects of the cancer or its therapy on one of the components of the hemostatic system. Because of the frequency and the special management problems caused by abnormalities in the hemostatic system in patients with cancer, it is important to consider the possible causes and corrective measures in detail.

1. Increased vascular fragility may be due to chronic corticosteroid therapy, chronic malnutrition, or “senile purpura.” Bleeding is usually not severe, but bruising, particularly around IV sites, is common. Hemostatic therapy is not necessary.

2. Thrombocytopenia may occur for a variety of reasons. Some of the more common causes are as follows.

a. Chemotherapy and radiotherapy regularly cause depression of platelet production. Serial blood cell counts must be monitored while patients are being treated.

b. Bone marrow invasion or replacement causing thrombocytopenia is commonly seen only with leukemias or lymphomas but may occur in other cancers that invade the bone marrow.

c. Splenomegaly with splenic sequestration is most common with leukemia or lymphoma.

d. Folate deficiency with decreased platelet production is common in patients with cancer because of poor nutrition. Dietary history should provide the clues to the diagnosis.

e. Neoplasm-induced immune thrombocytopenic purpura. Patients with lymphoproliferative malignancies (e.g., chronic lymphocytic leukemia, Hodgkin disease) often develop immune thrombocytopenic purpura (ITP). ITP may also be the presenting symptom of a nonhematologic malignancy. Usually, the ITP improves with prednisone 1 mg/kg/day followed by treatment of the malignancy.

f. Drug-induced immune thrombocytopenia. Many nonchemo-therapy medications used to treat patients with malignancy can cause immune thrombocytopenia. Offending agents to consider are heparin, vancomycin, H2-receptor antagonists, penicillins, cephalosporins, interferon, sulfa-containing antibiotics, diuretics, and hypoglycemic agents.

g. Graft-versus-host disease developed after bone marrow transplantation may produce a chronic (often isolated) immune-mediated thrombocytopenia. The platelet count may respond to increased immunosuppression.

3. Abnormalities of platelet function must be suspected in patients who have a normal or near-normal platelet count and signs or symptoms of bleeding and a documented prolonged bleeding time. Most cases are secondary to drug effects including aspirin and other nonsteroidal anti-inflammatory agents, antibiotics (e.g., ticarcillin), antidepressants (e.g., tricyclic drugs), tranquilizers, and alcohol. Consider any drug that the patient is taking as a possible offender until proved otherwise. The presence of fibrin degradation products is a common cause of platelet dysfunction in patients with malignancy who also have DIC. Platelet dysfunction may occur in patients with malignant paraproteinemias as a result of the coating of the platelet surfaces by the immunoglobulin. When renal failure develops or is present in such patients, the platelet dysfunction is magnified.

4. Coagulation factor deficiencies may develop in patients with malignancy for several reasons:

bull Acute (decompensated) DIC depletes most clotting factors but to variable degrees.

bull Liver failure causes deficiency of all clotting factors except factor VIII.

bull Malnutrition leads to deficiency of factors II, VII, IX, and X (the vitamin K–dependent factors).

bull Fibrinolysis may be due to the release of urokinase in prostate cancer or secondary to DIC. This may produce hypofibrinogenemia as well as fibrin split products, which act as circulating anticoagulants.

bull Functionally abnormal clotting factors are occasionally seen. The most commonly diagnosed abnormality is dysfibrinogenemia.

5. Acquired circulating anticoagulants may develop in patients with a number of different tumors. Many of these anticoagulants are heparinoid in nature. The most common associations are with carcinoma of the lung and myeloma. Other anticoagulants act as antithrombins; in this case, the most common association is with carcinoma of the breast.

6. Chemotherapy and other drug-induced bleeding

a. Mithramycin, although rarely used now, may lead to platelet dysfunction and a reduction in multiple coagulation factors. Hemorrhage due to these effects may occur in up to half of patients treated with mithramycin.

b. Anthracyclines may be associated with primaryfibrinolysis or fibrinogenolysis and hemorrhage.

c. Dactinomycin is a powerful vitamin K antagonist that causes defective synthesis of all vitamin K–dependent proteins ( factors II, VII, IX, and X, protein C, and protein S).

d. Melphalan, cytarabine, doxorubicin, vincristine, and vinblastine are all associated with platelet dysfunction.

e. Mitomycin, daunorubicin, cytarabine, bleomycin, cisplatin, tamoxifen, pentostatin, gemcitabine, atorvastatin, clopidogrel, ticlopidine, cyclosporine, sulfonamides, tacrolimus, sirolimus, “crack” cocaine, penicillin, rifampin, penicillamine, oral contraceptives, arsenic, quinine, and iodine are all associated with TTP.


About half of all patients with cancer and about 90% of those with metastases manifest abnormalities of one or more routine coagulation parameters (Table 28.2). These abnormalities may be minor early in the patient's disease, but as the disease progresses, the hemostatic abnormalities become more pronounced. Serial coagulation tests may offer the clinician a clue to response to therapy or recurrence of malignant disease and are more valuable than a single determination in patients with no symptoms of hemostatic disruption.


A. Screening tests for bleeding

The following tests provide an adequate screening battery: platelet count, bleeding time or whole blood platelet function screening testing, aPTT, prothrombin time (PT), thrombin time, and fibrinogen level.

B. Interpretation of screening laboratory studies

Abnormal results of the screening tests reflect hematologic problems caused by blood vessels, platelets, or coagulation factors. The following list provides clues to the interpretation of the screening test results that help determine the most likely cause or causes of the patient's bleeding.

1. Platelet count. A normal platelet count it 150,000 to 450,000/μL. If thrombocytopenia is less than 100,000/μ.L, consider the following:

bull Bone marrow failure

bull Increased consumption of platelets

bull Splenic pooling of platelets.

Thrombocytosis with a platelet count of more than 500,000/μ.L has the following characteristics:

bull It is common in patients with neoplasms.

bull It may be seen in association with iron deficiency (e.g., secondary to gut neoplasm).

bull It usually poses no risk of arterial thrombosis unless the patient has a myeloproliferative disorder.

2. Bleeding time. This is a useful screening test if the platelet count is normal and platelet dysfunction is suspected.

bull A normal bleeding time requires normal platelet number, normal platelet function, and normal function of the blood vessels and connective tissues.

bull A prolonged bleeding time may be due to thrombocytopenia, abnormal platelet function, and, rarely, inadequate vessel functions. The bleeding time may be spuriously prolonged in elderly people with “tissue-paper” skin.

bull The following formula is a rough rule of thumb to be used to estimate what the bleeding time should be in patients who have platelet counts between 10,000 and 100,000/(μL. Although it was derived using the Mielke template, the principle should still hold for contemporary bleeding time devices: bleeding time (min) = 30 – ([platelet count/(μL]/4000). (Whole blood platelet function screening testing is replacing the bleeding time in many laboratories. This method of screening for platelet function abnormalities appears to be a better predictor for the risk of bleeding due to platelet function abnormalities than the bleeding time.)

3. Prolonged PT. This is seen in the presence of the following:

bull Deficiency of one or more of the following clotting factors: VII, X, V, II (prothrombin), or I (fibrinogen); oral anticoagulant therapy leads to a deficiency of factors II, VII, IX, and X

bull Circulating anticoagulants against factor VII, X, V, or II

bull Dysfibrinogenemia.

4. Prolonged aPTT. This is seen in the presence of the following:

bull Deficiency of any of the following clotting factors: XII, XI, IX, VIII, X, V, II, or I. Factor XII deficiency is not associated with bleeding. Fletcher and Fitzgerald factor deficiencies (both rare) may also prolong the aPTT.

bull Circulating anticoagulants directed against the factors mentioned above or the lupus inhibitor.

bull Anticoagulant therapy with heparin or oral anticoagulants.

5. Prolonged thrombin time. Prolongation of the thrombin time may be due to the following:

bull Hypofibrinogenemia (fibrinogen less than 100 mg/dL)

bull Some forms of dysfibrinogenemia

bull Fibrin-fibrinogen split products

bull Heparin therapy

bull Paraproteins.

If the thrombin time is prolonged, further studies to clarify the cause may be required.

6. Low fibrinogen level. When evaluating the results of a fibrinogen assay, one must be familiar with the assay method used. Many laboratories use immunologic assays, which measure both functionally normal and abnormal fibrinogens. If such an assay is in use, the thrombin time can be used to evaluate the functional integrity of the fibrinogen. A low functional fibrinogen level means that production is decreased, consumption is increased, or a dysfibrinogen is present. Fibrinogen is an acute-phase reactant and is often elevated with advanced malignancy. A fibrinogen level in the normal range may actually be relatively low for the patient's physiologic state and thus may be a sign of DIC (see Table 28.1).

C. Laboratory findings in patients with DIC

Acute DIC is often associated with significant hemorrhage, whereas chronic DIC may be asymptomatic or associated with thromboses. Screening and confirmatory laboratory tests are shown in Table 28.1.

D. Review of peripheral smear

Review of peripheral smear for schistocytes and decreased numbers of platelets is needed if TTP is suspected.



A. Transfusion therapy

1. General guidelines

a. Regard elective transfusion with allogeneic blood as an outcome to be avoided. Blood is not risk-free. Consider the factors that will influence the use of blood products, including the following:

bull Alternative forms of therapy that could control bleeding (e.g., topical measures or desmopressin).

bull How symptomatic is the patient? Do not treat an abnormal laboratory test in a symptom-free patient. For example, patients with chronic DIC may demonstrate prolongation of both the PT and the aPTT and mild to moderate throm-bocytopenia. If there is no demonstrable bleeding, transfusion therapy is not necessary.

b. Use the specific blood component needed by the patient.

c. Minimize complications of transfusion by using the following:

bull Only the amount and type of blood product indicated for the patient in the specific clinical setting.

bull Leukocyte-reduced blood products, irradiated blood, or both, when indicated.

d. Obtain a class I human leukocyte antigen (HLA) type on the patient at diagnosis to allow for better support of platelet transfusions.

2. Blood component therapy

a. RBC transfusions

(1) Available forms of RBCs for transfusion. The primary component available for red cell transfusion is RBCs.

(2) Criteria for transfusing RBCs. Chemotherapy-induced anemia is common in patients with cancer and should preferably be managed by erythropoiesis-stimulating agents, such as recombinant erythropoietin and/or hematinics such as iron, folic acid, and vitamin B12, as applicable. RBCs are indicated for the treatment of symptomatic anemia to increase the oxygen-carrying capacity by increasing red cell mass. RBCs should not be used to increase well-being, promote wound healing, increase oncotic pressure, or as a source of blood volume.

Transfusion should be based on clinical assessment and not a laboratory value. In general, RBC transfusion may be indicated in the following circumstances:

bull Hemoglobin less than 7 g/dL if clinically indicated. Acceptable transfusion goal is to maintain hemoglobin at 7 to 9 g/dL.

bull Any hemoglobin level in the presence of acute life-threatening hemorrhage with evidence of hemodynamic instability or inadequate oxygen delivery.

bull Hemodynamically stable anemia with hemoglobin less than 10 g/dL with acute coronary syndrome (either angina or acute myocardial infarction).

bull Hemoglobin less than 10 g/dL causing or contributing to symptomatic anemia (not explained by other causes), with excessive fatigue, dyspnea on exertion, syncope, or with signs such as tachycardia, tachypnea, and postural hypotension.

(3) Expected response. One unit of RBCs will increase the hemoglobin by approximately 1.0 g/dL (hematocrit by 3%) in a stable nonbleeding patient. However, the response may vary depending on the hemoglobin content of the unit and the patient's blood volume.

(4) Check hemoglobin. Hemoglobin equilibrates in 15 minutes in a stable nonbleeding patient. Use the posttransfusion hemoglobin along with the patient's clinical status to determine if additional RBCs transfusion is needed.

b. Platelet transfusions

(1) Available forms of platelets for transfusion. Platelets may be ordered and transfused in various ways. Because most patients with an underlying malignancy have the potential for needing long-term platelet support, platelet products should be leukocyte reduced from the initiation of transfusion (see Section IV.A.3). In general, patients who need platelet support can be started with platelets (whole-blood–derived platelet concentrates) or apheresis platelets (single-donor platelets). Many blood centers have geared up production of apheresis platelets, and this product may be more readily available in some places than in others. There are no solid data to suggest that starting with apheresis platelets decreases the incidence of alloimmunization. In fact, the Trial to Reduce Alloimmunization to Platelets Study Group (1997) did not show any benefit in the use of platelet apheresis products over whole-blood–derived platelets. HLA-compatible platelets or crossmatched platelets should be reserved for patients who have become refractory to whole-blood–derived platelets or apheresis platelets and are alloimmunized [see Section IV.A.2.b.(4)]. Patients who are candidates for bone marrow transplantation should receive single-donor platelets (if available) from the initiation of platelet therapy. Patients who are candidates for transplantation with bone marrow from an HLA-matched sibling should not receive apheresis products from the potential donor before the transplantation.

(a) Platelets (whole-blood–derived platelet concentrates, random-donor platelets, or platelet concentrates). Four to six U (usually pooled in one bag) are considered an adequate dose for a 70-kg adult.

(b) Apheresis platelets obtained by apheresis (single-donor platelets). These come as a single pack and represent the platelets obtained by apheresis from a single donor. One unit of platelets obtained by apheresis is equivalent to 6 to 8 U of platelet concentrates.

(2) Check platelet count 10 minutes to 1 hour, and then 24 hours, after platelet transfusion to estimate recovery and survival of platelets in the patient. Each unit of platelet concentrate should increase the platelet count by about 7000/μL. The expected 1-hour posttransfusion rise in platelets is 15,000 platelets/μL divided by the patient's body surface area in square meters for each unit of platelet concentrate (thus, for a person of 2 m2, 6 U should produce a rise of 45,000/μL [6 × 15,000/2]). The corrected count increment (CCI) may also be used to determine the response. The CCI is determined by the platelet count increment (posttransfusion – pretransfusion count) multiplied by the body surface area in m2 and divided by the platelet count in the product. A CCI of greater than 7500 between 10 minutes to 1 hour posttransfusion is considered an adequate response and represents 20% to 30% platelet recovery. At 24 hours, a CCI of greater than 4500 is adequate. A reduced CCI at 18 to 24 hours after a normal CCI at 10 minutes to 1 hour is believed to represent increased consumption of platelets due to nonimmune clinical events [see Section IV.A.2.b(4)]; in some patients, it may represent immune destruction.

(3) Criteria for transfusing platelets

(a) For patients with reduced platelet production, criteria for transfusion are shown in Table 28.3.


(b) Increased platelet destruction. Platelet transfusions are of limited benefit in patients with thrombocytopenia due to increased destruction as a result of either antibodies or consumption. If potentially life-threatening bleeding complicates thrombocytopenia due to increased destruction, platelet transfusions may be given; however, only small increments in the platelet count usually occur. IV γ-globulin 1 g/kg IV daily × 2 days given before the platelet transfusions might improve the response.

(c) Dysfunctional platelets. One must stop any drugs known to cause platelet dysfunction. Although the use of platelet transfusions should be considered, pharmacologic methods of enhancing platelet function, such as desmopressin, should be used if possible (see Section IV.B.1).

(4) Refractoriness to platelet transfusions is defined as a CCI of less than 7500 at 10 minutes to 1 hour or less than 4500 at 18 to 24 hours after transfusion of 5 to 6 U of platelet concentrates or 1 U of apheresis platelets after two or more transfusions. Refractoriness may be the result of alloimmunization (i.e., formation of alloantibodies to HLA or rarely platelet-specific antigens) or to other immune or nonimmune causes such as fever, septicemia, DIC, splenomegaly, drugs, infections, or bleeding. Alloimmunization is the most difficult form to treat and therefore is best prevented (see Section IV.A.3).

(a) Evaluation. Patients who become refractory to platelet transfusions should have a laboratory evaluation for alloimmunization. They should also be evaluated for infection and DIC. Further, all potentially offending medications should be stopped.

(b) Therapy. The therapeutic modalities for ITP (corticosteroids, IV globulin, danazol) are generally ineffective for platelet refractoriness due to alloimmunization. Two therapeutic options exist, as follows.

(i) HLA-compatible platelets. These platelets may be provided using two approaches: the donor platelets have HLA class I antigens similar to the patient: HLA antigen matching; an alternative approach is to provide platelets that lack HLA class I antigens to which the patient has made antibodies: the antibody specificity prediction method. A combination of the two may provide the best match.

(ii) Crossmatched platelets. Because platelets are available at most blood centers, if a blood center performs crossmatching, it is often easier to obtain crossmatched platelets as no specific donor qualification is required. This product is as effective as HLA-compatible platelets in producing a platelet response in the alloimmunized patient. Either platelet concentrates or apheresed platelets can be crossmatched with the recipient. Nonreactive or, in extenuating circumstances, the least reactive platelets can then be selected for transfusion.

c. Coagulation factor support

(1) FFP contains all clotting factors (but not platelets) and should be used for multiple coagulation factor deficiencies. FFP requires 20 to 30 minutes to thaw and must be thawed at 37°C. Once thawed, FFP (thawed plasma) must be transfused within 5 days of thawing as long as it is maintained at 1 to 6°C.

(2) Plasma frozen within 24 hours of phlebotomy is often used interchangeably with FFP.

(3) Cryoprecipitated antihemophilic factor is a source of factor VIII-vWF complex, fibrinogen, and factor XIII. Each bag of cryoprecipitated AHF contains about 50% of the factor VIII-vWF complex (minimum of 80 U) and 20% to 40% of the fibrinogen (minimum of 150 mg) harvested from 1 U of blood. Cryoprecipitated AHF is stored in a frozen state and has the advantage of concentrating the clotting factors in a small volume (10 to 15 mL/bag). It is used primarily in deficiencies of fibrinogen. The goal is to keep the fibrinogen level higher than 100 mg/dL. The usual dosage of cryoprecipitated AHF to correct hypofibrinogenemia is one bag of cryoprecipitated AHF for every 10 kg of body weight. Because 50% is recovered after transfusion, this may raise the fibrinogen level only by about 50 mg/dL. Larger doses may be needed for severe hypofibrinogenemia or “flaming” DIC. The dose of cryoprecipitated AHF may be calculated using the following formula: number of bags = desired increase in fibrinogen level in mg/dL × plasma volume/average fibrinogen per bag (average fibrinogen content may be obtained from the blood supplier). The patient is evaluated to determine if the laboratory values have been corrected.

(4) Factor IX concentrates are available as factor IX complex concentrates, which contain factors II, VII, IX, and X, or as coagulation factor IX concentrates. The latter are highly purified factor IX concentrates with few or no other coagulation factors. Several precautions are worth noting regarding the factor IX concentrates:

bull This concentrate is made from pooled plasma but is treated with viral attenuation processes such as dry or vapor heat in the case of factor IX complex concentrates (therefore, the risk of hepatitis is significant) and solvent-detergent or monoclonal antibody in the case of coagulation factor IX concentrates. The dose depends on the preparation to be used. The goal is to bring the factor concentration to no more than 50% of normal.

bull There is a small risk of DIC resulting from the use of factor IX complex concentrates. Newborns and patients with liver dysfunction are at increased risk. The coagulation factor concentrates are far less thrombogenic and should be used in cases at increased risk for venous thrombosis or DIC.

bull Factor IX concentrates are stored in the lyophilized state. Do not shake when reconstituting.

3. Leukocyte reduction. Patients who have not previously received transfusions and who will need long-term blood product support should receive leukocyte-reduced (5 × 106 leukocytes/bag) blood products. Leukocyte reduction may prevent febrile transfusion reactions, prevent cytomegalovirus (CMV) infections, and delay or prevent alloimmunization. Controversy still exists as to whether tumor recurrence and infections are a result of immunomodulatory effects of blood transfusion and if they can be reduced by leukoreduction. There is also ongoing controversy as to whether leukocyte-reduced blood products should be provided to all patients, not just patients at risk. Leukocyte reduction does not prevent graft-versus-host disease, but irradiation does (see Section IV.A.5).

Two methods of leukocyte reduction by filtration are currently available: bedside and prestorage.

a. Bedside filtration involves leukocyte reduction at the time of transfusion. Disadvantages include plugging of the filter, the presence of leukocyte breakdown products, bag breakage, and lack of consistency of products. Filters are available for RBCs and platelets.

b. Prestorage leukocyte-reduced RBCs are RBCs that have been leukocyte-reduced generally within 8 to 24 hours of collection. Advantages are fewer leukocyte breakdown products, ease of administration, and consistent quality (95% of products with less than 5 × 106 leukocytes/bag). Cost may be perceived as a disadvantage. However, this is offset by the expense of stocking filters, training of staff in the use of filters, and breakage. Likewise, whole-blood-derived platelets may be prestorage leukocyte reduced by filtration. Some apheresis machines leukocyte reduce apheresis platelets during collection.

4. CMV-seronegative cellular blood products. Only patients known to be anti-CMV negative with impaired immunity should be considered for the use of CMV-negative screened blood. This group includes children, for the most part. Some institutions also consider provision of CMV-seronegative units for bone marrow transplant candidates. The use of CMV-seronegative blood seriously restricts the potential donor pool for these patients. Leukocyte-reduced blood products (5.0 × 106/bag) are considered CMV-“safe” products and may be used as an alternative to CMV-seronegative products based on institutional policy. Only cellular blood products transmit CMV.

White blood cell (WBC) depletion filters also remove CMV because CMV resides in the WBCs. Irradiation of blood products does not render them CMV-free. Frozen deglycerolized blood is considered free of CMV contamination.

5. Irradiated blood products. These prevent the development of graft-versus-host disease. Irradiated blood products, in the case of patients with cancer or hematologic malignancies, are indicated in the following situations:

bull Congenital immunodeficiency

bull Bone marrow, peripheral blood stem cell, or umbilical cord stem cell transplantation

bull Directed blood donations to blood relatives

bull HLA-compatible or crossmatched platelets

bull Granulocyte transfusions

bull High-dose chemotherapy with growth factor or stem cell rescue

bull Hodgkin lymphoma

bull Leukemia and non-Hodgkin lymphoma (relative indications)

bull Purine analogues, such as, fludarabine treatment

bull Significant lymphocytopenia with alemtuzumab treatment.

B. Other forms of therapy

1. Desmopressin 0.3 μg/kg IV over 30 minutes every 12 to 24 hours for 2 to 4 days may be used to elevate factor VIII and vWF levels as well as improve platelet function. Tachyphylaxis may occur if therapy is continued for longer periods. Intranasal desmopressin 0.25 mL twice a day using a solution containing 1.3 mg/mL has been given for minor bleeding episodes.

2. Fibrin glue. This is a topical biologic adhesive. Its effects imitate the final stages of coagulation. The glue consists of a solution of concentrated human fibrinogen, which is activated by the addition of bovine thrombin and calcium chloride. The resulting clot promotes hemostasis and tissue sealing. The clot is completely absorbed during the healing process. The best adhesive and hemostatic effect is obtained by applying the two solutions simultaneously to the open wound surface. Fibrin glue has been used primarily in surgical settings. It has been most effective when used for surface, low-pressure bleeding. There is a small risk of anaphylactic reaction because of the bovine origin of the thrombin.

3. Antifibrinolytic agents. ε-Aminocaproicacid(EACA)andtranexamic acid have been used to control bleeding associated with primary fibrinolysis as seen in patients with prostatic carcinoma and in a small number of patients with refractory thrombocytopenia. Great care must be taken in the use of these agents because of a possible increased risk of thrombosis. EACA may be used topically to control small-area, small-volume bleeding.

4. Oprelvekin (interleukin-11) maybe used for the treatment and prevention of chemotherapy-related thrombocytopenia. Oprelvekin is a thrombopoietic growth factor that directly stimulates the proliferation of hematopoietic stem cells and megakaryocyte progenitor cells as well as megakaryocyte maturation, resulting in increased platelet production. It may cause substantial fluid retention and should be used with caution in patients who have congestive heart failure (CHF), those with a history of CHF, and those being treated for CHF. One must also be cautious using this agent in patients who are receiving diuretic therapy or ifosfamide because sudden deaths have been reported as a result of severe hypokalemia. Oprelvekin should be used to prevent thrombocytopenia, which would be severe enough to require platelet transfusions. Therapy usually begins 6 to 24 hours after the completion of chemotherapy, and patients should be monitored for any signs or symptoms of allergic reactions or cardiac dysfunction.

Selected Readings

Anand SS, Wells PS, Hunt D, et al. Does this patient have deep vein thrombosis? JAMA. 1998;279:1094–1099.

Ansell J, Hirsh J, Hylek E, et al. Pharmacology and management of the vitamin K antagonists. Chest. 2008;133:160S–198S.

Baker WF Jr. Thrombosis and hemostasis in cardiology: review of pathophysiology and clinical practice. II. Recommendations for anti-thrombotic therapy. Clin Appl Thromb Hemost. 1998;4:143–147.

Blajchman MA, Goldman M, Freedman JJ, Sher GD. Proceedings of a consensus conference: prevention of post-transfusion CMV in the era of universal reduction. Transfus Med Rev. 2001;15:1–20.

Callum JL, Karkouti K, Lin Y. Cryoprecipitate: the current state of knowledge. Transfus Med Rev. 2009;23:177–188.

AABB, American Red Cross, America's Blood Centers, Armed Services Blood Program. Circular of Information for the Use of Human Blood and Blood Components. Bethesda, MD: AABB; 2009.

Dwyre DM, Holland PV Transfusion-associated graft-versus-host disease. Vox Sang. 2008;95:85–93.

Hillyer CD, Emmens RK, Zago-Novaretti M, et al. Methods for the reduction of transfusion transmitted cytomegalovirus infection: filtration versus the use of seronegative donor units. Transfusion.1994:34:929–934.

Hull RD, Pineo GF. Prophylaxis of deep vein thrombosis and pulmonary embolism: current recommendations. Clin Appl Thromb Hemost. 1998;4:96–104.

Humphries JE. Transfusion therapy in acquired coagulopathies. Hematol Oncol Clin North Am. 1994;8:1181–1201.

Klein HG, Spahn DR, Corson JL. Red blood cell transfusion in clinical practice. Lancet. 2007;370:415–426.

Lee AY. Thrombosis and cancer: the role of screening for occult cancer and recognizing the underlying biological mechanisms. Hematology Am Soc Hematol Educ Program. 2006;438–443.

Legler TJ, Fischer I, Dittman J, et al. Frequency and course of refractoriness in multiply transfused patients. Ann Hematol. 1997;74:185–189.

Napolitano LM, Kurek S, Luchette FA, et al. Clinical practice guideline: red blood cell transfusion in adult trauma and critical care. Crit Care Med. 2009;37:3124–3157.

National Comprehensive Cancer Network (NCCN). NCCN Clinical Practice Guidelines in Oncology. Cancer- and chemotherapy-induced anemia. www.nccn.org.

Preiksaitis JK. The cytomegalovirus-“safe” blood product: is leukoreduction equivalent to antibody screening? Transfus Med Rev. 2000;14:112–136.

Rizzo DJ, Somerfield MR, Hagerty KL, et al. Use of epoetin and darbepoetin in patients with cancer: 2007 American society of clinical oncology/American society of hematology clinical practice guideline update. J Clin Oncol.2008;26:132–149.

Schiffer CA, Anderson KC, Bennett CL, et al. Platelet transfusion for patients with cancer: clinical practice guidelines of the American Society of Clinical Oncology. J Clin Oncol. 2001;19:1519–1538.

Shlicter SJ. Platelet transfusion therapy. Hematol Oncol Clin NAm. 2007;21:697–729.

Shlicter SJ, Kaufman RM, Assmann SF, et al. Dose of prophylactic platelet transfusions and prevention of hemorrhage. N Engl J Med. 2010;362:600–613.

The Thrombosis Interest Group of Canada: http://www.tigc.org/eguidelines/cancer06.htm.

Vassallo RR. New paradigms in the management of alloimmune refractoriness to platelet transfusions. Curr Opin Hematol. 2007;14:655–663.

Wiesen AR, Hospenthal DR, Byrd JC, et al. Equilibration of hemoglobin concentration after transfusion in medical inpatients not actively bleeding. Ann Intern Med. 1994;121:278–280.