Published: November 2013
The accurate diagnosis and treatment of patients with bleeding requires some basic understanding of the pathophysiology of hemostasis. The process is divided into primary and secondary components. Primary hemostasis is the formation of a platelet plug at the site of injury and occurs within seconds of injury. Secondary hemostasis involves the coagulation system and results in fibrin formation. It requires several minutes to complete. The fibrin strands strengthen the hemostatic platelet plug. This process is particularly important in bleeding originating from large vessels and in preventing recurrent bleeding hours or days after the initial injury (Figure 1).
Bleeding disorders can thus be categorized into three groups: disorders of platelet function or number, disorders of clotting factors, and a combination of these. Platelet disorders, including von Willebrand's disease, are discussed elsewhere in this section ("Disorders of Platelet Function and Number"). A focused history, physical examination, and screening laboratory tests are of paramount importance in directing the evaluation in a time-saving and cost-effective manner.
The evaluation of the bleeding patient should be focused primarily on whether and how actively the patient is bleeding. This may not be apparent if blood loss has been intermittent or gradual. Hence, deciding whether the patient is bleeding should not be based solely on a declining hemoglobin level or sudden hypotension. Bleeding into a third space, particularly after surgery or trauma, needs to be considered. Hemolysis, sequestration or hemodilution represent other causes of a decrease in the hemoglobin level.
A complete history including a past medical history should be taken; a history of human immunodeficiency virus (HIV) infection, liver or kidney disease, or malabsorption is often associated with bleeding. A medication history should be obtained, with particular attention to anticoagulants, nonsteroidal anti-inflammatory drugs (NSAIDs), oral contraceptives, antibiotics, ethanol, and dietary vitamins K and C. The response to past hemostatic challenges, such as trauma, tooth extraction, pregnancy, surgery, sports, and menstruation, should be determined. A family history of bleeding disorders may be helpful for assessing pathologic bleeding. The physical examination should focus on identifying signs of bleeding (e.g., petechiae, mucosal bleeding, soft tissue bleeding, ecchymoses) as well as signs of systemic disease.
If bleeding is suspected, one must identify the site and severity, duration of bleeding, and clinical setting. Mucocutaneous bleeding suggests a platelet disorder. It manifests as petechiae, ecchymoses, epistaxis, and genitourinary and gastrointestinal bleeding. Bleeding into potential spaces (joints, fascial planes, and retroperitoneum) suggests a coagulation factor deficiency. In hospitalized patients, bleeding from multiple sites can be seen with disseminated intravascular coagulation (DIC) or thrombotic thrombocytopenic purpura (TTP).
Acute extensive mucocutaneous bleeding in a patient previously without symptoms should suggest immune thrombocytopenic purpura (ITP). Spontaneous soft tissue hematomata or extensive bruising and oozing from multiple sites in previously asymptomatic patients might suggest accidental warfarin ingestion or acquired factor VIII inhibitors. The latter may be seen particularly in older patients and in patients with rheumatologic diseases. Postoperative bleeding at a surgical site is usually related to a local surgical problem. Spontaneous or excessive posttraumatic (immediate or delayed) bleeding can indicate a localized pathologic process or a disorder of the hemostatic process.
Initial laboratory investigations should include a complete blood count (CBC), prothrombin time (PT), activated partial thromboplastin time (aPTT), platelet function screening, and examination of a peripheral blood smear. A preoperative screen for a patient with a negative history and examination should include a CBC, PT, and aPTT only. The following is a brief description of tests available for the evaluation of hemorrhagic disorders.
In the platelet count, always verify thrombocytopenia by reviewing a peripheral smear. For example, if there is platelet clumping with the usual ethylenediaminetetraacetic acid (EDTA), it does not represent true thrombocytopenia; an accurate platelet count can be obtained by using citrated or heparinized tubes.
The aPTT represents the time for clot formation after adding calcium, phospholipids, and kaolin to a sample of citrated blood. It is prolonged by heparin, direct thrombin inhibitors, a deficiency of or inhibitor of factors in the intrinsic and common pathways (e.g., factors II, V, VIII, IX, X, XI, and XII) as well as lupus anticoagulant, vitamin K deficiency, or severe liver disease. Lupus anticoagulants interfere with the activity of the phospholipids used to trigger the clotting in the laboratory, and results may vary for different commercial preparations.
The PT represents the time for clot formation after the addition of thromboplastin (tissue factor) and calcium to citrated blood. It is prolonged with deficiencies of factors II, V, VII, and X or fibrinogen, liver disease, vitamin K deficiency, and warfarin use. Currently, this time is accompanied by the International Normalized Ratio, which compares it to the laboratory's baseline normal. Table 1 illustrates causes of a prolonged PT, aPTT, or both.
|Prolonged PT||Prolonged aPTT||Prolonged PT and aPTT|
|Factor VII deficiency||vWF, factor VIII, IX, XI, or XII deficiency||Prothrombin, fibrinogen, factor V, X or combined factor deficiency|
|Vitamin K deficiency||Heparin use||Liver disease|
|Liver disease||Inhibitor of vWF, factors VIII, IX, XI or XII||DIC|
|Warfarin use||Antiphospholipid antibodies||Supratherapeutic heparin or warfarin|
|Factor VII inhibitor||
The thrombin time (TT) is the time to clot formation after the addition of thrombin to citrated blood. The TT is prolonged by heparin, direct thrombin inhibitors, fibrin degradation products (FDPs), paraproteins, and fibrinogen deficiency (qualitative and quantitative). Protamine is added to neutralize the heparin so that the TT can be interpreted without heparin interference. This assay has been used to establish the presence of adequate fibrinogen but is not being used as widely now.
Reptilase time measures the time to clot formation after the addition of reptilase, a thrombin-like snake enzyme, to citrated blood. Unlike the TT, it is not affected by heparin. It can be useful to determine whether heparin is the cause of the prolonged TT.
The 1:1 mixing study is done when the PT or aPTT is prolonged. The patient's plasma is mixed with normal plasma, and the test is repeated. If the mixing of normal plasma corrects the abnormal result (PT or aPTT), a factor deficiency is suggested; otherwise, an inhibitor is suspected. Similarly, an incubated mixing study is done 1 hour (and occasionally 2 hours) after mixing of the patient's plasma with normal plasma. It is used to differentiate a lupus anticoagulant from clotting factor inhibitors; the latter usually react immediately, whereas the former may have a delayed inhibition and become abnormal only after incubation).
The urea clot solubility test relies on the ability of urea to dissolve unstable clots, which are formed in the absence of factor XIII. Normal clots are not dissolved by urea or monochloroacetic acid, unlike clots in patients with factor XIII deficiency.
Fibrin degradation products are fragments resulting from the action of plasmin on fibrin or fibrinogen and reflect high fibrinolysis states (such as DIC), when their levels are elevated.
D dimers are formed when crosslinked fibrin is degraded. They can be measured specifically by enzyme-linked immunosorbent assay (ELISA). Their level is usually higher in DIC and in thrombotic conditions, such as deep venous thrombosis and pulmonary embolism. Their elevation in the absence of symptoms does not imply the presence of these disorders.
Platelet function screening is performed on the PFA-100, which is a platelet function analyzer that largely has supplanted the bleeding time in the clinical arena. It tests the ability of platelets to aggregate while flowing through two capillary tube cartridges. In one, they are exposed to collagen–adenosine diphosphate [ADP] and in the other to collagen-epinephrine. It has a reported sensitivity of approximately 95% and specificity of approximately 89% in detecting platelet dysfunction, and a 98% positive predictive value in detecting aspirin-induced platelet defects.
Screening platelet function for detecting von Willebrand disease (vWD) has a low negative predictive value and may require repeat testing. One also may measure von Willebrand factor (vWF) antigen (vWF:Ag) by immunoassay and vWF activity by measuring the ability of patient's vWF to agglutinate normal platelets in the presence of ristocetin (vWF:RCo) or by its ability to bind collagen (vWF:CB). Factor VIII:C activity is a functional assay for factor VIII that is measured by mixing normal plasma with factor VIII-deficient plasma. Levels of vWF:Ag and vWF:RCo may be elevated during pregnancy, oral contraceptive use, and liver disease. They may decreased with hypothyroidism and type O blood.
Platelet aggregation studies remain the most sensitive method for detecting and distinguishing platelet function defects. In these tests, platelet aggregation is tested by measuring changes in optical density as platelets respond to various agents: ADP, epinephrine, collagen, arachidonic acid, and ristocetin. Different disorders will show different patterns. For example, platelets of patients with Glanzmann thrombasthenia (dysfunctional or deficient glycoprotein IIb/IIIa complex in platelets) only aggregate with ristocetin, whereas platelets of patients with Bernard-Soulier syndrome (absent or decreased glycoprotein Ib complex on platelets) have no aggregation with ristocetin, reduced aggregation with collagen, and normal aggregation with ADP, arachidonic acid, and epinephrine. These are discussed thoroughly in the chapter Disorders of Platelet Function and Number.
Deficiencies or inhibitors of clotting factors, whether acquired or inherited, can result in bleeding disorders. Figure 1 illustrates the current understanding of the coagulation cascade. This section provides an overview of the hemophilias and of the less-common coagulation factor deficiencies and inhibitors.
Hemophilias are most commonly the X-linked recessive diseases characterized by deficiency either of factor VIII (hemophilia A) or factor IX (hemophilia B, or Christmas disease). The incidence is 1 per 5000 live births for hemophilia A and 1 per 30,000 live births for hemophilia B. In 30% of patients, hemophilia is the result of a de novo mutation, and no family history can be elicited. Males are most commonly affected; however, symptomatic females have been documented, and the proposed mechanisms include X chromosome inactivation or deletion, or the presence of a true homozygous offspring of an affected father and a carrier mother. Coinheritance of the factor V Leiden mutation occurs in about 5% of patients and results in a decreased bleeding tendency. The clinical severity correlates well with factor levels; they are clinically classified as mild (>5% of normal factor activity), moderate (1%-5% factor activity), and severe (<1% factor activity).
The most common bleeding sites are joints (80% of bleeding), muscles, and the gastrointestinal mucosa. Ankles are the most commonly affected joints in children, whereas knees and elbows are more often involved in adults. Quadriceps and iliopsoas bleeding are the most common sites of muscle hematomas. Abdominal wall bleeding and gastrointestinal mucosal bleeding can occur. In most children, the hemophilia is already known at the time of first bleeding because of previous screening for a positive family history. In severe disease, bleeding occurs in the first 2 years of life; this contrasts with patients with milder disease, whose hemophilia can go undiagnosed for years. Late complications include hemarthroses and joint destruction, blood-borne infectious complications such as HIV and hepatitis acquired from donor-derived replacement factors, and development of clotting factor inhibitors.
The diagnosis is suggested by an elevated aPTT level in a male patient with a positive family history. This is tyically followed by a 1:1 mixing study with normal plasma, which corrects the aPTT. Hemophilia B patients might have a normal aPTT. Factors VIII and IX levels are decreased in hemophilias A and B, respectively. A two-stage assay is recommended. It is more technically demanding but it is also more accurate, especially in patients with a mutation near the A domain of factor VIII, which makes the mutant factor less stable. Carrier detection often relies on DNA-based methodology rather than finding 50% factor activity. Genetic testing can identify patients at risk for inhibitor development; patients with a missense mutation or small deletion are less likely to develop inhibitors than patients with nonsense mutations or large deletions.
Prevention. Prevention of bleeding includes wearing bicycle helmets, avoidance of contact sports, good oral hygiene, careful immunization techniques, timely replacement therapy after trauma, and treatment of acute bleeding episodes. Primary prophylactic therapy has been shown to reduce the incidence of arthropathy in severely affected patients. However, considerable controversy surrounding factor use remains, especially with regard to the age at onset when this therapy is initiated and the expense.
Local Control. When the operative site is well visualized, such as with dental work or in-office gynecological procedures, topical thrombin preparations may be sufficient. Recombinant human thrombin avoids the problems of raising antibodies against bovine thrombin or the potential of viral transmission posed by human thrombin isolated from donor pools. Absorbable hemostatic agents such as gelatin sponges, oxidized regenerated cellulose, and microfibrillar collagen, only provide a framework on which clot may form and require an intact coagulation system.
Factor Replacement. Choices for replacing factor include recombinant versus plasma-derived factor VIII. Plasma-derived concentrates vary in purity and must undergo viral inactivation procedures. First-generation recombinant products (Bioclate, Helixate FS, Kogenate, Recombinate) have demonstrated efficacy in clinical trials; however, they continue to carry a theoretic risk of viral transmission because of the added human albumin necessary for factor stabilization. Second-generation recombinant products (Kogenate FS and B-domain deleted recombinant factor VIII [BDDrFVIII]) do not require albumin stabilization. Factor IX replacement has traditionally been with prothrombin complex concentrates (PCCs) that contain factors II, VII, and X, as well as IX, and were associated with thrombotic risk. Newer plasma-derived factor IX concentrates are effective, and they undergo viral inactivation. Recombinant factor IX concentrates such as Benefix are also effective and have no added albumin, hence eliminating a theoretic risk of viral infection.
The choice of replacement therapy depends on availability, safety, and cost, with the knowledge that plasma-derived products are becoming safer and recombinant products are less available and two or three times more costly. For patients with HIV infection, the use of recombinant products has been associated with a slower decline in CD4 cell count, and many experts recommend using recombinant products in that setting. However, it is unknown whether this advantage translates into improved clinical outcome. There is a suggestion that using recombinant products can result in a higher likelihood of inhibitor formation, which would be a theoretical benefit to the use of human products.
The desired factor level depends on the site and severity of bleeding: 30% to 40% factor activity is required for early joint or muscle bleeding, 50% factor activity is required for dental surgery or more-severe muscle bleeding, and 80% to 100% factor activity is required for life-threatening or serious bleeding (intracranial, intra-abdominal, or orthopedic surgery). Because the half-life of factor VIII is about 12 hours and that of factor IX 16 hours, factor levels should be checked at least every 12 hours and 16 hours, respectively.
The following formula is used to calculate the required dosage of factor VIII replacement:
Dose (units) = (desired factor level – baseline factor level) x (patient weight [kg]/2)
For example, if a 60-kg patient with 1% factor level needs to undergo dental surgery, which requires correction to 50% of factor VIII level, 1500 units of factor VIII must be administered initially. This is usually followed by maintenance dosing (every 12 hours for factor VIII–deficient patients) at one half the initial dose. For patients with factor IX deficiency, the initial dose is calculated using the following formula:
Dose (units) = (desired factor level – baseline factor level) x patient weight (kg) x 1.2
The maintenance dose is usually equal to one half the initial dose and is given daily. Monitoring through factor levels is usually recommended after major trauma, bleeding, or surgery.
Desmopressin. Desmopressin (DDAVP) is the treatment of choice in patients with mild hemophilia A and mild to moderate bleeding, but it has no role in hemophilia B. It increases the production of Factor VIII and von Willebrand Factor by the endothelium. The use of desmopressin is described for the treatment of von Willebrand disease (vWD) elsewhere in this section ("Disorders of Platelet Function and Number").
Antifibrinolytic Therapy. Antifibrinolytic therapy (in the form of tranexamic acid or ε-aminocaproic acid) is useful in controlling oral cavity bleeding and menorrhagia. For details on dosing, see elsewhere in this section for its use in patients with vWD ("Disorders of Platelet Function and Number").
Recombinant Factor VIIa. Hemophilia patients with high titers of inhibitors respond to recombinant factor VIIa. Although the recommended dosage schedule is 90 µg/kg every 2 hours until hemostasis is achieved, lower doses may be effective. Patients with lower titers of inhibitors might respond to higher doses of factor VIII replacement or factor VIII bypassing products such as activated prothrombin complex concentrates.
Treatment of Long-Term Complications. Treatment of the long-term complications of hemophilia requires a multidisciplinary approach. Chronic hemarthroses may be managed with short-term prophylaxis and, at times, requires synovectomy, which can use arthroscopy or radioactive phosphorus. Experimental treatments with gene-targeted therapies are undergoing testing.
Most clotting factors are synthesized by the liver parenchyma. A final terminal carboxylation step which depends on vitamin K is required for Factors II, VII, IX and X, as well as for procoagulant Proteins C and S. Factor VIII, von Willebrand factor (vWF), and tissue plasminogen activator are produced in the endothelium, including that of the liver. The liver reticuloendothelial system is responsible for metabolizing most clotting factors and fibrin degradation products. The characteristics of the coagulopathy of liver disease are presented in Table 2. Treatment involves correction of vitamin K deficiency, when present, and the judicious use of fresh-frozen plasma (FFP).
|Test||Mild Hepatocellular Injury||Severe Hepatocellular Injury||Cirrhosis||Vitamin K Deficiency|
|Factor VII||↓ N||↓||↓||↓|
|Factors II, IX, and X||N||↓||↓||↓|
ATIII, antithrombin III; FSPs, fibrin split products; N, normal; PT, prothrombin time; PTT, partial thromboplastin time; TT, thrombin time; ↑, increased; ↓, decreased
Factor XI deficiency is inherited as an autosomal trait. The incidence in the general population is estimated to be 1 in 1 million, but about 10% of Ashkenazi Jews are heterozygous. The bleeding tendency does not correlate with factor levels, and bleeding is worse from areas with high intrinsic fibrinolytic activity, such as the oral cavity or genitourinary tract. The propensity to bleed also seems to be increased in patients with a nonsense mutation compared with patients with a missense mutation. Treatment involves the use of FFP to achieve a factor XI level of 30% to 45% (despite the lack of clear consensus on the target factor XI activity). Factor XI concentrates are available in Europe and have the advantage of undergoing viral inactivation and having a smaller volume. However, they have been associated with DIC and increased thrombogenicity.
Factor X deficiency is a rare (1/1,000,000) autosomal recessive deficiency characterized by asymptomatic heterozygotes and by homozygotes with bleeding symptoms that correlate with factor activity. It can be acquired in association with amyloidosis, acute respiratory infections, and leukemias. The most common bleeding complications are hematomas, hemarthrosis, epistaxis, and menorrhagia. Treatment involves initial replacement with 10 to 15 mL/kg of prothrombin complex concentrate (PCC), which contain a variable concentration of factor X, to a target of 15% to 20% factor X activity. This level is sufficient to prevent bleeding problems, whereas correcting to higher concentrations, over 50%, may lead to thromboses. Maintenance can be with 3-6 mL/kg every 24 hours.
Factor VII deficiency is a rare (1/500,000) autosomal recessive deficiency that exhibits little correlation between bleeding risk and factor activity. In general, less than 1% activity produces severe bleeding similar to that seen in the hemophilias, and more than 5% activity produces mild bleeding that is often localized to mucous membranes. Treatment involves the use of recombinant factor VIIa at a dose of 22 to 26 µg/kg (in contrast to a dose of 90 µg/kg in patients with hemophilia and inhibitors) to normalize the prothrombin time.
Factor V deficiency is a rare (1/1,000,000) autosomal recessive deficiency in which patients can manifest platelet-type bleeding (easy bruising, epistaxis, and oral bleeding). Patient can have an increased risk of thrombosis. Treatment involves replacement with FFP—an initial dose of 20 mL/kg followed by 5 mL/kg every 12 hours, with monitoring of factor V levels and bleeding—for a goal of 25% factor activity. Platelet transfusion may be required in severe bleeding, because platelets account for 20% of the total pool of factor V.
Factor II deficiency (prothrombin deficiency) is a rare (1/2,000,000) autosomal recessive disorder associated with mucosal and deep tissue bleeding. Treatment involves the use of PCCs, which contain varying concentrations of prothrombin. Factor survival analyses, measuring the effect of varying amounts of PCCs, are required to ensure proper dosing.
Congenital afibrinogenemia is a rare (1/1,000,000) autosomal recessive bleeding disorder characterized by the absence of fibrinogen. A high rate of consanguinity is noted, and carriers often have decreased fibrinogen levels. Early symptoms include umbilical stump bleeding; bleeding later in life can be life-threatening and involve any organ system. Patients with hypofibrinogenemia have mild bleeding only. Low levels of fibrinogen are more commonly seen in clinical practice in liver disease or DIC, or after the use of thrombolytics. Treatment involves replacement with cryoprecipitate – 250 mg fibrinogen per 5- to 7-kg loading dose followed by a daily infusion of 250 mg/15 kg – to a target fibrinogen level of of 80 mg/L.
Factor XIII deficiency is a rare (1/1,000,000) autosomal recessive disorder. Acquired factor XIII deficiency has been noted in patients with Henoch-Schönlein purpura, erosive gastritis, and leukemia. Bleeding occurs early in life, with umbilical stump bleeding. Later in life, bleeding occurs in skin, muscles, and the oral cavity, and it is often delayed after a hemostatic challenge. Intracerebral bleeding occurs in 30% of patients and is a major cause of mortality. Hemarthroses are rare, and female patients have recurrent abortions if they do not receive replacement therapy. Diagnosis is by noting clot solubility in both 5 M urea and 1% monochloroacetic acid, and testing factor XIII activity. Treatment involves using factor XIII concentrates. Some experts have recommended primary prophylaxis with factor XIII concentrates, 1000 units every 6 weeks, or every 3 weeks in pregnant patients. FFP or cryoprecipitate can be used when these concentrates are not available; FFP contains varying concentrations of factor XIII.
Coagulation factor inhibitors are antibodies that neutralize a specific clotting factor's function. They are called alloantibodies when they occur in patients with inherited factor deficiency, and they are called autoantibodies when they arise in patients without an inherited factor deficiency. The most commonly inhibited factor in clinical practice is factor VIII. The management of inhibitors of other clotting factors follows the same general guidelines and is not discussed here.
Autoantibodies to factor VIII are characteristically oligoclonal non–complement-fixing immunoglobulin (Ig) G. Patients with lymphoproliferative disorders or multiple myeloma might have IgM or IgA antibodies. The incidence of factor VIII inhibitors is 0.2 to 1 in 1,000,000 person-years, with a higher incidence in older age groups. There is an equal sex distribution.
Associated conditions include connective tissue disorders (e.g., rheumatoid arthritis, systemic lupus erythematosus, myasthenia gravis, temporal arteritis, and pemphigus), drugs (e.g., penicillins, sulfa, interferon alfa), malignancy (e.g., lymphoproliferative disorders, graft-versus-host disease, prostate, renal, lung, or colon cancer), pregnancy (usually within 3 months of an uncomplicated first pregnancy and delivery), and idiopathic causes, especially in older adults.
Soft tissue bleeding, gross hematuria, and postsurgical hemorrhage can occur; however, fatal bleeds take place in 15% of patients. Hemarthroses are rare. Laboratory tests are notable for a prolonged aPTT, which is not corrected by a 1:1 mixing study with normal plasma. Quantification of inhibitor titers is done with a Bethesda inhibitor assay; 1 Bethesda unit (BU) is the amount of antibody in the patient's plasma that permits detection of 50% residual factor activity when mixed with normal plasma.
The treatment of factor VIII inhibitor is often directed at the cause, when known. For drug-induced inhibitors, discontinuing the drug responsible results in recovery within several months; most postpartum inhibitors resolve within 2 to 3 months after delivery.
For symptomatic patients, the treatment is aimed at managing the bleed and reducing the antibody titer. The latter involves immunosuppression with steroids, cyclophosphamide or azathioprine, biologic response modifiers (desmopressin), intravenous immunoglobulin (IVIg), or plasmapheresis. Prednisone 1 mg/kg/day for 3 to 6 weeks is the treatment of choice and results in about a 30% response rate. For patients who do not respond to steroids, cyclophosphamide 2 mg/kg/day for 6 weeks results in an added 30% response rate. Azathioprine has also been used as an immune suppressant. IVIg 0.4 g/kg/day for 5 days results in a 25% to 30% response rate. Individual reports and small case series show remissions, some durable, with rituximab, cyclosporine or mycophenylate, often in combination with each other, steroids or other immunosuppressive agents.
Plasmapheresis can lower high-titer antibodies. It can be combined with a staphylococcal protein A column, which attaches to the Fc portion of the antibody, and results in a 50% to 90% decrease in circulating antibodies. Disadvantages include cost, difficulty of performing it in unstable patients, the need for central venous access, and the potential for circulatory collapse with the use of staphylococcal protein A.
The management of the actively bleeding patient includes replacing factor to overwhelm the antibody for low-titer antibody (5 BU/mL) or using porcine factor VIII, activated thrombin complexes, or recombinant factor VIIa for patients with high-titer antibodies, because these patients would not respond to factor replacement in the form of factor VIII concentrates, FFP, or cryoprecipitate.