Hemorrhagic and Thrombotic Disorders

Published on 07/03/2015 by admin

Filed under Critical Care Medicine

Last modified 22/04/2025

Print this page

This article have been viewed 3333 times

Hemorrhagic and Thrombotic Disorders

Neil A. Lachant

Chapter Outline

Approach to a Critically Ill Patient with Hemorrhage or Thrombosis

In the patient with acute hemorrhage or thrombosis, the history taker must be a detective interviewing the patient, family members, and outside health care providers. Useful information includes a personal or family history of a known bleeding disorder; bleeding after previous trauma, surgery, or dental procedures; a personal history of recent bleeding, bruising, or menorrhagia; and a history of liver or kidney disease, malabsorption, or recent chemotherapy. Perhaps the most challenging is obtaining a medication history, particularly as it may relate to covert heparin exposure. The use of anticoagulants, antiplatelet agents, aspirin, and nonsteroidal anti-inflammatory drugs (NSAIDs) should be determined. Recently started or intermittently used medications (e.g., quinine) that may cause thrombocytopenia should be elicited. In a hypercoagulable patient, a personal or family history of a known thrombophilic defect, thrombosis, recurrent miscarriage, or stillbirth may be useful.

The physical examination can be used to assess active problems, although physical signs of hemorrhage or thrombosis can be subtle. As examples, petechiae or purpura may develop only in dependent areas in a bedridden patient. The digits and skin should be carefully examined for signs of ischemia or necrosis.

Laboratory Tests of Coagulation

Although not reflective of the current model of coagulation, the integrity of coagulation is routinely tested by a limited set of in vitro assays (Fig. 78.1). The prothrombin time (PT) reflects the cascade of reactions traditionally called the extrinsic pathway, whereas the activated partial thromboplastin time (aPTT) reflects the intrinsic pathway. They intersect in the common pathway. The thrombin time (TT) measures the rate of conversion of fibrinogen to insoluble fibrin polymer after thrombin is added to plasma. A prolonged TT may be due to an inhibitor of thrombin (e.g., heparin, direct thrombin inhibitor [DTI]), hypofibrinogenemia or dysfibrinogenemia, fibrin degradation products (FDPs), and rarely paraproteins. The only specific coagulation factor that is routinely measured is fibrinogen. These four tests can usually localize abnormalities in the coagulation cascade.

Figure 78.1 Coagulation factors evaluated in routine coagulation assays. aPTT, activated partial thromboplastin time; PT, prothrombin time; TT, thrombin time.

Once a coagulopathy has been identified, the next step is to determine whether it is due to a factor deficiency or a circulating inhibitor. In the inhibitor or mixing study, the patient’s plasma is mixed with an equal volume of normal pooled plasma. Normalization of the PT or aPTT in such a “mixing study” reflects a deficiency of one or more coagulation factors. It also implies that administration of fresh frozen plasma (FFP) should correct the coagulopathy. A further diagnosis requires specific factor assays. If there is partial or no correction in the mixing study, an inhibitor is suspected—most often contaminating heparin or a lupus anticoagulant.

The template bleeding time (BT) assesses in vivo platelet function in patients with a normal platelet count. It is used as a screen for disorders of platelet function. The concept of using the BT to predict surgical bleeding is archaic.

Disorders of Platelets

Thrombocytopenia

Mechanisms and General Management

Thrombocytopenia is common in the intensive care unit (ICU). It has been estimated that 23% to 41% of patients in the ICU have a platelet count less than 100,000/µL, and 10% to 17% have a count less than 5000/µL.¹ Common causes of thrombocytopenia in the ICU are shown in Box 78.1. In complex, acutely ill patients, many of these mechanisms may operate simultaneously. Severe sepsis is the most common cause of thrombocytopenia in the ICU.²

Frequently, thrombocytopenia must be managed without a specific diagnosis. Medications should be reviewed for potential offending agents.3,4 Inhibitors of platelet function should be avoided with platelet counts below 50,000/µL. If there is bleeding or if invasive procedures are anticipated, platelet transfusions should be given, unless contraindicated, to elevate the platelet count above 50,000/µL. In life-threatening situations, the goal should be a platelet count higher than 100,000/µL. In nonbleeding patients, maintenance of a platelet count above 10,000/µL (20,000/µL with fever or infection) with prophylactic transfusions is usually adequate.⁵

Pseudothrombocytopenia

Pseudothrombocytopenia is a laboratory artifact.⁶ The platelet count is factitiously lowered because of the presence of naturally occurring antibodies that cause platelet agglutination in the presence of ethylenediaminetetraacetic acid (EDTA) at room temperature. The diagnosis is suspected by finding platelet clumps on the peripheral blood smear. Repeating the platelet count with a different anticoagulant such as citrate will generally produce a normal platelet count.

Drug-Induced Thrombocytopenia

A limited number of drugs have evidence-based data to support a causal role in the development of thrombocytopenia.3,4,7,8 Medications commonly used in the ICU can cause thrombocytopenia (Box 78.2). Drug-induced thrombocytopenia most commonly occurs 7 to 21 days after exposure to the offending agent. Clinical manifestations can range from an asymptomatic decrease in platelets to life-threatening bleeding. The diagnosis is established by (1) finding a temporal relationship between starting the drug and the fall in the platelet count, (2) having no alternative diagnosis, and (3) having the platelet count recover after removal of the putative offending agent. Unfortunately, this is usually difficult to establish in the typical ICU patient. Treatment is based on removing the putative offending agent and initiating a drug of another class if possible. Though often used, steroids in general have not been shown to hasten the rate of platelet recovery. In severe thrombocytopenia with bleeding such as seen with quinine, intravenous immunoglobulin (IVIG) (1 g/kg/day for 2 days) and platelet transfusion are beneficial.

Glycoprotein IIb/IIIa Inhibitors

All platelet glycoprotein (GP) IIb/IIIa inhibitors have been associated with severe thrombocytopenia that can occur within hours of exposure (up to 2 weeks with abciximab).⁷ Heparin-induced thrombocytopenia (HIT) is the main differential diagnosis. Bleeding is very uncommon with HIT because of the strong prothrombotic state. Conversely, with GP IIb/IIIa inhibitor-associated thrombocytopenia, bleeding or hematoma formation may occur, especially at the site of the sheath. A platelet count less than 20,000/µL and clinical bleeding are indications for platelet transfusion. The use of IVIG and corticosteroids is not evidence based.

Immune Thrombocytopenic Purpura

Immune thrombocytopenic purpura results from the destruction of IgG-coated platelets in the reticuloendothelial system, primarily the spleen. There is no compensatory increase in thrombopoiesis. Anemia, if present, may be autoimmune (Evans syndrome) or due to bleeding and iron deficiency. Immune thrombocytopenia may be primary (idiopathic [ITP]) or secondary. The differential diagnosis includes pseudothrombocytopenia; immune thrombocytopenia secondary to systemic lupus, human immunodeficiency virus (HIV) infection, or hepatitis C; and drug-induced thrombocytopenia.

Guidelines for the diagnosis and management of ITP have been developed.9,10 Indications for therapy in ITP are platelet count less than 20,000 to 30,000/µL or clinical bleeding. Corticosteroids (prednisone, 1 mg/kg/day, or pulse dexamethasone, 40 mg/day for 4 days monthly) are the usual initial therapy for ITP.^9,10 Individuals with ITP in the ICU usually have severe or life-threatening bleeding. Several modalities of therapy should be used in concert to raise the platelet count in urgent situations (methylprednisolone, 1 g/day for 3 days, and IVIG, 1 g/kg/day for 2 days). Although platelets may be destroyed quickly, platelet transfusions should still be used as initial therapy. The response to platelet transfusion may improve after IVIG is given. Anti-Rh(D) IgG (WinRho), 50 to 75 µg/kg, has also been used.¹¹ Because the dose of anti-Rh(D) must be reduced in face of anemia, its use may be problematic in a patient with severe bleeding. Anti-Rh(D) IgG is ineffective in Rh-negative patients and after splenectomy. ε-Aminocaproic acid (4-5 g IV followed by either 2-4 g IV every 4 hours or 0.5-1.0 g/hour continuous IV infusion [maximum 24 g/24 hours]) may be useful for mucosal bleeding and severe menorrhagia.

Post-transfusion Purpura

Post-transfusion purpura (PTP) is a rare condition characterized by acute, severe immune-mediated thrombocytopenia. PTP occurs in human platelet antigen-1 (HPA-1) negative individuals who receive HPA-1 positive platelets (most commonly as a contaminant in packed red blood cells [RBCs]). These individuals have previously been sensitized to HPA-1 through transfusion or pregnancy. Usually 7 to 10 days after reexposure to HPA-1, an anamnestic response occurs, resulting in a precipitous fall in the platelet count. Petechiae and bleeding are common. IVIG and plasma exchange are effective.¹²

Acquired Platelet Dysfunction

Medication-Induced Abnormalities

Aspirin irreversibly acetylates cyclooxygenase (COX), inhibiting platelet function for the life of the platelet (7 to 10 days). The aspirin effect can be overcome with platelet transfusion or the infusion of desmopressin (DDAVP). The effect of NSAIDs is reversible and disappears as the drug is cleared, usually within 24 to 48 hours for ibuprofen. The risk of bleeding from NSAIDs is lowest with ibuprofen and greatest with ketorolac. The thienopyridine clopidogrel tightly binds the platelet ADP P2Y12 receptor. Clopidogrel should be withheld for 5 to 7 days before elective surgery or invasive procedures.¹³ In an emergency, platelet transfusion can be tried, but platelet function may not be fully restored because of circulating active clopidogrel metabolites, which have a half-life of 8 hours and bind to the transfused platelets. Prasugrel is a new thienopyridine whose active metabolites have a 4-hour half-life. COX-2 inhibitors don’t affect platelet function.

Renal Failure

The hemorrhagic diathesis of renal failure is the result of metabolic derangements related to uremic toxins.¹⁴ Bleeding may worsen when the hematocrit falls below 30% due to a rheologic phenomenon in which rapidly flowing RBCs gravitate to the center of the streaming blood and force the platelets toward the vessel wall. Uremic bleeding is uncommon in the modern era of dialysis.¹⁵ For clinical bleeding, intravenous DDAVP (0.3 µg/kg given over 30 minutes) is often therapeutic.¹⁴ If the hematocrit is less than 30%, the patient should be transfused with packed RBCs.^14,15 An alternative to DDAVP is cryoprecipitate (10 bags every 12 to 24 hours).¹⁴ Intravenous conjugated estrogen, 0.6 mg/kg/day for 5 days, is also effective.¹⁴ An unexpected “coagulopathy” (prolonged aPTT and TT) may develop as a result of delayed clearance of heparin after dialysis. If available, an anti–factor Xa assay will show the presence of heparin.

Thrombocytosis

Although not as common as thrombocytopenia, reactive thrombocytosis (splenectomy, surgery, or bleeding) may be seen in the ICU. Reactive thrombocytosis does not have an increased risk of thrombosis. Antiplatelet therapy is not needed even for platelet counts higher than 1,000,000/µL.

Complex Thrombohemorrhagic Disorders

Heparin-Induced Thrombocytopenia

Although commonly looked for, HIT is an uncommon cause of thrombocytopenia in the ICU.2,16 HIT is a paradoxical condition in which modest thrombocytopenia may be associated with devastating heparin-induced thrombocytopenia thrombosis (HITT). Thus, the entire health care team needs to be vigilant for the development of HIT in any patient receiving heparin. HIT has been associated with all types of heparin (unfractionated [UFH] and low-molecular-weight [LMWH]), at any dose, and by any route, including flushes and heparin-coated catheters.^1,17 The incidence of HIT has been estimated to be less than 1% of patients in the ICU.^2,16,18 HIT occurs in three time intervals. Classic HIT occurs 5 to 15 days after the initiation of heparin. Rapid-onset HIT develops hours to 1 to 2 days after heparin is started in individuals who have preformed circulating antibodies from a previous exposure to heparin, usually in the last 2 months. Classic and rapid-onset HIT may be manifested as thrombocytopenia with or without thrombosis [HIT(T)]. Delayed-onset HIT occurs an average of 12 days after the discontinuation of heparin and is manifested as isolated thrombosis. The thrombocytopenia of HIT is usually modest with an average platelet count of 50,000 to 60,000/µL.¹⁹ Severe thrombocytopenia (<10,000/µL) should suggest an alternative diagnosis. HIT may be associated with a normal platelet count if there is a baseline thrombocytosis (e.g., postoperatively). HIT may be manifested as isolated thrombocytopenia or thrombocytopenia with potentially devastating thrombosis. Venous complications include deep venous thrombosis (DVT), pulmonary embolism (PE), cerebral sinus thrombosis, infarctive adrenal hemorrhage, and skin necrosis at the heparin injection site. Arterial complications include iliofemoral artery thrombosis, digital ischemia, myocardial infarction, stroke, and mesenteric artery thrombosis.

HIT is an immune-mediated process in which the heparin–platelet factor 4 complex becomes immunogenic. Antibodies are formed that activate platelets, causing the release of thrombogenic microparticles.16,19 The antibodies can also activate monocytes and endothelial cells.

The diagnosis of HIT is clinicopathologic. HIT should be thought of in appropriate clinical situations (Box 78.3). The “4T” score has been used to predict the likelihood (pretest probability) of HIT (Table 78.1).^16,20 The HEP score, devised by an expert consensus panel, has recently been validated.²¹ Although the decision to initiate therapy should be based on the clinical likelihood of HIT, laboratory testing should be used for the retrospective confirmation of the pretest likelihood of HIT. The complexities of HIT antibody testing are reviewed elsewhere.^16,22 In general, a strongly positive HIT enzyme-linked immunosorbent assay (ELISA) and strongly positive serotonin release assay (SRA) with a high pretest probability confirm the diagnosis of HIT, whereas a negative ELISA and SRA make the diagnosis of HIT unlikely with a low pretest probability. In other situations, clinical judgment prevails in establishing or excluding a diagnosis of HIT, especially because ELISA and SRA testing from commercial labs has not been validated in clinical studies.^1,16,22

Table 78.1

Determining Pretest Probability of Heparin-Induced Thrombocytopenia (HIT): The “4 Ts”

Pretest probability score: 0-3 points = low (<5%), 4-5 points = intermediate (physician judgment), 6-8 points = high (>80%).

^*First day of heparin exposure = day 0.

Adapted from Lo GK, Juhl D, Warkentin TE, et al: Evaluation of pretest clinical score (4 T’s) for the diagnosis of heparin-induced thrombocytopenia in two clinical settings. J Thromb Haemost 2006:4:759.

Once the diagnosis of HIT is thought to be likely, all heparin should be discontinued. Heparin-coated pulmonary artery catheters should be replaced with noncoated catheters. Catheters for dialysis or apheresis should be locked with 4% citrate or tissue plasminogen activator. The patient’s chart and bedside should be labeled heparin allergy. Platelet transfusions should be avoided except for life-threatening bleeding, given the anecdotal observations of acute thrombosis occurring after platelet transfusion. A patient with life- or limb-threatening arterial thrombosis should be evaluated for surgical intervention. Anticoagulation with a DTI (Table 78.2) should be initiated in all patients unless contraindicated. LMWH is contraindicated in HIT caused by UFH. Conversion to warfarin can be considered after a minimum of 5 days of DTI therapy if the platelet count has returned to normal (suggesting that the process has cooled off and the patient is no longer hyperprothrombotic) and no future invasive procedures are planned. Simply discontinuing heparin therapy and not starting a DTI is inappropriate because occult thrombosis may have already developed.²³ Starting or continuing warfarin monotherapy is also contraindicated because of the risk of venous limb gangrene.²⁴ If warfarin has been given at the time HIT is diagnosed, vitamin K should be given to replenish proteins C and S. If invasive procedures are needed (e.g., tracheostomy, pacemaker), it is best to delay them, if medically safe, until the platelet count is normal to minimize the risk of developing thrombosis during the time that DTI therapy is withheld. Inferior vena cava (IVC) filters should be avoided because of the risk of vena cava thrombosis. Patients with active HIT may need percutaneous coronary intervention (PCI). Argatroban has been approved by the Food and Drug Administration (FDA) for use during PCI in patients with HIT.²⁵ Though not FDA approved, bivalirudin has been used safely in this situation.²⁶ In a patient who has active HIT or persistent HIT antibodies and who needs open heart surgery, medical management is recommended until the antibody becomes negative. If urgent surgery is needed, bivalirudin is most commonly used.²⁷ For the patient with a past history of HIT who needs open heart surgery and is currently HIT antibody negative, heparin can be used during bypass and a DTI started as soon as it is surgically safe postoperatively.²⁸

Table 78.2

Parenteral Anticoagulant Doses

aPTT, activated partial thromboplastin time; Ccr, creatinine clearance; CHF, congestive heart failure; CI, continuous infusion; ICU, intensive care unit; qd, once daily; SC, subcutaneously.

Data from Selleng K, Warkentin TE, Greinacher A. Heparin-induced thrombocytopenia in intensive care patients. Crit Care Med 2007;35:1165-1176.

Thrombotic Thrombocytopenic Purpura

Thrombotic thrombocytopenic purpura (TTP) is a relatively rare disorder whose hallmarks are thrombocytopenia, microangiopathic hemolytic anemia (MAHA), and neurologic dysfunction. The diagnosis of TTP should be considered in any patient with unexplained thrombocytopenia and MAHA.²⁹ TTP is the clinical manifestation of a heterogeneous group of underlying disorders driven by different pathophysiologic processes. Although most cases of TTP are idiopathic, TTP may be associated with exposure to drugs (e.g., ticlopidine, clopidogrel, quinine, mitomycin C, gemcitabine, cyclosporine), pregnancy, HIV infection, and bone marrow transplantation.³⁰ The morphologic hallmark of TTP is hyaline thrombi in precapillary arterioles composed primarily of platelets. Classic TTP is due to an immune-mediated deficiency of the metalloproteinase ADAMTS13, which cleaves ultralarge multimers of von Willebrand factor (vWF) into smaller forms that are less platelet reactive.³¹ Deficiency of this enzyme allows ultralarge forms of vWF that are normally sequestered in the endothelium to bind platelets and form microthrombi in the circulation. Conversely, ADAMTS13 activity is normal in TTP associated with bone marrow transplantation and the hemolytic-uremic syndrome.^30,31 Organ dysfunction is due to microvascular thrombosis. Commonly involved organs are the brain, kidneys, heart, and pancreas. It is critical to establish the diagnosis quickly because TTP can be rapidly fatal if not properly treated.

The core laboratory features of TTP are those of MAHA (more than two schistocytes per oil immersion field on a peripheral smear, increased lactate dehydrogenase, and very low haptoglobin). The differential diagnosis includes disseminated intravascular coagulation (DIC), severe vasculitis, eclampsia, HELLP (hemolysis, elevated liver enzymes, and low platelet count) syndrome, malignant hypertension, sepsis, and malignancy.

Once the diagnosis of TTP is thought to be likely, therapy should be initiated quickly because of the proclivity of the disease to progress rapidly. All patients should initially be treated in the ICU. Platelet transfusions should be avoided except for life-threatening bleeding because of anecdotal reports of acute decompensation after platelet transfusions. A large-bore catheter will need to be placed, even in the face of severe thrombocytopenia, because the mainstay and only evidence-based component of therapy for TTP is plasma exchange.32,32a Most commonly, 1.5 plasma volumes are exchanged daily with FFP. A clue that TTP is the correct diagnosis is the color of the plasma removed from the first plasmapheresis. A red or brown color suggests free hemoglobin from intravascular hemolysis. If plasma exchange cannot be initiated in a timely manner, infusion of 30 mL/kg of FFP daily can be a temporizing maneuver.³² Corticosteroids (e.g., prednisone, 1 mg/kg/day) are commonly used as well.^32a The use of antiplatelet agents, which may increase the risk of bleeding, and vincristine is controversial.³³ In patients who do not respond to initial therapy, rituximab is an attractive second-line therapy.^30,34

Hemolytic-Uremic Syndrome

Hemolytic-uremic syndrome (HUS), the triad of thrombocytopenia, MAHA, and acute kidney injury, is most commonly due to infection with Shiga toxin–producing Escherichia coli in children.³⁵ Adult atypical HUS (aHUS) is not Shiga toxin associated. It may be idiopathic or associated with calcineurin inhibitors (cyclosporin, tacrolimus), pregnancy, and HIV. aHUS is associated with low C3 and is a complement deposition disease. Mutations in complement factor H (CFH) are found in a minority of cases. Although not as efficacious as when used for TTP, plasma exchange is used for aHUS. The anti-C5 antibody eculizumab (Soliris) has been FDA approved for the treatment of aHUS.³⁶ Plasma exchange should be initiated while waiting to obtain eculizumab. Supplemental eculizumab dosing is required after each plasma exchange.³⁷ Patients need to receive quadravalent meningococcal vaccine before starting eculizumab.

Disseminated Intravascular Coagulation

DIC is always a manifestation of a severe underlying pathologic process (Box 78.4). The final common pathway is the generation of thrombin, which produces microthrombi that can lead to organ dysfunction.³⁸ In the process, natural anticoagulants are consumed and fibrinolysis is activated. Circulating FDPs, along with the consumption of platelets and coagulation factors, lead to a bleeding diathesis. The balance between these competing processes results in the clinical manifestations in any particular individual.

The clinical manifestations of DIC are protean. Microvascular thromboses produce ischemia in many tissues, including the brain (delirium, coma), skin (digital gangrene, purpura fulminans), kidney, lungs (acute respiratory distress syndrome), gastrointestinal tract (mucosal ulceration), and blood (MAHA). The bleeding associated with DIC is due to the combination of depletion of coagulation factors, thrombocytopenia, inhibitory effects of FDPs on coagulation and platelet function, and tissue necrosis with ulceration. Common manifestations are intracerebral hemorrhage, global oozing at lines and venipuncture sites, hematuria, epistaxis, and gingival and gastrointestinal bleeding.

There are no strict, evidence-based laboratory criteria for the diagnosis of DIC.³⁸ The diagnosis is based on identifying an underlying predisposition, the clinical features, and laboratory testing. The most common laboratory features of DIC are thrombocytopenia, elevated FDP or D-dimers, and decreased fibrinogen. The PT and aPTT may or may not be prolonged, depending on the sensitivity of the assay, degree of baseline elevation of coagulation factors as acute-phase reactants, the rate of consumption of coagulation factors, and the amount of FDPs released. An absolute decrease in fibrinogen is very suggestive of DIC. However, fibrinogen activity in the normal range may still be consistent with DIC because fibrinogen is an acute-phase reactant and its level may be inappropriately low for the underlying physiologic state.³⁹ Serial measurement of fibrinogen is often useful. Schistocytes are seen in half the cases. Although the natural anticoagulants antithrombin (AT) (previously called AT-III) and proteins C and S are decreased, and thrombin-antithrombin complexes are increased, their levels usually cannot be measured in real time and are of limited clinical value.

The International Society of Thrombosis and Haemostasis (ISTH) has developed a scoring system for the identification of overt DIC using four commonly available tests (platelets, prothrombin time, FDPs such as D-dimer or soluble fibrin monomer, and fibrinogen) (Table 78.3), which has been prospectively validated.^40,41 It is important to remember that all the laboratory abnormalities seen in DIC can be caused by other disease processes. Vitamin K deficiency,⁴² a heparin-contaminated blood sample, or a lupus anticoagulant can cause a coagulopathy, whereas elevations in FDPs can be seen with metastatic cancer or can be found after surgery or trauma. Other microangiopathies cause thrombocytopenia and RBC fragmentation. Liver failure can produce a constellation of laboratory abnormalities indistinguishable from those observed in DIC.⁴³

Table 78.3

Original ISTH Scoring System for Overt Disseminated Intravascular Coagulation (DIC) and the Modified ISTH Scoring System for Overt DIC in Severe Sepsis

Test	Score
Platelet count (per µL)
≥100,000	0
50,000-99,999	1
<50,000	2
D-dimer (µg/mL)
≤0.39 (upper limit of normal)	0
0.40-4.0	2
>4.0	3
Prothrombin time prolongation (s)
≤3	0
>3 but <6	1
≥6	2
Fibrinogen (mg/dL)
>100	0
<100	1

Original ISTH score: (platelets + D-dimer + PT + fibrinogen) ≥ 5 signifies overt DIC.

Modified ISTH score: (platelets + D-dimer + PT) ≥ 5 signifies overt DIC in severe sepsis.

ISTH, International Society of Thrombosis and Haemostasis; PT, prothrombin time.

Adapted from Taylor FB, Toh CH, Hoots WK, et al: Towards definition, clinical and laboratory criteria, and a scoring system for disseminated intravascular coagulation. Thromb Haemost 2001;86:1327; and Dhainaut J-F, Yan SB, Joyce DE, et al: Treatment effects of drotrecogin alfa (activated) in patients with severe sepsis with or without overt disseminated intravascular coagulation. J Thromb Haemost 2003;2:1924.

The most important principle in the management of DIC is to treat the underlying cause. If the driving process cannot be controlled, the DIC will progress unabated. Hypothermia and acidosis should be corrected because they interfere with the function of coagulation factors.

If the diagnosis is suspected only because of laboratory abnormalities, management should be conservative. Alternative, potentially correctable processes should be kept in mind, such as vitamin K deficiency and drug-induced thrombocytopenia.

If clinically significant bleeding becomes apparent, a patient with overt DIC should receive blood component therapy (cryoprecipitate if fibrinogen is less than 100 mg/dL, FFP, and platelets). A realistic goal is to keep the fibrinogen level 100 to 150 mg/dL, the aPTT close to normal, and the platelet count greater than 50,000/µL. Heparin is not routinely used for the treatment of DIC. Heparin may be helpful for ischemic manifestations (e.g., digital ischemia, purpura fulminans). A continuous infusion at a rate of 7 U/kg/hour is a reasonable starting dose.

Other treatment modalities have been tried in DIC. The use of AT concentrates showed no significant decrease in mortality rate.⁴⁴ Activated protein C (drotrecogin alfa) used for the treatment of sepsis has been withdrawn from the U.S. market. Because of the intense fibrinolytic activity in DIC, antifibrinolytic agents (ε-aminocaproic acid and tranexamic acid) have been tried in an attempt to reduce bleeding. In general, failure rates have been quite high, and serious thromboses have occurred because the lysis of diffuse microthrombi was suppressed.

Disorders of Hemostasis

It has been estimated that 16% of ICU patients have bleeding caused by a coagulation defect and another 66% have abnormal coagulation test results.

Artifacts

A common cause of an unexplained prolongation of the aPTT is unsuspected heparin in the line from which the blood was drawn. A diagnostic clue is a prolonged TT with a normal fibrinogen. A heparin level (anti–factor Xa assay) will confirm the diagnosis. Although venous access is often an issue in the ICU, blood for coagulation studies should be drawn by venipuncture if possible.

Coagulation Factor Abnormalities

Acquired Deficiencies of Procoagulants

Vitamin K Deficiency

Vitamin K is essential for the post-translational modification (γ-carboxylation) of factors II, VII, IX, and X and proteins C and S. The daily requirement of vitamin K is 50 µg. Normal daily intake of vitamin K is 200 µg from the diet and 200 µg from intestinal flora. Thus, vitamin K deficiency occurs commonly in the ICU in the patient who is not being fed and who is receiving antibiotics that alter the gut flora. Vitamin K stores last approximately 1 week.

Because factor VII has the shortest half-life (approximately 4 hours), early vitamin K deficiency is manifested by an isolated prolongation of the PT. As factors II, IX, and X become depleted, the aPTT will become prolonged. Vitamin K deficiency is common in critically ill patients and may result in serious bleeding.⁴² Treatment of vitamin K deficiency depends on the clinical situation. In a nonbleeding patient, vitamin K should be given enterally if the gut is working. Subcutaneous vitamin K should be avoided if the gut is working because of its erratic absorption, especially in face of impaired tissue perfusion and because anaphylactic reactions have been reported with subcutaneous vitamin K.⁴⁵ In a bleeding patient, FFP will correct the coagulopathy most quickly. In critically ill patients with an urgent need for vitamin K replacement, phytonadione may be administered intravenously if the benefits are thought to outweigh the risk of anaphylaxis. Vitamin K should be given no faster than 1 mg/minute with cardiorespiratory monitoring because anaphylactic reactions can occur even at this slow infusion rate.⁴⁵ One milligram of vitamin K twice a week can prevent vitamin K deficiency.

Liver Disease

The perturbation of hemostasis in liver disease is complex. All hemostatic factors except for vWF and plasminogen activator inhibitor-1 are synthesized in the liver. Loss of the ability to synthesize factor VII is disproportionately greater in patients with liver disease than in those receiving warfarin. Thus, the bleeding risk of a prolonged international normalized ratio (INR) from liver disease is not as great as the same INR with warfarin therapy.⁴⁶ As outlined later, changes in factor VIII/vWF, and decreased synthesis of natural anticoagulants and fibrinolytic proteins, may help keep the patient with liver disease in hemostatic balance,⁴³ making the autoanticoagulation of liver disease promulgated among house staff a myth.

Although hepatic dysfunction is most commonly associated with bleeding, the risk of thrombosis has been underappreciated.⁴³ Synthesis of hemostatic proteins becomes impaired when the albumin falls below 2.5 g/dL. Similar to vitamin K deficiency, the earliest abnormality is a reduction in factor VII. In more severe liver disease, the other coagulation factors become deficient. Factor V and fibrinogen do not usually decrease significantly until the liver failure is severe. In contrast, the activities of factor VIII and vWF are normal or elevated. AT, proteins C and S, and plasminogen are decreased. Hepatic clearance of activated procoagulants and FDP is reduced. Thrombocytopenia may develop due to decreased thrombopoietin production or splenic sequestration. Advanced liver disease may also be complicated by vitamin K deficiency from malnutrition and decreased vitamin K absorption due to cholestasis.

A variety of approaches have been used to treat the coagulopathy of liver disease. If the patient is not bleeding, a trial of vitamin K can be given for 3 days to see if the coagulopathy corrects. Because all patients with liver disease have some degree of biliary obstruction and fat malabsorption, the subcutaneous route is preferred over oral administration. Although the risk of anaphylaxis with intravenous vitamin K is low, it is not generally recommended empirically in a patient with liver disease because of the low likelihood that vitamin K will completely correct the coagulopathy.⁴⁵ If bleeding is a problem, FFP should be administered. Because the relationship between the levels of coagulation factors and the PT/PTT is nonlinear, the greater the prolongation of the PT and PTT, the greater the chance of correction with FFP. However, it is difficult to normalize the PT.⁴⁷ On average, a unit of FFP increases all coagulation factors by 6% to 7%. A typical patient with the coagulopathy of liver disease may have approximately 10% to 15% of the normal level of coagulation factors. If a factor level of 50% of normal is needed for hemostasis, 6 units of FFP would be expected to correct the bleeding diathesis. The PT and PTT may still be prolonged, however, depending upon the sensitivities of PT and PTT assays. The PT (and aPTT) should be checked immediately after the FFP is administered. Given the short half-life of factor VII, waiting until the next morning’s routine blood draw will produce a prolongation of at least the PT and give the impression that the coagulopathy is uncorrectable.

Massive Transfusion

Massive volume replacement (>1 blood volume in less than 24 hours) may not give the liver sufficient time to replace coagulation factors. In a hemodynamically stable patient, coagulation studies should be followed serially and FFP infused when a coagulopathy develops. Conversely, exsanguinating patients should be managed with balanced transfusion of RBC : FFP:platelets in a ratio of 1 : 1 : 1 (either 1 unit of single donor or 6 units or random donor platelets).⁴⁸

Acquired Circulating Inhibitors

Factor XI Deficiency

Factor XI deficiency is an autosomal recessive disorder most commonly seen in Ashkenazi Jews. The bleeding history is variable, ranging from menorrhagia to bleeding with delivery or surgery. Therapy depends on the clinical situation.⁵² For severe deficiency, FFP (10 to 20 mL/kg followed by 3 to 6 mL/kg every 6 hours) is used. The goal is factor XI activity greater than 40% to 45%. Individuals with factor XI deficiency may be seen in the ICU because factor XI deficiency is not protective against coronary artery disease. These individuals may require emergency cardiac catheterization and coronary intervention.⁵³ Clopidogrel has been given safely after coronary stenting.⁵⁴

Venous Thromboembolism

Venous thromboembolism (VTE) is common in critically ill patients. Twenty-nine percent of medical ICU patients have been found to have asymptomatic lower extremity DVT, although 5.8% and 4.2% have symptomatic DVT and PE, respectively.⁵⁵ Most patients admitted to the ICU have multiple risk factors including recent surgery, trauma, sepsis, malignancy, immobilization, stroke, advanced age, heart or respiratory failure, previous VTE, and indwelling catheters.^55–57

Deep Venous Thrombosis Prophylaxis

All patients admitted to the ICU should be evaluated for their risk for DVT. The perceived risk of thrombosis and bleeding needs to be assessed in each ICU patient.⁵⁵ In medical patients, options are LMWH and UFH.⁵⁵ In medical patients with a high bleeding risk, mechanical prophylaxis (graded compression stockings and intermittent pneumatic compression devices) is recommended.⁵⁵ Beyond the size limitations of this chapter, specific recommendations for DVT prophylaxis are available for trauma, abdominal/pelvic, thoracic, cardiac, and neurologic surgery.⁵⁸

Therapy for Venous Thromboembolism

In the noncritical care setting, VTE is routinely treated with at least 5 days of UFH, LMWH, or fondaparinux (see Table 78.2), followed by warfarin.⁵⁹ In the ICU, intravenous UFH is the safest mode of anticoagulation give its rapid reversabilty for procedures or bleeding. LMWH has the disadvantages of a longer half-life and being only partially reversed with protamine. As discussed in the section on anticoagulation, fondaparinux should not be used routinely in the ICU. ICU patients should not be transitioned to warfarin given their risk of bleeding and frequent need for invasive procedures. Warfarin is hard to regulate due to frequent drug interactions (especially with antibiotics) and erratic oral intake. Warfarin may stick to feeding tubes.⁶⁰ Warfarin is absorbed in the upper small bowel and may not be absorbed if given via J-tube.⁶¹

Inferior Vena Cava Filter

Approximately 80% of PE occur as a result of the lower extremity DVT. IVC filter placement is indicated in two situations: (1) acute DVT and a contraindication to anticoagulation (e.g., recent intracranial hemorrhage or stroke, recent neurologic or ophthalmologic surgery, active gastrointestinal hemorrhage, cerebral metastasis with a high risk of bleeding [e.g., melanoma, renal cell, seminoma]⁶²), and (2) recurrent PE despite therapeutic anticoagulation. Filters may have some benefit in face of PE causing hypotension.⁵⁹ IVC filters are not indicated for PE without DVT. IVC filters do not eliminate all risk of PE and do not decrease mortality rates.⁵⁹ There is an approximate 10% risk of thrombosis at the filter insertion site.⁵⁹

Catastrophic Antiphospholipid Syndrome

Catastrophic antiphospholipid syndrome is an uncommon disorder manifested as multiple organ dysfunction secondary to microvascular thrombi. The diagnosis should be considered in patients with an acute onset of more than three of the following: respiratory failure, stroke or other neurologic impairment, abnormal liver function, renal impairment, skin infarction, and thrombocytopenia. Laboratory testing should show evidence of antiphospholipid antibodies. Though not evidence based, empiric therapy has centered around the combined use of high-dose corticosteroids, anticoagulation, and plasma exchange or IVIG.

Anticoagulants

A variety of parenteral and oral anticoagulants acting through different mechanisms are commercially available with many more in development. Commonly used dosing schema are shown in Table 78.2.

Heparin, Heparin Derivatives, and Heparinoid

Heparin

Heparin is a glycosaminoglycan that anticoagulates blood by augmenting the activity of AT.⁶³ Complexes of heparin and AT inhibit thrombin, factor Xa, and other procoagulant proteases. The first five sugars of heparin bind to AT. The additional sugars determine the protein binding properties and pharmacokinetics of UFH and LMWH.⁶³

Unfractionated Heparin

Binding of heparin to AT varies markedly between individuals, making monitoring a requirement. The anti–factor IIa property of heparin is measured by the aPTT. Because the anti-IIa and anti-Xa properties of UFH are equal, UFH is usually monitored by the aPTT.

The goal of therapeutic heparin is to maintain the aPTT in a range that corresponds to a heparin level of 0.3 to 0.7 U/mL.⁶³ Given the marked variability in aPTT response to heparin because of differences in reagents and laboratory equipment, this range needs to be determined in each laboratory. Because some hospitals do not have a heparin protocol, an adaptation of the Raschke nomogram is shown in Table 78.4.^63,64 An aPTT ratio of 1.5 to 2.5 times the control value correlates very poorly with the anti-Xa heparin assay and should not be used. The dose of heparin is weight based. No adjustments are needed for obesity, hepatic dysfunction, or renal impairment. The hemoglobin and platelet count should be monitored daily as surveillance for occult bleeding and HIT. Overanticoagulation with heparin can usually be managed by simply interrupting the infusion. With serious bleeding, heparin can be reversed immediately with intravenous protamine sulfate.⁶³ One milligram of protamine will neutralize 100 units of heparin. If bleeding occurs during the constant infusion, the amount of heparin given during the previous 2 hours should be used for the calculation. Thus, a patient bleeding after a 5000-unit heparin bolus will need 50 mg of protamine, whereas a patient receiving a 1250 U/hour infusion will need only 25 mg of protamine. Protamine should be given slowly over 1 to 3 minutes to minimize the risk for hypotension and bradycardia. The risk for anaphylaxis is increased in those who have received protamine zinc insulin or who have had a vasectomy.

Table 78.4

Heparin Dosing Nomogram

Anti–Factor Xa (U/mL)	aPTT
Initial dose	80 U/kg bolus, then 18 U/kg/hour
<0.15	80 U/kg bolus, increase by 4 U/kg/hour
0.15-0.29	40 U/kg bolus, increase by 2 U/kg/hour
0.30-0.70	No change
0.71-0.85	Decrease infusion by 2 U/kg/hour
>0.85	Hold 1 hour, decrease infusion by 3 U/kg/hour

aPTT, activated partial thromboplastin time.

Adapted from Raschke RA, Reilly BM, Guidry JR, et al: The weight based heparin dosing nomogram compared with a “standard care” nomogram: A randomized controlled trial. Ann Intern Med 1993;119:874.

Low-Molecular-Weight Heparin

The properties of LMWH vary significantly from those of UFH.⁶³ The anti-Xa : anti-IIa ratio of LMWH is approximately 4 : 1. Thus, LMWH does not significantly affect the aPTT. There is little protein binding, allowing for predictable renal excretion. The properties of the available LMWHs (enoxaparin, dalteparin, tinzaparin) are different. Their kinetics and dosing are not interchangeable.⁶³ In a normal-sized adult (50 to 150 kg) with normal renal function, monitoring is not necessary. If monitoring is necessary (weight outside the usual range, impaired renal function, pregnancy), anti-Xa activity is measured 3.5 to 4.0 hours after an enoxaparin dose. The therapeutic range for enoxaparin (1 mg/kg every 12 hours) is 0.5 to 1.0 U/mL. LMWH is only 60% reversible by protamine.⁶³ If bleeding occurs within 8 hours of a LMWH injection, 1 mg of protamine per 100 anti-Xa units should be given. Because of the kinetics of LMWH, a second protamine dose of 0.5 mg per 100 anti-Xa units should be given if the bleeding persists.⁶³

Pentasaccharide

Fondaparinux (Arixtra) is a synthetic analog of the first five sugars of heparin. Due to its small size, it is a pure factor Xa antagonist. Fondaparinux is approved for treatment of DVT and PE and for DVT prophylaxis in orthopedic and abdominal surgery. Although not FDA approved, fondaparinux has been used off-label for the prophylaxis and treatment of HIT(T).¹⁶ As a result of its 18- to 21-hour half-life, its use in the ICU is limited because it is not rapidly cleared in a patient population that needs frequent procedures, that has a high risk of bleeding, and often has renal impairment (contraindicated with creatine clearance [Ccr] < 30 mL/minute). It does not prolong the PT, aPTT, or TT. It can be measured in the anti-Xa assay calibrated for fondaparinux. It is not reversed by protamine. Bleeding due to pentasaccharide should be supported by transfusion. rH-VIIa can be tried.

Heparinoid

Danaparoid is a mixture of sulfated glycosaminoglycans, distinct from heparin, that produces an anticoagulant effect primarily by enhancing the activity of AT. Danaparoid is primarily used in the setting of HIT. Dosing schema are available for most clinical situations.⁶⁵ If monitoring is necessary, such as during renal failure, anti-Xa activity must be determined. Though no longer available in the United States, danaparoid is available in Canada and Europe.

Direct Thrombin Inhibitors

As the name implies, DTIs directly bind to and inhibit the function of thrombin. Because of their expense, parenteral DTIs are generally reserved for the treatment of patients with HIT(T) or true heparin allergy. Although none of them have an antidote, rH-VIIa has anecdotally been used off-label to stop life-threatening bleeding. Oral DTIs have recently become available.

Argatroban is an arginine analog with a 40-minute half-life. It is metabolized in the liver, and its half-life increases to around 140 minutes with hepatic dysfunction. Because of its short half-life, an initial bolus is not needed. Although argatroban is monitored by the aPTT, it has a significant effect on the PT.⁶⁶ Argatroban monotherapy will approximately double the INR and can markedly prolong the INR with warfarin cotherapy. As long as the aPTT is therapeutic, this exaggerated increase in INR does not increase the bleeding risk.

Hirudin is a family of recombinant proteins originally extracted from the salivary glands of leeches.⁶⁷ Lepirudin is used subcutaneously or intravenously. The half-life of intravenous lepirudin is 80 to 180 minutes but extends to days with severe renal failure. Although lepirudin can be removed by hemofiltration, the filters are not approved for use in the United States.⁶⁸ It should not be used in those with sulfite sensitivity. Approximately 40% of individuals exposed to lepirudin will develop antilepirudin antibodies, which will impair drug excretion and which may cause anaphylaxis on reexposure. Lepirudin is no longer produced for the European Union.

Bivalirudin is a reversible synthetic analog of hirudin. Bivalirudine has the advantage of a 25-minute half-life. Although it is cleared primarily by intravascular proteolysis, the dose must be reduced with renal dysfunction. Although not FDA approved, bivalirudin is used off-label in many centers for systemic anticoagulation for HIT(T).⁶⁹ It is also used off-label for bypass surgery in individuals with HIT.²⁷

Dabigatran (Pradaxa) is the first FDA approved oral DTI. It is approved for prevention of stroke and systemic embolization with nonvalvular atrial fibrillation.⁷⁰ It should be used with extreme caution in the elderly, especially because of the risk of bleeding due to overestimation of the Ccr and risk of trauma in the infirm. The dose is 150 mg twice a day (if Ccr > 30 mL/minute). Although not FDA approved, dabigatran has shown efficacy for the treatment of VTE.⁷¹ It binds free and clot bound thrombin, and is primarily cleared via the kidneys with a half-life of 12 to 17 hours.⁷² Drugs that are metabolized via P-glycoprotein may alter metabolism of dabigatran, which is a prodrug. There is no standardized laboratory test to measure the dabigatran level. Thus, in a patient who is having bleeding or thrombotic events while taking dabigatran, there is no test to tell whether the event may be related to over- or underanticoagulation. Given its sensitivity, a normal TT suggests that there is no dabigatran present. How long dabigatran should be discontinued before procedures depends upon the renal function and the bleeding risk of the procedure.⁷² Dabigatran has no antidote. In the bleeding patient, in addition to supportive care and transfusion, the use of activated charcoal (if within 2 hours of ingestion), FEIBA, rH-VIIA, and dialysis have been suggested. Although antifibrinolytics have been shown to decrease bleeding after trauma,⁷³ they should be used with trepidation with FEIBA and rH-VIIa because of the risk of thrombosis.

Oral Factor Xa Inhibitors ⁷⁴

Oral factor Xa inhibitors are undergoing clinical trials for a variety of indications. Because there are no evidence-based data regarding the reversal of oral factor Xa inhibitors in the bleeding patient, they should be used with caution in the ICU.

Apixiban

Apixiban (Eliquis), 5 mg twice a day, has been shown to be superior to warfarin for stroke prevention in atrial fibrillation and is under FDA review.⁷⁵ Its elimination is fecal and renal with an 8- to 15-hour half-life. Platelet GP and strong CYP3A4 inhibitors significantly increase its plasma concentration.

Rivaroxaban

Rivaroxaban (Xarelto), 20 mg daily, is FDA approved for the prevention of stroke and systemic embolization with nonvalvular atrial fibrillation, and doses of 10 mg daily have been approved for the prevention of VTE with knee and hip replacement surgery.⁷⁶ Although not FDA approved, it has been shown to be efficacious for the treatment of DVT and PE.⁷⁷ It has a 5- to 9-hour half-life (9-13 hours in the elderly) and is metabolized primarily via the cytochrome system. Combined platelet GP and strong CYP3A4 inhibitors may significantly increase the plasma concentration of rivaroxaban. It should be avoided in face of significant hepatic dysfunction, and dose reduction is required for a Ccr less than 50 mL/minute. It should be discontinued at least 24 hours before surgery. Rivaroxaban activity can be measured by the anti-Xa assay used to monitor heparins. Unfortunately, at the time of this writing, standards to calibrate the assay are not available in the United States. Prolongation of the INR by rivaroxaban has been reversed with the use of PCC in healthy volunteers.⁷⁸ There are no data on the efficacy of PCC in bleeding patients. Rivaroxaban is not dialyzable.

Warfarin

The vitamin K antagonist warfarin decreases the activities of factors II, VII, IX, and X and proteins C and S. Although prolongation of the PT is often seen within the first 24 to 48 hours after initiation of warfarin due to the decrease in factor VII, 4 to 5 days are required for factors II and X to drop to antithrombotic levels.

The response to warfarin is variable and is affected by many factors common in the ICU, such as poor nutrition (vitamin K deficiency), liver dysfunction, and coadministration of medications that affect warfarin pharmacokinetics. It is not uncommon for patients to become over-anticoagulated with warfarin. An approach to warfarin reversal is shown in Box 78.5.^79,80

Box 78.5

Reversal of Warfarin

International Normalized Ratio (INR) < 4.5, No Bleeding

• Lower dose or

• Omit dose and restart at a lower dose

INR > 4.5 but <10.0, No Significant Bleeding

• Omit 1 or 2 doses and restart at a lower dose, or

• For rapid reversal (e.g., surgery), vitamin K, 3 to 5 mg orally

INR should decrease in 24 hours

Give another 1 to 2 mg if the goal is not reached

INR > 10.0, No Significant Bleeding

• Hold warfarin and

• Give vitamin K, 5 to 10 mg orally

The INR should be significantly reduced in 24 to 48 hours

Give additional vitamin K orally if needed

Resume warfarin when the INR is therapeutic

Any INR, Major Bleeding

• Hold warfarin

• Vitamin K, 5-10 mg by slow intravenous infusion

• Repeat vitamin K every 12 hours as needed

• If urgent correction needed (in order of preference)

Quadravalent prothrombin complex concentrate

Fresh frozen plasma

Recombinant human factor VIIa

Adapted from Hirsch J, Raschke R: Heparin and low molecular weight heparin: The seventh ACCP conference on antithrombotic and thrombolytic therapy. Chest 2004;126:188S; Ageno W, Gallus AS, Wittkowsky A, et al: Oral anticoagulant therapy: Antithombotic therapy and prevention of thrombosis, 9th ed. ACCP evidence-based clinical practice guidelines. Chest 2012;141;e44S-e88S.

Key Points

• Hemorrhagic and thrombotic complications in a critically ill patient can usually be anticipated and avoided by careful history taking and physical examination. Routine laboratory tests (platelet count, PT, aPTT, TT, fibrinogen) are helpful in evaluating hemorrhagic disorders but not thrombotic disorders.

• Vitamin K deficiency is a relatively common coagulopathy in critically ill patients, and it is easily treated if recognized.

• von Willebrand disease, the most common inherited coagulopathy, varies highly in clinical severity. Humate-P and Alphanate are the only “factor VIII” concentrates that are therapeutic.

• The bleeding diathesis of uremia can improve with several different treatments, including dialysis, DDAVP, intravenous administration of estrogens, and cryoprecipitate.

• Idiopathic thrombocytopenic purpura (ITP) in adults is usually chronic. If life threatening, several modalities are available for urgent elevation of the platelet count.

• Disseminated intravascular coagulation (DIC) is always a secondary disease process. Treatment must be focused on the underlying cause. If bleeding occurs, blood components should be given. Heparin is generally reserved for clinically evident thrombosis.

• Thrombotic thrombocytopenic purpura (TTP) is a clinical diagnosis associated with multiple inciting events. It is often fatal unless recognized early and treated with plasma exchange.

• Heparin-induced thrombocytopenia with thrombosis may be life threatening. It should be looked for in any patient with thrombocytopenia who has been exposed to heparin. It can be managed with DTIs.

• Deep venous thrombosis is common in critically ill patients. Prophylactic treatment should be considered for every patient in the ICU. Intravenous UFH is often the safest treatment of established DVT in the critical care setting.

• Novel oral direct thrombin and factor Xa inhibitors are being used with increasing frequency. They have no antidotes for bleeding and there are no evidence-based protocols for their reversal.