 |
|||||||||
|
|||||||||
| CECIL |
| TEXT BOOK of MEDICINE |
Section XIV Hematologic Diseases
| 182 THROMBOTIC DISORDERS: HYPERCOAGULABLE STATES Andrew I. Schafer • |
|
|
The hypercoagulable states, also referred to as thrombophilias, encompass a group of inherited or acquired conditions that cause a pathologic thrombotic tendency or risk for thrombosis.
The primary hypercoagulable states are caused by quantitative or qualitative abnormalities in specific coagulation proteins that induce a prothrombotic state. Most of these disorders involve inherited mutations that lead to either (1) deficiency of a physiologic antithrombotic factor or (2) increased level of a prothrombotic factor (Table 182-1). Particularly when combined with other prothrombotic mutations (multigene interactions), these primary hypercoagulable states are associated with a lifelong predisposition to thrombosis. The trigger for a discrete, clinical thrombotic event is often the development of one of the acquired, secondary hypercoagulable states superimposed on an inherited state of hypercoagulability. The secondary hypercoagulable states, a diverse group of mostly acquired conditions (see Fig. 182-2), cause a thrombotic tendency by complex, often multifactorial mechanisms.
Click here to view this table....
 |
| FIGURE 182-2 Secondary hypercoagulable states. The pathophysiologic basis of thrombotic risk in these diverse disorders is complex and multifactorial. Predominant mechanisms of thrombosis for the different secondary hypercoagulable states shown are based on Virchow's triad of thrombogenesis: abnormalities in blood flow, abnormalities in blood composition, and abnormalities of the vessel wall. (Modified from Schafer AI: The primary and secondary hypercoagulable states. In Schafer AI [ed]: Molecular Mechanisms of Hypercoagulable States. New York, Chapman & Hall, 1997, p 16.) |
PRIMARY HYPERCOAGULABLE STATES
Antithrombin III Deficiency
Epidemiology and Pathobiology
Inherited quantitative or qualitative deficiency of antithrombin III leads to increased fibrin accumulation and a lifelong propensity to thrombosis (Chapter 178). Antithrombin is the major physiologic inhibitor of thrombin and other activated coagulation factors; therefore, its deficiency leads to unregulated protease activity and fibrin formation.
The frequency of asymptomatic heterozygous antithrombin deficiency in the general population may be 1 in 350. Most of these individuals have clinically silent mutations and never have thrombotic manifestations. The frequency of symptomatic antithrombin deficiency in the general population has been estimated to be between 1 in 2000 and 1 in 5000. Among all patients seen with venous thromboembolism, antithrombin deficiency is detected in only about 1%, but it is found in approximately 2.5% of selected patients with recurrent thrombosis or onset of thrombosis at a younger age (<45 years old).
Patients with type I antithrombin deficiency have proportionately reduced plasma levels of antigenic and functional antithrombin that result from a quantitative deficiency of the normal protein. Impaired synthesis, defective secretion, or instability of antithrombin in type I antithrombin-deficient individuals is caused by major gene deletions, single nucleotide changes, or short insertions or deletions in the antithrombin gene. Patients with type II antithrombin deficiency have normal or nearly normal plasma antigen accompanied by low activity levels, characteristics indicative of a functionally defective molecule. Type II deficiency is usually caused by specific point mutations leading to single amino acid substitutions that produce a dysfunctional protein. More than 250 different mutations causing type I or type II antithrombin deficiency have been recognized to date.
The pattern of inheritance of antithrombin deficiency is autosomal dominant. Most affected individuals are heterozygotes whose antithrombin levels are typically about 40 to 60% of normal. These individuals may have the full clinical manifestations of hypercoagulability. Rare homozygous antithrombin-deficient patients generally have type II deficiency with reduced heparin affinity, a variant that is associated with a low risk for thrombosis in its heterozygous form; other forms of homozygous antithrombin deficiency are probably incompatible with life.
Protein C Deficiency
Protein C deficiency leads to unregulated fibrin generation because of impaired inactivation of factors VIIIa and Va, two essential cofactors in the coagulation cascade.
Epidemiology
The prevalence of heterozygous protein C deficiency in the general population is about 1 per 200 to 500. Protein C deficiency is found in 3 to 4% of all patients with venous thromboembolism.
Pathobiology
As with antithrombin deficiency, two general forms of protein C deficiency are recognized: type I, in which quantitative deficiency of the protein is associated with a proportionate decrease in protein C antigen and activity, and type II, in which qualitative defects in protein C are associated with disproportionately reduced protein C activity relative to antigen. More than 160 mutations are known to cause protein C deficiency. In the more common type I deficiency, frameshift, nonsense, or missense mutations cause premature termination of synthesis or loss of protein C stability. In type II deficiency, different mutations can cause abnormalities in protein C activation or function. The mode of inheritance of protein C deficiency is autosomal dominant. As in antithrombin deficiency, most affected individuals are heterozygotes.
Protein S Deficiency
Protein S is the principal cofactor of activated protein C (APC), and its deficiency mimics that of protein C in causing loss of regulation of fibrin generation by impaired inactivation of factors VIIIa and Va.
Epidemiology
The prevalence of protein S deficiency in the general population is unknown. Its frequency in all patients evaluated for venous thromboembolism (2 to 3%) is comparable, however, to that of protein C deficiency.
Pathobiology
Protein S circulates in plasma partly in complex with C4b binding protein; only free protein S, which normally constitutes about 35 to 40% of total protein S, can function as a cofactor of APC. As in antithrombin and protein C deficiencies, quantitative (type I) and qualitative (type II) forms of inherited protein S deficiency are known. In addition, type III protein S deficiency is characterized by normal plasma levels of total protein S but low levels of free protein S.
Relatively few specific mutations of the protein S gene have been described to date. Most involve frameshift, nonsense, or missense point mutations.
Activated Protein C Resistance (Factor V Leiden)
Epidemiology
The factor V Leiden mutation is remarkably frequent (3 to 7%) in healthy white populations but is far less prevalent in certain black and Asian populations. In various studies, APC resistance was found in a wide range of frequencies (10 to 64%) in patients with venous thromboembolism.
Pathobiology
Most subjects with functional APC resistance have a single, specific point mutation in the gene for factor V, which is a critical target of the physiologic anticoagulant action of APC. In this mutation, termed factor V Leiden, guanine is replaced with adenine at nucleotide 1691 (G1691A), which leads to the amino acid substitution of Arg504 by Gln and renders factor Va incapable of being inactivated by APC. Heterozygosity for the autosomally transmitted factor V Leiden mutation increases the risk for thrombosis by a factor of 5 to 10, whereas homozygosity increases the risk by a factor of 50 to 100.
Prothrombin Gene Mutation
The substitution of G for A at nucleotide 20210 of the prothrombin gene has been associated with elevated plasma levels of prothrombin and an increased risk for venous thrombosis. The allele frequency for this gain-of-function mutation is 1 to 6% in white populations, but it is much less prevalent in other racial groups. The prothrombin G20210A mutation is found in 6 to 8% of all patients with venous thromboembolism.
Other Primary Hypercoagulable States
Elevated levels of factor VIII coagulant activity are a significant risk factor for venous thrombosis, and family studies suggest that high factor VIII levels are often genetically determined. Increased levels of factors VII, IX, and XI, fibrinogen, von Willebrand's factor, and thrombin-activatable fibrinolysis inhibitor, as well as very low levels of tissue factor pathway inhibitor, may also confer increased risk. Many other inherited abnormalities of specific physiologic antithrombotic systems may be associated with a thrombotic tendency. Most of these conditions are limited to case reports or family studies, their molecular genetic bases are less well defined, and their prevalence rates are unknown, but are probably much lower than those of the disorders described earlier. The other primary hypercoagulable states include heparin cofactor II deficiency, dysfunctional thrombomodulin, and many fibrinolytic disorders that lead to impaired fibrin degradation, including hypoplasminogenemia, dysplasminogenemia, plasminogen activator deficiency, and certain dysfibrinogenemias that cause a thrombotic rather than a bleeding diathesis.
Clinical Manifestations
The primary hypercoagulable states are associated with predominantly venous thromboembolic complications (Chapter 81). Deep venous thrombosis of the lower extremities and pulmonary embolism are the most frequent clinical manifestations. More unusual sites of venous thrombosis include superficial thrombophlebitis and mesenteric and cerebral venous thrombosis (see Table 178-2). Arterial thrombosis involving the coronary, cerebrovascular, and peripheral circulations is not linked to any of the primary hypercoagulable states, although some reports have described their occurrence with protein S deficiency and homozygous antithrombin deficiency. Venous thrombosis can also result in arterial occlusion by paradoxical embolism across a patent foramen ovale.
The initial episode of venous thromboembolism can occur at any age in patients with primary hypercoagulable states, but it typically takes place in early adulthood. Positive family histories of thrombosis can frequently be elicited. The risk for thrombosis varies among the individual primary hypercoagulable states and is highest in patients with deficiencies of antithrombotic factors (see Table 182-1); it is markedly increased with the coexistence of multiple prothrombotic mutations. Patients with homozygous deficiency states tend to have more severe thrombotic complications. A peculiar manifestation of homozygous protein C or protein S deficiency is neonatal purpura fulminans. This serious, sometimes fatal syndrome is caused by ischemic necrosis secondary to widespread thrombosis of small cutaneous and subcutaneous vessels. Fatal purpura fulminans associated with a bleeding diathesis has also been described in a patient with an acquired IgG inhibitor of protein C. Warfarin-induced skin necrosis (Chapter 35 and see Fig. 182-1) may infrequently complicate the initiation of oral anticoagulant therapy in patients with heterozygous protein C or protein S deficiency. Because both these proteins depend on vitamin K for normal function, their plasma levels in patients with inherited deficiency states may drop to nearly zero within a few days of starting therapy with warfarin, a vitamin K antagonist, and lead to a transient prothrombotic imbalance and skin necrosis caused by dermal vascular thrombosis. Nevertheless, oral anticoagulation provides effective long-term antithrombotic prophylaxis in these individuals.
 |
| FIGURE 182-1 Acute skin necrosis in a patient with protein C deficiency who was treated with heparin and warfarin for deep vein thrombosis that occurred after elective hip surgery. Warfarin treatment was withdrawn and anticoagulation continued with heparin. Skin grafting of the affected area was required. (From Forbes CD, Jackson WF: Color Atlas and Text of Clinical Medicine, 3rd ed. London, Mosby, 2003.) |
In most patients with primary hypercoagulable states, discrete clinical thrombotic complications seem to be precipitated by acquired prothrombotic events (e.g., pregnancy, use of oral contraceptives, surgery, trauma, immobilization), many of which are the secondary hypercoagulable states discussed subsequently. In particular, thrombosis complicates pregnancy, especially during the puerperium, in about 30 to 60% of women with antithrombin deficiency, 10 to 20% with protein C or protein S deficiency, and almost 30% with APC resistance, unless prophylactic anticoagulation is administered during this period.
Diagnosis
Laboratory diagnosis (Chapter 178) of the primary hypercoagulable states requires testing for each of the disorders individually because no general screening test is available to determine whether a patient may have such a condition. At this time, functional, immunologic, or DNA-based assays are available to test for antithrombin deficiency, protein C deficiency, protein S deficiency, APC resistance (factor V Leiden), and the prothrombin G20210A mutation.
A reasonable diagnostic approach at this time is to screen at least all “strongly thrombophilic” patients after an initial episode of venous thromboembolism: individuals with (1) a documented event before 50 years of age, (2) a positive family history, (3) massive or submassive pulmonary embolism, or (4) spontaneous thrombosis at an unusual site (e.g., intra-abdominal or cerebral). Although indefinite anticoagulation is recommended for patients who have had two or more venous thromboembolic events, regardless of whether a primary hypercoagulable state is found, testing these individuals for thrombophilia may also be useful to guide family screening strategies. Individuals with arterial thrombosis generally should not be tested for any of these disorders because primary hypercoagulable states (see Table 182-1) are not clearly associated with an increased risk for arterial thrombosis. In contrast, some of the secondary hypercoagulable states, including hyperhomocysteinemia and the antiphospholipid syndrome (see later), are associated with an increased risk for arterial thrombosis.
In general, testing for primary hypercoagulable states is not recommended immediately after a major thrombotic event, but rather in clinically stable patients at least 2 weeks after completing oral anticoagulation following a thrombotic episode. Active thrombosis may transiently consume and deplete some of the proteins in plasma and lead to the erroneous diagnosis of inherited antithrombin, protein C, or protein S deficiency. In addition to acute thrombosis, pregnancy, estrogen use, liver disease, and disseminated intravascular coagulation (DIC) may cause acquired deficiencies of antithrombin, protein C, or protein S. Anticoagulation may also interfere with some of the functional tests for primary hypercoagulable states. Heparin treatment can cause a decline in antithrombin levels to the deficiency range even in normal individuals. In contrast, warfarin can elevate antithrombin levels into the normal range in patients who do have an inherited deficiency state. Warfarin therapy also reduces the functional levels and, less prominently, the immunologic levels of protein C and protein S, thereby potentially leading to a misdiagnosis of inherited deficiency. When testing is indicated in patients in whom interruption of prophylactic oral anticoagulation is considered to be too risky, protein C and protein S levels can be determined after warfarin therapy has been discontinued under heparin coverage for at least 2 weeks.
Functional assays are the best screening tests for antithrombin, protein C, and protein S deficiencies because these assays detect both quantitative and qualitative defects; antigenic (immunologic) assays detect only quantitative deficiencies of these proteins. Functional coagulation assays for protein C and protein S may yield spuriously low values, however, if APC resistance is present. APC resistance can be diagnosed by the newer high-sensitivity and high-specificity coagulation assays or by DNA analysis of peripheral blood mononuclear cells for the factor V Leiden mutation.
Treatment
The initial treatment of acute venous thrombosis or pulmonary embolism in patients with primary hypercoagulable states is not different from that in patients without genetic defects (Chapters 35 and 81). As in patients without known thrombophilia, thrombolytic therapy should be considered after massive venous thrombosis or pulmonary embolism. Acute management is initiated with at least 5 days of unfractionated or low-molecular-weight heparin. Oral anticoagulation with warfarin can be started on the first day of heparin use and continued for at least 6 months in patients with venous thromboembolism in the absence of triggering factors (e.g., postoperative state), with regulation of the dose to maintain an international normalized ratio (INR) of the prothrombin time between 2.0 and 3.0.1
Continuing oral anticoagulant prophylaxis beyond the initial 6 to 12 months after an acute episode of venous thromboembolism must be weighed against continued exposure of the individual patient to the significant risk for bleeding complications. Patients with primary hypercoagulable states who have had two or more thrombotic events should receive indefinite or lifelong prophylactic anticoagulation with warfarin (Chapter 81). Indefinite or lifelong anticoagulation is probably indicated for individuals with recurrent thrombosis even in the absence of identifiable primary hypercoagulable states.
The decision to continue prophylactic oral anticoagulation beyond the initial period after the first episode of thrombosis is more difficult (Table 182-2). After a single episode of thrombosis, patients with inherited hypercoagulable states should probably receive indefinite or lifelong anticoagulation if their initial episodes were life-threatening or occurred in unusual sites (e.g., mesenteric, cerebral venous thrombosis) or if they have more than one prothrombotic genetic abnormality (Chapter 81). Some authorities also recommend indefinite or lifelong anticoagulation after an initial venous thromboembolic event in patients whose risk of recurrence likewise appears to be increased: those with isolated heterozygous deficiencies of antithrombin, protein C, or protein S or patients with homozygous factor V Leiden. In the absence of these characteristics, particularly if the initial episode was precipitated by a transient acquired prothrombotic situation (e.g., pregnancy, postoperative state, immobilization), it is reasonable at this time to discontinue warfarin therapy after 6 to 12 months and administer subsequent prophylactic anticoagulation only during high-risk periods.
Click here to view this table....
Asymptomatic individuals with known thrombophilia who have not had previous thrombotic complications do not require prophylactic anticoagulation except during periods of high risk for thrombosis. Because about half of the first-degree relatives of a patient with a primary hypercoagulable state should be affected, these individuals should be counseled about the implications of making a diagnosis.
Management of pregnancy in women with primary hypercoagulable states requires special consideration because of the high risk for thrombosis, particularly during the puerperium. Women with thrombophilia who have previously had thrombosis—and probably also asymptomatic women with thrombophilia—should receive prophylactic anticoagulation throughout pregnancy and for 4 to 6 weeks postpartum, a particularly high-risk period. Coumarin derivatives cross the placenta and have the potential to cause both bleeding and teratogenic effects in the fetus; therefore, oral anticoagulants should not be used during pregnancy. Heparin does not cross the placenta and does not cause these fetal complications. Therefore, either unfractionated heparin or fixed-dose, low-molecular-weight heparin is the anticoagulant of choice during pregnancy. Neither warfarin nor heparin induces an anticoagulant effect in a breast-fed infant when the drug is given to a nursing mother, so either can be given safely when indicated in the postpartum period.
Because warfarin-induced skin necrosis (Fig. 182-1) is a rare problem, screening of all patients for inherited protein C or protein S deficiency, conditions that are known to predispose to this complication, is not indicated before starting warfarin therapy. Most cases can be avoided by not initiating warfarin therapy with high loading doses and by concomitant coverage with heparin. When the complication does occur, as manifested by painful red and subsequently dark, necrotic skin lesions within a few days of starting warfarin, such therapy must be discontinued immediately, vitamin K administered, and heparin started (Chapter 35). The use of fresh-frozen plasma or purified protein C concentrate to normalize protein C levels rapidly can improve results. Despite this rare complication, warfarin is an effective, long-term prophylactic anticoagulant in patients with inherited protein C or protein S deficiency.
Antithrombin III concentrate purified from normal human plasma may be a useful adjunct to anticoagulation in “heparin-resistant” patients, who represent unusual cases of type II antithrombin deficiency, and in antithrombin-deficient patients with recurrent thrombosis despite adequate anticoagulation. Infusion of antithrombin concentrate can also be considered in some perioperative or obstetric settings in which anticoagulation poses an unacceptable bleeding risk.
SECONDARY HYPERCOAGULABLE STATES
Definition
The secondary hypercoagulable states (Fig. 182-2) are diverse, mostly acquired disorders that predispose patients to thrombosis by complex, multifactorial pathophysiologic mechanisms. Many of these conditions also represent the acquired precipitating stimuli for clinical thrombotic events in individuals with a genetic predisposition (primary hypercoagulable states). Although each disorder causes thrombosis primarily through abnormalities in blood flow (rheology), the composition of blood (coagulation factors and platelet function), or the vessel wall, multiple overlapping mechanisms are operative in many of them.
Hyperhomocysteinemia
Hyperhomocysteinemia is an elevated blood level of homocysteine, a sulfhydryl amino acid derived from methionine by a transmethylation pathway (Fig. 182-3). Homocysteine is remethylated to methionine or catabolized to cystathionine. The major remethylation pathway requires folate and cobalamin (vitamin B12) and involves the action of methylenetetrahydrofolate reductase (MTHFR); a minor remethylation pathway is mediated by betaine-homocysteine methyltransferase. Alternatively, homocysteine is converted to cystathionine in a trans-sulfuration pathway catalyzed by cystathionine β-synthase (CBS), with pyridoxine used as a cofactor.
 |
| FIGURE 182-3 Intracellular metabolism of homocysteine occurs through remethylation to methionine or trans-sulfuration to cysteine. Numbered circles indicate the principal enzymes involved: (1) methionine synthase; (2) 5,10-methylenetetrahydrofolate reductase; (3) betaine-homocysteine methyltransferase; (4) cystathionine β-synthase. (Modified from De Stefano V, Finazzi G, Mannucci PM: Inherited thrombophilia: Pathogenesis, clinical syndromes, and management. Blood 1996;87:3531–3544.) |
Homozygous CBS deficiency states that lead to severe hyperhomocysteinemia (homocystinuria) (Chapter 228) cause premature arterial atherosclerotic disease and venous thromboembolism, as well as mental retardation, neurologic defects, lens ectopy, and skeletal abnormalities. By comparison, adults with heterozygous CBS deficiency, with resultant mild to moderate hyperhomocysteinemia, may have only venous or arterial thrombotic manifestations. Hyperhomocysteinemia resulting from inherited remethylation pathway defects usually involves reduced activity of MTHFR. In homozygous individuals with the autosomal recessive C677T mutation of the MTHFR gene, which occurs in 15% of certain populations, moderate hyperhomocysteinemia may occur and is correctable with folic acid, but its relationship to venous thrombosis is controversial. Acquired causes of hyperhomocysteinemia in adults most commonly involve nutritional deficiencies of the cofactors required for homocysteine metabolism, including pyridoxine, cobalamin, and folate.
Acquired and inherited hyperhomocysteinemia is a probable risk factor for both arterial and venous thrombosis (Chapters 49 and 81). The mechanism of homocysteine-induced thrombosis and atherogenesis involves complex, probably multifactorial effects on the vessel wall. Homocysteine can cause vascular endothelial injury, conversion of the endothelial surface of blood vessels from an antithrombotic to a prothrombotic state, and smooth muscle cell proliferation. These toxic effects of homocysteine on the vessel wall may be mediated by oxidant stress.
Vitamin supplementation with folate, pyridoxine, and cobalamin can normalize elevated blood levels of homocysteine. However, several recently reported prospective clinical trials of homocysteine-lowering therapy have failed to show reduced rates of vascular events in patients with established vascular disease. It remains to be determined whether this disappointing lack of clinical benefit with homocysteine-lowering vitamin therapy indicates that homocysteine is not a direct atherogenic factor, or that vitamin therapy in this setting might have other, offsetting deleterious effects, or that possibly other mechanisms are operative.
Malignancy
Multiple abnormalities of hemostasis are involved in the hypercoagulable state in cancer patients, many of which initiate a systemic process of chronic DIC (Chapters 181 and 189). The thrombotic tendency of patients with cancer may also be related to mechanical factors, such as immobility, indwelling central venous catheters, or a bulky tumor mass compressing vessels, and to comorbid conditions, such as sepsis, surgery, liver dysfunction secondary to metastases, and the prothrombotic effects of certain antineoplastic agents.
The incidence of thrombotic complications in cancer patients depends in part on the type of malignancy. Hypercoagulability seems to be most prominent in patients with pancreatic cancer (Chapter 204), adenocarcinoma of the gastrointestinal tract (Chapters 202 and 203) or lung (Chapter 201), and ovarian cancer (Chapter 209). The presence of underlying malignancy compounds the independent risk for thrombosis in the postoperative state. Thrombosis most commonly occurs in patients with established or concurrently diagnosed malignancy (Chapters 81 and 189).
The most frequent thrombotic manifestations in patients with neoplasms are deep vein thrombosis and pulmonary embolism, but more unusual and distinctive thrombotic complications are also found. Trousseau's syndrome, characterized by migratory superficial thrombophlebitis of the upper or lower extremities, is strongly linked to cancer. Nonbacterial thrombotic endocarditis involves fibrin-platelet vegetations on heart valves, which produce clinical manifestations by systemic embolization (Chapter 59). Of patients with nonbacterial thrombotic endocarditis, 75% have underlying malignancies at autopsy. Trousseau's syndrome and nonbacterial thrombotic endocarditis are highly associated with adenocarcinomas. The occurrence of either syndrome in patients without known cancer demands a more vigorous search for occult malignancy than in patients with deep vein thrombosis or pulmonary embolism. Thrombotic microangiopathy (Chapter 189), characterized by hemolysis with red blood cell fragmentation, thrombocytopenia, and microvascular thrombosis with involvement of target organs, occurs in about 5% of patients with metastatic carcinomas, most commonly those with gastric (Chapter 202), lung (Chapter 201), and breast (Chapter 208) primary sites.
Treatment
Treatment of acute venous thromboembolism in cancer patients should be initiated as in other patients, but subsequent prophylactic anticoagulation should be continued while active malignancy is present. Anticoagulation can be difficult in many cancer patients; these patients may be resistant to warfarin prophylaxis. Anticoagulation can also be complicated by bleeding into tumors. Long-term treatment of cancer patients with low-molecular-weight heparin after venous thromboembolism (Chapter 35) reduces recurrences and possibly decreases bleeding complications when compared with treatment with warfarin.2
Myeloproliferative Disorders and Paroxysmal Nocturnal Hemoglobinuria
Thrombosis and, apparently paradoxically, bleeding are major causes of morbidity and mortality in patients with myeloproliferative disorders (Chapters 172 and 177) and the related stem cell disorder paroxysmal nocturnal hemoglobinuria (Chapter 164). In uncontrolled polycythemia vera (Chapter 172), increased whole blood viscosity contributes to the thrombotic tendency. Thrombocytosis, abnormal platelet function, and other less well understood factors are also probably involved in the hemostatic defect of the myeloproliferative disorders and paroxysmal nocturnal hemoglobinuria.
In addition to deep vein thrombosis and pulmonary embolism, some distinctive thrombotic manifestations are seen. Hepatic vein thrombosis (Budd-Chiari syndrome) and portal and other intra-abdominal venous thromboses (Chapter 146) are associated with myeloproliferative disorders and paroxysmal nocturnal hemoglobinuria (Chapter 164) and may be the initial manifestations of the disease. Myeloproliferative disorders, particularly polycythemia vera (Chapter 172) and essential thrombocythemia (Chapter 177), may cause erythromelalgia (see Fig. 172-2), a syndrome of microvascular thrombosis manifested by intense pain accompanied by warmth, duskiness, and mottled erythema, sometimes resembling livedo reticularis, in a patchy distribution in the extremities, most prominently in the feet; digital microvascular ischemia progressing to vascular insufficiency and gangrene may ensue (Chapter 80). A wide spectrum of neurologic manifestations may be caused by cerebrovascular ischemia, especially in patients with essential thrombocythemia.
Treatment
Treatment of venous thromboembolism in patients with the myeloproliferative disorders and paroxysmal nocturnal hemoglobinuria should be initiated as in patients without these hematologic disorders. In patients with thrombosis associated with polycythemia vera, the hematocrit should be maintained in the normal range with phlebotomies or chemotherapy, or with both (Chapter 172). Low-dose aspirin (100 mg daily) can prevent thrombotic complications without increasing the incidence of major bleeding in patients with polycythemia vera who have no contraindications to such treatment.3 In patients with essential thrombocythemia, cytoreduction of the elevated platelet count should be achieved with chemotherapy (Chapter 177).
Antiphospholipid Syndrome
Antiphospholipid syndrome is characterized by venous and arterial thrombosis, recurrent spontaneous abortions (which may also be due to thrombosis), thrombocytopenia, and a variety of neuropsychiatric manifestations. The syndrome is associated with a heterogeneous group of autoantibodies that bind to anionic phospholipid-protein complexes, a protein cofactor of which is β2-glycoprotein I. Patients with this syndrome have any combination of positive tests to detect different plasma antiphospholipid-protein antibodies (e.g., anticardiolipin antibodies) and/or phospholipid-based clotting tests (lupus anticoagulants) (Chapter 287). The predominant prothrombotic effects of these antibodies are probably directed to the vessel wall.
Deep vein thrombosis and pulmonary embolism are the most frequent venous thrombotic events in these patients. Cerebrovascular events are the most common arterial thrombotic complications and are manifested as stroke, transient ischemic attacks (Chapter 431), multi-infarct dementia (Chapters 26 and 425), or retinal artery occlusion (Fig. 449-20). Peripheral and intra-abdominal vascular occlusion is encountered more rarely. About a third of these patients have nonbacterial heart valve vegetations (Libman-Sacks endocarditis). The most prominent obstetric complications are recurrent spontaneous abortions and fetal growth retardation, which are probably due to thrombosis of placental vessels. Patients are occasionally seen with “catastrophic” antiphospholipid syndrome involving a series of acute and sometimes fatal vascular occlusive events, or “thrombotic storm.” Thrombotic complications are limited largely to patients with primary antiphospholipid syndrome or patients in whom the antibodies are associated with collagen vascular disease, not with drugs or infections.
Treatment
Acute management of thrombosis in these patients is essentially the same as in other individuals. Monitoring of heparin anticoagulation is difficult in patients with a lupus anticoagulant because they already have a prolonged activated partial thromboplastin time at baseline; the use of low-molecular-weight heparin, which does not require monitoring, can circumvent this problem. Warfarin is effective in preventing recurrent thrombosis but usually requires prolonged or indefinite therapy with doses to achieve an INR of 2.0 to 3.0.4 No established treatment of women with antiphospholipid syndrome has been shown to prevent recurrent fetal loss. Treatment with prednisone and aspirin during pregnancy is not effective in promoting live birth and may increase the risk for prematurity.
Pregnancy, Oral Contraceptives, and Hormone Replacement Therapy
The pathophysiology of hypercoagulability associated with pregnancy (Chapter 259) involves a progressive state of DIC throughout the course of pregnancy. Activation of the coagulation system is initiated locally in the uteroplacental circulation, where the placenta is the source of increased thrombin generation. Platelet activation and increased platelet turnover also occur during normal pregnancy, and about 8% of healthy women have mild thrombocytopenia at term. Simultaneously, the fibrinolytic system is progressively blunted throughout pregnancy because of the action of placental plasminogen activator inhibitor type 2. The net effect of these coagulation changes is creation of a state of hypercoagulability that makes pregnant women vulnerable to thrombosis, particularly in the puerperium. These systemic alterations are compounded by prothrombotic mechanical and rheologic factors in pregnancy, including venous stasis in the legs caused by the gravid uterus, pelvic vein injury during labor, and the trauma of cesarean section. Oral contraceptives induce a prothrombotic state by increasing procoagulant effects and decreasing physiologic anticoagulant effects. The use of oral contraceptives is associated with an increased risk for venous thrombosis, myocardial infarction, stroke, and peripheral arterial disease, particularly during the first year of use (Chapter 262). Unexpectedly, third-generation oral contraceptives, which contain less estrogen and a different progestin, double the risk for venous thromboembolism in comparison to second-generation preparations. Postmenopausal hormone replacement increases the risk for deep venous thromboembolism by a factor of 2 to 3.5, at least during the first year. Hormone replacement therapy has no beneficial and possibly even a detrimental effect on the risk for arterial disease (Chapter 262).
Deep vein thrombosis and pulmonary embolism are the most common thrombotic complications of pregnancy and the use of oral contraceptives or hormone replacement therapy. Coexisting primary hypercoagulable states are an additive risk factor in all of these settings. In the absence of a clear family history of venous thromboembolism, there is little justification, however, to screen for prothrombotic mutations with pregnancy or before starting hormone replacement therapy or oral contraceptives. Increasing age, increasing parity, cesarean delivery, prolonged bedrest or immobilization, obesity, and previous thromboembolism are additional prothrombotic risk factors in pregnant women. Most thrombotic events associated with pregnancy occur in the peripartum period, especially after delivery. Special considerations for anticoagulation in the setting of pregnancy are noted in the section on treatment of primary hypercoagulable states.
Postoperative State, Immobilization, and Trauma
Postoperative thrombosis (Chapter 459) is caused by a combination of local mechanical factors, including decreased venous blood flow in the lower extremities, and systemic changes in coagulation (Chapter 81). The level of risk for postoperative thrombosis depends largely on the type of surgery performed. It is probably compounded by coexisting risk factors, such as an underlying inherited primary hypercoagulable state or malignancy, advanced age, and prolonged procedures. Postoperative deep vein thrombosis and pulmonary embolism, the most common thrombotic complications, are often asymptomatic but detectable by noninvasive studies. The incidence of deep vein thrombosis after general surgical procedures is about 20 to 25%, with almost 2% of these patients having clinically significant pulmonary embolism. The risk for deep vein thrombosis after hip surgery and knee reconstruction ranges from 45 to 70% without prophylaxis, and clinically significant pulmonary embolism occurs in 20% of patients undergoing hip surgery. Postoperative thrombosis risk after urologic and gynecologic surgery more closely approximates that found after general surgery. Although the process of thrombosis generally begins intraoperatively or within a few days of surgery, the risk for this complication can be protracted beyond the time of discharge from the hospital, particularly in hip replacement patients.
Patients who are bedridden or experiencing prolonged air travel are at increased risk for venous thromboembolism. Venous thromboembolism is also one of the most common causes of morbidity and mortality in survivors of major trauma, and asymptomatic deep vein thrombosis of the lower extremities has been detected by venography in more than 50% of hospitalized trauma patients (Chapter 113). The risk for venous thrombosis after trauma is increased by advanced age, need for surgery or transfusions, and the presence of lower extremity fractures or spinal cord injury.
Mechanical methods of prophylaxis against venous thromboembolism should be considered in high-risk postoperative patients and bedridden patients with medical conditions, either in combination with anticoagulant prophylaxis or instead of it in patients who have an unusually high risk for bleeding with anticoagulation. Such methods include graduated compression stockings, intermittent pneumatic compression devices, and venous foot pumps. For long-distance travelers with thrombophilia, either properly fitted below-knee graduated compression stockings or a single prophylactic dose of low-molecular-weight heparin injected before departure is recommended in addition to general measures such as avoidance of dehydration and frequent stretching of calf muscles.
|
|