Stroke

Fall 2000
Volume 11, Number 2

 Next Article  

Stroke is a leading cause of death and morbidity throughout the world. The role of hypercoagulable states in the development of stroke is controversial from reports in the literature. However, there does appear to be an association between stroke, especially in young patients, and a number of hypercoagulable conditions such as antiphospholipid antibody syndrome, Leiden Factor V mutation, activated protein C resistance, prothrombin gene G20210A mutation, protein C and protein S deficiencies. There also does appear to be an association between some of these conditions and cerebral vein thrombosis. Laboratory evaluation and clinical management of these patients is discussed.

Introduction. Stroke presently is the third leading cause of death in the United States, surpassed only by heart disease and cancer.1 There are a number of risk factors for stroke, classified as nonmodifiable and modifiable. Some of the nonmodifiable factors include age, male gender, and ethnicity. There are a number of risk factors which can be modified including hypertension, atrial fibrillation, coronary disease, diabetes, hypercholesterolemia, smoking, obesity, and others.

There are a number of hematologic causes of stroke.These include hemoglobinopathies, sickle cell disease, and myeloproliferative disorders, such as polycythemia rubra vera and essential thrombocytosis. It also has been recognized that a number of clotting disorders can be factors in the etiology of stroke. These include hereditary disorders such as Factor V Leiden mutation, which is the most frequent cause of activated protein C resistance, protein C deficiency, protein S deficiency, antithrombin-III deficiency, and G20210A prothrombin mutation.3

There are review articles3 describing these hypercoagulable states as well as reviews4 discussing some of these risk factors in cerebral thrombosis and stroke.

The role of hypercoagulable states in strokes is controversial; however, in patients who do not have other obvious traditional risk factors and etiologies for stroke, a hypercoagulable state should be considered. Identification of a hypercoagulable state may be even more important in the younger patient with stroke.

Antiphospholipid Antibodies. Antiphospholipid antibodies consist of several related, but somewhat clinically distinct subgroups, including lupus anticoagulants (LA), anticardiolipin antibodies (ACAs), and
a number of less well characterized and investigated antiphospholipid antibodies.5,6 ACAs and LA occur in approximately 5% and 4%, respectively, of the general population. Antiphospholipid antibodies have been associated with venous thromboembolism as well as arterial thrombosis, including coronary thrombosis, stroke, and transient ischemic attacks. The presence of antiphospholipid antibodies is considered a risk factor for stroke by some authors, however, some studies have shown no such association.5 

The presence of antiphospholipid antibodies is common in patients with systemic lupus erythematosus and other collagen vascular diseases. Clinical features seen in association with antiphospholipid antibodies include collagen vascular symptoms that do not meet all the criteria for a definitive diagnosis of a definable collagen vascular disorder. These include sun sensitivity, Raynaud’s phenomenon, and immune thrombocytopenia. Also, some patients will have a history of recurrent fetal loss and spontaneous abortion. It is the clinical impression of this author that on careful history-taking, many patients with antiphospholipid antibodies have some of the above symptoms, or often may have a family history of collagen vascular disease. Antinuclear antibodies are often found, usually in low titers and nonspecific patterns.

The laboratory evaluation for antiphospholipid antibodies includes the direct detection of antibodies by an enzyme-linked immunoassay (ELISA). There also are a number of functional tests to detect the presence of ACAs and LA. LA can often be suspected when a prolonged activated partial thromboplastin time (APTT) is found. However, a prolonged APTT is not present in many patients who have antiphospholipid antibodies. Another useful test in the detection of LA is the dilute Russell viper venom test. Definitive diagnosis of an antiphospholipid antibody syndrome usually requires a positive test for the determination of antibodies, or a positive functional test on at least two occasions, in the appropriate clinical setting.

It is generally recommended that patients who develop thrombosis associated with ACAs and LA be treated with heparin initially and then with warfarin.5 However, there is controversy as to what level of International Normalized Ratio (INR) is necessary for maintenance therapy. It has been reported by some investigators,7 and also it has been the clinical experience of this author, that recurrent thrombosis can occur in patients with ACAs and LA who are treated with the usual intensity of warfarin treatment, and that prevention of recurrences requires an INR exceeding 3.0. There are conflicting reports regarding the value of INR in the management of these patients.5

Activated Protein C Resistance. Activated protein C resistance is a common cause of hereditary predisposition to venous thrombosis.3 The most common cause of activated protein C resistance is the Factor V Leiden mutation, which is present in the heterozygous form in 5% of the general white population. It is less common in other ethnic groups. The role of activated protein C resistance and Leiden Factor mutation in arterial thrombosis is less clear. A number of studies suggest an association.8-10 Leiden Factor V mutation has been described in neonatal stroke.11

Activated protein C resistance can occur in the absence of the Leiden Factor V gene mutation. This abnormality has been reported in patients with stroke.12-14 Therefore, it may be important to perform a functional assay for activated protein C resistance in addition to testing for the Leiden Factor V mutation in these patients. The controversy as to the relevance of Leiden Factor V mutation in stroke occurs because there are a number of published reports which do not find this association between Leiden Factor V and stroke.15-17 A review by Huisman did not find that Leiden mutation was a major risk factor for myocardial infarction or stroke unless there was also another classical risk factor, such as diabetes, hypertension, and smoking.18 Press, et al, reported a lack of an association between Leiden Factor V mutation and ischemic stroke in the elderly.19 It may be that in the elderly, an extensive work-up for a hypercoagulable state is not indicated unless there is a history of previous recurrent venous embolism or a strong family history of a hereditary thrombophilia. Activated protein C resistance due to Leiden Factor V mutation is not a significant risk factor for stroke in young African-Americans.20 This may be due to the low prevalence of Leiden Factor V mutation in this ethnic group. However, in this study there were some young patients with stroke who did demonstrate functional activated protein C resistance, which may have been due to causes other than the Leiden Factor V mutation.

Prothrombin Gene G20210A Mutation. Fairly recently another common genetic variation, prothrombin gene G20210A, has been described. This mutation causes elevated plasma prothrombin concentrations and is a not uncommon cause of venous thrombosis. It has been found in the heterozygous form in 2.3% of normal controls.3 The prothrombin G20210A mutation has been found to be associated with increased risk of stroke and may be a factor in the etiology of cerebral ischemia in young patients.22 Other studies18 have not shown this association. Testing for prothrombin G20210A mutation is readily available by polymerase chain reaction. A study by Reuner23 showed an association with the prothrombin G20210A mutation and cerebral vein thrombosis, but not with acute ischemic stroke or transient ischemic attack. Likewise, Martinelli24 did show an association of cerebral vein thrombosis with the prothrombin gene mutation as well as with the Leiden Factor V mutation. This study also showed a strong and independent association of oral contraceptives to cerebral vein thrombosis. The presence of both prothrombin gene mutation and oral contraceptive use raised the risk of cerebral vein thrombosis additively.

Protein C and Protein S Deficiency. Thrombomodulin is expressed on endothelial cell surfaces. This binds thrombin and inactivates thrombin’s procoagulant activity.3 (Fig. 1) Protein C binds to the thrombomodulin-thrombin complex and becomes converted into activated protein C. Activated protein C along with protein S as a cofactor inhibits coagulation by degrading the activated Factor V and Factor VIII. Protein C and protein S deficiency occur in approximately 0.5% and 0.7% of the general population. Deficiencies in both of these factors are well recognized etiologies of venous thrombosis. The role of protein C and protein S deficiency in arterial thrombotic disease and stroke is less clear. There are, however, several studies that have demonstrated the presence of protein C deficiency in stroke, either alone or in combination with other causes of a hypercoagulable state.25-28

Since both protein C and S are consumed during acute thrombotic events, and their levels can be falsely decreased in the presence of acute phase reactants, the measurement of these factors in this setting is unreliable. When protein C or protein S deficient patients are identified, warfarin treatment must not be started until full heparinization has been achieved. Warfarin has been shown to depress the levels of protein C and protein S, and unless patients are on heparin with therapeutic activated partial thromboplastin times, they may develop a transient increased hypercoagulable state and there is the risk of the unique phenomenon of “Coumadin-induced skin necrosis.”

Figure 1

Miscellaneous and Less Common Risk Factors Associated with Arterial Thrombosis. Homocysteine is a sulfur-containing amino acid which is formed during the metabolism of methionine.29 The concentration of homocysteine in body fluids is regulated by two primary enzymes, cystathionine beta-synthase and methylene tetrahydrofolate reductase (MTHFR). Genetic deficiencies of MTHFR are common in the North American population.30 Elevated homocysteine levels also occur in the presence of B12, B6, and folate deficiencies, as well as in renal failure. Hyperhomocysteinemia is recognized as a risk factor for both arterial and venous thrombosis. Even mild elevations of homocysteine can be a risk factor for occlusive arterial disease.29, 30 Elevated levels of homocysteine have been found to be a risk factor in stroke and thrombotic events in patients with systemic lupus erythematosus 31 and also may be a risk factor for stroke in the general population. Recognition of hyperhomocysteinemia is important in that it can be easily treated by increased folic acid intake.31

Several other risk factors have been implicated in arterial thrombosis and may be a factor in stroke.30 These include elevated plasma fibrinogen, Factor VII, and Plasminogen Activator Inhibitor-I levels. It also has been suggested that antithrombin deficiencies may be a factor in ischemic stroke.14

Platelets also do play a significant in the development of thrombosis. A discussion of this is beyond the scope of this article.

Conclusion. Hypercoagulable states, both congenital and acquired, may be causes of both ischemic stroke and cerebral vein thrombosis. A work-up to identify one of the recognizable hypercoagulable states is indicated, especially in younger patients with stroke. Laboratory evaluation for hypercoagulable states may also often be indicated in those patients who do not have other obvious risk factors for their stroke.

In the acute setting, several studies may be obtained in the evaluation of a possible hypercoagulable state. These include the Leiden Factor V mutation, the prothrombin G20210A mutation, and anticardiolipin antibody studies. A baseline activated partial thromboplastin time also is useful, in that if it is elevated it may indicate the presence of a lupus anticoagulant. Other recognizable hematologic conditions also should be looked for including thrombocytosis, significant erythrocytosis, etc. It may also be useful to measure homocysteine levels.

Studies which are not helpful in the setting of an acute event, include determination of fibrinogen levels, protein C, protein S, and antithrombin-III levels.

It is also important to recognize other conditions which may in and of themselves be associated with a hypercoagulable state. These may be synergistic with one of the above described abnormalities causing the development of thrombosis. These include the use of oral contraceptives or hormones, systemic inflammatory disorders, and malignancies.

For those patients who are placed on oral anticoagulation with warfarin, it may be appropriate to reevaluate them off of anticoagulants after 4 to 6 months of treatment. If from clinical history, family history and/or laboratory studies, a patient is felt to have a hypercoagulable state, the decision for long term chronic anticoagulation needs to be individualized. This decision needs to occur after a thorough discussion of the risks of recurrent thrombosis, as well as the risks of long term anticoagulation. If a hereditary hypercoagulable state is found, it also may be appropriate to recommend screening of other family members. There may be recommendations that can be made to other affected family members that may be able to reduce their risk of thrombosis in the future.
  

Address comments and questions to:
 
David R. Trevarthen, MD
701 E. Hampden Ave.,  Suite 370
Englewood, CO 80113

 Next Article