Journal of Neuro-Ophthalmology, March 2003, Volume 23, Issue 1

The Coagulation System

Congenital and acquired hypercoagulable states arise from an imbalance between procoagulant and anticoagulant activity. Although these imbalances are present throughout the entire vascular tree, thrombotic lesions are usually localized in discrete segments of the veins or arteries and in certain organ systems. Thus, hypercoagulable states are likely to be associated with focal defects in the vascular wall to produce thrombosis. Many recently described factors are associated with hypercoagulability. Because thrombosis is a disease in which genetic and acquired risk factors interact dynamically, a thorough history, family history, and physical examination should be performed before ordering an extensive and costly coagulation panel.

NANOS SYMPOSIUM The Coagulation System Valerie Biousse, MD Abstract: Congenital and acquired hypercoagulable states arise from an imbalance between procoagulant and antico­agulant activity. Although these imbalances are present throughout the entire vascular tree, thrombotic lesions are usually localized in discrete segments of the veins or arter­ies and in certain organ systems. Thus, hypercoagulable states are likely to be associated with focal defects in the vascular wall to produce thrombosis. Many recently de­scribed factors are associated with hypercoagulability. Be­cause thrombosis is a disease in which genetic and acquired risk factors interact dynamically, a thorough history, family history, and physical examination should be performed be­fore ordering an extensive and costly coagulation panel. ( JNeuro- Ophthalmol 2003; 23: 50- 62) The coagulation of blood is mediated by cellular compo­nents and soluble plasma proteins. Thrombosis in­volves complex interactions between the endothelial sur­face, platelets, and several activated coagulation factors. The mechanisms underlying thrombosis are contained within Virchow's triad: ( 1) decrease in blood flow ( stasis); ( 2) injury to the vessel wall; and ( 3) imbalance between procoagulant and anticoagulant factors ( 1- 16). In response to vascular injury, circulating platelets adhere, aggregate, and provide cell- surface phospholipid for the assembly of blood- clotting enzyme complexes ( Fig. 1). The blood coagulation system has the ability to trans­duce a small initiating stimulus into a large fibrin clot. The potentially explosive nature of this cascade is offset by the natural anticoagulant mechanisms. The maintenance of ad­equate blood flow and the regulation of cell- surface activity limit the local accumulation of activated blood- clotting en­zymes and complexes. Departments of Ophthalmology and Neurology, Emory University, Atlanta, Georgia. Address correspondence to Valerie Biousse, MD, Emory University School of Medicine, Neuro- ophfhalmology Unit, Emory Eye Center, 1365 B Clifton Road, N. E. Atlanta, GA 30322, USA; E- mail: vbiouss@ emory. edu Peer reviewed and modified from an oral presentation at the 28th An­nual Meeting of the North American Neuro- Ophfhalmology Society, Cop­per Mountain, Colorado, February 9- 14, 2002. Congenital and acquired hypercoagulable states arise whenprothrombotic activities predominate over anticoagu­lant activities or when clot- lysing mechanisms are reduced ( 1- 16). According to this formulation, the loss of a circu­lating anticoagulant should cause a diffuse thrombotic phe­nomenon. However, systemic alterations in the hemostatic mechanism typically produce local thrombotic lesions, as in the central retinal vein, cerebral sinuses, lower extremity veins, or coronary arteries. The pathophysiology underly­ing this phenomenon is attributed to local defects in the vas­cular wall or in blood flow, perhaps related to signaling pathways specific to a particular vascular bed ( 8,9). Endo­thelial cell- derived procoagulant and anticoagulant activi­ties may be differentially expressed across the vascular bed ( Fig. 2) ( 8,9). Clinical studies have demonstrated that abnormalities in protein and cellular hemostatic regulatory elements are associated with coagulation specific to certain vascular beds. For example, congenital deficiencies of antithrombin III, protein C, and protein S are associated with an increased risk of deep vein thrombosis in the lower, not the upper, limbs. Antiphospholipid syndrome has a propensity toward the formation of clots in the retina and placenta. The factor V Leiden mutation, the most common genetic disorder of hemostasis, is predominantly associated with thrombosis of deep veins in the legs and brain, and not in arteries. This mutation does not confer an increased risk of acute myocar­dial infarction in young women unless they smoke, consis­tent with an increased requirement for additional prothrom-botic stimuli to promote thrombosis in the coronary arteries. Paroxysmal nocturnal hemoglobinuria is associated with a very high incidence of thrombosis in large arteries, particu­larly those of the heart and central nervous system. The lo­cal vascular bed predisposition to prothrombotic pathol­ogy also extends to platelet disorders. The microthrombotic lesions characteristic of thrombotic thrombocytopenic pur­pura notably spare the liver and lungs ( 1- 16). Focal thrombosis must therefore be viewed as the re­sult of complex interactions between congenital or acquired defects in the coagulation and fibrinolytic cascades, as well as alterations in platelets and endothelial signaling path­ways. In ordering a " coagulation work- up" in a patient with focal thrombosis, the clinician must consider the site of Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. 50 J Neuro- Ophthalmol, Vol. 23, No. 1, 2003 NANOS SYMPOSIUM JNeuro- Ophthalmol, Vol. 23, No. 1, 2003 *• Plasm in Thrombomodulin Protein S Protein C FIG. 1. The clotting cascade. Coagulation is initiated by the exposure of blood to tissue factor bound to cell mem­branes. Tissue factor interacts with factor Vila to convert factor IX to factor IXa and factor X to factor Xa ( only the activated forms are shown). Factor IXa converts factor X to factor Xa. Factor Xa generates factor 11 a ( thrombin) from factor II ( prothrombin) in a reaction that is accelerated by factor V. Once factor 11 a is generated, it cleaves plasma fibrinogen to generate fibrin monomers, which polymerize and link to form a chemically stable clot. Thrombin also feeds back to activate cofactors VIII and V, thereby ampli­fying the coagulation mechanism. The tissue- factor-pathway inhibitor is a lipoprotein- associated plasma protein that forms a quaternary complex with tissue factor, factor Vila, and factor Xa, thereby inhibiting the extrinsic coagu­lation pathway. The thrombomodulin- protein C- protein S pathway inactivates factors Va and Villa. Antithrombin III inactivates factors Xla, IXa, Xa, and IIa in a reaction that is accelerated by the presence of heparan sulfate. In the fibri­nolytic pathway, tissue- type plasminogen activator ( t- PA) and urokinase- type plasminogen activator ( u- PA) convert plasminogen to plasmin. Once generated, plasmin pro-teolytically degrades fibrin. Reprinted with permission ( 8). Blood vessel of the lung Blood vessel nf the heart C Blood vessel of the liver Blood vessel of the brain FIG. 2. Vascular bed- specific hemostasis. The interaction of various anticoagulant and procoagulant forces promotes overall hemostasis, but the actual components of this interac­tion differ from one vascular bed and organ to another. Thrombomodulin ( TM) is more important in maintaining the hemostatic balance of the lungs and heart ( A and B) than it is in the liver ( C), whereas the fibrinolytic pathway ( tissue- type plasminogen activator [ t- PA] and urokinase- type plasminogen activator [ u- PA] are important in mediating blood fluidity in all three vascular beds ( A, B, and C). Neither thrombomodulin nor fibrinolysis is essential in maintaining balanced hemostasis in the blood vessels of the brain ( D). The physiologically rel­evant natural anticoagulant mechanisms that are operative in this vascular bed have not yet been identified. Reprinted with Copyright © Lippincott Williams & Wilkins. UnauttocnisssrJ ^ production of this article is prohibited. 51 JNeuro- Ophthalmol, Vol. 23, No. 1, 2003 NANOS SYMPOSIUM thrombosis, the type of thrombosis ( arterial, venous, or both), and the size of the vessel involved ( large or small vessel) ( 1,2). COAGULATION ABNORMALITIES A patient who experiences repeated episodes of thrombosis is said to have " thrombophilia" or " hypercoagu­lability." Thrombosis may occur when the naturally-occurring anticoagulant proteins are defective or deficient ( usually an inherited state), or in association with condi­tions that affect vascular integrity, decrease blood flow, or cause an imbalance in blood proteins. Coagulopathies are usually classified as " congenital" or " acquired," although congenital and acquired factors are usually present in the same patient ( Table 1, 2) ( 1- 16). TABLE 1. Congenital and acquired hypercoagulable states Congenital Protein C deficiency Protein S deficiency Antithrombin III deficiency Activated protein C resistance ( Factor V Leiden) Prothrombin gene ( Factor II 20210A) mutation Heparin cofactor II deficiency Dysfibrinogenemia Plasminogen activator inhibitor ( PAI- 1) gene polymorphism Congenital plasminogen deficiency Thrombomodulin gene mutation Sickle cell disease Platelet defects Acquired Antiphospholipid syndrome Myeloproliferative disorder Paroxysmal nocturnal hemoglobinuria Thrombotic thrombocytopenic purpura Disseminated intravascular coagulation Malignancy Sepsis Hyperviscosity syndrome Trauma Immobilization Surgery Pregnancy Oral contraceptives Heparin- induced thrombocytopenia Congenital and acquired combinations Hyperhomocysteinemia Elevated factor VIII levels Elevated fibrinogen levels CONGENITAL ABNORMALITIES The frequency of the major inherited thrombophilias varies substantially within healthy populations and among patients with venous thrombosis. Factor V Leiden and the G20210A mutation in the prothrombin gene are common among healthy whites but are extremely rare among Asians and blacks. The frequency of all inherited thrombophilias is significantly higher in unselected patients with venous thrombosis than in healthy subjects. Since factor V Leiden and the G20210A mutation in the prothrombin gene are relatively common, their coinheritance with other thrombo­philias is not rare ( 6,15). In most patients with inherited thrombophilia, the first thrombotic event occurs before the age of 45 years. The first event occurs even earlier in patients who have more than one inherited thrombophilia or who are homozygous for factor V Leiden or the G20210A mutation in the pro­thrombin gene. Asymptomatic heterozygotes that are rela­tives of index patients with inherited thrombophilias have a significant risk of venous thrombosis. The highest risk, 0.87 to 1.6% per year, has been observed in persons who are heterozygous for antithrombin deficiency, and the lowest, 0.25 to 0.45% o per year, has been seen in persons who are heterozygous for factor V Leiden ( 6,13). Persons who are heterozygous for the G20210 A mutation in the prothrombin gene, or who have protein C or protein S deficiency, have an annual incidence of venous thrombosis of 0.55% ( G20210A), 0.43 to 0.72% ( protein C), and 0.5 to 1.65% ( proteinS) ( 15,16). Activated Protein C Resistance ( Factor V Leiden Mutation) Activated protein C ( APC) resistance is the most common known hereditary predisposition to venous throm­bosis ( 16). It accounts for 20% o of first events in unselected patients, 50% o of events in familial thrombosis, 60% o of events in pregnant women, and 60% o of events in patients known to have normal protein S, protein C, antithrombin, or antiphospholipid antibody levels. This defect is present in about 4.5% of apparently normal populations of whites, and is rare in black and Asian populations ( 13- 18). Activated protein C normally degrades activated fac­tor V and VIII by proteolytic cleavage at specific sites, thereby inhibiting coagulation ( Fig. 1) ( 1- 6,13- 18). Indi­viduals with APC resistance have a mutated factor V that is rendered less sensitive to the natural anticoagulant protein C- protein S system. This mutation leads to a gain in the procoagulant system. More than 95%> of cases are caused by a point mutation, known as the factor V Leiden mutation, at one of the arginine cleavage sites in the factor V gene. Two additional factor V mutations at another arginine cleavage site have recently been reported to be a very rare cause of APC- resistance( 15). Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. 52 © 2003 Lippincott Williams & Wilkins NANOS SYMPOSIUM JNeuro- Ophthalmol, Vol. 23, No. 1, 2003 TABLE 2. Hypercoagulable states and characteristic sites Oj Hypercoagulable state Protein C deficiency Protein S deficiency Antithrombin deficiency Heterozygous Homozygous for mutation of heparin- binding domain Factor V Leiden G20210A mutation in the prothrombin gene High concentration of factor VIII Sickle cell disease ( homozygous) Hyperhomocysteinemia MTHFR mutation ( 3- cystathionase gene synthetase mutation Antiphospholipid antibody syndrome Myeloproliferative diseases Paroxysmal nocturnal hemoglobinuria Disseminated intravascular coagulation Thrombotic thrombocytopenia purpura MTHFR, methylene tetrahydrofolate reductase. The factor V Leiden mutation is usually associated with an increased risk of deep vein thrombosis in the legs and brain ( 1- 6,13- 18). Individuals who are heterozygous for factor V Leiden have a seven- fold greater risk of throm­bosis; homozygous persons have a 20- fold greater risk. The risk of arterial thrombosis with the Factor V Leiden muta­tion is uncertain, but Rosendaal et al ( 13) recently demon­strated a 32- fold increased risk of myocardial infarction in young women smokers and compared with women non-smokers. In non- smoking women, the presence of factor V Leiden did not appear to increase the risk of myocardial infarction over those without the mutation ( 18,19). The high prevalence of this defect also means that patients who are heterozygous for the factor V Leiden mutation are often also heterozygous for other defects, such as antithrombin III, protein C and S, or prothrombin G20210A polymor­phism ( 16). Thus, it is important to perform investigations of other inherited coagulation disorders in patients known to be APC- resistant ( 13). Prothrombin Gene ( 20210A) Mutation A mutation in the 3'- untranslated region of the pro­thrombin ( factor II) gene ( G to A at position 20210) also leads to a gain in procoagulant function, an increased Characteristic sites of thrombosis Deep veins of legs, coumadin- induced skin necrosis Deep veins of legs Deep veins of legs, deep veins and arteries Deep veins of legs and brain, coronary arteries Deep veins of legs and brain, coronary and cerebral arteries Deep veins of legs All vessels, mostly small, arteries > veins All veins and arteries All arteries and veins Portal and hepatic veins Portal and hepatic veins All vessels All micro vessels with the exception of those of the liver and lung plasma concentration of prothrombin, and an increased risk of thrombosis ( 1- 6,13- 17,20- 22). The prothrombin G20210A mutation predisposes pa­tients to deep vein thrombosis, mostly in the legs and brain, and may be a genetic risk factor for both stroke and isch­emic heart disease ( 10,20- 22). Indeed, the risk of venous thrombosis is increased 2.8- fold and the risk of ischemic stroke is increased 3.8- fold in heterozygotes. It is higher in individuals who also carry the factor V Leiden mutation. The level of prothrombin itself also correlates with an in­creased risk of stroke, with an odds ratio of 2.1 for levels above 115% ( 1- 6,10,13- 18,20- 22). Because of its rela­tively high prevalence in the normal population ( 1- 4% of whites), the clinician should be mindful of the possibility of combined defects, especially of factor V Leiden, antiphos­pholipid antibodies, and hyperhomocysteinemia ( 13). Congenital Deficiencies or Defects of Antithrombin III, Protein C, and Protein S Numerous mutations have been described in patients with a deficiency of protein C, protein S, or antithrombin III. Type I defects ( low activity and low antigen level) pre­dominate in patients with a deficiency of protein C or S, whereas either type I or type II ( low activity and normal Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. 53 JNeuro- Ophthalmol, Vol. 23, No. 1, 2003 NANOS SYMPOSIUM antigen level) defects are common in patients with anti-thrombin III deficiency. Congenital deficiencies of antithrombin III, protein C, and protein S are associated with an increased risk of deep vein thrombosis of the lower, but not the upper, limbs. Because antithrombin III, protein C, and protein S are the main natural inhibitors of the procoagulant system, a het­erozygous deficiency of these proteins leads to excessive thrombin formation ( Fig. 1). Deficiency states involving these proteins do not confer a predisposition to arterial thrombosis ( 1- 6,14- 16). One interesting exception is a mu­tation of the heparin- binding site of antithrombin III. In this case, the abnormal antithrombin does not bind to heparin and is therefore much less efficient at inhibiting thrombin and more proximal enzymes of the coagulation cascade. The resulting thrombotic phenotype involves both arteries and veins ( 1- 5). Persons with a homozygous defi­ciency of protein C or S are exceedingly rare; they present soon after birth with purpura fulminans or massive venous thrombosis ( 4,5). Dysfibrinogenemia Dysfibrinogenemia is a rare cause of congenital hy­percoagulability ( Table 3), with an estimated prevalence of 1% among young adults with unexplained thrombosis. Thrombosis, which occurs only in a minority of affected subjects, may either be venous or arterial. Some inherited abnormal fibrinogen states have been associated with bleeding ( 1- 6,11,23). Plasminogen Activator Inhibitor ( PAI- 1) Gene Polymorphism Plasminogen activator inhibitor 1 inhibits tissue plas­minogen. Increased concentrations of plasminogen second­ary to polymorphisms in PAI- 1 gene have been associated with myocardial infarction but not with cerebral arterial or venous stroke ( 11). Congenital Plasminogen Deficiency Congenital plasminogen deficiency is a rare disorder inherited as an autosomal recessive trait. Thrombosis in­volves mainly the veins of the lower extremities ( 1- 6,13). Heparin Cofactor II Deficiency Heparin cofactor II is a plasma glycoprotein that acts as an inhibitor of thrombin. As a natural inhibitor of coagu­lation, its deficiency may promote thrombosis, mostly of veins ( 1- 6,13). Factor VIII Excess High concentrations of clotting Factor VIII are re­lated to an increased risk of venous thrombosis. Concentra­tions of Factor VIII are determined mostly by blood group, which accounts for the old observation of a relation be­tween non- 0 blood groups and the risk of thrombosis ( 1 - 6,13). Other factors such as factors II, V, VII, IX, X, and XI are associated with increased risk for thrombosis. Thrombomodulin Thrombomodulin is an endothelial cell membrane glycoprotein that acts as a thrombin receptor on the endo­thelial cell surface and promotes protein C activation. Vari­ous mutations in the thrombomodulin gene have been re­ported in patients with myocardial infarction, pulmonary embolism, or cerebral venous thrombosis, suggesting that some mutations may represent risk factors for throm­bosis ( 1- 6,13). Sickle Cell Disease Because of its high prevalence among blacks ( 1: 600), sickle cell anemia is a common cause of thrombosis ( 24- 26). Sickle cell- hemoglobin C and sickle cell ( 3- thalasse-mia, which are other common genotypes of sickle cell dis­ease, are jointly as common as sickle cell anemia. Sickle hemoglobin ( hemoglobin S, 2( 32S) accounts for over half the hemoglobin in patients with this disorder. Sickle cell trait, present among 8% of blacks, does not cause throm­botic episodes. Sickle hemoglobin forms polymers when deoxygen-ated ( 24- 26). The delay in forming hemoglobin S polymers is exquisitely dependent on the concentration of hemoglo­bin S. This observation explains why small reductions in the concentration of hemoglobin S might have important clini­cal benefits. In sickle cell trait, the cellular concentration of hemoglobin S is too low for polymerization to occur under most conditions. Vaso- occlusion, responsible for most of the severe complications of sickle cell disease, involves mainly small arteries and less commonly large arteries or veins ( 24- 26). Vaso- occlusion is initiated and sustained by interactions among sickle cells, endothelial cells, and con­stituents of plasma ( Fig. 3) ( 24). No single mechanism ex­plains the vaso- occlusion; its cause may be different from event to event, and its severity differs among patients. Treatment now allows a better and longer life ( 24- 26). Platelet Defects Although platelets play a very important role in the pathogenesis of thrombosis, especially arterial events, little is known about platelet defects predisposing to thrombosis. This may in part be due to the difficult assays for platelet function. The techniques for assessing platelet adhesion and platelet aggregation are labor intensive, expensive, and dif­ficult to standardize ( 11,27). Activated platelets initiate hemostatic plug formation and provide a scaffolding for coagulation activation. Plate­lets are activated by a number of stimuli, such as exposure Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. 54 © 2003 Lippincott Williams & Wilkins TABLE 3. Frequency of congenital hypercoagulable states Z I c in •< Frequency in patients with thrombosis (%); increased risk of thrombosis associated with this defect Hypercoagulable state Protein C deficiency Protein S deficiency Deficiency of antithrombin Factor V Leiden G2021OA mutation in the prothrombin gene Type II heparin- binding mutation ( antithrombin) High concentration of factor VIII ( M5000UI) Sickle cell disease Hyperhomocysteinemia Frequency in the general population (%) 0.14- 0.5 0.7 0.17% 5% ( whites)* 1- 4% ( whites) 0.01% 11% Heterozygous: 8% AA Homozygous: 1/ 600 AA 5- 10% Frequency in patients with thrombosis (%) 3% 1- 2% 1% 20% 6% Very rare 25% ? 10% Heterozygous Increased risk of thrombosis associated with this defect x7 x5 x5 x8 ( x35 with oral contraceptives) x2- 3 fold- venous; x3.8 fold- stroke ? x6 ? x2.5 Homozygous Frequency in patients with thrombosis (%) Very rare ( result of consanguinity) Very rare ( result of consanguinity) Very rare ( result of consanguinity) 1 per 5000 ? Very rare ? High risk - Increased risk of thrombosis associated with this defect Severe thrombosis at birth Severe thrombosis at birth Lethal x80 ( Thrombosis occurs in adulthood) ? ( Thrombosis occurs in adulthood) ? ? Thrombosis in childhood - (> 18.5iunol/ l) MTHFR C677T mutation ( 3- cystathionase gene synthetase mutation High risk Very rare Homocystinuria TO f * Up to 15% in southern Sweden and Middle East Adapted from ( 1). ? JNeuro- Ophthalmol, Vol. 23, No. 1, 2003 NANOS SYMPOSIUM / J- Glabin gene [ • M l codon) T • GAG ( glutamic acid) Hemoglobin S solution -*• GTG ( valine) Hemoglobin S polymer Hemoglobin S cell Cell heterogeneity Sickled Vaso- ocelusion FIG. 3. Pathophysiology of sickle cell disease. In hemoglo­bin S, a substitution of T for A in the sixth codon of the P- globin gene leads to the replacement of a glutamic acid residue by a valine residue. On deoxygenation, hemoglobin S polymers form, causing cell sickling and damage to the membrane. Some sickle cells adhere to endothelial cells, leading to vaso- occlusion. Reprinted with permission ( 24). of vascular endothelium, fibrin deposition, and abnormal surfaces such as atheromata. A number of observations, in­cluding the appearance of platelet thrombi on atheromata, indicate that platelet physiology is relevant to stroke ( 27). Hyperaggregable platelets have been found in a vari­ety of vascular diseases, such as unstable angina, diabetes mellitus, strokes, retinal artery thrombosis, and acute thromboembolic events ( 27). In patients with these disor­ders and hyperaggregable platelets, plasma levels of PF4, beta- thrombomodulin, and thromboxane A2 have also been elevated, suggesting that there is in vivo activation of plate­lets. There are reports of congenital defects in the prosta­glandin pathways associated with thrombosis, especially on the arterial side ( 11,27). The sticky platelet syndrome has been described in more than 200 patients with early- onset arterial occlusive disease, and is occasionally associated with venous throm­bosis ( 28). However, this entity remains debated. ACQUIRED ABNORMALITIES Variation in the concentration of clotting factors can also be explained by acquired factors ( 3,8,9,14). Any con­dition, including liver disease or endothelial dysfunction, that damages the organs where clotting factors are produced or mediated may affect concentrations of these factors, as may dietary intake of substrates and vitamins ( such as vita­min K), age, the use of oral contraceptive agents, tobacco, alcohol, or special circumstances such as pregnancy, meno­pause, surgery, trauma, malignancy, infection, or immobi­lization ( 14). Even when a congenital disorder is the cause of loss of function of a protein involved in the coagulation cascade, acquired factors may further predispose one to clotting. Clear indications of synergistic effects come from studies of thrombophilic families, where high risk has been found in pregnancy and the puerperium, and with the use of oral contraceptive agents ( 13- 15,29- 32). In several series of un-selected women with thrombosis during pregnancy, factor V Leiden was more common than in the general population ( 29- 32). The frequency of factor V Leiden among these women varied widely between studies, from 8% in Scotland to 50 to 60% in Sweden ( which partly reflects the geo­graphical differences in the population prevalence of factor V Leiden). These data suggest that a part of pregnancy-related thrombosis results from concomitant congenital ab­normalities in the hemostatic system ( 13- 15). A similar synergistic effect has been shown for factor V Leiden and the use of oral contraceptive agents. The estimated baseline risk of thrombosis for non- carriers who do not use oral con­traceptives was 0.8 per 10,000 people per year ( 29,30). The annual risk for women with factor V Leiden who did not use oral contraceptives was 5.7 per 10,000 people per year ( relative risk 6.9); for women who used oral contraceptives but did not carry factor V Leiden, it was 3.0 per 10,000 people per year ( relative risk 3.7); for women with factor V Leiden who used oral contraceptives, it was 28.5 per 10,000 people ( relative risk 34.7) ( 29,30). Inpatients with cerebral venous thrombosis, the combination of protein C defi­ciency, factor V Leiden, or prothrombin 20210A with the Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. 56 © 2003 Lippincott Williams & Wilkins NANOS SYMPOSIUM JNeuro- Ophthalmol, Vol. 23, No. 1, 2003 use of oral contraceptive agents led to a 30- fold to 150- fold increased risk, compared with women who did not use oral contraceptive agents and did not have such a defect ( 31,32). These combined risks are much higher than the individual risk conferred by the use of oral contraceptives or a throm-bophilic defect alone ( 14,29- 32). Hyperhomocysteinemia Hyperhomocysteinemia is a good example of an ab­normal plasma factor that results from both genetic and ac­quired factors ( Table 4) ( 10,13,14,33- 36). Homocysteine is an amino acid derived from methionine that can be con­verted into cysteine ( Fig. 4). Metabolic pathways involving homocysteine require vitamin B12, vitamin B6, and folate to function properly. Due to the existence of a cellular ho­mocysteine export mechanism, plasma normally contains a small amount of homocysteine ( averaging 10 mmol/ 1). This export mechanism complements the catabolism of ho­mocysteine through transsulfuration; together these mecha­nisms help maintain low intracellular concentration of this potentially toxic sulfur amino acid. In hyperhomocysteine­mia, plasma homocysteine levels are elevated and, barring impaired renal function, the occurrence of hyperhomocys­teinemia indicates that homocysteine metabolism has been disrupted and that the export mechanism is delivering ex­cess homocysteine into the blood. This export mechanism limits intracellular toxicity, but leaves vascular tissue exposed to possibly deleterious effects of excess homocysteine. Mutations in gene encoding enzymes of homocys­teine metabolism, such as cystathionine B- synthetase or TABLE 4. Classification of hyperhomocysteinemia Severe hyperhomocysteinemia High homocysteine levels at all times. Caused by deficiencies in cystathionine beta- synthetase ( CBS), methylenetetrahydrofolate ( MTHFR), or in enzymes of B12 metabolism. Mild hyperhomocysteinemia Fasting: Moderately high homocysteine levels under fasting conditions. Reflects impaired homocysteine methylation ( folate, B12, or moderate enzyme defects [ e. g. thermolabile methylenetetrahydrofolate]). After methionine load Abnormal increase in homocysteine after methionine load. Reflects impaired homocysteine transsulfuration ( heterozygous cystathionine beta- synthetase defects, B6 deficiency). Copyright © Lippincott Williams & Wilkins. methylene tetrahydrofolate reductase ( MTHFR), lead to in­creased concentrations of homocysteine. For example, con­genital homocystinuria is caused by homozygous defects in the gene encoding cystathionine B- synthetase. In individu­als with such defects, fasting plasma homocysteine concen­trations can be as high as 400 umol/ l. Most individuals with hyperhomocysteinemia, however, do not carry either ge­netic variant, but have impaired methionine metabolism, so that hyperhomocysteinemia is caused by insufficient dietary intake of folic acid, vitamin B6, or vitamin B12 ( 33- 38). Hyperhomocysteinemia has been associated with ve­nous thrombosis in several studies ( 33- 36), and is consid­ered an important risk factor for arterial disease ( 11). Hy­perhomocysteinemia may promote atherosclerosis as well as venous and arterial thrombosis by causing endothelial damage and secondarily increased endothelial tissue factor expression, activation of the coagulation cascade, increased platelet adhesiveness, and the conversion of TDT choles­terol into small, dense, proatherogenic forms ( 11,33- 36). A unique aspect of the hypercoagulability associated with hyperhomocysteinemia is that homocysteine levels can be decreased simply by administering vitamin B12 ( 100 ug daily), vitamin B6 ( 3 mg daily), and folate ( 400 ug daily) ( 37,38). Whether this leads to a concomitant reduction in the risk for arterial or venous thrombosis, and whether such an effect is mediated by lowering the homocysteine level, remains a question that awaits confirmation in randomized trials. However, a recent prospective study ( 38) with a 14- year follow- up showed that women with the highest intake of folate and vitamin B6 had the lowest risk for myocardial infarction and fatal coronary artery disease. Antiphospholipid Antibody Syndrome Antiphospholipid antibodies ( aPT) cause clot forma­tion, particularly within venous and arterial segments of the retina, brain, and placenta ( 39- 49). Antiphospholipid anti­bodies were originally described in patients with systemic lupus erythematosus ( Table 5). Tater, these antibodies were also found in the absence of a systemic autoimmune disor­der ( primary antiphospholipid antibody syndrome). To fa­cilitate studies, preliminary criteria for classification of the antiphospholipid antibody syndrome were recently pub­lished ( Table 6) ( 40). The antiphospholipid antibody ( Table 7, 8) syndrome is present if there is one positive aPT test- anticardiolipin ( aCT) or lupus anticoagulant ( TAC)- according to defined laboratory criteria, in at least two blood samples taken at least 6 weeks apart, together with one or more clinical episodes of arterial, venous, or small vessel thrombosis in any tissue or organ ( 40). Thromboembolic events in relation to aPT have been described for almost any vessel in the body, but the most frequently occurring events are thrombosis of deep leg thorized reproduction of this article is prohibited. 57 JNeuro- Ophthalmol, Vol. 23, No. 1, 2003 NANOS SYMPOSIUM MethyleneTHF Glycine MelhyllHF • Cysteine Alplia- Ketobutyrate Methyl acceptors Phosphotidyl ethanolamine Neurotransmitters ( dopamine, etc) Guanidoacetate Prcteins ( myelin, etc) DMA RNA Methylated acceptors Phtwphotidy] choline Methylated neurotransmitters Creatine Methylated proteins Methylated DNA Methylated UNA FIG. 4. Homocysteine metabolic path­ways. Homocysteine is a nonprotein- form-ing sulfur amino acid whose metabolism is at the intersection of two metabolic path­ways: remethylation and transsulfuration. In the methylation pathway ( upper pathway), homocysteine acquires a methyl group either from betaine ( a reaction that occurs mainly in the liver), or from 5- meth-yltetrahydrofolate ( a reaction that occurs in all tissues and is vitamin B12- dependent). In the transsulfuration pathway ( lower path­way), homocysteine condenses with serine to form cystathionine in a reaction cata­lyzed by cystathionine beta- synthetase ( CBS) and requiring pyrodoxal- 5'- phosphate ( PLP). Methionine is the sole source of homocysteine. veins, pulmonary emboli, ischemic stroke and transient ischemic attacks ( 39). Antiphospholipid antibodies are a heterogenous fam­ily of antibodies with diverse cross reactivity ( 39,41,45- 47). Anticardiolipin antibodies ( aCLs) are detected by stan­dard enzyme- linked immunoabsorbent assay ( ELISA); the lupus anticoagulant ( LA) prolongs phospholipid- dependent coagulation assays. These assays are the best characterized and standardized ( 40). There is partial concordance be­tween the two assays, but the LA assay is more specific for patients at risk for thromboembolic events ( 40,41). In con­trast, the aCL assay is more sensitive but less specific, and positivity can be found in various contexts ranging from health to certain medications, malignancies, and infectious diseases ( Table 5). The isotype usually implicated in throm­bosis is IgG, more specifically IgG2 ( 39,46). Recent data suggest that the presence of high titers of aCL immunore-activity, mainly IgG isotype, but also possibly IgM, corre­late with an increased risk of thrombosis ( 39,46). Generally, the titers of IgG aCL deemed pathologic are greater than 40 GPL, although this is a somewhat arbitrary cutoff point and is dependent on the test systems ( 39,40,42). Data accumulated over the past few years have radically changed our understanding of the antigenic speci­ficities of the autoantibodies associated with the antiphos­pholipid syndrome and the pathogenetic mechanisms asso­ciated with these antibodies. The concept of a protein target for aPLs evolved from a series of reports in 1990 that iden­tified ( 32- glycoprotein I (( 32- GPI; also named apolipopro-tein H) as a necessary plasma cofactor to bind cardiolipin in vitro on ELISA plates. ( 32- GPI has several anticoagulant functions and anti-( 32- GPI antibodies can help differentiate between autoimune aCLs that require ( 32- GPI and " benign" alloimmune aCLs that do not ( 39,45- 47). Anti-( 32- GPI an­tibodies were shown to be more specific for thrombosis than conventional aCLs ( 47) and can occasionally be the TABLE 5. Antiphospholipid antibody syndromes Primary antiphospholipid syndrome Secondary antiphospholipid syndrome Rheumatologic disorders Systemic lupus erythematosus Dermatomyositis, polymyositis Sjogren syndrome Rheumatoid arthritis Systemic sclerosis Temporal arteritis Psoriasis arthropathy Behcet syndrome Others Infections Viral Bacterial Parasitic Lymphoproliferative disorders Lymphoma Paraproteinemias Medication exposure Phenothiazines Quinidine Hydralazine Procainamide Phenytoin Miscellaneous Intravenous drug abuse Sickle cell disease Guillain- Barre syndrome Adapted from ( 39). Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. 58 © 2003 Lippincott Williams & Wilkins NANOS SYMPOSIUM JNeuro- Ophthalmol, Vol. 23, No. 1, 2003 TABLE 6. Criteria for the classification of the antiphospholipid syndrome * Clinical criteria 1. Vascular thrombosis One or more episodes of arterial, venous, or small vessel thrombosis, in any tissue or organ. Thrombosis must be confirmed by imaging or Doppler studies or histopathology, with the exception of superficial venous thrombosis. For histopathologic confirmation, thrombosis should be present without significant evidence of inflammation in the vessel wall. 2. Pregnancy morbidity One or more unexplained deaths of a morphologically normal fetus at or beyond the 10th week of gestation, with normal fetal morphology documented by ultrasound or by direct examination of the fetus, or One or more premature births of a morphologically normal neonate at or before the 34th week of gestation because of severe preeclampsia, or severe placenta insufficiency, or Three or more unexplained consecutive spontaneous abortions before the 10th week of gestation, with maternal anatomic or hormonal abnormalities and paternal and maternal chromosomal causes excluded. Laboratory criteria 1. Anticardiolipin antibodies of IgG and/ or IgM isotype in blood in medium or higher titer, on two or more occasions, at least 6 weeks apart, measured by a standardized enzyme- linked immunosorbent assay for ( 32- glycoprotein I- dependent anticardiolipin antibodies. 2. Lupus anticoagulant present in plasma on two or more occasions at least 6 weeks apart, detected according to the guidelines of the International Society on Thrombosis and Hemostasis ( 40), in the following steps: Prolonged phospholipid- dependent coagulation demonstrated on a screening test, e. g., activated partial thromboplastin time, kaolin clotting time, dilute Russell viper venom time, dilute prothrombin time, Textarin time. Failure to correct the prolonged coagulation time on the screening test by mixing with normal platelet- poor plasma. Exclusion of other coagulopathies, e. g. factor VIII inhibitor or heparin, as appropriate. * Definite diagnosis if at least one of the clinical criteria and one of the laboratory criteria are met. Adapted from Wilson ( 40). only positive assay associated with the antiphospholipid syndrome ( 47). ELISA kits for antibodies against ( 32- GPI are currently available and FDA- approved. Antiphospholipid antibodies occur in 3 to 5% of the general population. The rate of thrombosis in persons with an antiphospholipid antibody in a prospective study ( 43) was 1% per year in individuals with no history of thrombo­sis, 4% per year in patients with systemic lupus erythema­tosus, 5.5% per year in patients with a history of thrombo­sis, and 6% per year in individuals with a high- titer IgG anticardiolipin antibody ( exceeding 40 GPL units). Cur­rently, aCL testing and evaluation for the LA [ following specific criteria ( 40)] are the recommended screening tests. These tests are useful only in the appropriate clinical set­ting. Testing for the antibody is valid only if performed at least 6 weeks after the coagulative event. Data based on patients with the antiphospholipid syn­drome suggest that there remain approximately 10 to 15% of patients who, despite presenting with the clinical picture of the antiphospholipid syndrome, have negative tests for aCL and LA. Thus, in patients for whom there is high clini­cal suspicion, further testing is indicated for antibodies to TABLE 7. Recommended coagulation test panel Routine panel CBC, differential, platelets Fibrinogen Serum protein electrophoresis PT, PTT Protein C ( functional protein C assay) Protein S ( free protein S antigen) Antithrombin ( functional antithrombin assay) Activated protein C resistance and factor V Leiden mutation Prothrombin G20210A mutation Antiphospholipid antibodies Lupus anticoagulant, anticardiolipin antibodies ( IgG, IgM) If strong clinical suspicion of antiphospholipid syndrome, include anti( 32- GPI Homocysteine If homocysteine is elevated, test folate, B12, creatinine, MTHFR mutation Hemoglobin electrophoresis ( in blacks) Additional tests ( order when initial panel is negative and clinical suspicion for a hypercoagulable state remains high) Work- up for chronic compensated disseminated intravascular coagulation Work- up for sepsis and cancer Tests for dysfibrinogenemia Plasminogen antigen and activity Isolated measurement of clotting factors Platelet aggregation tests Confirmatory panel Immunological assay of protein C, S, antithrombin III as appropriate Repeat lupus anticoagulant and anticardiolipin antibodies after 6 weeks MTHFR, methylene tetrahydrofolate reductase. Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. 59 JNeuro- Ophthalmol, Vol. 23, No. 1, 2003 NANOS SYMPOSIUM TABLE 8. Tests affected by anticoagulants or recent thrombosis Heparin Decreases antithrombin Interferes with some lupus anticoagulant tests Warfarin Decreases proteins C and S Recent thrombosis (< 10 days) Consumes antithrombin and proteins C and S Increases factor VIII ( acute phase reactant) [ 32- GPI or to noncardiolipin phospholipids ( 39,40,45- 47). Antibodies against ( 32- GPI or other specific proteins may be used as more specific confirmatory tests in patients with positive aCL and potentially related thromboocelusive events. It is important for the clinician to be familiar with the test systems used by local or reference laboratories, and to be aware that none of these tests is yet standardized ( 40). Myeloproliferative Diseases Myeloproliferative diseases are characterized by an unusually high incidence of thrombosis in the hepatic, por­tal, and mesenteric veins. Polycythemia vera and essential thrombocytosis are also commonly associated with cerebral venous thrombosis and central retinal vein occlusions, and, more rarely, with arterial occlusions involving the brain, the eyes, and the heart ( 1,6). Paroxysmal Nocturnal Hemoglobinuria Paroxysmal nocturnal hemoglobinuria, which is characterized by intravascular hemolysis and episodes of hemoglobinuria and life- threatening venous thrombosis, is a rare acquired clonal defect in a single hematopoietic stem cell ( 48,49). Thrombosis in hepatic, portal, mesenteric, or cerebral veins is one of the most serious complications, oc­curring in about one- third of patients ( 48,49). The median survival is 10 years, and only about a quarter of patients remain alive more than 20 years after diagnosis ( 48,49). Thrombotic Thrombocytopenic Purpura Thrombotic thrombocytopenic purpura and the he­molytic uremic syndrome produce microthrombotic lesions that are detectable in all organs, with the notable exception of the liver and the lungs. Cerebral microthrombi are common ( 2). Disseminated Intravascular Coagulation Disseminated intravascular coagulation ( DIC) is also a classic cause of diffuse microthrombi involving both ar­teries and veins. It is most commonly associated with malig­nant disease and sepsis, but its causes are multiple ( 2,50,51). Malignancy Malignancy has been associated with thrombosis for over 100 years, and thrombosis may be its first sign. There is also evidence that the hemostatic system may be involved in the spread of malignant disease. Hemostatic abnormali­ties such as compensated, chronic DIC, or fibrinolytic acti­vation, are extremely common in such patients, and are usu­ally difficult to diagnose ( 51). Sepsis Sepsis is associated with hematologic changes pri­marily involving leukocytes, platelets, and the hemostatic system. Thrombosis is common and is secondary to DIC. Most patients have compensated chronic DIC, which is usu­ally not identified with the use of routinely available test procedures. This form of DIC can only be diagnosed with the aid of newer assay techniques ( 50,52). TABLE 9. Coagulation work- up: recommendations 1 2 3 4 5 6 7 Testing for coagulopathies is costly and should be performed only in patients who meet specific criteria, such as repeated thrombotic events, family history of thrombophilia, first episode of thrombosis before the age of 45 or unusual site of thrombosis. Because of synergy among risk factors for thrombosis, work- up should include screening for the most common hereditary hypercoagulable states, as well as thorough clinical evaluation of acquired risk factors for thrombosis. Coagulation testing should not be performed in a patient treated with anticoagulants. These tests should be delayed and obtained after heparin or warfarin therapy has been discontinued for at least 2 weeks. Platelet function testing should not be obtained in a patient treated with antiplatelet agents. Most coagulation tests should not be performed within 6 weeks of an episode of thrombosis. Results of coagulation testing should be interpreted with caution since " normal values" may vary from one laboratory to another. Clinicians ordering such tests should ascertain that the laboratory techniques are standardized. All abnormal results of coagulation testing for thrombophilia should be accompanied by a consult with a pathologist. Indeed, some tests may need to be repeated, family members may benefit from genetic counseling, and long- term treatment with anticoagulants or antiplatelet agents may be required. Specific counseling may also be necessary regarding the use of contraceptive agents and the management of pregnancy. The " coagulation panel" should be adapted to the type of thrombosis ( arteries/ veins; large/ small vessels) and the site of thrombosis ( see Table 7). Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. 60 © 2003 Lippincott Williams & Wilkins NANOS SYMPOSIUM JNeuro- Ophthalmol, Vol. 23, No. 1, 2003 Hyperviscosity Syndrome Hyperviscosity syndrome is common in patients with Waldenstrom macroglobulinemia, multiple myeloma, and hyperleukocytosis. It has been associated with recurrent vein thrombosis, commonly involving the brain and the eyes ( 1). CONCLUSIONS The growing number of defined hypercoagulable states makes it increasingly likely that laboratory testing will identify a predisposing condition in a patient with thrombosis. Because thrombosis is a disease in which ge­netic and acquired risk factors interact dynamically, physi­cians must perform a thorough history, family history, and physical examination before ordering an extensive and costly coagulation panel ( 53). The work- up of a patient with thrombosis does not lend itself to a uniform approach. The most important fac­tors to consider before ordering a " routine panel" are a fam­ily history of thrombosis and the organ and type of blood vessel involved ( Tables 2 and 9) ( 53- 56). Because patients with thrombophilia often have defects or deficiencies in multiple proteins, a large screening panel is necessary to evaluate the genetic risk and to identify acquired factors. The timing of the screening evaluation in relation to the clinical event is critical in interpreting results. For example, protein C, protein S, and antithrombin III may be consumed in a thrombotic event, so that testing should be delayed for at least 1 to 2 weeks after the event. Given that anticardio-lipin antibodies may be falsely high or low after a throm­botic event, retesting should be performed at least 6 weeks after the clinical event ( Table 8). REFERENCES 1. Biousse V. Coagulation abnormalities and their neuro- ophfhal-mologic manifestations. Curr Op Ophthalmol 1999; 10: 382- 93. 2. Thomas DP, Roberts HR. Hypercoagulability in venous and arterial thrombosis. Ann Intern Med 1997; 126: 638- 44. 3. De Stefano V, Finazzi G, Mannucci PM. Inherited thrombophilia: pathogenesis, clinical syndromes and management. Blood 1996; 87: 3531^ 14. 4. Lane DA, Mannucci PM, Bauer KA, et al. Inherited thrombophilia: parti. ThrombHaemost 1996; 76: 651- 62. 5. Lane DA, Mannucci PM, Bauer KA, et al. Inherited thrombophilia: part II. Thromb Haemost 1996; 76: 824- 34. 6. Federman DG, Kirsner RS. An update on hypercoagulable disor­ders. Arch Int Med 2001; 161: 1051- 6. 7. de Lage M, Snieder H, Ariens RA, et al. The genetic of haemostasis: a twin study. Lancet 2001; 357: 101- 5. 8. Rosenberg RD, Aird WC. Vascular- bed- specific hemostasis and hypercoagulable states. N Engl J Med 1999; 340: 1555- 64. 9. Edelberg JM, Christie PD, Rosenberg RD. Regulation of vascular bed- specific prothrombotic potential. Circ Res 2001; 89: 117- 24. 10. Sykes TC, Fegan C, Mosquera D. Thrombophilia, polymorphisms, and vascular disease. MolPathol 2000; 53: 300- 6. 11. Williams MS, Bray PF. Genetics of arterial prothrombotic states. Exp Biol Med 2001 ; 226: 409- 19. 12. Wu AH, Tsongalis GJ. Correlation of polymorphisms to coagula­tion and biochemical risk factors for cardiovascular diseases. Am J Cardiol 2001; 87: 1361- 6. 13. Rosendaal FR. Venous thrombosis: a multicausal disease. Lancet 1999; 353: 1167- 73. 14. Bertina RM, Rosendaal FR. Venous thrombosis: the interaction of genes and environment. N Engl J Med 1998; 338: 1840- 1. 15. Seligsohn U, Lubetsky A. Genetic susceptibility to venous throm­bosis. N Engl J Med 2001; 344: 1222- 31. 16. Martinelli I, Mannucci PM, De Stefano V, et al. Different risks of thrombosis in four coagulation defects associated with inherited thrombophilia: a study of 150 families. Blood 1998; 92: 2353- 8. 17. Price DT, Ridker PM. Factor V Leiden mutation and the risks for thromboembolic disease: a clinical perspective. Ann Intern Med 1997; 127: 895- 903. 18. De Stefano V, Chiusolo P, Paciaroni K, et al. Epidemiology of fac­tor V Leiden: clinical implications. Sem Thromb Hemostasis 1998; 24: 367- 79. 19. Rosendaal FR, Siscovick DS, Schwartz SM, et al. Factor V Leiden ( resistance to activated protein C) increases the risk of myocardial infarction in young women. Blood 1997; 89: 2817- 21. 20. Margaglione M, Brancaccio V, Giuliani N, et al. Increased risk for venous thrombosis in carriers of the prothrombin GA20210 gene variant. Ann Intern Med 1998; 129: 89- 93. 21. Rosendaal FR, Siscovick DS, Schwartz SM, et al. A common pro­thrombin variant ( 20210 G to A) increases the risk of myocardial infarction in young women. Blood 1997; 90: 1747- 50. 22. De Stefano V, Chiusolo P, Paciaroni K, et al. Prothrombin G20210A mutant genotype is a risk factor for cerebrovascular dis­ease in young patients. Blood 1998; 91: 3562- 5. 23. Roberts HR, Stinchcombe TE, Gabriel DA. The dysfibrinoge-naemias. Br JHematol 2001; 114: 249- 57. 24. Steinberg MH. Management of sickle cell disease. N Engl J Med 1999; 340: 1021- 30. 25. Ohene- Frempong K, Weiner SJ, Sleeper LA, et al. Cerebrovascular accidents in sickle cell disease: rates and risk factors. Blood 1998; 91: 288- 94. 26. Adams RJ, McKie VC, Hsu L, et al. Prevention of a first stroke by transfusions in children with sickle cell anemia and abnormal re­sults on transcranial doppler ultrasonography. N EnglJ Med 1998; 339: 5- 11. 27. del Zoppo GJ. The role of platelets in ischemic stroke. Neurology 1998; 51: S9- 14. 28. Mammen EF. The sticky platelet syndrome. Sem Thromb Hemost 1999; 25: 361- 5. 29. Vandenbroucke JP, Koster T, Briet E, et al. Increased risk of venous thrombosis in oral- contraceptive users who are carriers of factor V Leiden mutation. Lancet 1994; 344: 1453- 7. 30. Vandenbroucke JP, Rosing J, Bloemenkamp KW, et al. Oral con­traceptives and the risk of venous thrombosis. N EnglJ Med 2001; 344: 1527- 35. 31. De Bruijn SF, Stam J, Koopman MM, et al. Case control study of risk of cerebral sinus thrombosis in oral contraceptive users who are carriers of hereditary prothrombotic conditions. BMJ 1998; 316: 589- 92. 32. Martinelli I, Sacchi E, Landi G, et al. High risk of cerebral- vein thrombosis in carriers of a prothrombin- gene mutation and in users of oral contraceptives. N Engl J Med 1998; 338: 1793- 7. 33. den Heijer M, Rosendaal FR, Blom HJ, et al. Hyperhomocysteine-mia and venous thrombosis: a meta- analysis. Thromb Haemost 1998; 80: 874- 7. 34. Ridker PM, Manson JE, Buring JE, et al. Homocysteine and risk of cardiovascular disease among postmenopausal women. JAMA 1999; 281: 1817- 21. 35. Brattstrom L, Wilcken DE, Ohrvik J, et al. Common mefhylenetet-rahydrofolate reductase gene mutation leads to hyperhomocysteine-mia but not to vascular disease: the result of a meta- analysis. Cir­culation 1998; 98: 2520- 6. 36. Selhub J, D'Angelo A. Relationship between homocysteine and thrombotic disease. Am JMedSci 1998; 316: 129^ 11. Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. 61 JNeuro- Ophthalmol, Vol. 23, No. 1, 2003 NANOS SYMPOSIUM 37. Rimm EB, Willet WC, Hu FB, et al. Folate and vitamin B6 from diet and supplements in relation to risk of coronary heart disease among women. JAMA 1998; 279: 359- 64. 38. Jacques PF, Selhub J, Bostom AG, et al. The effect of folic acid fortification on plasma folate and total homocysteine concentra­tions. N Engl J Med 1999; 340: 1449- 54. 39. Greaves M. Antiphospholipid antibodies and thrombosis. Lancet 1999; 353: 1348- 53. 40. Wilson WA, Gharavi AE, Koike T, et al. International consensus statement on preliminary classification criteria for definite an­tiphospholipid syndrome: report of an international workshop. Ar­thritis Rheum 1999; 42: 1309- 11. 41. Triplett DA. Antiphospholipid antibodies and thrombosis. A conse­quence, coincidence, or cause? Arch PatholLab Med 1993; 117: 78- 88. 42. Levine SR, Salowitch- Palm L, Sawaya KL, et al. IgG anticardio-lipin antibody titer > 40 GPL and the risk of subsequent fhrombo-occlusive events and death: aprospective cohort study. Stroke 1997: 28: 1660- 5. 43. Finazzi G, Brancaccio V, Moia M, et al. Natural history and risk factors for thrombosis in 360 patients with antiphospholipid anti­bodies: a four- year prospective study from the Italian registry. Am J Med 1996; 100: 530- 6. 44. Tanne D, D'Olhaberriague L, Schultz LR, et al. Anticardiolipin an­tibodies and their associations with cerebrovascular risk factors. Neurology 1999; 52: 1368- 73. 45. Levine SR, Jacobs BS. 2001: a prospective, seasonal odyssey into antiphospholipid protein antibodies. Stroke 2001; 32: 1699- 1700. 46. Tanne D, Triplett DA, Levine SR. Antiphospholipid- protein anti­bodies and ischemic stroke: not just cardiolipin any more. Stroke 1998; 29: 1755- 8. 47. Brey RL, Abbott RD, Curb JD, et al. Beta2 Glycoprotein 1- depen­dent anticardiolipin antibodies and the risk of ischemic stroke and myocardial infarction: the honolulu heart program. Stroke 2001: 32; 1701- 6. 48. Rosse WF. Paroxysmal nocturnal hemoglobinuria as a molecular disease. Medicine 1997; 76: 63- 93. 49. Hillmen P, Lewis SM, Bessler M, et al. Natural history of paroxys­mal nocturnal hemoglobinuria. N Engl J Med 1995; 333: 1253- 8. 50. Bick RL. Disseminated intravascular coagulation: pathophysiologi­cal mechanisms and manifestations. Sem ThrombHemostasis 1998; 24: 3- 18. 51. Francis JL, Biggerstaff J, Amirkhosravi A. Hemostasis and malig­nancy. Sem Thromb Hemostasis 1998; 24: 93- 109. 52. Mammen EF. The hematological manifestations of sepsis. J Anti­microbial Chemotherapy 1998; 41( suppl A): 17- 24. 53. Francis JL. Laboratory investigation of hypercoagulability. Semin Thromb Hemost 1998; 24: 111- 26. 54. McKenzie SB, Clare CN, Smith LA, et al. Laboratory test utiliza­tion in the diagnosis of hypercoagulability. Clin Lab Sci 2000; 13: 215- 21. 55. Quinn C, Hill J, Hassouna H. A guide for diagnosis of patients with arterial and venous thrombosis. Clin Lab Sci 2000; 13: 229- 38. 56. Marques MB, Triplett DA. When to suspect hypercoagulability and how to investigate it. Ann Diagn Pathol 2001; 5: 177- 83. Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. 62 © 2003 Lippincott Williams & Wilkins