Venous thromboembolism: The intricaciesTK Dutta, V Venugopal
Department of Medicine, Jawaharlal Institute of Postgraduate Medical Education and Research, Pondicherry 605 006, India
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0022-3859.48442
Source of Support: None, Conflict of Interest: None
Venous thromboembolism (VTE) has been a subject of great interest of late. Since Rudolph Virchow described the famous Virchow's triad in 1856, there have been rapid strides in the understanding of the pathogenesis and factors responsible for it. Discovery of various thrombophilic factors, both primary and acquired, in the last 40 years has revolutionized prognostication and management of this potentially life-threatening condition due to its associated complication of pulmonary thromboembolism. Detailed genetic mapping and linkage analyses have been underlining the fact that VTE is a multifactorial disorder and a complex one. There are many gene-gene and gene-environment interactions that alter and magnify the clinical picture in this disorder. Point in case is pregnancy, where the risk of VTE is 100-150 times increased in the presence of Factor V Leiden, prothrombin mutation (Prothrombin 20210A) and antithrombin deficiency. Risk of VTE associated with long-haul air flight has now been well recognized. Thrombotic events associated with antiphospholipid syndrome (APS) are 70% venous and 30% arterial. Deep venous thrombosis and pulmonary embolism are the most common venous events, though unusual cases of catastrophes due to central vein thrombosis like renal vein thrombosis and Budd-Chiari syndrome (catastrophic APS) may occur.
Keywords: Activated protein C resistance, air travel, antiphospholipid antibody syndrome, antithrombin, Factor V leiden, protein C deficiency, protein S deficiency, thrombophilia.
The coagulation cascade is an intrinsic protective mechanism, which is responsible for limiting blood loss by precisely regulating interactions between components of vessel wall, platelets and plasma proteins.  However, unregulated activation of this cascade can have potentially deleterious consequences in the form of venous and arterial thrombosis, which may have life-threatening complications.
The incidence of venous thromboembolism (VTE) is estimated to be around one in 1000 as per the Western literature. It increases with increasing age to as much as one in 100.  It is responsible for 50,000 deaths annually in the United States alone. Virchow's paper in 1856 outlined the triad, which even till date forms the basis of our understanding of venous thrombosis.  The triad composed of the following factors, which contributed to formation of venous thrombosis-an endothelial lesion, venous stasis and a hypercoagulable state.
Conventional risk factors for venous thrombosis like pregnancy, immobilization, surgery, trauma etc. are operative by influencing one or all of the above three variables.
The hypercoagulable state has recently been a subject of intense scrutiny due to recognition of a new risk factor i.e. air travel. VTE has been reported with increased occurrence during long-haul air flights e.g. trans-Atlantic flights. 
Familial aggregation of thrombosis points to heritable conditions that may be predisposing to hypercoagulability. The identification of this group of disorders has reduced the proportion of thrombotic disorders, which were hitherto termed idiopathic.  The thrombosis in these conditions manifests typically as recurrent venous thrombosis often in the younger age group individuals, at atypical sites or with life-threatening features. This hypercoagulable state has been termed thrombophilia.
The thrombophilia is used to describe the familial or acquired disorders of the hemostatic system, which are likely to predispose to thrombosis [Figure 1].  Secondary or acquired thrombophilia refers to those systemic diseases that have been known and proven to be associated with thrombosis-like malignancy, paroxysmal nocturnal hemoglobinuria, myeloproliferative disorders etc.  JAK2 mutation is associated with a severe form of central venous thrombosis, namely Budd-Chiari syndrome More Details. Inherited thrombophilias, that are not associated with any other systemic illness, are also called primary thrombophilias. 
Detailed genetic mapping and linkage analyses have been underlining the fact that thrombophilia is a multifactorial disorder and a complex one. The scenario is further complicated by the fact that there are many gene-gene and gene-environment interactions that alter the clinical picture in these disorders. While studies from the West have focused attention on inherited thrombophilias, there is a dearth of similar studies from India. More and more thrombotic disorders, which perhaps went undetected earlier due to lack of investigational facilities, are being recognized now in India. Most setups in the country, however, still label a large proportion of thrombosis as 'idiopathic'. This paper is a review on the pathogenesis of VTE with current information on it.
In 1856, Rudolph Virchow conducted autopsies on 11 cases of fatal pulmonary embolism and identified a source in the iliofemoral veins. His findings subsequently led to the definition of the triad, which has got his name-endothelial lesion, venous stasis and a hypercoagulable state.  This work was central to the evolution of the concept of thrombo-hemorrhagic balance. This theory states that there is a fine balance between fibrin degradation and fibrin production and dominance of the latter leads to a hypercoagulable state. The earliest support for this hypothesis came with Egeberg's discovery of partial deficiency of antithrombin III in a Norwegian family in 1965.  This was followed by several other discoveries over the years [Table 1].
Environmental risk factors for venous thrombosis:
The various risk factors are enlisted below:
Pregnancy and puerperium
Drugs- e.g. estrogens and oral contraceptive pills, thalidomide, heparin
Surgery is an important clinical situation, which has been inconclusively linked to VTE over the years. Patients are at risk, not in the immediate postoperative, but for several months afterwards. The factors operative in thrombogenesis related to surgery are-trauma to major vessels, prolonged immobilization and anesthetic agents. The type of surgery also is a predictor of thrombotic risk with major abdominal and orthopedic surgery associated with maximum risk. 
In hip and knee surgery, the risk of thrombosis reaches 30-50%, while abdominal and gynecological surgeries, especially prostatectomy, are associated with a 30% risk.  Minor surgery lasting less than 30 min is associated with a low risk of deep vein thrombosis, the incidence being around 10%. Major trauma is another potent risk factor for deep vein thrombosis (DVT). Pulmonary embolism was the third most common cause of death in victims of significant trauma surviving more than 24 h, as reported by Clagett et al . 
Travel has been emphasized to be associated with DVT, of late. It is thought that venous stasis is increased by the high pressures from the edge of a person's seat on the back of the calves with prolonged sitting, which decreases venous flow and gradually increases the hematocrit with a concomitant rise in plasma protein concentration.  In a study in 1997, Ferrari et al., identified DVTs in journeys lasting more than four hours within previous two months in 40 (15.5%) of 258 patients with DVT.  In a later study, the same authors attributed an odds ratio of 3.98 for the development of venous thrombosis in patients with a recent history of travel. Modes of travel varied from road to air. Mechanisms postulated include hemoconcentration, induced reduction in spontaneous endothelial fibrinolysis, dehydration, low humidity and immobilization. , Reports over the years have found 18- 36% of DVTs related to air travel, with a few cases of thrombosis of even the subclavian and cerebral veins. 
Obesity has been occasionally reported to be associated with DVT. Various epidemiological studies have attributed relative risks of 2.9 to 3.92 for obesity in causation of DVT.  Immobilization is one of the most important precipitating factors in the causation of DVT. Primary and secondary thrombophilias as well as thrombogenic states like pregnancy, surgery and fractures often manifest their thrombogenic effects during periods of immobilization. 
Pregnancy has been identified as a hypercoagulable state for many decades. There is an increased incidence of both arterial and venous thrombosis during pregnancy and puerperium. Each of the components of Virchow's triad is present in pregnancy. Numerous changes in the coagulation system, which are seen during pregnancy, include increased fibrinogen and clotting factors, inhibition of fibrinolysis and raised levels of platelet activating factors I and II. Stasis develops in late pregnancy due to progesterone-related increase in venous capacitance added to inferior vena compression by the gravid uterus.  Koster et al. , in their report from the Leiden thrombophilia study stated that pregnancy confers a relative risk of 4 for development of VTE.  The thrombotic risk in puerperium is even greater and is estimated to be approximately 2.3-6.1 per 1000 women amounting to a relative risk of 3-5.  Taking the relative shortness of its duration into account, puerperium has been estimated to be 20 to 30 times more thrombogenic than the general population. 
Estrogens and oral contraceptive pills, thalidomide, heparin
Estrogens have been identified as important agents in the causation of thrombosis in young women in the reproductive age group. Their role in venous thrombosis is well established. Oral contraceptive pills (OCPs) are the most important cause of acquired thrombophilia in the West. However, this factor is not of much importance in our population. Especially in the lower socioeconomic classes, tubectomy or intrauterine contraceptive device is preferred over OCPs as means of contraception.
The thrombogenicity of the pill was first reported by Jordan in 1961 who reported a case of pulmonary embolism in a nurse who had just started taking OCPs. Early studies in the 1970s and 1980s reported oral contraceptive-induced relative risks of 4 to 11 for development of VTE.  However, studies over the last decade like the International WHO study (1995), Transnational study (1996) and the Leiden Thrombophilia study (1994) have reported risks of 4-4.5. ,,
Of late, renewed use of thalidomide for hematologic malignancies has been associated with incidence of venous thrombosis in several instances, to an extent that the drug has to be stopped or simultaneous anticoagulants have to be added. 
Thrombosis is a well-known complication of unfractionated heparin; and this complication is much less with low molecular weight heparin.  Heparin-induced thrombocytopenia (HIT) with thrombosis is a thrombogenic state caused by antibodies formed after exposure to heparin. These antibodies bind to the heparin-platelet Factor 4 (PF4) complex and secondarily promote thrombosis in vivo in both venous and arterial sites by aggregating platelets (HIT Type II). Venous thrombosis predominates arterial thrombosis (5:1). HIT is often a medical emergency and the mortality rate is 20%. In general, platelet counts begin to decrease five to nine days after heparin administration is initiated. ,
Prior history of DVT/pulmonary embolism (PE) remains a strong risk factor for new VTE disease. While Ferrari's study in the French population reported a past history in 21% of cases,  a prospective study by Prandoni et al. , revealed a 22% recurrence after cessation of three months of anticoagulant.  Heit et al. , found that this risk was highest immediately after hospitalization and gradually declined to a plateau at three years. The cumulative probability of recurrence at 10 years was 30.4%.  Underlying risk factors and age are associated with different rates of recurrence; for example, malignancy confers a high risk of recurrence while temporary risk factors like trauma, pregnancy etc. are associated with a lower risk. 
Inherited primary abnormalities of the hemostatic system that predispose to thrombosis are called primary thrombophilias or inherited thrombophilias. Since 1965, when antithrombin deficiency was the first primary thrombophilia to be reported, many other thrombophilias have been described till date. The discovery of activated protein C resistance and the Factor V Leiden mutation in 1995 has resulted in a profusion of studies on the subject with newer discoveries occurring at a frantic pace. This has made thrombophilia an ever-changing field.
The state of hypercoagulability results from alterations in the anticoagulant mechanisms, which prevent the production of thrombin and its progression to active thrombosis. There are two major anticoagulant systems  [Figure 2]:
I. Systems that attenuate the rate of prothrombin conversion
II. Inactivation of thrombin in the blood
The substances in the thrombin inactivation pathway include:
Antithrombin, by itself a slow inhibitor of thrombin inactivates the latter 1000 times faster in the presence of heparin. Antithrombin binds heparin irreversibly and prevents the action of Factors V, VIII, XIII and platelets on thrombin production. The underlying defects in certain thrombophilias have been conclusively proven. Their incidence is mentioned in [Table 2].  The important ones, with the data supporting their association with thrombotic risks, are reviewed below.
Protein C deficiency
Protein C is a 62 kDa, vitamin K-dependent glycoprotein produced by the liver. It has a half-life of 6-8 h. Protein C activity in normal volunteers is reported to be 0.61-1.32 U/ml. Levels of 0.55-0.65 may be consistent with either heterozygous deficiency or may represent the lower end of the normal distribution. 
Deficiency of Protein C can manifest as Type I deficiency in which both Protein C antigen and activity are decreased (absolute Protein C deficiency); and Type II deficiency in which low Protein C activity contrasts with normal antigen levels (functional deficiency). Type I defects are more common than Type II defects.  The human Protein C gene is situated in Chromosome 2. Homozygous Protein C deficiency has been described to be presenting the neonate with extensive thrombosis of visceral veins or with purpura fulminans, with untreated cases having a thrombosis in 15% (8/54) of their Protein C deficient patients.  However, a few cases, with low Protein C levels (< 20%), have late clinical expression. They present with venous thrombosis during early childhood or early adulthood. 
Protein S deficiency
Protein S is an essential cofactor for the activated protein C-mediated degradation of coagulation Factors V and VIII. The range of total Protein S antigen in normal volunteers is reported to be 0.67-1.25 U/ml and free Protein S to be 0.23-0.49 U/ml.  The Protein S gene has been mapped to Chromosome 3. Protein S deficiency presents as an autosomal dominant trait with complete penetrance and may be heterozygous, homozygous or compound heterozygous.
The prevalence of Protein S deficiency in the general population has not been reliably assessed in any study in the past. Fifty per cent of Protein S deficient patients would sustain a first thrombotic event before 45 years of age.  In unselected patients with VTE, Seligsohn et al., in their meta-analysis found a 2.3% prevalence of Protein S deficiency.  In selected patients the reported prevalence ranges from 1.5-6.3%.  Van den Belt et al., , nevertheless, in their study on recurrence of VTE found a 23% recurrence rate in carriers of Protein C/S deficiencies. 
Antithrombin is a natural inhibitor of thrombin as well as of Factors Xa, IXa, XIIa and kallikrien. The British Committee for standards in hematology has noted that while plasma levels of antithrombin in patients with familial antithrombin deficiency are 0.4-0.7 i.u./ml, levels higher than 0.8 i.u./ml are considered unlikely to predispose to thrombosis.  The gene for Antithrombin III is located on Chromosome 1q containing seven exons. Family studies in kindreds with antithrombin deficiency have revealed that antithrombin deficiency probably confers a higher risk of thrombosis than Protein C or Protein S deficiencies. 
The reported incidence of antithrombin deficiency in the general population varies from 0.2-0.17% as revised by De Stefano et al. ,  which amounts to one-tenth of that for Protein C deficiency. Despite this it has 1-2% prevalence in patients with thrombosis (as against 2.5% for Protein C deficient patients). Thus, Antithrombin III appears to confer a higher thrombotic risk than Protein C and S deficiencies. 
Homozygous antithrombin deficiency is not compatible with life. In heterozygous antithrombin deficiency, thrombosis is common before 16 years of age, and about 50% of asymptomatic family members develop a first thrombotic event before 25 years of age.  In unselected patients presenting with VTE, antithrombin deficiency was present in 1.9% in a meta-analysis by Seligsohn et al .  Report by Koster's Leiden thrombophilia study has predicted a 50-fold difference between the prevalence of antithrombin deficiency among patients with a first VTE as compared to the general population.  The prevalence in selected patients varies from 0.5-5% as reported by various studies and amounts to 4.3% in Seligsohn's meta-analysis. ,,
Acquired antithrombin deficiency
Acquired antithrombin deficiency is caused by liver disease, nephrotic syndrome, DIC, pregnancy etc. In contrast to Protein C or Protein S levels, antithrombin levels may increase with oral anticoagulant treatment. Full-dose heparin treatment causes a 30% reduction in antithrombin levels within several days,  however, it is not so much with low molecular weight heparin. 
Activated Protein C resistance and Factor V Leiden
Thrombin generation at sites of vascular injury on the one hand activates Factors V and VIII, while on the other hand activates Protein C. Activated protein C inactivates membrane-bound Factors V and VIII, while its action on free Factor V (FV) and Factor VIII is negligible. Cleavage of Factor Va by activated Protein C occurs at three sites, Arg 506, Arg 306 and Arg 679. Cleavage at Arg 506 has favorable kinetics as compared to cleavage at the other two sites by a factor of 10. The phenomenon of activated Protein C resistance described by Dahlback in 1993 was subsequently proven to be due to a G506A mutation which makes Arg 506 unfavorable to cleavage by activated Protein C. This mutation, which was named Factor V Leiden, gives rise to inadequate prolongation of activated partial thromboplastin time on addition of activated Protein C to the affected plasma.  Rodegheiro et al., however, in 1999 demonstrated that activated Protein C resistance (APCR) and the Factor V Leiden (FVL) carrier state are independent risk factors for VTE.  This suggested the possibility of other mutations, which would also be the cause of APCR. Indeed separate reports by Chan et al., and Williamson et al. , in 1998 described two different mutations in the Arg 306, designated FV-Hongkong and FV-Cambridge respectively. , However, FV Leiden (FVL) remains the underlying defect in 95% of cases with APCR.  The average prevalence of FVL/ APCR in the Western population is 10-15%.  The odds ratio for development of VTE was estimated to be 6.6 by Koster et al. , in the Leiden thrombophilia study for heterozygous carriers of the Factor V Leiden mutation.  Homozygous Factor V Leiden defects confer a risk as high as 30-140 fold, as summarized by Dahlback in a review.  The mean age at first thrombosis was reported as 25 years (10-40 years) in homozygotes and 36 years (18-71 years) in heterozygotes by Zoller et al . 
In unselected patients, the incidence of APCR/ FVL ranges from 11-12%. , Seligsohn et al., in their review have derived a 40% prevalence of APCR in selected patients from an analysis of existing data. 
There is also a reported association of the Factor V Leiden mutation with the Budd-Chiari syndrome and cortical venous thrombosis. , Though there is a lower absolute risk of thrombosis in patients with Factor V Leiden mutation, it is associated in a significant number of patients with both central and peripheral venous thrombosis. This is explained by the high prevalence of this defect in the general population. 
Fibrinogen and dysfibrinogenemia
Fibrinogen is the main determinant of plasma viscosity accounting for about 50% of the latter. Besides, it has been postulated to favor accelerated atherosclerosis and a thrombogenic state by directly influencing the amount of fibrin formed in response to a stimulus for coagulation. Increased levels of fibrinogen (greater than 300 mg/dl) are associated with an approximately two-fold risk for a variety of cardiovascular end-points such as unstable angina, nonfatal and fatal myocardial infarction, stroke and peripheral vascular disease. An increased level of plasma fibrinogen is postulated to be one of the factors responsible for the increased thrombotic risk in pregnancy, manifesting as deep venous thrombosis.
Congenital dysfibrinogenemia is a term used to describe a relatively rare condition wherein an inherited abnormality in the fibrin molecule results in defective fibrin clot formation. The complications associated with abnormal clot formation may range from asymptomatic to life-threatening ones. Fortunately, 40% of patients with congenital dysfibrinogenemia are asymptomatic; however, 50% of patients have a bleeding disorder and the remaining 10% have a thrombotic disorder or combined thrombotic and bleeding tendencies.  The prevalence of congenital dysfibrinogenemia in venous thrombosis is estimated to be 0.8%. 
Congenital afibrinogenemia, a bleeding disorder, when treated with fibrinogen may be associated with venous thrombosis also.
Hyperhomocysteinemia and homocystinuria
Homocysteine is a sulphur-containing amino acid that is gaining increasing attention due to its association with vascular disease, especially accelerated arterial atherosclerosis. Though association of hyperhomocysteinemia with arterial thrombosis is well known, it is a low risk for venous thrombosis.
The thrombogenicity of hyperhomocysteinemia is related to both accelerated atherosclerosis and hypercoagulability. , Thus, hyperhomocysteinemia plays a part in the causation of both arterial and venous thrombosis. Hyperhomocysteinemia is the apparent explanation for 10-25% of patients with venous thrombosis. , Falcon et al. , in their study of venous thrombosis in patients under 40 years of age found hyperhomocysteinemia in 19% of patients with DVT. 
Homocystinuria, though rare, is a high risk for venous thrombosis. The frequency of classic homocystinuria due to reduced activity of cystathionine β-synthase is 1:200,000. The patient presents with life-threatening vascular complications. 
This is a point mutation, which results in a substitution of Guanine for Adenine at the locus 20210 on the prothrombin gene. This in turn results in increased prothrombin production by the liver,  and high prothrombin levels predispose to thrombosis, both arterial and venous.
The prevalence of this mutation in the general population is estimated to be 2.3% for its heterozygous form.  In unselected patients of VTE, the prevalence of this mutation was 6.2% conferring a relative risk of 2.8. This risk appeared to correlate with increased prothrombin levels (with levels greater than 115% associated with an odds ratio of 2.1).  In selected patients with a family history of VTE, the prevalence of prothrombin G20210A is 18%. 
High factor VIII levels are common in the general population, with levels greater than 150 IU/dl found in 11% of the general population as determined by Koster et al .  The relative risk for VTE was six-fold as compared to those with levels less than 100 IU/dL. Non-O blood groups have higher levels of plasma factor VIII. 
High Factor VIII levels have also been associated with recurrent VTE. Kyrle et al., in a study of 360 patients with VTE found recurrence in 10.6% and reported that those with recurrence had significantly higher mean levels of Factor VIII (182 ± 66 vs. 157 ± 54 µg/dL). They found that the overall relative risk of recurrence was 6.7 for those with Factor VIII levels greater than the 90 th percentile. 
Other primary thrombophilias
There are many other thrombophilic defects whose association with thrombosis remains unproven.  These are plasminogen deficiency, defects in plasminogen activator synthesis or release, increased concentration of fibrinolytic inhibitors (e.g. plasminogen activator Inhibitor I), heparin Cofactor II deficiency, increased concentration of histidine-rich glycoprotein and Factor XII deficiency.
From the time of Conley and Hartmann's description of a unique coagulation inhibitor in patients with systemic lupus erythematosus (SLE) in 1952 and the subsequent designation of the term lupus anticoagulant to this substance by Feinstein and Rappaport in 1972, many studies have attempted to define the association between antiphospholipid antibodies and thrombosis.
The antiphospholipid syndrome (APS) manifests immunologically with the presence of one or more of the following abnormalities-the lupus anticoagulant (LAC), anticardiolipin antibodies (ACLA) and β2-glycoprotein-I antibodies.
Lupus anticoagulants and anticardiolipin antibodies are associated with many disorders like collagen vascular disorders, SLE, lymphoproliferative disorders, viral infections, drug intake (hydralazine and chlorpromazine) and have also been described as an isolated abnormality- primary APS. 
APS is associated with both arterial and venous thrombosis, recurrent fetal loss, livedo reticularis, thrombocytopenia, elevated levels of antibodies directed against membrane anionic phospholipids or there is evidence of a circulating anticoagulant.
Antiphospholipid antibodies (APLA) have been reported in 2-5% of the general population. , While in patients with SLE, the incidence of APLA is as high as 50-60%, non-SLE cases account for more than half of the cases of the antiphospholipid antibody syndrome. Thrombosis associated with ACLA is related to both the titer and the type of ACLA, IgG ACLA imparting a higher thrombotic risk. 
Many mechanisms are proposed for antiphospholid antibodies to induce thrombosis. These range from endothelial defects to platelet activation to alteration in plasma proteins.
These are summarized below: [Additional file 1]
Cumulative literature over the years suggests that the thrombotic events associated with APLS are 70% venous and 30% arterial. Deep venous thrombosis and PE are the most common venous events. 
Updated Sapporo Antiphospholipid Syndrome Diagnostic Criteria
It is adopted to make a diagnosis of APS, which is as follows:
I Persistently (12 weeks apart) positive antiphospholipid antibodies (lupus anticoagulant, moderate-to-high titer anticardiolipin antibodies, or moderate-to-high titer β2-glycoprotein-I antibodies)
Repeat antiphospholipid testing is necessary due to the naturally occurring presence of transient low levels of antiphospholipid following infections.
This is defined as:
I Laboratory confirmation of the presence of antiphospholipid antibodies
Cases of catastrophies due to central vein thrombosis like renal vein thrombosis and Budd-Chiari syndrome occur. 
The probability of first DVT with some hereditary and acquired risk factors are given in [Table 3] and [Table 4].  [Table 5] mentions probability of recurrence after first DVT. 
Interaction, also known as effect modification or synergism is present when the risk in the presence of two factors exceeds the sum of separate effects of the two factors. 
Interaction in thrombotic diatheses may occur at any of the following levels:
Gene-gene interaction; interaction between two primary thrombophilic defects
The first example of gene-gene interaction is the consequence of a homozygous gene defect. Homozygous Protein C and S deficiencies result in fatal neonatal purpura fulminans or visceral venous thrombosis, while homozygous antithrombin deficiency is incompatible with life.  Milder defects like the Factor V Leiden mutation result in a sevenfold risk in heterozygotes, but become a major risk with an 80-fold thrombotic risk in homozygotes. 
Due to the high incidence of this defect in the general population, Factor V Leiden is the primary thrombophilia most commonly found to interact synergistically with other primary thrombophilias.
In a family study, Koelmen et al., reported that among families with both Factor V Leiden and Protein C gene mutations, thrombosis was present in 31% of individuals with Protein C deficiency and 13% of individuals with Factor V Leiden. The incidence of thrombosis in individuals with both abnormalities was as high as 75%. They also found that the first thrombotic event in patients with both abnormalities occurred at an earlier age than in those with only one abnormality. 
The combination of Factor V Leiden with APLA gives rise to a severe thrombotic tendency, a prevalence of 22% being greater than the prevalence of either abnormality in isolation. 
Interaction between risk factors
Multiple acquired risk factors for thrombosis may be present in a single individual which can predispose to development of VTE. Probably the most studied is the association of pregnancy and oral contraceptive with other risk factors for VTE. The risk due to oral contraceptives and hetereozygous Factor V Leiden is fourfold and fivefold respectively when singly present, but the risk rises to 35 times when both the risks are combined. 
Advanced age and immobilization for more than three days are stated by Clagett et al., to be the risk factors which determine the incidence of post-traumatic DVT.  Similarly, immobilization, advanced age, surgical intervention and site of malignancy are determinants of malignancy-related DVT.  In the review by Andres et al. , Caesarian section was reported to increase pregnancy-related VTE, while a past history of oral contraceptive use and a thrombotic event during the previous pregnancy predicted an increased risk of 7.5-12% for a recurrent VTE.  Thus VTE is the end result of the interaction of a variety of risk factors and more than one factor may be contributing to a thrombotic event in a patient.
Gene environment interactions
Primary thrombophilias contribute in various degrees to an increased risk of thrombosis. While patients with fulminant thrombophilias suffer thrombotic episodes relentlessly without treatment, minor traits may be discovered only by laboratory testing. However, thrombosis in most patients is episodic, often separated by prolonged asymptomatic periods. This intermittent pattern of thrombosis suggests that each event is probably triggered by some factor. The trigger is usually in the form of an acquired risk factor.  Indeed in patients with thrombophilias, approximately one half of the thrombotic events occur when there is an increased risk from additional risk factors such as oral contraceptive use, surgery, trauma or pregnancy.  Tabernero et al. , in their study of hereditary disorders with venous thrombosis found that 66% (135/204) of their cases suffered the thrombotic event in the presence of precipitating factors. 
Pregnancy and thrombophilias show a synergistic effect on the risk for thrombosis [Table 6]. 
The detection of thrombophilias requires sophisticated tests, which in a developing country like India may not be widely available for routine thrombophilic screening. The costs of these tests further add to their being rather prohibitive. In such a scenario it becomes important to identify the correct candidates for such tests.
Most thrombophilic defects noted above manifest with thrombosis at an early age (<45 years for Protein C and Protein S deficiency, <25 years for Antithrombin III deficiency).  Increasing age has been reported to increase thrombotic risk. Malignancy-related venous thrombosis naturally occurs more commonly in the older age group. 
All the thrombophilic defects mentioned above result in recurrent episodes of thrombosis, in some cases even while on anticoagulant treatment.
A family history of thrombosis is present in at least 40% of individuals with thrombophilic defects.  Therefore, the presence of a family history should alert one to the presence of a heritable thrombophilia. Similarly, any episode of venous thrombosis that is life-threatening or occurs at atypical sites like cerebral or mesenteric veins should raise the possibility of an underlying thrombophilia.
In the 1995 guidelines for management of thrombophilias, Cavenagh et al., have suggested that patients should be investigated for thrombophilias in the following situations: 
Patients with any of the features given above are considered to have high likelihood of thrombophilia.  Patients with a first unprovoked event, pregnancy or OCP-related thrombosis or pulmonary embolism/proximal vein thrombosis (induced by surgery or immobilization) are considered to have an intermediate likelihood of thrombophilia; while those with thrombosis of a distal vein provoked by surgery or trauma or immobilization are unlikely to have a thrombophilia.
On the basis of the available literature, it would be desirable to evolve an approach to the diagnosis and management of thrombophilia, although a consensus on the issue is yet to emerge. The most comprehensive approach appears to be the one outlined by Seligsohn, which is briefly outlined below. 
Tests for thrombophilic screening can be divided into three categories.
Both the groups with high and intermediate likelihood of thrombophilia should be subjected to the high-priority tests after six months of anticoagulation. In addition, the high likelihood group should be subjected to the intermediate-priority tests as well. Other thrombophilias should be looked for their coexistence with commoner abnormalities like lupus anticoagulant, APCR or the prothrombin mutation. The individual test and management of each thrombophilic condition, however, is beyond the scope of this review.
In a study in western India, 25 of 86 members (28%) from the family of patients with familial history deep venous thrombosis were found to have positive markers for thrombophilia. There was a high prevalence of variant MTHFR C677T in that study, but the incidence of MTHFR C677T in our general population is also high.
Prothrombin gene polymorphism G20210A seemed to be nonexistent in our population and Antithrombin III deficiency also appeared to be low compared to other markers of thrombophilia. 
Summarizing, following factors come to light. The risk factors and thrombophilias associated with thrombosis are many. Important among them are pregnancy, immobilization, surgery, drugs like oral contraceptives, thalidomide and heparin, Factor V Leiden, prothrombin gene mutation, antiphospholipid antibodies and high Factor VIII level. Further, deficiency of Protein C, Protein S and antithrombin is also associated with VTE. Usually, a complex interaction between some of these factors causes an episode of thrombosis. Since investigations for thrombophilia are costly and time-consuming, selection criteria for investigation may be early age at onset, frequent recurrences, strong family history, unusual migratory or widespread locations, and severity out of proportion to any recognized stimulus. Further, there may be regional variations in epidemiological distribution of risk factors and primary thrombophilias, thus these are worth looking into systematically.
[Figure 1], [Figure 2]
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5], [Table 6]