Clinical Description
Clinical Manifestations of Factor V Leiden Thrombophilia
Venous thromboembolism (VTE) is the primary clinical manifestation of factor V Leiden thrombophilia. Deep vein thrombosis (DVT) is the most common VTE. The most common site for DVT is the legs, but upper-extremity thrombosis also occurs. Superficial venous thrombosis may also occur.
Pulmonary embolism (PE). Some evidence suggests that PE is less common than DVT in individuals with the Leiden variant. Multiple studies and a meta-analysis consistently found a higher risk of DVT than PE in individuals with a Leiden variant [Dentali et al 2012, van Langevelde et al 2012]. The different effect on the risk of DVT and PE is referred to as "the factor V Leiden paradox"; the mechanism is still not well understood.
Thrombosis in unusual locations such as cerebral veins and splanchnic veins may also occur, but less commonly.
Risk for VTE in adults. The clinical expression of the Leiden variant varies [Campello et al 2016]: many individuals with the Leiden variant never develop thrombosis [Heit et al 2005] and those who do typically experience their first thrombotic event as adults; however, some have recurrent thromboembolism before age 30 years.
Heterozygosity for the Leiden variant is not associated with an increase in mortality or reduction in normal life expectancy even in the presence of a history of VTE [Hille et al 1997, Heijmans et al 1998, Pabinger et al 2012].
While individuals homozygous for the Leiden variant have a higher risk for thrombosis than heterozygotes, the clinical course of an acute thrombotic episode is not more severe or more resistant to anticoagulation in homozygotes than in heterozygotes.
Heterozygotes. The relative risk for venous thrombosis is increased approximately three- to eightfold in Leiden variant heterozygotes. Lower relative risks are reported in heterozygotes identified from general population screening [Juul et al 2004, Heit et al 2005].
The risk for VTE is increased in Leiden variant heterozygotes:
Despite the increase in relative risk, the overall annual incidence of a first VTE is low in heterozygotes, approximately 0.5% [Middeldorp 2011].
A heterozygous Leiden variant was associated with the following:
A sixfold increased risk for primary upper-extremity thrombosis (not related to malignancy or a venous catheter) [
Martinelli et al 2004]
A sixfold increased risk of superficial vein thrombosis not associated with varicose veins, malignancy, or autoimmune disorders [
Martinelli et al 1999]
A Leiden variant:
Homozygotes. Compared to heterozygotes, homozygotes have a higher thrombotic risk and tend to develop thrombosis at a younger age.
The risk for VTE is increased in Leiden variant homozygotes:
Risk for VTE in children. The cause of VTE in children is multifactorial and results from the interaction between acquired clinical risk factors (see Table 2), one or more underlying medical conditions, and an inherited predisposition to thrombophilia [Klaassen et al 2015, van Ommen & Nowak-Göttl 2017].
The most important clinical risk factor for thrombosis in children is a central venous catheter (CVC). A Leiden variant was associated with CVC-related VTE in some [Neshat-Vahid et al 2016] but not all studies [Thom et al 2014].
A Leiden variant significantly increased the risk of cerebral venous thrombosis in children (odds ratio 2.74); see meta-analysis by Kenet et al [2010].
A Leiden variant was also reported to increase the risk of neonatal cerebral vein thrombosis [Kenet et al 2010, Laugesaar et al 2010].
In a prospective study, asymptomatic children – family members of symptomatic probands with the Leiden variant who were themselves heterozygous or homozygous for the Leiden variant – had no thrombotic complications during follow up that averaged five years [Tormene et al 2002]. Thus, the available data suggest that asymptomatic children with a Leiden variant are at low risk for thrombosis except in the setting of strong circumstantial risk factors (see Table 2).
Risk for VTE in pregnancy. Normal pregnancy is associated with a five- to tenfold increased risk of developing VTE.
Women heterozygous for the Leiden variant have a five to eight times greater risk of pregnancy-related VTE than women without the variant [Robertson et al 2006, Bleker et al 2014, Gerhardt et al 2016]. The risk is higher in women from families with a history of thrombosis and in women older than age 34 years. The highest risk of VTE occurs during the first six weeks post partum.
While heterozygosity for the Leiden variant increases the relative risk for pregnancy-associated VTE, the absolute risk is low in the absence of other predisposing factors. VTE is estimated to occur in 1% of pregnancies in women who are Leiden variant heterozygotes. The absolute risk increases to 3% in those with a positive family history of VTE [Bleker et al 2014, Campello et al 2016].
In women homozygous for the Leiden variant the relative risk is increased 17- to 34-fold [Robertson et al 2006, Gerhardt et al 2016]. The absolute risk of developing pregnancy-related VTE is estimated at 2.2%-4.8% of pregnancies. The risk is higher (14%) in homozygotes with a positive family history and in those older than age 34 years [Bleker et al 2014, Gerhardt et al 2016].
Women doubly heterozygous for the Leiden variant and the 20210G>A F2 variant are reported to have an eight- to 47-fold increased relative risk of pregnancy-related VTE [Jacobsen et al 2010, Gerhardt et al 2016]. The probability of VTE during pregnancy and the puerperium is lower (5.5%) in doubly heterozygous women younger than age 35 years than in older women (8.2%) [Gerhardt et al 2016].
Recurrent Thrombosis
In adults heterozygous for a Leiden variant alone. Evidence suggests that a heterozygous Leiden variant has at most a modest effect on risk for recurrent thrombosis after initial treatment of a first VTE.
A modest, approximately 1.5-fold increased risk of VTE recurrence was identified in several meta-analyses [Marchiori et al 2007, Segal et al 2009]; however, in several prospective cohort studies of unselected individuals with a first VTE the recurrence risk was not increased in Leiden variant heterozygotes [Christiansen et al 2005, Lijfering et al 2010].
The reported risk may be higher in studies of families prone to thrombosis than in unselected individuals. In a prospective study of families with a strong history of thrombosis, the incidence of recurrent VTE was 3.5 per 100 person-years in persons with the Leiden variant (heterozygotes and homozygotes) [Vossen et al 2005]; however, a large family study found the rate of recurrent VTE in relatives with a Leiden variant to be similar to those reported in the general population (7% after 2 years, 11% after 5 years, and 25% after 10 years) [Lijfering et al 2009].
In Leiden variant homozygotes and heterozygotes with other risk factors. Risks for recurrent VTE in Leiden variant homozygotes and double heterozygotes for the Leiden variant and the F2 20210G>A variant vary widely between studies.
Similar rates of VTE recurrence for both Leiden variant homozygotes and heterozygotes were found in a recent study [Perez Botero et al 2016], whereas an earlier systematic review found that homozygosity for the Leiden variant conferred a 2.6-fold increased risk of recurrent VTE [Segal et al 2009].
Not all studies found a high risk for recurrence in Leiden variant homozygotes and double heterozygotes even when the analysis was restricted to those with a first unprovoked VTE [Lijfering et al 2010].
The risk of VTE is unknown in individuals with the rare combination of Leiden variant and prothrombin 20210G>A variants (i.e., homozygous Leiden variant / heterozygous 20210G>A, homozygous Leiden variant / homozygous 20210G>A). In a retrospective study, the risk of recurrence was 13% at one year and 22% at five years, similar to that expected in individuals with an unprovoked VTE without thrombophilia. However, the recurrence risk in this group was significantly higher than expected after a first VTE provoked by a transient risk factor (6% at 1 year and 26% at 5 years) [Lim et al 2016].
In children. In children, inherited thrombophilia appears to have at most a modest effect on the risk of recurrence, similar to findings in adults [Klaassen et al 2015].
In pregnant women. During pregnancy women with a prior history of VTE have an increased recurrence risk, ranging from 0% to 15% in published studies. The risk is higher in women with a prior unprovoked episode or an estrogen-related VTE, and in those with coexisting genetic or acquired risk factors (Table 2). No studies have specifically evaluated the risk for recurrent VTE in pregnant women who have the Leiden variant.
In subgroup analysis of a prospective study of the safety of withholding anticoagulation during pregnancy in 125 women with a history of VTE, Brill-Edwards et al [2000] found the following:
Women with a prior unprovoked VTE and thrombophilia (especially factor V Leiden thrombophilia) had the highest recurrence rate during pregnancy (20% of pregnancies).
A Leiden variant was associated with an increased risk of antepartum recurrence (odds ratio 10).
Women with either thrombophilia or a prior unprovoked VTE (but not both) had recurrence rates of 13% and 7.7%, respectively.
Obstetric Complications
It is unlikely that a factor V Leiden thrombophilia (i.e., heterozygosity or homozygosity for the Leiden variant) is a major factor contributing to pregnancy loss and other adverse pregnancy outcomes (preeclampsia, fetal growth restriction, and placental abruption). At most, factor V Leiden thrombophilia is one of multiple largely unknown genetic and environmental predisposing factors contributing to these complications.
Pregnancy loss. Available data suggest that Leiden variant heterozygosity is at most a weak contributor to recurrent or late pregnancy loss. A meta-analysis evaluating only prospective cohort studies reported a slightly increased risk of pregnancy loss in women with the Leiden variant (4.2%) compared to those without the variant (3.2%) (odds ratio 1.52) [Rodger et al 2010]. A meta-analysis found that heterozygosity for the Leiden variant is associated with a twofold increased risk for a late unexplained fetal loss and a fourfold higher risk for loss in the second trimester compared to the first trimester [Robertson et al 2006]. Although maternal homozygosity for the Leiden variant was associated with stillbirth, presence of the Leiden variant was not associated with stillbirths in the subset of stillbirths resulting from placental insufficiency [Silver et al 2016].
Preeclampsia, fetal growth restriction, and placental abruption may also involve impaired placental perfusion; however, their association with inherited thrombophilia is more controversial [Greer et al 2014]. For example:
A systematic review focused on prospective cohort studies found no significant association of preeclampsia or placental abruption with factor V Leiden thrombophilia [
Rodger et al 2010].
A Danish case-cohort study found that heterozygosity for the Leiden variant increased the risk of severe preeclampsia (odds ratio 1.6), severe fetal growth restriction (odds ratio 1.5), and symptomatic placental abruption (odds ratio 1.7) [
Lykke et al 2012].
Such conflicting results may reflect the varying diagnostic and selection criteria, different ethnic groups, and small number of cases included. However, given that preeclampsia and placental abruption are heterogeneous disorders, it is unlikely that a single thrombophilic variant (such as the Leiden variant) plays a major causal role.
Arterial Thrombosis NOT Convincingly Associated with the Leiden Variant
The available evidence indicates that presence of a Leiden variant is not a major risk factor for arterial thrombosis of any sort, including myocardial infarction and stroke in fetuses, children, and adults, the majority of which occur in the presence of established cardiovascular risk factors (including hypertension, hyperlipidemia, diabetes mellitus, and smoking). For more information click here (pdf).
Clinical Expression of Factor V Leiden Thrombophilia
In addition to the number of Leiden variants (discussed above), the clinical expression of factor V Leiden thrombophilia is influenced by family history, coexisting genetic abnormalities, acquired thrombophilic disorders, and circumstantial risk factors.
A positive family history. Individuals with a Leiden variant who have a first-degree relative with a history of thrombosis had a threefold increased risk for VTE compared to those with a Leiden variant with a negative family history [Bezemer et al 2009]. The risk was increased to fivefold in those with a relative with a VTE before age 50 years and to 18-fold with two or more affected relatives. The family history had additional value in predicting risk even in those with a Leiden variant, suggesting the presence of unknown genetic risk factors.
Coexisting genetic abnormalities. The combination of Leiden variant heterozygosity and most other thrombophilic disorders (including protein C deficiency, protein S deficiency, antithrombin deficiency, and the F2 thrombophilia variant [c.*97G>A, commonly known as 20210G>A]) has a supra-additive effect on overall thrombotic risk [Ridker et al 1997a] (see Differential Diagnosis). For example:
Individuals
heterozygous for either the Leiden variant or the
F2 thrombophilia variant had a four- to fivefold increased thrombotic risk, compared to double heterozygotes, who had a 20-fold increased thrombotic risk [
Emmerich et al 2001].
The
F2 thrombophilia variant was four- to fivefold more common in Leiden variant homozygotes with VTE than in controls without a Leiden variant and with no thrombotic history [
Ehrenforth et al 2004].
Acquired thrombophilic disorders include antiphospholipid antibody (APLA) syndrome, paroxysmal nocturnal hemoglobinuria, myeloproliferative disorders, and increased levels of clotting factors. Despite the following observations, the effect of these acquired disorders on Leiden variant-associated thrombotic risk is not well defined.
Leiden variant heterozygotes with factor VIII levels greater than150% of normal had a two- to threefold increased incidence of VTE than heterozygotes for a Leiden variant alone [
Lensen et al 2001]. The reason for the association of high FVIII levels with VTE is unknown; factor VIII, an acute phase reactant, is high in the presence of inflammation and estrogen. Although suspected, a genetic basis has not been identified.
A Leiden variant was reported to contribute to increased risk for thrombotic complications in persons with polycythemia vera and essential thrombocytosis [
Trifa et al 2014].
Circumstantial Risk Factors for VTE
Circumstantial risk factors for VTE in Leiden variant heterozygotes or homozygotes are summarized in Table 2. Other risk factors that to date have not been studied in Leiden variant heterozygotes are the newer forms of combined hormonal contraception, transdermal and vaginal ring contraception, for which the risk of VTE is at least as great as the risk associated with combined oral contraceptives (COCs) [Dore et al 2010, Lidegaard et al 2012].
Table 2.
Circumstantial Risk Factors: Increased Risk for Thrombophilia in Persons with the Factor V Leiden Variant (heterozygosity and homozygosity not specified)
View in own window
| Circumstance | Relative Risk for VTE | Comment | Citation |
|---|
| Malignancy 1 | Unknown: probably modest | In persons w/cancer, no indication for:
| Farge et al [2013], Kovac et al [2015], Pabinger et al [2015] |
| Central venous catheter use | 2x-3x ↑ relative risk | |
Van Rooden et al [2004]
|
| 5x ↑ relative risk in those w/cancer |
Dentali et al [2008]
|
| Travel | 2x ↑ relative risk | Risk ↑:
W/travel duration; In those w/thrombophilia
|
Chandra et al [2009]
|
| 8x ↑ relative risk when traveling ≥4 uninterrupted hrs | |
Cannegieter et al [2006]
|
| COCs 1 | Much ↑ during 1st yr of use than subsequent yrs | |
Martinelli et al [2016]
|
| ↑ in COC w/desogestrel or drospirenone than COC w/levonorgestrel | |
Bergendal et al [2014]
|
| Unopposed progesterone contraception | Injectable medroxyprogesterone assoc w/16x ↑ risk | |
Bergendal et al [2014]
|
| Low-dose oral form has lowest thrombotic risk of hormonal contraception |
Tepper et al [2016]
|
| Oral HRT 1 | 7x-25x ↑ relative risk | Risk ↑ w/↑ estrogen dose | Straczek et al [2005], Douketis et al [2011] |
| Transdermal HRT 1 | Lower relative risk than oral HRT | Preliminary data suggest may not ↑ thrombotic risk | Straczek et al [2005], Canonico et al [2010], ACOG [2013a] |
| SERMS 1 | Not well defined | Risk likely >2x ↑ over risk assoc w/SERMS alone | Barrett-Connor et al [2006], Nelson [2013] |
| Obesity (BMI >30 kg/m2) | 5x-12x ↑ relative risk | Risk ↑ w/BMI | Severinsen et al [2010], Ribeiro et al [2016] |
| Overweight (BMI >25 - <30 kg/m2) | 4x-10x ↑ relative risk | Risk ↑ w/BMI | Severinsen et al [2010], Ribeiro et al [2016] |
| Organ transplantation | Not well defined | | Ghisdal et al [2010], Pereboom et al [2011], Parajuli et al [2016] |
| Minor leg injury | 23x-50x ↑ risk | | van Stralen et al [2008], van Adrichem et al [2014] |
| Surgery | Unclear | Any excess risk conferred by Leiden variant heterozygosity likely small, compared to risk assoc w/surgery | Joseph et al [2005], Charen et al [2015], van Adrichem et al [2015] |
| Age | After age 45 yrs lifetime risk = 17% (vs 8% in general population) | ↑ age is an independent risk factor for VTE. |
Bell et al [2016]
|
| Risk ↑ w/age, BMI, smoking, +FH |
Juul et al [2004]
|
| Highest risk in persons age >70 yrs w/+FH |
Karasu et al [2016]
|
- +
FH = positive family history; COCs = combined oral contraceptives; HRT = hormone replacement therapy; SERMS = selective estrogen receptor modulators
- 1.
See text that follows table for more details.
Malignancy. To what extent inherited thrombophilia increases the risk of VTE in persons with cancer remains controversial [Decousus et al 2007, Pabinger et al 2015]. Because malignancy is such a strong thrombotic risk factor, it may obscure the effect of mild thrombophilic disorders including factor V Leiden. Thrombophilia status was not considered in recent guidelines for prophylaxis and treatment of VTE in patients with cancer [Farge et al 2013].
Combined oral contraceptive (COC) use substantially increases the relative risk for VTE in women heterozygous for the Leiden variant.
The supra-additive effect of both a Leiden variant and use of COC was confirmed in multiple studies in which odds ratios for VTE ranged from 11 to 41 [Wu et al 2005, Dayan et al 2011, Bergendal et al 2014, van Vlijmen et al 2016]. For women who are either homozygous for the Leiden variant or doubly heterozygous for the Leiden variant and the prothrombin 20210G>A variant the odds ratios for VTE ranged from 17 to 110 [Mohllajee et al 2006, van Vlijmen et al 2016].
Despite the marked increase in relative risk for VTE, the absolute incidence of VTE may be low because of the low baseline risk for VTE in young women.The incidence of VTE in COC users with either a Leiden variant or the prothrombin 20210G>A variant ranged 0.49 to 2.0 VTE/100 pill-years compared to 0.19 to 0/100 pill-years in COC users without these variants. The absolute VTE risk is substantially higher in COC users doubly heterozygous for the Leiden variant and the prothrombin 20210G>A variant or homozygous for either variant (0.86 vs 0.19/100 pill years) [van Vlijmen et al 2011, van Vlijmen et al 2016].
The thrombotic risk of COC is at least as high in women older than age 50 years as in younger users [Roach et al 2013]. However, since the incidence of VTE increases with age, the absolute risk for VTE in women older than age 50 years is much higher than in younger COC users.
Oral hormone replacement therapy (HRT) is associated with a two- to fourfold increased relative risk for VTE in healthy postmenopausal users of HRT compared to non-users [Renoux et al 2010, Eisenberger & Westhoff 2014]. Data comparing the VTE risk of combined estrogen/progestin HRT and unopposed estrogen are inconclusive [Eisenberger & Westhoff 2014].
The risk increases with higher estrogen doses and may differ with the particular estrogen and progestin components [Renoux et al 2010, Canonico et al 2011, Smith et al 2014].
The risk of HRT is threefold increased in postmenopausal women with a factor V Leiden or prothrombin 20210G>A variant than in HRT users without thrombophilia [Roach et al 2013].
Transdermal HRT. Multiple observational studies consistently found that transdermal HRT did not increase the risk of VTE [Canonico et al 2010, Sweetland et al 2012, ACOG 2013a]. There is also evidence that transdermal estrogen is associated with a lower thrombotic risk than oral estrogen in women with inherited thrombophilic variants including the Leiden variant [Canonico et al 2010]. Women with a Leiden variant using transdermal estrogen had a risk similar to that of non-users with the variant. Among women with a Leiden variant the use of oral estrogen was associated with a fourfold increased risk for VTE over transdermal estrogen [Straczek et al 2005]. However, no prospective randomized trials have confirmed the safety of transdermal HRT in women with inherited thrombophilia.
Selective estrogen receptor modulators (SERMS). The risk for VTE in women with the Leiden variant who use SERMS is not well defined. Limited data suggest that SERMs such as tamoxifen and raloxifene are associated with a twofold increased risk for VTE, similar to the risk for HRT [Barrett-Connor et al 2006]. The thrombotic risk conferred by tamoxifen was higher than raloxifene in trials for primary prevention of breast cancer [Nelson 2013]. In light of the interaction of factor V Leiden with HRT, the risk is likely higher than that associated with SERMS alone.
