Thrombophilia in Children: Who to Test, How, When, and Why?

In 1956, Jordan and Nadorff described “the familial tendency in thromboembolic disease,” where over 40 patients with venous thrombosis and a family history that included multiple young relatives with venous thrombosis were reported, though the etiology for this familial tendency was not understood.¹ Antithrombin deficiency was the first inherited defect to be identified in 1965, but accounted for only a very small subset of patients with venous thrombosis.² Nearly 20 years later, two additional risk factors, low levels of either protein C or protein S, were associated with familial thrombosis.^3,4 The first infant with purpura fulminans due to homozygous protein C deficiency was described in 1983.⁵ These early studies involved evaluation of plasma levels of the specific proteins in extended family pedigrees to identify the inherited defect.⁶ Later, genetic analysis of the DNA of these anticoagulant proteins, by using polymerase chain reaction (PCR), demonstrated that multiple different mutations were responsible for each of these deficiencies.⁶ 

In the 1990s recognition of resistance to activated protein C led to the discovery of the factor V Leiden mutation, in which a single amino acid substitution of glutamic acid for arginine at position 506 resulted in the first described thrombophilic defect due to enhanced coagulation, rather than decreased anticoagulant activity.⁷ Another point mutation, the prothrombin 20210 mutation that enhanced coagulation through increased message RNA stability, was identified in 1996.⁸ Together these genetic mutations were far more prevalent than deficiencies of the anticoagulant proteins, accounting for a much higher population attributable risk in the Caucasian population.^8,9 Adult population-based studies have identified many other potential prothrombotic factors, but the clinical utility of testing for these risk factors has not been as well developed as the above five factors.⁶ Testing for inherited thrombophilias has increased significantly over the last 15 years.¹⁰ The Factor V Leiden mutation has become one of the most commonly performed genetic tests.¹¹ Recently, attention has focused on the clinical usefulness of thrombophilia testing in adults and children. As will be discussed below, the utility of such testing in childhood varies greatly depending on the clinical situation.

Thrombophilia refers to the propensity to develop thrombosis and can be applied clinically to patients who develop spontaneous venous thromboembolism (VTE), VTE with severity out of proportion to the stimulus, recurrent thrombosis, or VTE at a young age. It may also be used to describe an inherited risk factor for thrombosis, which is usually identified through laboratory testing. Individuals that demonstrate clinical “thrombophilia” do not necessarily have laboratory evidence of “thrombophilia,” and vice versa.

The most common laboratory thrombophilias and the tests used to establish the diagnosis are listed in Table 1 . The inherited defects in which the pathogenic link to thrombosis is well understood include deficiencies of protein C, protein S and antithrombin, the factor V Leiden mutation, and the prothrombin gene mutation. Other thrombophilias, which are less well characterized and not necessarily genetically determined, include elevated homocysteine, elevated lipoprotein(a), dysfibrinogenemias, and increased levels of factors VIII, IX and XI. Level 1 testing (see Table 1 ) includes the thrombophilia tests that are most prevalent in pediatric studies. If the results of these tests are normal and thrombophilia is strongly suspected, level 2 tests can be performed. There are numerous other proposed causes of thrombophilia. Some of these have been tested in only a small number of patients and none have gained widespread acceptance as risk factors for thrombosis. Though most of the prothrombotic alterations listed above may be inherited, several can also be acquired. Table 2  lists several clinical settings associated with acquired abnormal thrombophilia test results. Antiphospholipid antibodies (lupus anticoagulant, anti-β2 glycoprotein antibody and anticardiolipin antibody) are well-established acquired thrombophilic risk factors, although there are reports of familial cases.¹² 

Several issues regarding testing are unique to pediatric patients. Familiarity with the concept of developmental hemostasis is essential for the interpretation of thrombophilia tests in children, especially in those under the age of 6 months who account for the largest proportion of thrombotic events. Although demonstrated since the 1960s in pioneering work by William Hathaway and others, Maureen Andrew did much to promote awareness that the coagulation system in neonates differs greatly from older children and adults.^13,–16 While the basic pathways for coagulation, anticoagulation and fibrinolysis are maintained, the concentrations of many of the factors vary widely during fetal and postnatal development. Plasma levels of many procoagulants (factors II, VII, IX and X) and anticoagulants (protein S, protein C and antithrombin) are greatly reduced in neonates and even more so in premature neonates.^14,15 Dr. Andrew established reference ranges for coagulation proteins in healthy term and preterm Canadian infants, but this was nearly 20 years ago, using tests and reagents that are different from those in use today. A more recent study of coagulation testing in Australian neonates and children confirmed the concept of developmental hemostasis.¹⁷ However, normal pediatric ranges in the recent study were quite different from those previously published, demonstrating the difficulty of interpreting a borderline value.¹⁷ Ideally, each coagulation laboratory should establish age-related normal ranges from their local population, but in reality, it is not practical for each laboratory to obtain plasma samples from otherwise healthy infants and young children to establish these ranges. Therefore, when interpreting levels of protein C, S and antithrombin, particularly in infants in whom physiological values are changing rapidly over time, there are rarely “absolute cutoffs” that are useful. Often the normal range overlaps with values found in adults with heterozygous defects and retesting is required as the child matures.

The majority of young children who develop thrombosis have multiple coexisting risk factors. Several cohort studies suggest that inherited risk factors do not play a role in the majority of thrombotic events in children; however, small numbers and heterogeneous populations hamper these studies.^18,–20 The reported prevalence of thrombophilia in children with venous and arterial thrombotic events varies greatly, from as low as 13% to as high as 79%, and it is likely that this discrepancy has caused confusion regarding the role of testing.²¹ This tremendous variation is likely due to differences in study design, definition of congenital prothrombotic disorders, small sample sizes and different patient populations. To understand this further, one can look more carefully at two studies that have the most divergent results. In an unselected Canadian cohort of 171 children with non-cerebral VTE, the overall frequency of a prothrombotic risk factor was 13%, which was not higher than that of the healthy population. In contrast, the prevalence of prothrombotic risk factors in a German registry (285 patients) was 78%.²² The Canadian cohort was much younger than the German cohort (median age 2.3 months vs 6 years) and had a very high proportion of central venous catheter (CVC) VTE (77% vs 18%). Among older children with spontaneous VTE in the Canadian cohort, inherited thrombophilia was identified in 60%, confirming that this group has the highest likelihood of having an abnormal result within the larger studied population and therefore has the potential to benefit most from testing.

It is well established that the presence of a central catheter is the single most important risk factor for thrombosis in children.²³ The majority of CVC-VTE are sub-clinical and not accompanied by a swollen, painful, or erythematous extremity.^24,–26 In a prospective, multicenter cohort study of children with acute lymphoblastic leukemia (ALL), 29 of 85 (34%) developed CVC-VTE within the first month after CVC placement, but only 1 of these 29 patients had clinical symptoms. No patient tested positive for the factor V Leiden or prothrombin gene mutation.²⁷ 

A recent prospective study also did not find a contribution of prothrombotic risk factors in neonates with asymptomatic CVC-VTE.²⁸ These studies argue against thrombophilia testing in children with subclinical CVC-VTE. Testing can be considered in a patient who develops more than one subclinical event, though the value of such an analysis has not been studied.

Several studies suggest that the risk of symptomatic CVC-VTE is not increased by the presence of an inherited thrombophilia.^18,19,27 Pediatric thrombosis registries from both Canada and the Netherlands have reported that patients with catheter-related thrombosis do not have an increased prevalence of underlying thrombophilia.^18,19 In contrast, analysis of a large German thrombosis registry suggests that some inherited thrombophilias do increase the risk of symptomatic CVC-VTE .²⁹ Given these divergent results, there are not consistent data to make a firm recommendation for this patient group and analysis of a larger population may be needed. At the moment, one should consider the prevalence of prothrombotic risk factors in CVC-VTE in one’s regional population when making a decision of whether to pursue a thrombophilia workup in this setting.

Currently, thrombophilia testing of neonates and older children who have had a stroke is common, but as with venous thrombosis, the contribution of testing for a thrombophilic risk factor to improved clinical outcomes has not been demonstrated. The reported prevalence of prothrombotic conditions in pediatric stroke varies from 20% to 50%, for reasons that are similar to the variance reported in pediatric VTE.³⁰ Inherited thrombophilias (protein C deficiency, elevated lipoprotein (a), factor V Leiden mutation and prothrombin mutation) have been associated with an increased risk of recurrent stroke in older children, suggesting that testing in this setting may be useful.^31,32 A collaborative International Pediatric Stroke Registry is currently underway to address the role of prothrombotic disorders and stroke.³⁰ 

A common argument against thrombophilia testing in children is that currently there are no guidelines regarding how patient management should differ based on test results. Physicians are often left in a difficult position, unable to accurately or effectively counsel patients, because the definitive data as to whom to test and what to do with the results are not available. Given these important unanswered issues, thrombophilia testing should be strongly considered in patients who agree to participate in prospective longitudinal clinical studies to define prevalence, outcome and response to therapeutic intervention. Ideally, laboratory identification of a thrombophilic risk factor should aid in the clinical management of that patient. Potential opportunities for knowledge of an inherited thrombophilia to impact clinical outcome(s) are discussed below.

Identification of a thrombophilic marker will almost never influence the acute management of a patient with venous thrombosis, in which the mainstay of therapy is therapeutic anticoagulation. The exception would be a neonate or older child with severe (homozygous or compound heterozygous) deficiency of protein C or S or antithrombin deficiency who presents with purpura fulminans, extensive large vessel thrombosis or disseminated intravascular coagulation, which can be life-threatening. The estimated incidence of homozygous protein C deficiency is 1/ 250,000-1/500,000 births, and severe protein S and anti-thrombin deficiency are even more rare.³³ Early identification of one of these rare conditions is likely to influence treatment because replacement therapy with plasma-derived concentrate (protein C or antithrombin) or fresh frozen plasma is effective in the management of such a patient.⁵ 

Current guidelines regarding the duration of anticoagulant therapy for pediatric VTE are extrapolated from adult studies, and most recommendations are for 3 to 6 months of therapy for older children.³⁴ The paucity of data in neonates is reflected in the guidelines for neonatal VTE, where recommendations range from observation to anticoagulation from 2 weeks up to 3 months.³⁴ Given that optimal duration of therapy in children is not established, whether or not the presence of an underlying thrombophilia should influence the duration of therapy in children is simply not known. Recently, a risk-based strategy for treatment of pediatric VTE has been recommended, tailoring the duration of therapy based upon the risk for poor outcome (recurrent thrombosis or post-thrombotic syndrome).³⁵ In this strategy, risk assessment is determined using a number of patient characteristics (provoking condition, factor VIII activity, D-dimer, inherited and acquired thrombophilias) as well as characteristics of the thrombus itself (extent, location, and resolution).³⁵ 

Pooled data from adult studies demonstrate that patients with VTE who are heterozygous for either the factor V Leiden or prothrombin gene mutation have an increased risk of recurrent VTE, with odds ratios of 1.41 (95% C.I. 1.14–1.75) and 1.72 (95% C.I. 1.27–2.41), respectively.³⁶ This modestly increased risk of recurrence does not exceed the hemorrhagic risk of long-term anticoagulation.³⁶ However, patients who have a “high-risk” defect, such as anti-thrombin deficiency, homozygous factor V Leiden mutation, or more than one congenital risk factor, may benefit from long-term anticoagulation.³⁷ Therefore, one rationale for thrombophilia testing may be to identify the rare patient with a high recurrence risk genotype who may benefit from long-term anticoagulation.

There are limited data on the role of thrombophilia in recurrent VTE in pediatric patients with results that conflict due to the same variables discussed above.^18,19,38 A recent meta-analysis, which took into consideration the majority of published pediatric cohort studies, reported a significant association between protein S deficiency, anti-thrombin deficiency, prothrombin gene mutation and combined defects with recurrent VTE in children, suggesting that knowledge of an inherited thrombophilia may be clinically relevant.^39,40 Exactly which defects warrant long-term anticoagulation and the risk-benefit ratio of such an approach are not yet known, and long-term follow-up on sufficient numbers of patients is needed to more clearly define this risk so that patients can be appropriately counseled.

Patients who meet the criteria for antiphospholipid antibody syndrome are candidates for long-term anticoagulation, and so testing for antiphospholipid antibodies is warranted, particularly in patients with spontaneous thrombosis. Many pediatric patients with thrombosis have transient elevations of these antibodies, and the contribution of these transient antibodies to thrombosis is not known.

Thrombophilia testing is often performed to gain insight into why a young patient developed thrombosis, particularly in the case of unprovoked events. Given that there is still controversy regarding the causative role of prothrombotic risk factors in ill children with VTE, identification of a laboratory abnormality does not necessarily implicate that risk factor in the pathogenesis of the thrombosis.⁴¹ 

Identification of an inherited thrombophilia in a child with VTE may lead to the increased use of thromboprophylaxis in future high-risk situations, though many would argue that the history of VTE alone is enough to warrant early thromboprophylaxis.

Discovery of an inherited thrombophilia in an individual with VTE may lead to the identification of other family members who can be counseled regarding their risk. These asymptomatic individuals may be more likely to receive primary prophylaxis in the presence of transient risk factors. Young women who are identified in this way can make informed decisions regarding estrogen-containing contraception. It is also possible that knowledge of an inherited thrombophilia may lead to healthy lifestyle choices. At present, there are no studies that demonstrate the benefit of such familial testing and counseling.

In 2002 the Subcommittee for Perinatal and Pediatric Thrombosis of the Scientific and Standardization Committee (SSC) of the International Society of Thrombosis and Hemostasis (ISTH) recommended that all pediatric patients with venous or arterial thrombosis be tested for a full panel of genetic prothrombotic states.⁴² The rationale for this recommendation was that pediatric patients often have several risk factors for thrombosis, and that even if several acquired risk factors were present an evaluation for inherited risk factors should also be conducted.⁴² The committee also acknowledged that future studies involving diverse ethnic groups, neonates with catheter thrombosis, and the role of thrombophilia in recurrence might provide additional insights into the potential biological importance of these risk factors.

The field of pediatric thrombosis has undergone tremendous growth over the last 15 years. Pediatric patients who develop thrombosis are a heterogeneous group, and as more data has become available, it is increasingly clear that a single recommendation regarding thrombophilia testing that applies to all neonates and children with thrombosis may not be optimal. Most physicians would agree that thrombophilia testing should not be performed if the likelihood of an abnormal outcome is not increased in the population of interest. Recommendations for thrombophilia testing based on current understanding regarding the contribution of prothrombotic risk factors to pediatric thrombosis and the potential for benefit are outlined in Table 3 .

Thrombophilia testing in the acute setting may result in an incorrect diagnosis of inherited conditions. When interpreting the results of thrombophilia testing, it is important to remember that levels of antithrombin, protein C and protein S may transiently decrease during acute thrombosis. Similarly, factor VIII and lipoprotein(a) can be elevated in inflammatory conditions. Therefore, any test that is abnormal during the acute setting should be repeated later, ideally off anticoagulation. Levels of antithrombin may be decreased in patients who are on heparin, and vitamin K antagonists result in low levels of protein C and protein S. For these diagnostic assays, low levels must be confirmed when the infant or child is well, and testing both parents should also be considered before committing to a diagnosis of an inherited deficiency. The diagnostic accuracy of molecular mutation testing will not be affected, and so these tests can be sent even during an acute episode.

Thrombophilia testing in children who have a family member with a positive history of thrombophilia or VTE has become increasingly common, with little evidence to support this approach. Comprehensive thrombophilia testing in asymptomatic children in the absence of an identified inherited risk factor should be avoided. If the family member with VTE is alive, he or she should undergo testing prior to considering thrombophilia testing in the child. If an inherited thrombophilia is identified in a relative outside of the immediate family then the appropriate parent should be tested prior to deciding whether to pursue testing in the child. Increasingly, children whose mother or father has died at a young age from massive pulmonary embolism are referred for thrombophilia testing. In such a case, in which the contribution of inherited thrombophilia to the proband’s event is not known, identification of a thrombophilia defect in a child may result in increased anxiety, despite the fact that the absolute risk of thrombosis is still exceedingly low. Conversely, normal test results may provide false reassurance. The decision to perform thrombophilia testing should be made on an individual basis only after counseling the family regarding the potential benefits and limitations, and the results should be interpreted by a physician experienced in the management of children with thrombosis.⁴³ 

During the counseling, one should discuss how the results might affect the medical management of the child. The practice of genetic testing in children when the results have no clear benefit should be discouraged.⁴⁴ Fortunately, concerns in the United States regarding the potential impact of genetic risk factors on an individual’s future insurance coverage have lessened now that the Genetic Nondiscrimination Act of 2008 has been passed. Some have argued that testing for thrombophilia should be delayed until the subject is old enough to make their own decisions regarding testing. At the moment, there is very little opportunity for thrombophilia testing to benefit a young child. The incidence of venous thrombosis in healthy children is extremely low (0.07/100,000), so that it is unwarranted to consider long-term anticoagulation in an asymptomatic child.²³ To determine the risk of thrombosis in an asymptomatic child with inherited thrombophilia, Tormene et al followed children less than 15 years of age who had been identified with a single thrombophilic factor in a family with an individual with a VTE.⁴⁵ Eighty-one children were followed for 1 to 8 years (mean, 5 years) and no VTE occurred. In this study, some patients did encounter transient acquired risk factors such as surgery and trauma, which did not precipitate symptomatic VTE. The authors concluded that screening children under that age of 15 years was unjustified, but also that longer follow-up in a larger group was necessary.⁴⁵ 

There are some situations in which the presence of an inherited defect may influence medical decision-making, but in general this applies to older children. The first is in an adolescent female who may be considering oral contraceptive pills (OCPs). Knowledge of a congenital thrombophilia will allow her and her prescribing physician to discuss the increased risk of thrombosis associated with estrogen-containing contraception and third-generation progesterones. This discussion should include the baseline annual incidence of VTE, which is about 1 per 12,500 for all women of reproductive age and increases to 1 per 3,500 for those on OCPs, and how this risk is influenced by the presence of an inherited thrombophilia.⁴⁶ For subjects who are heterozygous for the factor V Leiden mutation and on OCPs, this baseline risk is increased 20- to 30-fold (relative risk) to approximately 1 per 500 women.⁴⁶ The young woman will be better informed regarding her choices and may choose lower-risk alternatives, such as progesterone-only preparations. In limited cases, the presence of a congenital thrombophilia may lead to targeted thromboprophylaxis in high-risk situations, e.g., after a femur fracture in an obese teenager, though there are few data to document the efficacy of this approach.

Adolescents identified with an inherited thrombophilia may benefit from avoiding high-risk situations (prolonged immobility, dehydration), pursuing healthy lifestyles (regular exercise and weight control), and recognizing early signs and symptoms of VTE. However, in reality, this education can be provided to all patients with a family history, without testing for thrombophilia.

Universal thrombophilia screening of young women prior to initiating OCPs is not recommended.⁴⁷ Similarly, there are no data to support screening of children who encounter high-risk situations, such as those with acute lymphoblastic leukemia or prior to CVC placement.

Thrombophilia testing is a controversial issue. Identification of inherited thrombophilias has contributed to our understanding of the pathophysiology of venous thrombosis. However, VTE is often a multifactorial disease, and acquired risk factors in sick children often play a more important role than inherited risk factors. The results of thrombophilia testing rarely influence management decisions, and understanding the limitations of testing is important. More data on long-term outcomes and management of pediatric patients with VTE and inherited thrombophilia are needed.