Executive Summary

Publication Details

Introduction

Venous thromboembolism (VTE) refers to pathologic thrombosis in the venous circulation. Although the most frequent venous thromboembolic event is deep venous thrombosis in the veins of the legs, thromboses can also occur in the veins of the upper extremities, pelvis, abdomen, and cerebral venous sinuses. Pulmonary embolism is the main life-threatening complication of deep vein thrombosis, in which a portion of the venous thrombus is carried to the pulmonary arteries by blood flow, potentially obstructing the pulmonary vasculature. Treatment begins with short-term use of a parenteral anticoagulant (and sometimes thrombolytic therapy) and then usually continues with a vitamin K antagonist, most commonly warfarin. The duration of therapy depends on whether the patient is considered to have continuing risk factors for recurrence.

Much effort has been devoted to quantifying the risk of recurrent thrombosis. If the patient has a persistent risk factor for thrombosis, anticoagulant therapy is often continued, sometimes for the life of the patient. The clinician and patient also try to reduce exposure to any modifiable risk factors and may use mechanical or pharmacological means of preventing thrombosis at high-risk times, such as during hospitalization or pregnancy.

The question of whether a patient has ongoing risk leads to the subject of this report, the value of testing individuals who have had a venous thromboembolic event for Factor V Leiden (FVL) and prothrombin G20210A. The FVL mutation is a base change (from G to A at position 1691) in the gene coding for the Factor V protein; the resulting amino acid substitution eliminates one of three activated protein C cleavage sites in Factor V. As a result, Factor V is inactivated to a lesser extent and persists longer in the circulation, leading to more thrombin generation. In the United States, a single FVL allele is present in about 5, 2.2, and 1.2 percent of the Caucasian, Hispanic, and African American populations, respectively. The prothrombin (Factor II) mutation is the second most common inherited risk factor for thrombosis. The mutation in the prothrombinG20210A gene is associated with an elevation of prothrombin levels to about 30 percent above normal in heterozygotes and to 70 percent above normal in homozygotes. In the United States, the prevalence of this allele is 1.1 percent in Caucasians and Hispanics and 0.3 percent in African Americans.

The Evaluation of Genomic Applications in Practice and Prevention (EGAPP) initiative was established by the Office of Public Health Genomics at the Centers for Disease Control and Prevention (CDC) to address the increasingly urgent need for timely and objective information that would allow health care providers and payers, policymakers, and consumers to identify genetic tests that are safe and useful and to provide guidance on their appropriate use in practice, based on available evidence. At their request, we reviewed the evidence regarding the value of testing for these mutations in two specific populations: (1) individuals who have had a venous thromboembolic event (probands), and (2) their family members.

Methods

The overarching question we were asked to address (Key Question [KQ] 1) was: Does FVL testing, alone or in combination with prothrombin G20210A testing, lead to improved clinical outcomes (e.g., avoidance of a recurrent VTE) in adults with a personal history of VTE or to improved clinical outcomes (e.g., avoidance of an initial VTE) in adult family members of mutation-positive individuals? Are testing results useful in medical, personal, or public health decision making? To address this question, we reviewed the literature regarding these tests’ analytic validity, clinical validity, and clinical utility when used in probands with VTE and in their family members.

The other KQs were as follows:

Our comprehensive search included electronic and hand searching. We searched five databases, MEDLINE® (1950 through May 2008), EMBASE® (1974 through December 2008), The Cochrane Library (Issue 2, 2008), the Cumulative Index to Nursing & Allied Health Literature (CINAHL®; 1982 through December 2008) and PsycInfo©, to identify primary literature on the analytic validity, clinical validity, and clinical utility of testing for FVL and prothrombin G20210A.

Two independent reviewers, from among six study team members, conducted title scans in parallel. The title review was designed to capture as many studies as possible that reported on the analytic validity, clinical validity, and clinical utility of testing for FVL and prothrombin G20210A. All titles potentially addressing these issues were promoted to the abstract review phase.

Abstracts were reviewed independently by two investigators. Abstracts were excluded if the investigators agreed that the article: (1) was not relevant to any of the key questions; (2) did not include any human data; (3) contained no original data; (4) was not conducted in adults; and (5) was not published in English. Differences of opinion were resolved through consensus adjudication.

Full articles selected for review underwent another independent parallel review by two investigators. In addition to the exclusion criteria used for the abstract review, there were additional exclusion criteria for each KQ. For the question about clinical validity of the mutations (KQ 3), we included only prospective studies of probands, although we permitted retrospective studies of family members because we anticipated few prospective studies. Each article underwent double review by study investigators for full data abstraction and assessment of study quality. We used a sequential review process in which the primary reviewer completed all data abstraction forms, and the second reviewer checked the first reviewer’s data abstraction forms for completeness and accuracy. Reviewer pairs were formed to include personnel with both clinical and methodological expertise. All information from the article review process was entered directly into the SRS 4.0 database.

The primary outcome extracted from the studies of analytic validity was concordance between the test and the reference test, as that was most often reported. When there were sufficient data (three or more studies) and the studies were qualitatively homogeneous with respect to key variables (population characteristics, study duration, mutation status, and length of follow-up), we conducted meta-analyses for the studies addressing the clinical validity of the tests. When it was inappropriate to combine studies quantitatively, we qualitatively summarized the results. For pooling, we used the number of events and count of the patients under observation in each group. We calculated a pooled estimate of the odds ratio for VTE in probands and separately in family members. We used a random effects model with the DerSimonian and Laird method for calculating between-study variance.

At the completion of our review, we graded the quantity, quality, and consistency of the best available evidence addressing the KQs by adapting an evidence-grading scheme recommended by the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) Working Group. To assess the quantity of evidence, we focused on the number of studies with the strongest design. We also assessed the quality and consistency of the best available evidence, including assessment of the limitations affecting individual study quality (using the individual study quality assessments), certainty regarding the directness of the observed effects in the studies, the precision and strength of the findings, and the availability (or lack) of data to answer the KQ. We classified evidence bodies as: (1) “high” grade, indicating confidence that further research is very unlikely to change our confidence in the estimated effect in the abstracted literature; (2) “moderate” grade, indicating that further research is likely to have an important impact on our confidence in the estimates of effects and may change the estimates in the abstracted literature; (3) “low” grade, indicating the further research is very likely to have an important impact on confidence in the estimates of effects and is likely to change the estimates in the abstracted literature.

Results

We reviewed 7,777 titles and included 124 articles in the review of one or more of the KQs.

Key Question 1

We identified no evidence to directly address KQ1, which required studies that directly tested the impact of testing on patient outcomes. The articles addressing the questions below, particularly KQ4, provided indirect evidence to answer KQ1.

Key Question 2

The conventional “gold standard” method for FVL and prothrombin G20210A detection is the sequencing of the specific genetic region of the gene of interest. However, a number of other reference methods are used instead because of the complexity and high costs associated with sequencing. Many studies that we reviewed used polymerase chain reaction and restriction fragment length polymorphism (PCR-RFLP) or allele-specific polymerase chain reaction (AS-PCR) assays as their reference standards, both of which are considered by the Food and Drug Administration to be acceptable reference standards.

Detection of FVL. Forty-one studies compared at least two methods for FVL detection. The majority of studies demonstrated 100 percent concordance between the experimental method and the reference method. The least concordance was seen in a study using electrochemical genosensors, which had a concordance of only 93 percent.

Detection of prothrombin G20210A. The concordance rates between the experimental methods and the reference standards for the detection of prothrombin G20210A were 100 percent in nearly all of the 23 studies.

Simultaneous detection of FVL and prothrombin G20210A. All 12 studies that employed multiplex technologies for the simultaneous detection of the two mutations reported 100 percent concordance between this approach and the matched reference standard.

Quality assurance. We identified three studies that addressed external quality assurance or laboratory performance relative to the gold standards. One described the results from the United Kingdom National External Quality Assessment Scheme (UK NEQAS) Thrombophilia Screening Program. Two hundred eighty centers participated in the thrombophilia screening exercises. For the centers performing genetic analysis for FVL and the prothrombin mutation, an error rate of 3 to 6 percent was identified, with both transcription-related and analytical errors observed.

Another study described results from the Royal College of Pathologists of Australasia external quality assurance (QA) program. This program sent 133 DNA samples with known mutations to laboratories in 10 separate surveys. Of 3,799 responses, the overall successful identification rate was 98.6 percent. Success rates in identifying specific mutations were 98.1 percent for FVL and 98.8 percent for prothrombin G20210A.

Finally, a survey was organized by the Subcommittee on Hemostasis of the Italian Committee for Standardization of Laboratory Tests (CISMEL). The authors concluded that regular quality control programs are warranted to identify the causes of failures to correctly identify specimens.

Key Question 3

Probands. Twenty-two articles examined the rates of recurrent venous thromboembolism in individuals with a history of VTE (probands) and the FVL mutation.

We pooled 13 studies comparing probands with heterozygous FVL to probands without this mutation, yielding an odds ratio for recurrence of 1.56 (95 percent confidence interval [CI], 1.14 to 2.12). The annualized event rates among heterozygous individuals were 3.1 percent to 7.5 percent in three studies. Seven studies described event rates in probands homozygous for FVL and those without mutations. The number of homozygous individuals ranged from 1 to 11 across studies. The odds ratio of recurrence was 2.65 (95 percent CI, 1.2 to 6.0).

Eight studies did not specify if the participants were heterozygous or homozygous for FVL. The pooled odds ratio (from 3 studies) associated with this unspecified status was 1.56 (95 percent CI, 0.75–3.2), very similar to the odds ratio for known heterozygous individuals.

We identified 18 articles that examined rates of recurrent VTE in probands with the prothrombin G20210A mutation. Nine of these compared individuals heterozygous for the mutation to individuals without mutation. The pooled odds ratio was 1.45 (95 percent CI, 0.96–2.2). Two studies described individuals homozygous for this mutation, but there were only 3 individuals with homozygosity. None of the three had a recurrent thrombosis.

Seven studies did not specify whether the individuals had homozygous or heterozygous prothrombin G20210A. The pooled odds ratio in the four studies with available data was 0.73 (95 percent CI, 0.37–1.4).

Doubly heterozygous individuals (with FVL and the prothrombin G20210A mutations) had an odds ratio of 4.8, with a wide confidence interval (95 percent CI, 0.5 to 46.)

When we separately evaluated patients with idiopathic VTE as the index event, we found that the odds ratio associated with heterozygous FVL was close to one (1.17 [95 percent CI, 0.63–2.18]).

Family members. We identified 17 articles that evaluated the occurrence of venous thrombosis in family members of probands. The majority of these were retrospective studies, although three included a prospective component. Nine studies described results for family members who were heterozygous for FVL. Six studies contributed to the pooled odds ratio, which was 3.5 (95 percent CI, 2.5 to 5.0). Homozygosity for FVL among family members had a pooled odds ratio of 18 (95 percent CI, 7.8 to 40) in the five studies pooled. The six studies that did not specify whether individuals were homozygous or heterozygous had a pooled odds ratio of 2.9 (95 percent CI, 1.8–4.8).

Only three studies evaluated heterozygosity for prothrombin G20210A in family members. The odds ratio was 1.9 (95 percent CI, 0.35–10). There was only a single study of family members with homozygosity for this mutation.

Doubly heterozygous family members had a pooled odds ratio of 6.7 (95 percent CI, 2.9 to 15) in the three studies with data to pool.

A sizeable subgroup of family members had VTE associated with pregnancies. Four articles exclusively addressed the risk of VTE attributable to FVL and prothrombin G20210A during pregnancies of family members, and one additional study of family members had useable pregnancy data.

Two of these studies, which included few women, evaluated the risk associated with homozygous FVL. The odds ratios for venous thrombosis were 16 (95 percent CI, 0.9 – 278) and 41(95 percent CI, 5.5–419). The odds ratios associated with heterozygous FVL in two studies were 5.4 (95 percent CI, 0.65–46) and 3.4 (95 percent CI, 0.35 to 33). The rates of events were low, at 2.5 percent per pregnancy (95 percent CI, 0.9–5.4) and 1.5 percent per pregnancy (95 percent CI, 0.5–4.3), respectively. Two studies evaluated the risk associated with heterozygosity for the prothrombin G20210A mutation. The rates of VTE were low, at 0.3 percent per pregnancy (95 percent CI, 0.1 to 1.6) and 1 percent per pregnancy (95 percent CI, 0.2–3.6), respectively. The odds ratios in these two studies were close to 1.

Individuals who were doubly heterozygous and pregnant were evaluated in two studies. The larger of the two studies reported an odds ratio similar to the odds ratios associated with heterozygosity for either mutation (odds ratio, 4.1; 95 percent CI, 0.37–46). The smaller study reported an odds ratio that was identical to that for homozygous FVL (odds ratio, 16; 95 percent CI, 0.9–278).

Key Question 4

Effect of testing on clinicians’ management. We found a single study addressing how physicians’ management decisions are affected by FVL testing. Canadian obstetrical care providers (N=662) were asked about management recommendations in response to four clinical scenarios involving pregnant women with FVL. For scenarios involving asymptomatic women, if the patient was described as having a family history of VTE, the percentage of doctors recommending thromboprophylaxis was twice the percentage recommending it for women lacking a family history (58 versus 26–34 percent).

Effect of management, stratified by test results, on VTE-related outcomes. No studies directly addressed the effect of testing on outcomes. We therefore also included the four studies that described VTE recurrence rates during anticoagulation among probands with FVL or prothrombin G20210A. Two studies investigated the effect of warfarin on recurrence rates, and one the effect of ximelagatran; one did not specify the treatment that patients received.

One study assessed thromboembolism recurrence rates among individuals with FVL or prothrombin G20210A who received either a low-intensity warfarin regimen or placebo. Low-intensity warfarin reduced the rate of recurrence among thrombophilic patients by 75 percent. This risk reduction, however, was not significantly different than the 58 percent reduction seen among patients without either mutation.

In one study, the recurrence rate for patients with FVL while receiving low- or conventional-intensity warfarin therapy was 0.8 percent per year (95 percent CI, 0.2 to 2.2). This rate was not statistically different from the rate among those participants without FVL (hazard ratio 0.7; 95 percent CI, 0.2 to 2.6).

In another study, 2 of 111 FVL carriers receiving ximelagatran had recurrences of VTE, as compared to 16 of the 125 with mutations who were assigned to receive the placebo; this difference in recurrence rates was described as statistically significant. Recurrence rates with treatment were very similar for individuals without FVL or prothrombin G20210A, There was no interaction between FVL or prothrombin G20210A and the effect of ximelagatran in preventing recurrent VTE (p-value for the interaction, 0.92 for FVL and 0.98 for prothrombin G20210A).

Finally, one group studied the rate of recurrence of VTE in a cohort of 304 patients in thrombophilic families, according to whether they were or were not receiving long-term anticoagulation. Long-term anticoagulation decreased the rate of VTE recurrence among these probands. A quantitative estimate of the risk reduction was not provided, nor were the details of the anticoagulation regimens available

Effect of testing and results on other outcomes. Four studies addressed how probands’ and family members’ knowledge, behaviors, and healthcare experiences were affected by their being tested for FVL or prothrombin G20210A. Two studies employed cross-sectional surveys of convenience samples of probands and family members to assess risk perception and behavioral effects following genetic testing. The other two studies used qualitative, structured interviews of probands and relatives to describe their experience during the process of testing as well as their interpretation of the results.

One group surveyed the perception of VTE risk and changes in behavior following testing for FVL or prothrombin G20210A among first-degree relatives of probands. More mutation carriers recognized trauma as a risk factor for VTE than did non-carriers, but otherwise there were no statistically significant differences between the two groups regarding their recognition of risk factors for VTE. Behavior changes following testing were uncommon in both groups.

One group investigated whether the type of thrombophilic mutation and history of VTE affect the perception of risk and level of worry among probands or their relatives with FVL, as compared to other thrombophilic mutations. Individuals with a history of VTE had an increased perception of risk and worried more about VTE than did individuals without prior VTE. However, worry and risk perception were not measured in non-carriers for comparison.

One study involved a qualitative study of asymptomatic relatives of probands with FVL to assess their overall experience with the testing process and how the results affected their daily lives. Among the 17 participants, most found that the testing process itself was not stressful; all had received written information about the test prior to testing. Although the majority of participants indicated that the testing had not altered their daily lives, many wanted to screen their children to decrease their risk of VTE from pregnancy or oral contraceptive use.

One group (in two studies) assessed the level of understanding of the testing process and the implications of the results among probands and relatives referred for FVL testing by their primary care doctor or specialist. Most participants did not consider thrombophilia testing to be different from other tests ordered by their providers. Most participants did not incorporate behaviors to reduce their risk for VTE into their daily routines as a result of the testing, although most participants who were aware of their positive status stated they had undergone testing to inform their decision about whether to take hormonal therapies, or to advise relatives on the matter.

Cost-effectiveness of FVL and prothrombin G20210A testing in the care of probands and their relatives. We identified six studies that assessed the cost-effectiveness of genetic testing and resultant changes in management and one study that assessed only effectiveness. The cost-effectiveness studies all used decision analytic models, which can provide support for further investigation of the utility of an intervention if the assumptions in the models are compatible with actual practice. The data ranges explored in the sensitivity analyses demonstrated the variables to which the cost-effectiveness of the interventions were most sensitive.

One group used decision analysis to assess the cost-effectiveness of testing for FVL and extending warfarin anticoagulation for 3 years or for life in carriers following a first VTE in a hypothetical cohort of 35-year-old women. If the rate of recurrence remained constant (7.3 percent/year), lifelong anticoagulation was the more cost-effective strategy (incremental cost-effectiveness ratio [ICER] = $16,823/quality-adjusted life year [QALY]) when compared to no testing and 6 months of anticoagulation). Lifelong anticoagulation was less cost-effective in patient populations with low FVL prevalence, low risk of recurrent VTE, or risk factors for bleeding on anticoagulant therapy.

Another group used decision analysis to assess the cost-effectiveness of testing for FVL and 2 years of warfarin anticoagulation in carriers following a VTE in a hypothetical cohort of 60-year-old men. FVL testing with 2 years of anticoagulation for carriers was a cost-effective strategy (ICER = $12,833/QALY) when compared to no testing and 6 months of anticoagulation. However, this intervention was not cost-effective for individuals with a high risk of fatal bleeding on warfarin, low VTE recurrence rate, low anticoagulation efficacy, or low anticoagulation compliance.

Building on the previous study, the author employed the same model to assess the cost-effectiveness of testing for double heterozygosity, followed by 2 years of warfarin anticoagulation for doubly heterozygous individuals. Testing for both mutations was cost-effective (ICER = $13,624/QALY) when compared to no testing and 6 months of anticoagulation. Testing was not cost-effective for patient populations with a high bleeding risk, low double-heterozygote prevalence, low levels of pulmonary embolism or mortality, or low anticoagulation efficacy.

Another group assessed the cost-effectiveness of a hypercoagulability testing panel and warfarin anticoagulation for 6, 12, 18, 24, or 36 months, or for life, following an apparently idiopathic deep venous thrombosis (DVT) in a hypothetical cohort of 40 year- olds. In the base case analysis, extending anticoagulation for 24 months following a positive test was the most cost-effective option (ICER = $11,100/QALY) when compared to the least costly option of not testing and treating for 24 months. The authors concluded that tests detecting disorders present in at least 5 percent of the population that confer a relative risk exceeding 1.25, including FVL and prothrombin G20210A, should be included.

Another decision analysis with a 5-year time horizon assessed the effectiveness of extending anticoagulation from 3 months for FVL carriers and non-carriers following an initial lower-limb DVT to 1, 2, 3, 4, or 5 years in a hypothetical cohort. The authors stated that the risk of bleeding must be below 2.5 percent/year in order for prolonged anticoagulation to be the more effective strategy.

Finally, one group used a decision-analytic model with a 12-month time horizon to assess the cost-effectiveness of universal or selective screening for FVL and resultant changes in management for carriers in four cohorts at high risk of VTE. In all four cohorts, selective screening was more cost-effective than universal screening.

One group used the cost and outcomes data from a prospective cohort of 967 pregnant women in the United Kingdom to assess the cost-effectiveness of FVL testing and enoxaparin anticoagulant prophylaxis to prevent pregnancy-related vascular complications over the 8-month time horizon, from 12 weeks of gestation to 6 weeks postpartum. No women actually received anticoagulant prophylaxis, but the hypothetical impact of treating FVL carriers with an assumed efficacy of 50 percent was modeled. Testing only those women with a personal or family history of VTE was the most cost-effective approach.

Discussion

Based on our review of the evidence, we graded the strength of the evidence for the key questions as follows:

Key Question 1

There was no direct evidence that testing for FVL or the prothrombin G20210A mutations improves clinical outcome in adults with a personal history of VTE or that it improves clinical outcomes in adult family members of mutation-positive individuals. The evidence supporting KQ 2 through 4 can be considered indirect evidence to answer this overarching question.

Key Question 2

  • There was high-grade evidence that tests to detect FVL have excellent analytic validity.
  • There was high-grade evidence that tests to detect prothrombin G20210A have excellent analytic validity.
  • There was high-grade evidence that most, but not all, clinical laboratories can test for FVL and prothrombin G20210A very accurately. There may be some laboratories that, for technical or administrative reasons, report inaccurate results.

We note that the majority of the tested assays are not presently in widespread use. The majority of U.S. laboratories use PCR or invader chemistry technologies.

Key Question 3

  • There was moderate-grade evidence that homozygosity for FVL in probands is predictive of recurrent VTE.
  • There was moderate-grade evidence that heterozygosity for FVL in probands is predictive of recurrent VTE.
  • The evidence is insufficient regarding the predictive value in probands of homozygosity for prothrombin G20210A, which is a rare genotype.
  • There was moderate-grade evidence that heterozygosity for prothrombin G20201A in probands is not predictive of VTE.
  • The evidence is insufficient regarding the predictive value in probands of double heterozygosity (FVL and prothrombin G20210A).

We note that there may be little predictive value in knowing the mutation status in patients with idiopathic VTE as the index event, since the odds ratios for this subgroup were close to one.

  • There was high-grade evidence that homozygosity for FVL in family members is predictive of VTE.
  • There was moderate-grade evidence that heterozygosity for FVL in family members is predictive of VTE.
  • The evidence is insufficient regarding the predictive value in family members of homozygosity for prothrombin G20210A, which is a rare genotype.
  • The evidence is insufficient regarding the predictive value in probands of heterozygosity for prothrombin G20210A.
  • There was low-grade evidence that double heterozygosity (FVL and prothrombin G20210A) in family members is predictive of VTE.
  • There was low-grade evidence that homozygosity for FVL in pregnant family members is predictive of VTE.
  • The evidence was insufficient regarding the predictive value in pregnant family members of heterozygosity for FVL.
  • The evidence was insufficient regarding the predictive value in pregnant family members of homozygosity for prothrombin G20210A.
  • The evidence was insufficient regarding the predictive value in pregnant family members of heterozygosity for prothrombin G20210A.
  • The evidence was insufficient regarding the predictive value in pregnant family members of double heterozygosity (FVL and prothrombin G20210A.)

For clinical context, we note that the annualized rate of venous thromboembolic events for family members without a mutation was approximately 0.1 percent per year. This translates to an event rate in heterozygous family members of 0.3 percent per year, or an absolute increase of 0.2 percent per year (a change from an average of 1/1000 person-years to 3/1000 person-years).

Key Question 4

  • There was no direct evidence that management guided by test resultsreduces VTE related-outcomes in individuals who have had VTE or in the probands’ family members who have been tested.
  • There was low-grade evidence that physicians may alter patient management based on the results of testing for FVL or prothrombin G20210A.
  • There was high-grade evidence that anticoagulation can reduce recurrent events in patients with FVL or prothrombin G20210A; however, there was only low-grade evidence that the relative reduction in risk is comparable to that seen in individuals without these mutations.
  • There was moderate-grade evidence that neither harms nor benefits have been conclusively demonstrated in individuals with VTE or in their family members when tested for FVL or prothrombin G20210A.
  • There was low-grade evidence, derived from models, that testing for FVL alone, prothrombin G20210A alone, or the two tests in combination may be cost-effective when caring for selected patients with VTE (those with a high risk of recurrence and/or low risk of bleeding) or their family members.

Limitations

In addition to the reported deficits in the literature, there are limitations to this report. In our assessment of clinical validity, we chose to pool odds ratios rather than time-dependent measures of recurrence (such as hazard ratios or incident rate ratios). This approach necessarily excluded some studies from the pooled estimates.

The odds ratios should approximate the relative rates of events in most studies, as these were relatively rare outcomes. We pooled the results using the DerSimonian and Laird random effects methods; this is a conservative method that often results in wide confidence intervals.

Many of these studies were observational studies, and physicians may have altered their management based on their knowledge of mutation status, thereby changing the likelihood of a particular outcome. Most studies mitigated this potential difficulty by excluding patients who were chronically anticoagulated or by using a pre-defined anticoagulation approach. In those studies that reported the duration of anticoagulation after the index event in mutation-positive and -negative subgroups, there was no obvious discordance in anticoagulation duration.

In these cohort studies, ascertainment bias is possible. In the studies of probands, the individuals were not blinded to their mutation status. Patients with mutations may be more likely to seek medical attention for symptoms consistent with VTE and might have been over-diagnosed with recurrence (as a result of false-positive tests), and those without mutations might have been under-diagnosed (because they did not seek medical attention for a thrombotic event that ultimately resolved without therapy). Ascertainment bias would tend to augment the association between the mutations and recurrent thrombosis. None of the studies we included had scheduled periodic radiographic testing to limit the potential for ascertainment bias.

The majority of the observational studies about family members were retrospective, with some notable exceptions. Retrospective studies are prone to important biases, including recall bias. Although such bias can be mitigated by interviewing participants before they have knowledge of their mutation status, this approach was variably described in these studies.

Implications for Future Research

Studies to directly address the overarching question (KQ 1) would ideally be designed as randomized trials, in which participants with venous thrombosis and/or their family members would be randomized to a test arm or a no-test arm. Individuals would be managed by their physicians on the basis of the results of the testing (with evidence-based recommendations). Sufficient follow-up time would be included so that venous thromboembolic events could be witnessed and compared between the tested and untested groups.

Analytic Validity

Although the mutation detection methods have been shown to have high analytic validity, a small minority of laboratories account for a disproportionate percentage of errors in the performance of these tests.

  • There is a need for ongoing programs aimed at monitoring molecular diagnostic laboratories, through quality assurance programs, to ensure the consistent provision of high-quality genetic testing services.

Clinical Validity

  • Future studies should report event rates over time (and relative rates of recurrence between specified groups), rather than just the number of events.
  • Studies should consistently differentiate between heterozygosity and homozygosity.
  • Studies should continue to use objectively measured thrombosis (radiographically proven) as a criterion for both the index and recurrent thromboses and should include more detail about both the index events and recurrences, such as the precipitants of these events.
  • Additional studies are needed to quantify the effect size more precisely with regard to the prothrombin G20210A mutation (alone and in conjunction with FVL).
  • Future research would be appropriate in Caucasian populations outside of Europe or in other populations with appreciable frequencies of mutations.
  • Future research could better explore the age-mutation interaction.

Clinical Utility

  • Studies should measure how actual clinician practices change in response to results of both FVL and prothrombin G20210A testing.
  • Studies should be powered sufficiently to evaluate the risks associated with prolonged anticoagulation, as they relate to patients with specific thrombophilic mutations.
  • Future studies in both probands and family members might focus on whether management decisions (duration of therapy, use of thromboprophylaxis) affect the rates of VTE, particularly during times of heightened thromboembolic risk.
  • Studies based in the United States may give a clearer understanding of how patients here might respond to the testing process and results.
  • Larger sample sizes should also be used to increase the ability to detect rarer events, such as stigmatization.
  • Efforts should be made to recruit representative patient populations, and relevant comparison groups should be included (e.g., carriers and non-carriers) to increase the practical applications of the study findings. Quantitative studies may be preferable, using standardized, validated questionnaires to evaluate patients’ experiences.
  • Clinical trials could include an assessment of the costs associated with a testing strategy, as compared to care without testing.

Our literature review included articles through December 2008. We do not anticipate any important secular changes in the event rate that would markedly change the event rates in upcoming years. We also do not expect major changes in the coming years in terms of the methods used to detect mutations. The most anticipated change would be an increasing in options to reduce risk as new drugs become available. Future research will need to include an evaluation of the risks and benefits associated with use of new anticoagulant drugs in probands and family members at high risk of thromboembolic events.