Venous thromboembolism (VTE), comprising a deep vein thrombosis (DVT) and a pulmonary embolism (PE), is a common potentially lethal condition [1,2]. Treatment with oral anticoagulation (OAC) therapy is effective at reducing recurrent VTE, but long-term oral anticoagulation therapy needs to be counterbalanced with the risk of major bleeding. Identification of subgroups of patients with unprovoked VTE that have lower and higher risk of recurrent VTE is important to help tailor the length of anticoagulation in this population.

Several genetic defects (i.e. thrombophilias) have been shown to be associated with an increased risk of VTE. However, it is likely that many inherited thrombophilias have not yet been identified, and as a consequence it has been suggested that a family history could be a better and potentially more comprehensive predictor for VTE and recurrent VTE. Several previous studies have shown that family history is a predictor of first VTE [3–7]. Few studies report on the value of family history as a predictor of recurrent VTE in those who have suffered a first unprovoked VTE. We sought to determine whether the presence of a positive family history is associated with a heightened risk of recurrent VTE in patients with unprovoked VTE from the Recurrent Venous thromboembolism Risk Stratification Evaluation (REVERSE) cohort.

We used data from the REVERSE study, a prospective cohort study which recruited patients with a first objectively proven unprovoked VTE treated with anticoagulation therapy for 5–7 months [8]. Unprovoked VTE was defined as occurring in the absence of a leg fracture or lower-extremity plaster cast, immobilization > 3 days, surgery using general anesthesia 3 months previous to the index event and no malignant disease in the past 5 years. Methods were previously described in 2008 [8]. Note that patients with a known ‘high-risk’ thrombophilia were excluded from the study. The latter included deficiency of protein S, protein C or antithrombin, persistently positive anticardiolipin antibodies (titers > 30 U mL−1) and lupus anticoagulant, and combined thrombophilic defects (e.g. homozygous factor V Leiden [FVL] or prothrombin gene mutation [PGM] or compound heterozygote for FVL and PGM).

At the baseline visit, each subject completed a detailed questionnaire inquiring about a family history of VTE in both first- and second-degree relatives. This information included the number of relatives who had been diagnosed with VTE, how many events each had suffered and whether the event was unprovoked or secondary.

Patients were seen every 6 months and were instructed to contact study personnel if they suspected a recurrent event. The criteria for the diagnosis of a recurrent VTE (primary outcome) were previously described [8] and validated [9]. The follow-up time was calculated according to the last follow-up, which corresponded to a recurrent event, the end of the REVERSE study (8 years follow-up completed), death or withdrawal from the study. If patients chose to withdraw their consent or if they had to restart anticoagulation therapy (for reasons other than a recurrent event), their data was censored.

Univariate and multivariate Cox regression analyses were performed to compare rates of recurrent VTE. In the multivariate model, adjustment for known risk factors of VTE recurrence was performed, including age, gender, body mass index, FVL, prothrombin gene variant, post-thrombotic syndrome signs and symptoms (hyperpigmentation, edema and redness), D-Dimer (at baseline, on OAC), hormone replacement therapy and oral contraceptives.

Patients self-reported their family history in a questionnaire provided by study personnel. In the event they did not know if their relative had suffered an unprovoked or secondary event, they simply indicated ‘unknown’. In all our analyses, unknown events were considered ‘secondary’ events. A sensitivity analysis was performed in order to determine if our results would be changed if unknown events were considered ‘unprovoked’.

Recurrence-free survival was estimated with Kaplan–Meier curves. A log-rank test was used to determine if survival varied between any two groups.

Between October 2001 and March 2006, 664 patients were enrolled in the study and 573 subjects were followed until the last follow-up date in December 2010. A total of 649 subjects had at least one follow-up and were therefore included in the analysis. Subjects in this cohort had a mean age of 53 years (min–max 18–95), the mean follow-up time averaged 3.8 years and 316 (48.7%) of the subjects were females. The most predominant ethnicity was Caucasian.

In total, 112 (17.3%) participants had a family history of VTE in a first-degree relative and 147 (22.7%) participants had a family history of VTE including at least one first- or second-degree relative. Among the 649 participants, 142 suspected VTE were adjudicated as recurrent VTE. The rates of recurrent VTE for participants with or without a family history of VTE are shown in Table 1. The rate of recurrent VTE for patients with a family history of VTE in at least one first-degree relative was 4.82 (95% confidence interval [CI]: 3.02–7.30) per 100 patient years. Conversely, the rate of recurrent VTE was 5.95 (95% CI: 4.93–7.11) per 100 patient years for patients without a history of VTE in first-degree relatives. Whether adjusted or unadjusted, there was no difference in the rate of recurrent VTE between patients with and without a family history (Table 1). Similarly, there was no difference in the rate of recurrent VTE when second-degree relatives were included in the analyses (data not shown). Further analyses sought to determine if a significant difference existed in the rate of recurrent VTE when a family history included only unprovoked events (Table 1), or when subjects had a family history with multiple family members affected by VTE (data not shown). Again, the hazard ratios, whether adjusted or unadjusted, showed no significant differences. The aforementioned sensitivity analysis, in which family events of an unknown nature were either considered unprovoked or secondary, did not provide any significant changes to the results.

Table 1.   Rate of recurrent venous thromboembolism (VTE) according to VTE history in first-degree relatives (FDR)
Family historyPatients without recurrent VTE (n = 507)Patients with recurrent VTE (n = 142)Recurrence rate per 100 patient-yearHazard ratio (95% CI)Adjusted hazard ratio (95% CI)*
  1. *Adjustment was done for age, gender, body mass index, factor V Leiden, prothrombin gene variant, hyperpigmentation, edema and redness (HER), D-Dimer levels, hormone replacement therapy and oral contraceptives.

  2. CI, confidence interval.

Negative (i.e. no first degree relative with any type of VTE)4171205.95 (4.93–7.11)ReferenceReference
 Any first degree relative with any type of VTE90224.82 (3.02–7.30)0.84 (0.54–1.33)0.86 (0.53–1.38)
 Any first degree relative with unprovoked VTE49125.18 (2.68–9.05)0.88 (0.49–1.60)0.88 (0.46–1.65)

In Kaplan–Meier analyses, the recurrence-free survival was not different among groups of patients with a family history compared with patients without a family history of VTE (data not shown).

Our study suggests that family history is not a predictor of recurrent VTE in patients with unprovoked VTE. These results support the findings of another previously published large prospective cohort study [10]. While the design of this study is comparable to REVERSE, the family history information reported is not as comprehensive as in our study. For example, only family history of first degree relatives was captured, and the nature of the event (i.e. provoked or secondary) was not reported.

In contrast to the present study, a recently published case–control study suggested an association between a family history of VTE and an increased risk of recurrence in African–American subjects [6]. In this study, a family history of VTE was greater than two-fold higher among African–American subjects with a recurrence compared with those with only one event. However, this association did not reach statistical significance in Caucasian subjects. Unfortunately, subjects in our cohort were mainly Caucasian; hence we could not validate this association.

The detailed family history collected for every REVERSE participant constitutes a strength for our study. We collected family history at the 5–7 month time-point rather than initial diagnosis which we believe likely increases the accuracy of family history data given that patients tend to gather more family information after being diagnosed with VTE. Defining whether a patient has a family history of provoked or unprovoked VTE is relevant clinically, in that a history of an unprovoked event would likely increase the suspicion of a genetic pre-disposition. Also, including second-degree relatives in our analysis could increase the chances of observing a familial pattern. These additional details allowed us to perform many subgroup analyses. Other strengths of this prospective cohort study come from its design. REVERSE remains the largest and most comprehensive study to evaluate risk factors for recurrence in patients with an unprovoked proximal DVT or PE.

This study had limitations. All VTE events were objectively documented; however, no objective documentation was required to confirm events of family members. Family history was reported by each participant, who also had to specify if the event was unprovoked or secondary. Nonetheless, in routine clinical practice, a family history is usually obtained in this fashion and rarely are family events confirmed to be objectively documented. Our study could have been strengthened by using a validated questionnaire for personal or a family history of VTE [11]. On another note, with only a limited number of participants with a family history and many subgroups possible, the size of many subgroups may have been too small to demonstrate an effect if one is present in subgroups (e.g. family history of unprovoked VTE in multiple family members). Furthermore, the prevalence of a family history of VTE in a first-degree relative was recorded for 112 (17.3%) patients in our cohort. In contrast with other previous studies reporting family history in patients with unprovoked VTE, this prevalence is lower. For instance, Hron et al. [10] reported a family history in 23% of patients with unprovoked VTE. It is unlikely but possible that this difference may be explained by the fact that patients with a family history of VTE wished to remain on anticoagulation therapy, potentially decreasing our chances of finding an association with family history and VTE recurrence; however, we do not suspect this selection bias. It is also important to consider that all baseline visits took place 5–7 months after the index event; hence patients who had a recurrent VTE within these months were not included in this study, which could lead to an underestimation of recurrent events.

Our findings do not apply to patients with ‘high-risk’ thrombophilia, as described above. As these patients are at higher risk of recurrence, it could be deemed unsafe for them to discontinue anticoagulation therapy. Because many of these defects are heritable, this could have led us to underestimate the value of family history in predicting VTE recurrence. Patients with ‘known’ FVL and PGM where included in the study and the majority of participants were screened for these thrombophlias before enrolment hence our study applies to this population.

Furthermore, many studies have shown that thrombophilias are weak predictors of recurrent VTE [12,13]. Because family history encompasses both known and unknown thrombophilias, our results suggest that unknown thrombophilias would also be weak predictors of recurrent VTE.

This study shows that a family history of VTE is not a predictor for recurrent VTE and therefore should not be used clinically to identify patients requiring indefinite anticoagulation therapy.

Disclosure of Conflict of Interest

  1. Top of page
  2. Disclosure of Conflict of Interest
  3. Acknowledgement
  4. References
  5. Appendix

The authors state that they have no conflict of interest.


  1. Top of page
  2. Disclosure of Conflict of Interest
  3. Acknowledgement
  4. References
  5. Appendix

The REVERSE study was funded by the Canadian Institutes of Health Research (grant no. MOP 64319) and bioMérieux (through an unrestricted research grant). Dr Marc Carrier is a recipient of the University of Ottawa Research Chair in Cancer and Thrombosis. Dr Marc Rodger is a Heart and Stroke Foundation of Canada Career Scientist, and has received support from the Ontario Ministry of Research and Innovation and holds a University of Ottawa Faculty of Medicine Research Chair in Thrombosis and Thrombophilia. Dr Susan Kahn is a recipient of a National Investigator Award from the Fonds de la Recherche en Santé du Québec. Dr Phil Wells is a recipient of the Canada Research Chair. Dr Mark Crowther holds a Career Investigator Award from the Heart and Stroke Foundation.


  1. Top of page
  2. Disclosure of Conflict of Interest
  3. Acknowledgement
  4. References
  5. Appendix
  • 1
    Anderson FA Jr, Wheeler HB, Goldberg RJ, Hosmer DW, Patwardhan NA, Jovanovic B, Forcier A, Dalen JE. A population-based perspective of the hospital incidence and case-fatality rates of deep vein thrombosis and pulmonary embolism. The Worcester study. Arch Intern Med 1991; 151: 9338.
  • 2
    Carrier M, Le Gal G, Wells PS, Rodger MA. Systematic review: case-fatality rates of recurrent venous thromboembolism and major bleeding events among patients treated for venous thromboembolism. Ann Intern Med 2010; 152: 57889.
  • 3
    Spannagl M, Heinemann LA, Dominh T, Assmann A, Schramm W, Schurmann R. Comparison of incidence/risk of venous thromboembolism (VTE) among selected clinical and hereditary risk markers: a community-based cohort study. Thromb J 2005; 3: 817.
  • 4
    Tosetto A, Frezzato M, Rodeghiero F. Prevalence and risk factors of non-fatal venous thromboembolism in the active population of the VITA project. J Thromb Haemost 2003; 1: 17249.
  • 5
    Dowling NF, Austin H, Dilley A, Whitsett C, Evatt BL, Hooper WC. The epidemiology of venous thromboembolism in Caucasians and African-Americans: the GATE Study. J Thromb Haemost 2003; 1: 807.
  • 6
    Mili FD, Hooper WC, Lally C, Austin H. The impact of co-morbid conditions on family history of venous thromboembolism in Whites and Blacks. Thromb Res 2011; 127: 30916.
  • 7
    Bezemer ID, van der Meer FJ, Eikenboom JC, Rosendaal FR, Doggen CJ. The value of family history as a risk indicator for venous thrombosis. Arch Intern Med 2009; 169: 6105.
  • 8
    Rodger MA, Kahn SR, Wells PS, Anderson DA, Chagnon I, Le Gal G, Solymoss S, Crowther M, Perrier A, White R, Vickars L, Ramsay T, Betancourt MT, Kovacs MJ. Identifying unprovoked thromboembolism patients at low risk for recurrence who can discontinue anticoagulant therapy. CMAJ 2008; 179: 41726.
  • 9
    Le Gal G, Kovacs MJ, Carrier M, Do K, Kahn SR, Wells PS, Anderson DA, Chagnon I, Solymoss S, Crowther M, Righini M, Perrier A, White RH, Vickars L, Rodger M. Validation of a diagnostic approach to exclude recurrent venous thromboembolism. J Thromb Haemost 2009; 7: 7529.
  • 10
    Hron G, Eichinger S, Weltermann A, Minar E, Bialonczyk C, Hirschl M, Stain M, Gartner V, Kyrle PA. Family history for venous thromboembolism and the risk for recurrence. Am J Med 2006; 119: 503.
  • 11
    Frezzato M, Tosetto A, Rodeghiero F. Validated questionnaire for the identification of previous personal or familial venous thromboembolism. Am J Epidemiol 1996; 143: 125765.
  • 12
    Christiansen SC, Cannegieter SC, Koster T, Vandenbroucke JP, Rosendaal FR. Thrombophilia, clinical factors, and recurrent venous thrombotic events. JAMA 2005; 293: 235261.
  • 13
    Baglin T, Luddington R, Brown K, Baglin C. Incidence of recurrent venous thromboembolism in relation to clinical and thrombophilic risk factors: prospective cohort study. Lancet 2003; 362: 5236.


  1. Top of page
  2. Disclosure of Conflict of Interest
  3. Acknowledgement
  4. References
  5. Appendix
All REVERSE investigators

Susan R. Kahn,¶** David A. Anderson,†† Isabelle Chagnon,‡‡ Susan Solymoss,¶ Mark Crowther,§§ Richard H. White,¶¶ Arnaud Perrier,*** Linda Vickars,††† Dr Marc Carrier*§ and Elham Sabri§


*Thrombosis Program, Division of Hematology, Department of Medicine, University of Ottawa, Ottawa, Ontario, Canada

Department of Internal Medicine and Chest Diseases and EA 3878, Brest University Hospital, Brest, France

Division of Hematology, Department of Medicine, University of Western Ontario, London

§Clinical Epidemiology Program, Ottawa Hospital Research Institute, The Ottawa Hospital, Ottawa, Ottawa, Ontario

Department of Medicine, McGill University, Montreal

**Centre for Clinical Epidemiology and Community Studies, Jewish General Hospital, Montreal

††Department of Medicine, Dalhousie University, Halifax, Nova Scotia

‡‡Department of Medicine, Hôpital du Sacré-Coeur de Montréal, University of Montreal, Montreal

§§Department of Medicine, Michael G. DeGroote School of Medicine, McMaster University, Hamilton, Ontario, Canada

¶¶Department of Medicine, UC Davis School of Medicine, Sacramento, CA, USA

***Department of Internal Medicine, Geneva University Hospital, and the Faculty of Medicine, Geneva, Switzerland

†††Department of Medicine, St. Paul’s Hospital, University of British Columbia, Vancouver, British Columbia, Canada