• epidemiology;
  • thrombophilia;
  • thrombosis risk factors;
  • venous thrombosis

See also Keeling D. Thrombophilia screening or screaming. This issue, pp 1191–2.

Summary. Background:Absolute risks of venous thromboembolism (VTE) in protein S-, protein C-, or antithrombin-deficient subjects are mainly based on retrospective data. Screening asymptomatic relatives of these patients is disputed, though studies addressing this issue have yet to be conducted. Methods: We prospectively followed 382 relatives of 84 probands. Participants were assessed for other thrombophilic defects and occurrence of exogenous risk factors (i.e. surgery/trauma/immobilization, malignancies, use of systemic estrogens, and pregnancy/puerperium). After screening, deficient subjects were advised to use thromboprophylaxis during exogenous risk factors; use of oral contraceptives was discouraged. Results: Overall annual incidence of VTE was 1.53% (95% CI, 1.00–2.34) in deficient vs. 0.29% (0.13–0.64) in non-deficient relatives; adjusted hazard ratio, 7.0 (95% CI, 2.7–18.0). Annual incidence of unprovoked VTE was 0.95% in deficient vs. 0.05% in non-deficient subjects; age-adjusted hazard ratio, 22.3 (= 0.003). In contrast, annual incidence of provoked VTE was 0.58% vs. 0.24%; age-adjusted hazard ratio, 2.8 (= 0.08). Fifty-five (37%) deficient and 80 (34%) non-deficient subjects experienced 91 and 143 exogenous risk factors, respectively, during which six vs. five VTEs (6.6% vs 3.5% per risk-period) occurred, despite the higher compliance with recommended thromboprophylaxis use in deficient (51%) vs. non-deficient (22%) subjects. In deficient subjects all provoked VTEs occurred when thromboprophylaxis was not used. Conclusions:Protein S, protein C or antithrombin deficiencies confer high absolute risk of VTE. Screening and subsequent augmentation of thromboprophylaxis use may result in reduction of provoked VTE, whereas risk of unprovoked VTE could not be affected by screening.