The antithrombin A384S mutation has a relatively high frequency in the British population but has not been identified in other populations. This variant has been associated with cases of thrombotic disease, but its clinical relevance in venous thrombosis remained unclear. We have conducted a secondary analysis of the prevalence of the mutation in a large case-control study, including 1018 consecutive Spanish patients with venous thromboembolism. In addition, we evaluated its functional consequences in 20 carriers (4 homozygous). This mutation, even in the homozygous state, did not affect anti-Xa activity or antigen levels, and it only slightly impaired anti-IIa activity. Thus, routine clinical methods cannot detect this anomaly, and, accordingly, this alteration could have been underestimated. We identified this mutation in 0.2% of Spanish controls. Among patients, this variant represented the first cause of antithrombin anomalies. Indeed, 1.7% patients carried the A384S mutation, but 0.6% had any other antithrombin deficiency. The mutated allele was associated with an increased risk of venous thrombosis with an adjusted OR of 9.75 (95% CI, 2.2-42.5). This is the first study supporting that antithrombin A384S mutation is a prevalent genetic risk factor for venous thrombosis and is the most frequent cause of antithrombin deficiency in white populations. (Blood. 2007;109: [4258][4259][4260][4261][4262][4263]

The protein Z-dependent protease inhibitor (ZPI) is a hemostatic serpin with anticoagulant activity. As for antithrombin, deficiency of ZPI could have relevant thrombotic consequences. We have studied 6 genetic modifications affecting the ZPI gene, identifying 5 haplotypes. Haplotype H5 is featured by a stop codon at position 67. The relevance of these genetic modifications and haplotypes in venous thrombosis was evaluated in a case-control study including 1018 patients and 1018 age-and sex-matched controls. Surprisingly, the H5 haplotype was found in 0.9% of controls, supporting that the Arg67Stop change is a low frequency nonsense polymorphism. The prevalence of this haplotype increased significantly in patients (3.0%), one of whom was in a homozygous state. Multivariate analysis confirms that carriers have a 3.3-fold risk of developing venous thrombosis (P ‫؍‬ .002; 95% CI: 1.5-7.1). Moreover, we observed a significant association of this polymorphism with familial history of thrombosis (P < .001). Our study supports that the ZPI Arg67Stop nonsense polymorphism might be an independent genetic risk factor for venous thrombosis. This polymorphism has slightly lower prevalence but similar thrombotic risk than the FV Leiden or prothrombin 20210A. Although further studies are required, all available data support that the ZPI is a candidate to play a significant role in thrombosis and should be evaluated in thrombophilic studies. (Blood. 2006;108:177-183)