Objective-Activated protein C (APC) resistance not related to the factor V Leiden mutation is a risk factor for venous thrombosis. Oral estrogen replacement therapy (ERT) has been reported to induce APC resistance. Little is known about the effect of transdermal estrogen. Methods and Results-We enrolled 196 postmenopausal women who were randomly allocated to receive either 1 mg 17␤-estradiol orally (nϭ63) or 50 g 17␤-estradiol transdermally per day (nϭ68), both associated with 100 mg progesterone daily or placebo (nϭ65) for 6 months. An activated partial thromboplastin time (APTT)-based test and the effect of APC on thrombin potential (ETP) were used. Oral ERT induced an ETP-based APC resistance compared with both placebo (Pϭ0.006) and transdermal ERT (PϽ0.001), but there was no significant effect of transdermal ERT compared with placebo (Pϭ0.191). There was no significant effect of ERT on the APTT-based APC sensitivity ratio. Prothrombin fragment 1ϩ2 plasma levels were significantly higher after 6 months of treatment in women allocated to oral ERT compared with those on placebo and transdermal ERT and were positively and significantly correlated with changes in ETP-based APC sensitivity ratio. Conclusions-Our data show that oral, unlike transdermal, estrogen induces APC resistance and activates blood coagulation. These results emphasize the importance of the route of estrogen administration. Key Words: hormone replacement therapy Ⅲ APC resistance Ⅲ blood coagulation Ⅲ randomized trial Ⅲ factor V S everal observational studies [1][2][3][4][5][6][7] found that oral estrogen replacement therapy (ERT) was associated with a 2-fold increased risk of venous thromboembolism (VTE). This finding was confirmed in 2 randomized clinical trials. 8,9 Furthermore, consistent data provided evidence that oral ERT resulted in coagulation activation. 10 -14 However, studies investigating the effect of transdermal estrogen on the thrombotic process are scarce. 3,15 Activated protein C (APC) resistance has recently emerged as a risk factor for venous thrombosis. 16,17 In most cases, this defect is related to the presence of the R506Q mutation in factor V (FV) Leiden. 18 However, an APC-resistant phenotype detected in the absence of FV Leiden is also an independent risk factor for venous thrombosis. 19 Observational studies 20 -22 and 1 randomized trial 23 showed that oral ERT could induce an acquired APC resistance. Little is known about the effect of transdermal ERT on APC resistance. Therefore, we conducted a randomized, placebocontrolled trial that investigated the effect of both oral and transdermal estrogen/progesterone regimens on the anticoagulant response to APC and on coagulation activation. Methods Study Design and SettingThis study was a randomized, double-blind, placebo-controlled, parallel-group trial that took place at the Hôpital de la Cavale Blanche, Brest, France, between September 1999 and August 2001. Participants were followed up by regular visits, at randomization (baseline), and after 6 months. A medical revi...

Summary Bleeding risk is not predictable in patients with factor XI (FXI; F11) deficiency. In this prospective study, our objectives were to determine the biological determinants for bleeding risk in patients with heterozygous FXI deficiency. Patients were classified as either bleeding patients or non‐bleeding patients by calculating the bleeding score (BS) described for von Willebrand disease. Primary haemostasis, thrombin generation, thromboelastometry, procoagulant proteins, inhibitors, fibrinolysis, and F11 gene mutations were compared between bleeding and non‐bleeding patients. Thirty‐nine patients were included. BS significantly correlated with clinical assessment (P = 0·001), and a score over 3 discriminated between bleeding (n = 15) and non‐bleeding (n = 24) patients (P = 0·034). Despite normal values, von Willebrand factor (VWF) and thrombomodulin (TM) plasma levels were significantly lower in bleeding patients than non‐bleeding patients [ristocetin cofactor activity (VWF:RCo) = 80·6 ± 29·7 iu/dl and 101·8 ± 29·5 iu/dl respectively, P = 0·043; and VWF antigen (VWF:Ag) = 84·0 ± 28·0 iu/dl and 106·3 ± 36·1 iu/dl respectively, P = 0·035; and TM = 17·7 ± 11·7 ng/ml and 23·6 ± 9·7 ng/ml respectively, P = 0·043]. When considering BS as a continuous variable, only VWF:RCo remained significant (P = 0·042), which accounted for 11% of the variability in BS.