Global Coagulation Assays in Transgender Women on Oral and Transdermal Estradiol Therapy

Abstract

Abstract Context The thrombotic effects of estradiol therapy in transgender women are unclear. Global coagulation assays (GCA) may be better measures of hemostatic function compared with standard coagulation tests. Objective To assess the GCA profiles of transgender women in comparison to cisgender controls and to compare how GCA differ between routes of estradiol therapy in transgender women. Design Cross-sectional case-control study. Setting General community. Participants Transgender women, cisgender male and cisgender female controls. Main outcome measures Citrated blood samples were analyzed for (i) whole blood thromboelastography (TEG®5000), (ii) platelet-poor plasma thrombin generation (calibrated automated thrombogram); and (iii) platelet-poor plasma fibrin generation (overall hemostatic potential assay). Mean difference (95% confidence intervals) between groups are presented. Results Twenty-six transgender women (16 oral estradiol, 10 transdermal estradiol) were compared with 98 cisgender women and 55 cisgender men. There were no differences in serum estradiol concentration (P = 0.929) and duration of therapy (P = 0.496) between formulations. Transgender women demonstrated hypercoagulable parameters on both thromboelastography (maximum amplitude + 6.94 mm (3.55, 10.33); P < 0.001) and thrombin generation (endogenous thrombin potential + 192.62 nM.min (38.33, 326.91); P = 0.009; peak thrombin + 38.10 nM (2.27, 73.94); P = 0.034) but had increased overall fibrinolytic potential (+4.89% (0.52, 9.25); P = 0.024) compared with cisgender men. No significant changes were observed relative to cisgender women. Route of estradiol delivery or duration of use did not influence the GCA parameters. Conclusion Transgender women on estradiol therapy demonstrated hypercoagulable GCA parameters compared with cisgender men with a shift towards cisgender female parameters. Route of estradiol delivery did not influence the GCA parameters.