Transgender-Affirming Hormone Therapies, QT Prolongation, and Cardiac Repolarization

Importance: Transgender women (assigned male at birth) usually take antiandrogens associated with estrogens (or are castrated) to induce feminization, whereas transgender men (assigned female at birth) take testosterone to induce masculinization. However, the cardiovascular outcomes of these gender-affirming hormone therapies (GAHTs) remain poorly studied. Objective: To examine the association between GAHT intake and cardiac repolarization alterations on electrocardiography in transgender individuals. Design, Setting, and Participants: In this cohort study, data from a prospective cohort of adult transgender individuals from a single center in France were collected from January 1, 2021, to January 1, 2023. GAHT consisted of injectable testosterone in transgender men and transdermal estradiol with mostly oral cyproterone acetate as antiandrogens in transgender women. Main Outcomes and Measures: Electrocardiographic features, including QTc, T-wave maximal amplitude (TAmp), and QT peak (QTp; distance between Q onset and T peak), were studied. Circulating sex hormones, including total testosterone, estradiol, progesterone, and gonadotrophins, were assessed concomitantly to electrocardiographic intake. Results: In the overall cohort of 120 transgender individuals (mean SD age, 29.7 11.9 years; 64 transgender men and 56 transgender women), mean (SD) QTc was similar between 35 transgender women receiving GAHT (406 20 milliseconds) and 23 transgender men before GAHT (400 16 milliseconds) but prolonged vs 41 transgender men receiving GAHT (378 19 milliseconds) (P < .001) or 21 transgender women before receiving GAHT (384 21 milliseconds) (P < .001). The start of GAHT in 15 transgender women was associated with increased QTc (mean SD, 20 12 milliseconds vs before receiving GAHT; P < .001) and decreased QTc in 18 transgender men (mean SD, -17 16 milliseconds vs before receiving GAHT; P < .001). No participant had a QTc greater than 480 milliseconds or QTc change greater than 60 milliseconds after the start of GAHT in this study. Nonlinear mixed models (eg, integrating age, calcemia, relevant circulating hormones levels, and torsadogenic drug intake) showed that QTc was associated with total testosterone in transgender men (mean SD estimate, -1.6 0.6 ms/ng/mL; P = .007) and prolactin (mean SD, 0.4 0.1 ms/ng/mL; P < .001). In transgender women, QTc was associated with total testosterone (mean SD estimate, -3.5 0.8 ms/ng/mL; P < .001). Variation of QTp and TAmp observed after the start of GAHT and associated hormonal alteration were globally associated with those observed with QTc, although in opposite directions for transgender women and transgender men. Conclusions and Relevance: In this cohort study, testosterone use in transgender men was associated with QTc and QTp shortening and increased TAmp. Androgen deprivation in transgender women was associated with opposite observations. The magnitude of QTc sexual dimorphism seen in cisgender adults was also observed in the transgender population. This work highlights that potential GAHT effects on cardiac repolarization warrant attention in the exponentially increasing transgender population, which is often exposed to coprescribed drugs prolonging QTc and at risk of TdP.