Gender‑Affirming Hormone Therapy and Acute Cardiovascular Events: Dutch Cohort Finds Lower MI but Higher VTE in Transgender Women, Elevated MI and Stroke Risk in Transgender Men

Highlight

– In a Dutch retrospective cohort (1972–2018 exposure window; registry outcomes 2012–2022), transgender women on gender‑affirming oestradiol had a lower standardized incidence ratio (SIR) for myocardial infarction (SIR 0.50; 95% CI 0.32–0.71) but an increased SIR for venous thromboembolism (SIR 1.81; 95% CI 1.33–2.35) compared with general population men.

– Transgender men on testosterone had markedly increased SIRs for myocardial infarction (SIR 4.20; 95% CI 2.72–6.01) and increased ischaemic cerebrovascular accident (SIR 1.55; 95% CI 1.01–2.20) versus general population women; VTE risk was not elevated.

– Adjustment for socioeconomic status made minimal differences; lifestyle factors generally mirrored the general population.

Background

Gender‑affirming hormone therapy (GAHT) is a cornerstone of medical care for many transgender people. Oestradiol-based regimens are commonly used to feminize transgender women, whereas testosterone is used to masculinize transgender men. Clinicians and patients rightly focus on both the mental health benefits and potential somatic risks, including cardiovascular disease (CVD).

Historical and mechanistic data have produced a nuanced, sometimes conflicting, picture. Exogenous oestrogens can have favourable effects on some cardiometabolic risk markers (e.g., HDL), but they also increase thrombogenicity through effects on coagulation. Testosterone may adversely affect lipids, haematocrit, and blood pressure in some patients, potentially increasing arterial event risk. Prior observational studies have suggested elevated risks of venous thromboembolism (VTE) and some arterial events in transgender people, but heterogeneous exposure definitions, small sample sizes, and limited adjustment for socioeconomic and lifestyle confounders complicate interpretation.

Study design

van Zijverden and colleagues conducted a retrospective cohort study of transgender women (n = 2714) and transgender men (n = 1617) who received GAHT at the Amsterdam gender clinic between 1972 and 2018. Person‑years of follow‑up totaled 23,907 for transgender women and 13,457 for transgender men. Medical diagnoses were ascertained from a national registry for the period 2012–2022.

The authors calculated standardized incidence ratios (SIRs) for three acute outcomes: myocardial infarction (MI), ischaemic cerebrovascular accident (stroke), and venous thromboembolism (VTE). Comparators were general population incidence rates of opposite‑sex reference groups, adjusted for socioeconomic status (SES) assessed via education, employment, and income. Lifestyle factors (body mass index, smoking, alcohol) were evaluated across age strata to assess whether lifestyle differences could explain observed outcome differences.

Key findings

The principal results are summarized below (SIR [95% CI]). All comparisons are versus general population reference groups after SES adjustment.

Transgender women (on oestradiol)

• Myocardial infarction: SIR 0.50 (0.32–0.71). This indicates approximately half the observed incidence of MI compared with general population men.
• Ischaemic cerebrovascular accident (stroke): SIR 0.94 (0.72–1.19). No statistically significant difference.
• Venous thromboembolism (VTE): SIR 1.81 (1.33–2.35). Nearly twofold higher VTE incidence.

Transgender men (on testosterone)

• Myocardial infarction: SIR 4.20 (2.72–6.01). A substantially elevated relative incidence versus general population women.
• Ischaemic cerebrovascular accident (stroke): SIR 1.55 (1.01–2.20). Modestly elevated risk.
• Venous thromboembolism: SIR 1.00 (0.53–1.61). No clear difference.

Adjustment for SES made minimal impact on these associations. Lifestyle measures (BMI, smoking, alcohol) largely resembled the general population, reducing the likelihood that such factors explained the observed differences.

Interpretation and biological plausibility

These results align with known pharmacologic effects of sex steroid therapies.

For transgender women, many forms of oestradiol reduce LDL cholesterol and may improve some metabolic markers; these effects could underlie the lower MI incidence observed. However, oestrogens increase coagulation factor changes that raise VTE risk. The observed dissociation—lower MI but higher VTE—fits this pharmacologic pattern. Importantly, type, dose, and route of oestrogen administration influence thrombosis risk; oral preparations and ethinyl oestradiol (now discouraged) carry higher VTE risk than transdermal formulations in postmenopausal women and are plausible modifiers here.

For transgender men, testosterone therapy can elevate haematocrit, increase LDL cholesterol, and raise blood pressure in susceptible individuals—changes that plausibly increase arterial thrombosis risk and may explain the observed elevated MI and stroke incidence. The absence of increased VTE risk is biologically plausible because androgenic therapies have not been consistently linked to prothrombotic changes of the same magnitude as oestrogens.

Strengths

• Large cohort size with many person‑years and use of national registry data for outcome ascertainment.
• Explicit adjustment for socioeconomic status (education, employment, income), addressing an important potential confounder that prior studies sometimes lacked.
• Consideration of lifestyle factors and stratified analyses by age group.

Limitations and caveats

• Observational design: Residual confounding and unmeasured variables (e.g., detailed smoking pack‑years, illicit drug use, precise physical activity) may remain.
• Heterogeneity of exposures: Regimens spanned decades (1972–2018) with evolving formulations, doses, and routes (oral vs transdermal), but registry outcomes were captured from 2012 onward. Detailed, time‑updated exposure data (dose, route, serum levels) were not presented for all participants, limiting causal attribution to specific formulations.
• Outcome capture limited to registry period 2012–2022; events before registry start may be missed, and differential survival bias is possible.
• Comparators used general population reference rates; ideal comparison groups of untreated transgender people are uncommon and difficult to construct, and some residual demographic differences could persist.
• Absolute risks were not emphasized in the summary data. Large relative increases (e.g., SIR 4.2 for MI in trans men) can still reflect small absolute risks if baseline incidence is low—clinically important context for counseling is absolute event rates and number needed to harm, which were not the focus of the report summary provided here.

Clinical implications

These findings have practical implications for clinicians caring for transgender patients:

• Risk assessment: Before initiating GAHT, clinicians should assess baseline cardiovascular risk (history of CVD, smoking status, blood pressure, lipids, diabetes, family history, thrombosis history) and reassess periodically.
• Route and formulation considerations for oestradiol: Given the elevated VTE signal, clinicians should avoid formulations with known high thrombogenicity (e.g., ethinyl oestradiol) and consider transdermal oestradiol in patients with VTE risk factors, extrapolating from postmenopausal HRT literature where transdermal delivery has lower VTE risk. However, high‑quality, transgender‑specific comparative data are still limited.
• Monitoring for testosterone therapy: For transgender men, monitor haematocrit, blood pressure, lipids, and glycaemic markers; address modifiable risk factors aggressively (smoking cessation, weight optimization, blood pressure/lipid control) and consider cardiology referral for high‑risk individuals.
• VTE counseling and thrombosis prevention: For transgender women with strong VTE risk factors (personal or strong family history, active malignancy, prolonged immobilization), engage in shared decision‑making about GAHT risk/benefit and consider thrombosis prevention strategies during high‑risk periods. Routine thrombophilia screening is not generally recommended unless family/personal history suggests.
• Shared decision‑making: Emphasize the mental health and quality‑of‑life benefits of GAHT and balance these against potential somatic risks. Individualized risk reduction and monitoring strategies should be co‑developed with patients.

Research and policy gaps

Key unanswered questions include the effect of specific oestradiol formulations and routes, the dose–response relationship for testosterone and arterial events, and absolute risk quantification across age groups and comorbidity strata. Prospective, multi‑centre registries that record time‑updated hormone type, dose, route, serum levels, surgical history, and granular lifestyle data are needed. Pragmatic trials are unlikely for hard CVD endpoints, but mechanistic studies and comparative safety studies of formulations (transdermal vs oral) would be informative.

Conclusion

The Dutch cohort reported by van Zijverden et al. provides important, contemporary data on acute cardiovascular events in transgender people receiving GAHT. After adjustment for socioeconomic status and with lifestyle broadly similar to the general population, transgender women on oestradiol experienced lower MI rates but higher VTE rates, while transgender men on testosterone experienced substantially higher MI rates and modestly increased stroke rates. These patterns are biologically plausible and broadly consistent with known effects of sex steroids on lipids, haematocrit, blood pressure, and coagulation.

Clinicians should continue to offer GAHT as part of gender‑affirming care while undertaking individualized cardiovascular risk assessment, selecting hormone regimens mindful of thrombosis and arterial risk factors, and monitoring appropriately. Further research should aim to refine absolute risk estimates, identify modifiable mediators, and compare the safety profiles of specific hormone formulations and routes.

Funding and clinicaltrials.gov

The primary article declares authorship and institutional affiliations per journal standards. No clinical trial registration was reported because this is an observational cohort study. Detailed funding declarations should be consulted in the original publication: van Zijverden LM et al., Eur Heart J. 2025 (doi: 10.1093/eurheartj/ehaf837) .

References

1. van Zijverden LM, Thijs A, van Diemen JJK, Wiepjes CM, den Heijer M. Transgender persons receiving gender‑affirming hormone therapy: risk of acute cardiovascular events in a Dutch cohort study. Eur Heart J. 2025 Nov 4:ehaf837. doi: 10.1093/eurheartj/ehaf837 IF: 35.6 Q1 B1. PMID: 41187090 IF: 35.6 Q1 B1.

2. Hembree WC, Cohen‑Kettenis P, Delemarre‑van de Waal HA, et al. Endocrine Treatment of Gender‑Dysphoric/Gender‑Incongruent Persons: An Endocrine Society Clinical Practice Guideline. J Clin Endocrinol Metab. 2017;102(11):3869–3903.

Author note

This article synthesizes the findings and clinical implications of the referenced cohort study for clinicians and policy makers. It aims to provide an evidence‑based, pragmatic interpretation that supports shared decision‑making and targeted monitoring in transgender care.