Association of Current and Long-Term Estradiol Use with Carotid Intima Media Thickness Among Transgender Women: A Cross-Sectional Study

Abstract

Purpose:

Epidemiologic studies suggest that the transgender population has a higher burden of cardiovascular (CV) disease. We aimed to assess CV risk and investigate the relationship between estradiol (E2) or ethinylestradiol (EE) use and carotid intima media thickness (cIMT) in transgender women.

Methods:

This is a cross-sectional analysis nested into a transgender-specific cohort in Rio de Janeiro, Brazil, from August 2015 to February 2018. Increased cIMT was defined as a measurement above the 75th percentile. We tested the association of E2, EE, or both with cIMT. We calculated odds ratios (ORs) using adjusted logistic regression models to assess the association of current use (use in the last 30 days) and long-term use (using for at least 365 consecutive days) of the hormone categories with cIMT.

Results:

We included 298 transgender women with a median age of 31 years (interquartile range [IQR]=25–38), 54.2% had human immunodeficiency virus (HIV) infection. Among transgender women currently on hormone therapy (44.9%), most were on estradiol (27.2%), a combination of E2/EE (12.7%), or EE alone (5.1%). Median cIMT was 0.57 mm (IQR=0.52–0.64). In the final adjusted models, current (OR=0.37; 95% confidence interval [95% CI]=0.14 to 0.93) and long-term (OR=0.20; 95% CI=0.04 to 0.7) E2 use was negatively associated with increased cIMT.

Conclusions:

Both current- and long-term E2 use had a negative association with increased cIMT in a young population of transgender women. Follow-up studies are needed to confirm its safety and support hormone recommendations for transgender women.

Introduction

Caring for transgender women, women who have a male sex assigned at birth, requires consideration of specific health issues related to corporeal changes that occur with feminizing hormones and procedures, as well as the higher incidence of human immunodeficiency virus (HIV) in this population.¹ In addition, they often face transphobia, discrimination, and social and economic exclusion, which significantly impact their overall health.^1,2 Transgender women have higher death rates than the overall population, and a major concern is the lack of appropriate evaluation of cardiovascular (CV) risk among transgender people.³

Based on the Gender Minority Stress and Resilience model, transgender people experience stressors such as stigmatization in multilevel social determinants (structural, interpersonal, and internalized stigma), discrimination, family stress, rejection, victimization, violence, transphobia, negative expectations, hypervigilance, and concealment of gender identity, among others. This results in higher stress levels, leading to depression, anxiety, and stigmatization burden, which may reshape CV health-related behaviors, such as diet, physical activity, alcohol use, smoking, substance abuse, and sleep and contribute to clinical CV risk factors (hypertension, obesity, dyslipidemia, hyperglycemia, HIV, and vascular dysfunction).^4,5

In addition, hormone use is common among transgender women and could impact their risk of cardiovascular disease (CVD) rates.⁶ Moreover, the use of injectable liquid industrial silicone, a common practice in low- and middle-income countries (LMICs), is associated with chronic inflammation and may enhance atherosclerosis.⁷ Furthermore, gender-affirming surgeries and procedures have been associated with higher carotid intima media thickness (cIMT) and increased mortality.^6,8

The use of feminizing hormone therapy is a fundamental component of the transition process for transgender women. Considering the many barriers to obtaining quality health care that transgender women often face, many resort to the use of nonprescribed hormones without medical supervision, exposing them to high levels of exogenous estrogens such as estradiol (E2) and ethinylestradiol (EE). Data on CVD in transgender women receiving cross-sex hormone therapy, mainly based on retrospective cohorts, have shown an increased thromboembolic risk, leading to a contraindication of EE use and recommending lower-dose transdermal estrogen formulations over high dose estrogen oral formulations.^3,9–12

CV risk may be assessed with cIMT, an important biomarker of subclinical arteriosclerosis, which correlates with early CV events.¹³ The association between cIMT and CV risk holds true for individuals with HIV.¹⁴

Data about transgender populations in LMICs, particularly clinical data, are scarce as they remain underserved, with poor health care access, despite standard of care recommendations being available.^1,15 Assessing early, subclinical CV subclinical disease in transgender women may further support adequate and transgender-specific health care recommendations. We aimed to evaluate the potential association between CV risk and the use of estrogens by assessing cIMT in transgender women.

Methods

This study was reviewed and approved by the Ethics Review Board of the Evandro Chagas National Institute of Infectious Diseases at Oswaldo Cruz Foundation (INI/Fiocruz), in Rio de Janeiro. All participants signed an informed consent form before study procedures.

The data supporting this study's findings are available from the corresponding author upon reasonable request.

Setting and study population

This is a cross-sectional study on baseline data from Transcendendo, an open prospective, clinic-based cohort of HIV-positive and -negative transgender women aged ≥18 years, living in Rio de Janeiro, Brazil. Cohort establishment and procedures have been described elsewhere.¹⁶ The current analysis enrolled any cohort participant with valid cIMT results and information on hormone use at study entry (from August 2015 to February 2018).

Study procedures and data collection

Participants engaged in face-to-face interviews using structured questionnaires conducted by trained professionals. We collected data on sociodemographic characteristics, tobacco use, illicit drug use, medical history (including hypertension, dyslipidemia, diabetes mellitus, CVD, acute myocardial infarction [AMI], and stroke), and family history of CVD. We retrieved medication history using direct questions regarding use of antihypertensive, antihyperglycemic, and lipid lowering medicine. We also assessed anthropometric measurements and resting blood pressure, which was measured thrice in the nondominant arm in the sitting position.

Laboratory testing was performed on the Transcendendo cohort at baseline using a nonfasting blood sample, including glucose, hemoglobin A1C (HbA1C), and lipid profile. Transgender women underwent HIV rapid testing, and for those with a negative result who reported condomless anal intercourse in the past 30 days, we offered a pooled HIV ribonucleic acid (RNA) testing to identify acute HIV infection. For transgender women living with HIV we collected clinical information, including date of first HIV-positive test, antiretroviral (ART) use, history of opportunistic infections, and laboratory test results (CD4⁺ T lymphocyte cells (CD4⁺) and CD8⁺ T lymphocyte cell (CD8⁺) count, plasma viral load values), as well as additional information regarding hormone use (age of initiation, report of medical guided hormonal therapy, and types of hormones used).

All participants were offered a carotid ultrasonography to evaluate cIMT. All examinations were performed by the same sonographer.

Main outcome

Our main outcome was the cIMT, measured by carotid ultrasonography of 1-cm portions of the distal left and right common carotid artery far walls with the use of a linear transducer (10-megahertz linear transducer [VIVID 7, GE]), with axial resolution of ∼0.10 mm and calculated automatically by Medical Imaging Applications software (MIA, COralville, IOWA) over three electrocardiographically gated cardiac cycles. The near and far walls of the common carotid artery, carotid bulb, and origin of the internal carotid artery bilaterally were evaluated longitudinally and transversally for plaque detection, following the Mannheim consensus and the Brazilian Longitudinal Study of Adult Health (ELSA-Brasil) procedures.¹⁷ The decision about the presence or absence of carotid plaque was made only by qualitative (visual) judgment, without measuring the thickness of the lesions.¹⁸

Increased cIMT was defined as a measurement above the 75th percentile. Current Brazilian recommendations for ultrasound evaluation on carotid atherosclerotic disease use the ELSA-Brasil cIMT values as the standard for the Brazilian adult population.¹⁸ Those values indicate a cIMT value greater than the 75th percentile for an individual's age, sex, and race/ethnicity as an indicative of higher subclinical atherosclerosis burden. Although binary-sex differences in cIMT have been reported, with higher baseline cIMT and higher rates of increased cIMT in cisgender men, those differences were not considered in this study.

Covariables definition

Explanatory variables were divided into demographic, traditional CV risk factors, HIV status, feminization, and history of gender related procedures. Self-report race/skin color was categorized as “White,” “Black,” “Mixed,” and “Other.” We dichotomized the number of years of formal education.

Family history of premature CVD was defined as history of sudden death or any episode of AMI or stroke among father or brothers younger than 55 years of age or mother or sisters younger than 65 years of age. The presence of hypertension, diabetes, and dyslipidemia was based on self-reporting or on current results/measurements, as follows. Hypertension: if the mean of the last two of three measurements showed systolic blood pressure of 140 mmHg or higher or diastolic blood pressure of 90 mmHg or higher.¹⁹ Diabetes: plasma nonfasting glucose of 200 mg/dL or higher or HbA1C of 6.5% or higher.²⁰ Dyslipidemia: Non-high-density lipoprotein (HDL) cholesterol higher than 160 calculated by the difference between total cholesterol and HDL cholesterol.²¹

HIV status was assessed by HIV rapid testing at baseline visit or prior HIV viral load >1000 copies/mL. T lymphocyte CD4⁺ counts were determined by flow cytometry (Becton-Dickinson), and HIV viral load quantification was measured by the NUCLISENS methodology, with a minimum detection limit of 40 copies/mL. Substance abuse was identified as use of cocaine, cannabis, or downers (defined as a depressive or sedative drug) in the last 12 months. Gender related surgeries included mammoplasty, facial surgery, penectomy, orchiectomy, and vaginoplasty.

We obtained data on all current (use in the last 30 days) and long-term (using for at least 365 consecutive days) reported hormonal medication used for feminization. The definition of current use as use in the last 30 days was due to the frequent use of a monthly injectable hormone. Feminizing hormones were later classified as: (1) “ethinylestradiol” (use of EE excluding those who were usually in combination with E2); (2) “E2” (use of E2 plus injectable progestogen or E2 plus oral progestogen, or isolated oral E2 or transdermal E2, excluding those using any form of E2 combined with EE); (3) “both” (use of EE combined with any type of E2). Those who reported conjugated estrogen use were excluded from analyses.

Statistical analysis

General descriptive characteristics of participants, shown as counts, percentages, medians, and interquartile ranges (IQRs), were compared to the presence of increased cIMT (yes/no). We applied nonparametric Mann–Whitney test for continuous variables and chi-squared tests for categorical variables, and the results are shown with the corresponding p-values. We performed logistic regression models with the calculated odds ratios (ORs) and 95% confidence intervals (CIs) to evaluate a potential association between current use of estrogen and cIMT. Estrogen use was categorized as E2, EE, both, or none (model 1); and any estrogen use or no estrogen use (model 2). Finally, we made additional assessments on current hormone use to check the overall robustness of the analysis, considering linear and logistic regression models, with the outcome paired for age in the latter (151 matched observations).

The final selection of covariates in each model was based on the best fit measured by the Akaike information criterion, using an automated model selection based on a genetic algorithm.²² The candidate set included traditional CV risk factors, such as age, race, history of CVD, family history of premature CVD, education, monthly per-person income, hypertension, diabetes, dyslipidemia, smoking status, body mass index (BMI), high-sensitivity C-reactive protein, and HIV. We also included transgender-specific factors that may potentially interfere with CV risk, such as gender related surgery and the use of industrial silicone, given their potential to cause inflammation. A similar approach compared “long-term” estrogen users (estrogen use in the last 365 consecutive days, which can be defined for both exposure categorizations) to not on estrogens in the last year.

A two-tailed level of significance was set at 5%. Analyses were performed using the R environment version 3.2.2.

Results

Among 357 transgender women enrolled in the cohort, 317 underwent carotid ultrasonography, and 308 had a valid cIMT measure. Out of the latter, 298 participants were included in the analysis and 10 were excluded due to the use of conjugated estrogen (n=6) or missing data on estrogen use (n=4). The median cIMT was 0.57 mm (IQR=0.52–0.64). A normal quantile-quantile (Q-Q) plot curve of cIMT (Supplementary Fig. S1) and linear regression diagnostics of residuals (Supplementary Fig. S2) showed a point of visual deflection proximal to the upper quartile, reinforcing the dichotomization into the 75th percentile. Among the 298 participants evaluated, 17 (5.8%) had carotid plaques.

Median age at enrolment was 31 years (IQR=25–38). Most participants were mixed race/skin color (51.9%), and 34% had less than 9 years of education. Overall, 46.1% of participants reported current smoking, 19% had family history of premature CVD, and only 5% had prior CVD history. Hypertension and dyslipidemia were common (16.0% and 19.1%, respectively). HIV infection was present in 54.2% of the participants. The median time since HIV diagnosis was 2.1 (IQR=0.0–7.2), while the median current and nadir CD4⁺ count were 583 cells/mm³ (IQR=386–776) and 329 cells/mm³ (IQR=202–529), respectively. Among the 167 transgender women living with HIV, 59.1% (97) were currently on ART, of whom 46.7% had an undetectable viral load. Prior AIDS defining illness occurred in 24.4% of the participants.

Most participants were not on feminizing hormone therapy (55.0%); 27.2% and 29.2% were currently using oral E2 and intramuscular estrogen-based regimen, respectively. Among participants currently on feminizing hormones, the median starting age was 17 years (IQR=15–20), mostly without medical guidance (75.2%). Prior use of liquid industrial silicone was common (49.8%), and 42.7% had undergone any gender-related surgery. Characteristics of the study population were compared between estrogen use groups (Supplementary Table S1); participants currently using E2 had lower cIMT levels (median 0.55 mm [IQR=0.51–0.62]) and were younger (median age 30 years [IQR=24–37]) than those not on current estrogen use (respectively, median cIMT 0.59 mm [IQR=0.53–0.67] and median age 34 years [IQR=28–41]).

Compared to participants with cIMT percentile below 75th, transgender women with increased cIMT were older (median age 38 years [IQR=31–50.5] vs. 29 years [IQR=24–36]) and presented with higher prevalence of traditional CV risk factors, such as hypertension, diabetes, higher levels of systolic and diastolic blood pressure, waist, and BMI (Table 1). Transgender women with increased cIMT had higher proportion of liquid industrial silicone use compared to those with cIMT below the 75th percentile.

Table 1.

Characteristics of Transgender Women Participants According to Carotid Intima Media Thickness 0.75 Percentile

	Total N=298	cIMT ≤P75 (0.64 mm)^a N=225	cIMT >P75 (0.64 mm)^a N=73	p
Age in years, median (IQR)	31 (25–38)	29 (24–36)	38 (31–51)	<0.001
Self-declared race/color, n (%)				0.8912
White	75 (25.42)	58 (26.01)	17 (23.61)
Black	67 (22.71)	51 (22.87)	16 (22.22)
Mixed/other	153 (51.86)	114 (51.12)	39 (54.17)
Years of education, n (%)^b				0.2723
<9	103 (34.92)	74 (33.18)	29 (40.28)
≥9	192 (65.08)	149 (66.82)	43 (59.72)
Per-person monthly income (US$), median (IQR)^c	181.82 (85,260)	173.2 (78–260)	196.97 (137,260)	0.6838
Smoking status, n (%)				0.2359
Current	136 (46.1)	106 (47.53)	30 (41.67)
Former	40 (13.56)	26 (11.66)	14 (19.44)
Never	119 (40.34)	91 (40.81)	28 (38.89)
Family history of premature CVD, n (%)^d	57 (19.32)	38 (17.04)	19 (26.39)	0.0807
History of CVD, n (%)	15 (5.08)	8 (3.59)	7 (9.72)	0.0592
Hypertension, n (%)	53 (19.06)	29 (13.81)	24 (35.29)	<0.001
Dyslipidemia, n (%)	47 (16.04)	31 (14.03)	16 (22.22)	0.0998
Diabetes mellitus, n (%)	5 (2.06)	1 (0.56)	4 (6.25)	0.0179
BMI, median (IQR)	24.6 (21.6–27.7)	24.0 (21.5–26.7)	26.7 (23.6–28.9)	<0.001
Waist, median (IQR)	83 (77–91)	81 (76–89)	89 (83–96)	<0.001
Systolic pressure, median (IQR)	110 (100–120)	110 (100–120)	120 (110–125)	<0.001
Diastolic pressure, median (IQR)	75 (65–80)	70 (65–80)	80 (70–85)	<0.001
Creatinine (mg/dL), median (IQR)	0.87 (0.76–0.97)	0.86 (0.76–0.96)	0.91 (0.80–1.02)	0.0339
Glucose (mg/dL), median (IQR)	89 (79–98)	88 (79–96)	95 (85–102)	0.0105
hs-CRP (mg/L), median (IQR)	0.2 (0.1–0.6)	0.2 (0.1–0.6)	0.3 (0.2–0.9)	0.3994
Total cholesterol, median (IQR)	162 (141–182)	159 (137–180)	174 (150–193)	0.0037
HDL (mg/dL), median (IQR)	41 (34–50)	41 (34–50)	41 (32–48)	0.3691
HIV infection, n (%)	44 (59.46)	164 (53.77)	120 (51.95)	0.2594
Current use of estrogen, n (%)				0.0116
Estradiol	81 (27.18)	67 (29.78)	14 (19.18)
EE	15 (5.03)	14 (6.22)	1 (1.37)
Mixed	38 (12.75)	32 (14.22)	6 (8.22)
No	164 (55.03)	112 (49.78)	52 (71.23)
Via/route of estrogen, n (%)
Intramuscular	87 (29.19)	74 (32.89)	13 (17.81)	0.0138
Oral	83 (27.85)	71 (31.56)	12 (16.44)	0.0123
Transdermal	33 (11.07)	24 (10.67)	9 (12.33)	0.6942
Substance abuse in the last 12 months, n (%)^e	170 (57.63)	140 (62.78)	30 (41.67)	0.0016
Liquid industrial silicon injection, n (%)	147 (49.83)	101 (45.29)	46 (63.89)	0.0061
Gender related surgery, n (%)	126 (42.71)	97 (43.50)	29 (40.28)	0.7773
Age at hormone initiation, median (IQR)	17 (15–20)	17 (15–20)	17 (15–21)	0.8675
Medically supervised hormone use, n (%)				0.4338
Never	222 (75.25)	169 (75.78)	53 (73.61)
Current	26 (8.81)	22 (9.87)	4 (5.56)
Former	14 (4.75)	10 (4.48)	4 (5.56)
Missing	33 (11.19)	22 (9.87)	11 (15.28)

Significant p-values in bold.

cIMT P75 (0.75 percentile)=0.64 mm.

Continuous variables were reclassified as categorical.

US$1.00=R$3.85.

AMI, stroke, or sudden death.

Cannabis, cocaine, or downers.

AMI, acute myocardial infarction; BMI, body mass index; cIMT, carotid intima media thickness; CVD, cardiovascular disease; EE, ethinylestradiol; HDL, high-density lipoprotein; HIV, human immunodeficiency virus; hs-CRP, high-sensitivity C reactive protein; IQR, interquartile range; P75, 0.75 percentile.

Risk factors for cIMT in transgender women

Overall, transgender women currently using E2 and estrogens had, respectively, 63% (OR=0.37; 95% CI=0.14 to 0.93, p=0.04) and 55% (OR=0.45; 95% CI=0.20 to 0.97, p=0.046) reduced odds of having a cIMT above the 75th percentile (Table 2). For each year of age, the odds of having a cIMT above the 75th percentile increased by 11% (OR=1.11; 95% CI=1.07 to 1.16, p≤0.001). Point estimates for E2 (β=−0.01; 95% CI=−0.04 to 0.01; p=0.314) and estrogen preparations (β=−0.01; 95% CI=−0.03 to 0.02; p=0.622) from the linear regression model were not statistically significant, although they indicated the same direction of the logistic regression model (Supplementary Table S2). Similar findings were obtained when estrogen groups were matched for age (E2: OR=0.49; 95% CI=0.19 to 1.16; p=0.116, and estrogen preparations: OR=0.49; 95% CI=0.23 to 1.03; p=0.064) (Supplementary Table S3).

Table 2.

Final Logistic Regression Model with Demographic and Clinically Relevant Predictors Associated with Carotid Intima Media Thickness in Transgender Women

Predictors	Univariable OR (95% CI)	Model 1-Sex hormone categories^a		Model 2-Current use of estrogen^a
Predictors	Univariable OR (95% CI)	Multivariable OR (95% CI)	p	Multivariable OR (95% CI)	p
Estrogen preparation
Estradiol	0.45 (0.22–0.85)	0.37 (0.14–0.93)	0.042
EE	0.15 (0.01–0.80)	0.27 (0.01–1.88)	0.257
Both	0.40 (0.15–0.96)	0.77 (0.22–2.42)	0.671
Current use of estrogen	0.40 (0.22–0.70)			0.45 (0.20–0.97)	0.046
Age in years	1.10 (1.07–1.14)	1.11 (1.07–1.16)	<0.001	1.11 (1.07–1.15)	<0.001
Gender related surgery	0.88 (0.51–1.50)	0.58 (0.25–1.29)	0.194	0.59 (0.26–1.30)	0.203
Hypertension	3.40 (1.80–6.43)	1.50 (0.57–3.87)	0.403	1.62 (0.62–4.12)	0.317
Monthly per-person income	1.00 (1.00–1.00)	1.00 (1.00–1.00)	0.597	1.00 (1.00–1.00)	0.690
hs-CRP (mg/L)	1.00 (0.85–1.13)	1.09 (0.91–1.28)	0.312	1.09 (0.91–1.28)	0.300
BMI (kg/m²)	1.08 (1.03–1.14)	1.06 (0.98–1.15)	0.141	1.06 (0.98–1.15)	0.156

Significant p-values in bold.

Hosmer and Lemeshow GOF test was 0.98 and 0.29 for Model 1 and Model 2, respectively.

CI, confidence interval; GOF, goodness of fit; OR, odds ratio.

Participants on long-term E2 use had reduced odds of having a cIMT above the 75th percentile compared to those not in hormone use in the last year (OR=0.20; 95% CI=0.04 to 0.70; p=0.02) (Table 3).

Table 3.

Final Logistic Regression Model for the Association Between Long Term Use of Estrogen and Carotid Intima Media Thickness in Transgender Women

Predictors	Univariable OR (95% CI)	Model 1-Sex hormone categories^a		Model 2-Long term use of estrogens^a
Predictors	Univariable OR (95% CI)	Multivariable OR (95% CI)	p	Multivariable OR (95% CI)	p
Estrogen preparation
Estradiol	0.27 (0.11–0.60)	0.20 (0.04–0.70)	0.020
Both	0.17 (0.05–0.46)	0.44 (0.09–1.73)	0.265
Gender related surgery	0.70 (0.36–1.35)	0.37 (0.11–1.07)	0.078	0.36 (0.11–1.04)	0.070
Dyslipidemia	1.39 (0.60–3.04)	0.30 (0.07–1.15)	0.098	0.31 (0.07–1.17)	0.102
Hypertension	2.41 (1.13–5.08)	1.18 (0.35–3.83)	0.781	1.18 (0.35–3.79)	0.785
Age in years	1.09 (1.06–1.13)	1.10 (1.05–1.16)	<0.001	1.09 (1.04–1.16)	<0.001
Monthly per-person income	1.00 (1.00–1.00)	1.00 (1.00–1.00)	0.332	1.00 (1.00–1.00)	0.329
Hs-CRP (mg/L)	0.86 (0.60–1.05)	1.09 (0.76–1.36)	0.484	1.08 (0.76–1.33)	0.544
BMI (kg/m²)	1.06 (1.00–1.13)	1.11 (0.99–1.25)	0.092	1.10 (0.98–1.24)	0.110
Current use of estrogen	0.23 (0.11–0.45)			0.27 (0.09–0.77)	0.018

Significant p-values in bold.

Hosmer and Lemeshow GOF test was 0.16 and 0.13 for model 1 and model 2, respectively.

Discussion

In this cross-sectional analysis we observed an association between current and long-term E2 use and decreased cIMT among transgender women enrolled in the Transcendendo cohort in Rio de Janeiro, Brazil. Our results are in agreement with prior reports from high-income countries.^23,24 To our knowledge, this is the first study in a LMIC and one of the largest to evaluate cIMT among transgender women.

Our results also corroborate current epidemiological data suggesting that estrogen use has a positive impact in CV health, particularly in younger cohorts. Cumulative data from observational and randomized trials indicate that the initiation of hormone replacement therapy in postmenopausal ciswomen aged less than 60 years or not long after menopause actually reduces the incidence of CVD.²⁵ Moreover, in a randomized-clinical trial, progression of atherosclerosis, measured by cIMT, was slower in postmenopausal ciswomen receiving oral E2 early after menopause compared to those in the placebo arm.²⁶

In parallel, prospective study indicates higher reported AMI history in transgender women compared to cisgender women (i.e., individuals who were assigned female at birth and who identify with the female gender), but did not have significant increase when comparing transgender women to cismen (i.e., individuals who were assigned male at birth and who identify with the male gender).²³

Age is a nonmodifiable CV risk factor, and, in our study, it was a strong predictor of increased CV burden. However, HIV infection was not associated with increased cIMT. Other studies identified that traditional CVD determinants had a stronger association with cIMT than HIV infection.^27,28

In addition, both current and long-term E2 use were associated with lower levels of cIMT in our population. The role of hormonal therapy on CV health in transgender women is still uncertain and controversial, with conflicting data supporting either an increased risk or protection.^3,24,29,30 This may be due to the complex interactions related to CVD, and as such, there is a need to carefully define the population assessed, as well as the hormone compound, dose, route, combination, and time of initiation.³¹ Moreover, Transgender women usually initiate hormone use early in life and at high doses, frequently without medical supervision, and continue their use for a long period or even lifelong.^3,29 Longitudinal studies and controlled trials could provide additional robust data on estrogen use and CV risk in this population.

Despite the association between exogenous estrogens and CV outcomes, the hormones are insufficient to explain the increased occurrence of CVD. Social determinants of health, which encompass many factors disproportionally affecting trans populations, play a significant role in its development, as well as in CVD morbidity and mortality.³²

cIMT measurements are less costly and less invasive to assess CVD risk than coronary artery calcium. Furthermore, cIMT has the potential to be more precise and accurate than risk prediction models (e.g., Framingham and atherosclerotic cardiovascular disease risk scores), which use binary sex categories and do not take hormone therapy into consideration.⁵

Limitations

Our study has several limitations. First, the cross-sectional design hinders our ability to determine causal association between hormones and cIMT. In addition, there are no standardized cutoffs of cIMT measures that could distinguish transgender individuals with and without CVD. However, the 75th percentile is adequate in providing some inference on populations with greater burden of subclinical atherosclerosis. The volume and weight of liquid industrial silicone injected were not considered to calculate BMI; however, it usually has a minimal impact on the total weight.

Since we enrolled individuals who were seeking care at an HIV/sexually transmitted infection clinic, there could be a selection bias toward transgender women engaged in care. In an attempt to minimize this bias, we enrolled all individuals presenting to our clinic into the study, regardless of previous care at our center. Our cohort was young and had low rates of hypertension, dyslipidemia, diabetes, and CVD, which may have impacted the strength of association of these variables, yet our results are similar to those in younger cohorts showing that hormone treatment reduces the risk of atherosclerosis. Finally, information on hormone use was self-reported, and detailed information on prior use can be affected by memory, limiting generalizability.

Conclusion

Transgender women have specific health considerations that emphasize the need for an adequate CVD evaluation. The inverse association between estradiol use without medical supervision and cIMT in a cohort of young transgender women shown here suggests that estradiol may have a cardioprotective effect, underscoring the importance of routine monitoring of CV risk factors in transgender women and of advising lifestyle changes as appropriate. Our results also suggest a more intensive CVD screening for those not undergoing estrogen or for older transgender women. Prospective studies are needed to better evaluate the long-term impact of estrogen use on the CV health of transgender women.

Footnotes

Acknowledgment

The authors thank the Transcendendo team and all study participants.

Authors' Contributions

A.C.G.F.: Conceptualization, Methodology, Formal analysis, Investigation, Writing—Original draft preparation, Project administration, R.C.M.: Conceptualization, Investigation, Methodology, Formal analysis, Writing—Review and Editing, L.E.C.: Conceptualization, Formal analysis, Writing—Review and Editing, E.M.J.: Conceptualization, Investigation, Writing—Review and Editing, Project administration, I.B.: Investigation, Writing—Review and Editing, R.K.F.: Investigation, Writing—Review and Editing, L.M.: Investigation, Writing—Review and Editing, L.F.E.: Investigation, Writing—Review and Editing, R.I.M.: Methodology, Formal analysis, Data Curation, Writing—Review and Editing, S.W.C.: Conceptualization, Investigation, Writing—Review and Editing, V.G.V.: Conceptualization, Supervision, Writing—Review and Editing, D.C.: Conceptualization, Supervision, Writing—Review and Editing, B.G.: Conceptualization, Supervision, Writing—Review and Editing, A.G.P.: Conceptualization, Supervision, Writing—Review and Editing. All authors read and approved the final article.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

This study was supported by Fiocruz PMA (Programa de Polìticas Públicas e Modelos de Atenção à Saúde) and the Brazilian Ministry of Health.

Supplementary Material

Abbreviations Used

References

Reisner

, Poteat

, Keatley

, et al. Global health burden and needs of transgender populations: A review. Lancet, 2016; 388(10042):412–436; doi: 10.1016/S0140-6736(16)00684-X

Baral

, Poteat

, Strömdahl

, et al. Worldwide burden of HIV in transgender women: A systematic review and meta-analysis. Lancet Infect Dis, 2013; 13(3):214–222; doi: 10.1016/S1473-3099(12)70315-8

Asscheman

, Giltay

, Megens

JAJ

, et al. A long-term follow-up study of mortality in transsexuals receiving treatment with cross-sex hormones. Eur J Endocrinol, 2011; 164(4):635–642; doi: 10.1530/EJE-10-1038

Caceres

, Streed

, Corliss

, et al. Assessing and addressing cardiovascular health in LGBTQ adults: A scientific statement from the American Heart Association. Circulation, 2020; 142(19):e321–e332; doi: 10.1161/CIR.0000000000000914

Streed

, Beach

, Caceres

, et al. Assessing and addressing cardiovascular health in people who are transgender and gender diverse: A scientific statement from the American Heart Association. Circulation, 2021; 144(6):e136–e148; doi: 10.1161/CIR.0000000000001003

Ciccone

, Loverro

, Scicchitano

, et al. Surgical and pharmacological reassignment: Influence on transsexual cardiovascular risk profile. Intern Med J, 2017; 47(11):1255–1262; doi: 10.1111/imj.13597

Ruyer-Thompson

, Guérin

, Borba

, et al. The downside of beauty: ASIA syndrome associated with local silicone injections: A literature review. Immunome Res, 2021; 17:7391; doi: 10.35248/1745-7580.21.s4.7391

Dhejne

, Lichtenstein

, Boman

, et al. Long-term follow-up of transsexual persons undergoing sex reassignment surgery: Cohort study in Sweden. PLoS One, 2011; 6(2):e16885; doi: 10.1371/journal.pone.0016885

Asscheman

, Gooren

, Eklund

. Mortality and morbidity in transsexual patients with cross-gender hormone treatment. Metab Clin Exp, 1989; 38(9):869–873; doi: 101016/0026-0495(89)90233-3

10.

Wierckx

, Mueller

, Weyers

, et al. Long-term evaluation of cross-sex hormone treatment in transsexual persons. J Sex Med, 2012; 9(10):2641–2651; doi: 10.1111/j.1743-6109.2012.02876.x

11.

Nota

, Wiepjes

, de Blok

CJM

, et al. Occurrence of acute cardiovascular events in transgender individuals receiving hormone therapy: Results from a large cohort study. Circulation, 2019; 139(11):1461–1462; doi: 10.1161/CIRCULATIONAHA.118.038584

12.

Streed

, Harfouch

, Marvel

, et al. Cardiovascular disease among transgender adults receiving hormone therapy: A narrative review. Ann Intern Med, 2017; 167(4):256–267; doi: 10.7326/M17-0577

13.

Carpenter

, Sinclair

, Kunadian

. Carotid intima media thickness and its utility as a predictor of cardiovascular disease: A review of evidence. Cardiol Rev, 2016; 24(2):70–75; doi: 10.1097/CRD.0000000000000077

14.

Grunfeld

, Delaney

JAC

, Wanke

, et al. Preclinical atherosclerosis due to HIV infection: Carotid intima-medial thickness measurements from the FRAM study. AIDS, 2009; 23(14):1841–1849; doi: 10.1097/QAD.0b013e32832d3b85

15.

Coleman

, Radix

, Bouman

, et al. Standards of care for the health of transgender and gender diverse people, version 8. Int J Transgend Health, 2022; 23(sup1):S1–S259; doi: 10.1080/26895269.2022.2100644

16.

Ferreira

ACG

, Coelho

, Jalil

, et al. The Transcendendo cohort study—Unveiling the social vulnerabilities of transgender women and travestis in Rio de Janeiro, Brazil. Transgend Health, 2019; 4(1):107–117; doi: 10.1089/trgh.20180063

17.

Touboul

P-J

, Hennerici

, Meairs

, et al. Mannheim carotid intima-media thickness and plaque consensus (2004-2006-2011). An update on behalf of the advisory board of the 3rd, 4th and 5th watching the risk symposia, at the 13th, 15th and 20th European Stroke Conferences, Mannheim, Germany, 2004, Brussels, Belgium, 2006, and Hamburg, Germany, 2011. Cerebrovasc Dis, 2012; 34(4):290–296; doi: 10.1159/000343145

18.

Santos

EHS

, Santos Neto

PJ dos

, Santos I de

. Carotid intima-media thickness in the Brazilian Longitudinal Study of Adult Health (ELSA-Brasil): A narrative review. Sao Paulo Med J, 2018; 136(1):64–72; doi: 10.1590/1516-3180.2017.0272141017

19.

Williams

, Mancia

, Spiering

, et al. 2018 ESC/ESH Guidelines for the management of arterial hypertension. Eur Heart J, 2018; 39(33):3021–3104; doi: 10.1093/eurheartj/ehy339

20.

American Diabetes Association (ADA). Introduction: Standards of medical care in diabetes—2018. Diabetes Care, 2018; 41(Supplement 1):S1–S2; doi: 10.2337/dc18-Sint01

21.

Garber

, Handelsman

, Grunberger

, et al. Consensus statement by the American Association of Clinical Endocrinologists and American College of Endocrinology on the comprehensive type 2 diabetes management algorithm—2020 executive summary. Endocr Pract, 2020; 26(1):107–139; doi: 10.4158/CS-2019-0472

22.

Calcagno

, Mazancourt

. glmulti: An R package for easy automated model selection with (generalized) linear models. J Stat Soft, 2010; 34(12):1–29; doi: 10.18637/jss.v034.i12

23.

Alzahrani

, Nguyen

, Ryan

, et al. Cardiovascular disease risk factors and myocardial infarction in the transgender population. Circ Cardiovasc Qual Outcomes, 2019; 12(4):e005597; doi: 10.1161/CIRCOUTCOMES.119.005597

24.

Maraka

, Singh Ospina

, Rodriguez-Gutierrez

, et al. Sex steroids and cardiovascular outcomes in transgender individuals: A systematic review and meta-analysis. J Clin Endocrinol Metab, 2017; 102(11):3914–3923; doi: 10.1210/jc.2017-01643

25.

Hodis

, Mack

. Postmenopausal hormone therapy and cardiovascular disease in perspective. Clin Obstet Gynecol, 2008; 51(3):564–580; doi: 10.1097/GRF.0b013e318181de86

26.

Harman

, Black

, Naftolin

, et al. Arterial imaging outcomes and cardiovascular risk factors in recently menopausal women: A randomized trial. Ann Intern Med, 2014; 161(4):249–260; doi: 10.7326/M14-0353

27.

Pacheco

, Grinsztejn

, da Fonseca M de

, et al. Traditional risk factors are more relevant than HIV-specific ones for carotid intima-media thickness (cIMT) in a Brazilian Cohort of HIV-infected patients. PLoS One, 2015; 10(2):e0117461; doi: 10.1371/journal.pone.0117461

28.

Hanna

, Post

, Deal

, et al. HIV infection is associated with progression of subclinical carotid atherosclerosis. Clin Infect Dis, 2015; 61(4):640–650; doi: 10.1093/cid/civ325

29.

Wierckx

, Elaut

, Declercq

, et al. Prevalence of cardiovascular disease and cancer during cross-sex hormone therapy in a large cohort of trans persons: A case-control study. Eur J Endocrinol, 2013; 169(4):471–478; doi: 10.1530/EJE-13-0493

30.

Getahun

, Nash

, Flanders

, et al. Cross-sex hormones and acute cardiovascular events in transgender persons: A cohort study. Ann Intern Med, 2018; 169(4):205–213; doi: 10.7326/M17-2785

31.

Morselli

, Santos

, Criollo

, et al. The effects of oestrogens and their receptors on cardiometabolic health. Nat Rev Endocrinol, 2017; 13(6):352–364; doi: 10.1038/nrendo.2017.12

32.

Powell-Wiley

, Baumer

, Baah

, et al. Social determinants of cardiovascular disease. Circ Res, 2022; 130(5):782–799; doi: 10.1161/CIRCRESAHA.121.319811

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.25 MB

0.37 MB

0.02 MB

0.01 MB

0.00 MB