BACKGROUND AND OBJECTIVES:

Many transgender youth experience gender dysphoria, a risk factor for suicide. Gender-affirming hormone therapy (GAHT) ameliorates this risk but may increase the risk for thrombosis, as seen from studies in adults. The aim with this study was to examine thrombosis and thrombosis risk factors among an exclusively adolescent and young adult transgender population.

METHODS:

This retrospective chart review was conducted at a pediatric hospital-associated transgender health clinic. The primary outcome was incidence of arterial or venous thrombosis during GAHT. Secondary measures included the prevalence of thrombosis risk factors.

RESULTS:

Among 611 participants, 28.8% were transgender women and 68.1% were transgender men. Median age was 17 years at GAHT initiation. Median follow-up time was 554 and 577 days for estrogen and testosterone users, respectively. Individuals starting GAHT had estradiol and testosterone levels titrated to physiologic normal. Multiple thrombotic risk factors were noted among the cohort, including obesity, tobacco use, and personal and family history of thrombosis. Seventeen youth with risk factors for thrombosis were referred for hematologic evaluation. Five individuals were treated with anticoagulation during GAHT: 2 with a previous thrombosis and 3 for thromboprophylaxis. No participant developed thrombosis while on GAHT.

CONCLUSIONS:

In this study, we examined thrombosis and thrombosis risk factors in an exclusively adolescent and young adult population of transgender people receiving GAHT. These data suggest that GAHT in youth, titrated within physiologic range, does not carry a significant risk of thrombosis in the short-term, even with the presence of preexisting thrombosis risk factors.

What’s Known on This Subject:

Gender dysphoria in transgender individuals place them at high risk of suicide. Gender-affirming hormone therapy ameliorates this risk but may increase thrombosis risk. Authors of previous studies in adults have suggested increased risk of thrombosis, especially with estrogen use.

What This Study Adds:

In this study, we examined the risk of thrombosis with gender-affirming hormone therapy in an exclusively adolescent and young adult population. We found that, despite preexisting thrombosis risk factors, there were no thromboses in this cohort.

Transgender or gender expansive youth (hereafter referred to as “transgender”) compose a significant portion of the population, from 1.8% to 2.7% of US high school students.13  Transgender youth and adults frequently experience gender dysphoria, which is associated with negative health outcomes, including suicide.4,5  Transitioning to the affirmed gender, through social and medical interventions, may alleviate gender dysphoria.6  One component of medical transition is the use of gender-affirming hormone therapy (GAHT).7  For transgender people who are assigned female at birth (transgender men [TGM]), such therapy includes the use of testosterone. For people who are assigned male at birth (transgender women [TGW]), bioidentical estrogen is used, often in combination with a testosterone antagonist, such as spironolactone.4,7  Because GAHT is often central to addressing gender dysphoria, it is critical to understand the risks associated with this therapy for health care providers to be able to appropriately counsel patients.

One potential risk of GAHT is development of arterial and venous thrombosis. Use of exogenous estrogen is associated with increased risk of thrombosis among cisgender women using estrogen-containing contraceptives and hormone replacement therapy, with the highest risk occurring in the first 6 to 12 months of therapy.8  Studies of the risk of thrombosis with use of estrogen for GAHT have shown inconsistent results. Multiple studies demonstrated an increased risk of venous thromboembolism (VTE) and stroke among TGW using estrogen GAHT.912  However, other studies reported minimal rates of thrombosis with estrogen GAHT.1317  Exogenous estrogen may produce a prothrombotic state through multiple mechanisms, including increased synthesis of hemostatic factors, suppression of inhibitors of coagulation, and/or inhibition of fibrinolysis.18,19 

Even less is known about the potential risk of thrombosis among transgender people using exogenous testosterone. In retrospective studies of GAHT in adult TGM, testosterone appears to have minimal risk for both VTE and stroke.9,10,15  However, studies reveal conflicting results among cisgender adult men receiving testosterone for treatment of hypogonadism, either demonstrating increased risk or no increased risk of VTE.2022  There are no studies examining the risk of testosterone use in adolescents, either in transgender or cisgender youth.

To date, all studies of risk of thrombosis among transgender people have been conducted among adults. Thus, little is known about the thrombotic risk associated with use of GAHT among youth in the United States, who are starting gender-affirming hormones at a younger age and are likely to be taking them for a longer time than adults. Only one publication, a case report, has addressed thrombosis in transgender youth.23  Thus, additional data are needed regarding risk of thrombosis in this population to guide medical decision-making for patients, their caregivers, and their clinicians. We sought to determine the incidence and risk factors for thrombosis in a population of transgender youth ages 13 to 24 years receiving GAHT as per current guidelines.4,7  Current recommendations for feminizing therapy include use of estrogen as 17β-estradiol (orally or transdermally) or estradiol valerate or cypionate (parenterally), with or without spironolactone and a gonadotropin-releasing hormone agonist. Current recommendations for masculinizing therapy include use of testosterone administered parenterally or transdermally (gel or patch). The aim with this study is to provide information about thrombosis risk factors among those exclusively using currently recommended treatment regimens in the United States. Given the increasing identification and treatment of younger transgender patients, key questions regarding the safety of GAHT in adolescents and young adults need to be addressed. This study included transgender patients solely aged <25 and included youth <18 years, an age group that historically has not been included in studies of thrombosis in the transgender population.

This retrospective cohort study was conducted by using a clinical database that includes all patients seen in the Cincinnati Children’s Hospital Medical Center (CCHMC) Transgender Health Clinic between its inception in July 2013 and the close of the current study (March 24, 2019). The CCHMC Institutional Review Board approved the protocol and waived the requirement for informed consent. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines were used in reporting this study.24 

From among all patients seen at the Transgender Health Clinic, inclusion criteria for this retrospective cohort were (1) initiation of GAHT before March 24, 2019, and (2) age 13 to 24 years at initiation of GAHT. The exclusion criterion was age <13 years at GAHT start.

Data were manually extracted from the charts of eligible subjects by using a standardized protocol and data abstraction tool (Supplemental Fig 1). The charts of all patients seen since the inception of the clinic were reviewed for eligibility for inclusion. To ensure consistency in data abstraction, a random sample of 10 charts was abstracted by each team member; the team then met to review the data and resolve inconsistencies. The manual chart review included review of TG clinic notes, hematology clinic notes, review of past medical history, problem list, social history, family history, laboratory studies, and medications. Demographic data, age and date of initiation of GAHT, and GAHT medication, dosage, and route were obtained. Additional risk factors for thrombosis, including migraine with aura, elevated BMI, tobacco use, medical diagnoses associated with increased risk of thrombosis, and family history of thrombosis (arterial or venous) and laboratory measures of risk factors for thrombosis, were extracted. Laboratory data included the most recent total testosterone and estradiol levels, complete blood counts, coagulation testing results, and results of any thrombophilia evaluation. For subjects who were referred for hematologic evaluation, recommendations for thromboprophylaxis were abstracted. Finally, arterial or venous thrombosis before and during GAHT, any treatment with therapeutic anticoagulation, any treatment with prophylactic anticoagulation concurrent with GAHT, and duration of anticoagulation treatment were collected.

For hormone levels, guidelines have not been established on timing of levels with respect to dosing (ie, random, trough, or peak levels).4,7  The general practice of the CCHMC TG clinic is to obtain levels when the patient was in clinic, especially for those on weekly injections (testosterone or estradiol), daily oral estradiol, and transdermal preparations (testosterone and estradiol). For patients receiving testosterone injections every other week, attempts are made to obtain levels on the “off” week of injection.

Descriptive analyses were used to examine patient demographics (including sex assigned at birth and affirmed gender), GAHT formulation and dose, risk factors for thrombosis, and abnormal laboratory results. Continuous and categorical data were summarized by median (interquartile range [IQR]) and frequency (percent), respectively. Differences in continuous variables between groups were assessed by using the Wilcoxon rank test. The number of youth who developed thrombosis was determined. Univariate and multivariate analyses were planned in the event of having sufficient number of thrombosis events; however, these analyses were not conducted because of the absence of thrombosis events in the cohort.

Among 1406 individual patients seen in the CCHMC Transgender Health Clinic, 611 subjects were eligible for inclusion in the study cohort. Median age was 17.0 years (IQR: 16.0–19.0) at first presentation to the clinic (Table 1). TGM were slightly younger at first presentation compared with TGW (17 [15–19] years vs 18 [15.5–20] years, respectively; P = .0019). Overall, 428 subjects were assigned female at birth, and 183 were assigned male at birth. For the cohort, 176 (28.8%) individuals identified as female, 416 (68.1%) as male, and 19 (3.1%) as nonbinary or gender nonconforming. The cohort was predominantly white (n = 544; 89%) and African American (n = 50; 8.2%); 14 subjects (2.3%) identified as Hispanic.

TABLE 1

Demographics

No. (%)Median (IQR)
Age at presentation, y   
 Total cohort — 17 (15–19) 
 GAHT: estrogen — 18 (15.5–20) 
 GAHT: testosterone  17 (15–19) 
Sex assigned at birth   
 Female 428 (70) — 
 Male 183 (30) — 
Affirmed gender   
 Female 176 (28.8) — 
 Male 416 (68.1) — 
 Other 19 (3.1) — 
Race   
 White 544 (89) — 
 African American 50 (8.2) — 
 Asian American 8 (1.3) — 
 Other 12 (1.9) — 
 Not documented 8 (1.3) — 
Ethnicity   
 Hispanic 14 (2.3) — 
 Non-Hispanic 595 (97.4) — 
 Not documented 2 (0.3) — 
No. (%)Median (IQR)
Age at presentation, y   
 Total cohort — 17 (15–19) 
 GAHT: estrogen — 18 (15.5–20) 
 GAHT: testosterone  17 (15–19) 
Sex assigned at birth   
 Female 428 (70) — 
 Male 183 (30) — 
Affirmed gender   
 Female 176 (28.8) — 
 Male 416 (68.1) — 
 Other 19 (3.1) — 
Race   
 White 544 (89) — 
 African American 50 (8.2) — 
 Asian American 8 (1.3) — 
 Other 12 (1.9) — 
 Not documented 8 (1.3) — 
Ethnicity   
 Hispanic 14 (2.3) — 
 Non-Hispanic 595 (97.4) — 
 Not documented 2 (0.3) — 

—, not applicable.

A variety of historical risk factors for thrombosis were present in the cohort (Table 2). Median BMI was 26.0 (IQR: 22.1–32.0); 148 subjects (24.2%) were overweight (BMI = 25–30), and 211 subjects (34.5%) were obese (BMI >30). Other risk factors included smoking tobacco (n = 94; 15.4%) and personal history of migraine with aura (n = 28; 4.6%). Overall, 49 subjects (8%) had documented a family member with thrombosis. Of these, 20 subjects (3.3%) had a parent with thrombosis and 23 (3.8%) had a grandparent with thrombosis, whereas the remainder (n = 6) did not have the affected relative specified. Five individuals (0.8%) had a family history of an inheritable thrombosis risk (eg, factor V Leiden or the prothrombin G20210A polymorphism).

TABLE 2

Risk Factors for Thrombosis

No. (%)Median (IQR)
BMI  26.0 (22.1–32.0) 
 Overall cohort —  
 <18.5 40 (6.5) — 
 18.5–25 212 (34.7) — 
 25–30 148 (24.2) — 
 >30 211 (34.5) — 
Tobacco use   
 Yes 94 (15.4) — 
 No 516 (84.5) — 
Migraine with aura   
 Yes 28 (4.6%) — 
 No 524 (85.8) — 
 Not documented 59 (9.7) — 
Family history of thrombosis   
 Yes 49 (8.0) — 
 No 388 (63.5) — 
 Not documented 174 (28.5) — 
Family history of risk factors for thrombosis   
 Yes 5 (0.8) — 
 No 374 (61.2) — 
 Not documented 232 (38.0) — 
Other personal risk factors   
 Inflammatory bowel disease 3 (0.5) — 
 Juvenile rheumatoid arthritis 1 (0.2) — 
No. (%)Median (IQR)
BMI  26.0 (22.1–32.0) 
 Overall cohort —  
 <18.5 40 (6.5) — 
 18.5–25 212 (34.7) — 
 25–30 148 (24.2) — 
 >30 211 (34.5) — 
Tobacco use   
 Yes 94 (15.4) — 
 No 516 (84.5) — 
Migraine with aura   
 Yes 28 (4.6%) — 
 No 524 (85.8) — 
 Not documented 59 (9.7) — 
Family history of thrombosis   
 Yes 49 (8.0) — 
 No 388 (63.5) — 
 Not documented 174 (28.5) — 
Family history of risk factors for thrombosis   
 Yes 5 (0.8) — 
 No 374 (61.2) — 
 Not documented 232 (38.0) — 
Other personal risk factors   
 Inflammatory bowel disease 3 (0.5) — 
 Juvenile rheumatoid arthritis 1 (0.2) — 

—, not applicable.

Table 3 summarizes the previous hormonal treatment history of subjects. More than half of subjects (n = 328; 53.7%) were previously treated with hormones, most frequently for menstrual suppression. The vast majority of subjects who had used previous hormonal treatment used a nonestrogen containing product (94.3%).

TABLE 3

Previous Hormone Use and Thrombosis Before GAHT

No. (%)Median (IQR)
Previous hormonal use   
 Yes 328 (53.7) — 
 No 262 (42.9) — 
 Not documented 21 (3.4) — 
Previous hormone used   
 Norethindrone contraceptive pill 148 (24.2) — 
 Depo-medroxyprogesterone acetate 113 (18.5) — 
 Combined oral contraceptive pill 35 (5.7) — 
 Norethindrone acetate 15 (2.5) — 
 LNG-IUS 15 (2.5) — 
 Etonogestrel implant 2 (0.3) — 
Thrombosis before GAHT   
 Yes 3 (0.5) — 
 No 608 (99.5) — 
Previous thrombosis type   
 Deep venous thrombosis 2 (66.6) — 
 Stroke 1 (33.3) — 
Treated with anticoagulation, yes 3 (100) — 
Duration of treatment, mo — 4.0 (3.0–10.5) 
No. (%)Median (IQR)
Previous hormonal use   
 Yes 328 (53.7) — 
 No 262 (42.9) — 
 Not documented 21 (3.4) — 
Previous hormone used   
 Norethindrone contraceptive pill 148 (24.2) — 
 Depo-medroxyprogesterone acetate 113 (18.5) — 
 Combined oral contraceptive pill 35 (5.7) — 
 Norethindrone acetate 15 (2.5) — 
 LNG-IUS 15 (2.5) — 
 Etonogestrel implant 2 (0.3) — 
Thrombosis before GAHT   
 Yes 3 (0.5) — 
 No 608 (99.5) — 
Previous thrombosis type   
 Deep venous thrombosis 2 (66.6) — 
 Stroke 1 (33.3) — 
Treated with anticoagulation, yes 3 (100) — 
Duration of treatment, mo — 4.0 (3.0–10.5) 

LNG-IUS, levonorgestrel intrauterine system; —, not applicable.

Three subjects (0.5%) had a thrombotic event before the start of GAHT. Two of these events were VTEs and 1 was an arterial stroke. One event was a provoked VTE occurring in a TGW, and anticoagulation was discontinued before initiation of GAHT. The second event of VTE occurred in a TGM who was maintained on anticoagulation with rivaroxaban during GAHT. The arterial stroke occurred in a TGW who was maintained on aspirin therapy throughout GAHT.

Among the 611 subjects, 182 TGW (29.8%) initiated estrogen for GAHT and 429 TGM (70.2%) initiated testosterone (Table 4). The median age at initiation of GAHT was 17 years (IQR: 16–19). The TGM cohort was slightly younger at initiation of GAHT compared with the TGW cohort (average age was 17 years vs 18 years, respectively; P = .004). All of the TGW used a form of estradiol (oral, transdermal, or intramuscular) with a median dosage of 4 mg (IQR: 2–6). Among the TGM cohort, the median dose of testosterone was 70 mg (IQR: 60–80), with dosage forms including subcutaneous, intramuscular, topical, and transdermal. The median duration of treatment with GAHT for the entire cohort was 574 days (IQR: 283–962). Among TGW, the median duration of treatment was 554 days (IQR: 283–1037) compared with 577 days (IQR: 283–923) for TGM. All subjects underwent monitoring of both estradiol and testosterone levels as they were treated with GAHT. Among estradiol users, the median estradiol level at most recent measurement was 47.2 pg/mL (IQR: 27.2–99.5), and the median testosterone level was 189.5 ng/dL (IQR: 15.2–367). Among testosterone users, the median estradiol level at most recent measurement was 33.2 pg/mL (IQR: 24.0–47.9), and the median testosterone level was 413.0 ng/dL (IQR: 225.0–581.2). Table 4 also shows estradiol and testosterone levels segregated by route of administration.

TABLE 4

GAHT

No. (%)Median (IQR)
Age at initiation, y   
 Overall cohort — 17 (16–19) 
 GAHT: Estrogen — 18 (16–20) 
 GAHT: Testosterone — 17 (16–19) 
GAHT hormone   
 Estrogen 182 (29.8) — 
 Testosterone 429 (70.2) — 
Treatment duration, d   
 Estrogen — 554 (283.0–1037.5) 
 Testosterone  577 (283.0–923.0) 
Estrogen formulation   
 Oral 165 (90.7) — 
 Transdermal 10 (5.5) — 
 Intramuscular 7 (3.8) — 
Testosterone formulation   
 Subcutaneous 312 (72.7) — 
 Intramuscular 105 (24.4) — 
 Gel 11 (2.8) — 
 Transdermal 1 (0.7) — 
Estrogen dose, mg — 4.0 (2.0–6.0) 
Testosterone dose, mg — 70.0 (60.0–80.0) 
Estrogen users   
 All: — — 
  Testosterone levels, ng/dL — 189.5 (15.2–367.0) 
  Estrogen levels, pg/mL — 47.2 (27.2–99.5) 
 Oral: — — 
  Testosterone levels, ng/dL — 193.0 (16.0–375.0) 
  Estrogen levels, pg/mL — 51.2 (29.5–100.0) 
 Transdermal: — — 
  Testosterone levels, ng/dL — 75.0 (30.0–288.0) 
  Estrogen levels, pg/mL — 26.3 (14.3–27.5) 
 Intramuscular: — — 
  Testosterone levels, ng/dL — 18.0 (8.5–408.5) 
  Estrogen levels, pg/mL — 112.0 (17.5–306.5) 
Testosterone users   
 All: — — 
  Testosterone levels, ng/dL — 413.0 (225.0–581.2) 
  Estradiol levels, pg/mL — 33.2 (24.0–47.9) 
 Subcutaneous: — — 
  Testosterone levels, ng/dL — 413.0 (237.8–560.5) 
  Estrogen levels, pg/mL — 33.1 (24.2–47.1) 
 Intramuscular: — — 
  Testosterone levels, ng/dL — 461.5 (219.0–695.2) 
  Estrogen levels, pg/mL — 34.4 (22.4–48.0) 
 Gel: — — 
  Testosterone levels, ng/dL — 373.0 (77.0–719.2) 
  Estrogen levels, pg/mL — 38.0 (26.7–55.1) 
No. (%)Median (IQR)
Age at initiation, y   
 Overall cohort — 17 (16–19) 
 GAHT: Estrogen — 18 (16–20) 
 GAHT: Testosterone — 17 (16–19) 
GAHT hormone   
 Estrogen 182 (29.8) — 
 Testosterone 429 (70.2) — 
Treatment duration, d   
 Estrogen — 554 (283.0–1037.5) 
 Testosterone  577 (283.0–923.0) 
Estrogen formulation   
 Oral 165 (90.7) — 
 Transdermal 10 (5.5) — 
 Intramuscular 7 (3.8) — 
Testosterone formulation   
 Subcutaneous 312 (72.7) — 
 Intramuscular 105 (24.4) — 
 Gel 11 (2.8) — 
 Transdermal 1 (0.7) — 
Estrogen dose, mg — 4.0 (2.0–6.0) 
Testosterone dose, mg — 70.0 (60.0–80.0) 
Estrogen users   
 All: — — 
  Testosterone levels, ng/dL — 189.5 (15.2–367.0) 
  Estrogen levels, pg/mL — 47.2 (27.2–99.5) 
 Oral: — — 
  Testosterone levels, ng/dL — 193.0 (16.0–375.0) 
  Estrogen levels, pg/mL — 51.2 (29.5–100.0) 
 Transdermal: — — 
  Testosterone levels, ng/dL — 75.0 (30.0–288.0) 
  Estrogen levels, pg/mL — 26.3 (14.3–27.5) 
 Intramuscular: — — 
  Testosterone levels, ng/dL — 18.0 (8.5–408.5) 
  Estrogen levels, pg/mL — 112.0 (17.5–306.5) 
Testosterone users   
 All: — — 
  Testosterone levels, ng/dL — 413.0 (225.0–581.2) 
  Estradiol levels, pg/mL — 33.2 (24.0–47.9) 
 Subcutaneous: — — 
  Testosterone levels, ng/dL — 413.0 (237.8–560.5) 
  Estrogen levels, pg/mL — 33.1 (24.2–47.1) 
 Intramuscular: — — 
  Testosterone levels, ng/dL — 461.5 (219.0–695.2) 
  Estrogen levels, pg/mL — 34.4 (22.4–48.0) 
 Gel: — — 
  Testosterone levels, ng/dL — 373.0 (77.0–719.2) 
  Estrogen levels, pg/mL — 38.0 (26.7–55.1) 

—, not applicable.

Overall, few subjects underwent testing for thrombophilia. In total, 17 subjects (2.8%) were referred for evaluation by hematology before initiation of GAHT because of known personal or family history of thrombosis or thrombosis risk or personal history of thrombophilia. Those found to have a laboratory-identified risk factor for thrombosis are included in Table 5. The most common laboratory identified risk factors were elevated plasminogen activator inhibitor-1 (PAI-1; n = 5; 31.3% of tests obtained were abnormal; 0.8% of cohort), PAI-1 gene polymorphism (n = 5; 35.7% of tests obtained were abnormal; 0.8% of cohort), and elevated factor VIII level (n = 4; 23.5% of tests obtained were abnormal; 0.7% of cohort). Ten subjects (2% of tests obtained were abnormal; 1.7% of cohort) had an elevated hemoglobin (>17.7 g/dL). The majority of subjects with an elevated hemoglobin were TGM receiving testosterone; however, one TGW had a hemoglobin of 18.4 g/dL.

TABLE 5

Hematologic Evaluation and Incidence of Thrombosis

No. (%)aNo. Patients With Testing Performed
Referred to hematology 17 (2.8) — 
Thrombophilia evaluation  — 
 Elevated factor VIII (>150%) 4 (23.5) 17 
 Erythrocytosis (>17.7 g/dL) 10 (2.0) 504 
 Activated protein C resistance ratio (<0.78) 1 (6.3) 16 
 PAI-1 (<16.3 IU/mL) 5 (31.3) 16 
 Factor V Leiden heterozygous 2 (11.8) 17 
 Prothrombin G20210A heterozygous 2 (12.5) 16 
 MTHFR 677 homozygous 3 (21.4) 14 
 PAI-1 4G homozygous 5 (35.7) 14 
 Elevated homocysteine (>10.7 µmol/L) 2 (20.0) 10 
Thromboprophylaxis before GAHT   
 Overall cohort 5 (0.8) — 
 History of thrombosis before GAHT 2 (0.3) — 
 No history of thrombosis before GAHT 3 (0.5) — 
Thrombosis on GAHT — 
No. (%)aNo. Patients With Testing Performed
Referred to hematology 17 (2.8) — 
Thrombophilia evaluation  — 
 Elevated factor VIII (>150%) 4 (23.5) 17 
 Erythrocytosis (>17.7 g/dL) 10 (2.0) 504 
 Activated protein C resistance ratio (<0.78) 1 (6.3) 16 
 PAI-1 (<16.3 IU/mL) 5 (31.3) 16 
 Factor V Leiden heterozygous 2 (11.8) 17 
 Prothrombin G20210A heterozygous 2 (12.5) 16 
 MTHFR 677 homozygous 3 (21.4) 14 
 PAI-1 4G homozygous 5 (35.7) 14 
 Elevated homocysteine (>10.7 µmol/L) 2 (20.0) 10 
Thromboprophylaxis before GAHT   
 Overall cohort 5 (0.8) — 
 History of thrombosis before GAHT 2 (0.3) — 
 No history of thrombosis before GAHT 3 (0.5) — 
Thrombosis on GAHT — 

MTHFR, methylenetetrahydrofolate reductase. —, not applicable.

a

For hematologic testing, results are reported as number of abnormal tests and percentage of patients who had testing performed for which results were abnormal.

Among subjects referred to hematology, 3 (0.5%) were recommended to initiate thromboprophylaxis before starting GAHT. All 3 patients had a family history of thrombosis. Additionally, 1 TGW had elevated factor VIII, 1 TGW had an elevated D-dimer, and 1 TGM had the prothrombin gene mutation. All 3 patients were started on prophylactic rivaroxaban before GAHT. Two of 3 individuals with previous thrombosis continued an anticoagulant during GAHT as described above.

No individual in the cohort, TGW or TGM, developed a VTE or arterial thrombosis (including stroke) while on GAHT.

In this retrospective cohort study of transgender youth cared for in a dedicated transgender clinic, we found no incidental occurrence of arterial or venous thrombosis associated with GAHT. This was true for both TGW using estrogen and TGM using testosterone. In contrast to previous studies that reported a higher risk of thrombosis in adult TGW using different formulations of estrogen,9,10  our cohort was uniformly treated with estradiol. Furthermore, serum estradiol levels were followed after initiation of GAHT, and estradiol doses were titrated to maintain serum levels at near physiologic levels. Testosterone levels also were followed among TGM, and similarly, doses were adjusted to maintain physiologic levels. Even among individuals with thrombosis risk factors, such as obesity and tobacco use, GAHT was not associated with thrombosis in our cohort. The demographics of our cohort are similar to those of a national sample of TG individuals, except for an increased ratio of TGM versus TGW.25  However, researchers of other studies of transgender youth also have noted a larger percentage of TGM than TGW.3,26,27 

A small number of youth were at higher risk for thrombosis than the majority of the cohort population. In general, individuals with known thrombophilia (ie, first-degree family member with thrombosis; personal history of factor V Leiden, prothrombin gene mutation) are recommended to avoid estrogen-containing contraceptives.28  However, individuals with these laboratory findings, and those with other risk factors for thrombosis, were able to start and continue GAHT safely in our cohort. These data, however, come with 2 caveats. First, the highest risk individuals in our cohort (with either a personal history of thrombosis or multiple risk factors for thrombosis) received thromboprophylaxis at the initiation of GAHT. Second, other studies demonstrated that the risk of thrombosis with GAHT may increase with longer duration of exposure.9,10 

The risk of thrombosis with use of exogenous estrogen is well documented in cisgender women.29,30  However, there are key differences in thrombosis risk between use of exogenous estrogen for contraception as compared with use of estrogen for GAHT. First, ethinyl estradiol is used in combined oral contraceptives, whereas estradiol is primarily used in GAHT. Second, the greatest risk of thrombosis with use of combined oral contraceptives is within the first 6 months of use.31  Conversely, GAHT is associated with ongoing risk of thrombosis throughout the duration of use.9  Finally, the risk of thrombosis associated with estrogen-containing contraceptives is directly proportional to dose of estrogen.32,33  Currently, recommendations for GAHT advise titrating estradiol dose to achieve physiologic estrogen levels.7  In our cohort, not all individuals had yet achieved physiologic hormone levels because they were in varying stages of gender transition. Our study reports the most recent hormone levels available; titration of hormone dosage to achieve physiologic levels occurring after the study end date would not have been captured. More information is needed to determine if higher estradiol levels for GAHT are associated with thrombosis in transgender youth. Route of administration may also play a role in risk of thrombosis. Oral estradiol, used by the majority of TGW in our cohort, has a first-pass effect in the liver, leading to greater potential changes in hemostatic factors.34  Thus, oral estradiol may have a higher risk of thrombosis compared with transdermal or intramuscular estradiol.35,36  In this cohort, users of intramuscular estradiol had higher estradiol and lower testosterone levels than users of oral estradiol; however, the small number of users of intramuscular estradiol (n = 7) limits the generalizability of these findings.

Risk of thrombosis with GAHT must be tempered with the risk of not treating underlying gender dysphoria, which itself is associated with significantly increased rates of suicidal ideation and suicide attempts. Forty percent of transgender persons report a history of a suicide attempt, a ninefold higher rate than in the general population.25  In adolescence, the risk may be higher, with >60% of transgender youth reporting suicidal ideation.3  GAHT for individuals with gender dysphoria is beneficial for mental health and is associated with a significant reduction in suicide rate.3739  Therefore, initiation of GAHT, even in the setting of thrombosis risk, is critical. Consultation with a hematologist who can counsel transgender youth and their families about the degree of risk may be helpful for informed consent for GAHT. Further studies are needed to assess safety of GAHT among transgender youth with known thrombosis risk and to determine the role of prophylaxis with anticoagulation during GAHT.

Our study has several limitations. First, the study was conducted at a single institution; however, the study included all youth who were started on GAHT since the inception of the Transgender Health Clinic. Second, as a retrospective study, the data were limited to that available through extraction of existing records in the electronic medical record (EMR). On the basis of the EMR, it was sometimes not possible to determine if a subject had discontinued GAHT or was lost to follow-up. Therefore, a risk assessment based on duration of exposure to GAHT is difficult to determine. Third, because of the limitations of data abstraction from an EMR, it was not possible to definitively determine when estrogen-based contraception or menstrual suppression was stopped, precluding our ability to exclude those who used testosterone for GAHT while simultaneously using estrogen-containing medications. However, no thromboses were reported even in the potential context of dual testosterone and estrogen therapy. Fourth, although all subjects were treated according to existing guidelines, the product used (either testosterone or estradiol) was not uniform among the cohort because of factors such as insurance coverage or patient preference. Therefore, determining if formulation of GAHT is associated with thrombosis risk is challenging. Fifth, the most recent testosterone and estradiol levels were reported for each subject. Some subjects had hormone levels below target physiologic levels, likely because they were still in the process of upward titration of the GAHT dose. Finally, the follow-up period for this cohort is limited. At a pediatric institution, as individuals age, they transition to adult care providers and are then lost to follow-up to our institution.

We assessed the risk of thrombosis in the setting of GAHT among adolescent and young adult transgender youth. No youth in our cohort, all of whom received either testosterone or estradiol, developed thrombosis when GAHT was dosed to maintain physiologic hormone levels, despite the existence of preexisting thrombosis risk factors. Further studies are needed to examine the longer-term risk of thrombosis and to evaluate the impact of GAHT on physiologic hemostasis among transgender youth.

We would like to thank Wendi Long, Justin McAdams, and Paula Mobberley-Schuman for their excellent technical assistance.

Drs E Mullins and T Mullins conceived and designed the research, performed the research, analyzed data, and wrote the manuscript; Dr Conard conceived and designed the research; Ms Geer, Ms Metcalf, and Ms Piccola performed the research; Dr Lane analyzed the data; and all authors reviewed the manuscript, approved submission, and agree to be accountable for all aspects of the work.

FUNDING: Funded by the Cincinnati Children’s Hospital Medical Center Academic Research Committee Award (principle investigator: Dr Conard). It was also made possible by support from the Clinical Management and Research Support Core (Cancer and Blood Diseases Institute). The funder did not play a role in the work.

EMR

electronic medical record

GAHT

gender-affirming hormone therapy

IQR

interquartile range

PAI-1

plasminogen activator inhibitor-1

TGM

transgender men

TGW

transgender women

VTE

venous thromboembolism

CCHMC

Cincinnati Children’s Hospital Medical Center

1
Flores
AR
,
Herman
JL
,
Gates
GJ
,
Brown
TNT
.
How many adults identify as transgender in the United States? 2016. Available at: http://williamsinstitute.law.ucla.edu/wp-content/uploads/How-Many-Adults-Identify-as-Transgender-in-the-United-States.pdf. Accessed February 12, 2019
2
Johns
MM
,
Lowry
R
,
Andrzejewski
J
, et al
.
Transgender identity and experiences of violence victimization, substance use, suicide risk, and sexual risk behaviors among high school students - 19 states and large urban school districts, 2017
.
MMWR Morb Mortal Wkly Rep
.
2019
;
68
(
3
):
67
71
3
Eisenberg
ME
,
Gower
AL
,
McMorris
BJ
,
Rider
GN
,
Shea
G
,
Coleman
E
.
Risk and protective factors in the lives of transgender/gender nonconforming adolescents
.
J Adolesc Health
.
2017
;
61
(
4
):
521
526
4
The World Professional Association for Transgender Health
.
Standards of care for the health of transsexual, transgender, and gender nonconforming people. 2011. Available at: https://wpath.org/publications/soc. Accessed February 18, 2019
5
Spack
NP
,
Edwards-Leeper
L
,
Feldman
HA
, et al
.
Children and adolescents with gender identity disorder referred to a pediatric medical center
.
Pediatrics
.
2012
;
129
(
3
):
418
425
6
Bauer
GR
,
Scheim
AI
,
Pyne
J
,
Travers
R
,
Hammond
R
.
Intervenable factors associated with suicide risk in transgender persons: a respondent driven sampling study in Ontario, Canada
.
BMC Public Health
.
2015
;
15
:
525
7
Hembree
WC
,
Cohen-Kettenis
PT
,
Gooren
L
, 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
8
Gialeraki
A
,
Valsami
S
,
Pittaras
T
,
Panayiotakopoulos
G
,
Politou
M
.
Oral contraceptives and HRT risk of thrombosis
.
Clin Appl Thromb Hemost
.
2018
;
24
(
2
):
217
225
9
Getahun
D
,
Nash
R
,
Flanders
WD
, et al
.
Cross-sex hormones and acute cardiovascular events in transgender persons: a cohort study
.
Ann Intern Med
.
2018
;
169
(
4
):
205
213
10
Wierckx
K
,
Elaut
E
,
Declercq
E
, 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
11
Asscheman
H
,
Gooren
LJ
,
Eklund
PL
.
Mortality and morbidity in transsexual patients with cross-gender hormone treatment
.
Metabolism
.
1989
;
38
(
9
):
869
873
12
van Kesteren
PJ
,
Asscheman
H
,
Megens
JA
,
Gooren
LJ
.
Mortality and morbidity in transsexual subjects treated with cross-sex hormones
.
Clin Endocrinol (Oxf)
.
1997
;
47
(
3
):
337
342
13
Schlatterer
K
,
Yassouridis
A
,
von Werder
K
,
Poland
D
,
Kemper
J
,
Stalla
GK
.
A follow-up study for estimating the effectiveness of a cross-gender hormone substitution therapy on transsexual patients
.
Arch Sex Behav
.
1998
;
27
(
5
):
475
492
14
Wierckx
K
,
Van Caenegem
E
,
Schreiner
T
, et al
.
Cross-sex hormone therapy in trans persons is safe and effective at short-time follow-up: results from the European network for the investigation of gender incongruence [published correction appears in J Sex Med. 2016;13(4):732]
.
J Sex Med
.
2014
;
11
(
8
):
1999
2011
15
Pyra
M
,
Casimiro
I
,
Rusie
L
, et al
.
An observational study of hypertension and thromboembolism among transgender patients using gender-affirming hormone therapy
.
Transgend Health
.
2020
;
5
(
1
):
1
9
16
Arnold
JD
,
Sarkodie
EP
,
Coleman
ME
,
Goldstein
DA
.
Incidence of venous thromboembolism in transgender women receiving oral estradiol
.
J Sex Med
.
2016
;
13
(
11
):
1773
1777
17
Ott
J
,
Kaufmann
U
,
Bentz
EK
,
Huber
JC
,
Tempfer
CB
.
Incidence of thrombophilia and venous thrombosis in transsexuals under cross-sex hormone therapy
.
Fertil Steril
.
2010
;
93
(
4
):
1267
1272
18
Abou-Ismail
MY
,
Citla Sridhar
D
,
Nayak
L
.
Estrogen and thrombosis: a bench to bedside review
.
Thromb Res
.
2020
;
192
:
40
51
19
Ellery
PER
,
Hilden
I
,
Sejling
K
, et al
.
Correlates of plasma and platelet tissue factor pathway inhibitor, factor V, and Protein S
.
Res Pract Thromb Haemost
.
2018
;
2
(
1
):
93
104
20
Kavoussi
PK
,
Machen
GL
,
Wenzel
JL
, et al
.
Medical treatments for hypogonadism do not significantly increase the risk of deep vein thrombosis over general population risk
.
Urology
.
2019
;
124
:
127
130
21
Li
H
,
Benoit
K
,
Wang
W
,
Motsko
S
.
Association between use of exogenous testosterone therapy and risk of venous thrombotic events among exogenous testosterone treated and untreated men with hypogonadism
.
J Urol
.
2016
;
195
(
4, pt 1
):
1065
1072
22
Martinez
C
,
Suissa
S
,
Rietbrock
S
, et al
.
Testosterone treatment and risk of venous thromboembolism: population based case-control study
.
BMJ
.
2016
;
355
:
i5968
23
Stanley
K
,
Cooper
J
.
Hormone therapy and venous thromboembolism in a transgender adolescent
.
J Pediatr Hematol Oncol
.
2018
;
40
(
1
):
e38
e40
24
von Elm
E
,
Altman
DG
,
Egger
M
,
Pocock
SJ
,
Gøtzsche
PC
,
Vandenbroucke
JP
;
STROBE Initiative
.
The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies
.
Lancet
.
2007
;
370
(
9596
):
1453
1457
25
James
SE
,
Herman
JL
,
Rankin
S
,
Keisling
M
,
Mottet
L
,
Anofi
M
.
The Report of the 2015 U.S. Transgender Survey
.
Washington, DC
:
National Center for Transgender Equality
;
2016
26
Thoma
BC
,
Salk
RH
,
Choukas-Bradley
S
,
Goldstein
TR
,
Levine
MD
,
Marshal
MP
.
Suicidality disparities between transgender and cisgender adolescents
.
Pediatrics
.
2019
;
144
(
5
):
e20191183
27
Kuper
LE
,
Mathews
S
,
Lau
M
.
Baseline mental health and psychosocial functioning of transgender adolescents seeking gender-affirming hormone therapy
.
J Dev Behav Pediatr
.
2019
;
40
(
8
):
589
596
28
Curtis
KM
,
Tepper
NK
,
Jatlaoui
TC
, et al
. U.S. Medical Eligibility Criteria for Contraceptive Use. In:
MMWR Recomm Rep
, vol.
65
.
2016
:
1
104
29
Vandenbroucke
JP
,
Rosing
J
,
Bloemenkamp
KW
, et al
.
Oral contraceptives and the risk of venous thrombosis
.
N Engl J Med
.
2001
;
344
(
20
):
1527
1535
30
Roach
RE
,
Lijfering
WM
,
Helmerhorst
FM
,
Cannegieter
SC
,
Rosendaal
FR
,
van Hylckama Vlieg
A
.
The risk of venous thrombosis in women over 50 years old using oral contraception or postmenopausal hormone therapy
.
J Thromb Haemost
.
2013
;
11
(
1
):
124
131
31
Bloemenkamp
KW
,
Rosendaal
FR
,
Helmerhorst
FM
,
Vandenbroucke
JP
.
Higher risk of venous thrombosis during early use of oral contraceptives in women with inherited clotting defects
.
Arch Intern Med
.
2000
;
160
(
1
):
49
52
32
de Bastos
M
,
Stegeman
BH
,
Rosendaal
FR
, et al
.
Combined oral contraceptives: venous thrombosis
.
Cochrane Database Syst Rev
.
2014
;(
3
):
CD010813
33
van Hylckama Vlieg
A
,
Helmerhorst
FM
,
Vandenbroucke
JP
,
Doggen
CJ
,
Rosendaal
FR
.
The venous thrombotic risk of oral contraceptives, effects of oestrogen dose and progestogen type: results of the MEGA case-control study
.
BMJ
.
2009
;
339
:
b2921
34
Buckman
MT
,
Johnson
J
,
Ellis
H
,
Srivastava
L
,
Peake
GT
.
Differential lipemic and proteinemic response to oral ethinyl estradiol and parenteral estradiol cypionate
.
Metabolism
.
1980
;
29
(
9
):
803
805
35
Vinogradova
Y
,
Coupland
C
,
Hippisley-Cox
J
.
Use of hormone replacement therapy and risk of venous thromboembolism: nested case-control studies using the QResearch and CPRD databases
.
BMJ
.
2019
;
364
:
k4810
36
Toorians
AW
,
Thomassen
MC
,
Zweegman
S
, et al
.
Venous thrombosis and changes of hemostatic variables during cross-sex hormone treatment in transsexual people
.
J Clin Endocrinol Metab
.
2003
;
88
(
12
):
5723
5729
37
de Vries
AL
,
McGuire
JK
,
Steensma
TD
,
Wagenaar
EC
,
Doreleijers
TA
,
Cohen-Kettenis
PT
.
Young adult psychological outcome after puberty suppression and gender reassignment
.
Pediatrics
.
2014
;
134
(
4
):
696
704
38
Tucker
RP
,
Testa
RJ
,
Simpson
TL
,
Shipherd
JC
,
Blosnich
JR
,
Lehavot
K
.
Hormone therapy, gender affirmation surgery, and their association with recent suicidal ideation and depression symptoms in transgender veterans
.
Psychol Med
.
2018
;
48
(
14
):
2329
2336
39
Achille
C
,
Taggart
T
,
Eaton
NR
, et al
.
Longitudinal impact of gender-affirming endocrine intervention on the mental health and well-being of transgender youths: preliminary results
.
Int J Pediatr Endocrinol
.
2020
;
2020
:
8

Competing Interests

POTENTIAL CONFLICT OF INTEREST: Dr E Mullins received honoraria from Takeda and Bayer for topics unrelated to this research; the other authors have indicated they have no potential conflicts of interest to disclose.

FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.

Supplementary data