Cross-sex hormone use does not increase cardiovascular risk in young male-to-female transsexuals: a study of self-medication by healthy Thai cabaret dancers

We prospectively recruited 102 MtF transsexual women to participate in this study. Participants were monitored for the effects of CSHs (estrogen, progesterone, and antitestosterone) on health. Approximately 65% of male subjects (n = 66, average age = 28 years) used CSH for the purpose of changing from masculine to feminine characteristics with the average hormone use period of 12.5 years. All hormone products were used based on self-administration dose and schedule. Hormones used consisted of commercially available contraceptive pills and injectables. Dosages varied arbitrarily. The control group consisted of healthy men without CSH use (n = 36, average age = 29.1 years). The CSH group clearly exhibited a lower level of smoking than the control group (19.6% vs 54.3%). For cardiovascular related parameters, HDL levels were significantly higher in the CSH group than in the control group; while BMI and CRP levels in the CSH group were significantly lower. Meanwhile, other parameters including systolic and diastolic blood pressure, total cholesterol, triglyceride, LDL and Fb were found to be comparable among the two groups (Table 1).

The CSH group exhibited significantly higher female sex hormone levels (17β estradiol and progesterone) and lower male sex hormone (testosterone). The average estrogen level in the CSH group was approximately five times the level in the control group (145.3 ± 417.2 vs. 28.2 ± 16.1 pg/mL). There were high variations in the CSH group, presumably because of various estrogen self-consumption regimens. Among those in the CSH group, there was one participant with an extremely high estrogen level (2,567 pg/mL) who had an injection of estrogen before the blood sample collection. Progesterone and testosterone were found within normal levels in both groups despite significantly lower levels in the CSH group. In the control group, all hormone levels were at baseline levels found in normal men (Table 2).

Linear correlation analysis (Table 3), under the classification of two groups of participants (CSH and control group), showed a negative correlation between total cholesterol and LDL, with 17β-estradiol in the CSH group. Whereas, testosterone and BMI levels were found to be positively correlated. In the control group, only the levels of 17β-estradiol showed a negative correlation with age. It is noted that an increase of female sex hormone level in the control group was found to correlate with an increase of CRP. However, progesterone showed no correlation with any markers in either the CSH or control group.

Presumed risk prevalence analysis, shown in Table 4, demonstrated standard health risk levels for cardiovascular diseases according to JNC 7 for age, blood pressure, lipid profile, and inflammatory biomarkers comparing between CSH and control groups [19]. Results from this study showed that participants in the control group had a predicted higher prevalence of disease susceptibility than those in the CSH group for all parameters. The only significant difference was found for CRP in which the control group contained 44% of the population proportion who had exceeding the levels of CRP (>1.8 mg%), while only 17% were found in the CSH group (P < 0.01).

Logistic regression was used to evaluate the effects of CSH on cardiovascular risk factors in MtF transsexuals (Table 5). Both a crude model, to reflect the effect of CSH alone, and an adjusted model for the additional effects of age and smoking with CSH use were considered in this study. Results accordingly showed the effect of CSH on CRP, but no other cardiovascular risk parameters. Odd ratios (OR) for the effect of CSH on CRP levels using a crude model and an adjusted model were found to be 0.26 (CI 0.1–0.68) (P < 0.01) and 0.34 (CI 0.13–0.93) (P < 0.05), respectively.

Results from this cross-sectional study demonstrated favorable effects of CSH use on the results of BMI, HDL, and CRP when comparing men who were constantly exposed to CSH and a non-using control group. Data showed that mean values of all three sex hormones and all clinical markers were within normal reference range, perhaps because of criteria that had excluded severe and/or chronic diseases, and the young age of participants. In addition, participants were actors in a cabaret setting where they constantly exercised and had a normal lifestyle. Despite the higher proportion of smoking in the control group, smoking exhibited no predicted effect on cardiovascular markers when analyzed with logistic regression and adjustment.

Changes in cardiovascular markers because of long-term use of CSH in MtF transsexuals have been reported in previous studies [10, 16, 20]. They reported no significant hazardous effects from long-term use of CSH except for ethinyl estradiol, which may relate to cardiovascular death [20]. The increase of HDL by 10%-24% (approximately 6 mg/dL from baseline) was described in CSH using subjects. This was consistent with our findings. We report an increase of approximately 20% (mean difference 8.3 mg/dL (CI 0.2-16.4) in HDL in the CSH group. These findings are similar in several previous studies with constant reports of increased HDL levels in males who used CSH, including those who had aromatase deficiency and MtF transsexuals despite various outcomes on other lipid profiles (total cholesterol, triglyceride, and LDL) [21-23]. An increase of HDL with sex hormone use was also observed in women [24]. BMI and CRP, by contrast, were previously reported to be increased in CSH groups, but were not significantly different [16]. We hypothesized that the differences between previous studies and the current study were likely to be a consequence of participant selection criteria. The current study was conducted in subjects who were homogeneous in terms of young age and regular exercise; whereas the earlier studies contained higher heterogeneous confounders including age, health status, type, and period of hormone use of the subjects.

Another long-term study of hormone use in MtF transsexuals showed an increased risk of cardiovascular-related death in CSH users of approximately threefold compared with normal men (4.1% in CSH group and 1.3% in never or former use) following the use of CSH for 18.6 years [20]. Interestingly, when analyzed by age, the cardiovascular risk was greater in the older group (older than 40 years) and no risk was found at younger ages despite a similar duration of CSH use. Therefore, age is likely to be the key distributing factor for the effects of CSH. This might explain the lack of negative health impact from CSH use found in the present study.

A study in women found that high-dose sex steroid hormones increases CRP inflammatory marker levels [25], whereas studies in MtF transsexuals and men with prostate cancer showed both increases and decreases in CRP [16, 26]. Studies on the effects of oral ethinyl estradiol (100 μg/d) in 15 MtF transsexuals on CRP [27] and in patients with prostate cancer [26] showed an association of estrogen treatment with increased CRP. Because these two studies studied men with estrogen interventions, the conflicting results on CRP levels may be accounted for by several factors. First, the previous study was conducted in MtF transsexuals who administered much higher doses of estrogen than those in our study. Second, cancer patients frequently had high CRP levels up to 100 μg/mL [28], which may contribute to the increase of CRP found by Lurer et al.

Fibrinogen levels in MtF transsexuals were also investigated in this study as a surrogate marker for the development of venous thromboembolisms (VTE), possibly related to CSHs. Despite a previous report of approximately 2% increased risk of VTE in MtF transsexuals who used CSHs [29], we did not observe this risk from measurement of fibrinogen. This avoidance of risk was likely because participants in this study were healthy and maintained a normal lifestyle. Furthermore, overall levels of sex hormones in our CSH using participants were relatively low, particularly progesterone, which ranged between low and normal. This is in accordance with studies that demonstrated correlation of VTE and progesterone levels rather than estrogen [30].

No records or actual written diaries of hormonal use were retrieved from our participants. Only memory of hormonal use was used and there may be recall bias in this study. CSH studies in developed countries had been performed in only sex change clinics and under the control of expert clinicians. The present study of self-medication with CSH by Thai men was conducted in a real-life field setting. The levels novel cardiovascular outcome markers, CRP and Fb were assessed in these MtF transsexuals. When compared with previous studies, the present study consisted of a larger number of transsexual participants, as distinct from other studies. However, weaknesses could be overcome with a stronger methodological design and more clinical measurements. A larger sample size, hormone utilization diaries, multiple clinical measurements (for average level), serum hormonebinding globulin (SHBG) measurement, more focus on outcomes, and longitudinal follow-up are needed in future field studies. To our knowledge, this study is the first report regarding cardiovascular risk factors in CSH using MtF transsexuals in Thailand. We were able to include a larger group of participants than in previous reports. The smaller MtF transsexual population in most studies included only from 10 to 30 participants per study.

Our findings were consistent with other studies that showed no serious risk or health problems with CSH use at a young age. Studies with longer follow ups and a stronger design would help to clarify this issue further.