|
Long-term testosterone administration
increases visceral fat in female to male transsexuals
Jolanda M. H. Elbers, Henk Asscheman, Jacob
C. Seidell, Jos A. J. Megens and Louis J. G. Gooren
Abstract
| Full Text [PDF]
The distribution of body fat appears to be
predictive of the health risks of obesity. Particularly, an increased amount
of intraabdominal or visceral fat is associated with insulin resistance and
an atherogenic lipid profile (1, 2).
Regional fat distribution differs between
men and women and can, therefore, be regarded as a secondary sex
characteristic.
Subcutaneous fat depots on hips and thighs
are considered typically female, whereas excess fat in men is predominantly
stored in the abdominal regions (3). After correction for the total body fat
mass, men have, on the average, more visceral fat than women (4). This sex
difference suggests a potential role for sex steroid hormones in determining
the site of fat deposition.
In women with high levels of total
testosterone and/or free testosterone, visceral fat accumulation or a high
waist to hip ratio, indicative of increased abdominal fat depots, is found
(5, 6). The association of high testosterone levels with insulin resistance
and an abdominal fat distribution in women has led to different, opposing
interpretations. According to some investigators, the underlying factor is
hyperinsulinemia, which leads to an increase in ovarian testosterone
production (7). Others argue that hyperandrogenicity is the primary event
leading to both accumulation of visceral fat and hyperinsulinemia (8).
In the present study, we investigated
prospectively the effect of long term testosterone administration on body
fat distribution in young, nonobese, female to male transsexuals undergoing
sex reassignment to a male status following a standardized regimen (9). This
enabled us to study the effect of (exogenous) hyperandrogenicity on the
distribution of body fat by quantification of both sc and visceral fat
depots in female subjects.
Subjects and Methods
Subjects
Ten young, nonobese, female to male
transsexuals (mean age 6 sd, 24 6 6 yr; range, 16–33 yr; mean body mass
index, 23.0 6 3.0 kg/m2; range, 18.7–26.9 kg/m2) participated in the
study. They were healthy, as assessed by their medical histories, physical
examinations, and laboratory measurements. They were studied before, after 1
yr, and after, on the average, 3.2 6 0.4 yr of testosterone administration
(range, 2.5–3.8 yr). They were treated with 250 mg testosterone esters (Sustanon
250, Organon, Oss, The Netherlands) every 2 weeks, im, until they were
ovariectomized as part of the cross-sex treatment after, on the average, 1.9
6 0.4 yr. After ovariectomy, they continued testosterone treatment with im
injections of 250 mg testosterone esters every 3 weeks. Two subjects
switched to 160 mg testosterone undecanoate (Andriol, Organon)/ day, orally,
during the last year. The study was approved by the ethical review board of
the Hospital Vrije Universiteit, and all subjects provided informed consent.
Anthropometry and body fat distribution
Body weight was measured to the nearest 0.1
kg, and height was measured to the nearest 0.1 cm with subjects wearing only
underwear. Magnetic resonance (MR) imaging was performed on a whole body
scanner with a magnetic field strength of 0.6 Tesla (Teslacon II, Technicare,
Solon, OH). An inversion recovery pulse sequence was used with a repetition
time of 524 ms, an echo time of 24 ms, and an inversion time of 190 ms. The
slice thickness was 10 mm, and the field of view was 410 or 450 mm.
Transverse MR images were obtained at the
level of the abdomen (lower edge of the umbilicus), the hip (upper margin of
the great trochanter), and the thigh (just below the gluteal fold). Three
images were taken simultaneously at each body region: one image at the level
of the marker, one above, and one below this position, with a gap between
the images of 0.25 cm. In all subjects, the same anatomical markers and
imaging parameters were used in repeatedMRacquisition.
Quantification of sc and visceral fat areas
was performed using an image-analyzing computer program (developed by our
Department of Biomedical Engineering) (10), based on a seed-growing
procedure. After a seed point is placed in a fat depot, this fat depot can
be circumscribed by selection of a pixel intensity range. The intensity
range is selected for each image separately according to the pixel intensity
histogram. The area of the circumscribed fat depot is quantified by
converting the number of pixels to square centimeters. Muscle area was
calculated from subtracting the areas of sc fat, bone, and connective tissue
from the total area on the image below the marker at thigh level. To reduce
variability, image analysis was performed by one observer, and the average
fat area of three images per level was used in the statistical analysis.
Coefficients of variation for intraobserver
variability in sc fat area measurements were 2.3% (abdomen), 2.4% (hip), and
2.2% (thigh); that for visceral fat area measurements was 9.8%. Due to a
technical error, the digital MR information of four subjects for one
measurement occasion was lost. The original MR information on film was
available and was redigitalized (Lumiscan 100, Lumisys, Sunnyvale, CA) to
quantify the fat depots. To study the variability introduced by the
digitalization step, MR data for all subjects were also analyzed using this
procedure. No statistical difference was observed between the two
measurements, and coefficients of variation appeared to be in the same
range, i.e. for sc fat area measurements, between 2–3%, and for the
visceral fat area, 10.7%.
Blood samples
Before and after 1 yr of testosterone
administration, blood samples were obtained after an overnight fast.
Standardized RIAs were used to measure serum levels of testosterone
(Coat-A-Count, Diagnostic Products Corp., Los Angeles, CA) and 17b-estradiol
(Double Antibody, Sorin Biomedica, Saluggia, Italy). Sex hormone-binding
globulin was measured using an immunoradiometric assay (Orion Diagnostica,
Espoo, Finland).
Statistics
Variables and changes in variables during
treatment did not differ statistically from a normal distribution. Baseline
variables are expressed as the mean 6 sd, and changes in variables during
treatment are presented as the mean change from baseline and 95% confidence
intervals.
Because of the small study population, the
paired Wilcoxon signed rank test was used to test the effect of 1 yr and 3
yr testosterone treatments vs. baseline values for the variables studied.
The Spearman rank correlation test was used to study correlations between
variables. Twosided P , 0.05 was considered statistically significant.
Results
After testosterone administration, serum
testosterone increased from 1.4 6 0.7 nmol/L before to supraphysiological
levels after 1 yr of treatment (34.4 6 32.1 nmol/L; P , 0.01), and the
levels of sex hormone-binding globulin in serum decreased from 56 6 31 nmol/L
at baseline to 25 6 10 nmol/L after 1 yr of treatment (n58; P,0.05). Serum
levels of 17b-estradiol decreased from 209 6 138 pmol/L before to 114 6 44
pmol/L after 1 yr of treatment (P 5 0.17). Mean body weight had not changed
significantly after 1 yr of treatment, and the increase in weight of 2.0 6
5.9 kg after 3 yr did not reach significance (Table 1). The changes in
weight ranged from 210.0 to 9.6 kg after 1 yr and from 29.0 kg to 13.9 kg
after 3 yr of treatment.
TABLE 1.
Baseline and changes in body weight, fat area, and muscle area measurements
during testosterone administration in 10 young nonobese female subjects.
|
|
Before treatment
|
Changes from baselinea
|
|
|
Baselineb
|
Range
|
After 1 yr
|
After 3 yr
|
|
Body Wt (kg)
|
67.9 ± 8.4
|
54.3–80.9
|
–0.2(–3.8, 3.5)
|
2.0(–2.2, 6.2)
|
|
Abdominal sc fat area (cm2)
|
194 ± 79
|
69–336
|
–50(–89,–11)c
|
–11(–51, 29)
|
|
Hip sc fat area (cm2)
|
195 ± 44
|
117–254
|
–57(–80,–35)c
|
–30(–72, 11)
|
|
Thigh sc fat area (cm2)
|
128 ± 56
|
182–358
|
–67(–89,–46)c
|
–28(–75, 20)
|
|
Visceral fat area (cm2)
|
32 ± 15
|
10–59
|
2(27,10)
|
13(4, 22)c
|
|
Muscle area (at thigh level; cm2)
|
266 ± 20
|
232–300
|
33(19,46)c
|
28(10, 47)c
|
a
Changes from baseline are expressed as means and 95% confidence intervals (in
parentheses).
b Baseline values are the mean 6 SD.
c Significant change from baseline (P , 0.05, by paired Wilcoxon signed
rank test).
After 1 yr of testosterone administration, sc
fat area measurements had decreased significantly at all levels measured,
whereas the mean visceral fat area had not changed (Table 1). Compared to
baseline, the area measurements after 3 yr of treatment showed small
nonsignificant reductions in sc fat. In contrast, an absolute increase of 13
cm2 (95% confidence interval, 4–22 cm2) was observed in the visceral fat
area (P 5 0.013). The mean relative percent change in the visceral fat area,
calculated from the individual changes from baseline, was 47% (95%
confidence interval, 8–91%).
A graphic presentation of the relative
changes, expressed as percentage from baseline, after 1 yr, and after 3 yr
of testosterone administration, is given in Fig. 1. Relative changes in body
weight were strongly correlated with relative changes in visceral fat and sc
fat at the abdominal level (Fig. 2). After both 1 and 3 yr of testosterone
administration, the relative changes in visceral fat were greater than those
in sc abdominal fat.
 FIG.
1. Relative changes, expressed as percent change from baseline, in
visceral and sc fat depots after 1 and 3 yr of testosterone administration in
10 female subjects. visc, Visceral fat; sc abd, sc abdominal fat; hip, sc hip
fat; thigh, sc thigh fat. *, P , 0.05. Data are the mean ± SD.
Discussion
The present study shows that long term
testosterone administration to young, nonobese female subjects induced a
redistribution of fat depots, with a preferential accumulation of visceral
fat if weight gain occurred. A shift from a typical female fat distribution
to a more male type of body fat distribution was observed. After 1 yr, a
relative redistribution of body fat had begun; at that time the visceral fat
area had not yet increased significantly, but significant reductions were
seen in the sc fat depots. Fat area measurements on MR images obtained after
3 yr of testosterone administration showed an absolute increase in visceral
fat, whereas sc fat depots were reduced, but were no longer significantly
different from baseline.
In a study by Lovejoy et al. (11), the
effect of exogenous androgens on regional fat distribution was investigated
in obese, postmenopausal women undergoing weight loss through caloric
restriction. In this study, an absolute increase in visceral abdominal fat
and a loss of sc fat at the levels of abdomen and thigh were observed after
9 months of administration of nandrolone decanoate, an anabolic steroid with
weak androgenic activity. The visceral fat depot had increased despite a
weight reduction. In the present study, we did not observe an absolute
increase in mean visceral fat after 1 yr of testosterone administration.
However, a strong positive correlation was
present between changes in body weight and changes in the visceral fat area.
Also, for the sc fat depots, there was a positive correlation with changes
in body weight after 1 yr, but all subjects lost sc fat regardless of
increases in body weight. Thus, weight gain resulted in a preferential
accumulation of fat in the visceral depot in hyperandrogenic female
subjects. After 3 yr of treatment, an absolute increase in mean visceral fat
was observed, and relative changes in this fat depot were larger than those
in the sc abdominal fat depot. It seems that long term exposure to high
levels of testosterone is required to increase the visceral fat depot in
young, nonobese female subjects.
It is not likely that a state of
hypoestrogenism after ovariectomy might (partially) explain our findings.
Part of the administered testosterone is peripherally aromatized to
estradiol.
In a study in which we investigated
ovariectomized female to male transsexuals receiving similar testosterone
dosages, serum estradiol levels were not different from levels in eugonadal
women in their early follicular phase (12) and were similar to those in
eugonadal men. Thus, after ovariectomy, testosterone administration to our
subjects still generated serum estradiol levels in a range comparable to
levels in eugonadal women in the follicular phase. Further, the association
between weight gain and visceral fat accumulation points in the same
direction after 1 yr of testosterone administration as after 3 yr of
testosterone administration, indicating that ovariectomy did not lead to a
change in this relationship.
A limitation of the present study is the
fact that it was not possible to compose a control group of young women with
the same degree of variation in weight over 3 yr of follow-up. To our
knowledge no detailed long term studies have been performed investigating
prospectively the changes in the visceral fat depot in young women. However,
cross-sectional studies in young women have shown that fat tissue is mainly
located in the sc fat depots and that excess fat is preferentially stored
sc, with a rather constant visceral fat depot (3, 4, 13).
We, therefore, compared our data with those
obtained for quantification of sc and visceral fat areas in 34 Dutch women
with a wide range in age and body mass index (13). It appeared in our
subjects that the increase in visceral fat area was larger and the change in
sc fat area was smaller than expected on the basis of findings in the
comparison group, although such a comparison with our study population
should be performed with caution.
In an earlier study we found that 4-month
administration of similar doses of testosterone to female to male
transsexuals (of comparable age) did not increase fasting insulin levels
significantly, but did lead to decreased insulin sensitivity (14). This
observation, combined with our present results, shows similarities with the
findings in women with high endogenous testosterone levels. The latter show
both increased abdominal fat depots and insulin resistance. In the subjects
of our study, nonobese, endocrine unremarkable, female subjects between the
ages of 16–33 yr, the primary event leading to abdominal fat accumulation
and insulin insensitivity was exogenous hyperandrogenism. This observation
might be relevant for determination of the primary event in women with a
combination of hyperandrogenism, insulin resistance, and abdominal fat
accumulation.
Some researchers believe that
hyperinsulinemia is the primary event (7). Of note, however, is the fact
that in our experiment testosterone levels were far above levels encountered
in most spontaneous hyperandrogenic states in women.
These findings in women are dissimilar with
some observations in men. Cross-sectional studies in men suggest an inverse
association between testosterone levels and abdominal fat distribution (15).
Oral testosterone treatment of middle- aged obese men with low normal
testosterone levels seemed to reduce visceral fat (16). The oral
administration might have been significant for the effects of the
testosterone preparation, as in a study comparing the effects of oral
anabolic steroids with those of parenteral testosterone treatment (as used
in our study), only the oral preparation had this effect, not the parenteral
testosterone preparation (17).
Further, factors such as age or duration of
testosterone exposure may determine the effect of testosterone on visceral
fat. The men in the above studies (16, 17) were at least 40 yr of age or
older, with a presumed exposure to their endogenous testosterone of some 25
yr.
FIG. 2. Relation between relative percent changes in body weight
(x-axis) and relative percent changes in visceral fat areas (solid circles)
and sc abdominal fat areas (sc abd fat; open squares; yaxis) after 1 yr (left
graphic) and after 3 yr (right graphic) of testosterone administration in 10
female subjects. *, P , 0.05 (Spearman rank correlations).
This study resolves some of the uncertainties
with regard to the association between testosterone and visceral fat in
women by demonstrating that long term testosterone exposure increases
visceral fat in young, nonobese female subjects.
Acknowledgements
We thank T. Schweigmann from the Department
of Diagnostic Radiology of the Hospital Vrije Universiteit for MR
acquisition, and F. Hoogenraad and H. van de Mortel from the Department of
Biomedical Engineering of the Hospital Vrije Universiteit for their
technical assistance with the image analysis. We also thank J. Welleweerd
from the Department of Radiotherapy of the University Hospital Utrecht for
the opportunity to redigitalize MR information.
References
1. Bjo¨rntorp P. 1991 Metabolic
implications of body fat distribution. Diabetes Care. 14:1132–1143.
2. Kissebah AH, Krakower GR. 1994 Regional
adiposity and morbidity. Physiol Rev. 74:761– 811.
3. Enzi G, Gasparo M, Biondetti PR, Fiore
D, Semisa M, Zurlo F. 1986 Subcutaneous and visceral fat distribution
according to sex, age, and overweight, evaluated by computed tomography. Am
J Clin Nutr. 44:739 –746.
4. Lemieux S, Prud’homme D, Bouchard C,
Tremblay A, Despre´s J-P. 1993 Sex differences in the relation of visceral
adipose tissue accumulation to total body fatness. Am J Clin Nutr.
58:463– 467.
5. Evans DJ, Barth JH, Burke CW. 1988 Body
fat topography in women with androgen excess. Int J Obesity.
12:157–162.
6. Wild RA. 1995 Obesity, lipids,
cardiovascular risk, and androgen excess. Am J Med. 98(Suppl
1A):27S–32S.
7. Poretsky L. 1991 On the paradox of
insulin-induced hyperandrogenism in insulin-resistant states. Endocr Rev.
12:3–13.
8. Bjo¨rntorp P. 1993 Hyperandrogenicity
in women–a prediabetic condition? J Intern Med. 234:579 –583.
9. Walker PA, Berger JC, Green R, Laub DR,
Reynolds CL, Wollman L. 1985 Standards of care. The hormonal and surgical
sex reassignment of gender dysphoric patients. Arch Sex Behav. 14:79
–90.
10. Elbers JMH, Haumann G, Asscheman H,
Seidell JC, Gooren LJG. 1995 Reproducibility of fat area measurements in
young non-obese subjects by computerized analysis of magnetic resonance
images [Abstract]. Int J Obesity. 19(Suppl 2):51.
11. Lovejoy JC, Bray GA, Bourgeois MO, et
al. 1996 Exogenous androgens influence body composition and regional body
fat distribution in obese postmenopausal women–a clinical research center
study. J Clin Endocrinol Metab. 81:2198 –2203.
12. Spinder T, Spijkstra JJ, Gooren LJG,
Hompes PGA, van Kessel H. 1989 Effects of long-term testosterone
administration on gonadotropin secretion in agonadal female to male
transsexuals compared with hypogonadal and normal women. J Clin
Endocrinol Metab. 68:200 –207.
13. Seidell JC, Oosterlee A, Deurenberg P,
Hautvast JGAJ, Ruijs JHJ. 1988 Abdominal fat depots measured with computed
tomography: effects of degree of obesity, sex, and age. Eur J Clin Nutr.
42:805– 815.
14. Polderman KH, Gooren LJG, Asscheman H,
Bakker A, Heine RJ. 1994 Induction of insulin resistance by androgens and
estrogens. J Clin Endocrinol Metab. 79:265–271.
15. Seidell JC, Bjo¨ rntorp P, Sjo¨ stro¨m
L, Kvist H, Sannerstedt R. 1990 Visceral fat accumulation in men is
positively associated with insulin, glucose, and C-peptide levels, but
negatively with testosterone levels. Metabolism. 39:897–901.
16. Mårin P, Holmång S, Jo¨nsson L, et
al. 1992 The effects of testosterone treatment on body composition and
metabolism in middle-aged obese men. Int J Obesity. 16:991–997.
17. Lovejoy JC, Bray GA, Greeson CS, et al.
1995 Oral anabolic steroid treatment, but not parenteral androgen treatment,
decreases abdominal fat in obese, older men. Int J Obesity.
19:614–624.
1
This work was supported by The Netherlands Organization for Scientific
Research (Grant 904–62-124). Presented in part at the Sixth European
Congress on Obesity, May 31 through June 3, 1995, Copenhagen, Denmark.
Citation: Division of Endocrinology/Andrology, Hospital Vrije Universiteit
(J.M.H.E., H.A., J.A.J.M., L.J.G.G.), Amsterdam;
and the Department of Chronic Disease and
Environmental Epidemiology, National Institute of
Public Health and Environmental Protection (J.C.A.),
Bilthoven, The Netherlands. J Clin
Endocrinol Metab 1997 Jul;82(7):2044-7
|
