Androgen honnone levels in women with male pattem baldness 
Clinical study 
ANDROGEN HORMONE LEVELS 
IN WOMEN WITH 
MALE PATTERN BALDNESS 
N. Martinovic, J. Micic, S. Ivanovic and A. Krunic 
ABSTRACT 
Background Adrogen hormones in women with androgenetic alopecia (AA) were studied 
Materials and methods. The prospective study included 50 women with androgenetic alopecia in fertile age. 
Blood and urine were taken on the 7th and 21st day of the same menstrual cycle. Free testosterone (T), 
androstanediole (A-DIOL) and dehydroepiandrosterone (DHA) were assayed in serum. In the 24-hour urine 
T, DHA, androsterone (A), etiocholanolone (E) and 17-ketosteroid (17-KS) levels were monitored. Results. 
Statistical deviation from physiological levels was observed for serum T. 
Conclusion. It is suggested that A-DIOL was possibly the principal androgen in etiopathogenesis of AA m 
women. The hormona! disorder seems to be expressed in the estrogen phase of the menstrual cycle. 
KEY WORDS 
alopecia, androgenetic, testosterone, androgen fractions, Jemale sex 
INTRODUCTION 
The studies of metabolism and mechanisms of 
androgen action on the hair follicle as one of 
significant target units of androgenetic steroids have 
not significantly contributed to the solution of the 
basic dilemma: why and how do androgenetic hormo-
nes cause hair follicle involution constantly in the 
same regions of capillitium which are predisposed 
for androgenetic alopecia (AA), without disturbing 
the follicle function in the remaining regions? Neither 
acta dennatovenerologica A.P.A. Vol 6, 97, No 3 
do in vitro study methods (incubation of isolated 
hair roots with particular androgens and determination 
of metabolite concentration in incubates or fibroblast 
culture as a model of studying the androgen binding 
capacity by the skin) enable an exact insight into 
complex metabolic and biological processes induced 
by androgen steroids. 
Special challenge in AA in women is due to the 
specificity of general hormona! milieu and to. the 
fact that alopecia nevertheless represents stigmata 
of certain degree of androgenization1• This high 
incidence of AA in women is puzzling. Some authors 
93 
Androgen honnone leve/s in women with male pattern baldness 
indicate the existence of metabolic disorders already 
on secretory level in women with AA2-4, while 
others assume an androgen metabolic disorder on 
cellular effector level5-13• Although it bas been proven 
that hair follicles posses enzymatic systems for 
metabolizing individual androgen steroids, metabolism 
is directed towards a-metabolic route. 
MATERIALS AND METHODS 
The purpose of the investigation was to study the 
levels of relevant androgen hormones in women 
with AA. The prospective study included 50 women 
in fertile age (Figure l.) in whom AA was diagnosed: 
Ludwig I type in 42% and Ludwig II type in 58% 
according to Ludwig's classification18. The study inclu-
ded female subjects without clinical signs of hirsutism 
in whom by gynecological examination an ovarian 
organic disease was excluded, and who did not 
receive oral hormonal contraceptives19•24 • 
The quantitative analysis of relevant hormones 
(free testosterone [f/, androstanediole /A-DIOL/, 
dehydroepiandrosterone /DHA/ in serum and DHA, 
No. of 
patient 
20· 
25 
15 
25-
30 
14 
30-
35 
Age intervals 
Figure l. Age distribution of 
androgenetic alopecia 
35- 40-
40 45 
50 females with 
androsterone /A/, etiocholanolone /E/, 17-ketosteroids 
/17-KS/ in 24-hour urine) were performed by adsorp-
tion thin layer chromatography25 • The material (blood 
Table l . Serum testosterone (T), dehydroepiandrosterone (DHA), androstanediole (A -DIOL) and urine 
dehydroepiandrosterone (DHA) levels in Jemale patients with androgenetic alopecia type I and II (n = 50) 
94 
hormone type 
(normal values) 
day of period 
X. mm 
X max 
X 
SE 
X 
KV% 
r 
t 
[T] 
(ng%) 
(18 - 22) 
7. 21. 
11 9,6 
60 37 
26,1 23,4 
11,0 7,9 
41,6 33,5 
0,12; NS 
1.396; NS 
[DHA] 
µg% 
( 40 - 60) 
7. 21. 
18 15 
105 100 
48,6 52,6 
18,5 20,6 
37,6 39,3 
0,25; NS 
1.011; NS 
[A-DIOL] 
µg % 
(20 - 40) 
7. 21. 
12 7 
72 90 
31,8 33,4 
15,2 16,2 
47,2 48,3 
0,60; p<0,01 
0.502; NS 
[DHA] 
(mg/24h) 
(1,5-2) 
7. 21. 
1,05 1,05 
5,52 3,62 
2,01 1,98 
0,84 0,58 
41,1 29,1 
0,44 p<0,05 
0.206 NS 
acta dennatovenerologica A.P.A. Vol 6, 97, No 3 
T 
mg% 
40 
35 
30 
25 
20 
15 
10 
5 
o 
Androgen hormone leve/s in women with male pattem baldness 
20 - 24 25 - 29 30 - 34 35 - 39 40 
Age intervals 
Figure 2. Mean values of testosterone ( T ) in sera of 50 females with androgenetic alopecia. The samples were 
taken at day 7 and 21 of the same menstrual cycle 
DHA 
mg% 
100 
80 
60 
40 
20 
o 
day 7 
~ ............ . ........................ ............ :/ 
1----------------------------'-------; ................ .. 
day 21 
1 
20 - 24 
1 
25 - 29 
1 
30 - 34 
1 
35 - 39 
1 
40 i 
Age intervals 
Figure 3. Mean values of dehydroepiandrosterone (DHA) in sera of 50 females with androgenetic alopecia. 
Samples were taken at day 7 and 21 of the same menstrual cycle 
A-DIOL 
mg% 
40 
35 
30 
25 
20 
15 
10 
5 
o 
······ 
··· ·· -····· ····· 
day 7 
day 21 
20 • 24 25 - 29 30 - 34 35 • 39 40 i 
Age intervals 
Figure 4. Mean values of androstanediol (A-DIOL) in sera of 50 females with androgenetic alopecia. Samples 
were taken at day 7 and 21 of the same menstrual cycle 
samples in the morning biorhythm) and 24-hour 
urine were taken on the 7th and 21st day of the 
same menstrual cycle. The results of densitograms 
acta dermatovenerologica A.P.A. Vol 6, 97, No 3 
were compared with physiological leve! parameters 
for women in reproductive period: 
T 18-22 ng/dl, A-DIOL 20-40 µg/dl, DHA 40-60 
95 
Androgen honnone leve/s in women with male pattem baldness 
Table 2. Dependence of dehydroepiandrosterone (DHA) and testosterone levels in the patients' sera 
7th 
7th 
7th 
21st 
21st 
21st 
decreased 
(:S; 39) 
physiologic 
(40 - 60) 
increased 
c~ 61) 
decreased 
(:S; 39) 
physiologic 
(40 - 60) 
increased 
c~ 61) 
3 
6 
2 
4 
7 
1 
day 7 of the period: X2 = 6.159; NS 
6 
12 
4 
8 
4 
12 
10 
2 
2 
10 
4 
20 
4 
4 
20 
8 
8 
12 
7 
8 
4 
10 
16 
24 
14 
16 
8 
20 
11 
28 
11 
14 
21 
15 
22 
56 
22 
28 
42 
30 
day 21 of the period: X2 = 11.142, p<0,05; c = 0,43; p<0,05 
µg/dl in serum and T 20-50 µg/dl, DHA 1.5 - 2 mg/ 
24h, A 1-3 mg/24h, E 3-6 mg/24h and 17- KS 6 -
14 mg/24h in urine. 
Since the literature data indicate the activities of 
o.-series, directing T metabolism towards A-DIOL in 
hair follicles in general, and particularly in regions 
predisposed to AA, the statistical analysis of results 
was directed towards correlative relationships: serum 
T • A-DIOL, serum T • DHA, serum DHA • 
urine DHA. The outlined parameters were monitored 
particularly for the 7th and 21st days of the same 
menstrual cycle. 
STUDY RESULTS 
The results of the statistical analysis of relevant 
steroids according to five-year age intervals showed 
that significant deviation from physiological levels 
was registered only for serum T (Figure 2). By F-test 
(variance analysis) it was evidenced that the female 
subjects' age did not influence either serum T levels 
or levels of other parameters. Thus, further analysis 
was performed for the female subject group as a 
96 
whole. Since the study was directed towards the a-
metabolic route of steroids and elevated serum T 
levels were registered, further statistical analysis of 
results was directed towards serum T, DHA and A-
DIOL levels and DHA in urine (Table 1, Figures 3, 
4). Statistical analysis (C- coefficient of association) 
showed that there was a praven association of 
serum T levels for the 7th and 21st day of the cycle 
(C= 0.55, p<0.01). For serum T, DHA (Table 2) 
and A levels there were no significant differences 
found in average values on the 7th and 21st day of 
the cycle. For serum A-DIOL levels (p<0.01) as well 
as urine DHA levels (p<0.05) there were found 
significant level dependencies between the 7th and 
21st day of the cycle (Tables 3, 4). Further statistical 
analysis (r - coefficient of correlation) was directed 
towards cci-relative relationships: serum DHA • T, 
serum T • A-DIOL and serum DHA • urine 
DHA. Significant level dependency of serum DHA 
and T (p<0.05) was registered, as well as serum T 
and A-DIOL (p<0.05) on the 21st day of menstrual 
cycle, while for serum and urine DHA there was not 
significant leve! dependency either for the 7th or the 
21st day ( coefficient of correlation - rx)· 
acta dennatovenerologica A.P.A. Vol 6, 97, No 3 
Androgen honnone leve/s in women with male pattern baldness 
Table 3. Dependence of testosterone (T) and androstanediol (A-DIOL) levels in the patients' sera 
7th 
7th 
7th 
21st 
21st 
21st 
decreased 
(:::; 19) 
physiologic 
(20 - 40) 
increased 
( ?'. 41) 
decreased 
(:::;19) 
physiologic 
(20 - 40) 
increased 
(?: 41) 
1 
6 
4 
2 
7 
3 
day 7 of the period: X2 = 3. 625; NS 
2 
12 
8 
4 
4 
6 
day 21 of the period: X2 = 11.003, p<0,05 
DISCUSSION 
The sex hormone binding globulin (SHBG) levels 
were not determined due to technical reasons, which 
does not reduce the value of this study since the 
unbound T fraction is the only one which is biologically 
active1• 26•27 . In healthy women 50-70% of T originates 
from peripheral androstenedione skin conversion. In 
54 % of female subjects there were found elevated 
serum T levels on the 7th day of the cycle, and in 
44% on the 21st day of the cycle. The association 
in T levels on the 7th day and 21st day correlated 
with the literature data28•29• In female subjects elevated 
T levels were not found in urine contrary to some 
other authors2•3• DHA is a precursor hormone for 
C19 and C18 steroids. The results did not suggest 
the presence of significant dependency between serum 
and urine DHA levels in women with AA. However, 
the significance of DHA for the studied pathology 
was confirmed by the analysis of the relationship of 
serum DHA • T, since DHA acted identically also 
on the leve! relation the 7th • 21st day of the 
menstrual cycle ( according to well known a - metabolic 
acta dennatovenerologica A.P.A. Vol 6, 97, No 3 
1 
1 
7 
4 
6 
9 
2 
14 
8 
2 
12 
18 
8 
13 
6 
18 
4 
16 
26 
12 
36 
8 
pathway of androgen hormones). 
10 
26 
14 
3 
31 
16 
20 
52 
28 
6 
62 
32 
Referring to the metabolism: DHA • androste-
nedione • T • A-DIOL, and having in mind high 
serum T levels in the majority of female subjects, it 
may be supposed that the ~lism is directed 
towards the leve! DHA • T in progesterone phase 
of the cycle. A significant dependency of DHA and 
T levels on the 21st day was 
1
registered. By statistical 
analysis T was shown as the targeted marker for 
AA in women, due to the praven leve! association 
for the 7th and 21st day of the cycle. 
There was noticed a different behavior of individual 
steroids depending on the cycle phase, on the basis 
of which it may be concluded that the estrogen 
phase in female subjects was followed by hormona! 
dysbalance which in progesterone phase receives its 
targeted route ( there was registered dependency at 
the leve! T • A-DIOL in the progesterone phase). 
This observation suggests the metabolic route of a-
series of C
19 
steroids in women with AA. 
97 
Androgen honnone leve/s in women with male pattern baldness 
CONCLUSION 
The results indicate that in women with AA the 
estrogen phase is the one followed by hormonal 
dysbalance which in progesterone phase receives its 
targeted route and suggest the possible significant 
biological activity of A-DIOL as a peripheral steroid 
in etiopathogenesis of AA in women. A possible 
way to prave this hypothesis would be an in vitro 
study which is unfortunately at present technically 
not possible to apply as a standardized method. 
REFERENCES 
l. Kveder C], Gibson M, Krušinski AP. Hirsutism: 
Evaluation and Treatment. J Am Acad Dermatol 1985; 
211: 215-25. 
2. Apostolakis M, Ludwig E, Voight KD. Testosteron -
Oestrogen und Gonadotropen - Ausscheidung bei diffusen 
unerblicher Alopecia. Klin Woschr 1965; 43: 9-12. 
3. Ludwig E. Uber das endokrine Substrat der diffusen 
weiblichen (androgenetischen) Alopecie. Arch Klin Exp 
Derm 1966; 227: 468- 77. 
4. De Villez RL, Dunn l. Female androgenetic alopecia. 
The 3 o; 17 [3-androstanediol glucuronide/sex hormone 
binding globulin ratio as a possible marker Jor Jemale 
pattern baldness. Arch Dermatol 1986; 122: 1011-6. 
5. Schweikert UH, Wilson DI. Regulation of human 
hair growth by steroid hormones. I Testosterone metabo-
lism in isolated hairs. J Clin Endocrinol Metah 1974; 
38: 811-9. 
6. Schweikert UH, Wilson DJ. Regulation of human 
hair growth by steroid hormones. II Androstenedione 
metabolism in isolated hairs. J Clin Endocrinol Metah 
1974; 39: 1012-9. 
7. Schweikert UH, Milewich, Wilson DJ. Amortization 
of androstenedione by isolated human hairs. J Clin 
Endocrinol Metah 1975; 40: 413-7. 
8. Schweikert UH, Milewich, Wilson DJ. Amortization 
oJ androstenedione by cultured human hairs fibroblasts. 
J Clin Endocrinol Metah 1976; 43: 785-95. 
9. Farthing GJM, Mattei MA, Edwards WRC, Dawson 
MA. The relationship between plasma testosterone and 
dihydrotestosterone concentrations and male facial hair 
growth. Br J Dermatol 1982; 107: 559-64. 
1 O. Schweikert UH, Wilson DJ. Androgen metabolism 
in isolated human hair roots. In: Hair research: status 
and future aspects, eds: O,fanos EC, Montagna W, 
Stuttgen G. Springer, Berlin, Heidelberg, New - York, 
1981, pp. 210-24. 
98 
11. Ebling FJ. Hormona! control oJ hair growth. 
ibidem, pp. 195-204. 
12. Bassas E. Genetic and androgens in male pattern 
alopecia. Physiopathologic basis of hair grafi. ibidem, 
pp. 686-90. 
13. Dawber RPR. Disorders of hair and nails in: 
Scientific basis of dermatology - a psychological approach. 
Eds. Thody J, Friedman SP, Churchill Livingstone, 
Edinburgh, 1986, pp. 330-48. 
14. Strauss Sl. Hormones and the pilosebaceous appa-
ratus. In: Hair research: status and Juture aspects, eds: 
O,fanos EC et al, Springer, Berlin, 1981, pp. 223-8. 
15. Adachi K, Kano M. A denyl-cyclase in human hair 
follicle: lts inhibition by dihydrotestosterone Biochem 
Biophys Res Commun 1970; 41: 884-90. 
16. Sansone-Bazzano G, Reisner MR, Bazzano G. Con-
version of testosterone - 1,23H to androstenedione3H in 
the isolated hair Jollicle in men. J Clin Endocrinol 
Metah 1972; 34: 512-5. 
17. Rook A . Endocrinal influence oJ hair growth. Br 
J Dermatol 1965; 77: 609-14. 
18. Ludwig E. Classification oJ the types of androgenetic 
alopecia (common baldness) occurring in the Jemale 
sex. Br J Dermatol 1977; 94: 247-54. 
19. Bardin CW, Lipset MB. Testosterone and andro-
stenedione blood productions rates in normal women 
with idiopathic hirsutism or polycystic ovaries. J Clin 
lnvest 1967; 46: 891-901. 
20. Price VH. Testosterone metabolism in the skin. 
Arch Dermatol 1975; 111: 1496-502. 
21. RosenJield RL, Ehrlich EN, Cla,y RE. Adrenal 
and ovarian contributions for the elevated free plasma 
androgen leve! in hirsute women. J Clin Endocrinol 
Metah 1972; 34: 92-8. 
22. Zaun H. Untersuchungen uber den Einfluss 
acta dennatovenerologica A.P.A. Vol 6, 97, No 3 
Androgen honnone leve/s in women with male pattern baldness 
antikonzeptiver Zweiphasen - hormonpriiparate auf das 
Wachstum der Kopfhaare. Arch Klin Exp Derm 1970; 
238: 197-206. 
23. Cormia F. Alopecia from oral contraceptives. JAMA 
1967; 201: 635-7. 
24. Horton R, Neiseler J. Plasma androgens in patients 
with polycystic ovarian disease. J Ciin Endocrinol 
Metah 1968; 28: 479-83. 
25. Ivanovic S. Ispitivanje androsterona i etioholanolona 
kod benignih i malignih tumora dojke. Doktorska 
disertacija, Beograd 1977 
26. Baird D, Horton R, Langcope JF. Steroid dynamics 
under steady stale conditions. Recent Prog Horm Res 
1969; 25: 611-29. 
27. Padridge WM. Transport of protein bound hormones 
into tissues in vitro. Endocrinol Rev 1981; 2: 103-6. 
28. Cipriani R, Puzza G, Foresta C, Veller Fomasa C, 
Peserico A. Sex hormone-binding globulin and saliva 
testosterone leve/s in men with androgenetic alopecia. 
Br J Dermatol 1983, 109: 249-52. 
29. Mortimer CH, Rushton H, James. Effective medica! 
treatment of common baldness in women. Ciin Exp 
Dermatol 1984; 914: 342-50. 
AUTHOR'S ADDRESSES 
Martinovic Nevenka, MD PhD, assistant professor of dermatology, Institute for skin and veneral diseases, 
Pasterova 2, 11000 Beograd, Yugoslavia 
Micic Jovan, MD PhD, professor of endocrinology, Institute of endocrinology, 
Doktora Subotica 16, Beograd, Yugoslavia 
Ivanovic Slavica, MD PhD, endocrinologist, Institute of oncology, Pasterova 6, Beograd, Yugoslavia 
Krunic Aleksandar MD, PhD, assistant professor of dermatology, Institute for skin and veneral diseases, 
Pasterova 2, Beograd, Yugoslavia 
acta dennatovenerologica A.P.A. Vol 6, 97, No 3 99