Journal of the Royal Society of Medicine Volume 79 August 1986 451
Spironolactone in the treatment of idiopathic hirsutism
and the polycystic ovary syndrome
D J Evans BSc MRCP C W Burke DM FRCP Department of Endocrinology, Radcliffe Infirmary, Oxford
Keywords: spironolactone, hirsutism, polycystic ovary syndrome
Summary LH levels and were classed as IH. Twenty-two sub-
Forty-eight hirsute women were treated with spiro- jects had oligomenorrhoea (cycle length >35 days)
nolactone 100mg twice daily for 3 to 12 months. and 4 had amenorrhoea of >3 months' duration.
Both facial and body hirsuties improved by 30-40%, All patients were advised of- the need for effective
and there was a threefold reduction in frequency of contraception because'of the possible effects of
local treatments such as waxing or shaving. Plasma antiandrogen treatment in feminizing the male
testosterone fell by 30%, though the improvement fetus1.
in hirsuties grading did not correlate with the fall In those women using local treatments to control
in plasma testosterone. Six subjects discontinued their hirsuties, the frequency of shaving, plucking,
treatment because of lack of effect, and 4 because of waxing or use of depilatory creams was recorded. In
menstrual disturbance. Spironolactone was equally those not using such methods, facial hirsuties was
effective in the treatment of idiopathic hirsutism graded on an arbitary scale of 1 to 5 based on the
and of the polycystic ovary syndrome. distribution and density of the hair growth over the
face (upper lip, chin and sideburns) and body (arms,
Introduction legs, chest, abdomen and back). Plasma testosterone
Hirsuties can be a distressing symptom in women was measured by radioimmunoassay.
and is often difficult to control, as reflected in the The results of treatment were assessed at 3 months
wide variety of drugs which have been used includ- and, in 16 subjects, again at 12 months. The studies
ing corticosteroids, oral contraceptives and the anti- were carried out with the approval of the hospital
androgens cyproterone acetate and cimetidine, all of ethical committe'e and informed consent of each sub-
which are often relatively ineffective or have un- ject. Results were analysed using Student's paired
desirable side effects'. The aldosterone antagonist and unpaired t tests and linear regression analysis
spironolactone also has an antiandrogenic action, for normally distributed data (welght, age, plasma
first seen during its use in the treatment of hyperten- testosterone) and the Wilcoxon matched pairs signed
sion as the unwanted side effects of gynaecomastia, ranks test, Mann-Whitney U test and Spearman
loss of libido and impotence in men and menstrual rank correlation otherwise (hirsuties gradings and
irregularity in women2. In a hypertensive obese treatment frequency).
woman who was also hirsute, a marked improvement
in the degree of hirsuties was seen following the Results
treatment of her hypertension with spironolactone3. Of 48 hirsute women treated with spironolactone, 42
This suggested its use in the treatment of hirsutism4'5. reported moderate to marked improvement in their
In the present study we have assessed the results subjective assessment of the degree of hirsuties.
of treatment with spironolactone of women with Facial hirsuties improved by 1.2 +0.2 grades (mean-
idiopathic hirsutism (IH) and the polycystic ovary + s.e.mean), a 39% reduction, while body hirsuties
syndrome (PCOS). diminished by 0.9±0.1 grades, a 29% reduction, with
all parts of the body being affected to a similar
Patients and methods extent (Table 2). The treatment frequency of the face
Forty-eight premenopausal women referred for the and legs fell by 3.4 and 2.7 fold respectively. Plasma
investigation and treatment ofhirsutism were treated testosterone fell by 0.9±0.2nmol/1 (P8 u/1 other initial plasma testosterone (r= 0.79, P<0.001). There
than at mid-cycle) with ultrasound evidence in 11 was no further decrease in plasma testosterone or in
and a history of infertility in 9), while 24 had normal body weight after 3 months. The improvement in
both facial and body hirsuties, however, was slightly
greater at 12 months than at 3 months (facial
Table 1. Patient characteristics 1.4 ±0.4 versus 0.8 ±0.2 grades, P= NS; body 1.0+0.2
versus 0.7 +0.2 grades, P<0.05). 0141-0768/86/
Mean Range When subjects with PCOS were compared to 080451-03/$02.00/0
those with IH, there were no significant differences @ 1986
Age (yr) 28.8 16-45 in the degree of facial or body hirsuties or in age,
Weight (kg) 76.9 52-164 The Royal
Body mass index (kg/m2) 28.8 19.6-51.2 weight or body mass index. Plasma testosterone, Society of
however, was significantly greater in the PCOS Medicine
452 Journal of the Royal Society of Medicine Volume 79 August 1986
Table 2. Effects of spironolactone treatment on hirsutism and plasma testosterone (mean ±s.e. mean)
Pretreatment Post-treatment n P
Hirsuties (grades): Facial 3.1+0.2 1.9+0.2 24 <0.001
Body 2.9+0.1 2.1+0.1 41 <0.001
Chest 3.0+0.3 1.9±0.2 16 < 0.001
Back 2.6+0.3 1.8+0.3 16 < 0.001
Abdomen 3.3 +0.2 2.3 +0.2 31 < 0.001
Legs 3.1+0.3 2.1+0.2 26 <0.001
Arms 2.4+0.2 1.7+0.2 25 <0.001
Treatment interval (days): Face 3.3 +0.6 11.4+2.2 25 < 0.001
Legs 4.9+0.6 13.5+2.1 19 < 0.001
Plasma testosterone (nmol/l) * 3.0+0.2 2.1+0.1 31 <0.001
0 Normal range for plasma testosterone in females 0.5-2.1 nmol/l.
group (3.4 + 0.3 nmol/l) than in IH (2.1 ± 0.1 nmol/l, The comparable improvement in hirsuties in
P <0.001) and there was a correspondingly greater PCOS and IH despite the greater fall in plasma
fall in plasma testosterone following spironolactone testosterone in PCOS, the lack of correlation
therapy (PCOS 1.2 + 0.2, IH 0.6 ± 0.2 nmol/l, P< 0.05). between the fall in testosterone and improvement
By contrast, there were no significant differences in in hirsuties, and the continued improvement in hir-
the decrease in facial hirsuties (PCOS 1.2±0.2, IH suties with no further fall in plasma testosterone
1.2 +0.4 grades) or fall in body hirsuties (PCOS with increasing duration of treatment, suggest that
0.8 + 0.2, IH 0.9 + 0.1 grades) on treatment. a peripheral action of spironolactone is as important
Three of the 22 women whose menses were as its action in reducing plasma testosterone levels.
initially regular developed polymenorrhoea, whilst Spironolactone is known to bind to androgen recep-
menses were unaltered in the remainder. Of the 22 tors in the skin as well as other peripheral tis-
women initially oligomennorhoeic, 2 became amenor- sues2"14'15, inhibiting androgen action peripherally,
rhoeic, menses were unaltered in 6 and became as well as reducing the testosterone production rate
regular in 10, while 4 developed polymenorrhoea or in both hirsute and non-hirsute women4; this is pre-
menorrhagia. Two of the 4 women who initially had sumably the result of its inhibition of cytochrome
amenorrhoea resumed menses, though remaining oli- P450 oxidase and hence 17-hydroxylase action in
gomenorrhoeic. No electrolyte imbalance occurred both gonads and adrenals2"16.
during treatment, and no other adverse effects apart Regardless of the mechanisms responsible, spiro-
from transient nausea and mild diuresis were nolactone proved an effective agent with few
reported. Ten patients (21%) (included in the analy- unwanted effects in the treatment of hirsutism in
sis) discontinued treatment after 2-9 months, 6 both PCOS and IH. Further studies are needed to
because of lack of effect and 4 because of the adverse determine whether a lower dose of spironolactone,
effect on menses. which might result in fewer menstrual irregularities,
may suffice to maintain the improvement in hirsuties
Discussion obtained on the higher dose.
Spironolactone in the dosage used in the present Acknowledgments: We thank the Nuffield Department of
study proved an effective treatment for hirsuties in Clinical Biochemistry, Radcliffe Infirmary for the measure-
the majority of patients, in both PCOS and IH. All ment of plasma testosterone.
regions of the body benefited to a similar extent.
Previous studies of smaller numbers of patients4 -12 References
have yielded conflicting views as to its efficacy; this 1 Coney PJ. Polcystic ovarian disease: current concepts
can probably be attributed to differences in the of pathophysiology and therapy. Fertil Steril
dose of spironolactone employed, those using 200 mg 1984;42:667-82
daily reporting similar responses to those in the 2 Loriaux DL, Menard R, Taylor A, Pita JC, Santen R.
present study and those using lower doses reporting Spironolactone and endocrine dysfunction. Ann
lower success rates. Most of the improvement in Intern Med 1976;85:630-6
3 Ober KP, Hennessy JF. Spironolactone therapy for
hirsuties was already manifest at 3 months, though hirsutism in a hyperandrogenic woman. Ann Intern
there was a further small improvement by 12 months. Med 1978;89:643-4
The effects may be delayed if a cyclical regimen of 4 Boisselle A, Tremblay RR. New therapeutic approach
spironolactone administration is employed5. There. to the hirsute patient. Fertil Steril 1979;32:276-9
were few adverse effects other than polymenorrhoea, 5 Shapiro G, Evron S. A novel use of spironolactone:
menorrhagia or oligo-amenorrhoea in 9 women, of treatment of hirsutism. J Clin Endocrinol Metab
sufficient severity to result in cessation of treatment 1980;51:429-32
in only 4. Menstrual regularity improved in 12 women 6 Cumming DC, Yang JC, Rebar RW, Yen SSC. Treat-
whose periods were initially irregular. Mild polyuria ment of hirsutism with spironolactone. JAMA
was the only other reported side effect and, unlike its 1982;247:1295-8
use in subjects with renal, cardiac or hepatic 7 Milewicz A, Silber D, Kirschner MA. Therapeutic
effects of spironolactone in polycystic ovary syn-
disease'3, there were no electrolyte disturbances. drome. Obstet Gynecol 1983;61:429-32
The fall in weight accompanying spironolactone 8 Messina M, Manieri C, Biffignandi P, Massucchetti C,
therapy was probably largely related to its diuretic Novi RF, Molinatti GM. Antiandrogenic properties of
action since no further weight reduction was seen spironolactone. Clinical trial in the management of
after the first 3 months. female hirsutism. J Endocrinol Invest 1983;6:23-7
Journal of the Royal Society of Medicine Volume 79 August 1986 453
9 Abramovici Y, Guillet F, Matte R, Chaieb L, Belanger 13 Greenblatt DJ, Koch-Weser J. Adverse reactions to
R. Spironolactone in idiopathic hirsutism. Clinical spironolactone: a report from the Boston Collabora-
and biological evaluations of treatment. Annales tive Drug Surveillance Program. JAMA 1973;225:40-3
d'Endocrinologie (Paris) 1983;44:393-6 14 Boiselle A, Dione FT, Tremblay RR. Interaction of
10 Spandri P, Gangemi M, Nardelli GB, et al. Testoster- spironolactone with rat skin androgen receptor. Can J
one, 17 KS, 17B2, FSH-LH variations and hirsutism Biochem 1979;57:1042-6
modifications during spironolactone therapy. Clin Exp 15 Eil C, Ekelson SK. The use of human skin fibroblasts
Obst Gyn 1984;11:49-54 to obtain potency estimates of drug binding to andro-
11 Lobo RA, Shoupe D, Serafini P, Brinton D, Horton R. gen receptors. J Clin Endocrinol Metab 1984;59:51-5
The effects of two doses of spironolactone on serum 16 Menard RH, Stripp B, Gillette JR. Spironolactone
androgens and anagen hair in hirsute women. Fertil and testicular cytochrome P-450: decreased testoster-
Steril 1985;43:200-5 one formation and in several species and changes in
12 Dorrington-Ward P, McCartney ACE, Holland S, et al. hepatic drug metabolism. Endocrinology 1974;94:
The effect of spironolactone on hirsutism and female 1628-36
androgen metabolism. Clin Endocrinol 1985;23:161-7 (Accepted 10 February 1986)