You are on page 1of 8

Clinical Endocrinology (2000) 52, 587±594

Spironolactone as a single agent for long-term therapy


of hirsute patients*

Poli Mara Spritzer, Karen O. Lisboa, Simone different between the groups after 6 and 12 months.
Mattiello and Francisco Lhullier LH levels decreased with CPA (72%) but not with
Gynecological Endocrinology Unit, Division of spironolactone.
Endocrinology, Hospital de ClõÂnicas de Porto Alegre; and CONCLUSION Our results suggest that spironolac-
Department of Physiology, Universidade Federal do Rio tone used as a single agent is as effective as cypro-
Grande do Sul, Brazil terone acetate combined with oestradiol for long-term
(Received 5 October 1999; returned for revision 16 November
treatment of patients with idiopathic hirsutism. In
1999; ®nally revised 22 December 1999; accepted 10 January PCOS patients, spironolactone is still effective for
2000) reducing hirsutism; however, for treatment of the
hormonal or metabolic manifestations associated
with PCOS, it may be necessary to combine spirono-
Summary lactone with either an antigonadotrophic agent or a
drug that improves peripheral insulin sensitivity.
OBJECTIVE To assess the androgen-suppressing
effect of spironolactone, and the use of this drug as
a single agent in the long-term therapy of hirsute
patients with either polycystic ovary syndrome Hirsutism is de®ned as excessive hair growth, especially in
(PCOS) or idiopathic hirsutism (IH). Standard cypro- women, in speci®c androgen-sensitive areas of the body.
terone acetate (CPA) treatment was used to evaluate Hirsutism can result from an overproduction of androgens by
the results obtained with spironolactone. the ovaries or adrenal glands, by increased sensitivity of the
DESIGN Prospective randomized study. pilosebaceous unit to circulating androgens, or by a combina-
PATIENTS Forty-six hirsute women were separated tion of both (Bardin & Lipsett, 1967; Kuttenn et al., 1977;
randomly into two groups, strati®ed for polycystic Rittmaster & Thompson, 1990).
ovary syndrome. For 12 months, Group 1 (21 patients, The two anti-androgens most commonly used for the
10 PCOS) received spironolactone only (200 mg/day). treatment of hirsutism are spironolactone and cyproterone
Group 2 (23 patients, nine PCOS) received CPA acetate (CPA). Although the general therapeutic effectiveness
(50 mg/day) with ethinyl oestradiol (35 mg/day). of these two anti-androgens has been con®rmed, extended
MEASUREMENTS Ferriman±Gallwey clinical score clinical trials conducted with either drug have produced
for hirsutism and serum testosterone, androstene- variable results (Cumming et al., 1982; Suraci et al., 1983;
dione, and LH levels. Thomas et al., 1985; Couzinet et al., 1986; O'Brien et al., 1991;
RESULTS In IH patients, hirsutism regressed equally
Cusan et al., 1994; Wong et al., 1995; Erenus et al., 1996). In
with spironolactone (21 6 2±14´5 6 2) and CPA addition, the anti-androgen drug is frequently chosen with no
(23 6 2±13 6 2). In PCOS patients, the mean score for concern for the aetiology of hirsutism. However, different
hirsutism after 12 months was signi®cantly lower with measures are required to treat idiopathic hirsutism (IH) and
CPA (12 6 1) than with spironolactone (16 6 1). Tes- hirsutism related to polycystic ovary syndrome (PCOS).
tosterone levels did not change with spironolactone; In IH, hirsutism is an isolated manifestation, and therefore an
with CPA there was a decrease from baseline in PCOS anti-androgen agent alone may be suf®cient to treat this
(47% and 51%, 6 and 12 months) and IH patients (31% condition. In turn, PCOS-related hirsutism is associated with
and 30%). Androstenedione levels also declined from anovulation and with other endocrinological manifestations,
baseline in CPA-treated PCOS patients (38% and 39%, such as increased ovarian androgen production, hyperinsuli-
6 and 12 months). Androgen levels were signi®cantly naemia and/or insulin resistance. Therefore, in PCOS, an anti-
androgen agent may inhibit hirsutism, but an antigonadotrophic
Correspondence: Poli Mara Spritzer MD, PhD, Departamento de agent is required to treat the associated manifestations.
Fisiologia, Universidade Federal do Rio Grande do Sul, Rua Furthermore, in the presence of obesity and/or metabolic
Sarmento Leite, 500, 90050±170 Porto Alegre, RS, Brazil. Fax: ‡ 55
51 3163166; E-mail: spritzer@vortex.ufrgs.br
disturbances, speci®c measures could be required for PCOS
*Presented in part as a poster at the 79th Annual Meeting of the patients, such as diet and physical exercise as well as the use of
Endocrine Society, Minneapolis, USA, June 1997. drugs that ameliorate peripheral insulin sensitivity.

q 2000 Blackwell Science Ltd 587


588 P. M. Spritzer et al.

Table 1 Baseline clinical and hormonal data for polycystic ovary syndrome (PCOS) patients, idiopathic hirsutism (IH) patients, and normal women²

Age Clinical score Body mass Testosterone Androstenedione


Group (years) for hirsutism index (kg/m2) (nmol/l) (nmol/l) LH (IU/l)

PCOS (n ˆ 19) 22 6 9 22 6 4* 24 6 5 3´02 6 1´73* 11´59 6 5´54 7´25 6 2´87*


IH (n ˆ 25) 25 6 8 21 6 6* 23 6 4 2´29 6 1´25 9´98 6 3´35 3´01 6 0´45
Normal women³ 28 6 8 562 23 6 5 1´87 6 1´17 6´42 6 1´32 3´85 6 1´86
(n ˆ 13)

² Values are expressed as mean 6 SD. * P < 0´05 vs. normal women. ³ Reference values refer to women with regular menstrual and ovulatory cycles
and without hirsutism, from the same geographical region and with socio-economic characteristics similar to the study group.

Although it has been suggested that spironolactone reduces The diagnosis of PCOS was based on the physical features of
androgen secretion (Shapiro & Evron, 1980) the existence of hyperandrogenism, disturbed menstrual cycles, elevated serum
such an effect is disputable, especially because this drug is LH levels or LH/FSH ratio, increased levels of serum
usually employed in association with contraceptives, which testosterone and/or androstenedione and no evidence of ovarian
inhibit the production of androgens. The present study was or adrenal neoplasm or Cushing's syndrome. The diagnosis of
therefore designed to assess the existence of an androgen- PCOS followed NIH consensus criteria (Zawadski & Dunaif,
suppressing effect in spironolactone when this drug is used as a 1992) and although transabdominal or transvaginal ovarian
single agent, and to assess the ef®cacy of spironolactone as a ultrasound was performed on most of the patients and enlarged,
single drug for treatment of hirsutism with different aetiologies, cystic ovaries were detected, ultrasound was not required for
namely IH and PCOS. Standard CPA treatment was used to the diagnosis of PCOS (Adams et al., 1986; Rittmaster &
monitor the results obtained with spironolactone. Thompson, 1990; Farquhar et al., 1994; Herter et al., 1996).
Idiopathic hirsutism was considered in hirsute patients with
regular, ovulatory cycles (luteal phase progesterone levels
Patients and methods higher than 12 nmol/l), normal androgen levels, and without
any known underlying disease. Table 1 shows the baseline
Patients
clinical features and hormone levels of the patients with PCOS
The study population included women treated consecutively or IH, and a comparison with reference features and hormone
during a 2-year period for hirsutism at the Gynecological levels in a group of 13 normal women. The study protocol was
Endocrinology Unit at Hospital de ClõÂnicas de Porto Alegre, approved by the local Ethics Committee and informed consent
Brazil. Late-onset (non-classic) congenital adrenal hyperplasia was obtained from every subject.
patients were excluded on the basis of a high plasma level of
17-hydroxyprogesterone (> 15´1 nmol/l) and/or its marked
Protocol
increase after ACTH stimulation (> 30 nmol/l) (Spritzer et al.,
1990; Aziz et al., 1994). Patients with hyperprolactinaemia Patients were observed for 1 year. After the diagnosis of the
(serum prolactin levels higher than 600 m U/l (20 mg/l) on two underlying aetiology, subjects were randomly separated into
different occasions) were also excluded. two treatment groups and strati®ed for the presence of PCOS.
Forty-six hirsute women were enrolled in the study. Two This was performed for two reasons: (1) it was part of the study
women were lost to follow-up; thus, 44 women, whose ages rationale to assess the effects of spironolactone on patients with
ranged from 14 to 42 years, participated in the study. None had different aetiologies of hirsutism; and (2) PCOS was thought to
received any drugs known to interfere with hormonal levels for in¯uence the patient's ability to respond to treatment. There-
at least 3 months before the study. Patients did not plan to fore, we used sequential treatment assignment to ensure that
become pregnant and agreed to use mechanical methods of treatment balance was achieved (Pocock & Simon, 1975).
contraception. The mean score for hirsutism, assigned by the Group 1 (21 patients, including 10 subjects with PCOS)
original method of Ferriman and Gallwey (F±G) (1961), was received spironolactone at a dose of 200 mg/day (Aldactone;
21 6 5 (range 11±35). Nineteen patients were diagnosed as Searle-Biolab, SaÄo Paulo, Brazil), 20 days per month for
having polycystic ovary syndrome (PCOS) and 25 as having 12 months. Group 2 (23 patients, including nine subjects with
idiopathic hirsutism (IH). Thirteen were overweight or obese, PCOS) received a CPA regimen modi®ed from Kuttenn et al.
with a body mass index greater than 25 kg/m2 (nine in the (1980): 50 mg/day of CPA (Androcur; Schering-Berlimed, Sao
PCOS group and four in the IH group. Paulo, Brazil), 20 days per month for 12 months, in addition to

q 2000 Blackwell Science Ltd, Clinical Endocrinology, 52, 587±594


Spironolactone in hirsute patients 589

ethinyl oestradiol (35 mg/day) over the last 10 days of CPA 12 months). All samples were obtained between 0800 and
(Spritzer et al., 1998). CPA treatment was used to monitor the 1000 h. Testosterone and androstenedione measurements were
results obtained with spironolactone. Both the antigonado- obtained by commercial radioimmunoassay (Diagnostic Pro-
trophic and the anti-androgenic effects of CPA are maintained ducts Corp., Los Angeles, CA, USA) with 6´5% and 10´3%
with this regimen (even if CPA is used with percutaneous 17-b intra- and interassay coef®cients of variation, respectively, for
oestradiol, which does not have an antigonadotrophic action) testosterone, and with 4´2% and 7´8% intra- and interassay
(Kuttenn et al., 1980), and the addition of ethinyl-oestradiol coef®cients of variation, respectively, for androstenedione. LH
ensures the maintenance of a regular menstrual cycle. was measured by an enzyme-labelled immunoabsorbent assay
For both treatment groups, medication was begun on the ®fth (Del®a, Pharmacia, Columbia, MD, USA) with a 4´5% and 11%
day of the menstrual cycle in patients with regular cycles; in intra- and interassay coef®cient of variation, respectively.
patients with abnormal cycles, medication was begun arbitra-
rily or after the best clinical estimation of the early follicular
Statistical analysis
phase. In the doses employed, both anti-androgen agents have
been shown to be effective for the treatment of hirsutism, and Results are presented as means 6 SEM, unless otherwise noted.
well tolerated (Kuttenn et al., 1980; Cumming et al., 1982; Comparisons between group means were analysed by Student's
Lobo et al., 1985; Barth et al., 1989; Spritzer et al., 1990). t-test. Two-way analysis of variance with repeated measures
To assess the therapeutic effects of each agent, hirsutism was followed by the Student±Newman±Keuls test was performed
graded independently by two observers (K.O.L. and S.M.) for comparing hirsutism mean scores and hormonal data within
using the original Ferriman and Gallwey clinical score (total each treatment group. Data were considered to be signi®cant at
maximum score reported, 44) (1961). The F±G score was P < 0´05.
determined every 3 months. Inter-observer reliability was
calculated by dual measurements of patients during the ®rst
Results
set of consultations. Pearson's correlation coef®cient for the
total score was 0´81 (P < 0´05). The mean difference between The clinical score for hirsutism and the hormonal data for
the scores for different measurements performed by the same Groups 1 and 2 were compared for PCOS and IH patients before
observer was 1´5. This value was not statistically different from treatment (Table 2). No signi®cant differences were observed in
0. Serum testosterone, androstenedione and LH levels were these parameters.
checked every 6 months. Signi®cant reductions in the clinical score for hirsutism were
observed with both anti-androgens. In patients with IH,
hirsutism regressed by the same magnitude with spironolactone
Hormone assays
(from 21 6 2 to 14 6 2) and CPA (from 23 6 2 to 13 6 2) (Fig.
Before starting therapy, blood samples for testosterone, 1a). In contrast, patients with PCOS treated with CPA showed a
androstenedione and LH assays were collected between days signi®cantly lower mean score for hirsutism after 12 months
2 and 10 of the menstrual cycle or on any day when the patients when compared to spironolactone-treated patients (12 6 1 for
were amenorrhoeic. During treatment, when patients were CPA and to 16 6 1 for spironolactone, P ˆ 0´009, Fig. 1a).
receiving only spironolactone or CPA, blood samples were Figures 1b and 1c show the changes in serum androgen levels
collected on days 6±9 of anti-androgen therapy (after 6 and after 6 and 12 months of treatment. Testosterone levels did not

Table 2 Clinical score for hirsutism and hormonal data before treatment in the two treatment groups (spironolactone and CPA)*

Clinical score Testosterone Androstenedione


Groups for hirsutism (nmol/l) (nmol/l) LH (IU/l)

Polycystic ovary syndrome


Spironolactone 22 6 2 3´05 6 0´45 11´62 6 2´16 4´34 6 0´86
CPA 21 6 1 2´94 6 0´38 11´41 6 1´39 7´28 6 0´94
Idiopathic hirsutism
Spironolactone 21 6 2 2´35 6 0´41 8´65 6 0´76 2´88 6 0´62
CPA 23 6 2 2´25 6 0´31 10´40 6 0´94 2´90 6 0´58

*Values are mean 6 SEM. No statistical differences were observed in the comparison of treatment type in PCOS or IH patients.

q 2000 Blackwell Science Ltd, Clinical Endocrinology, 52, 587±594


590 P. M. Spritzer et al.

(a)
30

25
Hirsutism score

20 a
aa
15 *a
10

0
0 3 6 12 0 3 6 12
(b)

4
Testosterone (nmol/l)

a a
2 *a *a
** **

1
Fig. 1 A Spironolactone; B CPA. (a)
Clinical scores for hirsutism with
0
0 6 12 0 6 12 spironolactone and CPA. *P < 0´05 vs.
spironolactone at 12 months (Student's t-test);
(c) a
P < 0´05 from baseline values (two-way
analysis of variance with repeated measures
16
Androstenedione (nmol/l)

and Student±Newman±Keuls test); (b)


14 testosterone levels with spironolactone and
12 CPA treatments. *P < 0´005 vs.
spironolactone at 6 months; **P < 0´05 vs.
10
*a *a spironolactone at 12 months (Student's t-test);
8 a
P < 0´05 from baseline values (two-way
6 analysis of variance with repeated measures
and Student±Newman±Keuls test); (c)
4
androstenedione levels with spironolactone
2 and CPA treatments. *P < 0´05 vs.
0 spironolactone at 6 and 12 months (Student's
0 6 12 0 6 12 t-test); aP < 0´05 from baseline values (two-
PCOS IH way analysis of variance with repeated
Time (months) measures and Student±Newman±Keuls test).

change in patients treated with spironolactone; however, 12 months) (Fig. 1c). The difference in androstenedione levels
patients treated with CPA had a signi®cant decrease from between patients with PCOS treated with spironolactone or
baseline levels (PCOS: 47% and 51%, after 6 and 12 months, CPA was also signi®cant after 6 (P ˆ 0´04) and 12 months
respectively; IH: 31% and 30%) (Fig. 1b). For hirsute women (P ˆ 0´019) (Fig. 1c).
with PCOS, the difference in testosterone levels with Table 3 shows LH serum levels before and during therapy
spironolactone and CPA treatment was signi®cant after 6 with spironolactone or CPA in hirsute patients with PCOS or
(P ˆ 0´001) and 12 months (P ˆ 0´044) (Fig. 1b). Androstene- IH. Regarding basal serum LH values, no statistical differences
dione levels were unchanged in patients treated with spirono- were observed between the treatment group types in PCOS or
lactone (Fig. 1c) but declined signi®cantly from baseline in IH patients. In patients with PCOS who received spironolac-
PCOS patients treated with CPA (38% and 39%, for 6 and tone, LH levels did not change during the period of treatment.

q 2000 Blackwell Science Ltd, Clinical Endocrinology, 52, 587±594


Spironolactone in hirsute patients 591

Table 3 Serum LH levels before and after 6 and 12 months of therapy with spironolactone or CPA

LH (IU/l)

Groups 0 6 months 12 months

Polycystic ovary syndrome


Spironolactone 4´34 6 0´86 4´30 6 0´73 6´03 6 0´77
CPA 7´28 6 0´94 1´77 6 0´72*a 2´01 6 0´75*a
Idiopathic hirsutism
Spironolactone 2´88 6 0´62 4´10 6 1´00 3´64 6 1´07
CPA 2´90 6 0´58 2´33 6 0´304 1´90 6 0´37

Values are mean 6 SEM. *P < 0´05 vs. spironolactone at 6 and 12 months (Student's t-test). aP < 0´05 from baseline values (two-way analysis of
variance with repeated measures and Student±Newman±Keuls test).

In contrast, there was a striking decrease in LH serum levels F±G method has been compared with other objective methods
among the CPA-treated patients with PCOS (76% and 72% of hair growth measurement, and assessed critically with
from baseline, respectively, after 6 and 12 months). The respect to reproducibility or sensitivity for detection of changes
difference in LH levels between spironolactone- and CPA- elicited with different types of treatment (Barth et al., 1991;
treated patients with PCOS was signi®cant after 6 (P ˆ 0´042) Heiner et al., 1995).
and 12 months (P ˆ 0´011). No signi®cant changes were In the present study, IH and PCOS patients in the two
observed in LH levels among patients with IH. treatment groups (spironolactone and CPA) had a very similar
Nine patients (42%) in Group 1 (four women with PCOS and clinical and hormonal pro®le prior to anti-androgen therapy.
®ve women with IH) receiving spironolactone developed During treatment, spironolactone and CPA were similarly
intermenstrual bleeding, especially during the ®rst 3 months effective in reducing hair growth in ovulatory patients
of treatment. After 12 months of treatment, only three patients presenting with hirsutism. Among IH women, who had
(with IH) maintained an intermenstrual bleeding pattern. Four normal baseline androgen levels, CPA-treated patients had a
patients (17%) in Group 2 (two women with PCOS and two signi®cant decrease in testosterone levels, whereas spironolac-
with IH) receiving CPA developed oligomenorrhoea. Therapy tone had no signi®cant effect.
did not have to be interrupted because of these side-effects. In hirsute patients with PCOS, CPA yielded signi®cantly
better clinical results than spironolactone after a year of
treatment. Furthermore, while no changes were observed in
Discussion
androgen levels in patients with PCOS treated with spirono-
CPA and spironolactone have both been shown to be effective in lactone, a consistent reduction of testosterone and androstene-
the treatment of hirsutism, but there have been no comparative, dione was observed in those treated with CPA. The reduction in
randomized studies taking into consideration the aetiology of androgen levels was coincident with the decrease in LH levels,
hirsutism. This study was designed to assess the ef®cacy of corresponding to the suppression of ovarian androgen secretion
spironolactone as a single agent for the treatment of women with through the antigonadotrophic effect of CPA. The fact that
either PCOS or idiopathic hirsutism, using a standard CPA spironolactone has no effects on LH levels suggests that
treatment to monitor the results obtained with spironolactone. spironolactone does not have a central effect on gonadotrophin
The F±G score is frequently used as a semiquantitative secretion. The absence of such a central effect is in agreement
measurement of hirsutism (Ferriman & Gallwey, 1961). In the with the lack of change in androgen levels observed among
present study, inter-observer reliability was high, with a spironolactone-treated PCOS women in this study.
signi®cant correlation of 0´81. In addition, repeatability was Spironolactone is an aldosterone antagonist used originally to
optimal, since the mean difference between the scores for treat hypertension. It is also a weak progestin (Shane & Potts,
measurements performed by the same observer was not 1978) and a weak inhibitor of testosterone biosynthesis (Menard
statistically different from 0. Several other groups have used et al., 1974). The inhibition of androgen action by spironolactone
F±G scores as the only criterion to assess changes in hirsutism is related to its ability to compete in vitro with testosterone and
elicited with different treatments (Suraci et al., 1983; DHT, by binding to the androgen receptor (Eil & Edelson, 1984).
Mowszowicz et al., 1984; Spritzer et al., 1990; Carmina et In the present study, a dose of 200 mg/day of spironolactone
al., 1994; Cusan et al., 1994; Tolino et al., 1996). Moreover, the was used for maximal anti-androgenic action. With this dose,

q 2000 Blackwell Science Ltd, Clinical Endocrinology, 52, 587±594


592 P. M. Spritzer et al.

no changes were observed either in serum potassium or in blood is a well-known anti-androgen (Hammerstein et al. 1975;
pressure levels (data not shown), an observation which is in Mowszowicz et al., 1984; Spritzer et al., 1990). As suggested
agreement with other reports (Cumming et al., 1982). The by the present results, the combination of these two properties is
200 mg/day dose was also used to test whether spironolactone, especially useful for the treatment of hirsutism associated with
employed as a single agent, would affect androgen levels, ovarian androgen overproduction, as in PCOS. Due to its
similarly to what happens when the drug is used in combination antigonadotrophic effect CPA suppresses androgen overpro-
with oral contraceptives (O'Brien et al., 1991; Erenus et al., duction, and the action of any remaining androgen is prevented
1996). However, our results indicate that spironolactone, used by the anti-androgenic action of CPA at the target cell site,
as a single agent at least in the usual dose, either for 20 days as namely the pilosebaceous gland.
in the present study, or for 30 days as in the study by Lobo et al. In this study, ethinyl oestradiol (35 mg/day), added on days
(1985), acts primarily as a peripheral anti-androgen and has no 16±24 of CPA administration, ensured adequate oestrogen
suppressive effect on androgen levels. input. It allowed regular bleeding and maintained target organ
There is some controversy regarding the effects of spirono- eutrophy, otherwise prevented by the use of CPA alone, which
lactone on androgen levels. Total serum levels of testosterone would induce endometrial atrophy. Therefore, the design of the
have been reported to decrease during spironolactone treatment present study does not allow us to rule out an additional effect
(Shapiro & Evron, 1980; Lobo et al., 1985). Lobo et al. (1985) of oestrogen therapy (although this therapy lasts only 10 days)
reported a signi®cant decrease in serum total testosterone with on gonadotrophin secretion. Nevertheless, it is important to
both 100 and 200 mg of spironolactone (for 30 days) with no stress that the antigonadotrophic effect of CPA has been well
changes in sex hormone binding globulin (SHBG) levels, but demonstrated by others (Hammerstein et al., 1975; Neumann et
observed an unexplained signi®cant increase in the unbound al., 1970; Kuttenn et al., 1980; Mowszowicz et al., 1984).
fractions of testosterone. Those authors concluded that the anti- Four of our patients presented with oligomenorrhoea despite
androgenic effects of spironolactone treatment were primarily the addition of oestrogen. However, this situation may be
related to its peripheral effect. It was suggested that the changes solved by changing oestrogen type (for example, ethinyl
in testosterone levels were possibly most re¯ective of increased oestradiol could be replaced by conjugated oestrogens or
metabolic clearance. In turn, Siegberg et al. (1987) demon- nonoral 17-b oestradiol), dose, or oestrogen treatment duration,
strated a tendency to decreased androgen levels with both since the management of PCOS patients (Kuttenn et al., 1980)
placebo and spironolactone treatment. Spironolactone treat- allows ¯exibility according to the convenience and/or tolerance
ment did not, however, cause any signi®cant change in serum of each patient.
androgen levels or SHBG. Similarly, Cumming et al. (1982) did In conclusion, the results of the present study suggest that
not ®nd any changes in SHBG levels. spironolactone and cyproterone acetate are both well tolerated
Previous publications have also reported an inhibition of and similarly effective in reducing hair growth in hirsute
steroidogenesis by spironolactone therapy occurring through an ovulatory patients with normal androgen levels, as in the
effect on the cytochrome P450 system. However, a decrease in inpatients with idiopathic hirsutism. In PCOS patients,
the concentration of the cytochrome was reported only at high spironolactone is still effective for reducing hirsutism; how-
doses (Menard et al., 1974) or under in vitro experimental ever, because these patients may have additional hormonal or
conditions, for example in bovine and human adrenal cortical metabolic alterations it may be necessary to combine
mitochondria (Chiau-Cheng et al., 1976). spironolactone with either an antigonadotrophic agent or a
Menstrual disturbances, particularly polymenorrhoea, may drug that improves peripheral insulin sensitivity. The combina-
be a problem during spironolactone treatment. As suggested by tion of conventional oral contraceptive pills with spironolac-
other researchers (Suraci et al., 1983; Siegberg et al., 1987; tone, not tested in the present study, might be an effective
Cusan et al., 1994), a cyclic spironolactone regimen may alternative for the treatment of hirsute patients who have
contribute to decrease this adverse effect. In the present study a ovarian hyperandrogenism. Further studies should also focus on
cyclical regimen was used, and spironolactone cycles were evaluating the metabolic tolerability of long-term treatment
found to be more regular than pretreatment cycles in PCOS with anti-androgen drugs in PCOS patients, especially in those
patients presenting with menstrual disorders prior to treatment. presenting with hyperinsulinaemia or insulin resistance.
This was not the case for IH patients whose cycles were regular
before treatment, although polymenorrhoea decreased during
Acknowledgements
the treatment period. Nevertheless, in general spotting occurred
more commonly during spironolactone cycles. This study was supported by grants from Financiadora de
CPA, which was ®rst synthesized as a progestational Estudos e Projetos (no. 41960949´00), Conselho Nacional de
compound, possesses a strong antigonadotrophic activity and Desenvolvimento Cientõ®co e TecnoloÂgico (no. 520544/960±0)

q 2000 Blackwell Science Ltd, Clinical Endocrinology, 52, 587±594


Spironolactone in hirsute patients 593

and FundacËaÄo de Amparo aÁ Pesquisa do Rio Grande do Sul (no. M., Young, D. & Judd, H.L. (1995) Comparison of a gonadotropin-
96/1765.7). releasing hormone agonist and a low oral contraceptive given alone
or together in the treatment of hirsutism. Journal of Clinical
Endocrinology and Metabolism, 80, 3412±3418.
References Herter, L.D., Magalhaes, J.A. & Spritzer, P.M. (1996) Relevance of the
determination of ovarian volume in adolescent girls with menstrual
Adams, J., Polson, D.W. & Franks, S. (1986) Prevalence of polycystic disorders. Journal of Clinical Ultrasound, 24, 243±248.
ovaries in women with anovulation and idiopathic hirsutism. British Kuttenn, F., Mowsowicz, I., Shaison, G. & Mauvais-Jarvis, P. (1977)
Medical Journal, 293, 355±359. Androgen production and skin metabolism in hirsutism. Journal of
Aziz, R., Dewailly, D. & Owerbach, D. (1994) Nonclassic adrenal Endocrinology, 75, 83±91.
hyperplasia: current concepts. Journal of Clinical Endocrinology and Kuttenn, F., Rigaud, C., Wright, F. & Mauvais-Jarvis, P. (1980)
Metabolism, 78, 810±815. Treatment of hirsutism by oral cyproterone acetate and percutaneous
Bardin, C.W. & Lipsett, M.B. (1967) Testosterone and androstenedione estradiol. Journal of Clinical Endocrinology and Metabolism, 51,
blood production rates in normal women and women with idiopathic 1107±1111.
hirsutism or polycystic ovaries. Journal of Clinical Investigation, 46, Lobo, R.A., Shoupe, D.S., Sera®ni, P., Brinton, D. & Horton, R. (1985)
891±902. The effects of two doses of spironolactone on serum androgens and
Barth, J.H., Cherry, C.A., Wojnarowska, F. & Dawber, R.P.R. (1989) anagen hair in hirsute women. Fertility and Sterility, 43, 200±205.
Spironolactone is an effective and well tolerated systemic anti- Menard, R.H., Stripp, B. & Gillette, J.R. (1974) Spironolactone and
androgen therapy for hirsute women. Journal of Clinical Endocri- testicular cytochrome p-450: decreased testosterone formation in
nology and Metabolism, 68, 966±970. several species and changes in hepatic drug metabolism. Endocri-
Barth, J.H., Cherry, C.A., Wojnarowska, F. & Dawber, R.P.R. (1991) nology, 94, 1628±1636.
Cyproterone acetate for severe hirsutism: results of a double-blind Mowszowicz, I., Wright, F., Vincens, M., Rigaud, C., Nahoul, K.,
dose-ranging study. Clinical Endocrinology, 35, 5±10. Mavier, P. & Mauvais-Jarvis, P. (1984) Androgen metabolism in
Carmina, E., Janni, A. & Lobo, R.A. (1994) Physiological estrogen hirsute patients treated with cyproterone acetate. Journal of Steroid
replacement may enhance the effectiveness of the gonadotropin- Biochemistry, 20, 757±761.
releasing hormone agonist in the treatment of hirsutism. Journal of Neumann, F., Berswordt-Wallrabe, R., Eiger, W., Steinbeck, H., Hahn,
Clinical Endocrinology and Metabolism, 78, 126±130. J.D. & Kramec, M. (1970) Aspect of androgen-dependent events as
Chiau-Cheng, S., Suzuki, K., Sadee, W. & Harding, B.W. (1976) Effect studied by anti-androgen s. Recent Progress in Hormone Research,
of spironolactone, canrenone and K-canrenoate on cytochrome P-450 26, 337±405.
and 11b- and 18 hydroxylation in bovine and human adrenal cortical O'Brien, R.C., Cooper, M.E., Murray, M.L., Seeman, E., Thomas, A.K.
mitochondria. Endocrinology, 99, 1097±1102. & Jerums, G. (1991) Comparison of sequential cyproterone acetate/
Couzinet, B., Le Strat, N., Brilly, S. & Schaison, G. (1986) Comparative estrogen versus spironolactone/oral contraceptive in the treatment of
effects of cyproterone acetate or a long-acting gonadotropin- hirsutism. Journal of Clinical Endocrinology and Metabolism, 71,
releasing hormone agonist in polycystic ovarian disease. Journal of 1008±1013.
Clinical Endocrinological Metabolism, 63, 1031±1035. Pocock, S.J. & Simon, R. (1975) Sequential treatment assignment with
Cumming, D.C., Young, J.C., Rebar, R.W. & Yen, S.S.C. (1982) balancing for prognostic factors in the controlled clinical trial.
Treatment of hirsutism with spironolactone. Journal of the American Biometrics, 3, 103±115.
Medical Association, 247, 1295±1298. Rittmaster, R.S. & Thompson, D.L. (1990) Effect of leuprolide and
Cusan, L., Dupont, A., Gomez, J.L., Tremblay, R.R. & Labrie, F. (1994) dexamethasone on hair growth and hormone levels in hirsute women:
Comparison of ¯utamide and spironolactone in the treatment of the relative importance of the ovary and the adrenal in the
hirsutism: a randomized controlled trial. Fertity and Sterility, 61, pathogenesis of hirsutism. Journal of Clinical Endocrinology and
281±287. Metabolism, 70, 1096±1102.
Eil, C. & Edelson, S.K. (1984) The use of human skin ®broblasts to Shane, H.P. & Potts, G.D. (1978) Oral progestational activity of
obtain potency estimates of drug binding to androgen receptors. spironolactone. Journal of Clinical Endocrinology and Metabolism,
Journal of Clinical Endocrinology and Metabolism, 59, 51±55. 47, 691±694.
Erenus, M., YuÈcelten, D., Gurbuz, O., Durmusogly, F. & Pekin, S. Shapiro, G. & Evron, S. (1980) A novel use of spironolactone:
(1996) Comparison of spironolactone±oral contraceptive versus treatment of hirsutism. Journal of Clinical Endocrinology and
cyproterone acetate±estrogen regimens in the treatment of hirsutism. Metabolism, 51, 429±432.
Fertility and Sterility, 66, 216±219. Siegberg, R., YloÈstalo, P., Laatikainen, T., Telkonen, R. & Stenman,
Farquhar, C.M., Birdsall, M., Mannign, P., Mitchell, J.M. & Farnace, U.H. (1987) Endocrine and clinical effect of spironolactone in female
J.T. (1994) The prevalence of polycystic ovaries on ultrasound hyperandrogenism. Archives of Gynecology, 240, 67±73.
scanning in a population of randomly selected women. Australian Spritzer, P.M., Billaud, L., Thalabard, J.C., Birman, P., Mowszowicz,
and New Zealand Journal of Obstetrics and Gynaecology, 134, 67± I., Raux-Demay, M.C., Kuttenn, F. & Mauvais-Jarvis, P. (1990)
72. Cyproterone acetate versus hydrocortisone treatment in late-onset
Ferriman, D. & Gallwey, J.D. (1961) Clinical assessment of body hair adrenal hyperplasia. Journal of Clinical Endocrinology and Meta-
growth in women. Journal of Clinical Endocrinology and Metabo- bolism, 70, 642±646.
lism, 21, 1140±1148. Spritzer, P.M., Lisboa, K.O., Mattiello, S. & Lhullier, F. (1998)
Hammerstein, J., Meckies, J., Leo-Rossberg, I., Moltz, L. & Zielske, F. Comparative effects of spironolactone and cyproterone acetate in the
(1975) Use of cyproterone acetate in the treatment of acne, hirsutism treatment of hirsute patients with polycystic ovary syndrome or
and virilism. Journal of Steroid Biochemistry, 6, 827±836. idiopathic hirsutism: a prospective controlled trial. Gynecological
Heiner, J.S., Greendale, G.A., Kawakami, A.K., Lapolt, P.S., Fisher, Endocrinology, 12 (suppl. 2), FC 118.

q 2000 Blackwell Science Ltd, Clinical Endocrinology, 52, 587±594


594 P. M. Spritzer et al.

Suraci, C., Costa, C. & De Pedrini, P. (1983) Ciproterone acetato e Wong, I.L., Morris, R.S., Chang, L., Spahn, M.A., Stanczyk, F.Z.
spironolattone a confronto nella terapia dell'irsutismo femminile. & Lobo, R.A. (1995) A prospective randomized trial comparing
Clinica Terapeutica, 104, 463±467. ®nasteride to spironolactone in the treatment of hirsute women.
Thomas, A.K., Slobodniuk, R., Taft, J., Cooper, M., Montalto, J. & Journal of Clinical Endocrinology and Metabolism, 80, 233±
Jerums, G. (1985) The treatment of hirsutism: experience with 238.
cyproterone acetate and spironolactone. Australian Journal of Zawadski, J.K. & Dunaif, A. (1992) Diagnostic criteria por polycystic
Dermatology, 26, 19±24. ovary syndrome: towards a rational approach. In Polycystic Ovary
Tolino, A., Petrone, A., Sarnacchiaro, F., Cirillo, D., Ronsini, S., Lombardi, Syndrome (ed. by J.R. Evans, S. Hersaltine & G.R. Marriam), pp.
G. & Nappi, C. (1996) Finasteride in the treatment of hirsutism: new 377±384. Blackwell, Boston.
therapeutic perspectives. Fertility and Sterility, 66, 61±65.

q 2000 Blackwell Science Ltd, Clinical Endocrinology, 52, 587±594