| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Departments of Obstetrics and Gynecology, and Medicine (R.A.), The University of Alabama at Birmingham, Birmingham, Alabama 35233; Department of Obstetrics and Gynecology, Columbia University, New York, New York 10032, and Department of Endocrinology, University of Palermo at Caltanissetta, 90139 Italy (E.C.); and Alopecia Research & Associated Technologies (ARATEC), and Department of Biochemistry and Molecular Biology, University of Miami School of Medicine (M.E.S.), Miami, Florida 34478
 |
Abstract |
|---|
 Top Abstract I. IntroductionII. Normal Regulation of...III. Hirsutism: How Much...IV. The Evolving Definition...V. Prevalence of IHVI. Pathogenesis of IHVII. Serum Markers of...VIII. Therapy of IHIX. ConclusionsReferences |
|---|
-reductase (5-RA)
determines to a great extent the production of dihydrotestosterone
(DHT), and consequently the effect of androgens on hair follicles.
While there are two distinct 5-RA isoenzymes, type 1 and type 2, the
activity of these in the facial or abdominal skin of hirsute women
remains to be determined. Although the definition of idiopathic
hirsutism (IH) has been an evolving process, the diagnosis of IH should
be applied only to hirsute patients with normal ovulatory function and
circulating androgen levels. A history of regular menses is not
sufficient to exclude ovulatory dysfunction, since up to 40% of
eumenorrheic hirsute women are anovulatory. The diagnosis of IH, when
strictly defined, will include less than 20% of all hirsute women. The
pathophysiology of IH is presumed to be a primary increase in skin
5-RA activity, probably of both isoenzyme types, and possibly an
alteration in androgen receptor function. Therapeutically, these
patients respond to antiandrogen or 5-RA inhibitor therapy.
Pharmacological suppression of ovarian or adrenal androgen secretion
may be of additional, albeit limited, benefit. New therapeutic
strategies such as laser epilation or the use of new biological
response modifiers may play an important role in offering a more
effective means of treatment to remove unwanted hair. Further
investigations into the genetic, molecular, and metabolic aspects of
this disorder, including only well defined patients, are needed.
-RA activity in IH -Androstane-3,17ß-diol glucuronide -RA inhibitors |
I. Introduction |
|---|
|
TopAbstractI. IntroductionII. Normal Regulation of...III. Hirsutism: How Much...IV. The Evolving Definition...V. Prevalence of IHVI. Pathogenesis of IHVII. Serum Markers of...VIII. Therapy of IHIX. ConclusionsReferences |
|---|
|
II. Normal Regulation of Hair Growth and Differentiation |
|---|
|
TopAbstractI. IntroductionII. Normal Regulation of...III. Hirsutism: How Much...IV. The Evolving Definition...V. Prevalence of IHVI. Pathogenesis of IHVII. Serum Markers of...VIII. Therapy of IHIX. ConclusionsReferences |
|---|
The earliest stage of human embryonic life at which organization of the primitive epidermis leads to the formation of hair structure is between 9 to 12 weeks (10, 11). During development of hair structures the primitive mesoderm forms the hair germ, with an associated down growth bringing ectoderm cells to the newly forming hair peg. The hair germ from the mesoderm ultimately forms the fibrous sheath of the follicle and the dermal papilla, while the hair peg becomes a solid core of epithelial cells that encloses the dermal papilla. The ectodermal and mesodermal elements remain in intimate contact and reflect an association that continues throughout the life of the hair follicle. These embryonic hair follicles may be producing hair by the 16th to 20th weeks of development.
Hair is composed of keratin proteins, which form the hair shaft. The hair shaft grows within the outer hair root sheath, which forms part of the epidermis. Structurally, there are three types of hair. Lanugo is a soft hair densely covering the skin of the fetus, which is shed between the first and the fourth month postpartum. Vellus hairs are also soft fine hairs, but larger than lanugo hairs. Vellus hairs are usually nonpigmented, generally measuring less than 2 mm in length, and cover the apparently hairless areas of the body. Histologically, vellus hairs have diameters that do not exceed 0.03 mm, smaller in diameter than that of the investing root sheath (12). Terminal hairs are longer, pigmented, and course in texture. This hair makes up the eyebrows, eyelashes, scalp hair, and pubic and axillary hair in both sexes, and much of the body and facial hair of men (8, 9). Terminal hairs are often described as being "medullated." The "medulla" of the hair follicle is the innermost area of terminal hairs and is thought to consist of "collapsed protein," although the exact composition of this area remains controversial. The smaller lanugo and vellus hairs are thought be nonmedullated (i.e., not having this inner pocket).
In skin, the hair follicles form groups, called follicular units (FU); each FU consists of approximately two to four hair follicles along with sebaceous glands and connective tissue sheaths (13, 14, 15). The concentration of FUs in skin, at least in the scalp, can vary between ethnic groups (13, 14). However, there is no gender difference in the number of FUs within each racial/ethnic group. Hence, the visible difference in body hair growth between men and women does not relate to the number of FUs, but to the type and quality of the hair within these follicles (8, 9).
B. Growth phases and regulation of hair growth
There are three phases to hair growth. An active growing phase
(anagen); followed by an involutional stage (catagen), in which the
hair stops growing and the hair bud shrinks; and finally, the telogen
phase in which the hair is resting, and then shed, as new hairs
displace it (8, 9). In humans, hair has the appearance of growing
continuously, the result of disynchrony in the growth phases between
the different hair follicles. In essence, while some hairs are in the
active phase of growth (anagen), others are in the resting phase
(telogen), and vice-versa, giving the impression of continuous growth.
The length of time of the growth phase will vary depending on where the
hair follicle is located. For example, on the scalp the anagen phase
may last 2 to 6 yr, while the anagen phase of body hairs may only last
3 to 6 months. Alternatively, the duration of the catagen and telogen
phases is similar in scalp and body hair, lasting 2 to 3 weeks and 3 to
4 months, respectively. The anagen-telogen ratio (the ratio of hairs in
anagen to the number of hairs in telogen) is often used to estimate
hair growth activity in specific areas of the skin, with a
higher ratio indicating a more active hair growth.
Sex steroids and a number of local and systemic factors can act
directly and indirectly on the dermal papilla to regulate hair growth.
In addition, these factors may also act on other parts of the hair
follicle, including the outer and inner root sheaths and the follicular
stem cells of the "bulge" area (16), which may be as important as
the dermal papilla in regulating hair growth (Fig. 1
).
|
Other hormones such as thyroid (24, 25) and GH (26) can also alter hair
growth. In general, a deficiency of hormones such as thyroid hormone or
GH, whether disease-associated or drug-induced, is associated with
changes in the anagen-telogen ratio in scalp and body hair (24, 25, 27). Hypothyroid patients treated with thyroid hormone replacement
generally begin to experience regrowth of scalp hair within
approximately 8 weeks (24). Investigators have demonstrated the
presence of thyroid hormone receptors in the outer sheath cells of hair
follicles, and a positive effect of
L-T3 on cellular proliferation in
cultured hair follicles (28). GH-deficient men treated with GH
substitution demonstrate an increase in body hair that occurs without
an observable increase in free androgen index (26), suggesting that GH
can directly stimulate body hair growth independent of an increase in
circulating androgens. Evidence of GH binding has been reported in all
layers of the lower one-third of the hair follicle, the outer sheath of
the upper two-thirds of the follicle, and in the dermal papilla (29).
Notwithstanding, a direct effect of GH on in vitro hair
follicle growth or morphology has not been observed (19). It is
possible that the effect of GH may be mediated through increased
production of insulin-like growth factor I (IGF-I). In human scrotal
skin fibroblasts, IGF-I, but not IGF-II or insulin, increased the
activity of 5-reductase (5-RA), an enzyme necessary for the
peripheral potentiation of the androgen effect (30). In
addition, in cultured hair follicles, IGF-I, more than IGF-II or
insulin, stimulated hair follicle growth (19).
2. Sex steroids. Of the sex steroids, androgens are the most important in determining the type and distribution of hairs over the human body. Under the influence of androgens, hair follicles that are producing vellus-type hairs can be stimulated to begin producing terminal hairs. In vivo experiments clearly document the effect of exogenous androgens on the differentiation of hair follicles in androgen-sensitive areas of the body (e.g., genitalia and facial beard area) in normal males (31), eunuchs (32), and female-to-male transsexuals (33). In addition to stimulating the production of terminal hairs in some skin areas, androgens prolong the anagen phase of body hairs, while shortening the anagen phase of scalp hairs (34, 35). Androgens also increase sebum secretion. Therefore, not only do androgens alter the type of hair present, but they may lengthen body hairs by increasing the length of the anagen phase, and will increase the oiliness of skin and hair.
Whether the ability of some hair follicles to produce terminal hairs
(i.e., "terminalize" vellus hairs) under the influence
of androgens is due to differences in androgen sensitivity or to other
(as yet undefined) primary differences in hair follicles remains
unclear. There is considerable variability among body hairs
(individually and between different skin areas) in their 5-RA
content and ability to metabolize androgens. It is very possible that
the variability in the ability of androgens to stimulate the production
and growth of terminal hairs may be regulated, to a large extent, by
these differences. Finally, the concentration of the androgen receptor
in determining the effect of androgens on terminal hair growth is
critical, as is clearly demonstrated by the sparse terminal body hair
of patients with incomplete and complete androgen insensitivity (36).
It should be noted that sex steroids may act on the hair follicle
independently of their circulating levels. This local or intracrine
effect is extremely important and may prove to play a significant role
in the development of androgen excess disorders, particularly IH. The
intracrinology of the hair follicle was recently demonstrated in
androgenetic alopecia (i.e., common male pattern hair loss),
which affects both men and women. In this disorder androgens shorten
the active anagen growth phase of the scalp hairs in genetically
susceptible men and women, producing miniaturized hairs through
subsequent hair cycles and a loss of grossly visible hairs. In men,
androgenetic alopecia is associated with a progressive loss of frontal
and occipital hair, followed by the loss of sagittal hairs (37). In
contrast, in women this disorder presents as a diffuse thinning of hair
growth in the frontal and sagittal scalp (38). It is possible that
differences in local sex steroid metabolism accounts, at least in part,
for the variations in clinical presentation between women and men with
this disorder (39). To test this hypothesis we studied 12 men and 12
women with androgenetic alopecia and noted that the content of
cytochrome P450 aromatase and 5-RA in the frontal hair follicles of
these women was 6-fold greater and 3-fold less, respectively, than in
the frontal hair follicles of the affected men (39). Hence, it is
possible that differences in local androgen metabolism could account
for the different clinical presentations of androgenetic alopecia
between genders. Nonetheless, it is still unclear whether these
differences are due to increased local clearance of androgens (via
conversion to estrogens) or from a direct estrogen effect. In fact,
some investigators have suggested that estrogens can directly cause
scalp hair loss by altering the anagen-telogen rate (40).
In contrast to scalp hair, a direct effect of estrogens on the
regulation of body hair growth is still uncertain (41). In the
rodent model, estrogens have been observed to inhibit the initiation of
growth and prolong the entire growth cycle of the hair follicle (42, 43). Progesterone and estradiol, at least in high doses, also inhibit
5-RA activity in human genital and pubic skin (44), possibly
decreasing the local production of the potent androgen
dihydrotestosterone (DHT). Little more is known regarding the direct
effects of progesterones and estrogens on hair growth in the human.
3. Skin 5-RA activity. Numerous experiments have documented
the 5-reduction of testosterone (T), androstenedione, and
dehydroepiandrosterone to DHT in vitro by skin and
individual hair follicles (34). In the intact organism, DHT is
primarily formed from the peripheral nonhepatic 5-reduction of T
in men and androstenedione in women (45, 46). In fact, the effect of
androgens on the development of terminal hairs appears to require, to a
significant degree, the action of 5-RA. This concept is supported by
the sparse beard development in patients with 5-RA deficiency (36, 47, 48). However, even among patients with 5-RA deficiency, the
metabolism of T to DHT by dermal papilla cells in vitro
correlates with terminal hair growth (49).
The important role of 5-RA and DHT in determining male-type hair
growth is further supported by the observations of Farthing and
colleagues (50) in men with coeliac disease, individuals who experience
a greater than normal divergence in their circulating concentrations of
T and DHT (50). In these men the growth rate of facial hairs correlates
directly with their plasma DHT levels, which are lower than normal.
Alternatively, hair density correlates with T (but not DHT) levels,
which in these patients are higher than normal. This dichotomous, yet
synergistic, effect of T and DHT is consistent with the partial
virilization observed in patients with 5-RA deficiency at puberty
(36, 47, 48). Overall, while circulating T in higher concentrations may
have a direct effect on the FU, DHT is the primary androgen active at
the target tissue (Fig. 2).
|
-RA determines, to a great extent, the
production of DHT and, consequently, the effect of androgens on hair
follicles. Progesterone and estradiol in high doses inhibit 5-RA
activity in human genital and pubic skin (44), while androgens were
found to stimulate and increase peripheral 5-RA activity (51).
However, not all androgens may be equal in their ability to affect
peripheral 5-RA activity and hair growth. For example, pubic skin
from patients with "adrenal hyperandrogenism" demonstrates levels
of 5-RA activity that are similar to those of controls and less than
those of patients with ovarian hyperandrogenism (52). In addition to
androgens, other factors may also control the distribution and activity
of 5-RA (Fig. 2
). For example, preliminary data indicate that local
(e.g., transforming growth factor-ß and epidermal growth
factor) and circulating growth factors (e.g., insulin-like
factor-I, activin A, and inhibin A) alter the activity of this enzyme
(30, 53, 54). Nonetheless, the specific role and mechanism of action of
these regulatory factors remain to be better defined.
5-RA activity is actually the product of the function of two
distinct isoenzymes, type 1 and type 2 5-RA (55) (Table 1). Each 5-RA isoenzyme is encoded by
separate genes located on chromosomes 5 and 2, respectively, and
demonstrate different biochemical characteristics and tissue
distributions (39). The type 1 isoenzyme has an optimum pH near 8.0, a
Michaelis-Menten constant (KM) of approximately
24 µM, and less cofactor specificity and is located
primarily in the microsomal subcellular fraction. The type 2 isoenzyme
has a more acidic pH range and a KM of
approximately 0.33 µM and is found in both the nuclear
and microsomal subcellular fractions. Therefore, type 2 5-RA can be
considered to be more active in synthesis of DHT, based on their
relative KM values (Table 1).
|
-RA isoenzymes are widely distributed throughout the body
(Table 1), although recent studies have noted that the type 2 isoenzyme
predominates in the testes, prostate, and the hair follicles of beard
and genital hair (55). Controversy still exists regarding which
isoenzyme is located in the hair follicles of the scalp. One
investigator has reported that only type 2 5-RA is present in scalp
hair follicles, whereas both types 1 and 2 were demonstrated in human
sebaceous glands (56). Other investigators (39, 57) have also found
both isoenzymes in scalp sebaceous glands, but they also reported
finding both types of 5-RA isoenzymes in scalp hair follicles (Fig. 1). Interestingly, both types of 5-RA can be localized to the outer
root sheath of the hair follicle, with less expression in the dermal
papilla (39) (Fig. 1). The type 1 isoenzyme is heavily expressed in the
distal portions of the sebaceous gland, whereas type 2 5-RA is found
primarily in the proximal duct of sebaceous glands (39). This location
of type 2 5-RA may be clinically important, since the proximal duct
of the sebaceous gland is where key pathological processes leading to
acne vulgaris take place.
In addition to variances in gender and skin location, the two 5-RA
isoenzymes may also differ functionally. For example, preliminary data
suggest that in skin the two isoenzymes may be regulated by androgens
to a different degree, with the type 1 isoenzyme more susceptible to
up-regulation by these steroids (58). Furthermore, type 2 5-RA is
present in skin beginning at birth, while the type 1 isoenzyme becomes
evident in skin only at the end of childhood or around the time of
puberty (55).
It should be noted that 5-RA activity has been primarily measured
biochemically in follicular cells of plucked and microdissected hairs
and in skin minces and primary cell cultures. Most studies determining
5-RA activity have been performed using genital or scalp skin
without determining the specific isoenzyme forms present. Furthermore,
the activity of these isoenzymes in the facial or abdominal skin of
hirsute women remains to be determined. It should be noted that the
recent availability of specific Northern blotting for identifying the
individual isoenzyme mRNAs has improved our capacity to detect the
presence of each of the 5-RAs (39). The development of these
techniques may lead to a better understanding of the differences in
hair growth between genders and the underlying pathophysiology of
hyperandrogenic disorders such as androgenetic alopecia, acne, and
hirsutism.
|
III. Hirsutism: How Much Is Too Much? |
|---|
|
TopAbstractI. IntroductionII. Normal Regulation of...III. Hirsutism: How Much...IV. The Evolving Definition...V. Prevalence of IHVI. Pathogenesis of IHVII. Serum Markers of...VIII. Therapy of IHIX. ConclusionsReferences |
|---|
Visual methods of determining the degree of hirsutism usually follow those originally described by Ferriman and Gallwey (1). In their study these investigators scored the density of terminal hairs at 11 different body sites (i.e., upper lip, chin, chest, upper back, lower back, upper abdomen, lower abdomen, arm, forearm, thigh, and lower leg) in 161 women ages 18 to 38 yr old attending a general medical outpatient clinic. In each of these areas a score of 0 (absence of terminal hairs) through 4 (extensive terminal hair growth) was assigned. In this study, hair growth over the forearm and lower leg was noted to be less sensitive or indifferent to androgens, and subsequent modifications of the Ferriman-Gallwey method have deleted scoring of these areas (59, 60). Scoring of hair growth in the sideburn area, lower jaw and upper neck, and buttocks have been included in more recent scoring systems (61).
B. Defining hirsutism
In their original report, Ferriman and Gallwey noted that if only
the nine "hormonal" skin areas (i.e., excluding the
lower leg and forearms) were considered, 9.9% of their 161 women had a
score above 5, 4.3% had a score above 7, and 1.2% had a score greater
than 10 (1). From these data a score of 8 or more has been considered
to represent hirsutism. It should be kept in mind that these studies
were performed in a predominantly white population. Although
racial/ethnic differences in the number, distribution, or androgen
sensitivity of hair follicles in normal individuals remains to be
better defined, information regarding the prevalence of hirsutism in
different racial groups is scant.
We prospectively studied 369 consecutive reproductive-aged black
(n = 195) and white (n = 174) women, examined at the time of
their preemployment physical (62). A previously described modification
of the Ferriman-Gallwey method (60), in which nine body areas are
assessed, was used. Of these women, 7.6%, 4.6%, and 1.9%
demonstrated a modified Ferriman-Gallwey score 
6, 8, or 10,
respectively. Obviously, the overall cutoff score used to define
hirsutism will decrease as the number of areas assessed (or the maximum
score assigned to each area) is reduced. For example, Lorenzo (59)
studied 300 unselected female medical patients using a modification of
the Ferriman-Gallwey score, in which only five areas of the body were
scored (chin, upper lip, chest, abdomen, and thighs). Using this
scoring method, this investigator did not observe a hirsutism score
over 5 in any of these women. While the exact numerical cutoff score
used to define hirsutism will vary according to the quantifying system
used, a value of 7 or greater is evident in only 5% of the general
population when a scoring system assessing nine body areas is used
(62).
In the study prospectively evaluating the prevalence of hirsutism we
did not note any significant racial differences: 8.0%, 2.8%, and
1.6% of white women and 7.1%, 6.1%, and 2.1% of black women had
hirsutism scores 6, 8, or 10, respectively (62). Alternatively,
various investigators have noted that, in comparison to white patients,
hirsutism in Asian women is relatively uncommon even in the face of
similar metabolic and endocrine abnormalities (63, 64, 65). Thus, in Asian
women the absence of hirsutism cannot be used to exclude the presence
of a hyperandrogenic disorder.
|
IV. The Evolving Definition of IH |
|---|
|
TopAbstractI. IntroductionII. Normal Regulation of...III. Hirsutism: How Much...IV. The Evolving Definition...V. Prevalence of IHVI. Pathogenesis of IHVII. Serum Markers of...VIII. Therapy of IHIX. ConclusionsReferences |
|---|
). The absence
of a clear definition of IH has been one of the principal sources of
continuing confusion and contradiction in the study of this disorder.
|
). In essence, IH was presumed if the presence or source of
androgen excess was not clear. Patients were included if they did not
have polycystic ovaries, congenital adrenal hyperplasia, or
androgen-secreting tumors, or if the limited battery of androgen assays
then available yielded normal results (66, 67, 68, 69, 70, 71). Importantly, at that
time many investigators defined patients as having IH regardless of
whether they had irregular menstruation/ovulation and/or elevated
circulating androgen levels, which today would be more consistent with
the diagnosis of PCOS (72). In addition, most of these studies did not
exclude patients with NCAH, as the diagnostic criteria for this
disorder had not yet been established.
B. As "hirsutism with a history of regular menstrual cycles"
Subsequently, IH was defined as affecting those patients with
hirsutism and regular menses (73, 74, 75), regardless of circulating
androgen levels (Table 2). However, a normal menstrual history in the
hirsute patient does not exclude ovulatory dysfunction (76), elevated
circulating androgen levels (77), or NCAH (78). For example, Mehta and
colleagues noted that while 40% of their hirsute women had regular
cycles, approximately one-half of these had elevated levels of one or
more androgens (77). Furthermore, in a prospective study of 64 hirsute
patients claiming to have regular monthly menses, approximately 40%
actually had oligoovulation when examined more closely (76). The
presence of oligoovulation in hirsute or hyperandrogenemic patients,
after the exclusion of related disorders (e.g., NCAH,
thyroid dysfunction, or hyperprolactinemia) is consistent with the
diagnosis of PCOS, as noted in the proceedings of a consensus
conference sponsored by the NIH/NICHD in April of 1990 (72). Hence, a
history of "regular menses" should not be used to exclude the
presence of ovulatory dysfunction or hyperandrogenemia and cannot be
used to determine IH.
C. As "hirsutism with normal ovulatory function and normal
circulating androgen concentrations"
The presence of oligoovulation in a hirsute patient is generally
considered to indicate the presence of androgen excess (as in PCOS or
NCAH) and to exclude IH. Nonetheless, it is possible that a patient
with IH (i.e., with hirsutism not related to androgen
excess) may incidentally be oligoovulatory due to other nonrelated
causes. It is also difficult to consider hirsutism as being
"idiopathic" if patients demonstrate clear-cut hyperandrogenemia.
In essence, in the patient with hirsutism, normoovulation, and
hyperandrogenemia, while the source of the elevated androgens may
remain a mystery, the physical finding of hirsutism is explained by the
increased circulating androgen levels. Consistent with this concept,
investigators have defined IH as hirsutism in the presence of a normal
total T, alone (67) or combined with normal levels of
androstenedione, dehydroepiandrosterone sulfate (DHEAS), LH/FSH,
and 17-hydroxyprogesterone levels, although the status of ovulatory
function was not considered (79).
To exclude those patients with both ovulatory dysfunction and
hyperandrogenemia, we (76, 80) and others (81, 82, 83) have defined IH more
strictly as diagnosable only in women who have 1) hirsutism
and 2) normal ovulatory function and 3) a normal
androgen profile (Table 2). Since a history of "regular menses"
does not accurately reflect ovulatory status in the hirsute patient
(76), ovulatory function must be confirmed by using a daily basal body
temperature (BBT) charting and/or a luteal phase (day 20–24 of the
menstrual cycle) progesterone level. Hirsute patients demonstrating
ovulatory dysfunction and no evidence of other related disorders
(e.g., NCAH, thyroid dysfunction) can be considered to have
PCOS (72). Nonetheless, the presence of polycystic-appearing ovaries
need not be used to exclude IH, as long as the patient has regular
ovulation and normal androgen levels. Since the measurement of a total
T serum level is insensitive (84, 85, 86, 87), we prefer to exclude
hyperandrogenemia by at least measuring DHEAS and total and free
T levels. In addition, the follicular phase basal
17-hydroxyprogesterone level should be measured, or an ACTH stimulation
test performed, to exclude 21-hydroxylase-deficient NCAH (88). Thyroid
dysfunction should also be excluded, generally by the measurement of
TSH. The ingestion of exogenous androgens should also be excluded.
In conclusion, the current diagnosis of IH should be one of exclusion, in which ovulatory dysfunction, hyperandrogenemia, and other defined androgen excess disorders are ruled out. This strict definition of IH excludes the hirsute patient with normal ovulatory function but with elevated circulating androgen levels. Although the underlying pathophysiology in these women remains unclear, a thorough discussion of these patients is beyond the scope of this review.
|
V. Prevalence of IH |
|---|
|
TopAbstractI. IntroductionII. Normal Regulation of...III. Hirsutism: How Much...IV. The Evolving Definition...V. Prevalence of IHVI. Pathogenesis of IHVII. Serum Markers of...VIII. Therapy of IHIX. ConclusionsReferences |
|---|
We recently studied the prevalence of IH in Alabama, defining the
disorder by the presence of significant hirsutism in the absence of
both ovulatory dysfunction and hyperandrogenemia (76). Of 132
consecutive hirsute women studied, at initial evaluation 48% had
cycles less than 35 days in length (i.e., "regular"). Of
these patients, 39% actually had oligo/anovulation as evidenced by
their BBT chart and luteal phase progesterone level. Hence, among our
hirsute patients, fully 71% had either overt or subtle ovulatory
dysfunction and could be considered as having PCOS (Fig. 3). Of the remaining 39 patients with
hirsutism and regular ovulatory function, 22 (17% of the total) with
total T, free T, and DHEAS levels within normal were diagnosed as
suffering from IH under the strict definition. A similar study in
southern Italy confirmed these results. Of 598 consecutive Italian
patients complaining of hirsutism, 298 (51%) reported "regular"
cycles (80). Nonetheless, of these women only 36 patients (or 6% of
all hirsute women seen) had normal circulating DHEAS and total and free
T levels, consistent with the diagnosis of IH.
|
|
VI. Pathogenesis of IH |
|---|
|
TopAbstractI. IntroductionII. Normal Regulation of...III. Hirsutism: How Much...IV. The Evolving Definition...V. Prevalence of IHVI. Pathogenesis of IHVII. Serum Markers of...VIII. Therapy of IHIX. ConclusionsReferences |
|---|
A. Exaggerated peripheral 5-RA activity in IH
It could be postulated that women with IH have a primary increase
in cutaneous 5-RA activity. Jenkins and Ash (90) reported that in
two of three women with "idiopathic hirsutism" the production of
DHT from T by minces of suprapubic skin was 50–100% above the upper
normal limit. Unfortunately, while these subjects with IH had normal
androgen levels, no mention was made of their ovulatory function.
Serafini and Lobo (91) studied 10 women with IH (defined by the
presence of hirsutism, normal androgen levels, and regular menstrual
cycles) and reported that these patients had higher levels of 5-RA
activity measured biochemically in genital skin, compared with
controls. Although circulating androgens are known to increase
peripheral 5-RA activity (51, 92), the circulating androgen levels
were normal in the IH women studied. Hence, the investigators
tentatively concluded that in the patients studied increased skin
5-RA activity could be a primary pathophysiological event (91).
It is possible that increased skin 5-RA may not be a factor unique
to IH and may simply reflect the development of hirsutism. For example,
in the studies by Jenkins and Ash (90) and Serafini and Lobo (91), a
similar increase in skin 5-RA activity was observed among patients
with PCOS. It was proposed that, in contrast to patients with IH, the
increase in peripheral 5-RA activity in women with PCOS occurs
secondarily to their higher androgen levels (91). In fact, these
investigators reported other data that suggest that peripheral events
play a role in the development of hirsutism, even among patients with
PCOS (93). Hence, it is possible that the increased peripheral 5-RA
activity observed in IH simply reflects the development of hirsutism,
regardless of cause. Overall, from the available data it is not
possible to conclusively establish that increased 5-RA activity
plays a primary role in the development of IH.
Likewise, it is still unclear which of the 5-RA isoenzymes, if any,
is predominant in the development of IH. Using molecular analysis of
isoenzyme-specific mRNAs, the concentrations of the type 1 5-RA
isoenzyme in sebaceous glands was higher in patients with acne (94, 95), with no significant differences in the concentration of the type 2
isoenzyme. Consistent with this finding, the use of finasteride (a type
2 5-RA inhibitor, see below) had no effect on sebum production in
men being treated for prostate hyperplasia (96). Nonetheless, in
contrast to acne, it is likely that the type 2 isoenzyme plays a
significant role in hirsutism since finasteride is at least partially
effective in its treatment (97, 98). Further studies using well defined
patients are required to establish the precise role of each of these
isoenzymes in the development of hirsutism, and specifically IH.
B. Androgen receptor polymorphisms
Although it has been postulated that a difference in the number of
peripheral androgen receptors results in IH, no quantitative
differences in androgen binding capacity (i.e., androgen
receptor content) between IH patients and controls have been found
(81). However, it is possible that qualitative, rather than
quantitative, differences in the androgen receptor may exist in women
with IH. Such a functional difference may result from genetic
alterations of the androgen receptor and may be sufficient to explain
the development of hirsutism in the absence of changes in circulating
androgen levels (99). In fact, genetic variations of the androgen
receptor have been associated with the development of other
androgen-sensitive disorders, such as prostate cancer (100).
Furthermore, it is known that sequence variations occur more often in
the androgen receptor gene than in other steroid receptor genes (101),
and it is possible that these polymorphisms affect androgen sensitivity
and phenotypic expression.
An example of a common genetic variation, possibly affecting the function of the androgen receptor, is the number of trinucleotide CAG repeats in exon 1 of the androgen receptor gene. These trinucleotide repeats have been found to vary widely (i.e., be polymorphic) among humans (102). The CAG codons encode for a long stretch of glutamines within the amino terminus of the transactivation domain of the androgen receptor. Expansion of this region of the androgen receptor has recently been implicated in the development of androgen-related skin disorders in men and women alike, including androgenetic alopecia and hirsutism (99, 103). In one report the number of CAG repeats of the androgen receptor gene were studied in women with peripheral signs of hyperandrogenism (99). Healthy controls and women with acne were found to have a similar mean (±SD) number of trinucleotide repeats (21 ± 3 and 20 ± 3, respectively), which was higher than that of women with androgenetic alopecia or hirsutism (17 ± 3 and 16 ± 3, respectively). Similarly, another study of 110 Hispanic women with IH found an inverse correlation between the hirsutism score and the size of the CAG repeat in the androgen receptor gene (103). Overall, it is possible that shorter CAG-repeat lengths in the N-terminal domain of the androgen receptor may be associated with the development of androgenetic alopecia and hirsutism.
These studies suggest that the development of hirsutism, and possibly IH, may be influenced to some degree by the molecular characteristics of the androgen receptor. However, the androgen receptor gene is located on the X chromosome and, according to the Lyon hypothesis, only one allele will be generally expressed in target tissues depending on the random pattern of X chromosome inactivation. It is then uncertain which X-linked allele is expressed in the tissues of women with hirsutism, since many women are heterozygous for the size of the CAG repeat region of the androgen receptor gene. Overall, further studies are required to determine the role of these androgen receptor gene polymorphisms in the development of IH.
C. Altered androgen metabolism
Since androgens themselves stimulate peripheral 5-RA activity
(44, 51), the presence of increased 5-RA activity is not sufficient
proof of a primary exaggeration in the function of this enzyme. In
fact, it remains unclear whether excessive exposure of the hair
follicle to androgens from the circulation plays a role in the
development of IH. In this regard, some investigators have reported
that the production rate of various androgens was increased in patients
with IH (66, 69). However, in these studies IH was defined as the
presence of "unexplained hirsutism," and a considerable fraction of
patients also had irregular menstrual cycles and increased androgen
levels.
Glickman and Rosenfield (104) studied androgen metabolism in the pubic hair follicles of four patients with IH. These investigators noted that only one of their patients had an abnormality in hair follicle metabolism. This patient had an exaggerated rate of inactivation of 17ß-hydroxysteroids (i.e., T and DHT) to 17-ketosteroids, which could not be considered to increase the effect of androgens on the hair follicle. Faredin and Toth (105) studied the lower abdominal skin of three patients with IH defined by the presence of regular menstrual cycles and normal androgen levels, although two of the women had increased circulating levels of androstenedione (105). These investigators reported a series of abnormalities in local androgen metabolism, although an exaggerated conversion of androstenedione to T appeared to be predominant, suggesting an exaggeration in 17ß-hydroxysteroid activity. Overall, the role of systemic or local abnormalities in androgen metabolism in IH remains to be better defined.
Insulin resistance and hyperinsulinism are now recognized as important features of PCOS, stimulating ovarian androgen secretion and suppressing sex hormone-binding globulin (SHBG) production (106). Insulin has been demonstrated to stimulate hair follicle growth in vitro (19). Nonetheless, it is unclear whether patients with IH also have abnormalities of insulin action. Paoletti and colleagues (107) noted that patients with both IH and PCOS had significantly higher fasting and oral glucose-stimulated levels of glucose, insulin, and C-peptide, compared with controls. Furthermore, the administration of the antiandrogen flutamide significantly blunted the fasting and stimulated levels of insulin in IH, but not PCOS, patients. Nonetheless, the investigators did not confirm the presence of normal ovulatory function in their patients with presumed IH, although these women had regular menses, an LH/FSH level less than 1, and a normal ultrasound of the ovaries. Hence, the presence of abnormalities of insulin action and its role and importance in the pathology of women with IH remain to be demonstrated.
Overall, it is clear that only an improved understanding of the control
and physiology of 5-RA and the AR will allow us to further elucidate
the pathophysiology of IH. Currently, with our limited understanding of
the pathogenesis of IH, the term "idiopathic" for those patients
who have regular ovulation, normal circulating androgens, and hirsutism
remains appropriate. Nonetheless, as this disorder is under active
investigation, the definition of the disorder is a dynamic one, subject
to revision as our understanding of the pathogenesis of the disorder
develops.
|
VII. Serum Markers of IH |
|---|
|
TopAbstractI. IntroductionII. Normal Regulation of...III. Hirsutism: How Much...IV. The Evolving Definition...V. Prevalence of IHVI. Pathogenesis of IHVII. Serum Markers of...VIII. Therapy of IHIX. ConclusionsReferences |
|---|
-RA
activity, increased circulating levels of DHT could be expected.
However, serum levels of this 5-reduced androgen are frequently
normal in hirsute women (93), probably because most DHT produced in
skin is not secreted into the circulation but acts locally before its
rapid metabolism.
B. 5-Androstane-3,17ß-diol glucuronide
The DHT metabolites, 3- and 3ß-androstanediol, have also been
studied as potential serum markers of IH. Similar to DHT, little or no
unconjugated androstanediol is found in the circulation (108). However,
the conjugated metabolites of these steroids have a much longer
half-life and are present in significant amounts in serum, allowing
their routine measurement. Unfortunately, although the circulating or
urinary levels of the androgen metabolites, 3,17ß-androstanediol
sulfate or androsterone glucuronide, may be increased in patients with
IH, they are also often normal. Overall, 3,17ß-androstanediol sulfate
or androsterone glucuronide are not useful as markers for the disorder
(109).
A number of investigators have suggested that the levels of
3-adiol-G may serve as a marker of peripheral 5-RA activity (72, 108, 110, 111). However, the serum level of this conjugate is not only
dependent on skin 5-RA activity, but also on the circulating levels
of androgen precursors (74, 112, 113). Consequently, higher levels of
circulating 3-diol-G have been reported in hirsute women with PCOS,
and those with adrenal or ovarian androgen excess (93, 114, 115, 116). Since
DHEAS and androstenedione are the major precursors of 3-adiol-G, the
level of this conjugate closely reflects adrenal androgen biosynthesis
(109). Thus, elevated serum levels of this conjugate either simply
reflect the presence of hirsutism, regardless of cause, or the excess
production of DHEAS by the adrenal or androstenedione from any source.
In fact, higher levels of circulating 3-diol-G suggest the presence
of IH only when all other circulating androgens are normal.
However, up to 20% of women with IH may demonstrate normal serum
levels of 3-adiol-G (111). The measurement of serum 3-adiol-G does
not appear to be of greater value than the physical observation of
excess hair growth in the diagnostic evaluation of the hirsute female.
In addition, the levels of this hormone do not predict therapeutic
response (74, 117), and we do not recommend the routine measurement of
3-adiol-G serum levels in the evaluation of IH or in other hirsute
patients.
|
VIII. Therapy of IH |
|---|
|
TopAbstractI. IntroductionII. Normal Regulation of...III. Hirsutism: How Much...IV. The Evolving Definition...V. Prevalence of IHVI. Pathogenesis of IHVII. Serum Markers of...VIII. Therapy of IHIX. ConclusionsReferences |
|---|
In addition to OCPs, androgen suppression can be achieved using
long-acting GnRH analogs. Defining IH as hirsutism in the face of
regular menstrual cycles, treatment with GnRH analogs appeared to be
generally effective in reducing hair growth in the small number of
patients studied (119, 120). Falsetti and Pasinetti (121) studied 16
patients with IH, defined by normal androgen values, although 75% also
had regular ovulatory cycles, and 16 patients had PCOS. All
women were randomized to receive either leuprolide acetate or
leuprolide plus an OCP containing 35 mg ethinyl estradiol (EE) and 2 mg
CPA. Both regimens resulted in a 20–25% decrease in the hirsutism
score and a reduction in hair diameter in either patient group. These
data suggest that reducing androgen levels, even when originally within
the normal range as in IH, can result in the amelioration of hirsutism.
It is possible that the reduction in circulating androgens leads to a
decrease in the availability of substrate for peripheral 5-RA, as
well as a reduction in the overall 5-RA activity, which is generally
stimulated by androgens (51).
B. Androgen receptor blockers
1. Spironolactone (SPA). The most common androgen blocker used
for the treatment of hirsutism in the United States is SPA, an
aldosterone antagonist structurally related to progestins. Overall, SPA
is an effective therapy for hirsutism (122, 123, 124), including IH (98). In
addition to antagonizing aldosterone, SPA competes with DHT for binding
to the androgen receptor. However, SPA has only 1/20th the binding
affinity of DHT for the androgen receptor, explaining why high doses of
the drug may be required for adequate suppression of hair growth. In
addition, SPA has an inhibitory effect on 5-RA and competes with
androgens for binding to SHBG (125, 126, 127, 128). SPA, or its 17-hydroxylated
metabolites, also inhibits the action of various enzymes involved in
androgen biosynthesis, although this effect is generally observed only
at doses greater than 200 mg/day (125, 126, 127, 128). SPA also demonstrates
variable progestational activity, which may decrease the circulating
LH/FSH ratio by decreasing the response of LH to GnRH. In turn, this
decrease may ameliorate the LH-stimulated secretion of ovarian
androgens (127, 129).
Since SPA acts through mechanisms different from that of OCPs, it may
be possible to improve the overall therapeutic effectiveness of this
drug by combining these medications, even in patients with IH
(130, 131, 132). As previously noted, the peripheral production of DHT is
both the result of 5-RA activity and of the circulating levels of
precursors, primarily T and androstenedione. The use of OCPs in
combination with SPA, while providing adequate contraception,
also helps to minimize the dysfunctional uterine bleeding or worsening
oligomenorrhea often observed in women using SPA alone.
SPA was initially approved and marketed as a diuretic, and primary side effects include polyuria, nocturia, and hypotension with associated headaches, fatigue, or even syncope. Nonetheless, patients rapidly develop a tolerance to this effect with chronic use, and few changes in serum electrolytes or blood pressure are seen with long-term therapy (122, 133). Because of its potassium-sparing nature, SPA should not be used in conjunction with other potassium-sparing diuretics, thiazides, in renal insufficiency, or with excess potassium intake, since patients may develop life-threatening hyperkalemia. Physicians may wish to evaluate serum electrolytes and blood pressure 2–4 weeks after treatment is started.
Other minor side effects commonly associated with SPA use include gastritis/dyspepsia and dry skin (122, 133). SPA should be taken with food, as this increases its absorption and reduces its potential for gastritis. In addition, the development of irregular ovulation and menstruation in women who previously had normal ovulatory function can occur (122, 133). The incidence of metrorrhagia or polymenorrhea in these patients may be decreased by administering SPA days 4 through 21 of the menstrual cycle (133) or by combining the medication with an OCP (see above) (130, 131). Minimization of other side effects may be also achieved by slowly increasing the dose by 25-mg increments, to a maximum of 100–300 mg/day. Before beginning therapy, patients should be educated regarding potential side effects and reassured that most will be temporary at best. While daily doses of 100 mg/day are generally effective for the treatment of hirsutism, higher doses (200 to 300 mg/day) may be preferable in very hirsute or markedly obese women.
Absolute contraindications to SPA use include renal insufficiency, anuria, chronic renal impairment, hyperkalemia, pregnancy, and abnormal uterine bleeding. The carcinogenic potential of SPA has been long debated. The package insert states that chronic toxicity studies in rats, using 25 to 250 times the usual human dose (on a body weight basis), resulted in benign adenomas of the thyroid and testes, malignant mammary tumors, and hepatic changes. Because of these findings in the rat, it has been recommended that SPA not be given to women with a genetic predisposition to breast cancer (134). SPA and its metabolites cross the placental barrier, and studies in rats indicate feminization of the male fetus. Therefore, SPA should be used in conjunction with an effective means of contraception.
2. Flutamide. Flutamide, a nonsteroidal antiandrogen, is available in the United States as adjunctive therapy for the treatment of prostate cancer. It is considered a pure androgen receptor blocker, although at high doses it may also reduce the synthesis of androgens or increase their metabolism (135). 2-Hydroxyflutamide is the principal metabolite of this drug which inhibits DHT binding to the androgen receptor. Flutamide has been found to be as effective or more effective than SPA for the treatment of hirsutism (82, 123, 124, 136, 137). In their study of patients with well defined IH, Couzinet and colleagues (82) noted a progressive decrease in the hirsutism score throughout 12 months of treatment with flutamide, 500 mg/day. None of the 10 patients experienced significant side effects. Nonetheless, there have been concerns regarding the possibility of serious adverse events, primarily hepatotoxicity (138, 139).
In the 5 yr after the marketing of flutamide, the U.S. Food and Drug Administration received reports of 20 deaths and 26 hospitalizations for flutamide-induced hepatotoxicity (139). While the vast majority of these were men, one of the hospitalized patients was a 45-yr-old female being given higher than standard doses of the drug (1,000 mg/day) for the treatment of alopecia and oily skin (139). Hence, we recommend monitoring serum markers of hepatic function at regular intervals during therapy. Other lesser side effects of flutamide include dry skin and a greenish tint to the urine. While doses of 250 mg twice daily are generally used, a single dose of 250 mg/day may be effective in some patients (137). Finally, in animal models, flutamide has a feminizing effect in the male fetus (140, 141), and nonsterilized patients of reproductive age should be advised to use effective contraception.
3. Cyproterone acetate. Cyproterone acetate is a 17-hydroxyprogesterone acetate derivative with strong progestogenic properties, similar in potency to megesterol acetate (142, 143). The contraceptive properties of CPA occur in part due to gonadotropin suppression. CPA is also categorized as an antiandrogen, since it competes with DHT for binding to the androgen receptor. It produces a decrease in circulating T and androstenedione levels through a reduction in circulating LH and has been used as an effective treatment for hirsutism (144). In a dose ranging study, Barth and colleagues (145) noted that an OCP containing 35 mg EE and 2 mg CPA per day was as effective in reducing hirsutism as the same OCP with the addition of 20 mg/day or 100 mg/day of CPA for the first 10 days of the cycle. Cyproterone acetate in doses of 50–100 mg/day, combined with 30–35 mg EE, is as effective as the combination of SPA, 100 mg/day, and an OCP in the treatment of hirsutism (146, 147). In another study a triphasic OCP containing CPA, 12.5 mg/day for the first 10 days of the cycle (plus varying doses of EE), was found to be as effective as flutamide, 250 mg/day (148). In contrast, an OCP containing CPA, 2 mg/day, in combination with EE, 35 mg/day, seems less effective than 100 mg/day of SPA (149).
Few studies investigating the efficacy of CPA in patients with IH are available. Jasonni and colleagues (150) treated 11 subjects with IH, defined by normal T and DHS levels and ovulatory cycles, and 15 women with PCOS with CPA, 50 mg/day, from days 1–15 in combination with 0.1 mg/day transdermal 17ß-estradiol. They reported that the hirsutism score fell by 50% over 6 months, although in reporting their results the investigators did not differentiate between women with IH and PCOS. Peereboom-Wynia and Boekhorst (151) treated another 11 well defined women with IH using CPA,100–200 mg, on days 5–15 and EE, 50 µg, on days 5–26 of the cycle, and reported a reduction in the hair density and diameter (151). Overall, it would appear that CPA is effective in treating patients with IH, although larger studies are still lacking.
This drug is currently not available in the U.S. but has been used for many years in other countries, sold as the OCP Diane-35 or Dianette (2 mg CPA and 35 µg EE/tablet, Schering AG, Germany) or Androcur (50 mg/tablet, Schering AG, Germany). Side effects may include loss of libido. Adrenal insufficiency is a rare complication of CPA therapy, seen primarily in children receiving high doses for the treatment of precocious puberty, and is unlikely to occur in adults (152). As with all antiandrogens, adequate contraception must be used to avoid the possibility of feminizing a male fetus, particularly when CPA is used at low doses in intermittent cycles (e.g., 2 mg/day for the first 10 days of the cycle).
C. 5-RA inhibitors
Since preliminary data suggest that IH is determined by increased
peripheral activity of 5-RA, ideal therapy should include agents
able to inhibit the activity of this enzyme. One such agent,
finasteride, is currently available in the United States for the
treatment of men with androgenetic alopecia (Propecia, 1 mg
finasteride, Merck & Co., Inc., West Point, PA) and benign prostate
hyperplasia (Proscar, 5 µg finasteride, Merck & Co., Inc.), and has
been found to be effective in the treatment of IH (97, 98). Finasteride
primarily inhibits type 2 5-RA activity (153).
In a 9-month prospective randomized trial, Erenus and colleagues (98)
noted a greater decrease in hirsutism score in 20 patients with IH
treated with SPA, 100 mg/day, compared with 20 patients treated with
finasteride, 5 mg/day (a 42 ± 12% vs. 15 ± 15%
decrease, respectively). Other investigators have also suggested that
finasteride is somewhat less effective than the antiandrogens for the
treatment of hirsutism. Venturoli and colleagues (148) compared the
effects of finasteride (5 mg/day) on hirsutism to flutamide (250
mg/day), ketoconazole (300 mg/day), and CPA (12.5 mg for the first 10
days of each cycle, plus doses of EE ranging from 0.01 mg/day to 0.02
mg/day) in 66 hirsute women. Flutamide and the CPA-estrogen combination
had efficacies similar to that of ketoconazole (-55%, -60%, and
-53%, respectively), which was modestly, but significantly, better
than that of finasteride (-44%). Although a significant number of
side effects were observed with ketoconazole, few problems were noted
with the remaining three treatments, in particular finasteride.
Overall, finasteride may be a useful candidate for treating women with
hirsutism at a dose of 5 mg/day. In fact, choosing between the various
antiandrogens/5-RA inhibitors available may depend more on side
effects and patient tolerance than on specific drug efficacy.
A 5-RA inhibitor still in clinical testing is dutasteride (GI198745,
GlaxoWellcome Co., Research Triangle Park, NC), a "dual" type 1 and
type 2 5-RA inhibitor. It is thought that this compound, although
similar in structure to finasteride, will be a more potent inhibitor.
Overall, it appears to suppress the type 2 isoenzyme 2–3 times more
strongly than finasteride, as well as inhibiting the type 1 form of
5-RA (154). This compound effectively inhibits DHT production by
99% approximately 24 h after oral administration and may be
potentially useful for the treatment of androgenetic alopecia, acne,
and hirsutism. Finally, it must be remembered that all of these 5-RA
agents have the potential of feminizing a male fetus (141, 155). All
women of reproductive age using these drugs must use effective
contraception.
D. Biological modifiers of hair follicular growth
Eflornithine hydrochloride is a new agent (Vaniqa,
DFMO/eflornithine 15% cream, Bristol Myers-Squibb Co., Buffalo, NY)
that has completed clinical phase III testing as a topical cream for
decreasing or arresting facial hair growth in women. Although this
agent is thought to affect hair follicle growth by inhibiting keratin
protein synthesis (156, 157, 158), the exact mechanism(s) of action and the
degree of efficacy are still inconclusive.
E. Mechanical control of hirsutism
In general, mechanical means of controlling, removing, or
destroying unwanted hairs should be considered complementary to medical
management in the treatment of the hirsute. Various options are
available including shaving, depilation, electrology, and laser
epilation, as follows.
1. Shaving and depilation. Plucking, waxing, or shaving facial and body sites are common means of hair removal for many women. However, these cosmetic methods of hair removal can irritate the skin, possibly resulting in folliculitis, pseudofolliculitis, and ingrown hairs. Although the majority of reports studying the effect of hair removal methods on hair growth rates have used male subjects (for review see Ref. 159), Peereboom-Wynia (159) studied the impact of various shaving and depilating methods on thigh hair growth in 15 women with IH. Overall, neither shaving nor depilation performed either once or over a series of 12 treatments had any effect on hair growth. Unfortunately, in this study IH was defined as hirsutism without obvious cause, with 20% of subjects having irregular cycles and circulating androgen levels that were not assessed. Further studies of the impact of mechanical means of hair removal on the growth of facial and body hair in IH and other hirsute women are needed.
2. Electrology. The role of electrologists in the management of patients with hirsutism must not be overlooked. Electrolysis (i.e., electroepilation) results in long-term, although gradual, hair destruction (160, 161). With repeated treatments, efficacy ranges from 15 to 50% permanent hair loss (162). Scarring can occur after electrolysis, especially if the procedure is inexpertly performed (163). It should be recognized that electrologists are often the first individuals to whom the hirsute patient turns for assistance. In one study of 779 consecutive new clients seeking electrology, 40% were noted to have potential risk factors for hyperandrogenism according to their response to a standardized questionnaire (164). Approximately 20% of women evaluated had a hirsutism score greater than 6, while PCOS was evident in more than 50%. Importantly, only 26% of clients referred for a free medical evaluation actually followed through, an indication of the high level of ignorance regarding the availability of medical treatment among hirsute women. Thus, it behooves all physicians caring for these women to establish and maintain communication with the electrology profession, referring patients for electroepilation to those who are properly trained. It should also be noted that certification and training requirements for electrology vary widely from state to state.
3. Laser epilation.Directed damage to hair follicles based on the theory of selective photothermolysis has been reported recently (165, 166). Various lasers have been evaluated for their effectiveness in treating body hair. High energy ruby light was shown to be effective as a tool for removal of unwanted hair growth in 1979 and was introduced for this clinical use in 1992, with the free running long pulse ruby laser introduced for routine use in 1995. The mode of action has been suggested to be selective hair thermolysis, which implies that light is selectively absorbed in certain structures of the FU. In recent years the 694-nm ruby lasers, the 1064-nm Q-switched Nd:YAG laser, the 755 nm long-pulsed alexandrite, and the 800-nm diode laser have been introduced for the removal of unwanted hair. For all instruments the principal mechanism underlying hair destruction has been photothermolysis of the darker pigmented hairs. Gray and fair-colored hair should not
