| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
EXTENSIVE PERSONAL EXPERIENCE |
Departments of Obstetrics and Gynecology (R.A., L.A.S., E.S.K., C.M., J.L., K.C.S., K.T., L.R.B.) and Medicine (R.A.), University of Alabama at Birmingham, Birmingham, Alabama 35233; Caracas Fertility Center (L.A.S.), Caracas, Federal District, Venezuela 1050; and the Health Research Council (C.M.), Mexican Institute of Social Security, Federal District, Mexico City, Mexico
Address all correspondence and requests for reprints to: Ricardo Azziz, M.D., M.P.H., M.B.A., Department of Obstetrics and Gynecology, Cedars-Sinai Medical Center, 8635 West Third Street, Suite 160 W, Los Angeles, California 90048. E-mail: azzizr{at}cshs.org.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
All patients presenting for evaluation of symptoms potentially related to androgen excess between October 1987 and June 2002 were evaluated, and the data were maintained prospectively in a computerized database. For the assessment of therapeutic response, a retrospective review of the medical chart was performed, after the exclusion of those patients seeking fertility therapy only, or with inadequate follow-up or poor compliance.
A total of 1281 consecutive patients were seen during the study period. Excluded from analysis were 408 patients in whom we were unable to evaluate hormonal status, determine ovulatory status, or find any evidence of androgen excess. In the remaining population of 873 patients, the unbiased prevalence of androgen-secreting neoplasms was 0.2%, 21-hydroxylase-deficient classic adrenal hyperplasia (CAH) was 0.6%, 21-hydroxylase-deficient nonclassic adrenal hyperplasia (NCAH) was 1.6%, hyperandrogenic insulin-resistant acanthosis nigricans (HAIRAN) syndrome was 3.1%, idiopathic hirsutism was 4.7%, and polycystic ovary syndrome (PCOS) was 82.0%. Fifty-nine (6.75%) patients had elevated androgen levels and hirsutism but normal ovulation. A total of 257 patients were included in the assessment of the response to hormonal therapy. The mean duration of follow-up was 33.5 months (range, 6–155). Hirsutism improved in 86%, menstrual dysfunction in 80%, acne in 81%, and hair loss in 33% of patients. The major side effects noted were irregular vaginal bleeding (16.1%), nausea (13.0%), and headaches (12.6%); only 36.6% of patients never complained of side effects.
In this large study of consecutive patients presenting with clinically evident androgen excess, specific identifiable disorders (NCAH, CAH, HAIRAN syndrome, and androgen-secreting neoplasms) were observed in approximately 7% of subjects, whereas functional androgen excess, principally PCOS, was observed in the remainder. Hirsutism, menstrual dysfunction, or acne, but not alopecia, improved in the majority of patients treated with a combination suppressive therapy; although more than 60% experienced side effects.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Notwithstanding their clinical importance, the prevalence of the different pathological conditions causing or associated with androgen excess remains unclear. Although the most recognizable clinical feature of androgen excess may be hirsutism, it should be noted that not all patients with hirsutism have overt evidence of androgen excess, with some women suffering from what we understand to be IH (4). Alternatively, not all patients with an androgen excess disorder have hirsutism, as in the Asian patient with PCOS (5). However, in most studies, hirsutism has been used as a surrogate marker for androgen excess, with the largest study to date of consecutive patients potentially affected by androgen excess including only subjects presenting with hirsutism and/or androgenic alopecia (6). This has made it difficult to accurately assess the frequency of the lesser common androgen excess disorders.
The response of hirsutism, ovulatory dysfunction, and other features of androgen excess to hormonal therapy is important to determine but has been studied primarily with the use of individual agents. However, combination therapy including oral contraceptives (OCs), antiandrogens, or metformin has been suggested to be superior to monotherapy (7, 8, 9). Unfortunately, reports evaluating the results of these regimens have generally included 50 or fewer patients (7, 8, 9, 10, 11, 12, 13), limiting our assessment of the efficacy of this therapeutic regimen.
The objective of the present study was to report on our experience evaluating over 1000 consecutive patients consulting for symptoms potentially related to androgen excess. From this data, we have estimated the prevalence of the different pathological conditions causing clinically evident androgen excess and further documented the long-term success of suppressive and/or antiandrogen hormonal therapy in the treatment of these patients.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
All patients presenting for the evaluation of symptoms potentially related to androgen excess to the reproductive endocrinology clinic (to R.A.) at the University of Alabama at Birmingham between October 1987 and June 2002 were included. The Institutional Review Board for Human Use at the University of Alabama at Birmingham approved the study. Patients evaluated included those presenting with oligo/amenorrhea, ovulatory dysfunction, excess hair growth, virilization, alopecia, or acne. The data were recorded and maintained prospectively in a computerized database (Alpha Four version 6.0; Alpha Software Corp., Burlington, MA).
Initial subject evaluation
All patients completed a uniform history form and underwent a complete physical examination. The following parameters were recorded prospectively: height, weight, race, age, gravidity, parity, degree of ovulatory/menstrual dysfunction, presence of acne, and hirsutism score. The body mass index (BMI) was calculated as kilograms per square meter. Beginning January 1993, the abdominal and hip circumferences were assessed as published previously (14), and the waist to hip ratio (WHR) was calculated. The presence of acne was recorded, but the severity was not generally scored. Excess body and facial terminal hair growth was assessed using a modified Ferriman-Gallwey (mF-G) hirsutism score (15).
The interval between bleeding episodes was assessed prospectively and classified as less than 26 d, 27–34 d, 35–44 d, 45 d to 3 months, and more than 3 months duration. This classification was established per the reported variations in menstrual cycle length observed by Treloar et al. (16). Patients with cycles greater than 35 d or less than 26 d were deemed to be oligo/anovulatory. In addition, beginning January 1995, ovulatory function was confirmed in all eumenorrheic (i.e. with a bleeding interval of 27–34 d) hirsute women by using the basal body temperature and measuring a cycle d 22–24 progesterone (P4) level. A level of P4 greater than 12.7 nmol/liter (4 ng/ml) was considered to represent ovulation, and eumenorrheic patients were then classified as "27–34 d plus ovulation" or "27–34 d plus oligo-ovulation."
In patients who had not received hormonal therapy for 3 months before their initial visit, the serum levels of total testosterone (T), free T, dehydroepiandrosterone sulfate (DHEAS), and 17-hydroxyprogesterone (17-HP) were obtained and recorded. Blood sampling for androgens was performed without regard to the time of the cycle or the day, although an effort was made to measure the 17-HP in the preovulatory phase of the menstrual cycle (17). Hyperandrogenemia was defined as an androgen value above the 95th percentile of 98 healthy control women [i.e. a total T 
2.94 nmol/liter (88 ng/dl), free T 0.026 nmol/liter (0.75 ng/dl), or DHEAS 6.64 µmol/liter (2750 ng/ml)], as reported previously (1).
Patients with a basal level of 17-HP greater than 6.0 nmol/liter (2 ng/ml) underwent either a repeat 17-HP test or proceeded directly to an ACTH stimulation test, as described previously (17). If the repeat 17-HP was 6.0 nmol/liter or greater, patients underwent an acute ACTH stimulation test (see below). Patients with levels of total T above 8.67 nmol/liter (250 ng/dl) on at least two separate occasions, or who were deemed to demonstrate clinical features suggestive of an ASN (e.g. virilization), underwent a transvaginal sonogram and a computerized tomography scan of the adrenals at 5-mm intervals to exclude ovarian and adrenal neoplasms, respectively (18).
Patients with evidence of ovulatory dysfunction also underwent measurements of serum prolactin and thyroid-stimulating hormone levels to exclude a prolactin-secreting adenoma and thyroid dysfunction, respectively. If clinically suspected, screening for Cushing’s syndrome was performed by either the overnight dexamethasone suppression test (i.e. the measurement of a cortisol level the morning after the administration of 1 mg dexamethasone orally at bedtime) through January 1990 or by measuring the 24-h urine-free cortisol content after that date.
In patients diagnosed with PCOS or HAIRAN syndrome (see below), fasting glucose (GLU) and insulin (INS) levels were obtained regularly after April 1996. Before those dates, basal and or GLU-stimulated INS and GLU levels were obtained selectively, primarily in patients clinically suspected of having INS resistance (e.g. by the presence of acanthosis nigricans on physical examination).
Acute ACTH stimulation test
When indicated (see above), the acute ACTH stimulation test was performed as described previously (17). In brief, all studies were started between 0730 and 0930 h in the fasting state and scheduled d 3–8 after a spontaneous vaginal bleed or after a withdrawal bleed induced by either 100 mg P4 in oil im or 300 mg/d oral micronized P4 for 7 d. Dexamethasone was not administered before the study. Three baseline samples were obtained 15 min apart and mixed (0-min sample). Immediately afterward, 0.25 mg ACTH-(1–24) (Cortrosyn; Organon Co., Orange, NJ) was administered iv over 60 sec, and blood was sampled 60 min later. Both the 0- and 60-min samples were assayed for 17-HP levels. If the stimulated 17-HP level was greater than 10 ng/ml, the patient was considered to have 21-hydroxylase (21-OH)-deficient NCAH (17). The diagnosis of NCAH was confirmed by the genotyping of CYP21, as described previously (17).
Differential diagnosis
Based on the above evaluation, two distinct types of androgen excess disorders, those with specific diagnoses ascertained by inclusion and those diagnosed by exclusion (defined as FAE disorders), were identified:
Specific diagnoses ascertained by inclusion testing: 1) ASNs, diagnosed by resection and histopathology of the tumor; 2) 21-OH CAH, diagnosed by review of prior records indicating evidence of congenital virilization in the presence of dramatically elevated basal levels of 17-HP, generally 90 nmol/liter (30 ng/ml) or greater (19); 3) 21-OH-deficient NCAH, diagnosed by an ACTH-stimulated 17-HP level greater than 10 ng/ml and confirmed by the genotyping of CYP21, as described previously (17); 4) HAIRAN syndrome, diagnosed by a fasting basal INS greater than 80 µIU/ml and/or an INS level during a 3-h oral glucose tolerance test greater than 300 µIU/ml, as described previously (20, 21).
Disorders of FAE, diagnosed by exclusion: 1) PCOS, diagnosed by the presence of ovulatory dysfunction in association with hirsutism and/or elevated androgen levels, after exclusion of the above specific disorders, as recommended by a 1990 National Institutes of Health/National Institute of Child Health and Human Development-sponsored conference on the subject (22); 2) IH, diagnosed by the presence of hirsutism accompanied by normal ovulation and normal androgens levels, after exclusion of the above specific disorders (23); and 3) hyperandrogenemia (HA) plus hirsutism, comprised of individuals with elevated androgen levels and hirsutism but normal ovulation.
Assessment of response to suppressive hormonal therapy
To determine the response to suppressive hormonal therapy in patients with androgen excess, we retrospectively reviewed the charts of all androgen excess patients seen initially between October 1987 and June 2001. Using a uniform form, the documented treatment outcome and side effects as assessed by the patient were recorded. Patients included were those who had PCOS, HAIRAN, IH, or HA plus hirsutism (see above). We excluded those patients: 1) seeking infertility therapy; 2) with inadequate follow-up (i.e. duration of the follow-up visits was less than 6 months, or who did not return after their initial visit); or 3) with poor compliance with treatment or follow-up (i.e. who were noncompliant in taking the medication according to prescribed directions, or who were seen for care at greater than 18-month increments).
All patients with menstrual or ovulatory dysfunction received OCs when possible. Patients with unwanted hair growth and evidence of excess facial or body terminal hair growth received spironolactone (SPA) (200 mg if mF-G 8; 100 mg/d if mF-G = 3–7) in combination with the OC, to minimize the risks of teratogenicity. SPA was rarely used alone, except in the occasional hirsute patient who had previously undergone a hysterectomy or tubal ligation. Other treatment regimens were occasionally used, including glucocorticoids, insulin sensitizers, GnRH analogs, flutamide, finasteride, and other estrogen-progestin combinations, alone or in combination; the majority of these were used as part of clinical trials (24, 25, 26).
When side effects were identified, they were managed as follows. First, if side effects attributable to SPA occurred, the dose of SPA was decreased. On occasion, the SPA was changed to flutamide or finasteride. Second, if persistent breakthrough bleeding (BTB) on the OC occurred, despite short-term supplementation with an oral estrogen, then the OC was changed to one containing 50 µg ethinyl estradiol plus 1 mg ethyndiol diacetate. Third, if other side effects attributable to the OC occurred (e.g. headaches or migraines, weight gain, increased breast size, etc.), then the OC was changed to one containing 20 µg ethinyl estradiol or discontinued altogether. Fourth, women who were hypertensive, smokers more than 35 yr of age, or those who had a previous thromboembolic event on OCs/hormones were considered for cyclic progestogen treatment only, potentially in combination with SPA.
Statistical analysis
INS resistance and ß-cell function were calculated by the homeostatic model assessment method (HOMA-IR and HOMA-%ß-cell, respectively), as described previously (27). Two-group comparison of continuous variables was performed using a two-sample t test with adjustment for nonconstancy of variance, when necessary. More than two group means were compared using the ANOVA with post hoc least squares means pairwise comparisons. All categorical end points were compared using the 
2 statistic or Fisher’s exact test, when appropriate.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
A total of 1281 consecutive patients were seen during the time period of the study. Of these, 408 (31.9%) subjects were excluded from analysis (Table 1
). In 157 (12.3% of total or 38.5% of excluded) patients, we were unable to evaluate hormonal status generally because they were already on hormonal therapy and were unwilling to discontinue the medication for the evaluation. In 113 (8.8% of total or 27.7% of excluded) patients, we were unable to confirm ovulatory status, and 138 (10.8% of total or 33.8% of excluded) women did not have evidence of androgen excess. In total, 873 subjects were included in the study.
|
Clinical features of androgen excess patients included in the study
Of the 873 patients included in the study 5.1, 16.0, 21.8, 23.7, 17.4, 11.1, 3.2, and 1.6% were 15 yr old or younger, 16–20, 21–25, 26–30, 31–35, 36–40, 41–45, and 46 yr old or older, respectively. Clinically, 75.5% had hirsutism and 14.2% had acne, whereas 29.7% complained of infertility. Patients with hirsutism or infertility were slightly older [28.2 ± 7.6 vs. 26.5 ± 7.1 yr (P < 0.005) and 29.4 ± 5.6 vs. 27.1 ± 8.4 yr (P < 0.001), respectively] than women without these complaints. Patients with infertility were also more obese than their non-infertile counterparts (34.0 ± 8.9 vs. 32.5 ± 9.2 kg/m2, P < 0.03, respectively). Alternatively, women with acne were younger and less overweight than those patients than non-acneic patients [23.6 ± 7.5 vs. 28.0 ± 7.0 yr (P < 0.001) and 29.5 ± 8.6 vs. 33.5 ± 9.1 kg/m2 (P < 0.001)].
Oligo-ovulation was present in 770 (88.2%) patients; 86.4% of oligo-ovulatory patients had overt menstrual dysfunction (12 with polymenorrhea and the remainder with oligomenorrhea). One hundred five (13.6%) oligo-ovulatory patients had apparent eumenorrhea until evaluated more closely. Of hirsute women, 35 (5.3%) had apparent eumenorrhea, of which 14 (40%) had ovulatory dysfunction when evaluated by a luteal P4 level. Of the 124 patients with acne, 33.9% (or 4.8% of the total) had acne only, without hirsutism, although all these patients had both hyperandrogenemia and oligo-ovulation.
Overall, 58.2% of patients included were obese (BMI, 30.0 kg/m2), with BMIs as follows: 1.8% were underweight (<19.0 kg/m2), 20.6% were of normal weight (19.0–24.9 kg/m2), 19.4% were preobese (25.0–29.9 kg/m2), 18.7% were mildly obese (30.0–34.9 kg/m2), 17.6% were moderately obese (35.0–39.9 kg/m2), and 21.9% were severely obese (40 kg/m2), according to 1998 World Health Organization and 1999 National Center for Health Statistics/Centers for Disease Control and Prevention criteria (28, 29).
Prevalence of abnormally elevated androgen measures
Overall, 77.8% of patients included in the analysis had hyperandrogenemia. Total T was elevated in 325 (37.9%), free T in 476 (55.5%), and DHEAS in 347 (40.4%) of the 858 subjects included. The prevalence of the different combinations of abnormal androgen levels is depicted in Table 2. Note that approximately one fifth of patients had no overt elevation in androgen levels, whereas another 10% had elevations in all three androgens evaluated.
|
Among the 873 patients entered into the study, five were on thyroid replacement for hypothyroidism at the time of their initial visit, and one additional patient was diagnosed with hypothyroidism at her evaluation (total prevalence of thyroid patients was diagnosed as such during her evaluation, for a total dysfunction of 0.7%). Two patients were receiving bromocriptine for a previous diagnosis of hyperprolactinemia, and one additional patient was diagnosed as such during her evaluation (for a total prevalence of hyperprolactinemia of 0.3%).
Of the 873 patients included in the study (Table 3), 59 (6.76%) had specific identifiable disorders including ASNs, 21-OH-deficient CAH, 21-OH-deficient NCAH, and HAIRAN syndrome diagnosed in 0.23, 0.69, 2.06, and 3.78% of the total population, respectively. In the remainder (814 or 93.24% of the total), disorders of FAE (i.e. of exclusion) were identified including PCOS, IH, and HA plus hirsutism in 82.02, 4.47, and 6.75% of patients, respectively.
|
Comparison of diagnostic groups
In comparing clinical and laboratory features between diagnostic groups, we again excluded patients with CAH and ASN because of their small numbers. We observed (Table 4) that there was significant racial difference between the different diagnoses, with NCAH patients being exclusively of the white race, whereas a greater proportion of HAIRAN patients were black (36.4%).
|
There was no difference between the groups regarding the prevalence of acne, whereas the differences observed in the frequency of oligo-ovulation and hirsutism were primarily related to the differences in the criteria for defining a diagnostic group (e.g. by definition, patients with IH and HA plus hirsutism have hirsutism with normo-ovulation). PCOS patients had the highest and HA plus hirsutism patients the lowest prevalence of infertility, respectively.
When comparing measures of INS action, including fasting glucose, INS, HOMA-IR, and HOMA-%ß-cell, we analyzed only those values obtained after April 1996, when these measures were obtained routinely. As expected, HAIRAN patients had higher fasting INS and HOMA-IR values than all other subjects (Table 4
). Patients with PCOS also had higher INS levels than women with IH or HA plus hirsutism. There were no differences in mean fasting GLU levels or HOMA-%ß-cell values between groups.
Long-term response to hormonal suppressive therapy
Subjects.
To assess the long-term response of androgen excess patients to hormonal suppressive therapy, we reviewed 736 consecutive patient charts. Of these, we excluded 102 patients (13.8% of the total) because we were unable to locate their full medical record. Of the remaining patients, 133 (18.1% of total or 21.0% of those patients in whom the chart was found) were excluded because they sought fertility therapy only. Another 244 (33.1% of total or 38.5% of those patients in whom the chart was found) patients were further excluded from analysis due to inadequate duration of follow-up (n = 172 or 23.4% of total), or for noncompliance with either treatment or follow-up (n = 72 or 9.8% of total). Patients who had inadequate duration of follow-up or were not compliant with either treatment or follow-up were more likely to be black but not different in age, BMI, or initial mF-G score than those subjects who were compliant with therapy and follow-up (Table 5). Overall, among patients with a chart available, 31.7% of black patients were compliant and had adequate follow-up compared with 54.1% of white patients.
|
Two hundred (77.8%) of the patients included in this analysis had PCOS, 27 (10.5%) had hirsutism and hyperandrogenemia but normal ovulatory function, 12 (4.7%) had HAIRAN syndrome, 10 (3.9%) had NCAH, and eight (3.1) had IH. Overall, the small numbers of subjects in some of these categories precluded a separate analysis of therapeutic success rates by diagnostic group. The initial treatment prescribed was OC plus SPA in 181 (70.4%) patients, OC only in 22 (8.6%) patients, SPA only in 30 (11.7%) patients, and other medications or combinations in 24 (9.3%) patients. The mean follow-up period was 33.5 ± 30.1 months (range, 6–129).
Outcome.
Self-reported treatment outcomes are depicted in Table 6. The success rate (i.e. patients reporting that they were better) in women who were compliant with therapy ranged from 80–86% for hirsutism, menstrual dysfunction, and acne. The success rate of treating alopecia, albeit with a small number of patients with this complaint, was significantly lower at 33%. There were no differences in mean BMI or age between patients experiencing improvement in either hirsutism or menstrual dysfunction. A meaningful analysis could not be performed for acne or alopecia due to the small number of subjects in each group.
|
6). The mean initial mF-G score was 10.5 ± 4.1, and the last recorded mF-G score was 4.7 ± 3.4, with a mean net change in mF-G score of 5.9 ± 4.1 in 3.4 ± 2.7 yr. The mean rate of decrease in the mF-G score was -2.8 ± 3.3 per year. Overall, 65.9% of patients had a mF-G score of 5 or less at their last recorded follow-up evaluation. If only those patients who received combination suppressive therapy (OC plus SPA) were analyzed, the mean net change in mF-G score was -6.2 ± 4.1, with a self reported rate of improvement of 89.0%. One hundred one (47.4% of the total) patients with a mF-G score or 6 or greater reported using electrology. Patients who used electrology concomitantly with hormonal suppression had a greater net change in their mF-G score compared with those hirsute women not using electrology (3.0 ± 3.0 vs. -2.6 ± 2.4; P < 0.05).
Side effects.
The most common side effects of hormonal suppressive therapy reported were headache, nausea, and irregular vaginal bleeding (Table 7). Only 93 (36.6%) subjects assessed did not report any side effects. Of patients having BTB, 38.2% stated that the hormonal therapy did not improve or actually worsened their menstrual dysfunction compared with 16.1% of patients without BTB (P < 0.005).
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
30.0 kg/m2), and 20% were severely obese (BMI, 40 kg/m2). Hence, the prevalence of obesity among androgen excess patients in this study is at least 2-fold higher than that of the general population. For example, data from the National Health and Nutrition Examination Survey project indicate that the prevalence of obesity in women 20–34 yr of age of all ethnicities/races was 25.8% in the 1999–2000 survey (30, 31).
In our population, clinically evident androgen excess, namely hirsutism, was observed in 75% of subjects. We should note that the prevalence of hirsutism would have been lower if the proportion of patients of Asian descent (<1% in the studied population) were greater, because these individuals tend to have lower degrees of excess hair growth (5). An improvement in hirsutism with combination hormonal suppression (an OC and SPN) was reported by approximately 90% of patients, and approximately 65% of these women had minimal excess hair growth (i.e. a mF-G score 
5) at their last recorded follow-up examination after a mean treatment period of approximately 3.5 yr. This success rate is consistent with previous reports evaluating antiandrogen therapy, alone or in combination with OCs, for the treatment of hirsutism (32, 33). Our data also indicated that patients who used electrology in addition to hormonal suppression had a greater decrease in the hair growth than those who did not, suggesting that this hair destruction technique should be encouraged as an adjuvant in the management of the hirsute woman.
Acne was the sole complaint or finding in 4.8% of hyperandrogenic patients evaluated, all of whom were both hyperandrogenemic and oligo-ovulatory. We and others have noted previously that the prevalence (34, 35, 36) of hyperandrogenemia among acneic patients is significant and that this clinical feature may be used as a marker for androgen excess, regardless of the patient’s age. However, we should note that our study design does not allow us to estimate the prevalence of androgen excess among acneic patients, because we included only hyperandrogenic patients into our study. The self-reported improvement in acne with hormonal suppression was approximately 80%. Similar success rates have been reported in randomized trials of acneic women treated with an OCP only (37, 38, 39, 40). Although spironolactone has also been found to have a positive effect on acne, a systematic review of published data could not determine the effectiveness of treatment and its value in clinical practice due to the small sample populations involved in previous trials (32).
Scalp hair loss was a complaint in 14 (4.0%) patients. All but one had ovulatory dysfunction, and all were also hirsute. Our data are consistent with that of others, indicating that the prevalence of androgen excess is high in alopetic women, particularly in those with ovulatory dysfunction (41). As for acne, our study design does not allow us to estimate the prevalence of androgen excess among alopetic patients, in general. Hormonal suppression, even in these hyperandrogenic patients, had limited effect on their alopecia, although the response of hirsutism and menstrual dysfunction in these patients was overall similar to that of the study population as a whole. The poor response of androgenic alopecia to androgen blockade and/or hormonal suppression is consistent with the results of a recent randomized trial (42).
Almost 90% of our patients had ovulatory dysfunction, approximately 85% with obvious oligomenorrhea and approximately 1.5% with polymenorrhea (cycles less than 26 d in length). Almost 15% of oligo-ovulatory patients had apparent eumenorrhea until evaluated more closely using a luteal (d 22–24) phase P4 level with or without basal body temperature monitoring. Of hirsute eumenorrheic women (5.3% of the total hirsute population), 40% were actually found to have ovulatory dysfunction when evaluated more closely. These data confirm the high prevalence of ovulatory dysfunction among hirsute women with regular menstrual cycles reported by others and ourselves (23, 43). Thus, direct assessment of ovulatory function should be a routine part of the evaluation of the hirsute woman with apparently regular episodes of vaginal bleeding. The success of hormonal therapy in subjectively improving menstrual dysfunction was approximately 80%. The high rate of BTB (
16%) may have precluded a higher success rate in the treatment of this clinical feature, because patients who experienced BTB were twice as likely as those who did not to state that hormonal therapy did not improve their menstrual dysfunction (38.2 vs. 16.1%).
Overall, approximately 78% of patients included in this study had detectable hyperandrogenemia, indicating that approximately 30% of patients did not have overt evidence of hyperandrogenemia, consistent with previous reports (1, 3, 44, 45). Free T was the most prevalent marker, solely elevated in 15% of patients and in combination with abnormalities of either total T or DHEAS in another approximate 40%. DHEAS also proved to be useful because it was the only laboratory abnormality detected in 17% of patients studied. However, total T was much less useful, the only androgen to be elevated in only 2% of subjects.
Because approximately 25% of our patients are nonhirsute and primarily diagnosed by having laboratory evidence of hyperandrogenemia, our data indicate that the measurement of free T and DHEAS are indicated in the evaluation of the nonhirsute woman suspected of suffering from androgen excess. Nonetheless, a few caveats are evident. First, the diagnostic value of free T is closely related to the assay method used. Recommended methods for the assessment of free T includes equilibrium dialysis (46, 47), calculation of free T from the measurement of SHBG and total T (48, 49), or ammonium sulfate precipitation (50). In general, current direct assays for the measurement of free T have limited value, particularly in the evaluation of the hyperandrogenic woman (46, 47). Second, the impact of the age-related decline in DHEAS levels, suggesting the need for age-related normative values, on the diagnostic value of this metabolite remains unclear (51).
In addition to the evaluation of the patient with uncertain androgenization, some investigators feel that the measurement of total T and DHEAS has some value in the detection of ASNs (52). However, more recent data suggest that the best predictor of these neoplasms is the clinical presentation, because androgen measures may be misleading and have a low positive predictive value (16, 53, 54, 55). In the population studied, both of our patients with ovarian androgen-secreting tumors had total T levels greater than 6.94 nmol/liter (200 ng/dl) when diagnosed. One patient was 65 yr old and had a rapid and sudden onset of severe hyperandrogenic signs and symptoms, suggesting the presence of an ASN, regardless of androgen levels (18). However, the second patient was originally mistaken for having 21-OH-deficient NCAH, and it was only after the circulating total T levels rose persistently to levels above 6.94 nmol/liter despite treatment with combination therapy that the correct diagnosis was suspected. Hence, although the positive predictive value of a total T level greater than 6.94 nmol/liter (200 ng/dl) is approximately 10% (18), this androgen may still have some value in identifying less clinically obvious patients with an ASN.
Thyroid dysfunction was found to be a relatively uncommon abnormality in our hyperandrogenic patients. Ferriman and Purdie (56) reported that none of their 467 hirsute women had a thyroid abnormality, whereas we diagnosed only one of our 873 patients as having hypothyroidism (although five others were already on thyroid replacement before their evaluation). This prevalence is similar or actually less than that reported by other investigators in the general population of women of similar age (57, 58, 59, 60, 61). Hyperprolactinemia was also rare, diagnosed in only one (0.3%) of our patients during their evaluation, a prevalence similar to that found by other investigators (6, 55, 62, 63, 64).
Approximately 7% of our androgen excess patients suffered from specific and definable disorders of inclusion, including ASNs (0.2%), CAH (0.7%), 21-OH-deficient NCAH (1.6%), and the HAIRAN syndrome (3.1%). Whereas ASNs are diagnosed by pathological examination and 21-OH-deficient NCAH by CYP21 genotyping, the diagnosis of HAIRAN syndrome is less clear. Although previous investigators have established diagnostic levels of basal or GLU-stimulated INS levels that are 3- to 5-fold the upper normal limits and some patients may have overt lipodystrophy (20, 21), it may be argued that many of these women simply represent a more metabolically affected group for PCOS. Only more intensive family and genetic studies will elucidate the true etiology and diagnostic criteria for this syndrome.
The remainder of patients evaluated had FAE or disorders of exclusion, including PCOS (82%), IH (4.7%), and hirsute and hyperandrogenemic patients with normal ovulatory function (6.8%). None of our patients were found to have Cushing’s syndrome or adrenal neoplasms. Also, although a few patients were seen with minimal clinical evidence of androgen excess (e.g. acne and/or seborrhea) after the administration of an androgenic drug, none were found to have sufficiently significant clinical and/or biochemical hyperandrogenism to meet the entry criteria into this study.
Notwithstanding significant variations in how the disorders were defined and the ethnicity of the populations studied, our prevalences are generally similar to those of other investigators (6, 55, 62, 63, 64). In these studies, the prevalence of ASNs ranged from 0.6–2.1% (ovarian ASNs, 0.3–1%; adrenal ASNs, 0–2.1%), NCAH from 1–3%, drug-related hyperandrogenism from 0–1%, and Cushings syndrome from 0–1%. The widest variations were found in the prevalences of PCOS (37.8–78%) and IH (15–38.7%), although the relative degrees were in general agreement with those of our study. We should note that previous reports differ from the present study by using criteria for the disorders diagnosed that was ambiguous or is currently outdated, and for not adjusting the prevalence rates reported for the potential bias inherent to the screening methods used.
Comparing the various diagnostic groups, we noted that HAIRAN syndrome patients were younger, more obese, more likely to be black, more hirsute, more INS resistant, and had the lowest SHBG levels compared with the other phenotypes. Patients with PCOS had the second highest degree of obesity and INS resistance and were the most likely to be infertile. Patients with IH were, on average, older and, as expected, had lower total T, free T, and DHEAS levels. Women with NCAH had the highest mean DHEAS and total and free T levels and were exclusively of the white race.
Approximately 7% of androgen excess patients had hirsutism and hyperandrogenemia, but with normal ovulation, and represent a diagnostic conundrum. Compared with PCOS patients, these women had a similar mean age but lower mean BMI, were less likely to be infertile, reflecting their normal ovulatory status; and had lower free T levels despite being more hirsute. They did, however, have higher DHEAS levels than PCOS patients. On average, these patients were slightly younger and had higher DHEAS and free T levels compared with IH patients. However, they had similar mean BMI and obesity rates, measures of INS action, and mF-G scores as patients with IH. These data suggest that patients with hirsutism and hyperandrogenemia, but with normal ovulation, may more closely resemble IH than PCOS patients and may represent individuals with predominantly adrenal androgen excess. Whether this clinical presentation represents a stable phenotype or is a transition form to PCOS remains to be determined.
Although we have discussed the overall results of hormonal therapy above, a few facts are worth highlighting. First, of the 501 patients who were seeking hormonal suppression of their androgenic symptoms and for whom the chart was located, approximately 50% had either less than 6 months of follow-up or were noncompliant with either treatment or follow-up. These patients were more likely to be black but did not differ in the apparent severity of their disorder. Impacting on compliance may be the high rate of side effects with therapy. Although few, if any, serious side effects were observed, only 37% of subjects studied did not report any side effects. Notably, in Alabama, few other practitioner options were available for patients desiring management of their hyperandrogenism. These data suggest that we must identify strategies for improving the level of education and compliance of these women.
The present report has a number of strengths: the large number of patients included; the fact that baseline features were obtained and maintained prospectively; the careful phenotyping implemented, including early recognition of the possibility of HAIRAN syndrome and NCAH and use of a standardized mF-G hirsutism scoring system; the recognition that diagnostic criteria evolved throughout the study; the uniformity of the assessment as performed by one investigator (R.A.); and the recognition that the investigator served a large catchment area. However, the study also has a few notable limitations: the fact that the protocol and definitions were modified over time, consistent with changes in our understanding of the disorders under consideration; the retrospective nature of our assessment of the presence of alopecia and of the therapeutic outcomes; and the increasing reputation of the examiner in this area of study throughout the time of the study, which may have increased the proportion of patients seen with rare causes of androgen excess. Nonetheless, it is unlikely that these limitations would significantly affect the results of the study.
In conclusion, in this large study of consecutive patients, PCOS was observed in approximately 80%, IH in approximately 5%, HAIRAN syndrome in approximately 3%, and NCAH in approximately 1.5% of patients, and ASNs were observed in approximately one of 500 androgen excess patients. The etiology of 7% of women who demonstrated hirsutism and hyperandrogenemia, but normal ovulation, remains to be determined. Over 80% of androgen excess patients compliant with their therapy experienced an improvement in hirsutism, menstrual dysfunction, and acne with a suppressive hormonal regimen, although only 36% were free of side effects. Alternatively, hair loss, although an infrequent complaint in our population, failed to improve in almost 70% of patients affected. Notably, almost 50% of patients did not appear to be compliant with therapy. These data indicate that PCOS is the most frequent cause of androgen excess, with all other remaining causes affecting a minority of patients, and that, with the exception of alopecia, therapeutic success is high in patients who are compliant with hormonal therapy.
|
Footnotes |
|---|
Abbreviations: ASN, Androgen-secreting neoplasm; BMI, body mass index; BTB, breakthrough bleeding; CAH, classic adrenal hyperplasia; DHEAS, dehydroepiandrosterone sulfate; FAE, functional androgen excess; GLU, glucose; HA, hyperandrogenemia; HAIRAN, hyperandrogenic insulin-resistant acanthosis nigricans; HOMA, homeostatic model assessment method; HOMA-%ß-cell, percentage of ß-cell function calculated by HOMA; HOMA-IR, insulin resistance calculated by HOMA; 17-HP, 17-hydroxyprogesterone; IH, idiopathic hirsutism; INS, insulin; mF-G, modified Ferriman-Gallwey; NCAH, non-classic adrenal hyperplasia; OC, oral contraceptive; 21-OH, 21-hydroxylase; PCOS, polycystic ovary syndrome; P4, progesterone; SPA, spironolactone; T, testosterone; WHR, waist to hip ratio.
Received July 1, 2003.
Accepted October 31, 2003.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  B. O. Yildiz, S. Bolour, K. Woods, A. Moore, and R. Azziz Visually scoring hirsutism Hum. Reprod. Update, June 30, 2009; (2009) dmp024v1. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Bidet, C. Bellanne-Chantelot, M.-B. Galand-Portier, V. Tardy, L. Billaud, K. Laborde, C. Coussieu, Y. Morel, C. Vaury, J.-L. Golmard, et al. Clinical and Molecular Characterization of a Cohort of 161 Unrelated Women with Nonclassical Congenital Adrenal Hyperplasia Due to 21-Hydroxylase Deficiency and 330 Family Members J. Clin. Endocrinol. Metab., May 1, 2009; 94(5): 1570 - 1578. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. B d'Alva, G. Abiven-Lepage, V. Viallon, L. Groussin, M. A. Dugue, X. Bertagna, and J. Bertherat Sex steroids in androgen-secreting adrenocortical tumors: clinical and hormonal features in comparison with non-tumoral causes of androgen excess Eur. J. Endocrinol., November 1, 2008; 159(5): 641 - 647. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S Cupisti, N Kajaia, R Dittrich, H Duezenli, M W Beckmann, and A Mueller Body mass index and ovarian function are associated with endocrine and metabolic abnormalities in women with hyperandrogenic syndrome Eur. J. Endocrinol., May 1, 2008; 158(5): 711 - 719. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. A. Martin, R. J. Chang, D. A. Ehrmann, L. Ibanez, R. A. Lobo, R. L. Rosenfield, J. Shapiro, V. M. Montori, and B. A. Swiglo Evaluation and Treatment of Hirsutism in Premenopausal Women: An Endocrine Society Clinical Practice Guideline J. Clin. Endocrinol. Metab., April 1, 2008; 93(4): 1105 - 1120. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. F. Escobar-Morreale, R. Sanchon, and J. L. San Millan A Prospective Study of the Prevalence of Nonclassical Congenital Adrenal Hyperplasia among Women Presenting with Hyperandrogenic Symptoms and Signs J. Clin. Endocrinol. Metab., February 1, 2008; 93(2): 527 - 533. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Azziz Diagnosing the diagnosis: why we must standardize the defining features of polycystic ovary syndrome Ann Clin Biochem, January 1, 2008; 45(1): 3 - 5. [Full Text] [PDF]
|
|
|
|
|
|
B. O. Yildiz, E. S. Knochenhauer, and R. Azziz Impact of Obesity on the Risk for Polycystic Ovary Syndrome J. Clin. Endocrinol. Metab., January 1, 2008; 93(1): 162 - 168. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Kajaia, H. Binder, R. Dittrich, P. G Oppelt, B. Flor, S. Cupisti, M. W Beckmann, and A. Mueller Low sex hormone-binding globulin as a predictive marker for insulin resistance in women with hyperandrogenic syndrome Eur. J. Endocrinol., October 1, 2007; 157(4): 499 - 507. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Luque-Ramirez, F. Alvarez-Blasco, J. I. Botella-Carretero, E. Martinez-Bermejo, M. A. Lasuncion, and H. F. Escobar-Morreale Comparison of Ethinyl-Estradiol Plus Cyproterone Acetate Versus Metformin Effects on Classic Metabolic Cardiovascular Risk Factors in Women with the Polycystic Ovary Syndrome J. Clin. Endocrinol. Metab., July 1, 2007; 92(7): 2453 - 2461. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Duxbury, L. Gallagher, and B. Keevil The Impact of Simultaneous Measurement of Testosterone and Androstenedione in Women with Suspected Androgen Excess Clin. Chem., April 1, 2007; 53(4): 804 - 805. [Full Text] [PDF]
|
|
|
|
|
|
R. Azziz, E. Carmina, D. Dewailly, E. Diamanti-Kandarakis, H. F. Escobar-Morreale, W. Futterweit, O. E. Janssen, R. S. Legro, R. J. Norman, A. E. Taylor, et al. Criteria for Defining Polycystic Ovary Syndrome as a Predominantly Hyperandrogenic Syndrome: An Androgen Excess Society Guideline J. Clin. Endocrinol. Metab., November 1, 2006; 91(11): 4237 - 4245. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Atmaca, F. Tanriverdi, K. Unluhizarci, F. Bayram, and F. Kelestimur Investigation of adrenal functions in patients with idiopathic hyperandrogenemia. Eur. J. Endocrinol., August 1, 2006; 155(2): 307 - 311. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. M. DeUgarte, K. S. Woods, A. A. Bartolucci, and R. Azziz Degree of Facial and Body Terminal Hair Growth in Unselected Black and White Women: Toward a Populational Definition of Hirsutism J. Clin. Endocrinol. Metab., April 1, 2006; 91(4): 1345 - 1350. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Carmina, F. Rosato, A. Janni, M. Rizzo, and R. A. Longo Relative Prevalence of Different Androgen Excess Disorders in 950 Women Referred because of Clinical Hyperandrogenism J. Clin. Endocrinol. Metab., January 1, 2006; 91(1): 2 - 6. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. A. Cataldo, F. Abbasi, T. L. McLaughlin, M. Basina, P. Y. Fechner, L. C. Giudice, and G. M. Reaven Metabolic and ovarian effects of rosiglitazone treatment for 12 weeks in insulin-resistant women with polycystic ovary syndrome Hum. Reprod., January 1, 2006; 21(1): 109 - 120. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. L. Rosenfield Hirsutism N. Engl. J. Med., December 15, 2005; 353(24): 2578 - 2588. [Full Text] [PDF]
|
|
|
|
|
|
R. Zhou, I. M. Bird, D. A. Dumesic, and D. H. Abbott Adrenal Hyperandrogenism Is Induced by Fetal Androgen Excess in a Rhesus Monkey Model of Polycystic Ovary Syndrome J. Clin. Endocrinol. Metab., December 1, 2005; 90(12): 6630 - 6637. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Azziz, C. Marin, L. Hoq, E. Badamgarav, and P. Song Health Care-Related Economic Burden of the Polycystic Ovary Syndrome during the Reproductive Life Span J. Clin. Endocrinol. Metab., August 1, 2005; 90(8): 4650 - 4658. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. L. San Millan, J. I. Botella-Carretero, F. Alvarez-Blasco, M. Luque-Ramirez, J. Sancho, P. Moghetti, and H. F. Escobar-Morreale A Study of the Hexose-6-Phosphate Dehydrogenase Gene R453Q and 11{beta}-Hydroxysteroid Dehydrogenase Type 1 Gene 83557insA Polymorphisms in the Polycystic Ovary Syndrome J. Clin. Endocrinol. Metab., July 1, 2005; 90(7): 4157 - 4162. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
