00n-972x/92/7402-0248$03.00/0 Journal of Clinical Endocrinology and Metabolism Copyright 8 1992 by The Endocrine Society
Vol. 74, No. 2 Printed in U.S.A.
The Role of Adrenocorticotropin Girls with Premature Adrenarche Oligomenorrhea* LYNN
A. HAWKINS?,
FRED
I. CHASALOW,
AND
SANDRA
Testing in Evaluating and Hirsutism/ L. BLETHEN
Department of Pediatrics, Schneider Children$ Hospital of Long Island, Jewish Medical Center, New Hyde Park, New York 11042; and the Long Island Campus for the Albert Einstein College of Medicine, New York. New York 11042
ABSTRACT. To identify biochemical predictors for future development of hirsutism and/or oligomenorrhea (H/O) in girls with premature adrenarche (PA), we performed dexamethasonesuppressed ACTH stimulation tests in girls with PA (n = 46), young women (n = 44) with H/O, and adult women (n = 31). Cortisol, androstenedione, dehydroepiandrosterone, and 17-hydroxyprogesterone were measured. Seven girls with PA (15%) and seven with H/O (16%) had evidence of nonclassical adrenal steroid biosynthetic defects [nonclassical congenital adrenal hyperplasia (NCAH)]. Twenty-five girls with PA (54%) and 28 girls with H/O (64%) had the moderately elevated 17-hydroxyprogesterone response to ACTH that has been reported in obli-
gate heterozygotes for 21-hydroxylase deficiency. There were no clinical features that distinguished the girls with NCAH from the others. ACTH testing is an important tool in distinguishing those girls with PA and H/O who have NCAH. Although we could find no differences in other adrenal steroid hormones that might predict which of the other girls with PA might later develop H/O, black girls comprised a substantially smaller fraction of the population with H/O than of the population with PA (2% us. 26%; x2 = 8.5; P < 0.005). This observation suggests that PA, in blacks who do not have NCAH, is more likely to be a benign condition/than in other ethnic groups. (J Clin Endocrinol Metab 74: 248-253, 1992)
A
DRENARCHE is defined as the development of sexual (pubic or axillary) hair. It is associated with an increase in serum levels of dehydroepiandrosterone sulfate (DHEA-S) and changes in the adrenal response to ACTH (1, 2). These hormonal changes are called biochemical adrenarche. Biochemical adrenarche results from maturational changes in the activities of the adrenal enzymes 3P-hydroxysteroid dehydrogenase (3/3HSD), 17hydroxylase, and 17,20-desmolase. The consequence of these changes in enzyme activity is increased production of adrenal androgens (3-5). In girls, if adrenarche occurs before 8 yr of age, it is considered to be premature adrenarche (PA). In addition to sexual hair, there may be acne and/or axillary odor. Simultaneous with the physical changes of PA, there are changes in adrenal androgen secretion, i.e. increased production of DHEA-S (2-6). In almost all cases, DHEA-
S levels in girls with PA are elevated compared to those in age-matched controls, but similar to those in older children with the same degree of pubic hair development. This observation suggests that the maturational events are similar in both PA and normal adrenarche (2, 3, 5). However, Forest et al. (7) administered DHEA-S to children with delayed adrenarche without inducing the growth of pubic hair. Thus, the increase in serum DHEAS levels, which is characteristic of biochemical adrenarche, is not the immediate cause of pubic hair growth. Previously, we have shown that most girls with PA have 17-hydroxyprogesterone (170HP) secretory responses to ACTH similar to those of obligate heterozygotes for 21-hydroxylase deficiency (8). Child et al. (9) first used the term type 2 response to describe this phenotype (9). The presence of the type 2 response in many girls with PA was puzzling because we could not elicit a history of PA in obligate heterozygotes for 21hydroxylase deficiency (8). Thus, the type 2 genotype itself was not always associated with PA. Other defects in the steroid biosynthetic pathways can be associated with the type 2 phenotype. Rosenfield et al. (10) proposed that the type 2 phenotype is actually a defect in regulation of cytochrome P45Oc17c~, the enzyme that synthesizes the androgens. Thus, the type 2 phenotype need
Received March 25, 1991. Address all correspondence and requests for reprints to: Sandra L. Blethen, M.D., Ph.D., Department of Pediatrics, Health Sciences Center T-11, State University of New York, Stony Brook, New York 117948111. * Presented in part at the 72nd Annual Meeting of The Endocrine Society, Atlanta, GA, June 1990. t Present address: Department of Pediatrics, Community Health Program of Queens-Nassau, 410 Lakeville Road, New Hyde Park, New York 11042. 248
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 15:20 For personal use only. No other uses without permission. . All rights reserved.
ACTH TESTING
not be associated with the 21-hydroxylase deficiency genotype. Hyperandrogenemia in adolescent and adult women is manifested in a variety of ways, including hirsutism and oligomenorrhea (H/O). Some women with H/O also have mild abnormalities in their 170HP responses to ACTH corresponding to the type 2 phenotype (2, 11). Thus, girls with PA have excess production of adrenal androgens and a type 2 phenotype, and young women with H/ 0 have excess production of androgens and a type 2 phenotype. It seems possible that, at least in some girls, PA may be an indication that a girl is at risk of developing H/O. Although the classical defects in adrenal steroid synthetic pathways leading to cortisol are usually symptomatic at birth, many nonclassical defects are detected only after ACTH testing. In addition to the type 2 phenotype with mild overproduction of 170HP, nonclassical forms of congenital adrenal hyperplasia (NCAH) have been observed in subjects with PA and H/O (2, 12, 13). Temeck et al. (14) found that 43% (10 of 23) subjects with PA had evidence for NCAH (7 with 21-hydroxylase deficiency and 3 with 3PHSD deficiency). The prevalence of these metabolic disorders varies with the population studied and has been reported as ranging from l-20% (ll-13,15-17). This study was performed to evaluate two questions. 1) What is the prevalence of NCAH in PA and H/O? 2) Are there clinical or biochemical features that can be used to predict which girls, if any, with PA will go on to develop H/O? To answer these questions, we measured 170HP, androstenedione (A), and DHEA after dexamethasone-suppressed ACTH stimulation testing in girls with PA, adolescent girls with H/O, and normal adult women (AW). If there was a biochemical difference between the two groups of women that was also present in some of the girls with PA, this difference might discriminate girls in whom PA was benign from those in whom it was a precursor for H/O. Subjects
and Methods
Subjects
Consecutive patients who were evaluated in the Pediatric Endocrinology clinic of the Schneider Children’s Hospital from March 1985 to September 1989 with PA (n = 46) and H/O (n = 44) were included in the study. The clinical profiles of these patients are summarized in Tables 1 and 2, respectively. None of the subjects were related. As part of a research protocol approved by the Long Island Jewish Medical Center Human Subjects Review Committee, AW (ages 18-54 yr) who had borne a child with aneuploidy and a control group who had not were evaluated. Subjects in this study had an increased frequency of type 2 response phenotype (18). In other populations, as expected, there was a 15% fre-
IN PA
249
quency of the type 2 phenotype. Informed consent was obtained from all participants. ACTH
stimulation
testing
ACTH testing was performed as previously described (8). Briefly, all subjects took dexamethasone (DEX; subjects 5 yr age, 1.0 mg DEX) at bedtime the evening before the test. The following morning, blood was collected before a rapid iv bolus injection of 250 rg synthetic ACTH (Cortrosyn, Organon, West Orange, NJ); additional blood samples were collected 30, 45, and 60 min post-ACTH injection. All serum samples were stored at -20 C until assayed. The subjects were subdivided on the basis of their responses to ACTH (8, 14). Type 1 response: stimulated
17OHP
level less than 3.5 nmol/L
This value is 3 SD above the mean observed in 40 normal men and women (8, 18). This range was confirmed in the normal group of this study. Type 2 response: stimulated L
17OHP
level between 3.5-20
nmol/
The type 2 response is found in obligate heterozygotes for classical 21-hydroxylase deficiency as well as in 5-15% of the general population (8, 14, 18). NCAH adrenal steroid biosynthetic defects: 21 -hydroxylase deficiency: stimulated 17OHP above 30 nmol/L
The basis for this criterion is more than 2 SD above the range for the type 2 response as determined with our protocol in our laboratory (14). 3f3HSD deficiency: PA subjects, stimulated DHEA above 30 nmol/L and DHEA/A ratio above 13; H/O subjects, stimulated DHEA above 75 nmol/L and DHEAIA ratio above 13
The DHEA response to ACTH in normal prepubertal girls is less than that in normal adult women (8). The levels of DHEA selected as being suggestive of 3/3HSD were 3 times the mean values observed for controls of the same age. Temeck et al. (14) measured both DHEA/A and 17-hydroxypregnenolone (17Preg) to 170HP ratios; the ratios were equivalent. Thus, because we were studying virilization, we chose to use the ratio of androgens (DHEA/A) and selected a ratio of 13 as the cutoff. Although there is no uniform agreement among investigators, this value seems to discriminate girls with at least partial 3PHSD deficiency. Assays
Cortisol was measured with kits purchased from Diagnostic Products Corp. (Los Angeles, CA). Serum 170HP, A, DHEA, and DHEA-S were determined as previously described (8). External quality control standards [Lyphochek Immunoassay Control Serum (Human) Levels 1, 2 and 3, Bio-Rad, Anaheim, CA] were included in each assay. Interassay coefficients of variation were all less than 12%.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 15:20 For personal use only. No other uses without permission. . All rights reserved.
HAWKINS,
250
CHASALOW,
AND BLETHEN
JCE& M.1992 Vol74.No2
1. Clinical profile of girls with PA
TABLE
ACTH response phenotype
[no. @)I The
Age at first reported symptoms h)”
Bone age (z-score)
Axillary
Acne
[no. (%)I
Pubic hair stage”
Hair
Odor
[no. (%)I
[no. (%)I
[no. (%)I -
1
14 (30)'
4.9 f 0.7
+0.86
Type2 25 (54)d
4.7 + 0.5
+0.82 f 0.3
4.6 + 2.3
+0.67 + 1.7
3fiHSD 3’
6.6 f 0.7
+1.01 f 0.7
1lPHSD 18
5.3
7 (50)
5 (36)
8 (57)
12 (86) T-2 2 (14) T-3
11 (44)
13 (52)
10 (40)
1 (4) T-l 20 (80) T-2 4 (16) T-3
0
1 (33)
1 (33)
1 (33) T-2 2 (67) T-3
3 (100)
2 (67)
1 (33)
2 (67) T-2 1 (33) T-3
1 (100)
1 (100)
1 uw
1 (100) T-2
f 0.3
NCAH 7 (15) 21-Hydroxylase 3’
deficiency
+2.33
a Seventeen of the 46 were less than 5 yr of age when symptoms were recognized. The overall age range was 6 months to 7 lo/12 yr. b Stages of adrenarche as described by Tanner (22). c Ethnic distribution: black, n = 4; Northern European, n = 5; Italian, n = 2; Greek, n = 1; Hispanic, n = 1; East Indian, n = 1. d Ethnic distribution: black, n = 8; Northern European, n = 10; Italian, n = 3; Greek, n = 2; Hispanic, n = 1; Ashkenazi Jewish, n = 1. eEthnic distribution: Northern European, n = 1; Italian, n = 1; Greek, n = 1. ‘Ethnic distribution: Northern European, n = 1; Italian, n = 1; Ashkenazi Jewish, n = 1. 8 Ethnic distribution: Italian, n = 1.
TABLE
2. Clinical profile of young women with H/O
ACTH response phenotype
[no. @)I Type 1 9 (2O)b Type2 28 (64)’ NCAH 3flHSD 7 (16)d
Acne
Hirsutism
[no. (%)I
[no. @)I
6 (67) 13 (46)
6 (86)
Age at menarche
Menses” None
Obesity
(yr)
fig
Oligo
(no.)
(no.)
(no.)
5 (56)
12.9 f 0.7
2
6
1
7 (78)
26 (93)
12.3 j, 0.3
8
14
1
11 (39)
4 (67)
12.3 + 0.7
1
5
0
[no. (%)I
7 (100)
Evidence of PC0 by ultrasound 4of5 11 of 21
3
of 7
The girls initially presented with symptoms after menarche and before age 18 yr. a Menses: Reg, regular; Oligo, cycles longer than 35 days; None, more than 1 yr without a menstrual period. b Ethnic distribution: Northern European, n = 4, Ashkenazi Jewish, n = 4, Hispanic, n = 1. c Ethnic distribution: Ashkenazi Jewish, n = 12; Northern European, n = 7; Italian, n = 5; Middle Eastern, n = 2; Hispanic, n = 1; Oriental, n= 1. d Ethnic distribution: black, n = 1; Northern European, n = 2; Italian, n = 4.
Statistics
Results
All results are presented as the mean f SEM. Statistical significance was determined using Students t test. Z-Scores were calculated for comparison of bone ages (BA) in PA patients as follows: Z-score = (actual BA - BA for chronological age)/sD for BA. Bone age determination, normative data on bone age, and bone age SD were made using the method of Greulich and Pyle (19).
DEX suppression and ACTH
stimulation
After overnight DEX suppression, cortisol levels were
less than @-)nmo; t/LIT in . all1, suL)Jects, 1 * 1 lnalcating . 1. I * aaequate 1 , . .^ . e--. . .. I^^\
adrenal suppression before AlYl’H stlmulatlon (ZU). There were no significant differences in baseline or DEX-suppressed levels of 170HP, DHEA, and A (data
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 15:20 For personal use only. No other uses without permission. . All rights reserved.
ACTH
TESTING
not shown). In response to ACTH, DHEA and A were significantly lower in PA subjects (P < 0.02) than in H/ 0 and AW subjects (Table 3); however, the DHEA/A ratio was not different. Of the girls with PA, 54% (25 of 46) had a type 2 170HP response to ACTH, as did 64% (28 of 44) of the girls with H/O. NCAH was diagnosed in some of the girls who presented with PA (n = 7) or H/O (n = 7); 10 subjects had 3fl-HSD, 3 had 21-hydroxylase deficiency, and 1 had DEX-suppressible hypertension without excess compound S (syndrome of apparent mineralocorticoid excess, presumably due to ll@HSD deficiency) (21). The incidence of these treatable causes of hyperandrogenemia was 15% in each group. Pubertal progression of girls with PA
Of the 18 girls with PA who were followed until thelarche, 13 had their first signs of breast development before 11 yr of age, i.e. earlier than 50% of all American girls, 2 matured more than 2 SD earlier, and none matured more than 2 SD later than other American girls (22). Thus, in spite of PA, further pubertal progression occurred within the normal age range, but somewhat earlier than average. DHEA-S
suppression by overnight
DEX treatment
DHEA-S synthesis can occur by either an ACTHdependent or an ACTH-independent pathway. The independent pathway (which cannot be suppressed by DEX) is dominant before adrenarche, whereas the dependent pathway is dominant after adrenarche. The relative contributions can be evaluated by DEX suppression (23). DHEA-S measurements were performed on sera from TABLE
3. Adrenal secretory profiles
Group Girls with PA Type 1 Type 2 NCAH PlOHase 3@HSD 1laHSD H/O Type 1 Type 2 NCAH 36HSD AW Type 1 Type 2
No. 14 25
170HP (nmol/L) 2.8 f 0.1 5.6 f 0.4
3 132.4 + 70.9 4.2 f 0.9 3 1 8.8 9 28
2.5 + 0.2 6.5 + 0.6
7
5.1 + 1.0
17 14
2.9 + 0.1 5.2 + 0.3
DHEA (nmol/L)
A (nmol/L)
10.8 + 2.8 12.8 + 1.7
1.3 + 0.2 1.5 f 0.2
9+2 921
14.2 f 6.2 45.4 + 1.7 28.5
4.7 + 2.4 2.8 + 0.7 2.9
4 fl 17 f 6 10
40.6 f 6.2 36.8 f 3.8
5.1 + 0.7 6.4 + 0.8
9+1 8+1
133.2 + 17.4 8.6 + 1.2 35.7 f 3.5 27.4 f 4.9
4.0 + 0.5 3.9 + 0.6
DHEA/A ratio
17 f 2 9*2 8fl
Data reported are the mean f SEM. ZlOHase, 21-Hydroxylase deficiency; IlfiHSD, syndrome of apparent mineralocorticoid excess presumably due to 1lflHSD deficiency (21).
IN PA
251
38 girls with PA obtained before and after overnight DEX suppression. In the group as a whole, before DEX suppression, DHEA-S was 4.8 + 0.7 pmol/L; after DEX suppression DHEA-S was 2.4 -I- 0.2 pmol/L, P < 0.01. However, DHEA-S was not suppressed by DEX in all of the girls with PA; the range of suppressed to random (S/ R) values obtained was 0.1-2.5 (0.68 + 0.09). In AW, DHEA-S before DEX suppression was 5.1 + 0.9 pmol/ L; DHEA-S after DEX suppression was 3.1 f 0.5 pmol/ L. The mean S/R value for AW was similar to that for PA subjects (0.62 + 0.03). Thus, in most girls with PA, the elevated DHEA-S levels are produced via the ACTHdependent pathway, just as they are in other postadrenarchal AW. To determine if the girls with PA in whom DHEA-S levels were not suppressible were different from the other girls with PA, clinical and biochemical data from the 10 subjects with the lowest S/R ratios were compared to the data from the 10 with the highest S/R ratios. There was no difference in the abundance of pubic hair or the incidence of hirsutism, acne, or obesity between the 2 groups. There was also no significant difference in stimulated 17-OHP, DHEA, A, or DHEA/A ratio between the 2 groups. Skeletal maturation was similar (z-score for the low S/R group, 0.92; z-score for the high S/R group, 1.00). In summary, there was no feature that distinguished girls with PA in whom DHEA-S synthesis was ACTH dependent from those in whom it was still independent. Discussion Incidence of the type 2 phenotype
in PA
In 1985 (8), on the basis of 10 cases of PA, we reported the high incidence of type 2 ACTH response phenotype and suggested that PA might be a precursor state for H/ 0 because both groups had the type 2 phenotype. In this study, of those subjects without NCAH, we confirmed the type 2 phenotype in 25 of 39 girls (64%) with PA and in 28 of 37 young women (76%) with H/O. Thus, we have confirmed our initial observation in an independent population. Incidence of NCAH
in PA and H/O
Of the girls whom we evaluated for PA, 15% had responses to ACTH consistent with a diagnosis of NCAH. There were 3 girls with nonclassical 21-hydroxylase deficiency, 3 with 3PHSD deficiency, and 1 with ll@HSD deficiency. The diagnosis of 3PHSD deficiency was made on the basis of an elevated DHEA response to ACTH and an elevated DHEA/A ratio. The DHEA/A ratio can be used to detect 3/3HSD deficiency just as well as the 17Preg to 170HP ratio because 3pHSD catalyzes
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 15:20 For personal use only. No other uses without permission. . All rights reserved.
252
HAWKINS,
CHASALOW.
the conversion of both DHEA to A and 17Preg to 170HP. We selected the DHEA/A ratio as our index because it seemed more appropriate to measure androgens than progestogens to study the process of virilization. Based on this sample, the 95% confidence limits for the incidence of NCAH in our population were 6% and 26%. At the time of diagnosis, girls with PA and NCAH did not have more severe virilization, more advanced bone ages, or other features that could distinguish them on clinical grounds from other girls with PA. Perhaps, if diagnosis had been delayed, the clinical symptoms would have progressed, and diagnosis could have been made on the basis of the extent of virilization (17). Of the young women with H/O, 16% had NCAH due to 3PHSD deliciency rather than 21-hydroxylase deficiency. Again, clinical evaluation of these women did not reveal features that distinguished those with NCAH from those with other causes of H/O. The high incidence of NCAH was not caused by a high proportion of Ashkenazi Jews in our population, as only 1 of the 14 subjects with NCAH was Jewish. Further, the prevalence of the type 2 response in Ashkenazi Jewish girls with H/O was greater than the prevalence of the type 2 response reported in unaffected Askenazi Jews (75% vs. 22%; x2 = 16.95; P < 0.0001) by others (24). In summary, because progressive virilization, which can be arrested with proper therapy, occurs in untreated girls with NCAH, the diagnostic value of ACTH stimulation testing should not be overlooked or postponed in girls with isolated PA or H/O (17).
Is there a characteristic pattern that can be used to predict which girls with PA will go on to develop H/O?
Based on the incidence of NCAH in the H/O group, girls with PA and NCAH (especially 3PHSD) would certainly seem to be at risk for H/O. The question of whether the girls with PA and type 2 phenotypes are also at increased risk for developing H/O is more complicated. If we had found 1) a significant difference between AW and the H/O group in one of the steroid responses to ACTH, and 2) some girls with PA showed a response to ACTH similar to that seen in H/O, then the latter subgroup might be at risk for developing H/O. Although the type 2 response phenotype was more common in H/O than in AW, there was no significant difference in adrenal androgens between the women with and without H/O in either the type 1 or type 2 response group. Thus, in the absence of NCAH, we could discern no biochemical markers that distinguished those girls with PA who were at risk for developing H/O from girls without such risk. Because of this inability to predict which girls with PA are at risk for developing H/O, we recommend that girls with PA, particularly those with
AND
BLETHEN
JCE & M .1992 Vol74.No2
NCAH or type 2 responses, receive continued follow-up. Another possible difference in girls with PA, which might predict risk of H/O in the future, is the relative roles of the ACTH-dependent and ACTH-independent pathways for DHEA-S secretion (23). However, retention of the ACTH-independent pathway was not associated with any difference in the clinical or biochemical features of PA. When we compared the steroid secretory patterns in PA to those in H/O, DHEA and A were both significantly lower in PA (P < 0.02). The DHEA response to ACTH was similar to that reported for normal prepubertal girls without PA (8). Even those PA girls with DHEA/A ratios indicative of 3PHSD deficiency had DHEA responses to ACTH that were lower than those of older girls with 3PHSD deficiency. Thus, despite the presence of 1) pubic hair, 2) levels of DHEA-S close to adult values, and 3) ACTH-dependent secretion of DHEA-S, girls with PA still have an immature 3PHSD (4). The ethnic background was strikingly different between the girls with PA and those with H/O. Although the percentage was similar for northern Europeans and Italians, there was a highly significant (x2 = 25.9; P < 0.001) difference in the proportion of Ashkenazi Jews and blacks. There are three possible explanations for this finding: 1) ascertainment bias, 2) differential androgen sensitivity, and 3) allelic differences. Whatever the explanation, our data suggest that PA in black girls, who do not have NCAH, may not be a precursor of H/O. In summary, the secretion of DHEA-S, DHEA, and A in women with PA or H/O is not different from that in normal adult women, regardless of whether individuals have the type 1 or type 2 ACTH response phenotype. Previously, when Forest administered DHEA-S to individuals with delayed adrenarche, there was no change in pubic hair development (7). Together, these findings lead us to speculate that perhaps DHEA-S is not the active hormone of adrenarche, but, instead, is a prohormone for the actual active hormone. In support of this hypothesis is a recent study evaluating different antibodies for the measurement of serum DHEA-S levels (25). Although each antibody adequately quantitated DHEA-S in nonvirilized patients or when authentic DHEA-S was added to steroid-free serum, the antibodies detected widely different amounts of DHEA-S in serum from virilized patients. This finding suggests the presence of DHEA-S metabolites or DHEA-S-like steroids in virilized patients. One of these metabolites could be the hormone responsible for PA and H/O. References 1. Rosenfield RJ. Plasma 17-ketosteroids and l7-beta hydroxysteroids in girls with premature development of sexual hair. J Pediatr. 1971;79:260-6.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 15:20 For personal use only. No other uses without permission. . All rights reserved.
ACTH
TESTING
2. Rich BH, Rosenfield RL, Lucky AW, Helke JC, Otto P. Adrenarche: changing adrenal response to adrenocorticotropin. J Clin Endocrinol Metab. 1981;52:1129-36. 3. Korth-Schutz S, Levine LS, New MI. Serum androgens in normal prepubertal and pubertal children and in children with precocious adrenarche. J Clin Endocrinol Metab. 1976;42:117-24. 4. Schiebinger RJ, Albertson BD, Cassorla FG, et al. The developmental changes in plasma adrenal androgens during infancy and adrenarche are associated with changing activities of adrenal microsomal 17-hydroxylase and 17,20-desmolase. J Clin Invest. 1981;67:1177-82. 5. Voutilainen R, Perheentupa J, Apter D 1983 Benign premature adrenarche: clinical features and serum steroid levels. Acta Paediatr Stand. 72:702-11. 6. Ghizzoni L, Virdis R, Ziveri H, et al. Adrenal steroid, cortisol, adrenocorticotropin, and @endorphin responses to human corticotropin-releasing hormone stimulation test in normal children and children with premature pubarche. J Clin Endocrinol Metab. 1989;69:875-80. 7. Forest MG, DePeritti E, David H, Sempe H. L’Adrenarche jouetelle vraiment un role determinant dans le developpement pubertaire. Ann Endocrinol (Paris). 1982;45:465-95. -8. Granoff AB, Chasalow FI, Blethen SL. 17-Hydroxyprogesterone responses to adrenocorticotropin in children with premature adrenarche. J Clin Endocrinol Metab. 1985;60:409-15. 9. Child DF. Bu’lock DE. Hillier VF. Anderson DC. Heteroeeneitv in adrenal steroidogenesis in normal men and women. Clin Endocrino1 (Oxf’). 1979;11:383-9. 10. Rosenfield RL, Barnes RB, Cara JF, Lucky AW. Dysregulation of cytochrome P45Oc17a as the cause of polycystic ovarian syndrome. Fertil Steril. 1990;53:785-91. 11. Lobo RA. Goebelsmann U. Adult manifestation of coneenital adrenal hyperplasia due to incomplete Pl-hydroxylase deficiency mimicking polycystic ovarian disease. Am J Obstet Gynecol. 1980;138:720-6. 12. Chrousos GP, Loriaux DL, Mann DL, Cutler GB. Late-onset 21hydroxylase deficiency mimicking idiopathic hirsutism or polycystic ovarian disease. An allelic variant of congenital virilizing adrenal hyperplasia with a milder enzymatic defect. Ann Intern Med. 1982;96:143-8. 19 Pang S, Lerner AJ, Stoner E, et al. Late-onset adrenal steroid 3/3A”.
IN PA
14. 15. 16.
17.
18. 19. 20. 21. 22. 23. 24. 25.
253
hydroxysteroid dehydrogenase deficiency. I. A cause of hirsutism in pubertal and postpubertal women. J Clin Endocrinol Metab 1985;60:428-39. Temeck JW, Pang S, Nelson C, New MI. Genetic defects of steroidogenesis in premature pubarche. J Clin Endocrinol Metab. 1987;64:609-17. Azziz R, Zacur HA. 21-Hydroxylase deficiency in female hyperandroaenism: screening and diaanosis. J Clin Endocrinol Metab. 198S;69:577-84. Zerah M, Ueshiba H, Wood E, et al. Prevalence of non-classical steroid 21-hydroxylase deficiency based on a morning salivary 17hydroxyprogesterone screening test: a small sample study. J Clin Endocrinol Metab. 1990;70:1662-7. Morris AH, Reiter EO, Geffner ME, Lippe BM, Itami RM, Hayes DM. Absence of nonclassical congenital adrenal hyperplasia in patients with precocious adrenarche. J Clin Endocrinol Metab. 1989;69:709-15. Chasalow FI, Heimler A, Stamberg J, Blethen SL. A hormonal response associated with an increased risk of aneuploid offspring [Abstract]. Am Sot Hum Genet. 1986;39:A56. Greulich WW, Pyle SI. Radiographic atlas of skeletal development of the hand and wrist, 2nd ed. Stanford: Stanford University Press; 196653. Blethen SL, Chasalow FI. Overnight dexamethasone suppression test: normal responses and the diagnosis of Cushing’s syndrome. Steroids. 1989;54:185-93. Ulick S, Tedde R, Mantero F. Pathogenesis of the type 2 variant of the syndrome of apparent mineralocorticoid excess. J Clin Endocrinol Metab. 1990;70:200-6. Tanner JM, Davies PS. Clinical longitudinal standards for height and height velocity for North American children. J Pediatr. 1985;107:317-29. Kreitzer PM, Blethen SL, Festa RS, Chasalow FI. Dehydroepiandrosterone sulfate levels are not suppressible by glucocorticoids before adrenarche. J Clin Endocrinol Metab. 1989;69:1309-11. Sherman SL, Aston CE, Morton NE, Speiser PW, New MI. A segregation and linkage study of classical and nonclassical 21hydroxylase deficiency. Am J Hum Genet. 1988;42:830-8. Chasalow FI, Blethen SL, Duckett D, Zeitlin S, Greenfield J. Serum levels of dehydroepiandrosterone sulfate as determined by commercial kits and reagents. Steroids. 1989;54:373-83.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 15:20 For personal use only. No other uses without permission. . All rights reserved.
Vol. 74, No. 2 Printed in U.S.A.
The Role of Adrenocorticotropin Girls with Premature Adrenarche Oligomenorrhea* LYNN
A. HAWKINS?,
FRED
I. CHASALOW,
AND
SANDRA
Testing in Evaluating and Hirsutism/ L. BLETHEN
Department of Pediatrics, Schneider Children$ Hospital of Long Island, Jewish Medical Center, New Hyde Park, New York 11042; and the Long Island Campus for the Albert Einstein College of Medicine, New York. New York 11042
ABSTRACT. To identify biochemical predictors for future development of hirsutism and/or oligomenorrhea (H/O) in girls with premature adrenarche (PA), we performed dexamethasonesuppressed ACTH stimulation tests in girls with PA (n = 46), young women (n = 44) with H/O, and adult women (n = 31). Cortisol, androstenedione, dehydroepiandrosterone, and 17-hydroxyprogesterone were measured. Seven girls with PA (15%) and seven with H/O (16%) had evidence of nonclassical adrenal steroid biosynthetic defects [nonclassical congenital adrenal hyperplasia (NCAH)]. Twenty-five girls with PA (54%) and 28 girls with H/O (64%) had the moderately elevated 17-hydroxyprogesterone response to ACTH that has been reported in obli-
gate heterozygotes for 21-hydroxylase deficiency. There were no clinical features that distinguished the girls with NCAH from the others. ACTH testing is an important tool in distinguishing those girls with PA and H/O who have NCAH. Although we could find no differences in other adrenal steroid hormones that might predict which of the other girls with PA might later develop H/O, black girls comprised a substantially smaller fraction of the population with H/O than of the population with PA (2% us. 26%; x2 = 8.5; P < 0.005). This observation suggests that PA, in blacks who do not have NCAH, is more likely to be a benign condition/than in other ethnic groups. (J Clin Endocrinol Metab 74: 248-253, 1992)
A
DRENARCHE is defined as the development of sexual (pubic or axillary) hair. It is associated with an increase in serum levels of dehydroepiandrosterone sulfate (DHEA-S) and changes in the adrenal response to ACTH (1, 2). These hormonal changes are called biochemical adrenarche. Biochemical adrenarche results from maturational changes in the activities of the adrenal enzymes 3P-hydroxysteroid dehydrogenase (3/3HSD), 17hydroxylase, and 17,20-desmolase. The consequence of these changes in enzyme activity is increased production of adrenal androgens (3-5). In girls, if adrenarche occurs before 8 yr of age, it is considered to be premature adrenarche (PA). In addition to sexual hair, there may be acne and/or axillary odor. Simultaneous with the physical changes of PA, there are changes in adrenal androgen secretion, i.e. increased production of DHEA-S (2-6). In almost all cases, DHEA-
S levels in girls with PA are elevated compared to those in age-matched controls, but similar to those in older children with the same degree of pubic hair development. This observation suggests that the maturational events are similar in both PA and normal adrenarche (2, 3, 5). However, Forest et al. (7) administered DHEA-S to children with delayed adrenarche without inducing the growth of pubic hair. Thus, the increase in serum DHEAS levels, which is characteristic of biochemical adrenarche, is not the immediate cause of pubic hair growth. Previously, we have shown that most girls with PA have 17-hydroxyprogesterone (170HP) secretory responses to ACTH similar to those of obligate heterozygotes for 21-hydroxylase deficiency (8). Child et al. (9) first used the term type 2 response to describe this phenotype (9). The presence of the type 2 response in many girls with PA was puzzling because we could not elicit a history of PA in obligate heterozygotes for 21hydroxylase deficiency (8). Thus, the type 2 genotype itself was not always associated with PA. Other defects in the steroid biosynthetic pathways can be associated with the type 2 phenotype. Rosenfield et al. (10) proposed that the type 2 phenotype is actually a defect in regulation of cytochrome P45Oc17c~, the enzyme that synthesizes the androgens. Thus, the type 2 phenotype need
Received March 25, 1991. Address all correspondence and requests for reprints to: Sandra L. Blethen, M.D., Ph.D., Department of Pediatrics, Health Sciences Center T-11, State University of New York, Stony Brook, New York 117948111. * Presented in part at the 72nd Annual Meeting of The Endocrine Society, Atlanta, GA, June 1990. t Present address: Department of Pediatrics, Community Health Program of Queens-Nassau, 410 Lakeville Road, New Hyde Park, New York 11042. 248
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 15:20 For personal use only. No other uses without permission. . All rights reserved.
ACTH TESTING
not be associated with the 21-hydroxylase deficiency genotype. Hyperandrogenemia in adolescent and adult women is manifested in a variety of ways, including hirsutism and oligomenorrhea (H/O). Some women with H/O also have mild abnormalities in their 170HP responses to ACTH corresponding to the type 2 phenotype (2, 11). Thus, girls with PA have excess production of adrenal androgens and a type 2 phenotype, and young women with H/ 0 have excess production of androgens and a type 2 phenotype. It seems possible that, at least in some girls, PA may be an indication that a girl is at risk of developing H/O. Although the classical defects in adrenal steroid synthetic pathways leading to cortisol are usually symptomatic at birth, many nonclassical defects are detected only after ACTH testing. In addition to the type 2 phenotype with mild overproduction of 170HP, nonclassical forms of congenital adrenal hyperplasia (NCAH) have been observed in subjects with PA and H/O (2, 12, 13). Temeck et al. (14) found that 43% (10 of 23) subjects with PA had evidence for NCAH (7 with 21-hydroxylase deficiency and 3 with 3PHSD deficiency). The prevalence of these metabolic disorders varies with the population studied and has been reported as ranging from l-20% (ll-13,15-17). This study was performed to evaluate two questions. 1) What is the prevalence of NCAH in PA and H/O? 2) Are there clinical or biochemical features that can be used to predict which girls, if any, with PA will go on to develop H/O? To answer these questions, we measured 170HP, androstenedione (A), and DHEA after dexamethasone-suppressed ACTH stimulation testing in girls with PA, adolescent girls with H/O, and normal adult women (AW). If there was a biochemical difference between the two groups of women that was also present in some of the girls with PA, this difference might discriminate girls in whom PA was benign from those in whom it was a precursor for H/O. Subjects
and Methods
Subjects
Consecutive patients who were evaluated in the Pediatric Endocrinology clinic of the Schneider Children’s Hospital from March 1985 to September 1989 with PA (n = 46) and H/O (n = 44) were included in the study. The clinical profiles of these patients are summarized in Tables 1 and 2, respectively. None of the subjects were related. As part of a research protocol approved by the Long Island Jewish Medical Center Human Subjects Review Committee, AW (ages 18-54 yr) who had borne a child with aneuploidy and a control group who had not were evaluated. Subjects in this study had an increased frequency of type 2 response phenotype (18). In other populations, as expected, there was a 15% fre-
IN PA
249
quency of the type 2 phenotype. Informed consent was obtained from all participants. ACTH
stimulation
testing
ACTH testing was performed as previously described (8). Briefly, all subjects took dexamethasone (DEX; subjects 5 yr age, 1.0 mg DEX) at bedtime the evening before the test. The following morning, blood was collected before a rapid iv bolus injection of 250 rg synthetic ACTH (Cortrosyn, Organon, West Orange, NJ); additional blood samples were collected 30, 45, and 60 min post-ACTH injection. All serum samples were stored at -20 C until assayed. The subjects were subdivided on the basis of their responses to ACTH (8, 14). Type 1 response: stimulated
17OHP
level less than 3.5 nmol/L
This value is 3 SD above the mean observed in 40 normal men and women (8, 18). This range was confirmed in the normal group of this study. Type 2 response: stimulated L
17OHP
level between 3.5-20
nmol/
The type 2 response is found in obligate heterozygotes for classical 21-hydroxylase deficiency as well as in 5-15% of the general population (8, 14, 18). NCAH adrenal steroid biosynthetic defects: 21 -hydroxylase deficiency: stimulated 17OHP above 30 nmol/L
The basis for this criterion is more than 2 SD above the range for the type 2 response as determined with our protocol in our laboratory (14). 3f3HSD deficiency: PA subjects, stimulated DHEA above 30 nmol/L and DHEA/A ratio above 13; H/O subjects, stimulated DHEA above 75 nmol/L and DHEAIA ratio above 13
The DHEA response to ACTH in normal prepubertal girls is less than that in normal adult women (8). The levels of DHEA selected as being suggestive of 3/3HSD were 3 times the mean values observed for controls of the same age. Temeck et al. (14) measured both DHEA/A and 17-hydroxypregnenolone (17Preg) to 170HP ratios; the ratios were equivalent. Thus, because we were studying virilization, we chose to use the ratio of androgens (DHEA/A) and selected a ratio of 13 as the cutoff. Although there is no uniform agreement among investigators, this value seems to discriminate girls with at least partial 3PHSD deficiency. Assays
Cortisol was measured with kits purchased from Diagnostic Products Corp. (Los Angeles, CA). Serum 170HP, A, DHEA, and DHEA-S were determined as previously described (8). External quality control standards [Lyphochek Immunoassay Control Serum (Human) Levels 1, 2 and 3, Bio-Rad, Anaheim, CA] were included in each assay. Interassay coefficients of variation were all less than 12%.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 15:20 For personal use only. No other uses without permission. . All rights reserved.
HAWKINS,
250
CHASALOW,
AND BLETHEN
JCE& M.1992 Vol74.No2
1. Clinical profile of girls with PA
TABLE
ACTH response phenotype
[no. @)I The
Age at first reported symptoms h)”
Bone age (z-score)
Axillary
Acne
[no. (%)I
Pubic hair stage”
Hair
Odor
[no. (%)I
[no. (%)I
[no. (%)I -
1
14 (30)'
4.9 f 0.7
+0.86
Type2 25 (54)d
4.7 + 0.5
+0.82 f 0.3
4.6 + 2.3
+0.67 + 1.7
3fiHSD 3’
6.6 f 0.7
+1.01 f 0.7
1lPHSD 18
5.3
7 (50)
5 (36)
8 (57)
12 (86) T-2 2 (14) T-3
11 (44)
13 (52)
10 (40)
1 (4) T-l 20 (80) T-2 4 (16) T-3
0
1 (33)
1 (33)
1 (33) T-2 2 (67) T-3
3 (100)
2 (67)
1 (33)
2 (67) T-2 1 (33) T-3
1 (100)
1 (100)
1 uw
1 (100) T-2
f 0.3
NCAH 7 (15) 21-Hydroxylase 3’
deficiency
+2.33
a Seventeen of the 46 were less than 5 yr of age when symptoms were recognized. The overall age range was 6 months to 7 lo/12 yr. b Stages of adrenarche as described by Tanner (22). c Ethnic distribution: black, n = 4; Northern European, n = 5; Italian, n = 2; Greek, n = 1; Hispanic, n = 1; East Indian, n = 1. d Ethnic distribution: black, n = 8; Northern European, n = 10; Italian, n = 3; Greek, n = 2; Hispanic, n = 1; Ashkenazi Jewish, n = 1. eEthnic distribution: Northern European, n = 1; Italian, n = 1; Greek, n = 1. ‘Ethnic distribution: Northern European, n = 1; Italian, n = 1; Ashkenazi Jewish, n = 1. 8 Ethnic distribution: Italian, n = 1.
TABLE
2. Clinical profile of young women with H/O
ACTH response phenotype
[no. @)I Type 1 9 (2O)b Type2 28 (64)’ NCAH 3flHSD 7 (16)d
Acne
Hirsutism
[no. (%)I
[no. @)I
6 (67) 13 (46)
6 (86)
Age at menarche
Menses” None
Obesity
(yr)
fig
Oligo
(no.)
(no.)
(no.)
5 (56)
12.9 f 0.7
2
6
1
7 (78)
26 (93)
12.3 j, 0.3
8
14
1
11 (39)
4 (67)
12.3 + 0.7
1
5
0
[no. (%)I
7 (100)
Evidence of PC0 by ultrasound 4of5 11 of 21
3
of 7
The girls initially presented with symptoms after menarche and before age 18 yr. a Menses: Reg, regular; Oligo, cycles longer than 35 days; None, more than 1 yr without a menstrual period. b Ethnic distribution: Northern European, n = 4, Ashkenazi Jewish, n = 4, Hispanic, n = 1. c Ethnic distribution: Ashkenazi Jewish, n = 12; Northern European, n = 7; Italian, n = 5; Middle Eastern, n = 2; Hispanic, n = 1; Oriental, n= 1. d Ethnic distribution: black, n = 1; Northern European, n = 2; Italian, n = 4.
Statistics
Results
All results are presented as the mean f SEM. Statistical significance was determined using Students t test. Z-Scores were calculated for comparison of bone ages (BA) in PA patients as follows: Z-score = (actual BA - BA for chronological age)/sD for BA. Bone age determination, normative data on bone age, and bone age SD were made using the method of Greulich and Pyle (19).
DEX suppression and ACTH
stimulation
After overnight DEX suppression, cortisol levels were
less than @-)nmo; t/LIT in . all1, suL)Jects, 1 * 1 lnalcating . 1. I * aaequate 1 , . .^ . e--. . .. I^^\
adrenal suppression before AlYl’H stlmulatlon (ZU). There were no significant differences in baseline or DEX-suppressed levels of 170HP, DHEA, and A (data
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 15:20 For personal use only. No other uses without permission. . All rights reserved.
ACTH
TESTING
not shown). In response to ACTH, DHEA and A were significantly lower in PA subjects (P < 0.02) than in H/ 0 and AW subjects (Table 3); however, the DHEA/A ratio was not different. Of the girls with PA, 54% (25 of 46) had a type 2 170HP response to ACTH, as did 64% (28 of 44) of the girls with H/O. NCAH was diagnosed in some of the girls who presented with PA (n = 7) or H/O (n = 7); 10 subjects had 3fl-HSD, 3 had 21-hydroxylase deficiency, and 1 had DEX-suppressible hypertension without excess compound S (syndrome of apparent mineralocorticoid excess, presumably due to ll@HSD deficiency) (21). The incidence of these treatable causes of hyperandrogenemia was 15% in each group. Pubertal progression of girls with PA
Of the 18 girls with PA who were followed until thelarche, 13 had their first signs of breast development before 11 yr of age, i.e. earlier than 50% of all American girls, 2 matured more than 2 SD earlier, and none matured more than 2 SD later than other American girls (22). Thus, in spite of PA, further pubertal progression occurred within the normal age range, but somewhat earlier than average. DHEA-S
suppression by overnight
DEX treatment
DHEA-S synthesis can occur by either an ACTHdependent or an ACTH-independent pathway. The independent pathway (which cannot be suppressed by DEX) is dominant before adrenarche, whereas the dependent pathway is dominant after adrenarche. The relative contributions can be evaluated by DEX suppression (23). DHEA-S measurements were performed on sera from TABLE
3. Adrenal secretory profiles
Group Girls with PA Type 1 Type 2 NCAH PlOHase 3@HSD 1laHSD H/O Type 1 Type 2 NCAH 36HSD AW Type 1 Type 2
No. 14 25
170HP (nmol/L) 2.8 f 0.1 5.6 f 0.4
3 132.4 + 70.9 4.2 f 0.9 3 1 8.8 9 28
2.5 + 0.2 6.5 + 0.6
7
5.1 + 1.0
17 14
2.9 + 0.1 5.2 + 0.3
DHEA (nmol/L)
A (nmol/L)
10.8 + 2.8 12.8 + 1.7
1.3 + 0.2 1.5 f 0.2
9+2 921
14.2 f 6.2 45.4 + 1.7 28.5
4.7 + 2.4 2.8 + 0.7 2.9
4 fl 17 f 6 10
40.6 f 6.2 36.8 f 3.8
5.1 + 0.7 6.4 + 0.8
9+1 8+1
133.2 + 17.4 8.6 + 1.2 35.7 f 3.5 27.4 f 4.9
4.0 + 0.5 3.9 + 0.6
DHEA/A ratio
17 f 2 9*2 8fl
Data reported are the mean f SEM. ZlOHase, 21-Hydroxylase deficiency; IlfiHSD, syndrome of apparent mineralocorticoid excess presumably due to 1lflHSD deficiency (21).
IN PA
251
38 girls with PA obtained before and after overnight DEX suppression. In the group as a whole, before DEX suppression, DHEA-S was 4.8 + 0.7 pmol/L; after DEX suppression DHEA-S was 2.4 -I- 0.2 pmol/L, P < 0.01. However, DHEA-S was not suppressed by DEX in all of the girls with PA; the range of suppressed to random (S/ R) values obtained was 0.1-2.5 (0.68 + 0.09). In AW, DHEA-S before DEX suppression was 5.1 + 0.9 pmol/ L; DHEA-S after DEX suppression was 3.1 f 0.5 pmol/ L. The mean S/R value for AW was similar to that for PA subjects (0.62 + 0.03). Thus, in most girls with PA, the elevated DHEA-S levels are produced via the ACTHdependent pathway, just as they are in other postadrenarchal AW. To determine if the girls with PA in whom DHEA-S levels were not suppressible were different from the other girls with PA, clinical and biochemical data from the 10 subjects with the lowest S/R ratios were compared to the data from the 10 with the highest S/R ratios. There was no difference in the abundance of pubic hair or the incidence of hirsutism, acne, or obesity between the 2 groups. There was also no significant difference in stimulated 17-OHP, DHEA, A, or DHEA/A ratio between the 2 groups. Skeletal maturation was similar (z-score for the low S/R group, 0.92; z-score for the high S/R group, 1.00). In summary, there was no feature that distinguished girls with PA in whom DHEA-S synthesis was ACTH dependent from those in whom it was still independent. Discussion Incidence of the type 2 phenotype
in PA
In 1985 (8), on the basis of 10 cases of PA, we reported the high incidence of type 2 ACTH response phenotype and suggested that PA might be a precursor state for H/ 0 because both groups had the type 2 phenotype. In this study, of those subjects without NCAH, we confirmed the type 2 phenotype in 25 of 39 girls (64%) with PA and in 28 of 37 young women (76%) with H/O. Thus, we have confirmed our initial observation in an independent population. Incidence of NCAH
in PA and H/O
Of the girls whom we evaluated for PA, 15% had responses to ACTH consistent with a diagnosis of NCAH. There were 3 girls with nonclassical 21-hydroxylase deficiency, 3 with 3PHSD deficiency, and 1 with ll@HSD deficiency. The diagnosis of 3PHSD deficiency was made on the basis of an elevated DHEA response to ACTH and an elevated DHEA/A ratio. The DHEA/A ratio can be used to detect 3/3HSD deficiency just as well as the 17Preg to 170HP ratio because 3pHSD catalyzes
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 15:20 For personal use only. No other uses without permission. . All rights reserved.
252
HAWKINS,
CHASALOW.
the conversion of both DHEA to A and 17Preg to 170HP. We selected the DHEA/A ratio as our index because it seemed more appropriate to measure androgens than progestogens to study the process of virilization. Based on this sample, the 95% confidence limits for the incidence of NCAH in our population were 6% and 26%. At the time of diagnosis, girls with PA and NCAH did not have more severe virilization, more advanced bone ages, or other features that could distinguish them on clinical grounds from other girls with PA. Perhaps, if diagnosis had been delayed, the clinical symptoms would have progressed, and diagnosis could have been made on the basis of the extent of virilization (17). Of the young women with H/O, 16% had NCAH due to 3PHSD deliciency rather than 21-hydroxylase deficiency. Again, clinical evaluation of these women did not reveal features that distinguished those with NCAH from those with other causes of H/O. The high incidence of NCAH was not caused by a high proportion of Ashkenazi Jews in our population, as only 1 of the 14 subjects with NCAH was Jewish. Further, the prevalence of the type 2 response in Ashkenazi Jewish girls with H/O was greater than the prevalence of the type 2 response reported in unaffected Askenazi Jews (75% vs. 22%; x2 = 16.95; P < 0.0001) by others (24). In summary, because progressive virilization, which can be arrested with proper therapy, occurs in untreated girls with NCAH, the diagnostic value of ACTH stimulation testing should not be overlooked or postponed in girls with isolated PA or H/O (17).
Is there a characteristic pattern that can be used to predict which girls with PA will go on to develop H/O?
Based on the incidence of NCAH in the H/O group, girls with PA and NCAH (especially 3PHSD) would certainly seem to be at risk for H/O. The question of whether the girls with PA and type 2 phenotypes are also at increased risk for developing H/O is more complicated. If we had found 1) a significant difference between AW and the H/O group in one of the steroid responses to ACTH, and 2) some girls with PA showed a response to ACTH similar to that seen in H/O, then the latter subgroup might be at risk for developing H/O. Although the type 2 response phenotype was more common in H/O than in AW, there was no significant difference in adrenal androgens between the women with and without H/O in either the type 1 or type 2 response group. Thus, in the absence of NCAH, we could discern no biochemical markers that distinguished those girls with PA who were at risk for developing H/O from girls without such risk. Because of this inability to predict which girls with PA are at risk for developing H/O, we recommend that girls with PA, particularly those with
AND
BLETHEN
JCE & M .1992 Vol74.No2
NCAH or type 2 responses, receive continued follow-up. Another possible difference in girls with PA, which might predict risk of H/O in the future, is the relative roles of the ACTH-dependent and ACTH-independent pathways for DHEA-S secretion (23). However, retention of the ACTH-independent pathway was not associated with any difference in the clinical or biochemical features of PA. When we compared the steroid secretory patterns in PA to those in H/O, DHEA and A were both significantly lower in PA (P < 0.02). The DHEA response to ACTH was similar to that reported for normal prepubertal girls without PA (8). Even those PA girls with DHEA/A ratios indicative of 3PHSD deficiency had DHEA responses to ACTH that were lower than those of older girls with 3PHSD deficiency. Thus, despite the presence of 1) pubic hair, 2) levels of DHEA-S close to adult values, and 3) ACTH-dependent secretion of DHEA-S, girls with PA still have an immature 3PHSD (4). The ethnic background was strikingly different between the girls with PA and those with H/O. Although the percentage was similar for northern Europeans and Italians, there was a highly significant (x2 = 25.9; P < 0.001) difference in the proportion of Ashkenazi Jews and blacks. There are three possible explanations for this finding: 1) ascertainment bias, 2) differential androgen sensitivity, and 3) allelic differences. Whatever the explanation, our data suggest that PA in black girls, who do not have NCAH, may not be a precursor of H/O. In summary, the secretion of DHEA-S, DHEA, and A in women with PA or H/O is not different from that in normal adult women, regardless of whether individuals have the type 1 or type 2 ACTH response phenotype. Previously, when Forest administered DHEA-S to individuals with delayed adrenarche, there was no change in pubic hair development (7). Together, these findings lead us to speculate that perhaps DHEA-S is not the active hormone of adrenarche, but, instead, is a prohormone for the actual active hormone. In support of this hypothesis is a recent study evaluating different antibodies for the measurement of serum DHEA-S levels (25). Although each antibody adequately quantitated DHEA-S in nonvirilized patients or when authentic DHEA-S was added to steroid-free serum, the antibodies detected widely different amounts of DHEA-S in serum from virilized patients. This finding suggests the presence of DHEA-S metabolites or DHEA-S-like steroids in virilized patients. One of these metabolites could be the hormone responsible for PA and H/O. References 1. Rosenfield RJ. Plasma 17-ketosteroids and l7-beta hydroxysteroids in girls with premature development of sexual hair. J Pediatr. 1971;79:260-6.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 15:20 For personal use only. No other uses without permission. . All rights reserved.
ACTH
TESTING
2. Rich BH, Rosenfield RL, Lucky AW, Helke JC, Otto P. Adrenarche: changing adrenal response to adrenocorticotropin. J Clin Endocrinol Metab. 1981;52:1129-36. 3. Korth-Schutz S, Levine LS, New MI. Serum androgens in normal prepubertal and pubertal children and in children with precocious adrenarche. J Clin Endocrinol Metab. 1976;42:117-24. 4. Schiebinger RJ, Albertson BD, Cassorla FG, et al. The developmental changes in plasma adrenal androgens during infancy and adrenarche are associated with changing activities of adrenal microsomal 17-hydroxylase and 17,20-desmolase. J Clin Invest. 1981;67:1177-82. 5. Voutilainen R, Perheentupa J, Apter D 1983 Benign premature adrenarche: clinical features and serum steroid levels. Acta Paediatr Stand. 72:702-11. 6. Ghizzoni L, Virdis R, Ziveri H, et al. Adrenal steroid, cortisol, adrenocorticotropin, and @endorphin responses to human corticotropin-releasing hormone stimulation test in normal children and children with premature pubarche. J Clin Endocrinol Metab. 1989;69:875-80. 7. Forest MG, DePeritti E, David H, Sempe H. L’Adrenarche jouetelle vraiment un role determinant dans le developpement pubertaire. Ann Endocrinol (Paris). 1982;45:465-95. -8. Granoff AB, Chasalow FI, Blethen SL. 17-Hydroxyprogesterone responses to adrenocorticotropin in children with premature adrenarche. J Clin Endocrinol Metab. 1985;60:409-15. 9. Child DF. Bu’lock DE. Hillier VF. Anderson DC. Heteroeeneitv in adrenal steroidogenesis in normal men and women. Clin Endocrino1 (Oxf’). 1979;11:383-9. 10. Rosenfield RL, Barnes RB, Cara JF, Lucky AW. Dysregulation of cytochrome P45Oc17a as the cause of polycystic ovarian syndrome. Fertil Steril. 1990;53:785-91. 11. Lobo RA. Goebelsmann U. Adult manifestation of coneenital adrenal hyperplasia due to incomplete Pl-hydroxylase deficiency mimicking polycystic ovarian disease. Am J Obstet Gynecol. 1980;138:720-6. 12. Chrousos GP, Loriaux DL, Mann DL, Cutler GB. Late-onset 21hydroxylase deficiency mimicking idiopathic hirsutism or polycystic ovarian disease. An allelic variant of congenital virilizing adrenal hyperplasia with a milder enzymatic defect. Ann Intern Med. 1982;96:143-8. 19 Pang S, Lerner AJ, Stoner E, et al. Late-onset adrenal steroid 3/3A”.
IN PA
14. 15. 16.
17.
18. 19. 20. 21. 22. 23. 24. 25.
253
hydroxysteroid dehydrogenase deficiency. I. A cause of hirsutism in pubertal and postpubertal women. J Clin Endocrinol Metab 1985;60:428-39. Temeck JW, Pang S, Nelson C, New MI. Genetic defects of steroidogenesis in premature pubarche. J Clin Endocrinol Metab. 1987;64:609-17. Azziz R, Zacur HA. 21-Hydroxylase deficiency in female hyperandroaenism: screening and diaanosis. J Clin Endocrinol Metab. 198S;69:577-84. Zerah M, Ueshiba H, Wood E, et al. Prevalence of non-classical steroid 21-hydroxylase deficiency based on a morning salivary 17hydroxyprogesterone screening test: a small sample study. J Clin Endocrinol Metab. 1990;70:1662-7. Morris AH, Reiter EO, Geffner ME, Lippe BM, Itami RM, Hayes DM. Absence of nonclassical congenital adrenal hyperplasia in patients with precocious adrenarche. J Clin Endocrinol Metab. 1989;69:709-15. Chasalow FI, Heimler A, Stamberg J, Blethen SL. A hormonal response associated with an increased risk of aneuploid offspring [Abstract]. Am Sot Hum Genet. 1986;39:A56. Greulich WW, Pyle SI. Radiographic atlas of skeletal development of the hand and wrist, 2nd ed. Stanford: Stanford University Press; 196653. Blethen SL, Chasalow FI. Overnight dexamethasone suppression test: normal responses and the diagnosis of Cushing’s syndrome. Steroids. 1989;54:185-93. Ulick S, Tedde R, Mantero F. Pathogenesis of the type 2 variant of the syndrome of apparent mineralocorticoid excess. J Clin Endocrinol Metab. 1990;70:200-6. Tanner JM, Davies PS. Clinical longitudinal standards for height and height velocity for North American children. J Pediatr. 1985;107:317-29. Kreitzer PM, Blethen SL, Festa RS, Chasalow FI. Dehydroepiandrosterone sulfate levels are not suppressible by glucocorticoids before adrenarche. J Clin Endocrinol Metab. 1989;69:1309-11. Sherman SL, Aston CE, Morton NE, Speiser PW, New MI. A segregation and linkage study of classical and nonclassical 21hydroxylase deficiency. Am J Hum Genet. 1988;42:830-8. Chasalow FI, Blethen SL, Duckett D, Zeitlin S, Greenfield J. Serum levels of dehydroepiandrosterone sulfate as determined by commercial kits and reagents. Steroids. 1989;54:373-83.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 November 2015. at 15:20 For personal use only. No other uses without permission. . All rights reserved.
