| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Other Original Articles |
Divisions of Endocrinology (M.R.P., J.F.C.) and Adolescent Medicine (M.J.M.), Department of Medicine, Children’s Hospital, Boston, Massachusetts 02115; Reproductive Endocrine Unit (M.R.P., W.F.C., P.A.B.) and Pediatric Endocrine Unit (P.A.B.), Massachusetts General Hospital, Boston, Massachusetts 02114; Biostatistics Center, General Clinical Research Center, Massachusetts General Hospital (D.L.H.), Boston, Massachusetts 02114; and Esoterix Endocrinology, Inc. (D.W.C.), Calabasas Hills, California 91301
Address all correspondence and requests for reprints to: Paul A. Boepple, M.D., Reproductive Endocrine Unit, Bartlett Hall Extension 5, Massachusetts General Hospital, Fruit Street, Boston, Massachusetts 02114. E-mail: boepple.paul{at}mgh.harvard.edu
Abstract
The physical changes that herald the onset of puberty result from
the combination of adrenarche and gonadarche. To examine adrenal
maturation and associated changes in growth without the confounding
effects of changes in the gonadal steroid milieu, we performed a
longitudinal study in 14 young girls with idiopathic central precocious
puberty during long-term pituitary-gonadal suppression. Beginning at
the mean age of 2.9 yr, dehydroepiandrosterone sulfate levels, linear
growth, skeletal maturation, body mass index, and secondary sexual
development were evaluated at 3- to 6-month intervals for up to 12.3
yr. In 12 of the girls, levels of dehydroepiandrosterone,
androstenedione, 17-hydroxypregnenolone, and 17
-hydroxyprogesterone
were determined before and after acute ACTH stimulation every 6 months
to investigate the maturation of adrenal steroidogenic enzyme
activity.
Serum dehydroepiandrosterone sulfate levels rose progressively throughout the study. An exponential model fit the longitudinal datasets well and indicated that dehydroepiandrosterone sulfate levels increased approximately 22%/yr from the youngest age onward. Increasing activity of 17–20 lyase (CYP17) and decreasing activity of 3ß-hydroxysteroid dehydrogenase were also evident in preadrenarchal subjects. When controlled for chronological age, no significant associations were noted between weight, body mass index, or body surface area and dehydroepiandrosterone sulfate levels. However, similar analyses revealed modest correlations of both height and growth velocity with dehydroepiandrosterone sulfate levels.
Our results suggest that adrenarche is not the result of sudden rapid changes in adrenal enzyme activities or adrenal androgen concentrations; rather, adrenarche may be a gradual maturational process that begins in early childhood.
THE PHYSICAL CHANGES of puberty result from a combination of two separate, but overlapping, processes (1, 2, 3, 4): adrenarche (increased production of androgens from the adrenal gland) and gonadarche (increased production of sex steroids from the gonad). Although increasing secretion of GnRH and pituitary gonadotropins underlie gonadarche, the factors that regulate adrenarche remain unknown (5, 6, 7, 8, 9). Much of the previous investigation of adrenarche has involved either the characterization of basal adrenal steroid production during childhood and pubertal development (5, 10, 11, 12, 13, 14) or the adrenal response to stimulation with ACTH (15, 16, 17). With few exceptions (3, 5, 18, 19), the interpretation of such studies has been confounded by the reliance upon cross-sectional data and/or by the coincident changes in the gonadal steroid milieu that accompany gonadarche in normal children.
The study of children treated for idiopathic central precocious puberty (CPP) provides an opportunity to circumvent these limitations. Here we report the longitudinal study of 14 young girls with CPP who were initially evaluated when they were preadrenarchal (average age, 2.9 yr) and were followed prospectively at 6-month intervals for up to 12.3 yr. We focused our analysis on the period during long-term pituitary-gonadal suppression, because this period afforded the unique opportunity to investigate adrenarche without the coincident influences of gonadarche. Baseline and post-ACTH (Cortrosyn) stimulation levels of adrenal steroids were obtained to delineate alterations in adrenal androgen production and steroidogenic enzyme activity that accompany adrenarche. The longitudinal nature of this study and the acquisition of careful auxological data also permitted us to test the hypothesis that adrenarche is influenced by alterations in body size (7, 10, 19, 20, 21, 22, 23, 24).
Subjects and Methods
Subject populations
The diagnosis of CPP was based upon the onset of breast development and/or menses associated with a pulsatile pattern of pituitary gonadotropin secretion and a pubertal response to exogenous GnRH in the absence of any identifiable adrenal or gonadal pathology (25). GnRH agonist (GnRHa) administration was begun after an initial evaluation. To permit the longitudinal study of adrenarche, subjects were included in this cohort if 1) CPP was idiopathic; 2) no confounding conditions existed that might influence growth or adrenal function (e.g. GH deficiency, congenital adrenal hyperplasia, or primary hypothyroidism); and 3) the girls were preadrenarchal after pituitary-gonadal suppression had been induced [age <6 yr, dehydroepiandrosterone sulfate (DHEAS) level <60 µg/dl after 6 months of GnRHa administration (+6m)]. This subset of subjects was selected from among a cohort of approximately 120 children with CPP. Uniform suppression of pituitary gonadotropin and gonadal sex steroid secretion was documented throughout the period of GnRHa administration, and compliance with the GnRHa regimen was verified by parental report and record of medication administration (for details, see Ref. 25).
Protocol
The protocol was approved by the human research committee of the two participating institutions (Massachusetts General Hospital, Boston, MA; Children’s Hospital, Boston, MA). Informed consent was obtained from parents before the enrollment of each subject in the study. Subjects were evaluated before, at 3- to 6-month intervals during, and at 6- to 12-month intervals after GnRHa administration. Each subject received daily sc injections of GnRHa using either deslorelin ([D-Trp6,Pro9-ethylamide]-GnRH; 4–8 mg/kg·d) or histrelin ([imBzl-D-His6,Pro9-ethylamide]-GnRH; 10 mg/kg·d). During each in-patient evaluation, confirmation of either active, pubertal gonadotropin secretion (before and after discontinuation of GnRHa) or pituitary desensitization (during GnRHa administration) was obtained (for details, see Ref. 25). Standing height (HT) was measured in the morning at least 30 min after the subject’s rising using a wall-mounted stadiometer; the average of three replicates is reported. During each admission breast and pubic hair development were assessed, and a left hand and wrist x-ray was obtained to monitor skeletal maturation.
To assess adrenal androgen secretion, serum was obtained the morning
after each admission (before, during, and after GnRHa administration)
for the measurement of DHEAS. The use of ACTH stimulation to define
precursor/product ratios for the assessment of steroidogenic enzyme
activity was added to the protocol after the study’s initiation.
From that point forward, samples were obtained for measurement
of dehydroepiandrosterone (DHEA), androstenedione
(
A4), 17-hydroxypregnenolone (17OH
Preg), and 17-hydroxyprogesterone (17OH Prog) before and 60 min
after iv cortrosyn administration [10 µg/kg (maximum, 250 µg) at
0830 h]. We were able to obtain data from serial ACTH stimulation
tests in 12 of the 14 subjects.
Methods
LH, FSH, estradiol, and DHEAS were measured using specific RIAs
as previously reported (25). DHEA,
A4, 17OH Preg, and 17OH Prog were measured
using solvent or chromatographic separation followed by RIA at Esoterix
Endocrinology, Inc. (Calabasas Hills, CA). Precursor to product ratios
were employed to assess steroidogenic enzyme activities, where an
increase in the ratio implied decreased enzyme activity, and a decrease
in the ratio implied increased enzyme activity. Two ratios were used to
investigate the function of 3ß-hydroxysteroid dehydrogenase
(3ßHSD): the ratio of the ACTH-stimulated levels of 17OH Preg to 17OH
Prog and the ratio of DHEA to A4.
The activity of C17,20 lyase (CYP17) was assessed using the ratios of
the ACTH-stimulated levels of 17OH Preg to DHEA and of
17OH Prog to A4. Bone age (BA) determinations
were made using the Tanner-Whitehouse radius-ulnar-short standards
(26, 27). Breast and pubic hair development were assessed
according to Tanner (28). SD scores for
height, weight, and body mass index (BMI) were calculated using the
method and standard curves released by the CDC in 2000
(http://www.cdc.gov/growthcharts/).
Statistical analysis
Methods of analyses were chosen to capitalize on the longitudinal nature of the study by permitting group analysis of the individual datasets. For statistical analyses, DHEAS concentrations were log transformed so that the values approximated a normal distribution. A random slopes model was used to analyze log DHEAS (n = 14) and the four ACTH-stimulated precursor-product ratios (n = 12) to look for significant trends over time. Initial analyses of DHEAS vs. chronological age suggested an exponential fit, but we also undertook analyses to compare the exponential fit of the data with linear models. In datasets limited to chronological age (CA) below 6 yr, a typically preadrenarchal age, linear and exponential fits of the data were compared, whereas in the whole dataset, the exponential fit was compared with a two-component linear model with a single change point. In both instances, residual variances were compared by the sign-rank test to assess the goodness of fit of the models.
To determine the association between the auxological measures (covariates) and log DHEAS among the 14 longitudinal datasets, two analyses were performed. In the first, each covariate was tested individually for significance employing the random slopes model to perform a forward regression. Each covariate was tested again while controlling for chronological age. In the second analysis, a random slopes model was used to seek a correlation between the residuals (observed - expected values) that stemmed from modeling the age trends in log DHEAS (linear fit) and each of the auxological measures (quadratic fit).
All descriptive data are presented as the mean ± SD; the estimates of mean slopes from the random slopes model are presented as the mean ± SEM. Data from all observations are included in the figures, but statistical analysis was limited to a period of documented complete pituitary-gonadal suppression [+6m through the visit when GnRHa administration was discontinued (D/C visit)]. Statistical significance was attributed to two-tailed P < 0.05. All analysis was performed using SAS version 8.1 (SAS Institute, Inc., Cary, NC).
Results
Fourteen preadrenarchal girls with idiopathic CPP were enrolled in
this study. Adrenal maturation was assessed at approximately 6-month
intervals during prolonged suppression of the pituitary-gonadal axis.
The mean period of evaluation while receiving GnRHa was 8.1 ± 1.1
yr (range, 6.0–9.9); the mean period of continued evaluation after D/C
visit (for 11 subjects) was 2.1 ± 0.9 yr (range, 0.6–4.1); and
the total duration of evaluation was 9.8 ± 1.8 yr (range,
6.8–12.3) for the 14 girls. To optimize the study of adrenarche
without the confounding effects of gonadarche, data analysis was
limited to a period of complete pituitary-gonadal suppression (+6m
through the D/C visit; see Table 1
).
|
60
µg/dl (1.63 µmol/liter)] (11, 12, 14, 17, 18), with
a DHEAS level of 60 µg/dl (1.63 µmol/liter) being reached in this
subset by CA 7.7 ± 1.2 yr. In this group DHEAS averaged 97
± 26 µg/dl (2.63 ± 0.71 µmol/liter) at the time of their D/C
visit at 11.3 ± 0.5 yr of age. In the remaining 6 girls, DHEAS
averaged 50 ± 7 µg/dl (1.36 ± 0.19 µmol/liter) at the
time of their GnRHa discontinuation at 10.8 ± 0.6 yr.
Our longitudinal data revealed a slow, but progressive, rise in DHEAS
levels from the initiation of the study onward (see Figs. 1-3). The progressive increases in DHEAS
levels exhibited in these young girls were seen well before they
attained biochemical evidence of adrenarche [DHEAS >60 µg/dl
(1.63 µmol/liter)] or reached the age of 6–8 yr, when adrenarche
is traditionally considered to begin (6, 10, 11, 12, 14, 17, 18). Analysis of the 14 individual datasets (DHEAS
vs. chronological age) indicated that an exponential
equation fit the aggregate data well. This assessment remained true
when the periods before and after CA 6 yr were analyzed separately
(P = 0.86 for difference between the two periods). The
estimated mean slope of log DHEAS vs. chronological age was
0.22 ± 0.02 (P < 0.0001), representing
approximately a 22% increase/yr. Although it is not clear that an
exponential model represents the best fit of the increases in DHEAS
over time, a linear fit of the increases in DHEAS observed under age 6
yr was not statistically better than the exponential fit
(P = 0.24), nor was a two-component linear model that
included all data and contained one change point, which would allow for
an early period of very modest increase followed by a later period of
more dramatic increase in DHEAS (P = 0.30).
|
Finally, the datasets were also analyzed to determine whether changes in BMI from one DHEAS measurement to the next were significantly correlated with serum DHEAS levels, as might have been expected from the data reported by Remer et al. (19). With age included in the forward regression model, no association was seen between change in BMI and serum DHEAS levels (P = 0.32). Similarly, in the second type of analysis, no correlation was found between the residuals of these two variables after having corrected for the underlying age trend that affects both (P = 0.55).
Data from serial ACTH stimulation tests were analyzed to examine the
changes in steroidogenic enzyme activity (decreased 3ßHSD and
increased CYP17 activity) that had been found in previous
cross-sectional studies (5, 16). None of our subjects had
abnormal baseline or stimulated steroid profiles (29).
Evidence for increasing CYP17 function was obtained from decreasing
ratios of both stimulated 17OH Preg to DHEA and 17OH Prog
to A4 levels (change in the ratio/change in CA
in years = -0.18 ± 0.03 and -0.35 ± 0.04,
respectively; both P < 0.0001). Conversely, evidence
for declining 3ßHSD activity was obtained from increasing ratios of
both stimulated 17OH Preg to 17OH Prog and DHEA to
A4 [change in the ratio/change in CA in
years = 0.20 ± 0.05; (P = 0.002) and
0.21 ± 0.04 (P = 0.0001), respectively] levels
during the course of the study (see Figs. 2 and 3 ). Changes in enzyme activities
began well before the clinical and biochemical criteria for adrenarche
were attained. As they were more statistically significant, the
alterations in CYP17 activity for each subject are displayed in Fig. 3. Interestingly, although the data
indicate that the direct conversion of 17OH Prog to
A4 is not evident in human adrenal tissue
studied in vitro (30), this precursor to
product ratio displayed the change of greatest magnitude in our
dataset.
|
|
We and others have found that very young girls (<6 yr of age) with idiopathic CPP are preadrenarchal at the time of initial evaluation (2, 4), one of the several clinical settings that support the distinct regulation of adrenarche and gonadarche (1, 2, 3, 4). Thus, once pituitary-gonadal suppression has been induced, this subject population provides a unique and powerful opportunity to evaluate adrenal maturation in the absence of the ongoing influence of gonadal sex steroids. Using this human model system, we report progressive increases in DHEAS concentrations from the +6m visit onward accompanied temporally by gradual alterations in the activity of key steroidogenic enzymes. These results suggest that adrenarche is not the result of sudden, rapid changes in adrenal enzyme activities or adrenal androgen concentrations. Rather, adrenarche appears to be a gradual and ongoing process that begins in early childhood.
Most previous studies have relied on cross-sectional data or have not included children young enough (5, 18) to document the pattern of gradually increasing DHEAS concentration that we have noted. However, although often not commented upon by the researchers, inspection of the data and graphs presented in many (11, 12, 13, 17), although not all (14), of these reports provides possible corroboration of the current findings. A recent longitudinal study using yearly 24-h urine collections to measure adrenal androgen production also provides evidence of gradual increases in adrenal androgen secretion that are earlier than previously expected (19).
Our longitudinal data also raise the possibility that DHEAS levels increase in an exponential pattern from early childhood through adolescence. Data in support of this possibility have been noted previously (19, 31), although these studies have lacked either the longitudinal design or sufficiently young subjects to establish firmly this pattern of increased adrenal androgen production. The pattern of progressively increasing DHEAS levels that we observed has the important implication that the levels of adrenal androgens seen at adrenarche and during early adolescence may be an extension of the same biological process(es) already evident in a careful examination of adrenal androgen levels during early childhood. A corollary is that developmental and/or genetic influences that modulate adrenal androgen concentrations in adolescence and adulthood may already be manifest early in childhood (19, 22, 24, 32, 33).
The clinical data in support of the association between body size and adrenal androgen levels have been conflicting. Methodologies have varied, and results are not fully comparable, but some data provide evidence for a correlation between measures of body size and/or fatness (10, 19, 20, 34), whereas other data do not (10, 21, 24, 34). Thus, although the clinical observation is often made that obese children seem to enter adrenarche early (10, 20), data conclusively demonstrating a relationship between indexes of body size and adrenal androgen levels are lacking. It should be noted, however, that factors leading to early or premature adrenarche may not be the same factors that modulate normally timed adrenal maturation (9).
The association between adrenal androgen production and body size was recently readdressed in a longitudinal study of children undergoing yearly urine collections for the assessment of adrenal androgen production (19). These researchers report that the largest increases in BMI were associated with significant increases in DHEAS production. Our data did not find such an association, but the studies are not fully comparable, as we measured serum DHEAS levels and cannot quantify production rates. It is also possible that the early rapid growth that characterized our patients’ CPP may have affected the relationships between body size and adrenal maturation. It seems unlikely that the administration of GnRHa affected the relationship, as our previous data indicate that GnRHa use does not lead to significant alterations in indexes of body composition (25). It is important to note that DHEAS and many of the growth parameters analyzed in this and other studies increase with CA. In any analysis of correlations, it must be remembered that CA may be a surrogate for important, but unidentified, developmental factors and that correlations between growth parameters and DHEAS levels may be spurious.
Alterations in adrenal steroidogenic enzymes (decreased 3ßHSD and increased CYP17 activity) would lead to increased DHEAS production from the adrenal gland. Previous data from in vivo (5, 16, 17, 35) and in vitro (36, 37, 38, 39, 40) studies have suggested that such alterations do occur and that they are correlated with increases in adrenal androgen production. However, these previous studies either have cross-sectional or have not included sufficient numbers of young subjects to permit delineation of the pattern of enzyme changes throughout adrenal maturation. Providing that other factors, such as altered metabolic clearance, have not confounded the analysis of precursor to product ratios, our longitudinal data indicate that 3ßHSD activity decreases and CYP17 activity increases beginning in early childhood. These changes may in part be a reflection of adrenal development and progressive expansion of the zona reticularis, which probably has decreased 3ßHSD activity (37, 38, 39) and increased CYP17 activity (37, 39).
Much research has focused on identifying a trigger for adrenarche. Our study paradigm has allowed us to study adrenal maturation without the confounding effects of gonadarche. The data from longitudinal DHEAS levels and from indexes of steroidogenic enzyme activity raise the possibility that adrenarche is not characterized by abrupt, marked increases in adrenal androgen production but, instead, may result from a progressive maturational process. A more complete understanding of adrenarche will result from future work that corroborates our findings and delineates the factors that control progressive adrenal maturation.
Acknowledgments
We thank the nursing staff of the participating General Clinical Research Centers for their dedicated care of these young patients during the evaluations, and the personnel of both the RIA Core Laboratory of the Reproductive Endocrine Sciences Center at Massachusetts General Hospital and the General Clinical Research Center Core Laboratory of Children’s Hospital for coordinating the assays. We also gratefully acknowledge the generous gift of the GnRHa from Drs. Wylie Vale and Jean Rivier of The Salk Institute (deslorelin, [D-Trp6,Pro9-NEt]-GnRH) and from Ortho Pharmaceuticals (histrelin, [imBzl-D-His6,Pro9-NEt]-GnRH).
Footnotes
This work was supported by NIH grants [HD-18169, General Clinical Research Center Grants RR-01066 and RR-02172, and K23-RR-15544-01 (to M.P.)] and the Lawson Wilkins Genentech Clinical Scholar Award (to M.P.).
Abbreviations: A4, Androstenedione; BA, bone
age; BMI, body mass index; CA, chronological age; CPP, central
precocious puberty; D/C visit, visit when GnRHa administration was
discontinued; DHEAS, dehydroepiandrosterone sulfate; GnRHa, GnRH
agonist; 3ßHSD, 3ß-hydroxysteroid dehydrogenase; HT, standing
height; 17OH Preg, 17-hydroxypregnenolone; 17OH Prog,
17-hydroxyprogesterone; WT, weight; +6m, 6 months of GnRHa
administration.
Received February 27, 2001.
Accepted May 10, 2001.
References
This article has been cited by other articles:
 |
|
 |
|
  Z Hochberg Juvenility in the context of life history theory Arch. Dis. Child., June 1, 2008; 93(6): 534 - 539. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Lazar, A. Padoa, and M. Phillip Growth Pattern and Final Height after Cessation of Gonadotropin-Suppressive Therapy in Girls with Central Sexual Precocity J. Clin. Endocrinol. Metab., September 1, 2007; 92(9): 3483 - 3489. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Sorensen, A. M. Andersson, N. E. Skakkebaek, and A. Juul Serum Sex Hormone-Binding Globulin Levels in Healthy Children and Girls with Precocious Puberty before and during Gonadotropin-Releasing Hormone Agonist Treatment J. Clin. Endocrinol. Metab., August 1, 2007; 92(8): 3189 - 3196. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. L. Rosenfield Identifying Children at Risk for Polycystic Ovary Syndrome J. Clin. Endocrinol. Metab., March 1, 2007; 92(3): 787 - 796. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F Domine, A-S Parent, G Rasier, M-C Lebrethon, and J-P Bourguignon Assessment and mechanism of variations in pubertal timing in internationally adopted children: a developmental hypothesis Eur. J. Endocrinol., November 1, 2006; 155(suppl_1): S17 - S25. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. D. Veldhuis, J. N. Roemmich, E. J. Richmond, and C. Y. Bowers Somatotropic and Gonadotropic Axes Linkages in Infancy, Childhood, and the Puberty-Adult Transition Endocr. Rev., April 1, 2006; 27(2): 101 - 140. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Remer, K. R. Boye, M. F. Hartmann, and S. A. Wudy Urinary Markers of Adrenarche: Reference Values in Healthy Subjects, Aged 3-18 Years J. Clin. Endocrinol. Metab., April 1, 2005; 90(4): 2015 - 2021. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. K. Ong, N. Potau, C. J. Petry, R. Jones, A. R. Ness, J. W. Honour, F. de Zegher, L. Ibanez, and D. B. Dunger Opposing Influences of Prenatal and Postnatal Weight Gain on Adrenarche in Normal Boys and Girls J. Clin. Endocrinol. Metab., June 1, 2004; 89(6): 2647 - 2651. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. D. Martin, R. Schweizer, C. P. Schwarze, M. W. Elmlinger, M. B. Ranke, and G. Binder The Early Dehydroepiandrosterone Sulfate Rise of Adrenarche and the Delay of Pubarche Indicate Primary Ovarian Failure in Turner Syndrome J. Clin. Endocrinol. Metab., March 1, 2004; 89(3): 1164 - 1168. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A.-S. Parent, G. Teilmann, A. Juul, N. E. Skakkebaek, J. Toppari, and J.-P. Bourguignon The Timing of Normal Puberty and the Age Limits of Sexual Precocity: Variations around the World, Secular Trends, and Changes after Migration Endocr. Rev., October 1, 2003; 24(5): 668 - 693. [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 |
