Precocious Puberty: Who Has It? Who Should Be Treated?
Karen Oerter Klein, M.D.
Department of Clinical Science
A.I. duPont Hospital for Children
Wilmington, Delaware
Address all correspondence and requests for reprints to: Karen Oerter Klein, M.D., Department of Clinical Science, A. I. du Pont Hospital for Children, P.O. Box 269, Wilmington, Delaware 19899. E-mail: kklein{at}nemours.org
 |
Introduction
|
---|
In this issue of JCEM, Palmert
et al. (see page 415) (1), address the issue of whether all
children with precocious puberty need treatment by reporting long-term
follow-up data on 20 girls with unsustained or slowly progressive early
puberty. This group of girls on the average achieved their genetic
height potential and normal adult heights without treatment. Ten of the
20 girls had onset of puberty after 5 yr of age. Thirteen of the 20
girls had a bone age (BA) advance greater than 2 yr above chronological
age (CA). None of the girls had GnRH-stimulated peak LH levels over 25
IU/L, and all had a GnRH-stimulated FSH peak greater than the LH peak.
Unsustained precocious puberty was defined as no further pubertal
development or regression of development as determined by a
questionnaire. Slowly progressive puberty was also defined by
questionnaire. This study once again demonstrates that not all girls
with precocious puberty need treatment, and it raises these questions:
What is precocious puberty? Who has precocious puberty? Which children
should be treated?
 |
What is precocious puberty and how is it treated?
|
---|
Precocious puberty has most commonly been defined as the onset of
puberty before the age of 8 yr in girls and before the age of 9 yr in
boys. It involves not only early physical changes of puberty, but also
linear growth acceleration and acceleration of bone maturation, which
leads to early epiphyseal fusion and short adult height. Precocious
puberty can be true or GnRH-dependent puberty, or it can be peripheral
or GnRH-independent puberty. The questions raised by Palmert et
al relate only to children with possible idiopathic true
precocious puberty. Most girls (95%) with precocious puberty have
idiopathic true precocious puberty. Boys more commonly (>50%) have an
identifiable etiology for precocious puberty. The causes of true
precocious puberty include central nervous system lesions, secondary to
GnRH-independent precocious puberty, and idiopathic precocious puberty.
The causes of GnRH-independent puberty include gonadal, adrenal,
ectopic, or exogenous sources of hormone production. Adrenal causes
include congenital adrenal hyperplasia (CAH) and tumors. Gonadal causes
include McCune-Albright syndrome, familiar male precocious puberty,
and tumors. Ectopic causes include human chorionic gonadotropin
secreting tumors. Hypothyroidism may also cause GnRH-independent
precocious puberty.
True precocious puberty involves activation of the
hypothalamic-pituitary-gonadal axis (HPGA). Standard treatment of
precocious puberty involves suppression of this axis with GnRH agonists
(2). When precocious puberty is not idiopathic, treatment is based
first on treating the underlying problem, and then may also involve
treatment with GnRH agonists. For example, children with non-salt
wasting CAH may present with precocious puberty. Appropriate management
of CAH is initiated first, followed by further evaluation of the need
for GnRH agonist therapy. A more detailed discussion of nonidiopathic
precocious puberty is beyond the scope of this editorial. The remainder
of this discussion relates to idiopathic precocious puberty.
 |
Why should we treat precocious puberty?
|
---|
Knowing why we treat idiopathic precocious puberty is
integral to understanding the questions of whom should we treat. The
only long-term complication of true idiopathic precocious puberty is
compromised adult height. Adult height is improved with treatment (2, 3). Therefore, part of the decision of who to treat involves estimating
adult height based on the childs current height and bone age, and
comparing this predicted height with mid-parental height (MPH) and with
the normal population. The other reasons to treat are psychosocial or
behavioral. For example, a 4-yr-old with menses, even if she has a good
predicted height, may deserve treatment. Or the 7-yr-old who is behind
in school may deserve treatment so that she is not so far ahead of her
peers and experiencing emotions related to hormonal changes that she
cannot understand.
 |
Is pubertal onset before age 8 yr in girls and 9 yr in boys an
appropriate definition for precocious puberty?
|
---|
The age definition of normal pubertal development is based on 95%
of the population or 2 standard deviations (SD) below the
mean age of pubertal onset in normal girls. Hence 2.5% of normal girls
have onset of puberty before age 8. When puberty in girls begins
between 6 and 8 yr, is it truly precocious puberty or simply at the
outer limits of normal? Are there some girls for whom earlier onset is
normal and others for whom it is precocious? Herman-Giddens et
al. (4) have recently studied 17,077 girls and reported normal
onset of puberty to be earlier than previously thought. In their
population the mean age of breast development in African-American girls
was 8.87 ± 1.93 yr (mean ± SD) and in white
girls was 9.96 ± 1.82 yr. If the definition of precocious is 2
SD below the mean, then, based on this data, precocious
puberty would be defined as onset less than 5 yr in African-American
girls and onset less than 6.3 yr in white girls. However, this study
did not have endocrine evaluations or follow-up data to determine if
any of the girls with earlier development actually had precocious
puberty.
 |
What defines the onset of puberty?
|
---|
Even after we define the time-frame for onset of precocious
puberty, we still must ask, what defines onset of puberty? The physical
changes of puberty are under the control of the HPGA. The HPGA activity
increases with the onset of puberty as evidenced by increasing number
and amplitude of spontaneous pulses of gonadotropins, luteinizing
hormone (LH), and follicle stimulating hormone (FSH), plus increased
peak LH and FSH levels in response to GnRH stimulation (5).
Unfortunately, there is overlap in spontaneous and stimulated
gonadotropin levels between prepubertal children and pubertal children,
making interpretation of gonadotropin testing sometimes difficult.
Interpretation of LH and FSH levels also depends on which gonadotropin
assay is used. The newer ICMA assays have much lower levels and much
greater sensitivity than the older RIAs. Because treatment of true
precocious puberty is suppression of the HPGA with GnRH agonists, the
axis must first have pubertal activity to warrant treatment. Thus, it
is important to define what constitutes an active HPGA in order to say
that true puberty has begun.
Simple premature thelarche involves only breast development, without
pubic hair growth, without accelerated bone maturation, and with a
normal height outcome, requiring no treatment. Simple premature
adrenarche involves only pubic hair development, without the other
manifestations of puberty, also requiring no treatment. It has been
suggested that there is a continuum of puberty from premature thelarche
through true precocious puberty (6). Pescovitz et al.
(6) divided girls into 6 groups: 1) isolated breast development, no
pubic hair, no growth spurt, and no bone age (BA) advancement; 2)
breast development and BA advance, but no pubic hair or growth spurt;
3) breast development and pubic hair, but no BA advance or growth
spurt; 4) breast development, growth spurt, and BA advance, but no
pubic hair; 5) breast development and pubic hair and either growth
spurt or BA advance; and 6) breast development and pubic hair and both
growth spurt and BA advance. Groups 1 and 2 had FSH predominant
responses to GnRH testing. Group 6 had LH predominant responses to GnRH
testing. However, there was variability in the LH and FSH responses to
GnRH testing in the intermediate groups. Some girls had LH predominant
responses, and some had FSH predominant responses. Palmert et
al. (1) describe girls with slowly progressing or unsustained
puberty. Some of their patients fall in the intermediate groups
described by Pescovitz et al. (6) which further
illustrates the spectrum of pubertal changes physically and
biochemically as well as the need for clinicians to distinguish between
patients when making decisions about treatment.
Palmert et al. report girls with both early breast
development and early pubic hair growth. Most of the girls were tall
for CA but short for BA, which usually predicts a shorter adult height
outcome. However, all girls had a prepubertal response to GnRH
stimulation and were, therefore, not considered candidates for
treatment with GnRH analogues. A prepubertal response to GnRH
stimulation was defined by Palmerts et al. criterion of
FSH peak greater than LH peak and LH peak less than 25 IU/L (measured
by RIA using the 2nd IRP-hMG LH standard). Most of the girls in their
study also had no detectable peaks of spontaneous nocturnal LH
secretion. Palmert et al. appropriately note this is an
older less sensitive LH assay and that newer gonadotropin assays may
help further define the gonadotropin secretion in similar girls. It is
also important to note that others might use different criteria to
interpret the GnRH tests. This illustrates another problem in defining
the onset of puberty. There is no single level of LH, FSH, or estradiol
with 100% specificity and 100% sensitivity for precocious puberty. We
previously reported a peak LH level of more than 15 IU/L or a peak
LH-to-peak FSH ratio of more than 0.66 as criteria for defining a
pubertal GnRH test (96% sensitivity, 100% specificity, no false
positives) (5). Palmerts patient no. 15 had a peak LH-to-peak FSH
ratio of 0.74, which is greater than 0.66. This patient also had a bone
age advance of 3.6 years above CA and a height SDS for CA of -2.0. She
had onset of menses at 8.8 yr. All of these things together would have
suggested a compromised adult height outcome in this girl, and she
probably would have been treated by some pediatric endocrinologists.
However, without treatment, she also achieved a final height greater
than her MPH and with a 0.2 SDS. The authors note some limitations of
doing a questionnaire follow-up. An additional limitation is the
accuracy of reported heights. If this reported height is accurate, this
girl illustrates how unclear the assessment of the need for treatment
can be.
Palmert et al. report that this groups of girls had good
final height outcome on the average. However, not all girls achieved
their MPH. Patients no. 11 and no. 14 achieved adult heights 10 cm and
12.6 cm below their MPH, respectively, and final height SDS of -1.9.
Neither of these girls had peak LH levels over 15 IU/L or peak
LH-to-peak FSH ratios over 0.66, so by Palmerts or our criteria would
not have been treated. They did have significant advancement of BA, and
they were two of the four girls who had measurable nocturnal LH pulses.
These two girls again illustrate the complexity of distinguishing who
needs treatment. Similar girls should undergo repeat evaluation
periodically, watching for rate of pubertal progression, changes in
predicted height, and determination if they meet biochemical criterion
for treatment, which might improve their adult height outcome.
Neely et al. (8) studied 49 girls with clinical signs of
true precocious puberty, and report that a baseline LH of more than 0.3
IU/L or a GnRH-stimulated LH peak of more than 5 IU/L (by the newer
ICMA using World Health Organization 2nd International Standard, Human
Pituitary LH 80/522) may be diagnostic of precocious puberty (7, 8). A
few girls with true precocious puberty were still missed using these
criteria. As stated by Palmert et al., more studies with the
newer assays may improve our definition of pubertal HPGA activity.
Estradiol levels have not been reliable for the diagnosis of precocious
puberty. While some girls present the obviously pubertal estradiol
levels, many girls present with estradiol levels below the detection
limit of available RIAs, as in the girls reported by Palmert et
al. We are investigating the potential usefulness of an
ultrasensitive estradiol assay in the evaluation of precocious puberty
(9, 10).
 |
Do all children with true idiopathic precocious puberty need
treatment?
|
---|
Fontoura et al. (11) described a slowly progressive
form of precocious puberty, in which pubertal development began early
but progressed slowly. They described 15 girls with clinical signs of
precocious puberty before age 8 yr, but with BA advance less than 2 yr
above CA. The girls were not treated and maintained predicted height
over 2 yr of follow-up. They compared this group to another 19 girls
with precocious puberty and BA advance greater than 2 yr above CA. The
second group had lower predicted heights at onset and were treated. The
untreated group had lower peak LH levels than the treated group.
Because the second group was treated, we do not know if all of them
would have progressed rapidly. Ten of the 19 had peak LH levels less
than 15 IU/L. Again, the conclusion is that not all girls need
treatment, but it cannot be concluded that the degree of BA advance can
always distinguish which cases of puberty should be treated. Thirteen
of 20 patients described by Palmert et al. had BA advance of
more than 2 yr.
Kreiter et al. (12) described 7 girls with precocious
puberty in whom predicted adult height was maintained over 2 yr of
follow-up without treatment. They treated 14 girls with precocious
puberty who had a loss in predicted height of at least 5 cm by two BA
determinations at least 5 months apart or a predicted height less than
152.5 cm. Seven girls who did not meet these criteria were not treated.
The groups did not differ in terms of gonadotropin testing, although
individual test results were not given. Both groups had improvement in
predicted height over the two yr. For comparison, note that 2 of the
girls in the Palmert et al. study achieved final heights of
less than 152.5 cm, a criterion Kreiter et al. would have
used to initiate treatment.
 |
Who should be treated?
|
---|
Any girl with precocious physical signs of puberty, significantly
advanced BA, decreased predicted height, and a pubertal response to
gonadotropin testing should be treated with GnRH agonists to suppress
pubertal progression and improve adult height. The problem remains that
each of these variables (age of appropriate puberty, definition of
significant BA advance, definition of decreased predicted height, and
definition of a pubertal response to gonadotropin testing) is still not
well defined and is approached differently by various investigators.
There are many girls with an unequivocal evaluation who deserve
treatment. It is the girls with a questionable evaluation of one or
more of the above variables who need close attention and awareness by
the clinician that treatment is not always necessary. No one has yet
defined absolute criteria for determining who will benefit from
treatment. Follow-up is extremely important in all girls, as it may
become clear over time who needs treatment. Some estimate of adult
height seems to be the most important predictor as reported by Fontoura
et al. (11) and Kreiter et al. (12). The girls in
the Palmert et al. study were selected by gonadotropin
testing alone. All girls who are not treated need follow-up to evaluate
pubertal progression, predicted height, pubertal gonadotropin levels,
and possible benefit from treatment. Girls with true precocious puberty
will continue to progress rapidly through puberty, so that repeat
physical examinations will raise the suspicion of who needs repeat
gonadotropin testing and possible treatment. Girls who dont need
treatment will progress slowly or will be unsustained as described
in this months report by Palmert et al. Because it is
important to treat girls with an unequivocal evaluation, it will be
difficult to obtain absolute criteria for treatment at this point, as
there will be no true comparison of untreated girls with more advanced
bone age or more decreased predicted height. However, more studies
should be done to try to distinguish what criteria to use. Palmert
et al. report follow-up in girls who have achieved final
height. Longer follow-up, including final height information in the
other untreated series (11, 12), may also help define criteria for
treatment.
 |
What can we conclude?
|
---|
Our definition of precocious puberty is still being
clarified, both from the perspective of age of physical changes and
interpretation of gonadotropin levels. In addition, not all children
with apparently true precocious puberty need medical intervention.
Palmert et al. have described a group of girls who have the
appropriate physical characteristics of precocious puberty, some with
advanced bone maturation and some without, but none with biochemical
evidence of pubertal HPGA activity, most of whom clearly did not
require treatment. In addition, others have used BA advance criteria
(11) or predicted height prognosis (12) to determine who to treat, even
when girls fit a biochemical definition of precocious puberty, and have
similarly shown that not all girls require treatment.
None of the untreated girls described by Fontoura (11), Kreiter (12),
or Palmert et al. (1) had rapid progression of their
puberty. In cases of possible precocious puberty that are equivocal (CA
between 58 yr, BA not as advanced, predicted height still close to
MPH, and/or GnRH testing unclear), it is important to obtain adequate
follow-up including careful examination of the rate of progression of
physical changes, linear growth, bone maturation, estimates of adult
height, and stimulated gonadotropin levels, to determine which girls
need treatment. More studies with the newer more sensitive gonadotropin
assays and possibly studies with more sensitive estradiol assays are
needed to determine if we can more accurately define puberty and to
determine which children will benefit from treatment.
Received December 2, 1998.
Accepted December 10, 1998.
 |
References
|
---|
-
Palmert MR, Malin HV, Boepple PA. 1999 Unsustained or
slowly progressive puberty in young girls: initial presentation and
long-term follow-up of 20 untreated patients. J Clin Endocrinol Metab. 84:415423.[Abstract/Free Full Text]
-
Comite F, Cassoria F, Barnes KM, et al. 1986 Luteinizing
hormone releasing hormone analogue therapy for central precocious
puberty: long-term effect on somatic growth, bone maturation, and
predicted height. JAMA 255:26132616.
-
Oerter KE, Manasco P, Barnes KM, Jones J, Hill S, Cutler GB
Jr. 1991 Adult height in precocious puberty after long-term
treatment with doslorelin. J Clin Endocrinol Metab. 73:12351240.[Abstract]
-
Herman-Giddens M, Slora EJ, Wasserman RC, et al. 1997 Secondary sexual characteristics and menses in young girls seen in
office practice: a study from the pediatric research in office settings
network. Pediatrics. 99:505512.[Abstract/Free Full Text]
-
Oerter KE, Uriarte MM, Rose SR, Barnes KM, Cutler GB Jr. 1990 Gonadotropin secretory dynamics during puberty in normal girls
and boys. J Clin Endocrinol Metab. 71:12511258.[Abstract]
-
Pescovitz OR, Hench KD, Barnes KM, Loriaux DL, Cutler GB
Jr. 1988 Premature thelarche and central precocious puberty:
the relationship between clinical presentation and the gonadotropin
response to luteinizing hormone-releasing hormone. J Clin
Endocrinol Metab. 67:474479.[Abstract]
-
Neely EK, Hintz RL, Wilson DM et. al. 1995 Normal ranges
for immunochemiluminometric gonadotropin assays. J Pediatr. 127:4046.[Medline]
-
Neely EK, Wilson DM, Lee PA, Stene M, Hintz RL. 1995 Spontaneous serum gonadotropin concentrations in the evaluation of
precocious puberty. J Pediatr. 127:4752.[Medline]
-
Klein K Oerter, Baron J, Colli MJ, McDonnell DP, Cutler GB
Jr. 1994 Estrogen levels in childhood determined by an
ultra-sensitive recombinant cell bioassay. J Clin Invest. 94:24752480.[Medline]
-
Klein KO, Baron J, Barnes KM, Pescovitz OH, Cutler GB Jr. 1998 Use of an ultrasensitive recombinant cell bioassay to determine
estrogen levels in girls with precocious puberty treated with an LHRH
agonist. J Clin Endocrinol Metab. 83:23872389.[Abstract/Free Full Text]
-
Fontoura M, Brauner R, Prevot C, Rappaport R. 1989 Precocious puberty in girls: early diagnosis of a slowly progressing
variant. Arch Dis Child. 64:11701176.[Abstract]
-
Kreiter M, Burstein S, Rosenfiel RL, et. al. 1990 Preserving adult height potential in girls with idopathid true
precocious puberty. J Pediatr. 117:354370.