- Email: [email protected]
Preserving adult height potential in girls with idiopathic true precocious puberty
Preserving adult height potential in girls with idiopathic true precocious puberty Mary Kreiter, MD, Stephen Burstein, MD, PhD, Robert L. Rosenfield, MD, George W. Moll, Jr,, MD, PhD, Jos~ F. Cara, MD, David K. Yousefzadeh, MD, Leona Cuttler, MD, and Lynne L. Levitsky, MD From the Departments of Pediatrics and Radiology, University of Chicago Pritzker School of Medicine, Chicago, Illinois We designed a prospective study of height potential in girls with idiopathic precocious puberty, comparing the presenting features of girls with and without evidence of reduced adult height potential. The 14 girls with impaired adult height prognoses (group 1) were reexamined after treatment with a gonadotropin releasing hormone agonist, nafarelin. The seven girls with the prognosis of unimpaired height (group 2) were followed without therapy. We found that the group could be distinguished at initial examination by the greater bone a g e / height age ratio of group I (mean _ SEM: 1.4 _ 0.06 vs 1.0 • 0.05; p <0.005) and by the greater difference between predicted height and target height in group 1. The mean predicted height in group I was significantly less than the mean target height (150.7 • 2.1 vs 165.4 • 3.0 cm; p <0.005), whereas the mean predicted and target heights in group 2 were similar (165.4 _+ 3.0 vs 164.3 • 2.1 cm). Initial estradiol levels were also greater in group I than in group 2 (21.6 vs 10.6 pg/ml; p <0.05), although this difference was not sustained during follow-up. In group I, nafarelin therapy suppressed the pituitary-gonadal axis, and although there was a transient reduction in height potential in girls with the youngest bone ages during the first 6 months of therapy, 2 years of treatment slightly improved predicted heights from 150.7 • 2.1 to 152.7 • 2.0 cm (p <0.05). Height predictions also increased without therapy during the 2-year observation period in group 2, from 165.4 _+ 3.0 to 168.7 _ 4.1 cm (p <0.05). Our data indicate that gonadotropin releasing hormone agonist therapy preserves height potential in girls with an initially impaired height prognosis, and that height potential is preserved without therapy in those with a g o o d initial height prognosis. (J PEDIATR1990;117:364-
70) True precocious puberty appears to be caused by premature maturation of the normal hypothalamic-pituitary-gonadal axis. As a consequence, secondary sexual characteristics develop early and skeletal maturation and linear growth accelerate. The only permanent physical sequela of precoSupported in part by U.S. Public Health Service grants RR-00055 (General Clinical Research Center), NIADDK 07011-15 (M.K.), and HD-06308 (R.L.R.) and in part by Syntex Corporation. Submitted for publication May 23, 1989; accepted April 11, 1990. Reprint requests: Robert L. Rosenfield, MD, Wyler Children's Hospital, Section of Pediatric Endocrinology, 5841 S. Maryland, Box 118, Chicago, IL 60637. 9/20/21523
364
cious puberty is attenuation of adult stature, and this effect is variable. 1-3 Recently, long-acting gonadotropin releasing hormone agonists have been reported to be effective in arBA CA FSH GnRH HA LH
Bone age Chronologic age Follicle-stimulating hormone Gonadotropin releasing hormone Height age Luteinizing hormone
resting pubertal progression and, unlike previous treatment modalities,47 in improving height potential. 8-17 However, all patients have been treated without consideration of their pretreatment prognosis for adult height. We therefore de-
Volume 117 Number 3 signed a prospective study of height potential in girls with idiopathic precocious puberty in which girls with evidence of impaired adult height potential were treated with the GnRH analog nafarelin and the remainder were followed without therapy. The first objective was to compare each group at presentation to identify discriminating characteristics, and the second, to examine the changes in each group after 2 years of follow-up. METHODS Patients. Twenty-one girls referred to our institution between January 1984 and July 1987 with previously untreated idiopathic true precocious puberty were examined at the University of Chicago General Clinical Research Center. Idiopathic true precocious puberty was defined as the onset of isosexual pubertal changes at a chronologic age of less than 8.0 years, with a sleep-associated rise in serum levels of luteinizing hormone and a pubertal response to GnRH or nafarelin stimulation testing, 18 in the absence of organic abnormality on abdominal ultrasound examination, computed tomography, or magnetic resonance imaging, the last becoming available at mid study. Of the 21 girls, 14 met our criterion for nafarelin treatment and chose to undergo treatment. The criterion for treatment was a deterioration in predicted height as indicated by a loss of ~ 5 cm on sequential bone age determinations at least 5 months apart (n = 3) or a predicted adult height of less than 152.5 cm (n = 11). We refer to these girls as sexually precocious and with a prognosis of impaired height (group 1). The seven girls with a prognosis of unimpaired height who did not meet the criterion for treatment at initial examination were observed without therapy (group 2). Evaluations. Initial evaluation included height, Tanner staging, determination of plasma levels of estradiol, vaginal smear, BA determination, pelvic ultrasonography, somatomedin C concentration, and GnRH or nafarelin stimulation testing. In addition, all patients in group 1, and five of seven in group 2, had nocturnal growth hormone determinations. After the institution of therapy, group 1 was followed at a minimum of 3-month intervals, with height, Tanner stage, plasma estradiol level, and vaginal smear monitored at each visit. Yearly height velocity and 2xBA/ACA ratio were computed at 6-month intervals in the first year and at the end of 12 months in the second year. At 6, 12, and 24 months of therapy, a BA radiograph, pelvic ultrasound examination, and determination of serum somatomedin-C level were performed. Basal levels of LH, follicle-stimulating hormone, and estradiol were assayed at the initial evaluation and after 6, 12, and 24 months of therapy. A GnRH or nafarelin stimulation test was performed at the initial evaluation and after 6 and 12 months of therapy. For an assessment of response to therapy by an individual patient, either
Preserving height potential in precocious puberty
365
a GnRH or a nafarelin test was consistently used. The peak L H and F S H responses during GnRH and nafarelin tests were pooled for intergroup comparisons because the peak responses to these agents are similar during puberty. 18 Urinalysis, hemogram, multiphasic serum chemistry pro fie, and serum prolactin level were obtained at each visit to monitor for toxic effects. Fasting insulin levels were monitored at the initial evaluation and at 6 and 12 months to examine the effect of nafarelin therapy on the insulin resistance of puberty. Group 2 was generally followed at 6-month intervals. Physical examination was performed and BA, vaginal smear, and plasma estradiol level were obtained at each visit. Treatment regimen. Group 1 was treated with nafarelin ([6-D-(2-naphthyl)alanine]-GnRH acetate; Synarel), generously supplied by Syntex Corp. Nafarelin therapy was initiated with an intranasal dosage of 800 # g / d a y in three patients and 1200 #g/day in the remaining 11, divided into two daily administrations. Dose adjustments were made at 3-month intervals if there was evidence of incomplete suppression of puberty. If control was not adequately achieved with 1600 ~g/day, nafarelin was administered subcutaneously at a dose of 4 #g/kg/day. The specific indications for dose adjustments were an elevated plasma estradiol level (9 to 23 pg/ml in six patients and 56 pg/ml in one patient; about half of these patients had vaginal eornification), vaginal cornification (n = 1) or LH >15 I U / L (n = 2) as isolated findings, and the clinical progression of puberty (n = 1). Only four patients required subcutaneous therapy (after 7 months in three and 20 months in one). Laboratory methods. Bone age was determined (by M.K., S.B., R.L.R.) from the Greulich and Pyle atlas by the method of Roche et a1.19 Height predictions were made according to the method of Bayley and Pinneau 2~because this has been shown to be the most accurate method of height prediction in precocious puberty. 21 In those patients with an initial BA of less than 7 years (n = 3), however, the RocheWainer-Thissen method a2 was used for each of that patient's height predictions to maintain consistency. Blinded scoring of BA by the three examiners showed good correlation. Target heights were calculated from mid-parental height by the method of Tanner et al. 23 Real-time pelvic ultrasound examinations were evaluated for average maximal ovarian diameter and ovarian architecture. A modification of the method of Salardi et al. 24 was used to classify ovarian architecture: class I--homogeneous appearance, no follicles identified; class II--nonhomogeneous, three or fewer small follicles 8 mm or less in diameter, per greatest diameter per ovary; class IlI--minifollicular, 4 to 9 follicles 8 mm or less in diameter; and class IV--follicular, 1 follicle greater than 8 mm but 20 mm or less in diameter. A vaginal maturation score was calculated
366
Kreiter et al.
The Journal of Pediatrics September 1990
T a b l e I. Clinical data Group 2
Group 4 Before therapy Mean • SEM Age (yr) Breast stage Pubic hair stage Height (cm) Height growth velocity (cm/yr) Bone age (yr) ABA/2xCA ratio BA/HA ratio Height prediction (cm)
6.1 3.2 2.4 123.0 9.9 9.8 1.8 1.4 150.7
• 0.5 • 0.1 • 0.2 • 2.5 + 0.8 • 0.4 ___ 0.4 • 0.06 • 2.1
After 2 yr of therapy
After 2 yr of observation
At entry
(n)
Mean • SEM
(n)
(14) (14) (14) (14) (14) (14) (14) (14) (14)
8.3 • 0.5 2.9 • 0.3 2.8 4- 0.2 135.4 • 2.4 5.2 • 0.6 11.5 • 0.4 0.7 • 0.1 1.2 + 0.04 152.7 • 2.0
(14) (14) (14) (14) (14)I" (13)I (13)* (13) (13)*
M e a n _+ $EM 5.8 3.0 1.6 121.5 10.5 7.9 1.5 1.0 165.4
• 0.8 ___0.0 • 0.3 + 6.5 • 1.1 • 1.1 _+ 0.4 • 0.05 • 3.0
(n) (7) (7) (7) (7) (7) (7) (7) (7)w (7)w
M e a n 4- $EM 7.9 • 3.1 • 2.7 • 140.7 • 8.7 • 10.5 • 1.2 • 1.0 • 168.7 •
0.5 0.3 0.5 3.1 0.6 0.4 0.2 0.03 4.1
(n) (7) (7) (7)* (7) (7)w (7) (7)
(7)11 (6)*,$
*p <0.05, within a group from initial evaluation. tP <0.005, within a group from initial evaluation. :~p <0.05, betweengroups for comparable observationtimes. w <0.02, betweengroups for comparableobservationtimes. liP <0.005, betweengroups for comparableobservationtimes.
T a b l e II. Endocrine data Group 2
Group I Before therapy Mean • SEM Basal estradiol (pg/ml) Basal LH (IU/L) ALH (IU/L) Basal FSH (IU/L) ~xFSH (IU/L) Vaginal maturation score Average maximal ovarian diameter (mm) Ovarian classification score Somatomedin C (U/ml)
46.7 7.4 68.6 3.8 11.7 49.8 26.6
• • • • • • •
16.9 1.6 10.6 0.3 1.6 4.3 2.0
2.6 +_ 0.2 2.2 • 0.3
After 2 yr of therapy
After 2 yr of observation
At entry
(n)
Mean • SEM
(n)
(14) (14) (14) (13) (13) (12) (10)
8.5 • 3.6 7.2 _ 0.8 -1.8 + 0.3 -27.8 • 6.5 28.0 + 2.3
(14)* (13)
(10) (14)
2.2 + 0.1 2.0 • 0.2
(9) (14)
(12)I (12):~ (9)
Mean • SEM 14.8 5.6 54.2 3.6 14.5 45.3 26.8
• 4.2 4- 1.2 • 24.9 • 0.5 • 3.7 • 9.2 ___ 5.0
3.4 • 0.5 2.2 • 1.1
(n)
Mean _+ SEM
(n)
(7) (6) (6)
15.7 • 10.2
(5)
33.5 • 13.0
(6)
1.8 • 0.6
(6)
(5) (5) (6)
(4) (4) (4)
*p <0.05, within a group from initial evaluation. l"p <0.02, within a group from initial evaluation. :~p <0.005, within a group from initial evaluation.
from the vaginal smear as the sum of the number of superficial cells and half the number of intermediate cells. 25
of Furlanetto. 27 Prolactin was measured by means of a ra-
Serum L H and F S H levels were determined with highly
geles, Calif.). Fasting insulin levels were measured by a double-antibody radioimmunoassay, 2s as were growth hormone levels with the use of polyelonal antiserum. 29
specific and sensitive radioimmunoassays modified to measure bioactive species preferentially. 26 The minimal detectable levels averaged 0.1 n g / m l (0.78 I U / L ) of the L H standard I-2 and 0.1 n g / m l (0.39 I U / L ) of the F S H standard 1-3 (National H o r m o n e and Pituitary Program, Na-
dioimmunoassay kit (Diagnostic Products Corp., Los An-
Consent. This study was approved by the University o f Chicago Hospitals and Clinics Clinical Investigation Committee. W r i t t e n informed consent was obtained from all
tional Institute of Diabetes and Digestive and Kidney
parents and assent from all children.
Diseases). Plasma estradiol levels were determined by a ra-
Statistical analysis. Student unpaired and paired t tests with the Bonferroni correction for multiple comparisons
dioimmunoassay the sensitivity of which averaged 2.7 p g / ml. 26 S o m a t o m e d i n C levels were determined by the method
were used as appropriate. 3~ The M a n n - W h i t n e y and
Volume 117 Number 3
Preserving height potential in precocious puberty
Wilcoxon signed rank tests were similarly used for nonparametric data and to normalize the distribution of parametric data, as indicated. Comparisons are presented as the mean ___SEM. T h e p values are two tailed. Actual sample size (n) varied slightly for some measurements because of incomplete data collection.
Height velocity was unchanged after 6 months of nafarelin therapy (9.9 _+ 0.8 vs 8.2 +_ 0.6 cm/yr) but decreased significantly during the second 6 months to the rate at which it remained during the second year of therapy (Table I) (p <0.002). Girls whose BA was less than the group mean of 9.8 years had a reduction in predicted height during the first 6 months of therapy, compared with those whose BA was greater than the group mean ( - 2 . 6 + 1.7 vs 2.1 _+ 0.9 cm; p <0.05). This difference was not related to hormonal factors. Thereafter the predicted height rose progressively during therapy. Overall, it was 150.7 _+ 2.1 cm before and i52.7 +_ 2.0 cm after 24 months of therapy (p <0.05, Wilcoxon signed rank test) (Table I). The ABA/AxCA ratio fell more than 50% in the course of the 2-year treatment period (p <0.05). The basal LH level was unchanged in response to therapy; however, the basal FSH level decreased significantly during nafarelin therapy. The LH and F S H responses to G n R H or nafarelin stimulation decreased by 6 months and remained low thereafter. The LH response was 68.6 + 10.6 I U / L at baseline and 8.7 + 2.5 I U / L at 6 months. The F S H response was 11.7 _+ 1.6 I U / L and 1.2 _+ 0.5 I U / L , respectively, a t the same time points. Estradiol levels decreased significantly in response to nafarelin therapy. The basal estradiol level at the start of therapy averaged 46.7 _+ 16.9 pg/ml and fell to 6.5 +_ 1.1 by 6 months (p <0.05), after which it remained suppressed. Vaginal maturation scores also decreased by 6 months of therapy, from 49.8 _+ 4.3 to 23.9 +__6.6 (p <0.001), and remained low thereafter. Ovarian long diameter and classification scores and somatomedin C levels were unchanged during nafarelin therapy (Table II). Group 2. Breast stage, plasma estradiol level, and vaginal maturation score showed no significant changes during the observation period in group 2. Pubic hair stage, however, continued to advance (p <0.05). None of the premenarcheal girls subsequently menstruated. Conversely, one of the postmenarcheal girls had intermittent waxing and waning of puberty even after results of serial ovarian ultrasound examinations and progesterone levels suggested that she had ovulated on one occasion. Bone age continued to advance (p <0.05) but growth velocity was unchanged. The ABA/ACA ratio was such that height predictions increased by 3.3 _+1.1 cm during the observation period (p <0.05). Toxic and other Side effects. Nafarelin therapy was well tolerated. There were no drug-related side effects, and routine laboratory screening showed no biochemical evidence of renal, hepatic, or hematopoietic toxic effects (data not shown). Prolactin levels did not change significantly during the treatment period. Fasting glucose levels (baseline vs 12 months of therapy: 87.7 + 2.2 vs 87.2 + 1.9 mg/dl) and
RESULTS Initial evaluation. Age and breast and pubic hair stages at entry were similar in the two groups (Table I). Menses occurred in two of the girls in group 1 before therapy, and in three in group 2. As expected from the criterion for entry into the treatment group, predicted height at entry into the study was significantly less in group 1 than in group 2 (p <0.01). Target heights were similar in group 1 and group 2 (163.1 _+ 1.1 vs 164.3 + 2.1 cm). Thus the mean predicted height at initial evaluation was significantly less than the mean target height in group 1 (p <0.005) but not in group 2. Predicted height was ~ 5 cm less than target height in 12 of the 14 girls in group 1 and in 3 of 7 girls in group 2. The B A / H A ratio at presentation was greater in group 1 girls than in group 2 (p1.2 in 13 of 14 girls jn group 1, in contrast to only 1 of the 7 girls in group 2. There was no significant difference in the height velocity of group 1 girls during their average pretreatment observation period of 5.2 months compared with the first 6 months of observation in group 2. Plasma estradiol levels at the initial evaluation were greater in group 1 than in group 2, averaging 21.6 pg/ml (range 7 to 66) and 10.6 pg/ml (range 3 to 32), respectively (p <0.05, Mann-Whitney test). An estradiol level >23 pg/ ml discriminated two thirds of the study population. However, the basal plasma estradiol levels and vaginal maturation scores during the remainder of the control period did not differ between the two groups (Table II). Basal and peak LH and FSH levels during either GnRH or nafarelin stimulation testing did not differ significantly between the groups. Ovarian size was similar in group 1 and group 2, as were the ovarian classification scores. All patients tested had nocturnal growth hormone peak levels ~ 6 ng/ml and normal somatomedin C levels, suggesting that they do not have classic growth hormone deficiency.
Study period Group 1. Nafarelin effectively arrested pubertal progression in group 1, in that breast stage, breast diameter, and pubic hair stage were unchanged after 24 months of nafarelin therapy. It was our impression that although breast measurements did not regress, breast tissue became atrophic, and most girls had a clear reduction of breast prominence. Menses ceased (after withdrawal bleeding in the postmenarcheal patients).
367
368
Kreiter et al.
insulin levels (15.1 ___ 1.8 vs 13.5 +_ 2.2 ~tU/ml) were unchanged by nafarelin therapy. Sinus clouding was reported on foUow-up magnetic resonance imaging in four patients, one of whom was receiving subcutaneous therapy, but none had clinical symptoms. One of these patients also had clouding on pretreatment films. DISCUSSION Treatment modalities for precocious puberty have sought to halt or reverse deterioration of height potential. Until recently, agents available for treatment of precocious puberty have been largely unsuccessful in improving the final height), 7 GnRH agonists have been reported to be well tolerated and effective. In girls with precocious puberty and impaired-height prognosis (group 1), we found that nafarelin therapy effectively suppressed the pituitary-gonadal axis and arrested breast development and menses, consistent with the findings of others using GnRH analogs), 9, 11-16Height velocity decreased and ABA/ACA ratio and predicted height improved significantly, as other investigators have found while employing other analogs. 9' I l, 12-16Manasco et al. 31 reported that height predictions increased steadily at the rate of 3 cm/yr as therapy was prolonged. In contrast to previous reports, we found that the decrease in height velocity did not occur until the second 6 months of therapy. Furthermore, during the first 6 months of therapy, predicted height in the subset of girls in group 1 with BAs less than 9.8 years initially deteriorated. Predicted height and ABA/ACA ratio improved during the second year of therapy. Others have reported that predicted height does not necessarily improve during the short term, 9, 32 but whether their patients had this transient decrease during the induction of treatment is unknown. This initial setback in predicted height could have been due to the stimulation of estradiol secretion that occurs during induction of GnRH agonist therapy. Alternatively, this simply may have represented a residual manifestation of unsuppressed puberty. It is unlikely that our GnRH agonist dose was inadequate, because we used intranasal doses fourfold to sixfold greater than are required to suppress pituitary function in adult women, 33 and control of puberty was as good as that of the other subset at the routine examination 3 months after the start of therapy. Unlike Pescovitz et al. 34 and Harris et al., 35 who found significant decreases in somatomedin C levels during GnRH analog therapy, we found that somatomedin C values did not change during nafarelin therapy. An important consideration in evaluating such differences between our results with GnRH analog therapy and those of others is that we selected for nafarelin treatment those girls most severely affected by their sexual precocity. Other studies may have included patients similar to those in our
The Journal of Pediatrics September 1990
group 2, in whom the neuroendocrine "drive" for puberty was relatively mild and thus more easily suppressible. It has long been recognized that idiopathic true precocious puberty only variably results in subnormal adult height. 1"3 We found that sexually precocious girls with an unimpaired-height prognosis (group 2) could be distinguished from those with an impaired-height prognosis in the early stages of evaluation. On an individual basis, a B A / H A ratio of less than 1.2 was almost always associated with a good height prognosis. In addition, a predicted height comparable to target height and a tendency toward lower plasma estradiol levels were more characteristic of this group. The pace of pubertal progression seemed to be slower as well; however, this was difficult to document because Tanner stage I I | describes most of pubertal breast development. Without therapy, group 2 girls maintained an acceptable height potential. In fact, height predictions increased significantly, although the mean B A / H A ratio remained constant at 1.0. The reasons for possible improvement in height prognosis are not readily apparent. One possible explanation is that low circulating levels of estradiol in a very narrow range may be stimulatory to growth. 36 However, although low doses of ethinyl estradiol increase height velocity, improvement in predicted height has not been confirmedS To test the possibility that the improved height predictions in our untreated girls may have been the result of lesser estrogen stimulation, we compared various estrogenic effects. Although plasma estradiol levels were lower in group 2 at the initial visit, this difference was not sustained and vaginal maturation indexes were similar. Recently Fontoura et al. 38 found significantly lower plasma estradiol levels in a group of girls with a slowly progressive variant of idiopathic precocious puberty and stable adult height potential. We found no clinically significant side effects from nafarelin spray therapy, other than possible chronic irritation of the nasal mucosa, about which our data were inconclusive. The long-term safety of GnRH analogs is still largely unknown; the potential effects of these agents on bone density, on subsequent recovery of the hypothalamicpituitary-gonadal axis and reproductive potential, as well as on possible oncogenicity, require ongoing investigation. Early diagnosis and institution of therapy in those patients with precocious puberty and a poor height prognosis seem indicated before substantial impairment of height potential occurs. When GnRH agonist therapy is instituted, our findings suggest the need to achieve rapid (within 6 weeks) and complete suppression of the pituitary-gonadal axis. Optimal treatment for precocious puberty is not yet defined but might include the use of higher GnRH analog doses for the first 6 months, followed by a decrease to the
Volume 117 Number 3 minimal effective dose that maintains suppression. 13 Alternatively, optimal initial therapy may be a combination of G n R H agonist and antigonadal or antisteroidal agents. We conclude that G n R H analog therapy is not indicated for every patient with precocious puberty. Although one may consider G n R H analog therapy for psychosocial reasons, such as repetitive menses at an extremely early age, behavioral changes, and vulnerability to sexual abuse, these situations require individualized and informed decisions. W e believe that the use of G n R H analogs for precocious puberty should be conservative until further studies confirm the long-term benefits and safety of these agents.
REFERENCES
1. Thamdrup E. Precocious sexual development. Springfield, 111.: Charles C Thomas, 1961. 2. Sigurjonsdottir TJ, Hayles AB. Precocious puberty: a report of 96 cases. Am J Dis Child 1968;115:309-21. 3. Mussam D, Dewhurst J, Grant DB. Precocious puberty: a follow-up study. Arch Dis Child 1984;59:77-8. 4. Kupperman HS, Epstein JA. Medroxyprogesterone acetate in the treatment of constitutional sexual precocity. J Clin Endocrinol Metab 1962;22:456-8. 5. Lee PA. Medroxyprogesterone therapy for sexual precocity in girls. Am J Dis Child 1981;135:443-5. 6. Kauli R, Pertzelan A, Prager-Lewin R, Grunebaum M, Laron Z. Cyproterone acetate in treatment of precocious puberty. Arch Dis Child 1976;51:202-8. 7. Sorgo W, Kiraly E, Homoki J, et al. The effects of cyproterone acetate on statural growth in children with precicious puberty. Acta Endocrinol (Copenh) 1987;115:44-56. 8. Comite F, Cutler GB, Rivier J, Vale WW, Loriaux DL, Crowley WF. Short-term treatment of idiopathic precocious puberty with a long-acting analogue of luteinizing hormonereleasing hormone: a preliminary report. N Engl J Med 1981;305:1546-50. 9. Comite F, Cassorla F, Barnes KM, et al. Luteinizing hormone releasing hormone analogue therapy for central precocious puberty: long-term effect on somatic growth, bone maturation, and predicted height. JAMA 1986;255:2613-6. 10. Styne DM, Harris DA, Egli CA, et al. Treatment of true precocious puberty with a potent luteinizing hormone-releasing factor agonist: effect on growth, sexual maturation, pelvic sonography, and the hypothalamic-pituitary-gonadal axis. J Clin Endocrinol Metab 1985;61:142-51. 11. Boepple PA, Mansfield M J, Wierman ME, et al. Use of a potent, long-acting agonist of gonadotropin-releasing hormone in the treatment of precocious puberty. Endocrinol Rev 1986; 7:24-33. 12. Lin TH, LePage ME, Henzl M, Kirkland JL. Intranasal nafarelin: an LH-RH analogue treatment of gonadotropindependent precocious puberty. J PEDIATR 1986;109:954-8. 13. Pescovitz OH, Comite F, Hench K, et al. The NIH experience with precocious puberty: diagnostic subgroups and response to short-term luteinizing hormone releasing hormone analogue therapy. J PEDIATR 1986;108:47-54. 14. Luder AS, Holland F J, Costigan DC, Jenner MR, Wielgosz G, Fazekas ATA. lntranasal and subcutaneous treatment of cen-
Preserving height potential in precocious puberty
369
tral precocious puberty in both sexes with a long-acting analog of luteinizing hormone-releasing hormone. 3 Clin Endocrinol Metab 1984;58:966-72. 15. Stanhope R, Adams J, Brook CGD. The treatment of central precocious puberty using an intranasal LHRH analogue (buserelin). Clin Endocrinol 1985;22:795-806. 16. Mansfield MJ, Beardsworth DE, Loughlin JS, et al. Long-term treatment of central precocious puberty with a long-acting analogue of luteinizing hormone-releasing hormone: effect on somatic growth and skeletal maturation. N Engl J Med 1983;309:1286-90. 17. Lee PA, Page JG, Leuprolide Study Group. Effects of leuprolide in the treatment of central precocious puberty. J PEDIATR 1989;114:321-4. 18. Rosenfield RL, Garibaldi LR, Moll GW, Watson AC, Burstein S. The rapid ovarian secretory response to pituitary stimulation by the gonadotropin-releasing hormone agonist nafarelin in sexual precocity. J Clin Endocrinol Metab 1986; 63:1386-9. 19. Roche AF, Eyman SL, Davila GH. Skeletal age prediction. J PI?DIATR 1971 ;78:997-1003. 20. Bayley N, Pinneau SR. Tables for predicting adult height from skeletal age: revised for use with the Greulieh-Pyle hand standards. J P~DtATR 1952;40:423-41. 21. Zaehmann M, Sobradilto B, Frank M, Friseh H, Prader A. Bayley-Pinneau, Roche-Wainer-Thissen, and Tanner height predictions in normal children and in patients with various pathologic conditions. J PEDIATR 1978;93:749-55. 22. Roche AF, Wainer H, Thissen D. The RWT method for the prediction of adult stature. Pediatrics 1975;56:1026-33. 23. Tanner JM, Goldstein H, Whitehouse RH. Standards for children's height at ages 2-9 years allowing for height of parents. Arch Dis Child 1970;45:755-62. 24. Salardi S, Orsini LF, Cacciari E, et al. Pelvic ultrasonography in girls with precocious puberty, congenital adrenal hyperplasia, obesity, or hirsutism. J PED1ATR 1988;112:880-7. 25. Meisels A. The maturation value. Aeta Cytol 1967;11:249. 26. Barnes RB, Rosenfield RL, Burstein S, Ehrmann DA. Pituitary-ovarian responses to nafarelin testing in polycystic ovary syndrome. N Engl J Med 1989;320:559-65. 27. Furlanetto RW. Pitfalls in somatomedin-C radioimmunoassay. J Clin Endocrinol Metab 1982;54:1084-8. 28. Starr JI, Horwitz DL, Rubenstein AH, Mako ME. Insulin, proinsulin and C-peptide. In: Jaffe B, Behrman HR, eds. Methods of hormone radioimmunoassay. 2nd ed. New York: Academic Press, 1979:613-42. 29. Burstein S, Grumbach MM, Kaplan SL, Li CH. Immunoreactive activity and receptor binding of mixed recombinants of human growth hormone and chorionic somatomammatropin. Proc Natl Acad Sci USA 1978;75:5391-4. 30. Snedecor GW, Cochran WG. Statistical methods. Ames: Iowa State University Press, 1967. 31. Manasco PK, Pesocvitz OH, Hill SC, et al. Six-year results of luteinizing hormone releasing hormone (LHRH) agonist treatment in children with LHRH-dependent precocious puberty. J PEDIATR 1989;115:105-8. 32. Crudo DF, Wilson BE, Poth MA. Failure of leuprolide to increase predicted height in some patients with precocious puberty [Abstract 446]. Pediatr Res 1988;23:275A. 33. Monroe SE, Blumenfeld Z, Andreyko JL, Schriock E, Henzl MR, Jaffe RB. Dose-dependent inhibition of pituitary-ovarian function during administration of a gonadotropin-releasing
370
Kreiter et al.
The Journal of Pediatrics September 1990
hormone agonist analog (nafarelin). J Clin EndocrinolMetab 1986;63:1334-41. 34. Pescovitz OH, Rosenfeld RG, Hintz RL, et al. SomatomedinC in accelerated growth of children with precocious puberty. J PED~ATR1985;107:20-5. 35. Harris DA, Van Vliet G, Egli CA, et al. Somatomedin-C in normal puberty and in true precocious puberty before and after treatment with a potent luteinizing hormone-releasing hormone agonist. J Clin Endocrinol Metab 1985;61:152-9. 36. Ross JL, Long LM, Skerda M, et al. The effect of low dose
ethinyl estradiol on six-month growth rates and predicted height of patients with Turner syndrome. J PEDIATR 1986; 109:950-3. 37. Martinez A, Heinrich J J, Domene H, et al. Growth in Turner's syndrome: long-term treatment with low-dose ethinyl estradiol. J Clin Endocrinol Metab 1987;65:253-7. 38. Fontoura M, Brauner R, Prevot C, Rappaport R. Precocious puberty in girls:early diagnosisof a slowlyprogressivevariant. Arch Dis Child 1989;64:1170-6.
BOUND VOLUMES AVAILABLE TO SUBSCRIBERS Bound volumes of the 1990 issues of THE JOURNAL OF PEDIATRICSare available to subscribers (only) from the Publisher, at a cost of $48.00 ($64.00 international) for Vol. 116 (January-June) and Vol. 117 (July-December), shipping charges included. Each bound volume contains subject arid author indexes, and all advertising is removed. Copies are shipped within 60 days after publication of the last issue in the volume. The binding is durable buckram, with the Journal name, volume number, and year stamped in gold on the spine. Payment must accompany all orders. Contact Mosby-Year Book, Inc., Circulation Department, 11830 Westline Industrial Dr., St. Louis, MO 63146-3318, USA/800-325-4177, ext. 7351. Subscriptions must be in force to qualify. Bound volumes are not available in place of a regular Journal subscription.
70) True precocious puberty appears to be caused by premature maturation of the normal hypothalamic-pituitary-gonadal axis. As a consequence, secondary sexual characteristics develop early and skeletal maturation and linear growth accelerate. The only permanent physical sequela of precoSupported in part by U.S. Public Health Service grants RR-00055 (General Clinical Research Center), NIADDK 07011-15 (M.K.), and HD-06308 (R.L.R.) and in part by Syntex Corporation. Submitted for publication May 23, 1989; accepted April 11, 1990. Reprint requests: Robert L. Rosenfield, MD, Wyler Children's Hospital, Section of Pediatric Endocrinology, 5841 S. Maryland, Box 118, Chicago, IL 60637. 9/20/21523
364
cious puberty is attenuation of adult stature, and this effect is variable. 1-3 Recently, long-acting gonadotropin releasing hormone agonists have been reported to be effective in arBA CA FSH GnRH HA LH
Bone age Chronologic age Follicle-stimulating hormone Gonadotropin releasing hormone Height age Luteinizing hormone
resting pubertal progression and, unlike previous treatment modalities,47 in improving height potential. 8-17 However, all patients have been treated without consideration of their pretreatment prognosis for adult height. We therefore de-
Volume 117 Number 3 signed a prospective study of height potential in girls with idiopathic precocious puberty in which girls with evidence of impaired adult height potential were treated with the GnRH analog nafarelin and the remainder were followed without therapy. The first objective was to compare each group at presentation to identify discriminating characteristics, and the second, to examine the changes in each group after 2 years of follow-up. METHODS Patients. Twenty-one girls referred to our institution between January 1984 and July 1987 with previously untreated idiopathic true precocious puberty were examined at the University of Chicago General Clinical Research Center. Idiopathic true precocious puberty was defined as the onset of isosexual pubertal changes at a chronologic age of less than 8.0 years, with a sleep-associated rise in serum levels of luteinizing hormone and a pubertal response to GnRH or nafarelin stimulation testing, 18 in the absence of organic abnormality on abdominal ultrasound examination, computed tomography, or magnetic resonance imaging, the last becoming available at mid study. Of the 21 girls, 14 met our criterion for nafarelin treatment and chose to undergo treatment. The criterion for treatment was a deterioration in predicted height as indicated by a loss of ~ 5 cm on sequential bone age determinations at least 5 months apart (n = 3) or a predicted adult height of less than 152.5 cm (n = 11). We refer to these girls as sexually precocious and with a prognosis of impaired height (group 1). The seven girls with a prognosis of unimpaired height who did not meet the criterion for treatment at initial examination were observed without therapy (group 2). Evaluations. Initial evaluation included height, Tanner staging, determination of plasma levels of estradiol, vaginal smear, BA determination, pelvic ultrasonography, somatomedin C concentration, and GnRH or nafarelin stimulation testing. In addition, all patients in group 1, and five of seven in group 2, had nocturnal growth hormone determinations. After the institution of therapy, group 1 was followed at a minimum of 3-month intervals, with height, Tanner stage, plasma estradiol level, and vaginal smear monitored at each visit. Yearly height velocity and 2xBA/ACA ratio were computed at 6-month intervals in the first year and at the end of 12 months in the second year. At 6, 12, and 24 months of therapy, a BA radiograph, pelvic ultrasound examination, and determination of serum somatomedin-C level were performed. Basal levels of LH, follicle-stimulating hormone, and estradiol were assayed at the initial evaluation and after 6, 12, and 24 months of therapy. A GnRH or nafarelin stimulation test was performed at the initial evaluation and after 6 and 12 months of therapy. For an assessment of response to therapy by an individual patient, either
Preserving height potential in precocious puberty
365
a GnRH or a nafarelin test was consistently used. The peak L H and F S H responses during GnRH and nafarelin tests were pooled for intergroup comparisons because the peak responses to these agents are similar during puberty. 18 Urinalysis, hemogram, multiphasic serum chemistry pro fie, and serum prolactin level were obtained at each visit to monitor for toxic effects. Fasting insulin levels were monitored at the initial evaluation and at 6 and 12 months to examine the effect of nafarelin therapy on the insulin resistance of puberty. Group 2 was generally followed at 6-month intervals. Physical examination was performed and BA, vaginal smear, and plasma estradiol level were obtained at each visit. Treatment regimen. Group 1 was treated with nafarelin ([6-D-(2-naphthyl)alanine]-GnRH acetate; Synarel), generously supplied by Syntex Corp. Nafarelin therapy was initiated with an intranasal dosage of 800 # g / d a y in three patients and 1200 #g/day in the remaining 11, divided into two daily administrations. Dose adjustments were made at 3-month intervals if there was evidence of incomplete suppression of puberty. If control was not adequately achieved with 1600 ~g/day, nafarelin was administered subcutaneously at a dose of 4 #g/kg/day. The specific indications for dose adjustments were an elevated plasma estradiol level (9 to 23 pg/ml in six patients and 56 pg/ml in one patient; about half of these patients had vaginal eornification), vaginal cornification (n = 1) or LH >15 I U / L (n = 2) as isolated findings, and the clinical progression of puberty (n = 1). Only four patients required subcutaneous therapy (after 7 months in three and 20 months in one). Laboratory methods. Bone age was determined (by M.K., S.B., R.L.R.) from the Greulich and Pyle atlas by the method of Roche et a1.19 Height predictions were made according to the method of Bayley and Pinneau 2~because this has been shown to be the most accurate method of height prediction in precocious puberty. 21 In those patients with an initial BA of less than 7 years (n = 3), however, the RocheWainer-Thissen method a2 was used for each of that patient's height predictions to maintain consistency. Blinded scoring of BA by the three examiners showed good correlation. Target heights were calculated from mid-parental height by the method of Tanner et al. 23 Real-time pelvic ultrasound examinations were evaluated for average maximal ovarian diameter and ovarian architecture. A modification of the method of Salardi et al. 24 was used to classify ovarian architecture: class I--homogeneous appearance, no follicles identified; class II--nonhomogeneous, three or fewer small follicles 8 mm or less in diameter, per greatest diameter per ovary; class IlI--minifollicular, 4 to 9 follicles 8 mm or less in diameter; and class IV--follicular, 1 follicle greater than 8 mm but 20 mm or less in diameter. A vaginal maturation score was calculated
366
Kreiter et al.
The Journal of Pediatrics September 1990
T a b l e I. Clinical data Group 2
Group 4 Before therapy Mean • SEM Age (yr) Breast stage Pubic hair stage Height (cm) Height growth velocity (cm/yr) Bone age (yr) ABA/2xCA ratio BA/HA ratio Height prediction (cm)
6.1 3.2 2.4 123.0 9.9 9.8 1.8 1.4 150.7
• 0.5 • 0.1 • 0.2 • 2.5 + 0.8 • 0.4 ___ 0.4 • 0.06 • 2.1
After 2 yr of therapy
After 2 yr of observation
At entry
(n)
Mean • SEM
(n)
(14) (14) (14) (14) (14) (14) (14) (14) (14)
8.3 • 0.5 2.9 • 0.3 2.8 4- 0.2 135.4 • 2.4 5.2 • 0.6 11.5 • 0.4 0.7 • 0.1 1.2 + 0.04 152.7 • 2.0
(14) (14) (14) (14) (14)I" (13)I (13)* (13) (13)*
M e a n _+ $EM 5.8 3.0 1.6 121.5 10.5 7.9 1.5 1.0 165.4
• 0.8 ___0.0 • 0.3 + 6.5 • 1.1 • 1.1 _+ 0.4 • 0.05 • 3.0
(n) (7) (7) (7) (7) (7) (7) (7) (7)w (7)w
M e a n 4- $EM 7.9 • 3.1 • 2.7 • 140.7 • 8.7 • 10.5 • 1.2 • 1.0 • 168.7 •
0.5 0.3 0.5 3.1 0.6 0.4 0.2 0.03 4.1
(n) (7) (7) (7)* (7) (7)w (7) (7)
(7)11 (6)*,$
*p <0.05, within a group from initial evaluation. tP <0.005, within a group from initial evaluation. :~p <0.05, betweengroups for comparable observationtimes. w <0.02, betweengroups for comparableobservationtimes. liP <0.005, betweengroups for comparableobservationtimes.
T a b l e II. Endocrine data Group 2
Group I Before therapy Mean • SEM Basal estradiol (pg/ml) Basal LH (IU/L) ALH (IU/L) Basal FSH (IU/L) ~xFSH (IU/L) Vaginal maturation score Average maximal ovarian diameter (mm) Ovarian classification score Somatomedin C (U/ml)
46.7 7.4 68.6 3.8 11.7 49.8 26.6
• • • • • • •
16.9 1.6 10.6 0.3 1.6 4.3 2.0
2.6 +_ 0.2 2.2 • 0.3
After 2 yr of therapy
After 2 yr of observation
At entry
(n)
Mean • SEM
(n)
(14) (14) (14) (13) (13) (12) (10)
8.5 • 3.6 7.2 _ 0.8 -1.8 + 0.3 -27.8 • 6.5 28.0 + 2.3
(14)* (13)
(10) (14)
2.2 + 0.1 2.0 • 0.2
(9) (14)
(12)I (12):~ (9)
Mean • SEM 14.8 5.6 54.2 3.6 14.5 45.3 26.8
• 4.2 4- 1.2 • 24.9 • 0.5 • 3.7 • 9.2 ___ 5.0
3.4 • 0.5 2.2 • 1.1
(n)
Mean _+ SEM
(n)
(7) (6) (6)
15.7 • 10.2
(5)
33.5 • 13.0
(6)
1.8 • 0.6
(6)
(5) (5) (6)
(4) (4) (4)
*p <0.05, within a group from initial evaluation. l"p <0.02, within a group from initial evaluation. :~p <0.005, within a group from initial evaluation.
from the vaginal smear as the sum of the number of superficial cells and half the number of intermediate cells. 25
of Furlanetto. 27 Prolactin was measured by means of a ra-
Serum L H and F S H levels were determined with highly
geles, Calif.). Fasting insulin levels were measured by a double-antibody radioimmunoassay, 2s as were growth hormone levels with the use of polyelonal antiserum. 29
specific and sensitive radioimmunoassays modified to measure bioactive species preferentially. 26 The minimal detectable levels averaged 0.1 n g / m l (0.78 I U / L ) of the L H standard I-2 and 0.1 n g / m l (0.39 I U / L ) of the F S H standard 1-3 (National H o r m o n e and Pituitary Program, Na-
dioimmunoassay kit (Diagnostic Products Corp., Los An-
Consent. This study was approved by the University o f Chicago Hospitals and Clinics Clinical Investigation Committee. W r i t t e n informed consent was obtained from all
tional Institute of Diabetes and Digestive and Kidney
parents and assent from all children.
Diseases). Plasma estradiol levels were determined by a ra-
Statistical analysis. Student unpaired and paired t tests with the Bonferroni correction for multiple comparisons
dioimmunoassay the sensitivity of which averaged 2.7 p g / ml. 26 S o m a t o m e d i n C levels were determined by the method
were used as appropriate. 3~ The M a n n - W h i t n e y and
Volume 117 Number 3
Preserving height potential in precocious puberty
Wilcoxon signed rank tests were similarly used for nonparametric data and to normalize the distribution of parametric data, as indicated. Comparisons are presented as the mean ___SEM. T h e p values are two tailed. Actual sample size (n) varied slightly for some measurements because of incomplete data collection.
Height velocity was unchanged after 6 months of nafarelin therapy (9.9 _+ 0.8 vs 8.2 +_ 0.6 cm/yr) but decreased significantly during the second 6 months to the rate at which it remained during the second year of therapy (Table I) (p <0.002). Girls whose BA was less than the group mean of 9.8 years had a reduction in predicted height during the first 6 months of therapy, compared with those whose BA was greater than the group mean ( - 2 . 6 + 1.7 vs 2.1 _+ 0.9 cm; p <0.05). This difference was not related to hormonal factors. Thereafter the predicted height rose progressively during therapy. Overall, it was 150.7 _+ 2.1 cm before and i52.7 +_ 2.0 cm after 24 months of therapy (p <0.05, Wilcoxon signed rank test) (Table I). The ABA/AxCA ratio fell more than 50% in the course of the 2-year treatment period (p <0.05). The basal LH level was unchanged in response to therapy; however, the basal FSH level decreased significantly during nafarelin therapy. The LH and F S H responses to G n R H or nafarelin stimulation decreased by 6 months and remained low thereafter. The LH response was 68.6 + 10.6 I U / L at baseline and 8.7 + 2.5 I U / L at 6 months. The F S H response was 11.7 _+ 1.6 I U / L and 1.2 _+ 0.5 I U / L , respectively, a t the same time points. Estradiol levels decreased significantly in response to nafarelin therapy. The basal estradiol level at the start of therapy averaged 46.7 _+ 16.9 pg/ml and fell to 6.5 +_ 1.1 by 6 months (p <0.05), after which it remained suppressed. Vaginal maturation scores also decreased by 6 months of therapy, from 49.8 _+ 4.3 to 23.9 +__6.6 (p <0.001), and remained low thereafter. Ovarian long diameter and classification scores and somatomedin C levels were unchanged during nafarelin therapy (Table II). Group 2. Breast stage, plasma estradiol level, and vaginal maturation score showed no significant changes during the observation period in group 2. Pubic hair stage, however, continued to advance (p <0.05). None of the premenarcheal girls subsequently menstruated. Conversely, one of the postmenarcheal girls had intermittent waxing and waning of puberty even after results of serial ovarian ultrasound examinations and progesterone levels suggested that she had ovulated on one occasion. Bone age continued to advance (p <0.05) but growth velocity was unchanged. The ABA/ACA ratio was such that height predictions increased by 3.3 _+1.1 cm during the observation period (p <0.05). Toxic and other Side effects. Nafarelin therapy was well tolerated. There were no drug-related side effects, and routine laboratory screening showed no biochemical evidence of renal, hepatic, or hematopoietic toxic effects (data not shown). Prolactin levels did not change significantly during the treatment period. Fasting glucose levels (baseline vs 12 months of therapy: 87.7 + 2.2 vs 87.2 + 1.9 mg/dl) and
RESULTS Initial evaluation. Age and breast and pubic hair stages at entry were similar in the two groups (Table I). Menses occurred in two of the girls in group 1 before therapy, and in three in group 2. As expected from the criterion for entry into the treatment group, predicted height at entry into the study was significantly less in group 1 than in group 2 (p <0.01). Target heights were similar in group 1 and group 2 (163.1 _+ 1.1 vs 164.3 + 2.1 cm). Thus the mean predicted height at initial evaluation was significantly less than the mean target height in group 1 (p <0.005) but not in group 2. Predicted height was ~ 5 cm less than target height in 12 of the 14 girls in group 1 and in 3 of 7 girls in group 2. The B A / H A ratio at presentation was greater in group 1 girls than in group 2 (p
Study period Group 1. Nafarelin effectively arrested pubertal progression in group 1, in that breast stage, breast diameter, and pubic hair stage were unchanged after 24 months of nafarelin therapy. It was our impression that although breast measurements did not regress, breast tissue became atrophic, and most girls had a clear reduction of breast prominence. Menses ceased (after withdrawal bleeding in the postmenarcheal patients).
367
368
Kreiter et al.
insulin levels (15.1 ___ 1.8 vs 13.5 +_ 2.2 ~tU/ml) were unchanged by nafarelin therapy. Sinus clouding was reported on foUow-up magnetic resonance imaging in four patients, one of whom was receiving subcutaneous therapy, but none had clinical symptoms. One of these patients also had clouding on pretreatment films. DISCUSSION Treatment modalities for precocious puberty have sought to halt or reverse deterioration of height potential. Until recently, agents available for treatment of precocious puberty have been largely unsuccessful in improving the final height), 7 GnRH agonists have been reported to be well tolerated and effective. In girls with precocious puberty and impaired-height prognosis (group 1), we found that nafarelin therapy effectively suppressed the pituitary-gonadal axis and arrested breast development and menses, consistent with the findings of others using GnRH analogs), 9, 11-16Height velocity decreased and ABA/ACA ratio and predicted height improved significantly, as other investigators have found while employing other analogs. 9' I l, 12-16Manasco et al. 31 reported that height predictions increased steadily at the rate of 3 cm/yr as therapy was prolonged. In contrast to previous reports, we found that the decrease in height velocity did not occur until the second 6 months of therapy. Furthermore, during the first 6 months of therapy, predicted height in the subset of girls in group 1 with BAs less than 9.8 years initially deteriorated. Predicted height and ABA/ACA ratio improved during the second year of therapy. Others have reported that predicted height does not necessarily improve during the short term, 9, 32 but whether their patients had this transient decrease during the induction of treatment is unknown. This initial setback in predicted height could have been due to the stimulation of estradiol secretion that occurs during induction of GnRH agonist therapy. Alternatively, this simply may have represented a residual manifestation of unsuppressed puberty. It is unlikely that our GnRH agonist dose was inadequate, because we used intranasal doses fourfold to sixfold greater than are required to suppress pituitary function in adult women, 33 and control of puberty was as good as that of the other subset at the routine examination 3 months after the start of therapy. Unlike Pescovitz et al. 34 and Harris et al., 35 who found significant decreases in somatomedin C levels during GnRH analog therapy, we found that somatomedin C values did not change during nafarelin therapy. An important consideration in evaluating such differences between our results with GnRH analog therapy and those of others is that we selected for nafarelin treatment those girls most severely affected by their sexual precocity. Other studies may have included patients similar to those in our
The Journal of Pediatrics September 1990
group 2, in whom the neuroendocrine "drive" for puberty was relatively mild and thus more easily suppressible. It has long been recognized that idiopathic true precocious puberty only variably results in subnormal adult height. 1"3 We found that sexually precocious girls with an unimpaired-height prognosis (group 2) could be distinguished from those with an impaired-height prognosis in the early stages of evaluation. On an individual basis, a B A / H A ratio of less than 1.2 was almost always associated with a good height prognosis. In addition, a predicted height comparable to target height and a tendency toward lower plasma estradiol levels were more characteristic of this group. The pace of pubertal progression seemed to be slower as well; however, this was difficult to document because Tanner stage I I | describes most of pubertal breast development. Without therapy, group 2 girls maintained an acceptable height potential. In fact, height predictions increased significantly, although the mean B A / H A ratio remained constant at 1.0. The reasons for possible improvement in height prognosis are not readily apparent. One possible explanation is that low circulating levels of estradiol in a very narrow range may be stimulatory to growth. 36 However, although low doses of ethinyl estradiol increase height velocity, improvement in predicted height has not been confirmedS To test the possibility that the improved height predictions in our untreated girls may have been the result of lesser estrogen stimulation, we compared various estrogenic effects. Although plasma estradiol levels were lower in group 2 at the initial visit, this difference was not sustained and vaginal maturation indexes were similar. Recently Fontoura et al. 38 found significantly lower plasma estradiol levels in a group of girls with a slowly progressive variant of idiopathic precocious puberty and stable adult height potential. We found no clinically significant side effects from nafarelin spray therapy, other than possible chronic irritation of the nasal mucosa, about which our data were inconclusive. The long-term safety of GnRH analogs is still largely unknown; the potential effects of these agents on bone density, on subsequent recovery of the hypothalamicpituitary-gonadal axis and reproductive potential, as well as on possible oncogenicity, require ongoing investigation. Early diagnosis and institution of therapy in those patients with precocious puberty and a poor height prognosis seem indicated before substantial impairment of height potential occurs. When GnRH agonist therapy is instituted, our findings suggest the need to achieve rapid (within 6 weeks) and complete suppression of the pituitary-gonadal axis. Optimal treatment for precocious puberty is not yet defined but might include the use of higher GnRH analog doses for the first 6 months, followed by a decrease to the
Volume 117 Number 3 minimal effective dose that maintains suppression. 13 Alternatively, optimal initial therapy may be a combination of G n R H agonist and antigonadal or antisteroidal agents. We conclude that G n R H analog therapy is not indicated for every patient with precocious puberty. Although one may consider G n R H analog therapy for psychosocial reasons, such as repetitive menses at an extremely early age, behavioral changes, and vulnerability to sexual abuse, these situations require individualized and informed decisions. W e believe that the use of G n R H analogs for precocious puberty should be conservative until further studies confirm the long-term benefits and safety of these agents.
REFERENCES
1. Thamdrup E. Precocious sexual development. Springfield, 111.: Charles C Thomas, 1961. 2. Sigurjonsdottir TJ, Hayles AB. Precocious puberty: a report of 96 cases. Am J Dis Child 1968;115:309-21. 3. Mussam D, Dewhurst J, Grant DB. Precocious puberty: a follow-up study. Arch Dis Child 1984;59:77-8. 4. Kupperman HS, Epstein JA. Medroxyprogesterone acetate in the treatment of constitutional sexual precocity. J Clin Endocrinol Metab 1962;22:456-8. 5. Lee PA. Medroxyprogesterone therapy for sexual precocity in girls. Am J Dis Child 1981;135:443-5. 6. Kauli R, Pertzelan A, Prager-Lewin R, Grunebaum M, Laron Z. Cyproterone acetate in treatment of precocious puberty. Arch Dis Child 1976;51:202-8. 7. Sorgo W, Kiraly E, Homoki J, et al. The effects of cyproterone acetate on statural growth in children with precicious puberty. Acta Endocrinol (Copenh) 1987;115:44-56. 8. Comite F, Cutler GB, Rivier J, Vale WW, Loriaux DL, Crowley WF. Short-term treatment of idiopathic precocious puberty with a long-acting analogue of luteinizing hormonereleasing hormone: a preliminary report. N Engl J Med 1981;305:1546-50. 9. Comite F, Cassorla F, Barnes KM, et al. Luteinizing hormone releasing hormone analogue therapy for central precocious puberty: long-term effect on somatic growth, bone maturation, and predicted height. JAMA 1986;255:2613-6. 10. Styne DM, Harris DA, Egli CA, et al. Treatment of true precocious puberty with a potent luteinizing hormone-releasing factor agonist: effect on growth, sexual maturation, pelvic sonography, and the hypothalamic-pituitary-gonadal axis. J Clin Endocrinol Metab 1985;61:142-51. 11. Boepple PA, Mansfield M J, Wierman ME, et al. Use of a potent, long-acting agonist of gonadotropin-releasing hormone in the treatment of precocious puberty. Endocrinol Rev 1986; 7:24-33. 12. Lin TH, LePage ME, Henzl M, Kirkland JL. Intranasal nafarelin: an LH-RH analogue treatment of gonadotropindependent precocious puberty. J PEDIATR 1986;109:954-8. 13. Pescovitz OH, Comite F, Hench K, et al. The NIH experience with precocious puberty: diagnostic subgroups and response to short-term luteinizing hormone releasing hormone analogue therapy. J PEDIATR 1986;108:47-54. 14. Luder AS, Holland F J, Costigan DC, Jenner MR, Wielgosz G, Fazekas ATA. lntranasal and subcutaneous treatment of cen-
Preserving height potential in precocious puberty
369
tral precocious puberty in both sexes with a long-acting analog of luteinizing hormone-releasing hormone. 3 Clin Endocrinol Metab 1984;58:966-72. 15. Stanhope R, Adams J, Brook CGD. The treatment of central precocious puberty using an intranasal LHRH analogue (buserelin). Clin Endocrinol 1985;22:795-806. 16. Mansfield MJ, Beardsworth DE, Loughlin JS, et al. Long-term treatment of central precocious puberty with a long-acting analogue of luteinizing hormone-releasing hormone: effect on somatic growth and skeletal maturation. N Engl J Med 1983;309:1286-90. 17. Lee PA, Page JG, Leuprolide Study Group. Effects of leuprolide in the treatment of central precocious puberty. J PEDIATR 1989;114:321-4. 18. Rosenfield RL, Garibaldi LR, Moll GW, Watson AC, Burstein S. The rapid ovarian secretory response to pituitary stimulation by the gonadotropin-releasing hormone agonist nafarelin in sexual precocity. J Clin Endocrinol Metab 1986; 63:1386-9. 19. Roche AF, Eyman SL, Davila GH. Skeletal age prediction. J PI?DIATR 1971 ;78:997-1003. 20. Bayley N, Pinneau SR. Tables for predicting adult height from skeletal age: revised for use with the Greulieh-Pyle hand standards. J P~DtATR 1952;40:423-41. 21. Zaehmann M, Sobradilto B, Frank M, Friseh H, Prader A. Bayley-Pinneau, Roche-Wainer-Thissen, and Tanner height predictions in normal children and in patients with various pathologic conditions. J PEDIATR 1978;93:749-55. 22. Roche AF, Wainer H, Thissen D. The RWT method for the prediction of adult stature. Pediatrics 1975;56:1026-33. 23. Tanner JM, Goldstein H, Whitehouse RH. Standards for children's height at ages 2-9 years allowing for height of parents. Arch Dis Child 1970;45:755-62. 24. Salardi S, Orsini LF, Cacciari E, et al. Pelvic ultrasonography in girls with precocious puberty, congenital adrenal hyperplasia, obesity, or hirsutism. J PED1ATR 1988;112:880-7. 25. Meisels A. The maturation value. Aeta Cytol 1967;11:249. 26. Barnes RB, Rosenfield RL, Burstein S, Ehrmann DA. Pituitary-ovarian responses to nafarelin testing in polycystic ovary syndrome. N Engl J Med 1989;320:559-65. 27. Furlanetto RW. Pitfalls in somatomedin-C radioimmunoassay. J Clin Endocrinol Metab 1982;54:1084-8. 28. Starr JI, Horwitz DL, Rubenstein AH, Mako ME. Insulin, proinsulin and C-peptide. In: Jaffe B, Behrman HR, eds. Methods of hormone radioimmunoassay. 2nd ed. New York: Academic Press, 1979:613-42. 29. Burstein S, Grumbach MM, Kaplan SL, Li CH. Immunoreactive activity and receptor binding of mixed recombinants of human growth hormone and chorionic somatomammatropin. Proc Natl Acad Sci USA 1978;75:5391-4. 30. Snedecor GW, Cochran WG. Statistical methods. Ames: Iowa State University Press, 1967. 31. Manasco PK, Pesocvitz OH, Hill SC, et al. Six-year results of luteinizing hormone releasing hormone (LHRH) agonist treatment in children with LHRH-dependent precocious puberty. J PEDIATR 1989;115:105-8. 32. Crudo DF, Wilson BE, Poth MA. Failure of leuprolide to increase predicted height in some patients with precocious puberty [Abstract 446]. Pediatr Res 1988;23:275A. 33. Monroe SE, Blumenfeld Z, Andreyko JL, Schriock E, Henzl MR, Jaffe RB. Dose-dependent inhibition of pituitary-ovarian function during administration of a gonadotropin-releasing
370
Kreiter et al.
The Journal of Pediatrics September 1990
hormone agonist analog (nafarelin). J Clin EndocrinolMetab 1986;63:1334-41. 34. Pescovitz OH, Rosenfeld RG, Hintz RL, et al. SomatomedinC in accelerated growth of children with precocious puberty. J PED~ATR1985;107:20-5. 35. Harris DA, Van Vliet G, Egli CA, et al. Somatomedin-C in normal puberty and in true precocious puberty before and after treatment with a potent luteinizing hormone-releasing hormone agonist. J Clin Endocrinol Metab 1985;61:152-9. 36. Ross JL, Long LM, Skerda M, et al. The effect of low dose
ethinyl estradiol on six-month growth rates and predicted height of patients with Turner syndrome. J PEDIATR 1986; 109:950-3. 37. Martinez A, Heinrich J J, Domene H, et al. Growth in Turner's syndrome: long-term treatment with low-dose ethinyl estradiol. J Clin Endocrinol Metab 1987;65:253-7. 38. Fontoura M, Brauner R, Prevot C, Rappaport R. Precocious puberty in girls:early diagnosisof a slowlyprogressivevariant. Arch Dis Child 1989;64:1170-6.
BOUND VOLUMES AVAILABLE TO SUBSCRIBERS Bound volumes of the 1990 issues of THE JOURNAL OF PEDIATRICSare available to subscribers (only) from the Publisher, at a cost of $48.00 ($64.00 international) for Vol. 116 (January-June) and Vol. 117 (July-December), shipping charges included. Each bound volume contains subject arid author indexes, and all advertising is removed. Copies are shipped within 60 days after publication of the last issue in the volume. The binding is durable buckram, with the Journal name, volume number, and year stamped in gold on the spine. Payment must accompany all orders. Contact Mosby-Year Book, Inc., Circulation Department, 11830 Westline Industrial Dr., St. Louis, MO 63146-3318, USA/800-325-4177, ext. 7351. Subscriptions must be in force to qualify. Bound volumes are not available in place of a regular Journal subscription.