0021-972X/96/$03.00/0 Vol. 81, No. 9 Journal of Clmcal and Metabolism Printed in U.S.A. Copyright 0 1996 by The Endocrine Society

Gonadotropin-Releasing Hormone Agonist Treatment of Girls with Constitutional Short Stature and Normal Pubertal Development

JEAN-CLAUDE CAREL, FRlkDliRIQUE HAY, RliGIS COUTANT, DANIlkLE RODRIGUE, AND JEAN-LOUIS CHAUSSAIN Downloaded from https://academic.oup.com/jcem/article/81/9/3318/2651102 by guest on 23 September 2021 INSERM U-342 and Department of Pediatric Endocrinology, University of Paris V, Hbpital Saint Vincent de Paul, Paris, France

ABSTRACT interruption of treatment, age was 14.9 2 1.3 yr (~13.5 yr in all GnRH agonists have been proposed to improve final height in patients), height was 149.1 k 4 cm, and final height prognosis was patients with constitutional short stature. We treated 31 girls, aged 150.6 2 3.6 cm. Final height prognosis was 1 2 2.3 cm greater than 11.9 i 1 yr (mean t- SD), with short stature, recent pubertal onset and pretreatment height prognosis (P < 0.02) and 1.2 k 2.2 cm below the predicted final height of 155 cm or less with depot triptorelin. During height predicted at the end of the treatment (P < 0.01). No major the 23 2 4 months of treatment, bone age progression was 0.6 ? 0.3 side-effect was observed. Height SD score decreased during treatment bone age yr/yr, and growth velocity declined from 7 k 2 to 4 2 0.8 with GnRH agonist from -2.3 ? 0.9 to -2.7 -C 0.7 SD score (P < cm/yr (P < 0.0001). Height prognosis, calculated by the Bayley-Pin- 0.0001). We conclude that 2 yr of depot triptorelin-induced pubertal neau method, progressed from 149.6 k 3.4 to 151.8 2 4 cm at the end delay has a limited effect on near-final height in girls with constitu- of treatment (+2.2 t 2.6 cm; P < 0.0001). When treatment was tional short stature and that the growth benefit observed does not interrupted, growth velocity slightly increased to 4.6 ? 1.6 cm/yr, and currently justify the use of GnRH agonists, given their cost and po- bone age maturation was accelerated: 1.3 2 0.4 bone age yr/yr during tential side-effects. (J Clin Endocrinol Metab 81: 33183322, 1996) the first posttreatment year. At the last visit, 26 k 9 months after

LJBERTY IS associated with increased growth velocity of the fourth treatment year. However, no follow-up after P and bone maturation, leading to fusion of growth cessation of treatment was presented. plates and achievement of final height. Extreme abnormal- We initiated an open trial of long acting GnRH agonist in ities of pubertal timing can affect final height. Premature girls with constitutional short stature, and we report here the exposure to sexual steroids in situations such as precocious results of 31 patients who have reached final or near-final or congenital adrenal hyperplasia is associatedwith height. a reduction of final height, averaging, in the case of preco- cious puberty, 2 SD (-10 cm) (1,2). Conversely, Subjects and Methods deficiency (idiopathic or Kallmann’s syndrome) is associated Patients with prolonged statural growth and a 3-cm increment in final Thirty-one girls were included with the following criteria: height -1 height. However, this increment is attributable to patients SD or less, height prognosis of 155 cm or less, recent pubertal onset treated after age 18 yr, i.e. after 5 yr of pubertal delay (3). defined by breast development (B2-B3) of less than 12-month duration, Several reports have documented the ability of treatment chronological age of 9-14 yr, bone age less than 12.5 yr, and pubertal with GnRH agonist to increase final height in patients with response of LH to GnRH (8). Preliminary data for nine patients have been previously reported (9). central precocious puberty (2, 4-6). However, genetic growth potential is not always fully restored (5), and some Treatment reports have suggestedthat the benefit on final height might be insignificant in patients with pubertal onset occurring Patients were treated with depot triptorelin (Decapeptyl, Ipsen-Bio- tech, France; 3.75 mg every 28 days) as previously described (10). Ad- close to the normal age of puberty (2). equate suppression of the gonadotroph was assessed clinically by the Given their action on final height in precocious puberty, interruption of pubertal development and suppressed LH response to GnRH agonists are candidates for increasing final height in GnRH. The planned treatment duration was 2 yr, but varied from 15-30 constitutional short stature. Municchi et al. (7) treated pa- months. After the end of treatment, growth and bone age were assessed every 6-12 months, and in most patients, a GnRH test was performed tients with short stature with deslorelin in a double blind 6-12 months after interruption of treatment. Data were analyzed in trial. Their preliminary data, concerning 16 patients, indi- patients followed for at least 12 months after discontinuation of the cated a 7.2-cm increase in predicted adult height at the end treatment who had reached a bone age of 13.5 yr or more.

Methods Received September 5, 1995. Revision received February 15, 1995. Rerevision received February 20, 1995. Accepted February 21, 1995. Bone age was evaluated by two of us (J.-L.C. and J.-CC.) according Address all correspondence and requests for reprints to: Dr. Jean- to the method of Greulich and Pyle (ll), and height prognosis was Claude Carel, INSERM U-342, Hapital Saint Vincent de Paul, 82 avenue calculated by the method of Bayley-Pinneau (12). Methods for measure- Denfert Rochereau, 75014 Paris, France. E-mail: [email protected]. ment of plasma estradiol and were previously described

3318 GnRH AGONISTS AND SHORT STATURE 3319

(13). Informed consent was given by the patients and their families to 0.6 ? 0.3 bone age yr/yr of treatment (Fig. 2). This reduction participate in the protocol, which was approved by the local ethical of bone age maturation resulted in a 2.2 + 2.6 cm increase in committee. height prognosis at the end of treatment (Table 1 and Fig. 2; Statistics P < 0.0001). Although the treatment was planned to last 24 months, its average duration was 23 -C4 months. Eight pa- Statistical tests were performed with the Statview- software (Abacus tients decided to interrupt the treatment before completing Concept, Berkeley, CA). Results are expressed as the mean i SD. Paired t tests were used. the 2-yr protocol, mainly becauseof poor growth velocity (2.7 2 2 cm / yr during the portion of the second year of treatment Results they completed). Their data have been maintained in the The initial characteristics of the 31 patients are described present analysis on the basis of the intention to treat. How- in Table 1. The patients were carefully evaluated to rule out ever, removal of these eight patients from the analysis would Downloaded from https://academic.oup.com/jcem/article/81/9/3318/2651102 by guest on 23 September 2021 defined causes of growth retardation; the peak plasma GH not affect the conclusions of our study (not shown). In three in response to a pharmacological stimulation test exceeded patients, treatment was slightly prolonged to 26, 27, and 30 10 ng/mL in all cases(mean GH peak, 22.1 ? 6.5 ng/mL). months, because of poor compliance with the schedule of When appropriate, other investigations ruled out defined out-patient visits. causes of short stature; a karyotype was performed in 5 The patients were followed for an additional 26 +- 9 patients with features suggestive of Turner’s syndrome, a months after interruption of treatment. Clinical pubertal de- jejunal biopsy was performed in 2 patients with abdominal velopment resumed in all patients, and a pubertal LH re- complaints compatible with celiac disease,and bone x-rays sponse to GnRH was documented in 23 patients 9.1 + 7 were performed in 2 patients with hyperlordosis clinically months after interruption of treatment (Table 2). Although suggestive of . All of these investigations bone age at the end of treatment was 11.8 + 0.6 yr, no clear were normal. Birth length adjusted for gestational age was growth spurt was observed (Fig. 1); growth velocity was 4.6 -1.8 t 1.2 SD, and a history of intrauterine growth retarda- t 1.6 cm/yr during the first posttreatment year, 1 ? 0.3 tion (IUGR; birth length, less than -2 SD) was found in 46% cm/yr more than that during the last year of treatment (P < (13 of 28) of the patients (12). Target height, calculated from 0.01). In contrast, bone age maturation was markedly in- midparental height, was 7.8 cm (1.4 SD) below the average creased, averaging 1.3 ? 0.4 bone age yr/yr during the first height of French women (14), indicating a genetic participa- posttreatment year and 1.5 + 0.3 bone age yr / yr during the tion in the short stature of our patients. A GnRH test was entire follow-up period (both P < 0.0001 VS. bone age pro- performed in 29 of 31 patients before treatment and con- gression during treatment; Fig. 2). Growth after interruption firmed the onset of pubertal development, with LH peak of treatment (difference between heights at last visit and at values greater than 5 IU/L in all cases (Table 2). the end of treatment) was inversely correlated with bone age As expected, treatment with GnRH agonist was associated at the end of treatment (r = 0.73; P < 0.0001). with an interruption of pubertal development, and the LH Final height prognosis was evaluated in those girls with a response to GnRH was suppressed (Table 2). Growth veloc- bone age of 13.5 yr or more (Table 1). Final height prognosis ity declined from 7 -C2 cm/yr before treatment to 4 ? 0.8 was 1 ~tr:2.3 cm superior to pretreatment height prognosis cm/yr during treatment (P < 0.0001). Growth velocity was (P < 0.02), but was 1.2 2 2.2 cm below the height predicted markedly decreased during the second year of treatment, at the end of treatment (P < 0.01; Table 1 and Fig. 2). Final averaging 3.6 -C 1 cm, with individual values as low as 1.7 height gain, calculated as the difference between pretreat- cm/yr (Fig. 1). This resulted in a reduction of height z score, ment height prognosis and final height prognosis, ranged declining from -2.3 -C 0.9 to -2.7 2 0.7 SD at the end of from -3.2 to +5.7 cm. We analyzed the predictive param- treatment (Table 1). Bone age maturation declined, averaging eter(s) of its variation in a simple linear regression model.

TABLE 1. Characteristics of the 31 patients during and after treatment

Before treatment At the end of treatment -1 yr after treatment At the end of follow-up Target ht (cm) 155.2 % 3.5 [148.5-161.51 Birth length (cm) 47.6 I! 2.1 [42-501 Birth length (SD score) -1.8 2 1.2 [-0.31-4.81 Chronological age (yr) 11.9 L 1 [9.7/141 13.7 2 0.9 [12/15.41 14.8 i 0.9 [12.9/16.41 15.9 2 1.2 [13.9/191 Ht (cm) 133.9 5 4.6 [124/1431 141.3 i 4.2 [131.5/1501 146 ? 4.3 [136.5/153] 149.1 ? 4 [141/156.51 Ht (SD) -2.3 -+ 0.9 [-4/-l] -2.7 f 0.7” L-4.7/-1.31 -2.4 k 0.6 [-4.3/-1.41 -2.2 2 0.6 I-3.4/-1.21 Ht velocity (cm&) 7%2 [4/121 4 2 0.8”ab L2.616.51 4.6 -c 1.6”,’ [1.8/8.31 3.7 k 1.4”~~ [1.2/8.2] Bone age (yr) 10.7 t 0.7 [9/12.51 11.8 2 0.6 [11/131 13.2 2 0.7 [12/151 14.9 L! 1.3 [13.5/181 Ht prognosis (cm) 149.6 2 3.4 [139.5/155.71 151.8 ? 4” [142.3/159.81 150.6 +- 3.6” [144/157.41 Body mass index (kg/m’) 17.4 F 1.5 [13.5/19.71 19.0 z 1.9 [15.9/25.41 19.6 ? 1.9 [15.7/22.61 20.2 k 2 [16.7/25.21 Body mass index (SD score) 0.06 i- 0.86 L-2.3/1.31 0.23 k 0.98 L-1.4/3.11 0.12 2 0.79 L-1.8/1.51 0.10 2 0.84 L-1.3/2.21 Values are the mean 2 SD, with extreme values in brackets. a P < 0.0001 US. pretreatment value. b During treatment. c During the first year after interruption of the treatment; NS us. during treatment. d During the entire posttreatment follow-up period, NS vs. during treatment. e P < 0.02 us. pretreatment value and P < 0.01 us. value at the end of treatment. CAREL ET AL. JCE & M . 1996 Vol81 . No 9

TABLE 2. Hormonal evaluation of the gonadotropic axis before, during, and after depot triptorelin treatment

Before treatment During treatment After treatment Basal plasma LH (III/L) 1.2 5 1.3 [0.2/7.71 0.5 -+ 0.1 [0.3-1.51 2.6 k 2.2 [0.5/9.2] Peak plasma LH after GnRH (III/L) 12.4 2 6.1[5/31.2] 0.6 k 0.3 [0.5/2.11 17.1 t- 13.9 [4.1/731 Basal plasma FSH (III/L) 2.7 I! 1.4 [0.6/5.51 0.8 2 0.5 [O&5/3.3] 3.5 5 1.2 [1.9/7.1] Peak plasma FSH after GnRH (III/L) 8.6 i 2.8 [3.9/14.1] 1.4 t- 1.3 [0.5/6.91 7.5 2 2.3 13.6D3.41 Plasma estradiol (pg/mL) 29 i 19 [10/701 12 2 3 [lo/251 62 2 41 [20/1631 Normal prepubertal values: estradiol, (5-25 pg/mL; LH baseline, 0.2-2.1 III/L; peak LH after GnRH, 1.7-5 III/L; FSH baseline, 0.6-3.4 IU/L; peak FSH after GnRH, 1.4-11.5 III/L. Values are the mean ? SD, with extreme values in brackets. . . Downloaded from https://academic.oup.com/jcem/article/81/9/3318/2651102 by guest on 23 September 2021 .___/II .‘/

” ‘.:lpy BEFORE DURING TREATMENT AFTER TREATMENT TREATMENT FIRST SECOND FIRST TOTAL YEAR YEAR YEAR FOLLOW- UP . e0.48; p

FIG. 3. Correlation between final height gain (difference between final and pretreatment height prognosis) and growth velocity during the second year of treatment in 24 patients who continued the treat- ment for more than 18 months; the regression line and 95% confidence interval are shown.

FIRST YEAR POST WRING AFTER TREATMENT A similar analysis was performed in 18 patients with a TREATMENT TREATMENT FOLLOW-UP bone age of 15 yr or more at the end of follow-up; height 153 B * e 1 T prognosis was 149.2 t- 3.7 cm before treatment, 151.6 t- 3.6 ; 152 - cm at the end of treatment, and 150.2 t 3.6 cm at the end of 2 follow-up (P < 0.02 us. end of treatment; nonsignificant VS. g 151 pretreatment). Actual height at the end of follow-up was E g 150 149.4 2 3.5 cm, identical to pretreatment height prognosis. E We also evaluated whether the therapeutic response dif- 149 7 BEFORE END OF END OF fered between patients with and without IUGR (Table 3). TREATMENT TREATMENT FOLLOW-UP None of the pretreatment variables (age, height, bone age, or FIG. 2. Bone age progression and height prognosis during treatment. height prognosis) differed between the two subgroups. Final A, Bone age progression calculated in bone age year per chronological height gain tended to be smaller in patients with IUGR (0.5 yr. *, P < 0.0001 us. bone age progression during treatment. B, Height +- 2.2 VS. 1.4 + 2.4 cm). However, due to the small number prognosis according to Bayley-Pinneau. *, P < 0.001 us. before treat- ment; **, P < 0.02 us. before treatment and P < 0.01 us. end of of patients in each subgroup, this difference was not treatment. significant. Body massindex increased during the trial (Table 1). How- Growth velocity during the second treatment year was pos- ever, when the results were expressed as a SD score for age itively correlated with final height gain (Fig. 3; r = 0.48; P < (15), no significant variation in body mass index was 0.02), suggesting that a subgroup of patients, who continue observed. to grow despite pubertal blockade, are better responders. This analysis was only possible in patients treated for more Discussion than 18 months. Bone age progression during treatment was We have treated 31 girls with idiopathic short stature with negatively correlated to final height gain (r = 0.35; P = 0.052; a long acting GnRH agonist for an average of 23 months. not shown). However, these two variables are not indepen- Interruption of pubertal development decreasedgrowth ve- dent, as both rely on pretreatment bone age. Pretreatment locity and, to a greater degree, bone age maturation, resulting growth velocity, chronological age, bone age, height or in a significant increase of 2.2 2 2.6 cm in height prognosis height prognosis, treatment duration, growth velocity dur- at the end of the treatment. However, when treatment was ing the first year of treatment, bone age maturation, and interrupted, bone age maturation was accelerated, and half growth velocity after treatment were not significantly cor- of the apparent benefit was lost. related with height gain (not shown). Our conclusions rely in part on the accuracy of height GnRH AGONISTS AND SHORT STATURE 3321

TABLE 3. Effect of treatment on growth prognosis in patients subdivided according to their birth length

All patients Intrauterine growth retardation (n = 13) Normal birth length (n = 15) Birth length (SD score) -1.8 2 1.2 -2.8 t 0.9 [-4.8/-21 -0.9 t 0.5 L-1.5/0.31 Ht prognosis before treatment (cm) 149.6 -c 3.4 [139.5/155.71 148.7 ?I 4.2 [139.5/153.8] 150 + 2.4 [146.3/155.7] Difference in ht prognoses between: End of treatment and before treatment (cm) 2.2 k 2.6 [-5J7.11 1.5 i 2.8 [-5.U5.91 2.8 -+ 2.6 [-0.1/7.11 End of follow-up and before treatment (cm) 1 i 2.2 [-3.2/5.71 0.5 i 2.2 [-3.2/4.6] 1.4 k 2.4 [-2.6/5.7] Values are the mean t SD, with extreme values in brackets. None of the differences between groups is statistically significant. prediction by the Bayley-Pinneau method. Zachmann et al. cm, significantly lower than the pretreatment growth veloc- (16) evaluated the accuracy of several prediction methods ity. This suggests that pharmacological interruption of pu-

and concluded that in normal children and children with berty does not simply revert steroid secretion and growth Downloaded from https://academic.oup.com/jcem/article/81/9/3318/2651102 by guest on 23 September 2021 short stature, the Bayley-Pinneau method is the most accu- plate receptivity to a prepubertal state and that when secre- rate. However, in normal girls aged lo-12 yr, the Bayley- tion of steroid resumes, the growth response is different from Pinneau method underestimated final height by 1.2-2.3 cm, that of a naive prepubertal child. Although the reasons for which, applied to our results, would mean a further reduc- this phenomenon are not clear, it indicates that reports of tion of the growth benefit we observed. In contrast, Municchi predicted height in patients still receiving treatment with et al. (7) observed a 2.7-cm overestimation of growth prog- GnRH analogs should be regarded with caution. nosis in seven placebo-treated patients. Similarly, Bramswig The relatively wide range of the calculated height gain et al. (17) found a 0.8 -+ 3.6-cm overestimation of growth (-5.1 to +7.1 cm; SD = 2.3 cm) in our patients is probably due prognosis in untreated girls with short stature, delayed pu- in part to the relatively wide variation in predicted to ob- berty, and a 2.7-yr delay of bone age over chronological age served height in untreated patients (SD = 3.6 cm) (17). How- (compared to 1.2 yr in our patients). According to these data, ever, growth velocity during the second year of treatment the improvement in final height in our patients could actu- was a factor contributing to height gain. The addition of GH ally average 2-3 cm rather than 1 cm, as determined from to GnRH agonist treatment in patients with constitutional predicted height. short stature might increase growth velocity and final height, What are the reasons for the failure of pubertal blockade but no data concerning final height have been presented to to increase final height markedly in these girls with consti- date (21). tutional short stature? Although significant deviations in In our opinion, the modest, although significant, increase final height from genetic target height are associated with in final height (1 cm, or even 2 cm if we consider the inac- extreme variations in pubertal development (i.e. precocious curacy of the Bayley-Pinneau method) does not justify treat- puberty and ), it is not clear whether smaller ment with GnRH agonists compared to their potential in- variations in the onset of pubertal development contribute to conveniences and side-effects (22). They induce a short term the dispersion of statural height in the general population. decrease in bone mass (23-25), and their long term effects on reduces the peak pubertal growth rate and bone are unknown. The psychological effects of treatment- total pubertal height gain (18), and normal boys entering induced pubertal delay have not been evaluated, but un- puberty at age 11 or 13 yr achieve the same final height (19). treated constitutional delay of puberty is associated with Therefore, the inverse variation of pubertal age and pubertal significant distress (20). We conclude that the effect of GnRH height gain limits the effect of treatments altering the tempo agonists on near-final height in girls with constitutional short of puberty. Longer treatment protocols could result in dif- stature is limited and does not currently justify their use, ferent outcomes. In their study, Municchi et al. (7) conclude given their cost and potential side-effects. that 4 yr of pubertal blockade increase predicted height by 7.2 cm compared to 2.2 cm in our study, but a longer fol- References low-up in this group of patients is needed. However, one factor limiting the duration of treatment is the chronological 1. Sigurjonsdottir TJ, Hayles AB. 1968 Precocious puberty. A report of 96 cases. Am J Dis Child. 115:309-321. age until which puberty can be blocked without inducing 2. Kletter GM, Kelch RP. 1994 Clinical review 60: effects of gonadotropin- psychological distress (20). We limited the duration of treat- releasing hormone analog therapy on adult stature in precocious puberty. J Clin Endocrinol Metab. 79:331-334. ment to 2 yr in our protocol, because we believe that pubertal 3. Uriarte MM, Baron J, Garcia HB, Barnes KM, Loriaux DL, Cutler Jr GB. 1992 development should not be blocked after the age of 14-15 yr The effect of oubertal d&v on adult heieht in men with isolated hvooeona- in girls. In addition, treatment-induced pubertal delay is dotropic hypbgonadism. J’Clin Endocrinzl Metab. 74:436-440. ‘I ” 4. Brauner R, Adan L, Malandry F, Zantleifer D. 1994 Adult height in girls with associated with a transient decrease in the height SD score, idiopathic true precocious piberty. J Clin Endocrinol Metab.w79:41?-420. increasing the difference from peers. 5. Oerter KE, Manasco PK, Barnes KM, Jones J, Hill S, Cutler Jr GB. 1991 Adult One of the unexpected findings of our study was the rapid height in precocious puberty after long-term treatment with deslorelin. J Clin Endocrinol Metab. 731235-1240. progression of bone age after interruption of GnRH agonist 6. Paul DL, Conte FA, Grumback MM, Kaplan SL. 1995 Long-term effect of treatment, causing a rapid decrease in height prognosis. In gonadotropin-releasing hormone agonist therapy in children with true pre- cocious puberty treated at a median age of less than 5 years. J Clin Endocrinol addition, although the average bone age at interruption of Metab. 73:1235-1240. treatment was 11.8 yr, corresponding to the middle of the 7. Municchi G, rose S, Pescovitz OH, Barnes KM, Cassorla FG, Cutler Jr GB. normal pubertal growth spurt [normal growth velocity in 1993 Effects of deslorelin-induced pubertal delay on the growth of adolescents with short stature and normally timed puberty: preliminary results. J Clin 12-yr-old girls, 7 +- 1.6 cm in the French general population Endocrinol Metab. 77~1334-1339. (14)], the peak growth velocity after treatment was 4.6 t 1.6 8. Roger M, Lahlou N, Lindner D, Chaussain J-L. 1992 Gonadotropin-releasing JCE & M . 1996 3322 CAREL ET AL. Vol81 . No 9

hormone testing in pediatrics. In: Ranke MB, ed. Functional endocrinologic and constitutional delay of growth and puberty: accuracy of five different diagnostics in children and adolescents. Mannheim: J & J Verlag; 229-247. methods of height prediction. J Pediatr. 1171886-891. 9. Lindner D, Job J-C, Chaussain J-L. 1993 Failure to improve height prediction 18. Bourguignon JP. 1988 Linear growth as a function of age at onset of puberty in short-stature pubertal adolescents by inhibiting puberty with luteinizing and sex steroid dosage: therapeutic implications. Endocr Rev. 9:467-488. hormone-releasing hormone analogue. Eur J Pediatr. 1521393-396. 19. Marti-Henneberg C, Vizmanos B, Moreno A, et al. 1995 Tempo of puberty 10. Roger M, Chaussain J-L, Berlier I?, et al. 1986 Long-term treatment of male and and final height in boys. Horm Res. 4451:5. female precocious puberty by periodic administration of long-acting prepa- 20. Rowne EC, Shalet SM, Wallace WH, Eminson DM, Price DA. 1991 Final ration of o-Trp%teinizing hormone-releasing hormone microcapsules. J Clin height in girls with untreated constitutional delay in growth and puberty. Eur Endocrinol Metab. 62:670-677. J Pediatr. 150:708-712. 11. Greulich WW, Pyle Sl. 1959 Radiographic atlas of skeletal development of the 21. Saggesse G, Cesaretti G, Barsanti S, Rossi A. 1995 Combination treatment hand and wrist. Stanford: Stanford University Press. with growth hormone and gonadotropin-releasing hormone analogs in short 12. Usher RH, McLean FH. 1969 Intra-uterine growth of liveborn Caucasian normal girls. J Pediatr. 126:468-473. infants at the sea level. J Fed&r. 74:801-806. 22. Rosenfield RL. 1994 Selection of children with precocious puberty for treat- 13. Care1 J-C, Lahlou N, Guazzarotti L, et al. 1995 Treatment of central precocious ment with gonadotropin releasing hormone analogs. J Pediatr. 124:989-991. puberty with depot leuprolide acetate. Eur J Endocrinol. 132699-704. 23. Saggese G, Bertelloni S, Baroncelli GI, Battini R, Franchi G. 1993 Reduction Downloaded from https://academic.oup.com/jcem/article/81/9/3318/2651102 by guest on 23 September 2021 14. Sempe M, Pedron G, Roy P. 1979 Auxologie, methodes et &quences. Paris: of : an effect of gonadotropin releasing hormone analogue treat- Theraplix. ment in central precocious puberty. Eur J l’ediatr. 152717-720. 15. Rolland Cachera MF, Cole TJ, Sempe M, Tichet J, Rossignol C, Charraud A. 24. Scharla SH, Mime HW, Waibel Treber S, et al. 1990 Bone mass reduction after 1991 Body mass index variations: centiles from birth to 87 years. Eur J Clin deprivation by long-acting gonadotropin-releasing hormone agonists Nutr. 45113-21 and its relation to pretreatment serum concentrations of 1,25-dihydroxyvita- 16. Zachmann M, Sobradillo B, Frank M, Frisch H, Prader A. 1978 Bayley- min D,. J Clin Endocrinol Metab. 70:1055-1061. Pinneau, Roth-Wainer-Thissen, and Tanner height predictions in normal chil- 25. Goldray D, W&man Y, Jaccard N, Merdler C, Chen J, Matzkin H. 1993 dren and in patients with various pathologic conditions. J Pediatr. 93:749-755. Decreased bone density in elderly men treated with the gonadotropin-releas- 17. Bramswig JH, Fasse M, Holthoff ML, con Lengerke HJ, van Petrykowski, ing hormone agonist decapeptyl (d-Trp6-GnRH). J Clin Endocrinol Metab. Schellong G. 1990 Adult height in boys and girls with untreated short stature 76:288-290.