0021-972X/92/7402-0436$03.00/0 Journal of Clinical Endocrinology and Metabolism Copyright 0 1992 by The Endocrine Society
Vol. 74, No. 2 Printed in U.S.A.
The Effect of Pubertal Delay with Isolated Hypogonadotropic
on Adult Height Hypogonadism
M. MERCEDES URIARTE, JEFFREY BARON,* HERNAN KEVIN M. BARNES, D. LYNN LORIAUX,* AND GORDON Developmental Endocrinology Branch, National Institute of Child Health National Institutes of Health, Bethesda, Maryland 20892
ABSTRACT. To determine whether delay of puberty alters adult height, we retrospectively evaluated the adult height of 41 patients who met rigorous criteria for the diagnosis of isolated hypogonadotropic hypogonadism. The adult height of these patients was compared with the mean height of normal American men at age 18 in the 1979 National Center for Health Statistics survey and with the mean adult height of 50 male normal volunteers who had been studied on the same wards as the patients with hypogonadotropic hypogonadism. The mean adult height of the men with isolated hypogonadotropic hypogonadism was 179.7 cm, which exceeded the height of the 50 control subjects and the normal American men (both 176.8 cm) by 2.9 cm (P < 0.05 and P < 0.006, respectively). The mean age at treatment to induce puberty was 20.0 yr, which corresponded to a mean delay in the onset of puberty of more than 8 yr relative to normal males. The final height of the men
B. GARCIA, B. CUTLER, and Human
in Men
JR.*
Development,
with hypogonadotropic hypogonadism correlated significantly with the duration of pubertal delay (r = 0.32, P = 0.04). Most of the enhanced mean adult height of the patients with isolated hypogonadotropic hypogonadism was attributable to those patients in whom treatment to induce puberty had been delayed to age 18 or greater. The mean adult height of these patients was 182.4 cm, or 5.6 cm greater than the mean height of the 50 controls or of normal American men (P C 0.001). The mean adult height of patients whose treatment began between 10 and 17 yr of age was 175.0 cm, which did not differ significantly from that of normal men. We conclude that prolonged delay of puberty (6 or more yr) in men with isolated hypogonadotropic hypogonadism is associated with a modest increase (-5 cm) of adult height. (J Clin Endocrinol Metab 74: 436-440,1992)
T
ism (IHH) often experience a delay of puberty as a result of delay in diagnosis and treatment. These patients thus provide a natural experiment on the effect of pubertal delay on adult height. We therefore studied the final height of 41 men who met rigorous criteria for the diagnosis of IHH and who had been evaluated at the NIH between 1959 and 1987.
HE pubertal growth spurt causes an approximate doubling of the prepubertal growth rate and contributes 1518% to adult height (1,2). However, puberty also sets in motion the events that lead to epiphyseal fusion and the cessation of growth. Disorders that advance the onset of puberty, such as congenital adrenal hyperplasia and LHRH-dependent precocious puberty, cause early epiphyseal fusion and impairment of adult height (3,4). Conversely, treatment of LHRH-dependent precocious puberty with a longacting LHRH agonist results in a significant improvement in the adult height when compared to pretreatment predicted height (5-7). These observations raise the possibility that delay of normal puberty might also increase final height. If this hypothesis were correct, then treatment with long-acting LHRH agonists might increase the duration of growth and the final height of children with short stature. Patients with isolated hypogonadotropic hypogonad-
Subjects
and Methods
Subjects We retrospectively analyzed the charts of 41 NIH patients who were evaluated between 1959 and 1987, who had a diagnosis of IHH, and who had achieved adult height at the time of the study. Twenty-eight of the 41 patients had hypogonadotropic hypogonadism and anosmia (Kallmann’s syndrome) (8,9). The remaining 13 patients were considered to have isolated hypogonadotropic hypogonadism based upon prolonged delay of puberty, low or normal basal plasma gonadotropin levels, normal thyroid function, normal adrenal function (in the 12 patients in whom this was tested), normal plasma PRL level, and
Received June 27,1989 Address requests for reprints to: Dr. Gordon B. Cutler, Jr., Building 10, Room 10-N-262, National Institutes of Health, Bethesda, Maryland 20892. * Commissioned officers in the United States Public Health Service.
normal GH response to stimulation
tests (~7 rg/L in the 5
patients in whom this was tested). Thirty-eight of the 41 patients had delayed puberty (failure to enter puberty by age 14). Three of the patients, who had 436
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 01 July 2014. at 02:36 For personal use only. No other uses without permission. . All rights reserved.
PUBERTAL
DELAY
AND ADULT
anosmia and were presumed to have Kallmann’s syndrome because of a positive family history (2 patients) or cryptorchidism (1 patient), received treatment to induce puberty before age 14. Testicular volume in the patients with IHH averaged2.1 + 1.5 ml (range,0.2-5.5 ml) at the time of initial evaluation (10). Five of the patients had unilateral cryptorchidism. Pretreatment testosteronelevels, which were available in 36 of the 41 patients, averaged0.9 f 0.7 nmol/L (range, 0.3-3.2 nmol/L; normal adult range, I-35 limol/L). Pretreatment FSH and LH levels, which were available in 37 patients, averaged2.7 + 1.4 IU/L (range, 0.7-7.0 IU/L) and 3.7 + 2.4 IU/L (range, 0.2-11.0 IU/L), respectively (normal adult ranges:5-25 IU/L for FSH, and 6-26 IU/L for LH). The patients with IHH had a meanageat treatment of 20.0 f 6.2 yr (range, lo-42 yr), and, thus, had a mean delay of puberty of 8.4 yr when comparedto the normal meanpubertal age of onset in boys of 11.6 yr (11). The treatments to induce puberty compriseda variety of regimensof testosterone,hCG, or hCG plus human menopausalgonadotropin. Treatment was often started before referral to NIH. Since most of our patients had considerabledelay in diagnosisand treatment, the goal of therapy was usually to promote virilization or fertility rather than to maximize adult height. Thus, the regimensemployed tended to promote more rapid virilization than would occur during spontaneouspuberty. The control men were 50 normal volunteers [age, 21.5 f 2.3 (SD) yr; range, 19-291who were studied at the NIH between 1978and 1983on the sameinpatient wardsasthe IHH patients. Height measurement A single height measurementby stadiometer was made at each NIH visit for 35 of the 41 subjects. The remaining 6 patients received 10 successiveheight measurementsat their most recent visit. Patients were consideredto have achievedadult height when one or more of the following criteria were met: 1) bone age L 18 yr (12); 2) no increasein height over a period of at least 6 months; 3) 5 yr or more of treatment with androgensor hCG; or 4) chronological age~40 yr. If more than one height measurement was available after achievement of adult height, the average of all values was used except for the six subjects in whom we usedthe meanof 10 successiveheight measurements at their most recent visit. An averageof 5 f 4 (SD) measurements (range, 1-19) wasusedto calculate adult height. The SD for these multiple measurementswithin each individual averaged1.2cm. Due to the retrospective nature of this study, data on body proportions (upper to lower segmentratio) at attainment of final height were not available. Reported parental heights were available for 25 of the 41 subjects.These parental heights were usedto calculate a sexadjusted midparental height (target height) for each subject according to the formula: l/z x (father’s height + mother’s height +13 cm) (13). Adult heights were evaluated in 50 normal volunteers, who were hospitalized on the samewards asthe patients with IHH, to serve as a control for the validity of the retrospective height
HEIGHT
437
analysis. These control adult heights were all basedupon a singleheight measurement. Statistical analysis The adult height of the IHH patients wascomparedwith the height of the 50 normal volunteers by the two-tailed Student’s t test, with the target height for eachsubject (from midparental height adjusted for male sex) by the two-tailed paired t test, and with the averageheight of American malesat age18, from the 1979standardsof the National Center for Health Statistics (14), by z test (15). The dependenceof final adult height in the IHH patients on ageat which treatment to induce puberty was begun was analyzed by linear regressionanalysis.All data are the mean + SD unlessotherwise stated.
Results The mean adult height in the 41 patients with IHH was 179.7 + 7.4 cm (range, 158.9-192.8 cm), which significantly exceeded the height at age 18 of American males (176.8 f 6.7 cm, P < 0.006) and of the 50 NIH normal male volunteers (176.8 f 6.3 cm, P < 0.05). The final height of the 28 patients with IHH and anosmia (179.6 f 7.9 cm) did not differ significantly from the final height of the 13 patients with IHH and normal olfaction (179.9 * 6.6 cm). Final adult height in the IHH patients correlated significantly with the age at initiation of treatment, and hence with the duration of pubertal delay (r = 0.32, P = 0.04) (Fig. 1).
0-i l NWMI I
1
10
I
Olfaction
1
20 30 Age at Treatment (years)
0
40
FIG. 1. The effect of age at initiation of treatment to induce puberty on final adult height in 41 men with isolated hypogonadotropic hypogonadism. The open symbols indicate patients with anosmia (Kallmann’s syndrome); the closed symbols indicate patients with normal olfaction. The solid line is the linear regression line by least-squares analysis of the combined data from both patient groups. The inner dashed lines denote the 95% confidence interval for the regression line; the outer dashed lines denote the 95% confidence interval for a single observation.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 01 July 2014. at 02:36 For personal use only. No other uses without permission. . All rights reserved.
URIARTE
438
Most of the enhanced height of the IHH patients was attributable to those patients in whom puberty had been delayed to age 18 or greater. The mean adult height of this group of patients was 182.4 + 5.8 cm, or 5.6 cm greater than the mean height of the 50 NIH adult male volunteers or of normal American men (P < 0.001). The mean adult height of patients whose treatment began between ages 10 and 17 yr was 175.0 f 7.8 cm, which did not differ significantly from that of normal men, and which was significantly less than the height of patients whose treatment began after age 18 (P < 0.005). Midparental height (by patient report) was available in 14 of the 26 patients whose treatment began after age 18. The adult height of these patients significantly exceeded their target height based upon reported midparental height (183.1 + 6.5 us. 178.4 + 4.07 cm, P < 0.01). The 25 available target heights of the IHH patients, based upon the reported midparental height adjusted for male sex, averaged 178.3 f 4.7 cm, which did not differ significantly from the mean height at age 18 of normal American men (176.8 f 6.7 cm). The mean age of the IHH patients at which they were first observed to have achieved one or more of the final height criteria was 27.4 f 4.8 yr (range, 19.2-41.7 yr). As expected, the mean age at which patients were observed to have attained final height was significantly greater in patients who began treatment after age 18 than in patients who began treatment between 10 and 17 yr of age (28.7 + 4.2 us. 25.1 + 5.0 yr, P < 0.025). The mean age at the time of height measurement in the 50 normal male volunteers was 21.5 + 2.3 yr (range, 19-29 yr). Because this was a retrospective study, we were unable to reconstruct full growth curves for most of the patients. However, three patients for whom these data were available are shown in Fig. 2. Two of these three patients continued to grow well beyond age 18. Discussion The adult height of NIH patients with IHH was significantly greater than the height of normal American men at age 18 or of adult male normal volunteers studied at NIH. The adult height of patients with IHH also correlated significantly with the age at which treatment to induce puberty was begun. The adult height of patients treated after age 18 exceeded the height of normal American men by 5.6 cm and of their own target height, based upon reported midparental height, by 4.7 cm. The available target heights for the patients with IHH did not differ significantly from the height of normal American men, which indicated that the patients with IHH did not come from unusually tall families. All of these data are consistent with the hypothesis that delay of puberty in these patients enhanced ultimate height relative to what
ET AL.
JCE & M -1992 Vol74.No2
180
-160 E s 2 .$140
100
c h
012
14
16
18
20
22
24
26
28
30I
321
Age (years) FIG. 2. Growth pattern of three patients with isolated hypogonadotropic hypogonadism for whom serial height data before epiphyseal fusion were available. The numbers at the right-hand margin indicate the normal American height percentiles from the 1979 survey of the National Center for Health Statistics (14). The normal percentiles from age 18-34 were extrapolated from the values at age 18. The arrows indicate the age at which treatment to induce puberty was begun.
would have occurred if the onset of puberty had been normal. The mechanism of this increased height is presumed to be delayed epiphyseal fusion and a prolonged growth period (16). Since the greatest heights were achieved in those patients in whom therapy was delayed unusually long, one would not predict such substantial increments in the more typical patient with IHH who presents to medical attention earlier and is treated sooner. Due to the retrospective nature of this study, we were unable to obtain sufficient sequential height measurements to characterize fully the growth curves of patients with IHH. However, in certain patients for whom these data were available, we found that growth continued well after 18 years of age. This prolonged period of growth may provide the explanation for the increased final stature. Additionally, it emphasizes the need in studies of growth in IHH to observe patients well into adulthood and to establish rigorous criteria for achievement of final height. To our knowledge only one previous study has shown significantly increased adult height in patients with hypogonadotropic hypogonadism. Raboch and Reisenauer (17) showed an adult height of 176.3 cm in 28 Czechoslovakian men with hypogonadotropic eunuchoidism compared to an adult height of 172.3 cm in 500 normospermit, apparently healthy men who were evaluated
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 01 July 2014. at 02:36 For personal use only. No other uses without permission. . All rights reserved.
PUBERTAL
DELAY
because of an infertile marriage. The relative increase in final height of these hypogonadotropic patients is similar to that observed in our patients. However, this study did not provide the precise diagnosis of the patients (there was no mention of anosmia or of hormonal tests of pituitary function) and did not evaluate final height in relation to established national norms, parental height, or age at onset of treatment. Several other studies have found normal growth in patients with isolated hypogonadotropic hypogonadism (M-20). Copeland et al. (18) described 4 boys, who were still growing at the time of the report and whose final heights were therefore not available. Kaushanski and Laron (19) observed normal prepubertal growth in 19 boys with isolated hypogonadotropic hypogonadism. Their patients, however, were also still growing at the time they were reported, and thus no conclusion about final adult height could be reached. Van Dop et al. (20) studied 20 boys with isolated hypogonadotropic hypogonadism who were treated with testosterone between ages 14 and 19. Those who had achieved final height appeared to be near the 50th percentile for normal men. These data are not inconsistent with those of the current study since our patients who were treated between the ages of 10 and 17 also had an adult height that did not differ from that of normal men. Thus, the observation of increased adult height in the NIH patients with IHH, in contrast to the normal height of the IHH patients in these previous studies, appears to be explained by the greater average delay in initiating treatment, by the longer period of follow-up (which ensured that adult height had been achieved), and by the larger number of subjects (which increased the statistical power of the study). Our findings are consistent with those of Hibi et al. (21), who examined the final height in patients who received GH therapy for idiopathic GH deficiency. They compared those patients who underwent a normal spontaneous puberty with those who also had gonadotropin deficiency and underwent a delayed, therapeutically induced puberty. The boys with combined deficiencies were treated at a mean age of 19.5 f 2.1 yr and were significantly taller (by 11.9 cm) than those who underwent spontaneous puberty earlier. Similarly, the girls with combined deficiency, who were treated at a mean age of 18.6 f 1.8 yr, were significantly taller (by 9.6 cm) than the girls who underwent an earlier, spontaneous puberty. The current study compared the final adult height of IHH patients to the height of normal 1%yr old American men. The use of 18-yr olds for this comparison was based upon the unavailability of normal height standards for older American men. However, this comparison group results in a slight overestimate of the height increase of IHH patients, since normal men grow an average of 0.8
AND ADULT
HEIGHT
439
cm after age 18 (22). When this is taken into account, the height increase of IHH patients treated after age 18 remains highly significant (height increase relative to normal men, 4.8 cm, P < 0.0001, assuming that the SD for final height in normal men is no greater than the SD at age 18). Since Kallman’s syndrome is a genetic disorder which can have manifestations other than pubertal delay, it could conceivably increase height by a mechanism other than pubertal delay. If this were the case, one would predict that the mean final height of patients with anosmia would be greater than those without anosmia. In fact, the two groups did not differ significantly in final height. Furthermore, as shown in Fig. 1, the relationship between age of treatment and final height appeared similar in the two groups. Taken together, these observations suggest that Kallmann’s syndrome did not affect final height by a mechanism other than pubertal delay. The puberty in our IHH patients was induced pharmacologically by a variety of regimens, often before referral to NIH. These regimens were intended to promote virilization rather than to maximize height and often achieved adult levels of plasma testosterone much more rapidly than during normal puberty. Such regimens have been shown by others to promote more rapid attainment of peak height velocity, and a smaller total pubertal height gain, than occur in subjects who undergo spontaneous puberty (23, 24). Thus, the gain in height attributable to delay of puberty may have been counteracted in part by failure to optimize height gain during puberty. Taken together, our data suggest that 6 or more yr of pubertal delay in boys with isolated hypogonadotropic hypogonadism is associated with an adult height that is approximately 5 cm greater than that of normal American men. A similar delay in puberty among patients with idiopathic GH deficiency and gonadotropin deficiency is associated with an even greater height increase (-10 cm) compared to children with isolated GH deficiency (21). Additionally, the preliminary analysis of 4 yr of LHRH agonist-induced pubertal delay in children with idiopathic short stature indicates a lo-cm increase in predicted adult height compared to the randomized, placebo control group (25). None of these data answers the question, however, of whether a pharmacological delay of puberty in children with short stature will result in increased final adult height. The answer to this question must await the completion of the ongoing clinical trials. Acknowledgment We thank Dr. Ora H. Pescovitz for helpful discussions.
References 1. Tanner JM, Whitehouse RH, Marubini E, Resele LF. The adolescent growth spurt of boys and girls of the Harpenden Growth
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 01 July 2014. at 02:36 For personal use only. No other uses without permission. . All rights reserved.
440
URIARTE
Study. Ann Hum Biol. 1976;3:109-26. 2. Frisch RE, Revelle R. The height and weight of girls and boys at the time of initiation of the adolescent growth spurt in height and weight and the relationship to menarche. Hum Biol. 1971;43:14059. 3. Urban MD, Lee PA, Migeon CJ. Adult height and fertility in men with congenital virilizing adrenal hyperplasia. N Engl J Med. 1978;299:1392-6. 4. Lee PA. Medroxyprogesterone therapy for sexual precocity in girls. Am J Dis Child. 1981;135:443-5. 5. Oerter KE, Manasco P, Barnes KM, Jones J, Hill S, Cutler Jr GB. Adult height in precocious puberty after long-term treatment with deslorelin. J Clin Endocrinol Metab, in press. 6. Boepple PA, Mansfield MJ, Crawford JD, Crigler Jr JF, Blizzard RM, Crowley Jr WF. Final heights in girls with central precocious puberty (CPP) following GnRH agonist (GnRHa)-induced pituitary-gonadal suppression. Pediatr Res. 1991;29374A. 7. Paul DL, Conte FA, Grumbach MM, Kaplan SL. Longterm effects of GnRH agonists on heights and gonadotropin-gonadal function in children with true precocious puberty. Pediatr Res. 1991;2983A. 8. Kallmann FJ, Schoenfeld WA, Barrera SE. The genetic aspects of primary eunuchoidism. Am J Ment Defic. 1944;48:203-36. 9. Lieblich JM, Rogol AD, White BJ, Rosen SW. Syndrome of anosmia with hypogonadotropic hypogonadism (Kallmann syndrome). Clinical and laboratory studies in 23 cases. Am J Med. 1982;73:50619. 10. Zachmann M, Prader A, Kind HP, Hafliger H, Budliger H. Testicular volume during adolescence. Cross-sectional and longitudinal studies. Helv Paediatr Acta. 1974;29:61-72. 11. Marshall WA, Tanner JM. Variations in the pattern of pubertal changes in boys. Arch Dis Child. 1970;45:13-23. 12. Greulich WW. Pvle SI. Radiomanhic atlas of skeletal develonment of the hand and wrist, 2nd ed. Stanford: Stanford University Press; 1959. 13. Tanner JM, Goldstein H, Whitehouse RH. Standards for children’s height at ages 2-9 years allowing for height of parents. Arch Dis ,
”
”
.
ET AL.
JCE&M.199!2 Voll4*No2
Child. 1970;45:755-62. 14. Hamill PVV, Drizd TA, Johnson CL, Reed RB, Roche AF, Moore WM. Physical growth: National Center for Health Statistics nercentiles. Am J Clin Nutr. 1979;32:607-29. 15. Snedecor GW. Cochran WG. Statistical methods. , 6th ed. ~~ Ames: ~~~--Iowa State University Press; 1967:56. 16. Roche AF, Davila GH. Late adolescent growth in stature. Pediatrics. 1972;50:874-80. 17. Raboch J, Reisenauer R. Analysis of body height in 829 patients with different forms of testicular pathology. Andrologia. 1976;8:265-8. 18. Copeland KC, Paunier L, Sizonenko PC. The secretion of adrenal androgens and growth patterns of patients with hypogonadotropic hypogonadism and idiopathic delayed puberty. J Pediatr. 1977;91:985-90. 19. Kaushanski A, Laron 2. Growth pattern of boys with isolated gonadotropin deficiency. Isr J Med Sci. 1979;15:518-21. 20. Van Dop C, Burstein S, Conte FA, Grumbach MM. Isolated gonadotropin deficiency in boys: clinical characteristics and growth. J Pediatr. 1987;111:684-92. 21. Hibi I, Tanaka T, Tanae A, et al. The influence of gonadal function and the effect of gonadal suppression treatment on final height in growth hormone (GH)-treated GH-deficient children. J Clin Endocrinol Metab. 1989;69:221-6. 22. Roche AF. Adult stature prediction: a critical review. Acta Med Auxol. 1984;16:5-28. 23. Bourguignon JP, Vandeweghe M, Vanderschueren-Lodeweyckx M, et al. Pubertal growth and final height in hypopituitary boys: a minor role of bone age at onset of puberty. J Clin Endocrinol Metab. 1986;63:376-82. 24. Bourguignon JP. Linear growth as a function of age at onset of puberty and sex steroid dosage: therapeutic implications. Endocr Rev. 1988,9:467-88. 25. Municchi G, Rose SR, Barnes KM, Cassorla F, Cutler Jr GB. A double-blind, placebo-controlled trial of LHRH agonist (LHRH,) treatment in patients with short stature and normal onset of puberty: preliminary results. Pediatr Res. 1991;29:82A.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 01 July 2014. at 02:36 For personal use only. No other uses without permission. . All rights reserved.
Vol. 74, No. 2 Printed in U.S.A.
The Effect of Pubertal Delay with Isolated Hypogonadotropic
on Adult Height Hypogonadism
M. MERCEDES URIARTE, JEFFREY BARON,* HERNAN KEVIN M. BARNES, D. LYNN LORIAUX,* AND GORDON Developmental Endocrinology Branch, National Institute of Child Health National Institutes of Health, Bethesda, Maryland 20892
ABSTRACT. To determine whether delay of puberty alters adult height, we retrospectively evaluated the adult height of 41 patients who met rigorous criteria for the diagnosis of isolated hypogonadotropic hypogonadism. The adult height of these patients was compared with the mean height of normal American men at age 18 in the 1979 National Center for Health Statistics survey and with the mean adult height of 50 male normal volunteers who had been studied on the same wards as the patients with hypogonadotropic hypogonadism. The mean adult height of the men with isolated hypogonadotropic hypogonadism was 179.7 cm, which exceeded the height of the 50 control subjects and the normal American men (both 176.8 cm) by 2.9 cm (P < 0.05 and P < 0.006, respectively). The mean age at treatment to induce puberty was 20.0 yr, which corresponded to a mean delay in the onset of puberty of more than 8 yr relative to normal males. The final height of the men
B. GARCIA, B. CUTLER, and Human
in Men
JR.*
Development,
with hypogonadotropic hypogonadism correlated significantly with the duration of pubertal delay (r = 0.32, P = 0.04). Most of the enhanced mean adult height of the patients with isolated hypogonadotropic hypogonadism was attributable to those patients in whom treatment to induce puberty had been delayed to age 18 or greater. The mean adult height of these patients was 182.4 cm, or 5.6 cm greater than the mean height of the 50 controls or of normal American men (P C 0.001). The mean adult height of patients whose treatment began between 10 and 17 yr of age was 175.0 cm, which did not differ significantly from that of normal men. We conclude that prolonged delay of puberty (6 or more yr) in men with isolated hypogonadotropic hypogonadism is associated with a modest increase (-5 cm) of adult height. (J Clin Endocrinol Metab 74: 436-440,1992)
T
ism (IHH) often experience a delay of puberty as a result of delay in diagnosis and treatment. These patients thus provide a natural experiment on the effect of pubertal delay on adult height. We therefore studied the final height of 41 men who met rigorous criteria for the diagnosis of IHH and who had been evaluated at the NIH between 1959 and 1987.
HE pubertal growth spurt causes an approximate doubling of the prepubertal growth rate and contributes 1518% to adult height (1,2). However, puberty also sets in motion the events that lead to epiphyseal fusion and the cessation of growth. Disorders that advance the onset of puberty, such as congenital adrenal hyperplasia and LHRH-dependent precocious puberty, cause early epiphyseal fusion and impairment of adult height (3,4). Conversely, treatment of LHRH-dependent precocious puberty with a longacting LHRH agonist results in a significant improvement in the adult height when compared to pretreatment predicted height (5-7). These observations raise the possibility that delay of normal puberty might also increase final height. If this hypothesis were correct, then treatment with long-acting LHRH agonists might increase the duration of growth and the final height of children with short stature. Patients with isolated hypogonadotropic hypogonad-
Subjects
and Methods
Subjects We retrospectively analyzed the charts of 41 NIH patients who were evaluated between 1959 and 1987, who had a diagnosis of IHH, and who had achieved adult height at the time of the study. Twenty-eight of the 41 patients had hypogonadotropic hypogonadism and anosmia (Kallmann’s syndrome) (8,9). The remaining 13 patients were considered to have isolated hypogonadotropic hypogonadism based upon prolonged delay of puberty, low or normal basal plasma gonadotropin levels, normal thyroid function, normal adrenal function (in the 12 patients in whom this was tested), normal plasma PRL level, and
Received June 27,1989 Address requests for reprints to: Dr. Gordon B. Cutler, Jr., Building 10, Room 10-N-262, National Institutes of Health, Bethesda, Maryland 20892. * Commissioned officers in the United States Public Health Service.
normal GH response to stimulation
tests (~7 rg/L in the 5
patients in whom this was tested). Thirty-eight of the 41 patients had delayed puberty (failure to enter puberty by age 14). Three of the patients, who had 436
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 01 July 2014. at 02:36 For personal use only. No other uses without permission. . All rights reserved.
PUBERTAL
DELAY
AND ADULT
anosmia and were presumed to have Kallmann’s syndrome because of a positive family history (2 patients) or cryptorchidism (1 patient), received treatment to induce puberty before age 14. Testicular volume in the patients with IHH averaged2.1 + 1.5 ml (range,0.2-5.5 ml) at the time of initial evaluation (10). Five of the patients had unilateral cryptorchidism. Pretreatment testosteronelevels, which were available in 36 of the 41 patients, averaged0.9 f 0.7 nmol/L (range, 0.3-3.2 nmol/L; normal adult range, I-35 limol/L). Pretreatment FSH and LH levels, which were available in 37 patients, averaged2.7 + 1.4 IU/L (range, 0.7-7.0 IU/L) and 3.7 + 2.4 IU/L (range, 0.2-11.0 IU/L), respectively (normal adult ranges:5-25 IU/L for FSH, and 6-26 IU/L for LH). The patients with IHH had a meanageat treatment of 20.0 f 6.2 yr (range, lo-42 yr), and, thus, had a mean delay of puberty of 8.4 yr when comparedto the normal meanpubertal age of onset in boys of 11.6 yr (11). The treatments to induce puberty compriseda variety of regimensof testosterone,hCG, or hCG plus human menopausalgonadotropin. Treatment was often started before referral to NIH. Since most of our patients had considerabledelay in diagnosisand treatment, the goal of therapy was usually to promote virilization or fertility rather than to maximize adult height. Thus, the regimensemployed tended to promote more rapid virilization than would occur during spontaneouspuberty. The control men were 50 normal volunteers [age, 21.5 f 2.3 (SD) yr; range, 19-291who were studied at the NIH between 1978and 1983on the sameinpatient wardsasthe IHH patients. Height measurement A single height measurementby stadiometer was made at each NIH visit for 35 of the 41 subjects. The remaining 6 patients received 10 successiveheight measurementsat their most recent visit. Patients were consideredto have achievedadult height when one or more of the following criteria were met: 1) bone age L 18 yr (12); 2) no increasein height over a period of at least 6 months; 3) 5 yr or more of treatment with androgensor hCG; or 4) chronological age~40 yr. If more than one height measurement was available after achievement of adult height, the average of all values was used except for the six subjects in whom we usedthe meanof 10 successiveheight measurements at their most recent visit. An averageof 5 f 4 (SD) measurements (range, 1-19) wasusedto calculate adult height. The SD for these multiple measurementswithin each individual averaged1.2cm. Due to the retrospective nature of this study, data on body proportions (upper to lower segmentratio) at attainment of final height were not available. Reported parental heights were available for 25 of the 41 subjects.These parental heights were usedto calculate a sexadjusted midparental height (target height) for each subject according to the formula: l/z x (father’s height + mother’s height +13 cm) (13). Adult heights were evaluated in 50 normal volunteers, who were hospitalized on the samewards asthe patients with IHH, to serve as a control for the validity of the retrospective height
HEIGHT
437
analysis. These control adult heights were all basedupon a singleheight measurement. Statistical analysis The adult height of the IHH patients wascomparedwith the height of the 50 normal volunteers by the two-tailed Student’s t test, with the target height for eachsubject (from midparental height adjusted for male sex) by the two-tailed paired t test, and with the averageheight of American malesat age18, from the 1979standardsof the National Center for Health Statistics (14), by z test (15). The dependenceof final adult height in the IHH patients on ageat which treatment to induce puberty was begun was analyzed by linear regressionanalysis.All data are the mean + SD unlessotherwise stated.
Results The mean adult height in the 41 patients with IHH was 179.7 + 7.4 cm (range, 158.9-192.8 cm), which significantly exceeded the height at age 18 of American males (176.8 f 6.7 cm, P < 0.006) and of the 50 NIH normal male volunteers (176.8 f 6.3 cm, P < 0.05). The final height of the 28 patients with IHH and anosmia (179.6 f 7.9 cm) did not differ significantly from the final height of the 13 patients with IHH and normal olfaction (179.9 * 6.6 cm). Final adult height in the IHH patients correlated significantly with the age at initiation of treatment, and hence with the duration of pubertal delay (r = 0.32, P = 0.04) (Fig. 1).
0-i l NWMI I
1
10
I
Olfaction
1
20 30 Age at Treatment (years)
0
40
FIG. 1. The effect of age at initiation of treatment to induce puberty on final adult height in 41 men with isolated hypogonadotropic hypogonadism. The open symbols indicate patients with anosmia (Kallmann’s syndrome); the closed symbols indicate patients with normal olfaction. The solid line is the linear regression line by least-squares analysis of the combined data from both patient groups. The inner dashed lines denote the 95% confidence interval for the regression line; the outer dashed lines denote the 95% confidence interval for a single observation.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 01 July 2014. at 02:36 For personal use only. No other uses without permission. . All rights reserved.
URIARTE
438
Most of the enhanced height of the IHH patients was attributable to those patients in whom puberty had been delayed to age 18 or greater. The mean adult height of this group of patients was 182.4 + 5.8 cm, or 5.6 cm greater than the mean height of the 50 NIH adult male volunteers or of normal American men (P < 0.001). The mean adult height of patients whose treatment began between ages 10 and 17 yr was 175.0 f 7.8 cm, which did not differ significantly from that of normal men, and which was significantly less than the height of patients whose treatment began after age 18 (P < 0.005). Midparental height (by patient report) was available in 14 of the 26 patients whose treatment began after age 18. The adult height of these patients significantly exceeded their target height based upon reported midparental height (183.1 + 6.5 us. 178.4 + 4.07 cm, P < 0.01). The 25 available target heights of the IHH patients, based upon the reported midparental height adjusted for male sex, averaged 178.3 f 4.7 cm, which did not differ significantly from the mean height at age 18 of normal American men (176.8 f 6.7 cm). The mean age of the IHH patients at which they were first observed to have achieved one or more of the final height criteria was 27.4 f 4.8 yr (range, 19.2-41.7 yr). As expected, the mean age at which patients were observed to have attained final height was significantly greater in patients who began treatment after age 18 than in patients who began treatment between 10 and 17 yr of age (28.7 + 4.2 us. 25.1 + 5.0 yr, P < 0.025). The mean age at the time of height measurement in the 50 normal male volunteers was 21.5 + 2.3 yr (range, 19-29 yr). Because this was a retrospective study, we were unable to reconstruct full growth curves for most of the patients. However, three patients for whom these data were available are shown in Fig. 2. Two of these three patients continued to grow well beyond age 18. Discussion The adult height of NIH patients with IHH was significantly greater than the height of normal American men at age 18 or of adult male normal volunteers studied at NIH. The adult height of patients with IHH also correlated significantly with the age at which treatment to induce puberty was begun. The adult height of patients treated after age 18 exceeded the height of normal American men by 5.6 cm and of their own target height, based upon reported midparental height, by 4.7 cm. The available target heights for the patients with IHH did not differ significantly from the height of normal American men, which indicated that the patients with IHH did not come from unusually tall families. All of these data are consistent with the hypothesis that delay of puberty in these patients enhanced ultimate height relative to what
ET AL.
JCE & M -1992 Vol74.No2
180
-160 E s 2 .$140
100
c h
012
14
16
18
20
22
24
26
28
30I
321
Age (years) FIG. 2. Growth pattern of three patients with isolated hypogonadotropic hypogonadism for whom serial height data before epiphyseal fusion were available. The numbers at the right-hand margin indicate the normal American height percentiles from the 1979 survey of the National Center for Health Statistics (14). The normal percentiles from age 18-34 were extrapolated from the values at age 18. The arrows indicate the age at which treatment to induce puberty was begun.
would have occurred if the onset of puberty had been normal. The mechanism of this increased height is presumed to be delayed epiphyseal fusion and a prolonged growth period (16). Since the greatest heights were achieved in those patients in whom therapy was delayed unusually long, one would not predict such substantial increments in the more typical patient with IHH who presents to medical attention earlier and is treated sooner. Due to the retrospective nature of this study, we were unable to obtain sufficient sequential height measurements to characterize fully the growth curves of patients with IHH. However, in certain patients for whom these data were available, we found that growth continued well after 18 years of age. This prolonged period of growth may provide the explanation for the increased final stature. Additionally, it emphasizes the need in studies of growth in IHH to observe patients well into adulthood and to establish rigorous criteria for achievement of final height. To our knowledge only one previous study has shown significantly increased adult height in patients with hypogonadotropic hypogonadism. Raboch and Reisenauer (17) showed an adult height of 176.3 cm in 28 Czechoslovakian men with hypogonadotropic eunuchoidism compared to an adult height of 172.3 cm in 500 normospermit, apparently healthy men who were evaluated
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 01 July 2014. at 02:36 For personal use only. No other uses without permission. . All rights reserved.
PUBERTAL
DELAY
because of an infertile marriage. The relative increase in final height of these hypogonadotropic patients is similar to that observed in our patients. However, this study did not provide the precise diagnosis of the patients (there was no mention of anosmia or of hormonal tests of pituitary function) and did not evaluate final height in relation to established national norms, parental height, or age at onset of treatment. Several other studies have found normal growth in patients with isolated hypogonadotropic hypogonadism (M-20). Copeland et al. (18) described 4 boys, who were still growing at the time of the report and whose final heights were therefore not available. Kaushanski and Laron (19) observed normal prepubertal growth in 19 boys with isolated hypogonadotropic hypogonadism. Their patients, however, were also still growing at the time they were reported, and thus no conclusion about final adult height could be reached. Van Dop et al. (20) studied 20 boys with isolated hypogonadotropic hypogonadism who were treated with testosterone between ages 14 and 19. Those who had achieved final height appeared to be near the 50th percentile for normal men. These data are not inconsistent with those of the current study since our patients who were treated between the ages of 10 and 17 also had an adult height that did not differ from that of normal men. Thus, the observation of increased adult height in the NIH patients with IHH, in contrast to the normal height of the IHH patients in these previous studies, appears to be explained by the greater average delay in initiating treatment, by the longer period of follow-up (which ensured that adult height had been achieved), and by the larger number of subjects (which increased the statistical power of the study). Our findings are consistent with those of Hibi et al. (21), who examined the final height in patients who received GH therapy for idiopathic GH deficiency. They compared those patients who underwent a normal spontaneous puberty with those who also had gonadotropin deficiency and underwent a delayed, therapeutically induced puberty. The boys with combined deficiencies were treated at a mean age of 19.5 f 2.1 yr and were significantly taller (by 11.9 cm) than those who underwent spontaneous puberty earlier. Similarly, the girls with combined deficiency, who were treated at a mean age of 18.6 f 1.8 yr, were significantly taller (by 9.6 cm) than the girls who underwent an earlier, spontaneous puberty. The current study compared the final adult height of IHH patients to the height of normal 1%yr old American men. The use of 18-yr olds for this comparison was based upon the unavailability of normal height standards for older American men. However, this comparison group results in a slight overestimate of the height increase of IHH patients, since normal men grow an average of 0.8
AND ADULT
HEIGHT
439
cm after age 18 (22). When this is taken into account, the height increase of IHH patients treated after age 18 remains highly significant (height increase relative to normal men, 4.8 cm, P < 0.0001, assuming that the SD for final height in normal men is no greater than the SD at age 18). Since Kallman’s syndrome is a genetic disorder which can have manifestations other than pubertal delay, it could conceivably increase height by a mechanism other than pubertal delay. If this were the case, one would predict that the mean final height of patients with anosmia would be greater than those without anosmia. In fact, the two groups did not differ significantly in final height. Furthermore, as shown in Fig. 1, the relationship between age of treatment and final height appeared similar in the two groups. Taken together, these observations suggest that Kallmann’s syndrome did not affect final height by a mechanism other than pubertal delay. The puberty in our IHH patients was induced pharmacologically by a variety of regimens, often before referral to NIH. These regimens were intended to promote virilization rather than to maximize height and often achieved adult levels of plasma testosterone much more rapidly than during normal puberty. Such regimens have been shown by others to promote more rapid attainment of peak height velocity, and a smaller total pubertal height gain, than occur in subjects who undergo spontaneous puberty (23, 24). Thus, the gain in height attributable to delay of puberty may have been counteracted in part by failure to optimize height gain during puberty. Taken together, our data suggest that 6 or more yr of pubertal delay in boys with isolated hypogonadotropic hypogonadism is associated with an adult height that is approximately 5 cm greater than that of normal American men. A similar delay in puberty among patients with idiopathic GH deficiency and gonadotropin deficiency is associated with an even greater height increase (-10 cm) compared to children with isolated GH deficiency (21). Additionally, the preliminary analysis of 4 yr of LHRH agonist-induced pubertal delay in children with idiopathic short stature indicates a lo-cm increase in predicted adult height compared to the randomized, placebo control group (25). None of these data answers the question, however, of whether a pharmacological delay of puberty in children with short stature will result in increased final adult height. The answer to this question must await the completion of the ongoing clinical trials. Acknowledgment We thank Dr. Ora H. Pescovitz for helpful discussions.
References 1. Tanner JM, Whitehouse RH, Marubini E, Resele LF. The adolescent growth spurt of boys and girls of the Harpenden Growth
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 01 July 2014. at 02:36 For personal use only. No other uses without permission. . All rights reserved.
440
URIARTE
Study. Ann Hum Biol. 1976;3:109-26. 2. Frisch RE, Revelle R. The height and weight of girls and boys at the time of initiation of the adolescent growth spurt in height and weight and the relationship to menarche. Hum Biol. 1971;43:14059. 3. Urban MD, Lee PA, Migeon CJ. Adult height and fertility in men with congenital virilizing adrenal hyperplasia. N Engl J Med. 1978;299:1392-6. 4. Lee PA. Medroxyprogesterone therapy for sexual precocity in girls. Am J Dis Child. 1981;135:443-5. 5. Oerter KE, Manasco P, Barnes KM, Jones J, Hill S, Cutler Jr GB. Adult height in precocious puberty after long-term treatment with deslorelin. J Clin Endocrinol Metab, in press. 6. Boepple PA, Mansfield MJ, Crawford JD, Crigler Jr JF, Blizzard RM, Crowley Jr WF. Final heights in girls with central precocious puberty (CPP) following GnRH agonist (GnRHa)-induced pituitary-gonadal suppression. Pediatr Res. 1991;29374A. 7. Paul DL, Conte FA, Grumbach MM, Kaplan SL. Longterm effects of GnRH agonists on heights and gonadotropin-gonadal function in children with true precocious puberty. Pediatr Res. 1991;2983A. 8. Kallmann FJ, Schoenfeld WA, Barrera SE. The genetic aspects of primary eunuchoidism. Am J Ment Defic. 1944;48:203-36. 9. Lieblich JM, Rogol AD, White BJ, Rosen SW. Syndrome of anosmia with hypogonadotropic hypogonadism (Kallmann syndrome). Clinical and laboratory studies in 23 cases. Am J Med. 1982;73:50619. 10. Zachmann M, Prader A, Kind HP, Hafliger H, Budliger H. Testicular volume during adolescence. Cross-sectional and longitudinal studies. Helv Paediatr Acta. 1974;29:61-72. 11. Marshall WA, Tanner JM. Variations in the pattern of pubertal changes in boys. Arch Dis Child. 1970;45:13-23. 12. Greulich WW. Pvle SI. Radiomanhic atlas of skeletal develonment of the hand and wrist, 2nd ed. Stanford: Stanford University Press; 1959. 13. Tanner JM, Goldstein H, Whitehouse RH. Standards for children’s height at ages 2-9 years allowing for height of parents. Arch Dis ,
”
”
.
ET AL.
JCE&M.199!2 Voll4*No2
Child. 1970;45:755-62. 14. Hamill PVV, Drizd TA, Johnson CL, Reed RB, Roche AF, Moore WM. Physical growth: National Center for Health Statistics nercentiles. Am J Clin Nutr. 1979;32:607-29. 15. Snedecor GW. Cochran WG. Statistical methods. , 6th ed. ~~ Ames: ~~~--Iowa State University Press; 1967:56. 16. Roche AF, Davila GH. Late adolescent growth in stature. Pediatrics. 1972;50:874-80. 17. Raboch J, Reisenauer R. Analysis of body height in 829 patients with different forms of testicular pathology. Andrologia. 1976;8:265-8. 18. Copeland KC, Paunier L, Sizonenko PC. The secretion of adrenal androgens and growth patterns of patients with hypogonadotropic hypogonadism and idiopathic delayed puberty. J Pediatr. 1977;91:985-90. 19. Kaushanski A, Laron 2. Growth pattern of boys with isolated gonadotropin deficiency. Isr J Med Sci. 1979;15:518-21. 20. Van Dop C, Burstein S, Conte FA, Grumbach MM. Isolated gonadotropin deficiency in boys: clinical characteristics and growth. J Pediatr. 1987;111:684-92. 21. Hibi I, Tanaka T, Tanae A, et al. The influence of gonadal function and the effect of gonadal suppression treatment on final height in growth hormone (GH)-treated GH-deficient children. J Clin Endocrinol Metab. 1989;69:221-6. 22. Roche AF. Adult stature prediction: a critical review. Acta Med Auxol. 1984;16:5-28. 23. Bourguignon JP, Vandeweghe M, Vanderschueren-Lodeweyckx M, et al. Pubertal growth and final height in hypopituitary boys: a minor role of bone age at onset of puberty. J Clin Endocrinol Metab. 1986;63:376-82. 24. Bourguignon JP. Linear growth as a function of age at onset of puberty and sex steroid dosage: therapeutic implications. Endocr Rev. 1988,9:467-88. 25. Municchi G, Rose SR, Barnes KM, Cassorla F, Cutler Jr GB. A double-blind, placebo-controlled trial of LHRH agonist (LHRH,) treatment in patients with short stature and normal onset of puberty: preliminary results. Pediatr Res. 1991;29:82A.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 01 July 2014. at 02:36 For personal use only. No other uses without permission. . All rights reserved.
