0021-972X/90/7105-1251$02.00/0 Journal of Clinical Endocrinology and Metabolism Copyright © 1990 by The Endocrine Society

Vol. 71, No. 5 Printed in U.S.A.

Gonadotropin Secretory Dynamics During Puberty in Normal Girls and Boys KAREN E. OERTER, M. MERCEDES URIARTE, SUSAN R. ROSE, KEVIN M. BARNES, AND GORDON B. CUTLER, JR. Developmental Endocrinology Branch, National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, Maryland 20892

ABSTRACT. To determine the most useful index of pubertal gonadotropin secretion we measured spontaneous LH and FSH levels every 20 min for 24 h and the LH and FSH responses to LHRH in a total of 37 girls and 30 boys representing each of the 5 stages of puberty. Mean 24-h LH and FSH levels rose significantly with increasing pubertal stage in both girls and boys. LH peak amplitude increased significantly with increasing pubertal stage for both sexes, whereas FSH peak amplitude did not. LH and FSH peaks were present throughout the 24-h period in all children, but the frequency did not change significantly with increasing pubertal stage. Mean gonadotropin levels, peak amplitudes, and peak frequencies tended to be higher at night from pubertal stages 14 of puberty. There were no significant sex differences in mean

LH, LH peak amplitude, or LH peak frequency. The LHRH-stimulated peak LH to peak FSH ratio was greater in boys than girls during pubertal stages 1-3 and was less useful in distinguishing pubertal from prepubertal boys. For girls, the most accurate index of pubertal gonadotropin secretion was a LHRH-stimulated peak LH to peak FSH ratio greater than 0.66, which detected 96% of the pubertal girls with no false positives. For boys, the most accurate index was a maximum spontaneous nighttime LH level of 12 IU/L or more, which detected 90% of the pubertal boys with no false positives. We conclude that there are important sex differences in the gonadotropin responses to LHRH during puberty, and that criteria for the onset of pubertal gonadotropin secretion should be sex specific. (J Clin Endocrinol Metab 7 1 : 1251-1258, 1990)

"NTORMAL pubertal development in humans involves -L ^1 activation of the hypothalamic-pituitary-gonadal axis, which usually begins between the ages of 9-13 yr in girls and between 10-14 yr in boys. Previous studies have shown that gonadotropin levels have a diurnal rhythm during the prepubertal years, with higher levels at night (1-4). Early puberty is characterized by a large increase in the amplitude and a smaller increase in the frequency of nighttime LH peaks (5-9). Additionally, the gonadotropin response to exogenous LHRH changes from a FSH-predominant response during the prepubertal years to a LH-predominant response during puberty (10). Although the broad outlines of these pubertal changes have been established previously, the detailed pattern of changes at each stage of puberty has not previously been described. Additionally, most previous studies have included insufficient numbers of girls and boys to determine whether there are sex differences in gonadotropin secretion during puberty. This has caused difficulties in the interpretation of hormonal testing in children with precocious puberty, because we lack precise knowledge Received April 3, 1990. Address all correspondence and requests for reprints to: Dr. Karen E. Oerter, Building 10, Room 10N262, National Institutes of Health, Bethesda, Maryland 20892.

of the diagnostic efficiency of spontaneous and stimulated gonadotropin levels as indices of pubertal activation of the LHRH neuron. To address this issue, we have measured spontaneous and LHRH-stimulated gonadotropin secretion in 37 girls and 30 boys representing each of the 5 stages of puberty.

Materials and Methods Subjects The subjects were 37 girls (aged 4-17 yr) and 30 boys (aged 6-18 yr; Table 1). All participants were normal volunteers recruited specifically for this study. All had normal physical examinations and no history of significant past medical illness. The study protocol was approved by the NICHHD Institutional Review Board, and informed consent was given by each child and by one parent. Pubertal staging Pubertal stage in girls was the mean of pubic hair and breast stages according to Tanner (11,12). Pubertal stage in boys was based on the mean of pubic hair stage according to Tanner and a modified genital staging method, based on testicular volume (13), that approximates the photographic stages used by Tanner. Specifically, testicular volumes of 1-4 mL were defined as stage 1, 5-9 mL as stage 2, 10-14 mL as stage 3, 15-19 mL as

1251

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 06 September 2014. at 08:01 For personal use only. No other uses without permission. . All rights reserved.

1252

OERTER

stage 4, and greater than or equal to 20 mL as stage 5. To simplify the presentation of the data, pubertal stages of 1.5 (n = 2), 2.5 (n = 2), 3.5 (n = 5), and 4.5 (n = 2) were rounded up to the next stage. Three of the eight stage 4 girls were postmenarchal, as were all of the stage 5 girls. Study evaluations were performed randomly throughout the cycle, since many of the subjects were not yet cycling regularly. Protocol The children were admitted to the Clinical Center at the NIH. Blood samples were drawn for LH, FSH, and PRL measurements every 20 min for 24 h starting at 0800 h. After the 24-h period of blood sampling was completed, LHRH was given iv in a dose of 100 ng. LH and FSH were measured at -30, - 1 5 , 0, 15, 30, 45, 60, 90, 120, and 180 min relative to administration of LHRH at time zero. Not all subjects completed both the 24-h sampling study and the LHRH test. Among the 37 girls, 7 received only the 24-h sampling study, and 5 received only the LHRH test. Among the 30 boys, 4 received only the 24-h sampling study.

ETAL.

JCE & M • 1990 Vol71»No5

correspond to the period 2000-0800 h. Peak frequencies and amplitudes were analyzed using the computer program Detect (16), with the probability of false positive peaks set at 1%. The effect of time of day on peak frequency was analyzed by x2 test. Comparisons between and among groups were made using analysis of variance. The two-tailed unpaired Student's t test with the Bonferroni adjustment was used for comparisons between two individual groups. To determine which index of gonadotropin secretion best distinguished the pubertal children from those who were prepubertal, we constructed separate ROC (receiver operating characteristic) curves for girls and boys for each of the following measures (17): the ratio of LHRH-stimulated peak LH to peak FSH, the LHRH-stimulated peak LH level, the maximum LH increment above baseline after LHRH stimulation, the maximum spontaneous nighttime LH level, the mean nighttime LH level, and the mean nighttime LH peak amplitude.

Results Spontaneous Gonadotropin Secretion

Hormone assays All measurements were made in duplicate at Hazleton Biotechnologies (Vienna, VA) with modifications of previously described methods (14, 15). The detection limit and the intraand interassay coefficients of variation for the LH, FSH, and PRL assays (at the ED50) were as follows: 0.6 IU/L, 2% and 9% for LH; 0.6 IU/L, 4% and 11% for FSH; and 1.8 Mg/L, 7%, and 12% for PRL. The interassay coefficients of variation at the ED90 and ED15 were 25% and 11.6% for LH, 26% and 10% for FSH, and 12.4% and 13.3% for PRL. The reference standards for the gonadotropin RIAs were pooled postmenopausal sera that were cross-referenced against the Second International Reference Preparation of human menopausal gonadotropin. Statistical analysis All data are shown as the mean ± SD. Daytime values correspond to the period 0800-2000 h, and nighttime values TABLE 1. Clinical features of the subjects Pubertal stage

No.

Age (yr)

Girls

1 2 3 4 5

7 4 9 8 9

8.8 ± 2.4 11.0 ± 0.2 11.8 ±1.0 13.1 ± 1.4 14.9 ± 1.2

-0.1 ± 0.2 ± 1.1 ± 0.0 ± -0.1 ±

0.3 0.8 1.2 0.4 0.3

-0.3 ± 0.6 ± -0.3 ± -0.5 ± 0.3 ±

0.8 2.2 0.2 0.2 1.2

Boys

1 2 3 4 5

9 3 6 6 6

8.8 ± 1.6 12.3 ± 1.0 13.8 ± 1.0 13.9 ± 0.7 15.7 ± 1.7

1.0 ± 0.4 ± 0.6 ± 0.6 ± 0.9 ±

1.3 0.7 1.0 1.2 1.0

0.6 ± -0.6 ± -0.2 ± 0.2 ± 0.6 ±

1.6 0.3 0.7 1.0 1.0

Sex

Ht

(SD)°

BMI (SD)6

Values are the mean ± SD. Height shown as SD for age. 6 Body mass index = weight (kg)/height (m2); shown as SD for age.

0

Mean 24-h LH and FSH levels rose significantly with increasing pubertal stage in both girls and boys [P < 0.01, by analysis of variance (ANOVA); Fig. 1]. Comparisons of mean LH and FSH levels among the individual pubertal stages indicated significantly greater mean LH levels in stage 5 girls and boys compared to stage 1 subjects (P < 0.01), and significantly greater mean FSH levels in stage 4 girls compared to stage 1 girls (P < 0.5) and in stage 5 compared to stage 1 boys (P < 0.01). LH peak amplitude increased significantly with increasing pubertal stage for both sexes (P < 0.01), whereas FSH peak amplitude did not change significantly (by ANOVA, P = 0.20 in the girls, 0.06 in the boys). LH and FSH peaks were present throughout the 24-h period in all children, including those who were prepubertal. The mean 24-h LH peak frequency did not change significantly with pubertal stage (P = 0.69, by ANOVA for both girls and boys). The mean FSH peak frequency in the girls did not change significantly with pubertal stage (P = 0.81), but appeared to peak at midpuberty for the boys (P = 0.05, by ANOVA). Mean gonadotropin levels tended to be higher at night from stages 1-4 of puberty in both sexes (Figs. 2 and 3 and Tables 2 and 3). Within the individual pubertal stages, these differences achieved statistical significance for LH at stages 1 and 4 in both sexes (P < 0.05) and for FSH at stage 1 in girls (P < 0.01) and at stages 1 and 4 in boys (P < 0.01). The stage 5 girls and boys exhibited no significant day-night difference in either LH or FSH levels. Mean gonadotropin peak amplitude was also higher at night from pubertal stages 1-4 in both sexes. Within the individual pubertal stages, these differences achieved

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 06 September 2014. at 08:01 For personal use only. No other uses without permission. . All rights reserved.

1253

GONADOTROPIN SECRETION DURING PUBERTY • 0

girls boy• MEAN 2 4 HOUR LH LEVEL

• girls D boys MEAN 2 4 HOUR FSH LBVBL

S 10

15

LH PEAK FREQUENCY

2 30

1 2

3

4

5

PUBERTAL STAGE FIG. 1. Mean 24-h level, peak amplitude, and peak frequency of LH and FSH during each stage of puberty in girls and boys. Data are presented as the mean ± SD.

statistical significance for LH at stage 1 in the girls (P < 0.05) and at stages 1, 3, and 4 in the boys (P < 0.01). The stage 5 children exhibited no significant day-night difference in peak amplitude for either LH or FSH. Mean gonadotropin peak frequencies during the early to midstages of puberty tended to be slightly higher at night than in the day. Within the individual stages, these differences achieved statistical significance for LH at stage 1 in the girls and stage 3 in the boys (P < 0.05). Prepubertal children averaged 0.39 ± 0.09 LH peaks/ h, which corresponds to an interpeak interval of 2.65 ± 0.59 h, with the highest frequency between 2400-0200 h (0.53 ± 0.12 peaks/h; Fig. 4). Stage 2-3 pubertal children averaged 0.39 ± 0.12 LH peaks/h (interpeak interval, 2.94 ± 1.47 h), with the lowest frequency between 12001400 h (0.28 ± 0.35 peaks/h) and the highest frequency between 2200-0200 h (0.56 ± 0.29 peaks/h). Stage 4-5 pubertal children averaged 0.49 ± 0.16 LH peaks/h (interpeak interval, 2.22 ± 0.62 h), with the highest frequency between 1200-1400 h (0.52 ± 0.30 peaks/h). These differences in peak frequency throughout the day were not statistically significant by x2 analysis. No consistent sex differences were observed for the mean 24-h level, the peak amplitude, or the peak frequency of LH and FSH. The only statistically significant sex differences were a higher mean 24-h FSH level in the stage 1 girls (P < 0.05) and a higher daytime FSH peak amplitude and peak frequency in the stage 5 girls (P < 0.05).

8AM

8PM

8AM 8AM TIME OF DAY

8PM

8AM

FIG. 2. LH (left) and FSH (right) profiles for each girl over 24 h. Girls in each pubertal stage (indicated at upper left corner of each panel) are graphed together to convey the overall temporal pattern of gonadotropin secretion. See Fig. 1 for analysis of mean 24-h level, peak amplitude, and peak frequency at each pubertal stage. FSH levels are not available for one of the stage 2 girls.

LHRH-stimulated gonadotropin secretion LHRH-stimulated peak LH levels increased from stage 1 to stage 4 in both girls and boys (P < 0.005, by ANOVA; Table 4). Within the individual pubertal stages, these increases achieved statistical significance in stage 3, 4, and 5 girls and boys compared to stage 1 children (P < 0.05) and in stage 4 compared to stage 2 girls (P < 0.05). Peak FSH levels varied significantly among the pubertal stages (P = 0.03, by ANOVA), but did not consistently increase or decrease. We also examined the ratio of the peak LH and FSH levels, because this ratio is used as a diagnostic test of pubertal activation of the LHRH neuron (10, 18). The ratio of peak LH to peak FSH (expressed in international units per L) has been considered diagnostic of pubertal onset if it exceeds 1 (a LH-predominant response). All of the prepubertal girls had a ratio of peak LH to peak FSH that was less than 1. However, 3 of the 9 prepubertal boys had a ratio greater than 1, 3 of the 10 girls in pubertal stages 2 and 3 had a ratio less than 1, and 1 of the boys in pubertal stage 2 had a ratio less than 1. In

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 06 September 2014. at 08:01 For personal use only. No other uses without permission. . All rights reserved.

OERTER ETAL.

1254

W Z

10

o S g 30 o

Z m

Z

20

CD _)

20

30

2 2°

J Cb 30

8AM

8PM

8AM

8AM

8PM

8AM

TIME OF DAY FlG. 3. LH (left) and FSH (right) profiles for each boy over 24 h. Boys in each pubertal stage (indicated at upper left corner of each panel) are graphed together to convey the overall temporal pattern of gonadotropin secretion. See Fig. 1 for analysis of mean 24-h level, peak amplitude, and peak frequency at each pubertal stage.

both girls and boys, the ratio increased significantly with increasing pubertal stage (P = 0.005, by ANOVA; Table 4). Within the individual pubertal stages, these increases achieved statistical significance in stage 3 and 4 girls compared to stage 1 girls (P < 0.05), in stage 3, 4, and 5 boys compared to stage 1 boys (P < 0.05), and in stage 4 compared to stage 2 girls (P < 0.05). There were no statistically significant differences in peak LH between girls and boys of the same pubertal stage. However, peak FSH was higher in girls compared to boys (P = 0.0001, by ANOVA). Within the individual pubertal stages, this difference achieved statistical significance at stages 3, 4, and 5 (P < 0.05). As a result of the higher peak FSH levels in the girls, the ratio of peak LH to peak FSH was higher in the prepubertal and early pubertal boys. This comparison achieved statistical significance at stages 1 and 3 (P < 0.05). Indices of pubertal gonadotropin secretion The conventional hormonal criterion for detecting pubertal onset is a LHRH-stimulated peak LH to peak

JCE & M • 1990 Vol 71 • No 5

FSH ratio of 1.0 or more (10,18). Although this criterion detected 95% of the pubertal children in our study population, 35% of the children not yet in puberty also met this criterion. To improve specificity, we analyzed the data separately for each sex. The most accurate index of pubertal gonadotropin secretion in girls was a LHRHstimulated peak LH to peak FSH ratio greater than 0.66, which detected 96% of the pubertal girls with no false positives. A peak LH level after LHRH stimulation of 15 IU/L or more also detected 96% of the pubertal girls with no false positives. Other measures of gonadotropin secretion had intermediate sensitivity in distinguishing pubertal from prepubertal girls with no false positives (100% specificity): mean nighttime LH peak amplitude of 6.7 IU/L or more (88% sensitivity), maximum LH increment after LHRH stimulation of 12 IU/L or more (88% sensitivity), mean nighttime LH level of 5 IU/L or above (80% sensitivity), and maximum nighttime LH level greater than 10 IU/L (92% sensitivity; see Fig. 5 for the maximum nighttime LH level; other data not shown). In the boys, a LHRH-stimulated peak LH to peak FSH ratio of 3.6 or more detected 65% of pubertal boys with no false positives. Although ratios lower than 3.6 gave improved sensitivity for the detection of pubertal boys, progressively lower ratios were exceeded by increasing numbers of prepubertal boys. To improve the efficiency of distinguishing pubertal from prepubertal boys, we examined other measures of gonadotropin secretion. The most accurate index was a maximal nighttime LH level greater than 12 IU/L, which detected 90% of the pubertal boys with no false positives (Fig. 5). Other measures of gonadotropin secretion had intermediate sensitivity in distinguishing pubertal from prepubertal boys with no false positives (100% specificity): mean nighttime LH peak amplitude of 9 IU/L or more (86% sensitivity), maximum LH increment after LHRH stimulation of 25.5 IU/L or above (83% sensitivity), mean nighttime LH level of 7 IU/L or more (81% sensitivity), and peak LHRH-stimulated LH level of 30 IU/L or more (78% sensitivity; data not shown). Since the ability to distinguish pubertal from prepubertal gonadotropin secretion seemed likely to depend on the stage of puberty, we also performed the above analysis using only the stage 1-3 children. The optimal criteria to distinguish pubertal from prepubertal children with 100% specificity were unchanged, since these criteria depend on the highest gonadotropin levels observed in the prepubertal children. However, the sensitivities for the detection of the stage 2 and 3 children were lower. A LHRH-stimulated peak LH to peak FSH ratio of 0.66 or greater detected 90% of the stage 2 and 3 girls compared to 96% of the stage 2-5 girls, and a LHRHstimulated peak LH of 15 or more detected 90% of stage

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 06 September 2014. at 08:01 For personal use only. No other uses without permission. . All rights reserved.

GONADOTROPIN SECRETION DURING PUBERTY

1255

TABLE 2. Comparison of day vs. night gonadotropin and PRL levels, peak amplitude, and peak frequency by pubertal stage in girls

Hormone

stage

Day

Night 3.5 ± 1.0° 5.3 ± 4.1 8.7 ± 5.1 11.0 ± 3.2" 8.4 ± 3.7

LH (IU/L)

1 2 3 4 5

2.9 ± 2.1 ± 5.7 ± 7.9 ± 10.2 ±

0.6 0.6 2.2 2.6 4.0

FSH (IU/L)

1 2 3 4

5

3.8 ± 5.0 ± 7.3 ± 7.2 ± 7.8 ±

1.4 3.6 2.3 1.4 2.5

4.8 ± 6.8 ± 7.2 ± 7.5 ± 7.3 ±

1 2 3 4 5

3.2 ± 3.0 ± 5.0 ± 6.4 ±

0.9 0.9 2.6 2.3

PRL (^g/L)c

1.4

Peak frequency (peaks/ 12 h)

Peak amplitude

Hormone level

Pubertal

Day

0.5 1.6 3.4 3.8 16.1

1.7° 4.8 3.1 1.7 2.6

4.5 ± 6.1 ± 8.2 ± 8.2 ± 9.3 ±

2.9 4.4 2.9 1.5 3.0

6.4 ± 8.0 ± 9.2 ± 8.9 ± 7.5 ±

8.1 ± 1.3° 7.5 ± 3.1° 9.1 ± 3.3* 9.4 ± 2.9"

8.8 ± 4.5 ± 7.4 ± 9.8 ±

6.0 1.3 3.0 3.4

14.2 ± 10.6 ± 12.4 ± 12.9 ±

4.7

2.4 2.1 2.8 1.1 3.9

Night 5.4 ± 2.4° 5.8 ± 2.5 5.8 ± 1.9 6.1 ± 1.7 5.7 ± 2.5

2.5 5.8 3.9 1.5 3.8

2.8 ± 4.0 ± 3.7 ± 4.4 ± 4.7 ±

3.0 2.9 1.2 1.8 2.5

3.3 ± 4.5 ± 4.7 ± 3.6 ± 4.1 ±

0.7 6.8 5.6 3.76

6.0 ± 4.4 ± 4.6 ± 5.8 ±

1.4 1.1 1.0 1.5

6.5 ± 6.0 ± 4.9 ± 6.8 ±

8.4

6.8

Day

4.0 ± 4.5 ± 4.0 ± 5.6 ± 6.7 ±

Night 4.5 ± 1.4° 7.9 ± 5.3 14.3 ± 12.2 14.1 ± 5.5 14.4 ± 5.0

3.5 ± 3.1 ± 5.6 ± 12.9 ± 19.2 ±

3.5

2.4 3.1 2.0 1.9 3.2

7

2.1 1.2 1.6 1.1

Values are the mean ± SD. ° P < 0.05, night vs. day. 6 P< 0.01, night vs. day. c PRL levels were measured in a subset of the girls: one in pubertal stage 1, two in pubertal stage 2, five in pubertal stage 3, seven in pubertal stage 4 and five in pubertal stage 5. TABLE 3. Comparison of day vs. night gonadotropin and PRL levels, peak amplitude, and peak frequency by pubertal stage in boys

Hormone

stage

Day

Night

Day

3 4 5

3.1 ± 3.3 ± 6.8 ± 7.6 ± 6.6 ±

0.8 1.1 2.6 3.7 1.3

3.9 ± 1.4° 7.8 ± 6.1 11.7 ± 6.4 10.7 ± 4.2" 7.9 ± 2.0

3.3 ± 6.4 ± 7.2 ± 10.9 ± 22.9 ±

1.5 1.9 4.8 4.7 40.0

FSH (IU/L)

1 2 3 4 5

2.2 ± 3.6 ± 5.3 ± 5.8 ± 5.8 ±

0.7 1.9 2.6 2.3 1.4

2.9 ± 1.3° 4.1 ± 2.2 5.8 ± 2.7 6.3 ± 2.3" 5.6 ± 1.3

2.8 ± 4.1 ± 6.0 ± 6.9 ± 4.4 ±

1.4 2.1 2.8 2.4 3.6

PRL W L ) "

1 3 4 5

3.3 ± 3.3 ± 3.6 ± 5.7 ±

2.6 1.5 0.7 2.8

8.0 ± 7.2 ± 6.2 ± 7.7 ±

4.7 ± 4.6 ± 4.8 ± 7.2 ±

3.8 2.1 1.2 3.2

LH (IU/L)

1 2

Peak frequency(peaks/ 12 h)

Peak amplitude

Hormone level

Pubertal

2.5° 1.0° 1.4" 2.6°

Night 4.8 ± 2.6° 12.8 ± 11.3 20.4 ± 9.4° 15.6 ± 4.6" 11.5 ± 3.1 4.6 ± 4.9 ± 6.5 ± 7.2 ± 6.9 ±

4.4 2.7 2.8 2.5 5.2

10.9 ± 3.3" 10.6 ± 1.5° 8.0 ± 1.8° 10.3 ± 4.2°

Day

Night

4.6 ± 2.3 ± 4.4 ± 5.3 ± 6.0 ±

3.6 2.3 3.2 2.4 1.8

4.6 ± 5.0 ± 6.4 ± 4.9 ± 6.0 ±

3.0 1.7 2.4" 1.9 2.1

1.9 ± 3.7 ± 4.2 ± 3.4 ± 1.5 ±

1.9 1.5 2.5 2.3 1.9

2.8 ± 4.3 ± 6.0 ± 4.7 ± 3.5 ±

2.4 1.5 1.2 1.6 2.3

5.0 ± 4.8 ± 4.2 ± 5.4 ±

0.7 1.0 1.8 1.1

6.0 ± 5.5 ± 6.6 ± 5.4 ±

0.7 1.3 1.5" 1.7

Values are the mean ± SD. " P < 0.01, night us. day. * PRL levels were measured in a subset of the boys: five in pubertal stage 1, four in pubertal stage 3, five in pubertal stage 4, and five in pubertal stage 5.

2 and 3 girls compared to 96% of the stage 2-5 girls. In the boys, a maximal nighttime LH level of 12 IU/L or above detected 83% of the stage 2 and 3 boys compared to 90% of the stage 2-5 boys. PRL secretion The 24-h mean PRL level, mean peak amplitude, and mean peak frequency did not change significantly with

pubertal stage in either girls or boys (P > 0.2, by ANOVA; data not shown). The mean PRL levels were higher at night than during the day in both sexes and at all stages of puberty (P < 0.05; Tables 2 and 3). The mean PRL peak amplitude was higher at night for both sexes in all stages. Within individual pubertal stages, these differences achieved statistical significance

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 06 September 2014. at 08:01 For personal use only. No other uses without permission. . All rights reserved.

OERTER ETAL.

1256

JCE & M • 1990 Vol 71 • No 5

3

O •** > u OJ

An A \

y ' X

0.5-

Max. night LH level

LH:F5H P

o>

^

BOYS

Peak LH 10.6 ± 3.5 16.7 ± 18.3 38.9 ± 23.7 109.4 ± 62.9 81.7 ± 62.3

Peak FSH Peak LH/peak FSH 23.3 ± 15.7 ± 14.6 ± 20.4 ± 14.0 ±

12.1 6.9 4.8 6.5 2.8

0.5 ± 0.9 ± 2.9 ± 5.6 ± 6.0 ±

0.2 (0.3-0.7) 0.8 (0.3-1.4) 2.0 (0.8-6.0) 3.2 (2.6-12.2) 5.4 (1.1-18.1)

13.5 ± 8.8 11.9 ±6.1° 1.4 ±1.1 (0.4-3.6)° 18.1 ± 9.2 9.3 ± 10.3 6.4 ± 8.2 (0.7-12.3) 38.5 ± 9.4 7.3 ± 2.6° 5.6 ± 1.7 (3.5-7.1)° 55.1 ± 14.8 12.6 ± 3.7° 4.5 ± 1.1 (2.9-5.4) 36.9 ± 6.6 9.6 ±2.1° 4.1 ± 1.5 (2.4-6.2)

Values are the mean ± SD (range). ° P < 0.05, boys us. girls of same pubertal stage.

at stage 5 for girls (P < 0.02) and at all stages tested for boys (P < 0.01). Mean PRL peak frequencies tended to be slightly higher at night in both sexes and at all stages, except for the stage 5 boys. These differences achieved statistical significance only in the stage 4 boys (P < 0.03). No sex differences were observed for the mean 24-h level, the peak amplitude, or the peak frequency of PRL (data not shown).

Discussion The data from the current study, which to our knowledge comprises the largest study of 24-h gonadotropin secretion at all stages of puberty in normal girls and boys, support the following generalizations about pubertal maturation of human gonadotropin secretion. First, mean LH and FSH levels increase throughout puberty

FIG. 5. ROC curves for the LHRH-stimulated peak LH to peak FSH ratio (LH:FSH) and for the maximum spontaneous nighttime LH level in girls and boys. The ROC curves depict the relationship between fraction of true positives and the fraction of false positives for all possible criteria for a positive and a negative result. True positives are pubertal children correctly identified as pubertal. False positives are prepubertal children incorrectly identified as pubertal.

due to a large increase in gonadotropin peak amplitude. Gonadotropin peak frequency did not change significantly with pubertal stage in this study. Second, higher nighttime than daytime LH peak amplitude occurs throughout the childhood years and the first 3 (girls) or 4 (boys) stages of puberty, but disappears in both sexes by stage 5. Stages 2 and 3 of puberty are characterized by a marked increase in the nocturnal augmentation of LH peak amplitude in both sexes. Additionally, the circadian variation of the LH peak frequency is most marked at these stages, with an apparent increase in nighttime LH peak frequency. Third, boys have a much greater LHRH-stimulated peak LH to peak FSH ratio than girls during pubertal stages 1-3, which necessitates the use of sex-specific criteria for the interpretation of this test. Moreover, sex-specific measures of gonadotropin secretion are required to best distinguish pubertal from prepubertal children: the LHRH test for girls, and the maximum spontaneous nighttime LH level for boys. The data presented here confirm and extend earlier findings that LH and FSH increase in girls and boys with increasing pubertal stage (5-9). Lee et al. (5), August et al. (6), Lee et al. (7), and Beck and Wuttke (9) all reported increased gonadotropin levels with increasing pubertal stage, based on single measurements of gonadotropins. Penny et al. (8) reported increases in mean LH and FSH levels and in mean LH and FSH peak amplitude, with no increase in LH or FSH peak frequency, as pubertal stage increased. We observed day-night differences in mean LH level and/or peak amplitude before puberty and through stage 4 of puberty. The day-night difference disappeared at stage 5. LH and FSH peaks were seen throughout the daytime and the nighttime in all pubertal stages and in

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 06 September 2014. at 08:01 For personal use only. No other uses without permission. . All rights reserved.

GONADOTROPIN SECRETION DURING PUBERTY

1257

both sexes. These data from normal children confirm the conclusions of several earlier studies, many of which included children with delayed puberty. These earlier studies include the observation of a sleep-associated rise in mean LH levels in prepubertal girls (3) and boys (1, 2, 4) and in pubertal girls (19) and boys (1, 2, 20, 21). Although Boyar et al. (19) did not observe sleep-induced augmentation of LH levels in prepubertal children, this probably reflected the small number of children studied and the smaller day-night difference in this group. Our observation of an apparent circadian variation in LH peak frequency in the stage 2 and 3 girls and boys agrees closely with the observations of Kelch et al. in pubertal boys (22), who found the highest LH peak frequency at 0100-0200 h in prepubertal children and at 2200-2400 h in pubertal children. However, in our children the circadian variation of LH peak frequency was not statistically significant by x 2 analysis. This nighttime increase in LH peak frequency disappeared completely in our pubertal stage 4 and 5 children, indicating that this apparent nocturnal augmentation of peak frequency is limited to prepubertal and early pubertal children. Many investigators have used the LHRH test in an attempt to detect pubertal activation of the LHRH neuron (18, 23-28), which is important in the differential diagnosis of precocious puberty, precocious thelarche, and precocious adrenarche (18, 24, 25). The LHRH test has also been used in an attempt to distinguish other disorders of puberty, including delayed puberty, isolated gonadotropin deficiency, and gonadal dysgenesis (23, 25). Although several investigators have suggested sex differences in the gonadotropin responses to LHRH (24, 25), to our knowledge no one has previously developed sexspecific criteria for the interpretation of this test. Additionally, several measures of response have been used to interpret this test, such as the peak LH level (18, 2 3 28), the maximum LH increment above baseline (25-28),

a maximum spontaneous nighttime LH of 12 IU/L or more detected 90% of pubertal boys with no false positives compared to 83% for the best measure of stimulated gonadotropin secretion. Our study does not address the mechanism for the sex difference in gonadotropin responses to LHRH. However, the observation that inhibin levels are twice as high in boys as in girls at each stage of puberty (29) suggests that the lower peak FSH levels in boys may result from their higher levels of inhibin. The clinical significance of these observations is that criteria for the onset of pubertal gonadotropin secretion should probably be revised on the basis of these normative data and, in addition, should be sex specific. This conclusion is based on the hypothesis that clinical disorders of puberty that lack activation of the LHRH neuron, such as premature thelarche, premature adrenarche, McCune-Albright syndrome and its variants, and delayed puberty, will have gonadotropin secretion that does not exceed that of normal prepubertal children, and that disorders involving activation of the LHRH neuron, such as central precocious puberty, will have gonadotropin secretion similar to that of normal pubertal children. The available data appear consistent with this hypothesis, since all children from our previous study of premature thelarche had a LHRH-stimulated peak LH to peak FSH ratio greater than 0.66, and two children with central precocious puberty that we previously considered to have a prepubertal response to LHRH based upon a peak LH to peak FSH ratio below 1 would now be classified as having a pubertal response based upon the revised criterion from this study (18). However, although these criteria for pubertal gonadotropin secretion appear to be applicable to clinical settings, their usefulness in differential diagnosis must ultimately be evaluated in patients with disorders of puberty.

and the ratio of peak LH to peak FSH (10,18); however, the relative usefulness of these measures has not previously been established. We observed marked sex differences in the gonadotropin response to LHRH, specifically a higher peak FSH response and, hence, a lower peak LH to peak FSH ratio, in the prepubertal and early pubertal girls. Moreover, the measure of response to LHRH stimulation that was most useful in distinguishing pubertal from prepubertal children differed between the sexes; a peak LH to peak FSH ratio above 0.66 was most useful in girls (96% sensitivity; 100% specificity), whereas the maximum LH increment above baseline was most useful in boys (83% sensitivity; 100% specificity). However, although LHRH-stimulated gonadotropin levels were most efficient in distinguishing pubertal from prepubertal girls, spontaneous gonadotropin levels were most efficient for the boys. Specifically,

References 1. Lee PA, Plotnick LP, Steele RE, Thompson RG, Blizzard RM. Integrated concentrations of luteinizing hormone and puberty. J Clin Endocrinol Metab. 1976;43:168-72. 2. Parker DC, Judd HL, Rossman LG, Yen SSC. Pubertal sleep-wake patterns of episodic LH, FSH and testosterone release in twin boys. J Clin Endocrinol Metab. 1975;40:1099-109. 3. Jakacki RI, Kelch RP, Sauder SE, Lloyd JS, Hopwood NJ, Marshall JC. Pulsatile secretion of luteinizing hormone in children. J Clin Endocrinol Metab. 1982;55:453-8. 4. Judd HL, Parker DC, Yen SSC. Sleep-wake patterns of LH and testosterone release in prepubertal boys. J Clin Endocrinol Metab. 1977;44:865-9. 5. Lee PA, Xenakis T, Winer J, Matsenbaugh S. Puberty in girls: correlation of serum levels of gonadotropins, prolactin, androgens, estrogens and progestins with physical changes. J Clin Endocrinol Metab. 1976;43:775-84. 6. August GP, Grumbach MM, Kaplan SL. Hormonal changes in puberty. III. Correlation of plasma testosterone, LH, FSH, testicular size, and bone age with male pubertal development. J Clin Endocrinol Metab. 1972;34:319-26. 7. Lee PA, Jaffe RB, Midgley Jr AR. Serum gonadotropin, testoster-

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 06 September 2014. at 08:01 For personal use only. No other uses without permission. . All rights reserved.

1258

8. 9.

10. 11. 12. 13. 14. 15. 16.

17. 18.

19.

OERTER ETAL.

one and prolactin concentrations throughout puberty in boys: a longitudinal study. J Clin Endocrinol Metab. 1974;39:664-72. Penny R, Olambiwonnu NO, Frasier SD. Episodic fluctuations of serum gonadotropins in pre- and post-pubertal girls and boys. J Clin Endocrinol Metab. 1977;45:307-ll. Beck W, Wuttke W. Diurnal variations of plasma luteinizing hormone, follicle-stimulating hormone, and prolactin in boys and girls from birth to puberty. J Clin Endocrinol Metab. 1980;50:6359. Cutler Jr GB. Precocious puberty. In: Hurst JW, ed. Medicine for the practicing physician. Boston: Butterworth; 1988;526-30. Marshall WA, Tanner JM. Variations in pattern of pubertal changes in girls. Arch Dis Child. 1969;44:291. Marshall WA, Tanner JM. Variations in pattern of pubertal changes in boys. Arch Dis Child. 1970;45:13. Zachmann M, Prader A, Kind HP, Haflinger H, Budlider H. Testicular volume during adolescence. Helv Pediatr Acta. 1974;29:61. Cutler GB Jr, Glenn M, Bush M, Hodgen GD, Graham CE, Loriaux DL. Adrenarche: a survey of rodents, domestic animals and primates. Endocrinology. 1978;103:2112. Ruder HJ, Guy RL, Lipsett MB. Radioimmunoassay for cortisol in plasma and urine. J Clin Endocrinol Metab. 1972;35:219. Oerter KE, Guardabasso V, Rodbard D. Detection and characterization of peaks and estimation of instantaneous secretory rate for episodic pulsatile hormone secretion. Comput Biomed Res. 1986;19:170-91. McNeil BJ, Keeler E, Adelstein SJ. Primer on certain elements of medical decision making. N Engl J Med. 1975;293:211-5. Pescovitz OH, Hench KD, Barnes KM, Loriaux DL, Cutler Jr GB. Premature thelarche and central precocious puberty: the relationship between clinical presentation and the gonadotropin response to luteinizing hormone-releasing hormone. J Clin Endocrinol Metab. 1988;67:474-9. Boyar R, Finkelstein J, Roffwarg H, Kapen S, Weitzman E, Hellman L. Synchronization of augmented luteinizing hormone secretion with sleep during puberty. N Engl J Med. 1972;287:582-6.

JCE & M • 1990 Vol 71« No 5

20. Corley KP, Valk TW, Kelch RP, Marshall JC. Estimation of GnRH pulse amplitude during pubertal development. Pediatr Res. 1981;15:157-62. 21. Boyar R, Rosenfeld RS, Kapen S, et al. Human puberty: simultaneous augmented secretion of luteinizing hormone and testosterone during sleep. J Clin Invest. 1974;54:609-18. 22. Kelch RP, Hopwood NJ, Sauder S, Marshall JC. Evidence for decreased secretion of gonadotropin-releasing hormone in pubertal boys during short-term testosterone treatment. Pediatr Res. 1985;19:112-7. 23. Vourtsis K, Illig R. The use of a low-dose LHRH test for the distinction of delayed adolescence and isolated gonadotropin deficiency. Helv Paediatr Acta. 1982;37:437-47. 24. Reiter EO, Kaplan SL, Conte FA, Grumbach MM. Responsivity of pituitary gonadotropes to luteinizing hormone-releasing factor in idiopathic precocious puberty, precocious thelarche, precocious adrenarche and in patients treated with medroxyprogesterone acetate. Pediatr Res. 1975;9:lll-6. 25. Zipf WB, Kelch RP, Hopwood NJ, Spencer ML, Bacon GE. Suppressed responsiveness to gonadotropin-releasing hormone in girls with unsustained isosexual precocity. J Pediatr. 1979;95:38-43. 26. Job JC, Gamier PE, Chaussain JL, Milhaud G. Elevation of serum gonadotropins (LH and FSH) after releasing hormone (LH-RH) injection in normal children and in patients with disorders of puberty. J Clin Endocrinol Metab. 1972;35:473-6. 27. Roth JC, Kelch RP, Kaplan SL, Grumbach MM. FSH and LH response to luteinizing hormone-releasing factor in prepubertal and pubertal children, adult males and patients with hypogonadotropic and hypergonadotropic hypogonadism. J Clin Endocrinol Metab. 1972;35:926-30. 28. Roth JC, Grumbach MM, Kaplan SL. Effect of synthetic luteinizing hormone-releasing factor on serum testosterone and gonadotropins in prepubertal, pubertal and adult males. J Clin Endocrinol Metab. 1973;37:680-6. 29. Burger HG, McLachlan RI, Bangah M, et al. Serum inhibin concentrations rise throughout normal male and female puberty. J Clin Endocrinol Metab. 1988;67:689-94.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 06 September 2014. at 08:01 For personal use only. No other uses without permission. . All rights reserved.

Gonadotropin secretory dynamics during puberty in normal girls and boys.

To determine the most useful index of pubertal gonadotropin secretion we measured spontaneous LH and FSH levels every 20 min for 24 h and the LH and F...
1012KB Sizes 0 Downloads 0 Views