0021-972x/92/7502-0536$03.00/0 Journal of Clinical Endocrinology and Metabolism Copyright 0 1992 by The Endocrine Society
Vol. 75, No. 2 Printed in LL9.A.
Corticoid-Induced Growth GH-Deficient and Normal PEDRO MARTUL, JAIME PINEDA, AND FELIPE F. CASANUEVA
CARLOS
Hormone Children*
(GH) Secretion
in
DIEGUEZ,
Pediatric Endocrinology Section, Hospital de Cruces (P.M., J.P.); Endocrine Section, Hospital General Galicia; the Departments of Medicine (F.F.C.) and Physiology (C.D.), Faculty of Medicine, Compostela University. Bilbao and Santiago de Compostela, Spain ABSTRACT
under the secretory curve were 1130 f 55 and 616 + 54 for the control and GH-deficient children, respectively. GH release in children after dexamethasone administration followed the pattern previously described for adults, i.e. there were no modifications of basal GH levels in the first 2 h, the GH peak appeared around the third hour, and the GH levels remained increased until the fourth hour after dexamethasone administration. Individually considered, practically all control children, but only 2 of 12 GH-deficient children, presented a dexamethasone-induced GH peak over the 10 pg/L level. In both normal and GH-deficient patients, corticoids appeared just as potent a stimulus as propranolol-exercise and clonidine, and more potent than hypoglycemia. This new stimulus showed a pattern similar to that of the hypothalamic stimuli, but clearly different with respect to the pituitary one (GHRH), suggesting that corticoids activate GH secretion by acting at hypothalamic level. In conclusion, acute administration of corticoids could be a suitable test in the diagnostic armamentaria of GH-deficient states. (J Clin Endocrinol Metab 75: 536-539, 1992)
Acute administration of corticoids is a potent stimulus of GH secretion in man. To ascertain their mechanism and point of action as well as the suitability of this novel test in the diagnosis of GH-deficient states, normal controls and GH-deficient children were studied. They were selected based on auxological criteria and the GH response to provocative stimuli. The GH-deficient children presented a blunted GH (mean z? SEM; pg/L) discharge after insulin-induced hypoglycemia (2.9 + 0.4), propranolol-exercise (7.4 + 1.5). and clonidine (6.5 & 0.8) compared with ialues in the normal children (7.2 If: 2.2, 15.8 f 2.4, and 15.6 + 1.8, resuectivelv). As exnected. GH-releasing hormone (GHRH)-induced Gfi discha&e in GH’-deficient children (33.2 + 4.9) was similar to that in the control children (35.1 f. 6.0). Administration of 2 mg/m’ body surface dexamethasone, iv, to normal children induced, 3 h later, a mean GH peak of 14.1 + 1.2 pg/ L. This was significantly higher that the corticoid-induced GH peak in GH-deficient children (6.7 f 1.1 pg/L). The corticoid-induced areas
I
de
T IS WELL known that chronic exposure to corticoids leads to growth delay and a state of blunted GH secretion in man (l-3). Moreover, we have recently reported a new observation, i.e. that acutely administered corticoids can induce a powerful GH secretion response in man (4-6). The observed stimulatory action of such steroid compounds has somepeculiar features; for example, after iv administration, the GH stimulatory effect takes nearly 3 h to become evident, and then the net secretory discharge is very significant (6). Acute administration of corticoids appears to be one of the most potent and the most delayed GH stimuli known so far. The precise mechanism and locus of action of these corticoidsin inducing GH releaseare unknown, and the potential clinical applications of such a new GH stimulus have not been explored. In the present work, corticoids were employed as GH stimulants in both normal and GH-deficient children. The two aims of the study were 1) to evaluate whether corticoids could be a suitable stimulus to incorporate into the diagnostic armamentaria of GH-deficient states, and 2) to attempt to understand more fully the corticoid mechanism and point of action. The results of this study suggest that corticoid-induced GH secretion is blunted in GH-deficient
states and that corticoids probably act at the hypothalamic level.
Received April 30, 1991. Address all correspondence and requests for reprints to: F. F. Casanueva, M.D., Ph.D., P.O. Box 563,15780 Santiago de Compostela, Spain. *This work was supported by Grants 89-0127.3 and 91-0134 from Fondo de Investigation Sanitaria (FIS), the Spanish Ministry of Health, and a research grant from Xunta de Galicia.
Subjects and Methods Twelve children (five girls and seven boys) of short stature due to GH secretory deficiency were studied (Table 1). Their ages ranged from 8.2-16.1 yr (mean age, 11.0 f 0.7 yr), and as a group they were more than 1.8 SD below the mean height for their age, with a height velocity 2.5 SD below the mean and a delay in bone age of more than 2 yr. None had multiple pituitary hormone deficiencies, especially ACTH deficiency, or had had previous treatment with GH or anabolic or sex steroids, and none had historical or physical evidence of nutritional deficiency, intrauterine growth retardation, systemic disease, dysmorphology, or psychological deprivation. Patient 1 had received radiotherapy for a retinoblastoma, patient 10 has a harelip, and patient 11 has a neurofibroblastoma; the rest consulted only for delayed growth. All subjects had abnormal responses to at least two pharmacological GH stimulatory tests, and after the diagnosis of GH secretory deficiency of hypothalamic origin was established, they underwent the dexamethasone test. All tests were performed within 3 months. After diagnosis, the subjects were treated with recombinant human GH at a dose-of 0.7 ILJ/ka. week bv dailv SC iniection for 1 vr (Table 1). ‘Th”e controigroub consisted of 36 &l&en of both sexes (mean age, 10.1 + 0.5 yr), all with normal growth and growth velocities and normal responses to standard stimuli of GH secretion. Approval for this study was obtained from the Hospital Committee after receiving informed parental consent. All tests were performed after habituation to the hospital setting. The experiments started at 0900 h after an overnight fast with the subjects recumbent. A butterfly needle (21 gauge) was-placed in a forearm ‘vein and kept patent with saline-heparin solution. The insulin-induced hypoglycemia test was undertaken in 12 GH-deficient and 7 control
536
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 20 November 2015. at 03:17 For personal use only. No other uses without permission. . All rights reserved.
CORTICOID-INDUCED TABLE
1. Clinical
characteristics
of GH-deficient
children
at
velocity
(cm/yr)
GH SECRETION 11 , 10
diagnosis ‘;fnt 1 2 3 4 5 6 7 8 9 10 11 12
zr;
Sex
Bone we (Yd
Ht (cm)
;ioE
8.2 10.4 16.1 10.8 13.0 8.4 12.2 9.6 8.9 10.2 14.4 9.7
F F M M M M M F M M F F
4.6 8.9 11.0 8.0 10.6 6.0 10.6 8.4 6.6 10.0 12.0 9.0
110.3 127.0 142.4 127.5 137.7 115.6 131.0 122.5 115.8 123.4 139.1 120.5
-2.7 -1.9 -4.6 -2.1 -2.2 -2.2 -2.6 -1.8 -2.4 -2.3 -3.6 -2.7
The increase included.
in
growth
velocity
after
Growth
At diagnosis 0.8 2.0 3.3 3.0 2.8 3.8 3.5 4.0 3.4 4.1 1.9 4.3 1 yr
of GH
9
i
Posttreatment
8
!?
5.8 7.5 9.6 11.1 7.0 6.9 7.4 7.0 7.2 7.8 6.4 8.0 treatment
537
1 T
01
is
children. At 0 min, 0.1 IU/kg regular insulin (Velosulin, Novo-Nordisk, Madrid, Spain) was administered iv, and blood sampling followed for 120 min. The exercise-propranolol test was performed on 12 GHdeficient and 9 control children. Propranolol (Sumial, ICI, Madrid, Spain) was administered at a dose of 40 mg, orally, and after 90 min of rest followed by 20 min of physical exercise, a blood sample was obtained. The clonidine test was performed in 6 GH-deficient and 14 control children. Clonidine (Catapresan, Boehringer Mannheim, Madrid, Spain) at a dose of 0.150 mg, orally, was administered at 0 min, and blood sampling was performed for 120 min. The GH-releasing hormone (GHRH) test was undertaken in 11 GH-deficient and 7 control children. GHRH-(l-29)NH2 (Geref, Serono, Madrid, Spain), was administered at a dose of 1.5 pg/kg, iv, at 0 min, and blood samples were obtained for 120 min. Finally, the dexamethasone test was carried out in 12 GHdeficient and 8 control children. At 0 min, dexamethasone (Fortecortin, Merck, Madrid, Spain) was administered iv at a dose of 2 mg/m* body surface, and blood sampling was performed every 30 min during the first 2 h, every 15 min during the third and fourth hours, and every 30 min during the last hour; the last blood sample was obtained 5 h after dexamethasone administration (6). The time elapsed between the realization of the classical stimuli and the dexamethasone stimulus was always less than 45 days. Human GH was measured by RIA, using commercial kits (DPC, Los Angeles, CA) with a sensitivity of 1 ng/mL. The intraassay coefficient of variation was 4.9% for a GH concentration of 3.6 pg/L (n = 18), and 3.7% for a concentration of 19.7 pg/L (n = 18). The interassay coefficients of variation were 6.2%, 4.3%, and 4.5% for GH concentrations of 3.1, 7.9, and 18.2 @g/L, respectively (n = 30). The areas under the secretory curve were calculated by a trapezoidal method. The mean GH peak, GH increases over basal values, and areas under the curve were compared between groups by the Mann-Whitney test. P < 0.05 was considered significant.
Hypoglycemia
0
20 40
60
90
120
Minutes FIG. 1. Mean in a group
+ SEM GH secretion after of normal (0) and GH-deficient
insulin-induced (0) children.
hypoglycemia
4o38 36 3432 J ; = $ E 2 5 6 a
GHRH
30 28262422201816141210 06-
Results Insulin administration induced a significant drop in blood glucose, which was similar in both groups of children and greater than 45 mg/dL from the basal value. The hypoglycemia-induced GH secretion in the control group was 7.2 f 2.2 pg/L at 40 min and afterward returned to near-basal values (Fig. 1). The hypoglycemia-stimulated GH secretion in the GH-deficient group was considerably dampened, with a peak of 2.9 rt 0.4 pg/L. The mean GH peaks were significantly different for the control and GH-deficient children (8.3 + 1.5 and 3.7 + 0.4 pg/L, respectively; P < 0.05). The second hypothalamic stimulus employed, propranolol-exercise (seeFig. 4), induced a mean GH peak of 15.8 + 2.4 pg/ L at 110 min in the control group, significantly higher (P
Vol. 75, No. 2 Printed in LL9.A.
Corticoid-Induced Growth GH-Deficient and Normal PEDRO MARTUL, JAIME PINEDA, AND FELIPE F. CASANUEVA
CARLOS
Hormone Children*
(GH) Secretion
in
DIEGUEZ,
Pediatric Endocrinology Section, Hospital de Cruces (P.M., J.P.); Endocrine Section, Hospital General Galicia; the Departments of Medicine (F.F.C.) and Physiology (C.D.), Faculty of Medicine, Compostela University. Bilbao and Santiago de Compostela, Spain ABSTRACT
under the secretory curve were 1130 f 55 and 616 + 54 for the control and GH-deficient children, respectively. GH release in children after dexamethasone administration followed the pattern previously described for adults, i.e. there were no modifications of basal GH levels in the first 2 h, the GH peak appeared around the third hour, and the GH levels remained increased until the fourth hour after dexamethasone administration. Individually considered, practically all control children, but only 2 of 12 GH-deficient children, presented a dexamethasone-induced GH peak over the 10 pg/L level. In both normal and GH-deficient patients, corticoids appeared just as potent a stimulus as propranolol-exercise and clonidine, and more potent than hypoglycemia. This new stimulus showed a pattern similar to that of the hypothalamic stimuli, but clearly different with respect to the pituitary one (GHRH), suggesting that corticoids activate GH secretion by acting at hypothalamic level. In conclusion, acute administration of corticoids could be a suitable test in the diagnostic armamentaria of GH-deficient states. (J Clin Endocrinol Metab 75: 536-539, 1992)
Acute administration of corticoids is a potent stimulus of GH secretion in man. To ascertain their mechanism and point of action as well as the suitability of this novel test in the diagnosis of GH-deficient states, normal controls and GH-deficient children were studied. They were selected based on auxological criteria and the GH response to provocative stimuli. The GH-deficient children presented a blunted GH (mean z? SEM; pg/L) discharge after insulin-induced hypoglycemia (2.9 + 0.4), propranolol-exercise (7.4 + 1.5). and clonidine (6.5 & 0.8) compared with ialues in the normal children (7.2 If: 2.2, 15.8 f 2.4, and 15.6 + 1.8, resuectivelv). As exnected. GH-releasing hormone (GHRH)-induced Gfi discha&e in GH’-deficient children (33.2 + 4.9) was similar to that in the control children (35.1 f. 6.0). Administration of 2 mg/m’ body surface dexamethasone, iv, to normal children induced, 3 h later, a mean GH peak of 14.1 + 1.2 pg/ L. This was significantly higher that the corticoid-induced GH peak in GH-deficient children (6.7 f 1.1 pg/L). The corticoid-induced areas
I
de
T IS WELL known that chronic exposure to corticoids leads to growth delay and a state of blunted GH secretion in man (l-3). Moreover, we have recently reported a new observation, i.e. that acutely administered corticoids can induce a powerful GH secretion response in man (4-6). The observed stimulatory action of such steroid compounds has somepeculiar features; for example, after iv administration, the GH stimulatory effect takes nearly 3 h to become evident, and then the net secretory discharge is very significant (6). Acute administration of corticoids appears to be one of the most potent and the most delayed GH stimuli known so far. The precise mechanism and locus of action of these corticoidsin inducing GH releaseare unknown, and the potential clinical applications of such a new GH stimulus have not been explored. In the present work, corticoids were employed as GH stimulants in both normal and GH-deficient children. The two aims of the study were 1) to evaluate whether corticoids could be a suitable stimulus to incorporate into the diagnostic armamentaria of GH-deficient states, and 2) to attempt to understand more fully the corticoid mechanism and point of action. The results of this study suggest that corticoid-induced GH secretion is blunted in GH-deficient
states and that corticoids probably act at the hypothalamic level.
Received April 30, 1991. Address all correspondence and requests for reprints to: F. F. Casanueva, M.D., Ph.D., P.O. Box 563,15780 Santiago de Compostela, Spain. *This work was supported by Grants 89-0127.3 and 91-0134 from Fondo de Investigation Sanitaria (FIS), the Spanish Ministry of Health, and a research grant from Xunta de Galicia.
Subjects and Methods Twelve children (five girls and seven boys) of short stature due to GH secretory deficiency were studied (Table 1). Their ages ranged from 8.2-16.1 yr (mean age, 11.0 f 0.7 yr), and as a group they were more than 1.8 SD below the mean height for their age, with a height velocity 2.5 SD below the mean and a delay in bone age of more than 2 yr. None had multiple pituitary hormone deficiencies, especially ACTH deficiency, or had had previous treatment with GH or anabolic or sex steroids, and none had historical or physical evidence of nutritional deficiency, intrauterine growth retardation, systemic disease, dysmorphology, or psychological deprivation. Patient 1 had received radiotherapy for a retinoblastoma, patient 10 has a harelip, and patient 11 has a neurofibroblastoma; the rest consulted only for delayed growth. All subjects had abnormal responses to at least two pharmacological GH stimulatory tests, and after the diagnosis of GH secretory deficiency of hypothalamic origin was established, they underwent the dexamethasone test. All tests were performed within 3 months. After diagnosis, the subjects were treated with recombinant human GH at a dose-of 0.7 ILJ/ka. week bv dailv SC iniection for 1 vr (Table 1). ‘Th”e controigroub consisted of 36 &l&en of both sexes (mean age, 10.1 + 0.5 yr), all with normal growth and growth velocities and normal responses to standard stimuli of GH secretion. Approval for this study was obtained from the Hospital Committee after receiving informed parental consent. All tests were performed after habituation to the hospital setting. The experiments started at 0900 h after an overnight fast with the subjects recumbent. A butterfly needle (21 gauge) was-placed in a forearm ‘vein and kept patent with saline-heparin solution. The insulin-induced hypoglycemia test was undertaken in 12 GH-deficient and 7 control
536
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 20 November 2015. at 03:17 For personal use only. No other uses without permission. . All rights reserved.
CORTICOID-INDUCED TABLE
1. Clinical
characteristics
of GH-deficient
children
at
velocity
(cm/yr)
GH SECRETION 11 , 10
diagnosis ‘;fnt 1 2 3 4 5 6 7 8 9 10 11 12
zr;
Sex
Bone we (Yd
Ht (cm)
;ioE
8.2 10.4 16.1 10.8 13.0 8.4 12.2 9.6 8.9 10.2 14.4 9.7
F F M M M M M F M M F F
4.6 8.9 11.0 8.0 10.6 6.0 10.6 8.4 6.6 10.0 12.0 9.0
110.3 127.0 142.4 127.5 137.7 115.6 131.0 122.5 115.8 123.4 139.1 120.5
-2.7 -1.9 -4.6 -2.1 -2.2 -2.2 -2.6 -1.8 -2.4 -2.3 -3.6 -2.7
The increase included.
in
growth
velocity
after
Growth
At diagnosis 0.8 2.0 3.3 3.0 2.8 3.8 3.5 4.0 3.4 4.1 1.9 4.3 1 yr
of GH
9
i
Posttreatment
8
!?
5.8 7.5 9.6 11.1 7.0 6.9 7.4 7.0 7.2 7.8 6.4 8.0 treatment
537
1 T
01
is
children. At 0 min, 0.1 IU/kg regular insulin (Velosulin, Novo-Nordisk, Madrid, Spain) was administered iv, and blood sampling followed for 120 min. The exercise-propranolol test was performed on 12 GHdeficient and 9 control children. Propranolol (Sumial, ICI, Madrid, Spain) was administered at a dose of 40 mg, orally, and after 90 min of rest followed by 20 min of physical exercise, a blood sample was obtained. The clonidine test was performed in 6 GH-deficient and 14 control children. Clonidine (Catapresan, Boehringer Mannheim, Madrid, Spain) at a dose of 0.150 mg, orally, was administered at 0 min, and blood sampling was performed for 120 min. The GH-releasing hormone (GHRH) test was undertaken in 11 GH-deficient and 7 control children. GHRH-(l-29)NH2 (Geref, Serono, Madrid, Spain), was administered at a dose of 1.5 pg/kg, iv, at 0 min, and blood samples were obtained for 120 min. Finally, the dexamethasone test was carried out in 12 GHdeficient and 8 control children. At 0 min, dexamethasone (Fortecortin, Merck, Madrid, Spain) was administered iv at a dose of 2 mg/m* body surface, and blood sampling was performed every 30 min during the first 2 h, every 15 min during the third and fourth hours, and every 30 min during the last hour; the last blood sample was obtained 5 h after dexamethasone administration (6). The time elapsed between the realization of the classical stimuli and the dexamethasone stimulus was always less than 45 days. Human GH was measured by RIA, using commercial kits (DPC, Los Angeles, CA) with a sensitivity of 1 ng/mL. The intraassay coefficient of variation was 4.9% for a GH concentration of 3.6 pg/L (n = 18), and 3.7% for a concentration of 19.7 pg/L (n = 18). The interassay coefficients of variation were 6.2%, 4.3%, and 4.5% for GH concentrations of 3.1, 7.9, and 18.2 @g/L, respectively (n = 30). The areas under the secretory curve were calculated by a trapezoidal method. The mean GH peak, GH increases over basal values, and areas under the curve were compared between groups by the Mann-Whitney test. P < 0.05 was considered significant.
Hypoglycemia
0
20 40
60
90
120
Minutes FIG. 1. Mean in a group
+ SEM GH secretion after of normal (0) and GH-deficient
insulin-induced (0) children.
hypoglycemia
4o38 36 3432 J ; = $ E 2 5 6 a
GHRH
30 28262422201816141210 06-
Results Insulin administration induced a significant drop in blood glucose, which was similar in both groups of children and greater than 45 mg/dL from the basal value. The hypoglycemia-induced GH secretion in the control group was 7.2 f 2.2 pg/L at 40 min and afterward returned to near-basal values (Fig. 1). The hypoglycemia-stimulated GH secretion in the GH-deficient group was considerably dampened, with a peak of 2.9 rt 0.4 pg/L. The mean GH peaks were significantly different for the control and GH-deficient children (8.3 + 1.5 and 3.7 + 0.4 pg/L, respectively; P < 0.05). The second hypothalamic stimulus employed, propranolol-exercise (seeFig. 4), induced a mean GH peak of 15.8 + 2.4 pg/ L at 110 min in the control group, significantly higher (P
