0021 -972X/90/7003-0585$02.00 Journal of Clinical Endocrinology and Metabolism Copyright © 1990 by The Endocrine Society

Vol. 71, No. 3 Printed in U.S.A.

Pulsatile Growth Hormone Secretion in Patients with Acromegaly and Normal Men: The Effects of Growth Hormone-Releasing Hormone Infusion MARIE C. GELATO, EDWARD OLDFIELD, D. LYNN LORIAUX, AND GEORGE R. MERRIAM Developmental Endocrinology Branch, National Institute of Child Health and Human Development; and Surgical Neurology Branch, National Institute of Neurological Disorders and Stroke, National Institutes of Health, Bethesda, Maryland 20892

ABSTRACT. Twenty-four GH secretory patterns were studied before and during continuous infusions of GHRH in six patients with active acromegaly and in six normal adult men. GH release was episodic in both groups. Control subjects showed a normal diurnal variation in GH release, with the majority of GH released at night (2200-0800 h); mean levels were 1.5 ± 0.4 (SE) ng/mL (day) and 4.2 ± 0.8 ng/mL (night). Acromegalics had no diurnal variation in GH; levels were 45.3 ± 13.7 ng/mL (day) and 39.8 ± 12.2 ng/mL (night). Acromegalics demonstrated an increased frequency of GH pulses compared to normals (11.8 ± 0.8 vs. 2.2 ± 0.3/24 h). During continuous 24-h infusions of GHRH, the normal subjects continued to show a diurnal variation in GH release, but GH pulse frequency increased to a rate (11.7 ± 1.4 pulses/24 h) very similar to that of the patients with acromegaly. In contrast, GHRH infusion did not alter the GH pulse frequency in the acromegalics. GHRH increased the mean levels of

GH in both groups (patients 80.2 ± 20.3 vs. 41.0 ± 12.1 ng/mL, x ± SE. P < 0.05; controls 10.2 ± 2.0 vs. 3.33 ± 0.5 ng/mL, P < 0.01). Some of the patients with acromegaly showed a progressive decline in GH levels during the infusion period, suggesting desensitization or exhaustion of releaseable stores; however, GH levels remained above basal values in all patients. After the 24h GHRH infusions, the GH response to a bolus of GHRH was diminished in the normal subjects (2.1 ± 0.9 vs. 16.8 ± 5 ng/mL, x ± SE; P < 0.01) but not in the acromegalic patients (30.2 ± 8.9 vs. 35.5 ± 12.5 ng/mL; NS). These results indicate that GH release is episodic under basal conditions and during continuous GHRH infusion in both acromegalic and normal subjects, indicating the importance of other modulators of GH release, such as somatostatin, which may remain pulsatile even in acromegaly. (J Clin Endocrinol Metab 7 1 : 585-590, 1990)

I

N normal individuals, GH secretion is episodic, with very low basal GH levels punctuated by bursts of secretion which are more numerous during periods of slow-wave sleep and may also follow meals, exercise, stress, or other stimuli (1). This pattern is not intrinsic to the pituitary but results from influences of two hypothalamic peptides: GHRH, which stimulates GH secretion, and SRIF, which inhibits GH release (2). Even when GHRH is infused continuously to provide constant pituitary stimulation, GH release remains pulsatile (3, 4), suggesting ongoing episodic inhibition through secretion of SRIF and perhaps other factors. In contrast, patients with acromegaly due to GHproducing pituitary tumors appear to have a continuous release of GH, with levels usually detectable at all times although with wide variations, some with ultradian flue-

tuations of large amplitude. The pattern of these changes has not been fully clarified (5). These fluctuations in GH secretion may be intrinsic to the behavior of clusters of tumor cells. Alternatively, they might represent the activation of physiological feedback mechanisms which attempt to reduce the abnormally elevated levels of GH and insulin-like growth factor-I for example by suppressing GHRH and perhaps by stimulating SRIF secretion. Such feedback effects of GH and IGF-I have been described in animal studies, although there has been no direct demonstration of such regulation in humans (68). In an attempt to characterize these changes and gain insight into the mechanisms underlying them, we studied patterns of episodic GH release in patients with acromegaly and compared them to those of normal subjects basally and during continuous 24-h GHRH infusion. We hypothesized that if the pattern of episodic GH release in the acromegalics were modulated by increased GH and IGF-I production, increasing GH release in normals might also evoke elevations of SRIF and result in a

Received June 1,1989. Address requests for reprints to Dr. G. Merriam, DEB, NICH HD, Bldg. 10 Rm 10N262 Bethesda, MD 20892. Present address of Dr. Marie C. Gelato, Division of Endocrinology, State University of New York at Stony Brook, Stony Brook, New York 11794.

585

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 22 December 2016. at 17:07 For personal use only. No other uses without permission. . All rights reserved.

JCE & M • 1990 Vol 71 • No 3

GELATO ET AL.

586

similar GH pattern. In acromegalics, if SRIF is already elevated, one might find that GHRH would stimulate an increase in the amount of GH released but not further alter the character of episodic secretion.

Materials and Methods Synthetic GHRH(l-44) NH2 was purchased from Bachem, (Torrance, CA), and formulated for infusion as previously described (9). Subjects Six patients were studied. Their ages ranged from 22-41 yr. The diagnosis of acromegaly was based on both clinical and biochemical criteria (see Table 1). All patients had evidence of pituitary tumors on sellar CT scans. Five patients had received no prior therapy; one patient (patient 2), had had a transfrontal craniotomy with subtotal tumor secretion 1 yr before participating in the study. Measurements of serum levels of GHRH were made in all patients. The values were at or slightly above the detection limits of the assay (10); none of these patients had values falling into the range observed in patients with ectopic hypersecretion of GHRH. Six normal adult men, ages 21-34, were also studied. They were free of any known illnesses and were taking no medications.

jug/kg iv, at 0900 h. Three hours later a continuous infusion of GHRH was started, at a rate of 1 ^g/kg- h, via a Harvard pump. At the completion of the 24-h infusion, another 1 ng/kg GHRH bolus was administered. The study concluded 3 h after the last injection of GHRH. The normal subjects also underwent sampling during 24-h basal and GHRH infusion periods; in their case the two sample periods were separated by at least 2 weeks, and each 24-h sampling period was preceded and followed by the bolus IV injection of 1 Mg/kg GHRH. In the normals, basal and GHRH infusion studies were performed in random order. During sampling, patients and normal volunteers were kept at bedrest and asked to refrain from napping during the day. They were allowed free access to water and given three meals each day. Assays Serum GH was measured by a standard RIA technique (11). Intraassay and interassay coefficients of variation for the GH assay were 7.8% and 22%, respectively, at a mean measured value of 0.9 ng/mL, 7.8% and 12% at 5.1 ng/mL, and 9.8% and 10% at 20.3 ng/mL. All GH samples from each subject were measured in the same assay. IGF-I levels were measured by RIA on extracted serum at Endocrine Sciences (Tarzana, California) (12, 13). Statistical Analysis

Protocol The protocol was approved by the Clinical Research Subpanel of the National Institute of Child Health and Human Development. All subjects gave informed consent before participating in the study. An iv line was placed in an antecubital vein 2 h before beginning the study and kept open with a slow saline infusion for sampling. Blood samples for GH measurement were obtained at 20-min intervals throughout the study. In five of the six patients, GH levels were also drawn for a 24-h period before GHRH testing to establish the pattern of basal GH secretion. Twenty four-hour GH sampling was performed during GHRH infusion. All patients received a bolus injection of GHRH, 1

The pattern of GH secretion was characterized using two methods: the Pulsar program (14) and our modification of the method of Santen and Bardin (15). Mean GH levels and GH peak frequency and amplitude were characterized. In addition to the 24-h means, values were extracted for two 10-h time windows during the day (1200-2200 h) and night (2200-0800 h). Formal sleep recordings were not made. The Pulsar method had peak detection cutoffs set as Gi = 3.8, G2 = 2.6, G3 = 1.9, G4 = 1.5, G5 = 1.2, so that the expected false-positive rate is approximately 1%. The modified Santen and Bardin method accepted a rise as a peak when the nadir-to-peak rise exceeded four intraassay coefficients of variation of the GH assay. Only the results from the Pulsar program analysis are given in the

TABLE 1. Clinical and endocrine features of patients

Pt

Age

Sex

1) 2)

34 30

M M

3) 4) 5) 6)

41 23 27 22

F F F F

Sellar CT scan

Macroadenoma Macroadenoma w/suprasellar ext Microadenoma Macroadenoma Microadenoma Macroadenoma w/suprasellar ext

GH during

GH after

OGTT (mcg/L)4

TRH (mcg/L)'

IGF-I U/ml°

Ant pit function

7.5 3.5

nL Hypogonadal

70/32 14/7

67/58 8/8.7

3.7 7.3 6.1 6.4

nL Amenorrhea Amenorrhea Amenorrhea

23/27 99/102 15/12 90/194

27/30 95/1130 23/39 90/1000

0

Normal adult range 0.5-2.2 U/mL. * Shown as basal values and maxima after glucose. c Shown as basal and peak values.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 22 December 2016. at 17:07 For personal use only. No other uses without permission. . All rights reserved.

GHRH IN NORMAL SUBJECTS VERSUS ACROMEGALIC PATIENTS text. The patterns of GH release were qualitatively the same with the two programs but pulse frequency was slightly higher in both patients and the normal volunteers using the modified Santen and Bardin method. Significance of differences between groups was assessed with Student's paired or unpaired t test as appropriate.

Results The mean GH responses during saline and GHRH infusions for the group of six normal subjects are shown in Fig. 1. There was a significant increase in mean integrated GH during GHRH infusions (3.33 ± 0.5 vs. 10.2 ± 2.0 ng/mL, x ± SE, P < 0.01) (Table 2). After 24 h of GHRH infusions, the GH responses to a GHRH bolus were decreased (2.1 ± 0.9 ng/mL) as compared to the response seen either before the infusions (12.0 ± 3 ng/mL, P < 0.05) or that following 24 h of saline (16.8 ± 5 ng/mL, P < 0.05). GH release during the GHRH infusions was episodic (Fig. 1), as has also been noted by others (3,4). Compared to GH release during saline infusions, there was an increase in the number of GH pulses (Table 2). The additional GH release was not evenly distributed throughout the day; although there was wide variation among individuals, most of the increase was concentrated during nighttime hours, and the daytime rise at approximately 1500 h also appeared to be an enhancement of spontaneous release at that time. Thus there was the appearance of an overall increase in GH with preservation of the qualitative features of the basal 24-h release pattern. GH release varied widely in the acromegalic patients but was pulsatile in all subjects. Because of the great differences in basal GH levels in the different patients, a group average of GH levels during saline and GHRH infusions would have little meaning. Therefore, for purposes of combining results among all patients, each patient's GH values were expressed as a fraction of the mean basal GH value, and these fractional values were

FIG. 1. Mean ± 1 SD. Levels of GH during saline (left) and GHRH {right) infusions in six normal young adult men. Infusions were preceded and followed by bolus injections of GHRH (arrows).

587

then averaged (Fig. 2). As seen in the normal men, mean integrated GH release was significantly increased during GHRH infusions as compared to baseline (80.2 ± 20.3 us. 41.0 ± 12.1 ng/mL, x ± SE, P < 0.05). However, in contrast to the normals, there was no diminution of the peak GH responses to a bolus of GHRH after vs. before the GHRH infusion (30.2 ± 8.9 vs. 35.5 ± 12.5 ng/mL, NS). GH release was episodic in the acromegalics during both saline and GHRH infusions. In contrast to normal men, the number of pulses in the patients did not increase during the GHRH infusions (Table 2) and the patients showed no diurnal variation in GH secretion either basally or during administration of GHRH (Table 2). The mean GH and the amplitude of the pulses in the patients were significantly increased (P < 0.05) during GHRH infusion (Table 2). In some patients, there was a tendency for the GH response to constant GHRH infusion to drift downward after an initial rise; however, other patients showed unremittingly increased pulsatile GH secretion throughout the infusions (data not shown). Even in those patients in whom a diminution of the response was seen, GH levels remained higher at the end than at the beginning of the infusion period.

Discussion These studies demonstrate that GH secretion is episodic under basal conditions and during infusions of GHRH in both acromegalic patients and normal subjects. Apart from the elevated levels of GH in the acromegalics, the major qualitative difference in the pattern observed is the presence of a consistent diurnal variation in GH in the normals which is absent in the acromegalics. The frequency of episodic GH release is higher in the acromegalics than in the normals. During GHRH infusion, this frequency increases in the normals but remains the same in the acromegalics. There are several possible alternative physiological interpretations to these results;

20 -

10 -

900 1200 15001800 2100 2400 300 600 900 1200 1500

900 1200 1500 18002100 2400 300 600 900 1200 1500

CLOCK TIME

CLOCK TIME

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 22 December 2016. at 17:07 For personal use only. No other uses without permission. . All rights reserved.

GELATO ET AL.

588

JCE & M • 1990 Vol 71 • No 3

TABLE 2. GH secretion during saline and GHRH infusions (means ± SE)

Normal subjects A. Saline (n = 6) 24 h Day (1200-2200 h) Night (2200-0800 h) B. GHRH (n = 6) 24 h Day (1200-2200 h) Night(2200-0800 h) Acromegalic patients A. Saline (n = 5) 24 h Day

Night B. GHRH (n = 6) 24 h Day Night

Mean GH (mcg/L)

No. of GH peaks (n/time period)

GH peak amplitude (mcg/L)

Area under curve (mcg-min/L)

3.3 ± 0.5 1.5 ± 0.4 4.2 ± 0.8*

2.2 ± 0.3 0.7 ± 0.3 1.5 ± 0.26

14.3 ± 3.6

5,976 ± 979 896 ± 234° 2,545 ± 466

10.2 ± 2.0c 8.8 ± 2.8 16.9 ± 3.7

11.7 ± 1.4C 5.0 ± 0.9" 4.3 ± 0.6

15.8 ± 4.5

18,366 ± 3,663 5,309 ± 1,692° 10,237 ± 2,197

41.0 ± 12.1 45.3 ± 13.7 39.8 ± 12.2

11.8 ± 0.8 5.0 ± 8.6 5.2 ± 0.4

80.2 ± 20.3d 77.2 ± 17.3 93.1 ± 25.8e

9.6 ± 0.8 4.2 ± 0.8 3.2 ± 0.3

15.39 ± 5

58,239 ± 1,701 27,208 ± 824 28,300 ± 1,046 114,002 ± 29,149" 46,290 ± 10,382 55,896 ± 15,487

32.3 ± 9"

0

P < 0.01 day us. night. P < 0.05 day us. night. c P < 0.01 saline us. GHRH. " P < 0.05 saline us. GHRH. e NS Day us. night; saline us.GHRH. 0

Pulsatile growth hormone secretion in patients with acromegaly and normal men: the effects of growth hormone-releasing hormone infusion.

Twenty-four GH secretory patterns were studied before and during continuous infusions of GHRH in six patients with active acromegaly and in six normal...
751KB Sizes 0 Downloads 0 Views