| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Clinical Studies |
Department of Medicine (L.S.), Radiology (L.L.) and Research Center for Endocrinology and Metabolism (K.S., J.-O.J., B.-Å.B.), Sahlgrenska University Hospital, S-413 45 Göteborg, Sweden
Address all correspondence and requests for reprints to: Lars Sjöström, Department of Medicine, Sahlgrenska University Hospital, Vita straket 15, S-413 45 Göteborg, Sweden.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
EE was determined in a chamber for indirect calorimetry. Body composition was determined with dual-energy x-ray absorptiometry and computed tomography using a four-scan technique. Blood samples were examined using well-established RIAs.
During the first 2 weeks, 24-h EE increased by 6 ± 3% (range 1–8%) from 40.9 ± 4.8 to 42.9 ± 4.8 kcal/24 h·kg (P < 0.05), sleeping metabolic rate by 14 ± 3% (range 10–18%) from 28.4 ± 1.9 to 32.9 ± 2.2 kcal/24h·kg (P < 0.001), and basal metabolic rate by 11 ± 7% (range 0–18%) from 29.6 ± 2.4 to 33.3 ± 2.6 kcal/24h·kg (P < 0.05). No change was found in daytime EE. The increase in EE covaried with changes in insulin-like growth factor 1, the free T3/free T4 ratio, insulin-like growth factor-binding protein-3, and the aminoterminal procollagen III peptide but not with changes in body composition.
It is suggested that the stimulating effect of GH on EE occurs gradually during a 2-week period and is only detectable during night and morning hours, when significant levels of GH occur.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We recently have shown that GH-deficient adults (GHD) have a basal metabolic rate (BMR) in the lower part of the normal range but not significantly different from healthy controls (5). This was evident when the absolute BMR value was used or when BMR was expressed per kg of fat-free mass (FFM) or adjusted for age, body fat (BF), and FFM.
It is known that BMR and BMR/FFM increases during treatment with GH (5, 6). It has been suggested that the increase in BMR is mediated via changes in body composition and increases in protein synthesis and/or the deiodination of T4 to T3 (see Ref.23). However, it is still unclear whether GH also has a direct effect on energy expenditure (EE). Furthermore, the within- and in-between day variability in the EE response to GH are largely unknown; and 24-h EE (24hEE), sleeping metabolic rate (SMR), BMR, and daytime EE (DayEE) have never been examined in a chamber for indirect calorimetry.
The aims of the present study were to evaluate EE in GHD further and to examine differences in EE within and in-between days during the first 4 weeks of GH treatment in GH-deficient subjects. We also aimed to study time-dependent changes in EE in relation to changes in body composition and metabolic markers such as insulin-like growth factor 1 (IGF-I), thyroid-related hormones, and the aminoterminal procollagen III peptide (PIIIp).
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The study was designed as an open study. Before the start of treatment, all the subjects were examined on 2 consecutive days. During GH treatment, all the subjects were examined on days 1, 2, 5, 8, and 15 and after 1 month in the chamber. (On day 5, chamber results were not available for subject no. 4 because of acute gastro-enteritis.) The subjects were not allowed to leave the chamber between the 2 baseline examination days or between the 1st and 2nd examination day. The mean of the two baseline values was used as a pretreatment value.
Study subjects
Four men and one woman, 23–67 yr old, were examined. All the subjects had been GH-deficient for more than 1 yr and were included in the study according to the following criteria: 1) a maximal GH concentration peak of less than 5 mU/L after an iv insulin tolerance test (0.1 IU insulin/kg BW) (all the patients displayed symptoms of hypoglycemia and had a blood glucose level less than 2.2 mmol/L); 2) the patients had been on stable replacement therapy with corticosteroids, L-T4, and gonadal steroids for at least 1 yr before the study.
The individual characteristics of the five patients are shown in Table 1
. Three male patients had been treated with irradiation
after pituitary surgery. One male patient had hypopituitarism caused by
an apoplexy in a pituitary adenoma. One woman had had idiopathic
hypopituitarism since childhood. Two of the subjects included in the
study had received treatment with GH during childhood. None of the
patients was a current smoker. One patient was being treated for
hypertension with a calcium blocker (Isradipin).
|
Treatment
The dosage of GH was 0.25 U/kg BW·week (12 µg/kg BW·day), administered as a daily evening sc injection by the patient (Pharmacia-Upjohn, Uppsala, Sweden). The first injection was given on day 1. Between the injections, the vials were stored for a maximum of 7 days at 5–12 C, protected from the light. No reduction in the dose because of side-effects was necessary during the study period.
Basal medication
The patients were on stable medication with glucocorticoids
(cortisone acetate 12.5 mg twice a day), L-T4 (0.10–0.20
mg/day), and sex hormones. Two subjects were being treated with ADH
(Desmopressin) (Table 1
). The dose of cortisone acetate was not changed
during the GH study, except in the case of intercurrent illnesses.
Calorimetric methods
EE was determined by indirect calorimetry in a 30-m3 metabolic chamber, as previously described (7). The mathematical procedures followed those described by Brown et al. (8). The protein oxidation was calculated from urinary nitrogen excretion. For repeated changes, the time resolution of the chamber was approximately 15 min, as shown by intermittent infusions of a nitrogen/CO2 mixture (80:20). The oxygen and CO2 analyzers were calibrated before each experiment by blending fresh air with N2 or CO2 using a gas blender (Environics Inc, Willington, CT). The overall response of the chamber was regularly validated by butane combustion experiments. The system was shown to be accurate within ±1.2% in 95% of the tests (7). The precision error calculated from double determinations in the same subjects on consecutive days before GH treatment was 4.1% for 24hEE, 4.6% for BMR, and 3.0% for SMR. The study participants entered the chamber at 1000 h. During the day, subjects were encouraged to be as active as they wished, but no vigorous exercise was allowed. The subjects were instructed to go to bed at 2200 h and were awakened at 0630 h. The subjects fasted from 2200 h until after the BMR measurement in the morning. Participants were removed from the chamber at 1100 h. During the chamber experiment, all the subjects filled in an activity diary. All the subjects received dietary instructions before starting the study and were instructed not to change their dietary intake over the study period. During the chamber examinations, an isocaloric diet was served.
EE measurements. EE data were obtained for the following periods:
Total 24hEE. EE during the total 24-h stay in the chamber.
SMR. EE between 0100 and 0600 h, excluding all periods of awakening related to toileting, for example.
BMR. EE measured between 0700 and 0740 h, with the subject resting on the bed in the chamber after an overnight fast.
DayEE. EE between 1300 and 1800 h.
Single injection of GH. The effects of one sc injection of GH were studied during the first day of treatment. EE measured during the hour preceding the first GH injection was compared with the EE measured during each consecutive hour for 5 h.
Measures of dispersion. The root of the mean squared successive differences (RMSSD) and the SD were calculated for the above described periods. The RMSSD has previously been used primarily in studies of blood pressure and has been shown to be a sensitive and almost pure measure of short-term variability (9). It also has been shown to be resistant to slow rhythms and linear trends, and it was therefore used in this study as a complement to SD to characterize the shape of the EE curve.
Predicted BMR. An estimation of BMR from weight, height, age, and sex was made according to standard equations described in Ref. 10.
Body composition
All measurements of body composition were made before each chamber experiment on days -3, -2, 2, 8, 15, and after 30 days.
Body weight was measured with the patients dressed in underwear and was determined to the nearest 0.1 kg on a calibrated scale. Body height was measured without shoes, in the standing position, to the nearest 0.1 m. Body mass index (BMI) was calculated as BW divided by height2.
Dual-energy x-ray absorptiometry. All the subjects were examined with PDX-L equipment produced by Lunar Corp., Madison, WI. Determinations of total body, BF, lean weight, and bone mineral content (BMC) were determined. FFM was calculated as lean weight plus BMC. The calculated error of a double determination in the same subject was 0.95% for BMC, 1.6% for lean weight, and 1.5% for BF.
Computed tomography. One longitudinal and four transsectional scans were obtained. All the subjects were examined at baseline and after 4 weeks on GH. The transsectional scans were obtained at L3, the iliac crest, thigh, and calf. Visceral and sc AT and muscle areas were determined by defining AT as tissues in the attenuation interval -190 to -30 HU, as previously described (11). Muscle and sc areas were determined as the mean of all scans. Visceral adipose area was determined as the mean of the L3 and the iliac crest scan.
Biochemical methods
GH concentrations in serum were determined using a polyclonal immunoradiometric assay method (Pharmacia), and IGF-I was determined by an RIA (Nichols Institute, Wijehen, the Netherlands). IGF-binding protein-3 (IGFBP3) was determined using an RIA (Nichols Institute). Serum-free T4(free T4) and free T3 were analyzed using ligand analog RIAs (Amerlex M, Amersham International plc, Amersham, Bucks, UK), serum total T4 using a single-antibody RIA (Farmos Diagnostica, Turku, Finland), total T3 using a polyethylene-glycol assisted double-antibody RIA (Diagnostic Products Corp, Los Angeles, CA), reverse T3 was determined by an RIA (Ares Serono, Milano, Italy), and PIIIp using an immunoradiometric assay (Hoechst-Behringwerke, Marburg, Germany). Blood samples were drawn in the morning of days -3, -2, 2, 5, 8, 15, and day 30 after an overnight fast. All the samples from each patient were run in the same assay and in complete blocks.
Statistical methods
Statistical calculations were carried out on a Vax 4100 computer using the Minitab statistical program, version 9. Comparisons between baseline (the mean value for the two baseline examinations) and consecutive examinations were made using a paired t test. Descriptive statistics and Spearman’s rank correlation coefficients were used at baseline.
In addition to the baseline observations, the study included repeated
simultaneous measurements of EE, body composition, and biochemical
variables on six different occasions with time intervals according to
Fig. 1. A two-stage method was used for the
time-dependent statistical analysis. First, the repeated measurement
from each patient was reduced to one summary variable reflecting the
time-dependent association between two variables. Second, the summary
variable (Spearman’s rank correlation coefficient) was analyzed in a
univariate fashion (12) by paired t tests. A
P-value less than 0.05 was considered statistically
significant.
|
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
At baseline, no significant difference was found between BMR measured in the chamber and the calculated reference BMR (BMR-predicted), adjusted for sex, age, body weight, and height [1610 ± 410 vs. 1700 ± 410 kcal/24 h; NS (not significant)]. 24hEE, SMR, and BMR were all correlated to BW and FFM (not shown). 24hEE and SMR also tended to be correlated to BF (not shown). All the subjects had a BMR/SMR ratio above 1.0.
Effects of treatment
EE. The treatment with GH increased 24hEE, SMR, and BMR in all
subjects. No alteration in DayEE was found over the study period (Table 2). The maximum increase in 24hEE, SMR, and BMR was seen
within the first 2 weeks of treatment (Fig. 1). Thereafter, no further
significant change could be detected. SMR and BMR increased more
markedly than 24hEE (Fig. 2). During the first 2 weeks
of treatment, 24hEE increased by a mean of 6 ± 3% (range 1–8%)
from 40.9 ± 4.8 to 42.9 ± 4.8 kcal/24 h·kg
(P < 0.05), SMR by 14 ± 3% (range 10–18%)
from 28.4 ± 1.9 to 32.9 ± 2.2 kcal/24h·kg
(P < 0.001), and BMR by 11 ± 7% (range 0–18%)
from 29.6 ± 2.4 to 33.3 ± 2.6 kcal/24h·kg
(P < 0.05). (Table 2). The lipid oxidation had
increased by 61 ± 36 mg/min (136%) at day 15 (P
= 0.02) (Table 2).
|
|
No significant effect from the first single injection of GH on EE or RQ was found (not shown).
Body composition. BW increased by a mean of 1.5 ± 0.6 kg after 2 weeks and was then virtually unchanged at 4 weeks (77.5 ± 22.1 vs. 78.6 ± 21.8 kg; NS). At that time, FFM displayed a tendency to increase by a mean of 2.0 ± 1.8 kg (54.3 ± 12.7 vs. 56.3 ± 12.9 kg; P = 0.072). A 4% decrease in dual-energy x-ray absorptiometry-measured BF did not reach significance (22.8 ± 11.4 vs. 21.9 ± 12.4 kg; NS). On average, the muscle area increased by 5% (198 ± 43 vs. 208 ± 46 cm2; P < 0.05), and visceral and sc AT areas (determined by computed tomography) decreased by 9.9 (142 ± 54 vs. 128 ± 50 cm2; NS) and 7.5 (160 ± 72 vs. 148 ± 69 cm2; P < 0.05) percent, respectively, over 4 weeks of treatment with GH.
Serum IGF-I, IGFBP-3, and PIIIp. IGF-I increased from a
mean of 79 ± 61 µg/L at pretreatment to a mean of 220 ±
130 µg/L within the first 2 days of treatment. A similar pronounced
increase also was found in IGFBP-3, which increased from 1.6 ±
1.0 mg/L at baseline to 2.4 ± 1.3 mg/L during the first 2 days. A
marked, but more gradual, increase was found in PIIIp during the first
2 weeks of treatment (Fig. 1 and Table 3).
|
). The ratio of free T3/free
T4 increased markedly during the first week of treatment,
and the maximum increase was found within the first 2 weeks of
treatment (Fig. 1). After 2 weeks, there was a tendency towards a
transient decrease in reverse T3 (Table 3). Heart rate and
blood pressure were unchanged over the study period (not shown). Correlation analysis. The correlation coefficient between delta values for changes in EE and delta values for changes in FFM, BF, FT3/FT4, PIIIp, IGF-I, and IGFBP-3 were calculated separately for each examined subject. The individual correlation coefficients were then averaged, and the possibility of a covariation was evaluated by paired t tests. Delta SMR covaried significantly with delta FT3/FT4 (r = 0.72 ± 0.17; P < 0.001), delta IGF-I (r = 0.70 ± 0.27; P < 0.01), delta PIIIp (r = 0.65 ± 0.17; P = 0.011), and delta IGFBP-3 (r = 0.65 ± 0.41; P < 0.01) but not with delta FFM (r = 0.40 ± 0.34; NS) and delta BF (r = -0.08 ± 0.71; NS). Similar covariations also were found between delta SMR adjusted for FFM and delta FT3/FT4, delta IGF-I, delta PIIIp, and delta IGFBP-3 (not shown).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In studies of subjects with GH deficiency, EE has been reported to be normal or low-normal (13, 5, 14), as well as increased (after adjustment for FFM) (15). In the present study, BMR was not significantly lower than the predicted value. In our previous report, subjects with GHD had a mean BMR that was in the low-normal range if standard prediction equations were used. Furthermore, BMR and BMR/FFM were unchanged, compared with controls matched for sex, age, and BMI (5). Similar observations were made by Hoffman et al. (13). Using doubly-labeled water (DLW), Chong et al. (14) found similar total EE in GH-deficient subjects as compared with normal subjects. In contrast, Salomon et al. (see discussion in Ref.5) observed an increase in BMR/FFM in GHD subjects, possibly caused by a poorly selected control group.
The metabolic stimulatory potential of GH has been established in several studies (1, 5, 15). GH treatment in normal, obese, and GHD has been found to increase BMR and to increase lipid oxidation (5), resulting in a decrease in BF and an increase in muscle and visceral organ weight (1). Using the DLW technique, Chong et al. (14) found that 2 weeks of GH treatment increased total EE by approximately 13%. We found a slightly lower increase in 24hEE (8%). The difference may be related to a larger degree of restriction in spontaneous physical activity in the chamber, but also to much larger errors in the DLW technique (16), compared with the chamber technique (7).
The first injection of GH failed to produce any significant effect on EE. In contrast, short-term studies have reported an acute effect by GH (17) on EE, whereas others have failed to show any such effect (18, 19). Our study demonstrated a detectable increase in EE after 2 days of treatment. The response time may, however, be dose-dependent, considering that larger doses have been shown to increase EE more rapidly (20).
The increase in EE seemed to occur progressively over a 2-week period, and no further increase was then observed.
We used a protocol involving repeated examination during the first 4 weeks, and the relationship between changes in EE and changes in body composition and certain metabolic variables was examined using a two-step procedure (12). The increase in EE covaried with the free T3/free T4 ratio, IGF-I, IGFBP-3, and PIIIp, but no covariation was found with changes in body composition. These results support previous findings that the effect of GH on energy metabolism is not solely explained by changes in body composition (4).
IGF-I has been shown to increase EE in the same way as GH (21). IGF-I also has an additive effect on EE in addition to the effect of GH (22). It is, however, unclear whether IGF-I has a direct effect on EE. It has been suggested that the effect of IGF-I on EE is mediated by secondary metabolic changes (21).
Like others, we observed an increase in free and total T3 concentrations and a tendency towards a decrease in free T4, suggesting the increased deiodination of T4 to T3 during GH treatment (6). The maximum response was found after 2 weeks, and 50% of the maximum response was detectable after 5 days. GH treatment, therefore, seems to have an acute effect on thyroid function. Studies using larger doses have shown effects even after 4 days (23), indicating a dose-response effect. However, the increase in T3 concentrations probably does not explain fully the calorigenic effect of GH. Experimental studies in normals and obese subjects have shown that the addition of GH to T3 treatment further increased EE (24, 25). GH thus seems to mediate its effects on EE through several different mechanisms.
A more marked increase was found in SMR and BMR than in 24hEE, and no change was found in DayEE. These findings suggest that the calorigenic effect of GH varies during the day, with a more pronounced increase during the night and morning hours, possibly because of higher serum GH concentrations after the evening injection (26). This could suggest a direct effect by GH on EE.
The predominant consumer of energy in the cell is the Na+, K+ATPas, and it has been suggested that GH alters EE through changes in the activity of the Na+, K+ pump (either at the expression or posttranslational level) (27). We therefore hypothesize that the observed predominately nightly increase in EE might be mediated through GH-induced alterations in Na+, K+ ATPas activity.
To summarize, we have examined 24hEE changes during the first 4 weeks of treatment with GH. We found that GH increases 24hEE, BMR, and SMR (but not DayEE). The main increase in EE seemed to occur gradually over the first 2 weeks of treatment and covaried with changes in several metabolic factors, but no relationship was found between changes in EE and changes in body composition.
|
Acknowledgments |
|---|
rd and the statistical advice of Anders
Odén. |
Footnotes |
|---|
Received August 26, 1996.
Revised December 6, 1996.
Accepted December 16, 1996.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  H. Mersebach, O. L. Svendsen, A. Astrup, and U. Feldt-Rasmussen Abnormal Sympathoadrenal Activity, but Normal Energy Expenditure in Hypopituitarism J. Clin. Endocrinol. Metab., December 1, 2003; 88(12): 5689 - 5695. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Gibney, M. L. Healy, M. Stolinski, S. B. Bowes, C. Pentecost, L. Breen, C. McMillan, D. L. Russell-Jones, P. H. Sonksen, and A. M. Umpleby Effect of Growth Hormone (GH) on Glycerol and Free Fatty Acid Metabolism during Exhaustive Exercise in GH-Deficient Adults J. Clin. Endocrinol. Metab., April 1, 2003; 88(4): 1792 - 1797. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. N. Roemmich, P. A. Clark, V. Mai, S. S. Berr, A. Weltman, J. D. Veldhuis, and A. D. Rogol Alterations in Growth and Body Composition During Puberty: III. Influence of Maturation, Gender, Body Composition, Fat Distribution, Aerobic Fitness, and Energy Expenditure on Nocturnal Growth Hormone Release J. Clin. Endocrinol. Metab., May 1, 1998; 83(5): 1440 - 1447. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
, P < 0.10; *,
P < 0.05; **, P < 0.01, ***,
P < 0.001.