| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Departments of Endocrinology (J.C.M., H.B., L.S., E.A.V.) and Cardiology (O.K.), University Hospital Vrije Universiteit, Amsterdam, The Netherlands
Address all correspondence and requests for reprints to: Dr. Jan C. ter Maaten, Department of Internal Medicine, University Hospital Groningen, P.O. Box 30.001, 9700 RB Groningen, The Netherlands.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
A large number of short term studies have reported beneficial effects of GH replacement therapy. It is now well established that short term GH treatment restores abnormalities in body composition and metabolism and improves physical performance, cognitive function, and general well-being (1, 3). To date, the long term effects of GH replacement therapy are not exactly known. The main questions are whether the beneficial effects of GH substitution are maintained over the long term and whether long term treatment is safe. The latter is of great importance because of the potential risks of chronic GH excess. Observations in patients with acromegaly indicate that GH excess may cause hypertension, cardiac hypertrophy, congestive heart failure, and increased mortality from cardiovascular and malignant diseases (4, 5). In the present study, we report the long term effects of GH replacement therapy on body composition, cardiac function, and bone mineral density (BMD) for periods of 39–69 months.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Fifty adult men with childhood-onset GHD were initially included in the study, as described previously (6, 7, 8). Previous GH treatment for short stature had been discontinued for at least 1 yr (7.5 ± 4.5 yr). GHD was confirmed before the study by a serum insulin-like growth factor I (IGF-I) concentration of at least 2 SD below the age-related normal mean and a maximal GH response to 100 µg/kg GHRH or insulin-induced hypoglycemia of less than 7 µg/L in all subjects. Twelve patients were excluded from analysis because of incomplete follow-up. Two of them showed poor compliance, 8 withdrew due to lack of motivation, 1 dropped out because of a diagnosis of Crohn’s disease, and 1 died due to generalized convulsive seizures despite antiepileptic treatment. The mean age of the remaining 38 patients was 28.0 ± 4.0 yr (range, 20–35). Twelve patients had isolated GHD. The 26 patients with multiple pituitary hormone deficiencies received stable, conventional hormone replacement: T4 (n = 24), testosterone undecanoate (n = 20), hydrocortisone (n = 19), antidiuretic hormone (n = 4), and CG (n = 2). The cause of pituitary failure was idiopathic or related to perinatal hypoxia in 33 patients and was related to treatment for craniopharyngioma in 5 patients. Informed consent was obtained from all subjects, and the protocol had been approved by the local ethics committee.
GH replacement therapy
The study protocol dictated that during the first 2 yr the patients were to be randomized to receive GH substitution (Norditropin, Novo Nordisk, Gentofte, Denmark) in a dose of 1 (n = 10), 2 (n = 18), or 3 (n = 10) IU/m2·day (9, 18, and 27 µg/kg), respectively. After the second year, the dose was changed to 2 IU/m2·day in all patients as this was believed to be the physiological replacement dose at that time. During the first year, doses were reduced by one third in the case of clinically relevant side-effects. Thereafter, the GH dose was titrated based on serum IGF-I levels. The study was terminated on December 31, 1996.
Study parameters
Anthropometric measurements included measurement of weight and skinfold thicknesses. Skinfold thickness measurements (Harpenden skinfold caliper) were performed on the right side of the body at seven different sites, according to standard procedures (8, 9). Computed tomographic scanning with the Siemens Somatom was used to evaluate changes in muscle mass of the upper leg (measured midway between the superior borders of great trochanter and patella) and intraabdominal fat (measured at the midlevel of the fourth lumbar vertebra). Calculation of intraabdominal fat area was performed with the software program provided by the manufacturer, using standard procedures (8, 10). Observed changes were evaluated in relation to expected changes using cross-sectional data in age-matched normal subjects (8).
The BMD at the lumbar spine, femoral neck, and trochanter and the total bone mineral content were measured by dual energy x-ray absorptiometry (Norland XR-26), according to standardized procedures (11). The long term precision of this method is 2.4% for the lumbar spine and 2.3% for the femoral neck. BMD was calculated by the software program and is presented as the areal density expressed in grams per cm2 or as the difference in SDs from the mean of age- and sex-matched healthy subjects (z-score). Because most of the patients were short, and BMD correlates with body height, we also calculated the height-corrected z-scores (12).
Transthoracic echo cardiographic measurements were performed with a Hewlett-Packard Co. (Sono 1500, Palo Alto, CA) machine and 2.5- or 3.5-mHz transducers, according to the recommendations of the American Society of Echo Cardiography (13). M-mode echo cardiography was used to measure left atrial and ventricular dimensions and left ventricular wall thickness, and thus left ventricular mass was calculated. Stroke volume and cardiac output were measured using pulsed Doppler echo cardiography at the left ventricular outflow tract using quantitative two-dimensional echo cardiography (Simpson’s rule). Left ventricular volumes were calculated from apical four- and two-chamber views. Stroke volume and cardiac output were adjusted for body surface area and expressed as indexes. All echo cardiographic data were evaluated at the end of the study by one examiner.
Exercise capacity was measured by bicycle ergometry (Medgraphics, St. Paul, MN). The patients started cycling at a workload of 40 watts for 3 min. Subsequently, the load was increased by 20 watts every minute until exhaustion. Exercise capacity was recorded as the maximal workload (watts) and maximal oxygen uptake (milliliters per min) achieved during the test. As exercise capacity declines with age and the study lasted for several years, we evaluated the maximal workload and oxygen consumption in relation to the expected normal values using standard formulas (14, 15).
Follow-up protocol
The patients visited our out-patient clinic every 3 months. Measurements of body composition and bone mineral density are reported at baseline and after each year of replacement therapy, whereas BMD is also reported after 6 months of treatment. Cardiac function was evaluated at baseline; after 1, 2, and 3 yr of treatment; and, in the patients who had been treated for at least 4 yr, at the end of the study.
Analytical methods
The serum IGF-I concentration was measured by RIA (Medgenix Diagnostics, Fleurus, Belgium) after acid-ethanol extraction of IGF-binding proteins. Intra- and interassay coefficients of variation were 6% and 10%. Free T4 and free T3 concentrations were measured with commercially available RIAs [Becton Dickinson and Co. (New York, NY) and Amerlite, Amersham (Aylesbury, UK), respectively].
Statistical analysis
Data are expressed as the mean ± SD unless otherwise stated. Changes in body composition, BMD, and echo cardiographic parameters are expressed as the percent increase or decrease from baseline. All variables were analyzed by ANOVA for repeated measurements to detect differences over time followed by paired t tests. All statistical tests were two sided. Correlation analysis was applied when appropriate. P < 0.05 was considered significant.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The mean follow-up period for the study group of 38 patients was 55 months (range, 39–69) by the end of the study. Seventeen patients were followed for a minimum of 60 months, and 32 patients were followed for at least 48 months.
GH dose and hormonal changes
During the first 2 yr of the study, GH doses were reduced in
21 patients. The mean GH doses after 12 and 24 months of treatment were
1.64 ± 0.66 and 1.51 ± 0.69 IU/m2·day (15 and
13 µg/kg), respectively. The predetermined dose of 2
IU/m2·day (18 µg/kg) that was instituted after the
second year had to be reduced in 32 patients: in 14 patients because of
side-effects, and in 18 patients because of supraphysiological serum
IGF-I levels. After 36, 48, and 60 months, the mean GH doses were
1.52 ± 0.39, 1.38 ± 0.40, and 1.30 ± 0.38
IU/m2·day (13, 12, and 11 µg/kg), respectively. During
the first 3 yr, serum IGF-I levels were in the upper normal range or
borderline supraphysiological (Table 1
).
Initially, free T4 levels decreased and free T3
levels increased, but both levels stabilized during prolonged GH
replacement (Table 1).
|
Baseline values of study parameters are given in Table 2. As shown in Fig. 1, mean body weight and leg muscle area
progressively increased during the study and exceeded baseline values
by 17.2% and 28.7% after 5 yr (Fig. 1). In contrast, maximal
decreases in the sum of skinfold thicknesses and intraabdominal fat
area (30.9% and 46.0%, respectively) were reached in the first
year.
|
|
After 4 and 5 yr, measurements were able to be evaluated in
only 27 and 13 subjects, respectively, because the dual energy x-ray
absorptiometer lost precision at the end of the study. At baseline, BMD
was subnormal at the lumbar spine (z-score, -1.7 ± 1.2), femoral
neck (z-score, -1.2 ± 1.1), and trochanter (z-score, -0.9
± 1.0). BMD at the lumbar spine and total bone mineral content
initially declined during GH replacement (Fig. 2). After 5 yr, BMD exceeded baseline
values by 9.6 ± 8.0% at the lumbar spine, 11.1 ± 12.8% at
the femoral neck, and 16.2 ± 9.4% at the trochanter
(P < 0.001 for all comparisons). Total bone mineral
content exceeded baseline values by 13.8 ± 7.9% after 5 yr. The
final z-scores were -0.6 ± 1.5 in the lumbar spine, -0.2
± 1.3 in the femoral neck, and -0.3 ± 1.0 in the trochanter.
The height-corrected z-scores increased from -0.66 ± 1.08 to
0.23 ± 1.35 in the lumbar spine and from -0.42 ± 1.14 to
0.43 ± 1.31 in the femoral neck. Correlation analysis between the
BMD z-scores at baseline and GH-induced changes in BMD after 4 yr
showed significant inverse correlations for the measurements obtained
at the femoral neck (r = -0.58; P = 0.002) and
the trochanter (r = -0.62; P < 0.001), but not
at the lumbar spine (r = -0.23; P = NS).
|
Two patients started antihypertensive treatment during the third year because of rising blood pressure. In the other 36 patients, mean supine blood pressure did not differ at the end of the study (113.1 ± 11.6/58.8 ± 7.8 mm Hg) from baseline (113.6 ± 9.0/60.6 ± 7.1 mm Hg). Heart rate increased from 64.0 ± 10.9 to 68.9 ± 10.3 beats/min after 1 yr (P = 0.007) and to 71.6 ± 8.5 beats/min at the end of the study (P < 0.001).
Echo cardiographic data were able to be evaluated in 37 subjects during
the first 3 yr and in 30 subjects at the end of the study. During the
study, left atrial end-systolic diameter gradually increased from
31.4 ± 3.9 to 35.0 ± 3.6 mm (P < 0.001),
and left ventricular end-diastolic diameter increased from 47.7 ±
4.2 to 50.8 ± 4.6 mm (P = 0.006). The
interventricular septum wall thickness and left ventricular mass
increased during the first year, but returned to baseline thereafter
(Fig. 3). During the study, left
ventricular end-diastolic and end-systolic volumes increased by 15.1%
(P < 0.001) and 18.2% (P = 0.003),
respectively. The ejection fraction did not change, but stroke volume
and cardiac output increased by 16.3% (P = 0.002) and
33.4% (P < 0.001), respectively. The stroke volume
index exceeded baseline values by 8.5% (P = 0.02)
after 1 yr and by 5.0% (P = 0.19) at the end of the
study, whereas cardiac index exceeded baseline values by 18.1%
(P = 0.001) after 1 yr and by 20.5% (P
< 0.001) at the end of the study.
|
).
|
The majority of side-effects occurred during the first months of GH replacement and included muscle or joint stiffness, swelling of hands/feet, manifest edema, paresthesia, gynecomastia, and thirst, as reported previously (7). Similar dose-dependent side-effects occurred after the second year when the dose was increased to 2 IU/m2·day as described by the protocol: paresthesia (n = 7), muscle or joint stiffness (n = 4), swelling of hands or feet (n = 2), arthralgia (n = 2), and hyperhydrosis (n = 2). One patient died after 49 months of GH replacement therapy due to generalized convulsive seizures. He had preexistent epilepsy. IGF-I levels in this patient were within the normal range during the preceding period.
Subgroup analysis
The patients who prematurely discontinued GH replacement
had a higher body weight, sc fat mass, and total bone mineral content
at baseline than the patients who completed the study (Table 2
). The
remaining variables did not differ between the two groups. The observed
differences were caused by three extremely obese subjects in the first
group (body mass indexes, 36.2, 38.7, and 39.8 kg/m2,
respectively).
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The observation that long term GH replacement improves cardiac performance without inducing cardiac hypertrophy is important, because GH increased cardiac mass in previous studies of shorter duration (16, 17, 18), albeit not all (19, 20, 21). GH-induced increases in cardiac mass have been found during both supraphysiological (18) and physiological (17) GH replacement and most likely reflect increases in the size of cardiac myocytes (22). The present study indicates that somewhat supraphysiological GH does not increase left ventricular mass over the long term. GH induced a significant increase in stroke volume during the first year and a more steady increase thereafter in proportion to the increases in body weight. The increases in stroke volume can be attributed to fluid retention and an increased preload (the Starling effect), as indicated by the rise in left ventricular end-diastolic volume, or to direct inotropic properties of GH (17). In contrast to stroke volume index, cardiac index increased significantly over the long term because heart rate increased at the same time. GH-induced increases in heart rate are well known, but presently unexplained (4, 16, 17, 18, 19).
The improvement in cardiac performance corresponded to a marked improvement in exercise performance. Increases in skeletal muscle mass and strength could also account for the improved exercise performance (23). However, GH may have improved exercise performance indirectly by a higher degree of physical activity or training over the course of the study.
The increase in skeletal muscle mass is mainly responsible for the gradual increase in body weight during our study. The cross-sectional muscle area increased by approximately 28% after 5 yr, whereas during short term treatment periods of 6 months, increases in muscle mass of only 5–8% have been reported (23, 24, 25). In contrast to the steady increase in muscle mass, the changes in fat mass showed a different time pattern. We found a considerable decline in fat mass during the first year of GH substitution, especially of the truncal or visceral region, which corresponds with previous studies (25). Visceral adipose tissue was partially regained later in the course of the study, reflecting the predicted age-related changes in intraabdominal fat (8).
It is now well established that GH treatment will temporarily decrease BMD (26), most likely due to an increase in the remodeling space (27). The subsequent increase in bone mass during our study compares favorably to changes in bone mass that have been observed during shorter periods of GH replacement (28). The favorable effects of GH on bone mass in our study can be explained by previous observations that GH increases bone mass more in subjects with childhood-onset GHD than in subjects with adult-onset GHD (28). GH-induced increases in bone mass were more marked in subjects with lower pretreatment z-scores, in analogy to previous observations in adult-onset GHD (29).
A major limitation of the present study is the lack of control subjects, but it is hardly feasible to perform a placebo-controlled study for a 5-yr period. As we studied young male adults with childhood-onset GHD, some of the beneficial effects of GH may be less in other GH-deficient subjects. It has been shown that GH-deficient men are more responsive to the effects of GH on body fat than are women (30). Also, GH may increase lean body mass more at a younger age (31). However, changes in body fat and lean body mass during a 1-yr substitution period did not differ between subjects with childhood- and adult-onset GHD (31, 32).
It can be argued that the beneficial effects of GH in our study are partly due to selection bias, because 12 subjects dropped out in the first 3 yr of the study. These patients showed marked individual differences in their responses to GH replacement that tended to be less than the mean changes in the group that continued therapy. It illustrates that not all GH-deficient subjects will experience sufficient benefit to warrant continuation of GH substitution. Likewise, in a recent study, 35 of 148 patients with adult-onset GHD stopped treatment within 2 yr because of insufficient subjective improvement (33). Remarkably, additional statistical analysis showed that the dropouts from our study had more psychological complaints and a higher anxiety level at baseline than the patients who continued therapy (Deijen, J. B., personal communication).
The initial GH substitution doses used in the present study were definitely too high (34). In almost all subjects the GH dose had to be reduced because of adverse effects and/or supraphysiological IGF-I levels. Starting with a low dose with individualized dose titration causes less side-effects, appears to have a similar efficacy, and, therefore, seems preferable (35, 36). All in all, the results of our study show that prolonged GH replacement therapy is beneficial and safe.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received July 20, 1998.
Revised February 12, 1999.
Accepted April 2, 1999.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  A. Giustina, G. Mazziotti, and E. Canalis Growth Hormone, Insulin-Like Growth Factors, and the Skeleton Endocr. Rev., August 1, 2008; 29(5): 535 - 559. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. L. Hartman, A. Weltman, A. Zagar, R. L. Qualy, A. R. Hoffman, and G. R. Merriam Growth Hormone Replacement Therapy in Adults with Growth Hormone Deficiency Improves Maximal Oxygen Consumption Independently of Dosing Regimen or Physical Activity J. Clin. Endocrinol. Metab., January 1, 2008; 93(1): 125 - 130. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Gotherstrom, B.-A. Bengtsson, I. Bosaeus, G. Johannsson, and J. Svensson A 10-Year, Prospective Study of the Metabolic Effects of Growth Hormone Replacement in Adults J. Clin. Endocrinol. Metab., April 1, 2007; 92(4): 1442 - 1445. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Bollerslev, J. Hallen, K. J Fougner, A. P. Jorgensen, C. Kristo, H. Fagertun, O. Gudmundsen, P. Burman, and T. Schreiner Low-dose GH improves exercise capacity in adults with GH deficiency: effects of a 22-month placebo-controlled, crossover trial Eur. J. Endocrinol., September 1, 2005; 153(3): 379 - 387. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Colao, C. Di Somma, A. Cuocolo, L. Spinelli, W. Acampa, S. Spiezia, F. Rota, M. C. Savanelli, and G. Lombardi Does a Gender-Related Effect of Growth Hormone (GH) Replacement Exist on Cardiovascular Risk Factors, Cardiac Morphology, and Performance and Atherosclerosis? Results of a Two-Year Open, Prospective Study in Young Adult Men and Women with Severe GH Deficiency J. Clin. Endocrinol. Metab., September 1, 2005; 90(9): 5146 - 5155. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Gola, S. Bonadonna, M. Doga, and A. Giustina Growth Hormone and Cardiovascular Risk Factors J. Clin. Endocrinol. Metab., March 1, 2005; 90(3): 1864 - 1870. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Colao, C. Di Somma, A. Cuocolo, M. Filippella, F. Rota, W. Acampa, S. Savastano, M. Salvatore, and G. Lombardi The Severity of Growth Hormone Deficiency Correlates with the Severity of Cardiac Impairment in 100 Adult Patients with Hypopituitarism: An Observational, Case-Control Study J. Clin. Endocrinol. Metab., December 1, 2004; 89(12): 5998 - 6004. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. R. Christ, M. H. Cummings, N. Jackson, M. Stolinski, P. J. Lumb, A. S. Wierzbicki, P. H. Sonksen, D. L. Russell-Jones, and A. M. Umpleby Effects of Growth Hormone (GH) Replacement Therapy on Low-Density Lipoprotein Apolipoprotein B100 Kinetics in Adult Patients with GH Deficiency: A Stable Isotope Study J. Clin. Endocrinol. Metab., April 1, 2004; 89(4): 1801 - 1807. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Maison and P. Chanson Cardiac Effects of Growth Hormone in Adults With Growth Hormone Deficiency: A Meta-Analysis Circulation, November 25, 2003; 108(21): 2648 - 2652. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. E. Underwood, K. M. Attie, and J. Baptista Growth Hormone (GH) Dose-Response in Young Adults with Childhood-Onset GH Deficiency: A Two-Year, Multicenter, Multiple-Dose, Placebo-Controlled Study J. Clin. Endocrinol. Metab., November 1, 2003; 88(11): 5273 - 5280. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Marleau, N. Lapointe, J. Massicotte, C. Cemeus, G. Jasmin, L. Dumont, M. G. Sirois, J.-L. Rouleau, P. du Souich, and H. Ong Effect of Chronic Treatment with Bovine Recombinant Growth Hormone on Cardiac Dysfunction and Lesion Progression in UM-X7.1 Cardiomyopathic Hamsters Endocrinology, December 1, 2002; 143(12): 4846 - 4855. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. M. Cook Shouldn't Adults with Growth Hormone Deficiency Be Offered Growth Hormone Replacement Therapy? Ann Intern Med, August 6, 2002; 137(3): 197 - 201. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Ezzat, S. Fear, R.-C. Gaillard, C. Gayle, H. Landy, S. Marcovitz, T. Mattioni, S. Nussey, A. Rees, and E. Svanberg Gender-Specific Responses of Lean Body Composition and Non-Gender-Specific Cardiac Function Improvement after GH Replacement in GH-Deficient Adults J. Clin. Endocrinol. Metab., June 1, 2002; 87(6): 2725 - 2733. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Colao, C. di Somma, R. Pivonello, A. Cuocolo, L. Spinelli, D. Bonaduce, M. Salvatore, and G. Lombardi The Cardiovascular Risk of Adult GH Deficiency (GHD) Improved after GH Replacement and Worsened in Untreated GHD: A 12-Month Prospective Study J. Clin. Endocrinol. Metab., March 1, 2002; 87(3): 1088 - 1093. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Colao, C. di Somma, A. Cuocolo, L. Spinelli, N. Tedesco, R. Pivonello, D. Bonaduce, M. Salvatore, and G. Lombardi Improved Cardiovascular Risk Factors and Cardiac Performance after 12 Months of Growth Hormone (GH) Replacement in Young Adult Patients with GH Deficiency J. Clin. Endocrinol. Metab., May 1, 2001; 86(5): 1874 - 1881. [Abstract] [Full Text]
|
|
|
|
|
|
R. Lanes, P. Gunczler, E. Lopez, S. Esaa, O. Villaroel, and R. Revel-Chion Cardiac Mass and Function, Carotid Artery Intima-Media Thickness, and Lipoprotein Levels in Growth Hormone-Deficient Adolescents J. Clin. Endocrinol. Metab., March 1, 2001; 86(3): 1061 - 1065. [Abstract] [Full Text]
|
|
|
|
|
|
H. C. Hoeck, P. Vestergaard, P. E. Jakobsen, J. Falhof, and P. Laurberg Diagnosis of Growth Hormone (GH) Deficiency in Adults with Hypothalamic-Pituitary Disorders: Comparison of Test Results Using Pyridostigmine Plus GH-Releasing Hormone (GHRH), Clonidine Plus GHRH, and Insulin-Induced Hypoglycemia as GH Secretagogues J. Clin. Endocrinol. Metab., April 1, 2000; 85(4): 1467 - 1472. [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 |
) changes in body composition parameters from baseline
values in men with GHD receiving GH for 5 yr. Data are shown for body
weight, leg muscle area (measured by computed tomographic scanning),
sum of skinfolds (measured at seven sites with a Harpenden skinfold
caliper), and intraabdominal fat (measured by computed tomographic
scanning). n = 38 unless stated otherwise. *,
P < 0.001 for comparison of changes from
baseline.
, P < 0.01; 
,
P < 0.001 (for comparison of changes from
baseline).
