| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Franz Volhard Klinik am Max Delbrück Centrum für Molekulare Medizin, Universitätsklinikum Charité, Humboldt Universität (K.J.O., O.S., R.D.), 13125 Berlin; Lilly Deutschland GmbH (W.F.B.), 61350 Bad Homburg; and University Children’s Hospital (W.F.B.), 35392 Giessen, Germany
Address all correspondence and requests for reprints to: Dr. Karl Josef Osterziel, Franz Volhard Klinik, Wiltbergstrasse 50, 13125 Berlin, Germany. E-mail: osterziel{at}fvk-berlin.de
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Treatment with GH may therefore be considered an additional therapeutic option in patients with heart failure. Volterani et al. reported that acute infusion of GH increased cardiac output by more than 50% in 12 patients with heart failure due to either ischemic or dilated cardiomyopathy (10). Chronic GH treatment, however, revealed conflicting results in patients with dilated or ischemic cardiomyopathy (11, 12, 13, 14, 15, 16). At present, the causes of the conflicting cardiovascular effects of GH in patients with heart failure are not known.
Heart failure itself and/or medical therapy may lead to subtle alterations in the somatotropic axis. Recently, Broglio et al. (17) observed that baseline IGF-I levels were significantly reduced in patients with heart failure compared to those in an age-matched control group. They did not find an alteration of GH secretion in a subgroup of six patients. In contrast, Giustina et al. reported decreased nocturnal GH secretion in patients with severe heart failure due to dilated cardiomyopathy (18). Medical treatment of heart failure may have additional influence on GH secretion. Angiotensin-converting enzyme (ACE) inhibitors and angiotensin II antagonists have been shown to increase GH and IGF-I in obese hypertensive patients (19) and in patients with congestive heart failure (20). ß2-Adrenergic agonists such as clonidine increase GH release, so that therapy with centrally acting ß-receptor blockers may decrease GH secretion (21, 22).
To further evaluate the somatotropic axis in heart failure, we examined urinary GH excretion, serum IGF-I and IGF-binding protein-3 (IGFBP-3) as markers of GH efficacy at baseline, and the endocrine response in a dose-finding study in 20 noncachectic male patients with heart failure due to coronary artery disease. The IGF-I response was then related to the severity of LV dysfunction.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Twenty male patients with heart failure due to coronary artery disease who fulfilled all of the inclusion criteria were randomized to either a low or high dose titration group. The inclusion criteria were 1) age between 25–70 yr, 2) previously documented LV ejection fraction (LV-EF) below 45% measured by either echocardiography or LV angiography, 3) confirmation of diagnosis by previous selective coronary angiography with at least 70% diameter reduction of at least one of the main coronary arteries, 4) optimized and stable medical therapy with ACE inhibitors (or in case of intolerance with angiotensin II receptor antagonists) and/or diuretics as well as stable clinical status for at least 2 weeks. Exclusion criteria were the presence of diabetes mellitus, known malignant disease, previous treatment with GH, or known alcohol or drug abuse. The study was approved by the ethics committee of the Max Delbrück Center for Molecular Medicine (Berlin, Germany), and written informed consent was obtained from all patients.
Protocol
Baseline measurements included physical examination; 12-lead electrocardiogram; M-mode and two-dimensional echocardiography; 24-h blood pressure monitoring; 3 6-min walking tests; determination of blood chemistry and serum GH, IGF-I, and IGFBP-3; and 24-h urine sampling. Subsequently, patients were randomly assigned single blind to a low dose (group A; n = 10) or a high dose (group B; n = 10) regimen of sc daily injections of human recombinant GH at 2200 h. Group A received a dose of 5 µg/kg·day for 4 days followed by 10 µg/kg·day for another 4 days, and group B received 10 and 20 µg/kg·day, respectively. Blood samples were taken in the morning on days 5 and 9. Echocardiography and blood pressure monitoring were repeated on day 9.
Echocardiography and 24-h blood pressure monitoring
The echocardiographic examinations were performed using a 3.5-mHz transducer (Toshiba Sonolayer SSH-160 A). M-mode recordings were taken in the parasternal long axis, and measurements were performed according to the recommendations of the American Society of Echocardiography (23). LV-EF was calculated by the modified Simpson’s rule (24).
Blood pressure was recorded with a standardized sphygmomanomer in the sitting position. Twenty-four-hour blood pressure and average heart rate were determined by the oscillometric method (Spacelab 90207) before and on day 9 of GH application. The recordings were obtained every 20 min from 0600–2200 h and every 30 min from 2200–0600. The average number of measurements was 62 ± 2.
Hormonal measurements
Blood samples were collected from an antecubital vein in the morning between 0800–0900 h after a fasting period of more than 12 h. After coagulation and centrifugation, serum aliquots were stored at -20 C until analysis. Serum levels of GH (assay sensitivity, 0.3 ng/mL) and IGF-I (assay sensitivity, 2 ng/mL) were measured with specific RIAs (25, 26) as described previously. The IGF-I assay uses an excess of IGF-II to eliminate interferences with IGFBPs (RIA; Mediagnost, Tubingen, Germany; reference range for age, 50–60 yr; 97–292 ng/mL) (26, 27). Intra- and interassay coefficients of variation were 3.2% and 7.6%, respectively. Serum levels of GH-binding protein (GHBP) were measured by the RIA described by Kratzsch et al., using an antibody directed against exon 3 of GHBP (28). The intra- and interassay variabilities were below 8.5% and 12%, respectively. The exon 3-GHBP correlates highly to the undifferentiated functional GHBP forms (28) and therefore reflects the liver density of GH receptors. IGFBP-3 was measured using a specific RIA with intra- and interassay coefficients of variation of 3.5% and 7.3%, respectively (assay sensitivity, 30 ng/mL; reference range for age, 50–60 yr; 1960–4650 ng/mL) (29). All samples for GH, GHBP, IGF-I, and IGFBP-3 determinations were measured in one assay. Twenty-four-hour urine was collected for measurement of urinary GH using a commercially available assay (BioMérieux, Nurtingen, Germany). The intraassay variability is 3.3%, and the interassay variability is 5.6%. For comparison with GH excretion of our patients, the 24-h mean ± SD (9.5 ± 6.0 pg/24 h) and reference range (0.74–26.24 pg/24 h) established by Braschi et al. were used (30). The laboratory was blinded as to the dose and time of the samples. The biochemical parameters were compared to those of healthy controls previously examined by us by calculating age-adjusted SD scores (27). All other parameters were analyzed by routine analysis in our hospital.
Statistics
Normally or symmetrically distributed data were compared by the
independent samples t test or paired t test. The
significance of differences was tested by using ANOVA for repeated
measures or Student’s t test with Bonferroni correction.
Categorical variables were evaluated by the 
2
test. Univariate regression analysis using the Pearson and Spearman
correlation coefficients were employed to analyze the relation between
variables (StatView 4.5 for Macintosh; Abacus Concepts, Inc.,
Berkeley, CA). All data, with the exemption of SD
scores, are presented as the mean and SEM.
P < 0.05 was taken as significant.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
All patients had depressed LV function (mean LV-EF, 30.0 ±
8.4%) due to coronary artery disease (8 patients with 1-vessel
disease, 8 patients with 2-vessel disease, and 4 patients with 3-vessel
disease). All patients had at least one previously documented
myocardial infarction (MI; 16 patients, anterior MI; 4 patients,
posterior MI). Four patients with anterior MI had 1 additional MI.
Fourteen patients were limited by dyspnea, and 6 patients were without
limitations of physical activity. This resulted in a mean New York
Heart Association (NYHA) class of 2.0 ± 0.8 (NYHA I:6, NYHA II:9,
NYHA III:5) and an average 6-min walking distance of 415 ±
82 m. All patients had had stable body weight during the previous
6 months. The severity of heart failure, as assessed by LV
end-diastolic dimension, EF, and walking distance, was not dependent on
age. The clinical and biochemical characteristics of the 2 patient
groups did not differ at baseline (Tables 1
and 2). There was a
significant difference in the mode of anticoagulation and the use of
aspirin, but the remaining medication did not differ between
groups.
|
|
Baseline insulin (normal range, 58–172 pmol/L), C peptide (normal range, 0.9–4.0 µg/L), fasting blood glucose (normal range, 3.4–5.6 mmol/L), and glycosylated hemoglobin (normal range, 3.9–5.7%) were in the normal range in both groups. Mean urinary GH (uGH) excretion and serum concentrations of IGF-I and IGFBP-3 for all patients at baseline were 1.4 ± 0.5 pg/24 h (range, not detectable to 10.4 ng/24 h), 117 ng/mL (range, 82–194), and 2648 ng/mL (range, 1656–4329). Baseline uGH excretion and serum IGF-I and IGFBP-3 were significantly diminished compared to the reference ranges derived from healthy age-matched subjects (27, 30): uGH, -1.38 ± 0.41 SD score (P < 0.0001); IGF-I, -1.11 ± 0.74 SD score (P < 0.0001); and IGFBP-3, -0.57 ± 0.78 SD score (P < 0.05). Baseline uGH excretion was below the detection limit (0.5 pg/mL) in 9 of 20 patients. Baseline uGH and IGF-I were not correlated to LV dimensions or fractional shortening. Baseline serum levels of IGF-I, but not uGH, correlated weakly to EF (r = 0.43; P = 0.058). Age, body mass index, and baseline insulin levels did not correlate to IGF-I or uGH.
GH treatment
GH treatment was tolerated in all patients, and functional classification of heart failure remained unchanged. Cardiac size and function, as assessed by echocardiography, were not altered after 8 days of GH application. Sitting blood pressure and heart rate did not change after all three doses of GH compared to their respective baseline values. Mean blood pressure determined by the 24-h recordings also remained unchanged in the low dose group, but decreased after the highest GH dose significantly, however to a small extent (-3 ± 2 mm Hg; P = 0.03).
The average daily doses of GH were 402 ± 19 µg (1.2 ± 0.1
IU) and 803 ± 38 µg (2.4 ± 0.1 IU) in group A and
776 ± 47 µg (2.3 ± 0.1 IU) and 1551 ± 94 µg
(4.7 ± 0.3 IU) in group B. There was a dose-dependent increase in
both serum IGF-I and IGFBP-3 during GH application (Fig. 1). The low dose group showed significant
increases in IGF-I from -1.3 ± 0.3 to 0.3 ± 0.1
SD score with 5 µg/kg GH·day (P <
0.001) and to 1.2 ± 0.2 SD score with 10
µg/kg GH (P < 0.001). IGFBP-3 increased in this
group from -0.7 ± 0.2 to 0.1 ± 0.2 and 0.4 ± 0.1
SD score, respectively (P <
0.001 for both). The changes were more pronounced in the high dose
group. IGF-I increased from -0.9 ± 0.2 to 1.0 ± 0.2
SD score with 10 µg/kg GH (P <
0.001) and to 1.8 ± 0.2 SD score with 20
µg/kg GH (P < 0.001). IGFBP-3 increased from
-0.4 ± 0.3 to 0.4 ± 0.2 SD score
(P < 0.001) and 0.8 ± 0.2
SD score (P < 0.001)
respectively. These hormonal changes represent alterations from the low
normal to the upper normal range and, most likely with the highest GH
dose, to slightly supranormal values.
|
).
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Baseline GH/IGF system
Decreased serum IGF-I and IGFBP-3 levels in these patients can be due to impaired GH secretion or impaired GH sensitivity. The following findings support the view that impaired GH secretion is the predominant cause. 1) uGH excretion was diminished compared to that in healthy controls. The measurement of uGH excretion correlates with GH secretion during the urine collection period (31, 32, 33, 34). Therefore, our measurements of GH excretion over 24 h indicate decreased 24-h GH secretion. Measurement of serum GH was not performed because single serum GH levels would not provide reliable information on GH secretion, as secretion is pulsatile (35). Our finding has to be confirmed by measurements of GH secretion, because uGH excretion shows significant, but not very tight, correlations to GH secretion (34). 2) Low GH doses, comparable to those used in replacement therapy in adults with GH deficiency, produced significant increases in both IGF-I and IGFBP-3 (36, 37, 38). 3) The time course of the IGF-I and IGFBP-3 responses was as rapid as that in patients with GH deficiency (39, 40). Further, the occurrence of side-effects after only 8 days of GH treatment suggests preserved GH sensitivity.
Our results are in accordance with those of an earlier study in patients with dilated cardiomyopathy in which decreased nocturnal GH secretion was found in six patients with EFs below 20% (18). In another study, however, a change in nocturnal GH secretion could not be shown in a small number of patients with dilated cardiomyopathy (17), but basal serum IGF-I levels were decreased in the patients with severe heart failure due to coronary artery disease or to dilated cardiomyopathy compared to those in age-matched controls. Normal serum IGF-I levels, however, were reported by Anker et al. (41) in noncachectic congestive heart failure with mixed etiology, whereas cachectic patients had higher serum GH but low IGF-I levels, suggesting GH resistance. Genth-Zotz et al. (16) reported normal IGF-I levels in seven patients with a severity of ischemic cardiomyopathy comparable to that in our patients. In both studies, however, IGF-I levels were not analyzed compared to age-adjusted normal values as we did by evaluation of SD scores.
ACE inhibitors, angiotensin II antagonists, and even ß-receptor-blocking agents may increase GH secretion (19, 20, 21, 22). However, all studies of heart failure did not consider the likely influence of medical treatment on GH secretion and IGF-I levels. All patients in our study received an ACE inhibitor or an angiotensin II antagonist, so that a possible influence of these drugs on GH secretion could not be evaluated. To examine the influence of ß-blockers on GH secretion, we compared the 15 patients with and the 5 patients without treatment with ß-blockers, but significant differences in uGH excretion or serum IGF-I could not be found. Even if there was an influence of drug treatment on GH secretion, IGF-I levels would be increased. The low IGF-I levels of our patients despite treatment with inhibitors of the renin-angiotensin and the sympathetic nervous system may thus support the assumption of decreased baseline GH secretion. Aside from age, multiple other factors, such as nutritional status, insulin, steroid hormones, T4, and different methods of IGF-I assays may in addition have contributed to the serum IGF-I levels reported in other studies (12, 15, 16, 41).
The cause of the decreased GH secretion in ischemic cardiomyopathy is unknown. The borderline relation of baseline IGF-I levels to EF indicates that the degree of cardiac dysfunction may influence GH secretion. The large intra- and interindividual variability in uGH (42) and also serum GH (43) may have obscured a possible relation of GH secretion to baseline cardiac function. Accordingly, Giustina et al. did not find significant relations of spontaneous serum GH levels to hemodynamic data, but they reported a positive correlation of IGF-I to cardiac index and a negative correlation to pulmonary artery pressure in 12 patients with dilated cardiomyopathy (43). It is therefore possible that a pathophysiological link exists between the severity of heart failure and GH secretion. The progression of heart failure is accompanied by an increasing extent of sympathetic activation (44) with an increase in central sympathetic outflow (45). Increased central sympathetic activity also activates neurons of the paraventricular nucleus and the nucleus coeruleus of the hypothalamus (46). Stimulation of these nuclei can result in increased somatostatin release, which blocks GH secretion (47, 48). This assumption is supported by a reduced GH response to GHRH but a normal response to hexarelin in patients with heart failure (17). The normal response to hexarelin demonstrates a normal releasable GH pool. Therefore, the reduced response to GHRH may be due to increased activity of the somatostatinergic system (17). Increased somatostatin release or other mechanisms acting via a change in neurotransmitters, such as a decreased parasympathetic drive (see Ref. 22 for review) may then lead to impairment of GH secretion in heart failure.
GH treatment effects
The IGF-I response to a GH dose of 10 µg/kg·day was significantly related to the degree of LV dysfunction, which means that the sensitivity of the somatotropic system becomes lower with increasing severity of cardiac dysfunction. Moreover, the negative association of IGF-I changes to LV dimensions agrees with the concept that the GH/IGF system may also be involved in the process of LV remodeling. To our knowledge this is the first report of an interaction between cardiac function and GH sensitivity. Broglio et al. could not find a relation of basal or GH-stimulated IGF-I levels to the severity of heart failure in a smaller number of patients with either ischemic or dilated cardiomyopathy (17). This may have been due to the larger variance in IGF-I levels compared to our study (baseline IGF-I level, 148 ± 50 compared to 117 ± 33 pg/mL). The observed relationship between the magnitude of GH-induced serum IGF-I elevations and LV dysfunction could be a direct effect of a reduced sc blood flow causing diminished absorption of sc administered GH. However, patients with comparable severity of heart failure had a GH-induced IGF-I increase similar to the increase in controls (17). This argues against delayed absorption of GH in our patients with compensated moderate heart failure. Liver function, as assessed by serum albumin and alanine-aminotransferase, and renal function, as assessed by serum creatinine, were normal and remained unchanged. Therefore, impaired hepatic function or altered clearance of IGF-I and its binding proteins is an unlikely cause of the altered IGF-I response (49, 50). In addition, the increase in IGF-I was not related to serum levels of GHBP, which probably reflects hepatic GH receptor density (51). Therefore, an as yet unknown pathomechanism may interfere with the GH-induced production of IGF-I. GH resistance, as reported in cachectic patients with heart failure (41), may thus be the one end of a spectrum of decreasing GH sensitivity paralleling the degree of LV dysfunction. Careful evaluation of the somatotropic axis and monitoring of serum IGF-I appear to be helpful measures to adjust GH doses to the requirements of the individual patient and to avoid undesirable side-effects in forthcoming clinical trials in patients with heart failure.
In conclusion, we have shown diminished GH secretion in patients with ischemic cardiomyopathy. GH treatment induces a rapid, dose-dependent increase in serum IGF-I. The magnitude of GH-induced serum IGF-I elevations is related to the severity of LV dysfunction.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received October 11, 1999.
Revised December 29, 1999.
Accepted January 7, 1999.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
1-Noradrenergic inhibition of growth
hormone secretion is mediated through the paraventricular hypothalamic
nucleus in male rats. Neuroendocrinology. 59:29–34.[Medline]
This article has been cited by other articles:
 |
|
 |
|
  P. Le Corvoisier, L. Hittinger, P. Chanson, O. Montagne, I. Macquin-Mavier, and P. Maison Cardiac Effects of Growth Hormone Treatment in Chronic Heart Failure: A Meta-Analysis J. Clin. Endocrinol. Metab., January 1, 2007; 92(1): 180 - 185. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. C. Cuneo GH and Cardiac Failure J. Clin. Endocrinol. Metab., October 1, 2001; 86(10): 4635 - 4637. [Full Text] [PDF]
|
|
|
|
|
|
J. W. A. Smit, Y. J. H. Janssen, H. J. Lamb, E. E. van der Wall, M. P. M. Stokkel, E. Viergever, N. R. Biermasz, J. J. Bax, H. W. Vliegen, A. de Roos, et al. Six Months of Recombinant Human GH Therapy in Patients with Ischemic Cardiac Failure Does Not Influence Left Ventricular Function and Mass J. Clin. Endocrinol. Metab., October 1, 2001; 86(10): 4638 - 4643. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. H. Wang Roads to Survival : Insulin-Like Growth Factor-1 Signaling Pathways in Cardiac Muscle Circ. Res., March 30, 2001; 88(6): 552 - 554. [Full Text] [PDF]
|
|
|
|
|
|
K. Yamashita, J. Kajstura, D. J. Discher, B. J. Wasserlauf, N. H. Bishopric, P. Anversa, and K. A. Webster Reperfusion-Activated Akt Kinase Prevents Apoptosis in Transgenic Mouse Hearts Overexpressing Insulin-Like Growth Factor-1 Circ. Res., March 30, 2001; 88(6): 609 - 614. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
