© 2000 American Heart Association, Inc.
ATVB Electronic Pages |
Insulin-Like Growth Factor–Binding Protein-3 Is Associated With the Presence and Extent of Coronary Arteriosclerosis
From the Institut für Arterioskleroseforschung an der Universität Münster (S.S.-L., H.S., G.B., G.A., A.v.E.), the Medizinische Klinik und Poliklinik C (Kardiologie, Angiologie) (G.M., A.E., G.B., S.K.), the Institut für Klinische Chemie und Laboratoriumsmedizin, Zentrallaboratorium, and the Interdisziplinäres Zentrum für Klinische Forschung (IZKF), Medizinische Fakultät (A.v.E.), Westfälische Wilhelms-Universität Münster, Münster, Germany.
Correspondence to Dr Arnold von Eckardstein, Institut für Klinische Chemie und Laboratoriumsmedizin, Zentrallaboratorium, Westfälische Wilhelms-Universität Münster, Albert-Schweitzer-Strasse 33, D-48149 Münster, Germany. E-mail vonecka{at}uni-muenster.de
Abstract
Abstract—Aging is associated with
the progression of arteriosclerosis and the decline
of several endocrine functions. We therefore investigated the
association of coronary arteriosclerosis
with hormones, the serum concentrations of which change during aging.
Coronary angiograms of 189 men <70 years old were evaluated by
3 semiquantitative score systems to estimate the extent of focal and
diffuse vessel wall alterations. Fasting sera were analyzed for
levels of glucose, lipids, thyroid-stimulating hormone, insulin,
insulin-like growth factor I (IGF-I), IGF-binding protein-3 (IGFBP-3),
dehydroepiandrosterone sulfate (DHEAS), testosterone, and sex
hormone–binding globulin (SHBG). After adjustment for age, body mass
index, and waist-to-hip ratio, 92 patients with 
1 stenoses
>70% differed from 97 patients without such focal lesions by higher
serum levels of glucose, total and LDL cholesterol, and
apolipoprotein (apo) B, as well as by lower serum levels of IGFBP-3.
Multivariate analyses revealed significant and
independent correlations of all 3 coronary scores with LDL
cholesterol (or apoB) and IGFBP-3; of 2 coronary
scores with age, glucose, and insulin; and of 1 score with IGF-I. No
significant correlations existed for waist-to-hip ratio (or body mass
index) and DHEAS (or testosterone or SHBG). IGFBP-3 explained 9% to
14% and 3.5% to 10% of the variances of focal and diffuse lesions,
respectively. In conclusion, IGFBP-3 and, with much less strength and
consistency, insulin and IGF-I, but not markers of
hypothyroidism, adrenopause, and andropause, have statistically
significant and independent associations with coronary
arteriosclerosis in men.
Key Words: aging • androgens • growth hormone • insulin resistance • quantitative coronary angiography
Aging is characterized by a gradual decrease of many
biological functions. Thus, the progression of
atherosclerosis is preceded and paralleled by the
impairment or loss of function in other general
physiological systems of the organism, for example,
the endocrine system.1 The endocrinology of aging is
characterized by the increased prevalence of autoimmune thyroiditis and
decreased conversion of thyroxin into triiodothyronine, which result in
thyroid dysfunction; by insulin resistance and ß-cell failure, which
result in glucose intolerance; by somatopause, with decreasing serum
levels of growth hormone and insulin-like growth factor I (IGF-I); and
by adrenopause, with declining serum levels of dehydroepiandrosterone
sulfate (DHEAS). In addition, menopause leads to the loss of estrogens
in women and andropause to the decrease of testosterone serum levels in
50% of men >60 years old.1
Insulin resistance and insulin deficiency contribute to the manifestation and progression of coronary arteriosclerosis both directly and through their metabolic sequelae, eg, diabetes mellitus, hypertension, and dyslipidemia.2 3 The protective effects of estrogens on the biology of vascular cells and the clinical course of atherosclerotic vessel disease in women have been documented in great detail.4 5 Much less is known about the relationships between arteriosclerosis and testosterone, DHEAS, or the growth hormone/IGF-I axis. Serum levels of testosterone and DHEAS were found to have significant associations with the incidence or presence of coronary heart disease, which, however, were related inversely in men and positively in women.6 7 Animal studies and in vitro studies have also yielded contradictory results on the role of androgens in arteriosclerosis.6 7 Several clinical studies have reported associations of growth hormone deficiency as well as of low serum levels of (free) IGF-I with premature coronary heart disease.8 9 10 11 12 However, growth hormone deficiency and low serum levels of IGF-I, high serum levels of thyroid-stimulating hormone (TSH) as an indicator of (latent) hypothyroidism, and low serum levels of testosterone or DHEAS in men or high serum levels of these androgens in women are confounded by various metabolic disorders, including obesity, insulin resistance, dyslipidemia, and impaired fibrinolysis.7 8 13 14
In the present study, we investigated the relationships of the above-mentioned endocrine systems to the severity of coronary arteriosclerosis in men, which was quantified with 3 semiquantitative angiographic algorithms.15
Methods
Subjects
A total of 189 men <70 years old (54.7±11.1 years) were
selected out of 2346 consecutive patients who underwent
coronary angio-graphy or percutaneous
transluminal coronary angiography (PTCA).15
Selection criteria were (1) first coronary angiography or PTCA
and (2) the absence of any lipid-lowering therapy. Indications to
undergo coronary angiography were either verification of
coronary heart disease (77%) in patients with angina pectoris,
previous myocardial infarction older than 2 months, abnormal exercise
or thallium tests, or planned first PTCA, or exclusion of
coronary artery disease (23%) in patients with other cardiac
diseases. These indications and exclusion criteria have been described
in detail previously.15
Anthropometric Data
Body height was measured in centimeters with a statometer. For
the determination of body weight, the probands were dressed only in
underwear. Body mass index (BMI) was calculated as the ratio of weight
(kg) to the square of height (m2). The
circumferences of waist and hip were measured in millimeters at the
level of the umbilicus and at the largest circumference at the
buttocks, respectively. Waist-to-hip ratio (WHR) was calculated by
division of the 2 parameters.
Coronary Angiography
The cine films were evaluated independently by 2 cardiologists
blinded to the patients’ clinical or laboratory findings. The
evaluation included 3 different scoring systems, as
follows.15
Vessel Score
Vessel score was 0 to 3 points. Each of the main
coronary arterial branches (left anterior
descending, left circumflex artery, right coronary artery)
having 1 stenoses of 70% was given 1 point each. If the
left main stem, which was regarded as 1 vessel, and the left anterior
descending and/or the left circumflex artery were affected, this was
counted as 2 points.
Stenosis Score
Stenosis score was 0 to 32 points. The maximum diameter
reduction of 8 coronary segments (left main stem, left anterior
descending artery, main diagonal branch, main septal branch, left
circumflex artery, main marginal branch, right coronary artery,
right posterior descending branch) was scored with 1 to 4 points
according to a luminal narrowing of 1% to 49% (1 point), 50% to 74%
(2 points), 75% to 99% (3 points), or a total occlusion (4
points).
Extent Score
Extent score was 0 to 100 points. This score was developed
by Sullivan et al.16 According to the proportional
length of each vessel segment in the coronary artery tree,
segments were graded with different maximum numbers of points: 5 points
for the left main stem, 20 for the left anterior descending artery, 10
for the main diagonal branch, 5 for the first septal branch, 20 for the
left circumflex artery, 10 for the obtuse marginal and posterolateral
vessels, 20 for the right coronary artery, and 10 for the right
posterior descending branch. The number of points for each segment
(irrespective of the degree of diameter reduction) was expressed as the
percentage length of visible lesions within the total segment. Occluded
vessels that were filled with contrast medium by collateral flow were
scored according to the visible irregularities of the vessel wall. If
no collateral flow existed, the mean value of all other vessel segments
of this angiogram was given to this occluded vessel segment.
Sample Collection
Blood was taken after 12 to 14 hours of fasting in the morning
and before coronary angiography. Sera were obtained after 30
minutes of clotting by centrifugation at
2000g for 15 minutes. Sera were removed and either used
directly for the measurements of metabolic
parameters and TSH or frozen at -70°C for later
determination of other endocrinological parameters within
the series.
Measurements of Metabolic Parameters
Serum levels of glucose, total cholesterol,
triglycerides, and HDL cholesterol (HDL-C) were
determined by enzymatic tests (Roche Diagnostics). HDL-C
was determined after precipitation of apoB-containing lipoproteins with
phosphotungstic acid/MgCl2. LDL
cholesterol (LDL-C) was calculated by the Friedewald
formula.17 Apolipoproteins A-I (apoA-I) and B (apoB) were
measured with turbidimetric immunoassays from Roche
Diagnostics, and lipoprotein(a) [Lp(a)] with a
latex-enhanced turbidimetric immunoassay from Immuno. All measurements
were performed with the Hitachi 917 Autoanalyzer (Roche
Diagnostics). Interassay imprecisions were <5% for all
tests.
Measurements of Hormones
Serum levels of insulin and insulin-like growth factor–binding
protein-3 (IGFBP-3) were determined by use of microtiter plate enzyme
immunoassays from Dako and Diagnostic Systems Laboratories
Inc, respectively. These hormone measurements were done in duplicate,
and results were accepted if the coefficients of variation were <10%.
Automated chemiluminescence immunoassays were used for the
quantification of IGF-I and DHEAS (Advantage from Nichols
Diagnostics), testosterone and TSH (ACS 180 from Chiron
diagnostics), and sex hormone–binding globulin (SHBG)
(Immulite system from DPC Biermann GmbH). Interassay coefficients of
variation of the latter hormone measurements were <10%.
Statistics
An exploratory analysis was performed with the
Statistical Package for the Social Sciences (SPSS-X).18
Comparisons of mean values between patients with coronary
arteriosclerosis (vessel, stenosis, or
extent score>0) and patients without coronary
arteriosclerosis (vessel, stenosis, or
extent score=0) were done by t test. This analysis
was performed on unadjusted data and on data that were adjusted for
age, BMI, and WHR by standardization for the median of the respective
parameters. Univariate regression
analyses were performed according to Pearson. Multiple
regression analyses were performed only on
parameters that on univariate analysis
had significant correlations with 1 vessel score. All indicated
parameters were entered simultaneously into the
various models. The variance in arteriosclerotic
lesions explained by the various multivariate models
and the various parameters within the model was estimated
by R2 and the square of the
correlation coefficient ß, respectively. For some
parameters [triglycerides, Lp(a), insulin,
IGF-I, testosterone, SHBG, testosterone/SHBG ratio, DHEAS, TSH],
logarithmic (loge) transformation was necessary
to obtain a Gaussian frequency distribution, which is mandatory for
these parametric tests.
Results
Association of Metabolic Risk Factors and Hormones With
the Presence of Coronary Arteriosclerosis
Tables I and
II compare the mean values of
anthropometric parameters, metabolic risk
factors, and hormones in patients with and without angiographically
detectable coronary arteriosclerosis.
Absence of coronary arteriosclerosis was
defined either as the absence of stenoses >70% in the main
coronary arterial branches, ie, vessel score=0
(Table
I) or as the absence of any diameter reduction or diffuse
lesion in the 8 coronary segments, ie, stenoses and
extent score=0 (Table II). In both scoring systems, patients
with angiographically detectable coronary
arteriosclerosis were characterized by
significantly higher age (P<0.001) and higher serum levels
of glucose (P<0.001 and P<0.05), total
(P<0.001 and P<0.05) and LDL
(P<0.001) cholesterol, and apoB
(P<0.01), as well as by lower serum levels of IGFBP-3
(P<0.001 and P<0.01) (Tables I and
II). In addition, the presence of focal lesions, ie, vessel
score 1, 2, or 3, was associated with significantly higher BMI
(P<0.05), and lower level of DHEAS (P<0.01)
(Table I) and presence of diffuse lesions, ie, stenoses
and extent scores >0, were also associated with significantly higher
WHR and significantly lower apoA-I (P<0.05) and IGF-I
(P<0.05) levels (Table II). After adjustment for
age, BMI, and WHR, the differences in glucose, cholesterol,
LDL-C, apoB (all scores), apoA-I (stenoses and extent scores),
and IGFBP-3 (vessel score) remained significant.
|
|
Correlations Between Coronary Scores and Clinical and
Laboratory Findings
Tables III and IV show the results of a multiple
regression analysis of relationships between coronary
scores and potential risk factors. In these models, only those
parameters were considered that had significant
associations with the presence of focal or diffuse
arteriosclerotic lesions (Tables I and
II) or had significant bivariate correlations with 1 vessel
scores (not shown). In some cases, 2 parameters that are
functionally related and therefore strongly correlated with one another
had significant associations with the presence of angiographically
detectable lesions or correlations with coronary scores, eg,
BMI and WHR, HDL-C and apoA-I, and LDL-C and apoB. In these cases, we
entered only 1 variable. We primarily tested a model that
encompassed the variables age, WHR, glucose, LDL-C, HDL-C, insulin,
IGF-I, IGFBP-3, and DHEAS (Table III). This model explained
28% to 32% of the variation in coronary scores. LDL-C and
IGFBP-3 had significant correlations with all 3 vessel scores. Age and
glucose had significant correlations to the stenosis and extent
scores. Multivariate correlations between insulin and
vessel score or stenosis score were also statistically
significant (P<0.05). In this model, IGFBP-3 explained
9.2%, insulin 2.8%, and LDL-C 7.3% of the variance in focal lesions.
The variances in diffuse lesions explained by IGFBP-3, insulin, and
LDL-C were 3.5% to 5.7%, 1.6% to 2.9%, and 6.0% to 8.9%,
respectively.
|
|
When apoB and apoA-I were entered into the multiple regression model
instead of LDL-C and HDL-C, respectively, the models also explained
30% of the variation in coronary vessel scores (Table
IV). The correlations of apoB to the coronary scores
tended to be stronger than the correlations of LDL-C. Moreover,
and interestingly, in this model compared with the previous model, the
correlations of coronary scores with IGFBP-3 became stronger
and the correlations of coronary scores with glucose and
insulin became weaker (Tables III and IV). In this
model, IGFBP-3 explained 13.8%, insulin 1.8%, and apoB 7.8% of the
variance in focal lesions. The same parameters accounted
for 6.0% to 9.9%, 1.0% to 1.8%, and 6.8% to 9.5% of the variances
in diffuse lesions, respectively.
IGF-I had no or only borderline statistically significant correlations with any coronary score. These correlations did not become stronger when IGFBP-3 was omitted from the models. Conversely, however, IGFBP-3 kept most of its statistically significant correlations when IGF-I was omitted. In none of these models did HDL-C, apoA-I, WHR, or DHEAS show significant and independent correlations to any coronary score. Likewise, neither BMI nor testosterone (nor SHBG nor testosterone ratio) had significant correlations with any vessel score when entered instead of WHR or DHEAS, respectively.
Discussion
Interindividual variability in the decay of various endocrine functions may contribute to differences in the clinical onset and further progression of arteriosclerosis, either directly by the regulation of vascular functions or indirectly through the interaction with risk factors. To unravel the role of various endocrine systems in arteriosclerosis whose function is reduced by aging, we investigated the relationships between the presence or severity of coronary arteriosclerosis and various hormones or conventional risk factors in 189 men <70 years old.
IGFBP-3 was the only endocrine parameter that on adjusted and multivariate statistical analyses consistently had significant associations with the presence and extent of coronary arteriosclerosis. On multivariate regression analyses, IGFBP-3 had significant correlations with both the number of focal lesions (vessel score), the number of diffuse lesions (stenoses score), and the extent of diffuse lesions (extent score), which were independent of age, BMI or WHR, LDL-C or apoB, HDL-C or apoA-I, and insulin or glucose, as well as DHEAS, SHBG, or testosterone (Tables III and IV). IGFBP-3 explained 9% to 14% of the variance in focal lesions and 3.5% to 10% of the variance in diffuse lesions, compared with 7% to 8% and 6% to 10%, respectively, explained by LDL-C or apoB, and 2% to 3% and 1% to 3%, respectively, explained by insulin (Tables III and IV).
What is the physiological basis of the associations between IGFBP-3 and the presence and extent of coronary arteriosclerosis? The association may be causal and reflect an involvement of the somatotropic axis in the pathogenesis of arteriosclerosis but may also be a surrogate for the actions of other factors that affect both IGFBP-3 serum levels and arteriosclerosis, for example, insulin resistance, obesity, and sex hormones. Finally, arteriosclerosis itself may influence IGFBP-3 levels.
IGFBP-3 is one of 7 binding proteins that prolong the half-life of
IGF-I and IGF-II in plasma and that modulate the interaction of IGFs
with their receptors on target cells and thereby their
mitogenic and metabolic
actions.19 20 21 IGFBP-3 accounts for 75% of the
circulating IGFBPs and transports 70% to 90% of IGF-I.19
IGFBP-3 inhibits cell growth, both directly by inhibition of DNA
synthesis and indirectly by sequestration of IGFs, which stimulate cell
growth.19 20 21 To the best of our knowledge, no data on
direct interactions of IGFBP-3 with vascular cells have been reported.
However, one of its ligands, IGF-I, was shown to stimulate the
proliferation of arterial smooth muscle cells if
present at low concentrations and to inhibit smooth muscle cell
proliferation if present at high concentrations.22 23
Moreover, IGF-I stimulates nitric oxide production in both
smooth muscle cells and endothelial
cells.24 Thus, IGF-I exhibits both proatherogenic and
antiatherogenic properties. In agreement with an antiatherogenic effect
of IGF-I, we (this study) and other researchers found inverse
associations between IGF-I and free IGF-I with coronary heart
disease8 10 and premature
arteriosclerosis in growth hormone–deficient
hypophysectomized patients.11 12 However, it is important
to note that in our study, most of the associations and bivariate
correlations (not shown) of IGF-I lost their statistical significance
after adjustment for age and obesity (Tables I and II)
and on multivariate analysis (Tables
III and IV). Moreover, because IGFBP-3 sequesters
bioactive free IGF-I, the inverse association of IGFBP-3 with
arteriosclerosis also at first sight argues against
antiatherogenic actions of IGF-I. However, high levels of IGFBP-3 in
serum may also reflect a large pool of bound IGF-I that can be released
by proteases within the arterial wall.25 26 27
It will hence be interesting to study the direct effects of both IGF-I
and IGFBP-3 on vascular cells. Alternatively, because both IGF-I and
IGFBP-3 serum levels correlate with growth hormone secretion and
because growth hormone exerts some of its actions independently of
IGF-I,19 20 21 the inverse association of both IGF-I and
IGFBP-3 may reflect the involvement of this hormone in
arteriosclerosis. Unfortunately, the pulsatile
secretion of growth hormone does not allow us to test this hypothesis
directly by the measurement of growth hormone serum levels. In this
context, it is noteworthy that growth hormone exerts beneficial effects
on some cardiovascular risk factors independently of
IGF-I, for example, lowering LDL-C and Lp(a) and increasing
HDL-C.13 14 28 29 30
IGFBP-3 concentration is also regulated by proteolytic cleavage, which in turn is regulated by insulin.19 25 26 27 For example, proteolysis of IGFBP-3 was found to be increased in patients with type 2 diabetes.31 Thus, low IGFBP-3 in coronary heart disease patients may reflect insulin resistance. In our study, however, insulin levels did not correlate with IGFBP-3 (not shown), and multivariate correlations of IGFBP-3 with the extent of coronary arteriosclerosis were stronger than and independent of insulin (Tables III and IV). IGFBP-3 had inverse correlations with BMI, WHR, and glucose (not shown), which may indicate the relationship of IGFBP-3 with insulin resistance. Nevertheless, the association of IGFBP-3 with the presence of disease (Tables I and II) and its correlations with the 3 coronary scores were independent of measures of obesity and glucose (Tables III and IV). It is important to note, however, that the strength of correlations between the 3 coronary scores and glucose or insulin on the one hand and IGFBP-3 on the other hand varied at mutual expense. This may be a subtle indication that low IGFBP-3 levels are part of the insulin resistance cluster, which contributes to arteriosclerosis.
Finally, some of the proteases that degrade IGFBP-3 are also secreted by macrophages, eg, metalloproteinases.19 26 It is hence important to consider that low IGFBP-3 may simply indicate the presence of vascular disease.
Insulin and/or glucose also had significant and independent associations with various coronary scores. These relationships are in agreement with the previously demonstrated associations of glucose intolerance/diabetes mellitus and insulin resistance/hyperinsulinemia with coronary arteriosclerosis.2 3 32 Traditionally, these associations have been explained with the manifestation of a proatherogenic risk factor profile in many glucose-intolerant and insulin-resistant patients.32 In our study, however, several components of the insulin resistance syndrome, ie, obesity, triglycerides, and blood pressure, had no significant or independent associations with coronary arteriosclerosis. Moreover, the correlations of insulin and glucose with the severity of coronary arteriosclerosis were independent of HDL-C or apoA-I. Thus, it appears that fasting insulin and glucose are more specific markers of the atherogenic metabolic syndrome and/or reflect direct interactions with the atherosclerotic process. In this context, it is also important to note that insulin exerts several direct effects on vascular wall cells, eg, induction of nitric oxide release from endothelial cells and stimulation of ion pumps in smooth muscle cells, resulting in endothelium-dependent and -independent vasodilation, respectively.2 3
On univariate analysis, we confirmed inverse
associations between DHEAS and the presence and extent of
coronary arteriosclerosis (Table I)
that were also assessed in several previous studies.6 7
However, these associations were not stable after adjustment for
confounders (age, obesity) and on multivariate
analysis, probably because of the correlations of DHEAS with
age, glucose, IGF-I, IGFBP-3, and LDL-C. Also in other studies, DHEAS
was confounded with components of the metabolic syndrome so
that its role as an independent cardiovascular risk
factor is questionable.6 7 Likewise, it is important to
note that neither testosterone nor the testosterone/SHBG ratio as a
surrogate marker for free and bioactive testosterone levels had
significant associations, which have been found in some previous
studies.6 7 33
In conclusion, IGFBP-3 and, with much less strength and consistency, insulin and IGF-I, but not markers of hypothyroidism, adrenopause, and andropause, have statistically significant and independent associations with coronary arteriosclerosis in men.
Acknowledgments
This project was supported by a grant from the Interdisziplinäres Zentrum für klinische Forschung (IZKF) Münster to Dr von Eckardstein (Project A3). Dr Schuler-Lüttmann is a recipient of a research fellowship from Deutsche Infarktforschungshilfe. We gratefully acknowledge the excellent technical assistance of Gaby Klapdor and Michael Stennecken.
Received August 26, 1999; accepted January 19, 2000.
References
1.
Lamberts SWJ, van den Beld AW, van der Lely AJ.
The endocrinology of aging. Science. 1997;278:419–424.
2. Hsueh W, Law RE, Saad M, Dy J, Feener E, King G. Insulin resistance and macrovascular disease. Curr Opin Endocrinol Diabetes. 1996;3:346–354.
3. Cleland SJ, Petrie JR, Ueda S, Elliott HL, Connell JMC. Mechanisms and pathophysiological significance of insulin’s vascular action. Curr Opin Endocrinol Diabetes. 1998;5:217–222.
4.
Barrett-Connor E, Bush TL. Estrogen and
coronary heart disease in women. JAMA. 1991;265:1861–1867.
5. Lobo RA, ed. Treatment of the Postmenopausal Woman: Basic and Clinical Aspects. New York, NY: Raven Press; 1994.
6. Alexandersen P, Haarbo J, Christiansen C. The relationship of natural androgens to coronary heart disease in males: a review. Atherosclerosis. 1996;125:1–13.[Medline] [Order article via Infotrieve]
7. von Eckardstein A. Androgens, cardiovascular risk factors and atherosclerosis. In: Nieschlag E, Behre HM, eds. Testosterone: Action, Deficiency, Substitution. 2nd ed. Berlin, Germany: Springer Verlag; 1998:229–258.
8.
Janssen JAMJL, Stolk RP, Pols HAP, Grobee DE, Lamberts
SWJ. Serum total IGF-I, free IGF-I, and IGFBP-1 levels in an elderly
population: relation to cardiovascular risk factors and
disease. Arterioscler Thromb Vasc Biol. 1998;18:277–282.
9. Ferns GAA, Morani AS, Anggard EE. The insulin-like growth factors: their putative role in atherogenesis. Artery. 1991;18:197–225.[Medline] [Order article via Infotrieve]
10. Spallarosa P, Brunelli C, Minuto F, Caruso D, Battistini M, Caponnetto S, Cordera R. Insulin-like growth factor 1 and angiographically documented coronary artery disease. Am J Cardiol. 1996;336:200–202.
11. Rosen T, Bengtsson BA. Premature mortality due to cardiovascular disease in hypopituitarism. Lancet. 1990;336:285–288.[Medline] [Order article via Infotrieve]
12. Markussis V, Beshyah SA, Fischer C, Sharp P, Nicolaides AN, Johnston DG. Detection of premature atherosclerosis by high resolution ultrasonography in symptom-free hypopituitary adults. Lancet. 1992;340:1188–1192.[Medline] [Order article via Infotrieve]
13. Bänsch D, Dirkes-Kersting A, Schulte H, Assmann G, von Eckardstein A. Growth hormone is an important determinant of high density lipoprotein cholesterol levels especially in women. Metabolism. 1997:46:1039–1043.
14. Bänsch D, Chen-Haudenschild C, Dirkes-Kersting A, Schulte H, Assmann G, von Eckardstein A. Growth hormone is correlated with high density lipoprotein cholesterol and apolipoprotein A-I independently of insulin-like growth factor in women. Metabolism. 1998;47:339–344.[Medline] [Order article via Infotrieve]
15. Enbergs A, Dorszewski A, Luft M, Mönnig G, Kleemann A, Schulte H, Assmann G, Breithardt G, Kerber S. Failure to confirm ferritin and caeruloplasmin as risk factors for the extent of coronary arteriosclerosis. Coron Artery Dis. 1998;9:119–124.[Medline] [Order article via Infotrieve]
16. Sullivan DR, Marwick TH, Freedmann SB. A new method of scoring coronary angiograms to reflect extent of coronary atherosclerosis and improve correlation with major risk factors. Am Heart J. 1990;119:1262–1267.[Medline] [Order article via Infotrieve]
17. Friedewald WT, Levy J, Fredrickson DS. Estimation of the concentration of low-density-lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem. 1972;18:499–502.[Abstract]
18. Nie NH. SPSSX User’s Guide. New York, NY: McGraw-Hill; 1983.
19. Spagnoli A, Rosenfeld RG. Insulinlike growth factor binding proteins. Curr Opin Endocrinol Diabetes Mellitus. 1997;4:1–9.
20. Grant M. Insulin like growth factor 1. Curr Opin Endocrinol Diabetes Mellitus. 1996;3:335–345.
21.
Jones JI, Clemmons DR. Insulin-like growth factor and
their binding proteins: biologic actions. Endocr Rev. 1995;16:3–34.
22. Delafontaine P. Growth factors and vascular smooth muscle cell responses. Eur Heart J. 1998;19(suppl G):G18–G22.
23. Schriever C, Schmidt A, Breithardt G, Buddecke E. Human recombinant insulin-like growth factor-I and -II stimulate the expression of basic fibroblast growth factor but suppress the division of bovine coronary smooth muscle cells. Atherosclerosis. 1998;122:255–263.
24. Walsh MF, Barazi M, Pete G, Muniyappa R, Dunbar JC, Sowers JR. Insulin-like growth factor-1 diminishes in vivo and in vitro vascular contractility: role of vascular nitric oxide. Endocrinology. 1996;137:1798–1803.[Abstract]
25. Bang P. Serum proteolysis of IGFBP-3. Prog Growth Factor Res. 1995;6:285–292.[Medline] [Order article via Infotrieve]
26. Collet-Solberg PF, Cohen P. The role of the insulin-like growth factor binding proteins and the IGFBP proteases in modulating IGF action. Endocrinol Metab Clin North Am. 1996;25:591–614.[Medline] [Order article via Infotrieve]
27. Twigg SM, Baxter RC. Regulation of serum insulin-like growth factor bioavailability. Curr Opin Endocrinol Diabetes Mellitus. 1999;6:84–90.
28. Angelin B, Rudling M. Growth hormone and hepatic lipoprotein metabolism. Curr Opin Lipidol. 1994;5:160–165.[Medline] [Order article via Infotrieve]
29. Rudling M, Olivecrona H, Eggertsen G, Angelin B. Regulation of rat hepatic low density lipoprotein receptors: in vivo stimulation by growth hormone is not mediated by insulin-like growth factor. J Clin Invest. 1996;97:292–299.[Medline] [Order article via Infotrieve]
30.
Olivecrona H, Johannson AG, Lindh E, Ljungall S,
Berglund L, Angelin B. Hormonal regulation of serum lipoprotein(a)
levels: contrasting effects of growth hormone and insulin like growth
factor-1. Arterioscler Thromb Vasc Biol. 1995;15:847–849.
31. Bang P, Brismar K, Rosenfeld RG. Increased proteolysis of insulin-like growth factor binding protein-3 (IGFBP-3) in noninsulin-dependent diabetes mellitus serum with elevation of a 29-kilodalton (kDa) glycosylated IGFBP-3 fragment contained in the approximately 130- to 150 kDa ternary complex. J Clin Endocrinol Metab. 1994;78:1119–1127.[Abstract]
32. Després JP, Marette A. Relation of components of insulin resistance to coronary disease risk. Curr Opin Lipidol. 1994;5:274–289.[Medline] [Order article via Infotrieve]
33.
Phillips GB, Pinkernell BH, Jing TY. The association of
hypotestosteronemia with coronary artery disease in men.
Arterioscler Thromb. 1994;14:701–706.
This article has been cited by other articles:
 |
|
 |
|
  S.-i. Kawachi, N. Takeda, A. Sasaki, Y. Kokubo, K. Takami, H. Sarui, M. Hayashi, N. Yamakita, and K. Yasuda Circulating Insulin-Like Growth Factor-1 and Insulin-Like Growth Factor Binding Protein-3 Are Associated With Early Carotid Atherosclerosis Arterioscler. Thromb. Vasc. Biol., March 1, 2005; 25(3): 617 - 621. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Delafontaine, Y.-H. Song, and Y. Li Expression, Regulation, and Function of IGF-1, IGF-1R, and IGF-1 Binding Proteins in Blood Vessels Arterioscler. Thromb. Vasc. Biol., March 1, 2004; 24(3): 435 - 444. [Abstract] [Full Text]
|
|
|
|
 |
|
 F. C. W. Wu and A. von Eckardstein Androgens and Coronary Artery Disease Endocr. Rev., April 1, 2003; 24(2): 183 - 217. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||