| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Department of Endocrinology and Metabolism (P.V., A.P.H., L.M.) and the Institute of Experimental Clinical Research (H.Ø.), Aarhus University Hospital, DK-8000 Aarhus C, Denmark
Address all correspondence and requests for reprints to: Dr. Peter Vestergaard, Osteoporosis Clinic, Aarhus Amtssygehus, Tage Hansens Gade 2, DK-8000 Aarhus C, Denmark.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Five hundred and ninety-five perimenopausal women (median age, 50.0 yr; range, 45–56 yr) participating in the Danish Osteoporosis Prevention Study were investigated in a cross-sectional study, and a randomly selected subgroup of 110 was followed after 5 yr in a longitudinal study for changes in serum IGFs and BMD of lumbar spine, femoral neck, and ultradistal forearm during (n = 46) or without HRT (n = 64). In the cross-sectional study, serum IGF-I correlated positively to distal forearm BMD and spine BMD, but not to femoral neck BMD, after stratification for age, body mass index, and other variables. In the follow-up study, HRT decreased IGF-I and IGF-II, but did not influence the age-related decline in IGF-binding protein-3 significantly. Serum alkaline phosphatase and urinary hydroxyproline/creatinine ratio both decreased during HRT, whereas BMD increased compared to control values. After adjustment for age, body mass index, treatment, and other factors, IGF-I correlated positively to changes in forearm and femoral neck BMD, but not to changes in spine BMD.
We conclude that serum IGF-I was positively associated to bone mineral density. Oral HRT decreases IGF-I and IGF-II.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The interaction among estrogen, endogenous GH production, serum IGFs, and bone mineral is not fully elucidated (10). The accelerated perimenopausal bone loss is induced by the loss of endogenous estrogen production (1). An acute increase in circulating IGF-I and IGF-II has also been observed in premenopausal women who were made estrogen deficient through administration of GnRH agonists (11). Estrogen has a stimulatory effect on GH secretion, but decreases in endogenous GH production have been reported during short term estrogen administration in children (10).
In ovariectomized and estrogen-treated rats, circulating IGF-I correlates to cortical bone mass (12). In accordance with this, positive correlations have been found between serum IGF-I and forearm BMD in perimenopausal women (6) as well as in osteoporotic patients (5). Sugimoto et al. (5) also observed significantly lower serum values of IGF-I and IGF-binding protein-3 (IGFBP-3) in patients with osteoporotic spinal fractures than in control subjects. No correlation was found between IGFBP-2 and -3 and forearm BMD in perimenopausal women (6), whereas a positive correlation was observed between IGFBP-3 and forearm BMD in osteoporotic patients (5). Furthermore, perimenopausal women had a decline in serum IGF-I 1 yr after menopause with an increase in bone turnover markers (6). Frystyk et al. (13) demonstrated that total IGF-I was independent of body mass index (BMI), whereas IGF-II and IGFBP-3 both increased with increasing BMI. BMI per se is a predictor of high BMD (14). Furthermore, excessive alcohol intake, malnutrition, and liver disease reduce serum IGF-I levels in vivo (8).
In combination, these findings demonstrate a very complex interaction among age, endogenous estrogen and GH secretion, serum levels of IGFs, BMD, and various confounding factors. Increasing age is associated with a decline in GH secretion and BMD. GH deficiency lowers IGFs (15), which, in turn, increases after the administration of GH (15). The decline in IGF-I in postmenopausal women observed by Nasu et al. (6) in a cross-sectional study may at least in part be linked to the known age-related decrease in IGF-I (4).
The aim of the present study was to investigate the effects of estrogen on bone mass and IGFs in perimenopausal women in both a cross-sectional and a longitudinal design. We measured relations among IGFs, BMD, and serum estradiol in untreated perimenopausal women and changes in IGF-I; IGF-II; IGFBP-3; BMD of lumbar spine, femoral neck, and ultradistal forearm; and biochemical markers of bone turnover over a period of 5 yr in perimenopausal women who were allocated to hormone replacement therapy (HRT), or not. Several variables were considered for the analysis to control for potential confounders, as indicated above (16).
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Inclusion criteria were 1) women with intact uterus, aged 45–58 yr, and 3–24 months past last menstrual bleeding or experiencing perimenopausal symptoms (including irregular menstruations) combined with elevated serum FSH; and 2) hysterectomized women, aged 45–52 yr, with elevated FSH. Exclusion criteria were 1) metabolic bone diseases, including osteoporosis, defined as nontraumatic vertebral fractures on x-ray; 2) current estrogen use or estrogen use within the past 3 months; 3) current or past treatment with glucocorticoids for more than 6 months; 4) current or past malignancy; 5) newly diagnosed or uncontrolled chronic disease; or 6) alcohol or drug addiction.
The present cross-sectional study included 595 women attending 1 of the 4 Danish Osteoporosis Prevention Study centers. Of the 525 women attending the 5-yr follow-up visit, a subgroup of 110 randomly selected subjects (46 who had been continuously receiving the same type of HRT and 64 who had not received any type of HRT) were used to study longitudinal changes in serum IGF-I, IGF-II, and IGFBP-3 during HRT. None of these 110 women was taking any other medication.
Women with intact uteri received oral estradiol (2 mg/day) for the first 12 days of each 28-day cycle, estradiol (2 mg/day) and norethisterone acetate (1 mg/day) for the following 10 days, and estradiol (1 mg/day) for the last 6 days (Trisekvens, Novo Nordisk A/S, Glostrup, Denmark). Hysterectomized women received oral continuous estradiol (2 mg/day; Estrofem, Novo Nordisk).
At inclusion and after 5 yr, bone mineral content and BMD were measured in the lumbar spine (L2–L4), the femoral neck, and the ultradistal part of the nondominant forearm by Hologic DEXA scanners (Hologic, Inc., Waltham, MA) in all participants. At inclusion, measurements were performed by a 1000/w scanner, and at 5 yr, they were made by a 2000/w scanner. Cross-calibration was insured through the use of double measurements and a common phantom, as described in detail by Abrahamsen et al. (17). Precision for BMD was 1.5% for the lumbar spine, 2.1% for the femoral neck, and 1.9% for the ultradistal forearm. These figures included cross-over calibration, change of hardware, change of technicians, and long term stability. Long term stability was high, with changes of less than 0.2%/yr (17).
At inclusion and after 5 yr, serum albumin, serum albumin-adjusted calcium, serum total alkaline phosphatase, and were measured by standard laboratory methods in all participants. Urinary hydroxyproline (OHP) was measured by Hypronosticon (Organon Teknika, Holland) on a low collagen diet in a morning spot urine (second void). Urinary creatinine was measured by routine laboratory methods.
Serum levels of IGF-I, IGF-II, IGFBP-3, and intact PTH were measured at inclusion in all 595 subjects and in the subgroup of 110 subjects at 5 yr. Serum IGF-I and IGF-II were measured as previously described in detail (18). Acid-ethanol extracts were devoided of IGFBPs as assessed by Western ligand blot and immunoassays (IGFBP-1, -2, and -3), and results were identical to those obtained by acid-gel chromatography. Intraassay coefficients of variation (CVs) were less than 5%, and interassay CVs were less than 10%. The obtained specificities and sensitivities were high, with less than 0.0002% cross-reactivity in heterologous assays with recombinant human proinsulin, and recombinant human insulin cross-reacting less than 0.01% in the IGF-I assay. Serum IGFBP-3 was measured by an immunoradiometric assay (DSL-6600 Active, Diagnostics Systems Laboratories, Inc., Webster, TX), with intra- and interassay CVs of less than 2% and 4%, respectively. Serum intact PTH was measured by DPC Immulite (chemiluminescense) with an interassay CV of 11% and an intraassay CV of 6%. The ratio between IGF-I (or IGF-II) and IGFBP-3 was used as an index of free IGF, as IGF binds to IGFBP-3.
Furthermore, serum levels of estradiol (AutoDELFIA, Wallac OY, Turku, Finland; intraassay CV, 5.2%; interassay CV, 8.5%), osteocalcin [Bone Gla protein (BGP) by RIA (19); intraassay CV, 5%; interassay CV, 10%], and bone-specific alkaline phosphatase [BAP by lectin precipitation (19); intraassay CV, 8%; interassay CV, 25%] were measured at inclusion in all participants. Moreover, it was registered at inclusion whether the woman was a smoker or consumed alcohol on a daily basis. Family history of fractures and her daily intake of coffee and tea were also noted. Finally, the intake of calcium and vitamin D was recorded using a diet history method by a clinical dietitian. The CV for intake of calcium from one measurement to the next was 15%, and for vitamin D intake it was 28%.
Statistics
Normal distribution was ensured through logarithmic transformation of skewed parameters (serum estradiol, total alkaline phosphatase, IGF-I, IGF-II, IGFBP-3, and OHP). Groups were compared by t test for two samples and by paired t test when appropriate. Correlations with BMD were tested by multiple stepwise linear regression, with age, BMI, intake of alcohol, current smoking, BAP, BGP, urinary OHP, serum estradiol, serum PTH, serum IGF-I, serum IGF-II, and serum IGFBP-3 as independent variables. Treatment effect was included in the stepwise multiple linear regression in the analysis of changes from baseline to 5 yr. Bivariate correlations were examined using Pearson’s correlation coefficient. Median, mean, SD, and SD of the mean were used as descriptive statistics. All calculations were performed using SPSS 6.1.3 for Windows (SPSS, Inc., Chicago, IL).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Table 1
gives baseline
characteristics for all participants and the subgroup of 110 subjects
who had remeasurements of IGFs after 5 yr. Only the number of
hysterectomized women and the BAP deviated borderline significantly
between the two groups.
|
). Neither serum IGF-II nor serum
IGFBP-3 correlated significantly to regional BMD. At all three sites,
BMD correlated inversely to biochemical markers of bone turnover.
|
HRT for 5 yr decreased biochemical markers of bone turnover
and maintained BMD at all three sites (Table 3). HRT reduced serum levels of IGF-I and
IGF-II. Serum IGF-I decreased 21% [double-sided P value
(2P) < 0.001] and IGF-II 7.5% in the HRT-treated
group, whereas no changes were observed in the untreated group. Serum
IGFBP-3 decreased in treated (20%; 2P < 0.001) as
well as untreated (21%; 2P < 0.001) subjects without
any difference between the groups (P = 0.40). The
ratios between IGF-I and IGFBP-3 and between IGF-II and IGFBP-3
increased more in the untreated than in the treated groups in
accordance with the observation that total IGFs decreased in the
treated women. IGF-I decreased significantly more in hysterectomized
women treated with continuous estrogen (n = 6) than in women with
intact uterus (n = 40) treated with sequential estrogen plus
norethisterone (median change, -53.5 vs. -29.5;
2P = 0.043), whereas there were no significant
differences for IGF-II (-100.5 vs. -198.5;
2P = 0.22) or IGFBP-3 (-640.5 vs. -733.0;
2P = 0.58).
|
).
Correlations with IGF-I were positive in the femoral neck and
ultradistal forearm, meaning that the decrease in BMD was greater with
lower initial serum IGF-I levels. Exclusion of serum BGP, serum BAP,
and urinary OHP from the equations did not change the observed
correlations.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In the cross-sectional study, we found positive correlations between serum IGF-I and BMD in areas with a high proportion of trabecular bone, such as the spine and the ultradistal forearm, taking known confounders into account. These findings are in accordance with the studies by Nasu et al. (6) and Sugimoto et al. (5), who reported positive correlations between serum IGF-I and forearm BMD in perimenopausal women and osteoporotic patients. Contrary to Sugimoto et al. (5), we did not find a relationship between serum IGF-I and femoral neck BMD. However, the observed correlations were weak, and only a minor part of the total variation in BMD was explained by the applied models. The poor correlation between IGF-I and BMD may be the reason for the difference between our observations regarding femoral neck BMD and the findings of Sugimoto et al. (5). Another explanation would be that cross-sectional BMD measurements represent accumulated changes in net bone balance over many years and are only marginally influenced by actual serum IGF-I concentrations. In this setting recent serum IGF-I levels most likely will affect skeletal sites with abundant trabecular bone, such as the spine and the ultradistal forearm, with high bone turnover. Furthermore the production of IGF-I is much larger in the liver than in the bones (8), indicating that serum levels mostly reflect changes in the liver (8). The lack of correlation between BMD and serum IGF-II or serum IGFBP-3 may also be explained by the mentioned discrepancy in the time frame between the two types of measurements. The significant inverse relations observed between bone markers (BGP, BAP, and OHP) and BMD at all three measuring sites probably reflect that the rise in bone turnover after menopause is followed by an increase in the remodeling space and over a couple of months by a decrease in the amount of bone (20). Although cross-sectional spine BMD was linked to IGF-I, there was no correlation between IGF-I and the 5-yr change in spine BMD. A negative association between initial serum estradiol and change in spine BMD was observed. The difference between the cross-sectional and follow-up study could be explained by the hypothesis that IGF-I in estrogen-replete premenopausal women is of significance to peak bone mass (i.e. the amount of bone), whereas the estrogen deficiency plays a larger role in trabecular bone loss in perimenopause (1). On the other hand, IGF-I was linked to loss of BMD in the femoral neck in the follow-up study, as those with a low initial IGF-I lost more BMD than those with a high initial IGF-I, pointing toward a link between cortical bone turnover and IGF-I (4, 12).
In the 5-yr follow-up study, BMD increased significantly in the HRT group compared to no difference in the lumbar spine and the femoral neck, and was maintained in the ultradistal forearm, as seen in other studies (21). Furthermore, bone resorption and bone formation decreased in the HRT group, as estimated by total alkaline phosphatase and urinary OHP, which is in accordance with the results of other studies (22). Serum albumin-adjusted calcium increased in the untreated group compared to that in the HRT group without a significant difference in serum PTH levels between the groups. This supports previous findings that estrogen may influence calcium homeostasis by attenuating the effect of PTH (23).
Serum levels of IGF-I and IGF-II were both suppressed in the HRT group, whereas identical changes were observed in serum IGFBP-3 in HRT-treated and untreated subjects. It has been observed that oral (but not transdermal) estrogen decreases IGF-I, with a larger decline with increasing basal IGF-I (24). The reason for the reduction in serum total IGF-I and IGF-II during HRT is not fully elucidated. Estrogen inhibits hepatic IGF-I synthesis (8, 24). Furthermore, estrogen reduces bone turnover, which decreases the number of osteoclasts and osteoblasts (25). The reduction in bone cell number could inhibit the total skeletal synthesis of IGFs and thereby contribute to the decrease in their serum levels. The estrogen-induced reduction in hepatic IGF-I synthesis may further reduce serum IGF and thus indirectly bone cell activity by decreasing local IGF. However, the positive association between serum IGF-I and changes in femoral neck and ultradistal forearm BMD during HRT may indicate that serum IGF-I modulates bone remodeling by inducing a slightly more positive bone balance per remodeling cycle. Although bone turnover as well as average serum IGF-I diminish during HRT, individuals with the lower baseline IGF-I levels had the larger decrease in BMD. The higher serum total IGF-I levels at the 5-yr follow-up in the untreated subjects may reflect a compensatory regulation to mitigate the accelerated perimenopausal bone loss. However, theoretically this compensation may fail with time due to diminishing IGF-I production with increasing age, an increased need that cannot be met, or an increased resistance to IGFs in the perimenopause.
In conclusion, the cross-sectional study showed a positive correlation between serum IGF-I and BMD of spine and ultradistal forearm. Baseline serum IGF-I correlated positively with changes in BMD in the femoral neck and the ultradistal forearm.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received November 11, 1998.
Revised March 12, 1999.
Accepted April 1, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  Y. Bremnes, G. Ursin, N. Bjurstam, S. Rinaldi, R. Kaaks, and I. T. Gram Insulin-like Growth Factor and Mammographic Density in Postmenopausal Norwegian Women Cancer Epidemiol. Biomarkers Prev., January 1, 2007; 16(1): 57 - 62. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. M. Morimoto, P. A. Newcomb, E. White, J. Bigler, and J. D. Potter Variation in Plasma Insulin-Like Growth Factor-1 and Insulin-Like Growth Factor Binding Protein-3: Genetic Factors Cancer Epidemiol. Biomarkers Prev., June 1, 2005; 14(6): 1394 - 1401. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. M. Morimoto, P. A. Newcomb, E. White, J. Bigler, and J. D. Potter Insulin-like Growth Factor Polymorphisms and Colorectal Cancer Risk Cancer Epidemiol. Biomarkers Prev., May 1, 2005; 14(5): 1204 - 1211. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. P A Rooman, L. O. De Beeck, M. Martin, J. van Doorn, S. Mohan, and M. V L Du Caju Ethinylestradiol and testosterone have divergent effects on circulating IGF system components in adolescents with constitutional tall stature Eur. J. Endocrinol., April 1, 2005; 152(4): 597 - 604. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K.-C. Leung, G. Johannsson, G. M. Leong, and K. K. Y. Ho Estrogen Regulation of Growth Hormone Action Endocr. Rev., October 1, 2004; 25(5): 693 - 721. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. H. Gravholt, R. W. Naeraa, K. Brixen, K. W. Kastrup, L. Mosekilde, J. O. L. Jorgensen, and J. S. Christiansen Short-Term Growth Hormone Treatment in Girls With Turner Syndrome Decreases Fat Mass and Insulin Sensitivity: A Randomized, Double-Blind, Placebo-Controlled, Crossover Study Pediatrics, November 1, 2002; 110(5): 889 - 896. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Chang, X. Wu, H. Yu, and M. R. Spitz Plasma Concentrations of Insulin-like Growth Factors among Healthy Adult Men and Postmenopausal Women: Associations with Body Composition, Lifestyle, and Reproductive Factors Cancer Epidemiol. Biomarkers Prev., August 1, 2002; 11(8): 758 - 766. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. H. Gravholt, A. L. Lauridsen, K. Brixen, L. Mosekilde, L. Heickendorff, and J. S. Christiansen Marked Disproportionality in Bone Size and Mineral, and Distinct Abnormalities in Bone Markers and Calcitropic Hormones in Adult Turner Syndrome: A Cross-Sectional Study J. Clin. Endocrinol. Metab., June 1, 2002; 87(6): 2798 - 2808. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Rinaldi, H. Dechaud, C. Biessy, V. Morin-Raverot, P. Toniolo, A. Zeleniuch-Jacquotte, A. Akhmedkhanov, R. E. Shore, G. Secreto, A. Ciampi, et al. Reliability and Validity of Commercially Available, Direct Radioimmunoassays for Measurement of Blood Androgens and Estrogens in Postmenopausal Women Cancer Epidemiol. Biomarkers Prev., July 1, 2001; 10(7): 757 - 765. [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 |
