HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Poli, A. Articles by Crosignani, P. G. Search for Related Content PubMed PubMed Citation Articles by Poli, A. Articles by Crosignani, P. G.

Plasma Low-Density Lipoprotein Cholesterol and Bone Mass Densitometry in Postmenopausal Women

Andrea Poli, MD, Fiorenza Bruschi, MD, Bruno Cesana, MD, Monica Rossi, MD, Rodolfo Paoletti, MD and Pier Giorgio Crosignani, MD

From the Department of Pharmacological Sciences and 1st Department of Obstetrics and Gynecology, University of Milan; and Epidemiology Unit, Ospedale Maggiore di Milano, IRCCS, Milan, Italy.

Address reprint requests to: Andrea Poli, MD, Department of Pharmacological Sciences, University of Milan, Via Balzaretti, 9 – 20133 Milan, Italy; E-mail: poli.nfi{at}tin.it.

	ABSTRACT

TOP ABSTRACT PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
OBJECTIVE: The prevalence and the clinical and social importance of osteopenia and osteoporosis are increasing in western societies. To improve knowledge of the risk factors associated with these conditions, we assessed the relationship between bone mass density and plasma lipid profile in a cohort of postmenopausal women.

METHODS: We studied 1303 postmenopausal women who attended a menopause outpatient clinic. All women underwent bone mineral density determination at the level of the lumbar spine. Plasma lipids and lipoproteins and bone metabolic markers were assessed on a blood sample obtained after a 12-hour fast.

RESULTS: Statistically significant associations were found by univariate analysis between prevalence of osteopenia and age, time since menopause, body mass index, and low-density lipoprotein (LDL) cholesterol. Specifically, women with plasma LDL cholesterol levels of at least 160 mg/dL had a more than doubled probability of being osteopenic compared with women with lower LDL cholesterol (47.9% versus 21.2%, respectively). Time since menopause, body mass index, and LDL cholesterol were the only variables significantly associated with the prevalence of osteopenia, by multivariable analysis.

CONCLUSION: Postmenopausal women with increased plasma LDL cholesterol levels had a greater probability of being classified as osteopenic than women with normal plasma LDL cholesterol levels. Our data, if confirmed, suggest that elevated levels of plasma LDL cholesterol should be regarded as an additional risk factor for reduced bone mineral density.

Osteoporosis and osteoporosis-related fractures are a major problem in postmenopausal women. The identification of new risk factors associated with an increased probability of having a reduced bone mass is of interest and may help improve the identification of high-risk subjects and the implementation of preventive strategies. The existence of a possible link between plasma lipids and bone mineral density (BMD), in this context, has received limited attention. Data in the literature are scanty and contradictory. In 427 healthy men, Adami et al¹ found reduced bone mass to be associated with a favorable lipid profile, whereas in a large cohort of diabetic women, a negative association was found between low-density lipoprotein (LDL) cholesterol and BMD ( Koshiyama H, Wada Y, Nakamura Y. Hypercholesterolemia as a possible risk factor for osteopenia in type 2 diabetes mellitus [letter]. Arch Intern Med 2001;161:1678[Free Full Text]). Recently, in a Japanese cohort, Yamaguchi et al² found a positive association between high-density lipoprotein (HDL) cholesterol and BMD, whereas a negative association between LDL cholesterol and BMD was observed at some BMD determination sites. In a small Italian cohort, on the other hand, a negative association was found between HDL cholesterol and BMD.³ The aim of our study was to assess the relationship between BMD and plasma lipid profile in a large cohort of postmenopausal women.

	PATIENTS AND METHODS

TOP ABSTRACT PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
Women referred or presenting for menopausal symptoms to the menopause outpatient clinic of the 1st Obstetrics and Gynecology Department of the University of Milan from January 1999 to November 1999 were considered for admission to the study if the following criteria were met: menopausal status (defined as amenorrhea for a minimum of 12 months), age 45–65 years, plasma triglyceride levels less than 400 mg/dL, absence, for at least 6 months before clinical evaluation, of treatment with hormone replacement therapy (HRT) or with drugs affecting lipid metabolism or bone metabolism. In all, 1303 women met these criteria.

All women had a routine BMD measurement at the level of the lumbar spine (L2–L4) by dual-energy x-ray absorptiometry (Norland XR-26 [Mark II densitometer, Norland Scientific Instruments, Fort Atkinson, WI]; coefficient of variation of the determination 1.1%). Osteopenia was defined as a BMD value less than the lower value of the standardized reference interval (mean value minus one standard deviation) for healthy young adult women ⁴; osteoporosis was defined as a BMD value less than the mean minus two standard deviations.

Blood samples for the determination of lipids and bone markers were collected after a 12-hour fast. Serum total cholesterol, HDL cholesterol, and triglycerides were measured by enzymatic tests (Boehringer, Mannheim, Germany). LDL cholesterol was calculated from measured values by the Friedewald formula. LDL cholesterol values were classified as normal (129 mg/dL or below), moderately elevated (130–159 mg/dL), and elevated (160 mg/dL and over). Osteocalcin plasma levels were determined by a radioimmunoassay kit, using an antibody against human osteocalcin (CIS, Gif-sur-Yvette, France). Alkaline phosphatase was determined by colorimetric assay in accordance with a standardized method (Roche Diagnostic, Mannheim, Germany).

The study protocol was approved by our local ethics committee. Descriptive statistics were calculated (means, standard deviation, median, minimum, and maximum for quantitative variables and absolute and relative frequencies for qualitative ones); continuous variables were categorized in two or more classes as follows: age less than or equal to 50 years, 51–55 years, 56–60 years, and 61 or more years; time since last menstrual period 1, 2–6, and 7 or more years; body mass index (BMI, calculated as weight/height [m]²) as less than 25 and 25 or more; total cholesterol as 220 or less, 221–250 and 251 mg/dL or more; HDL cholesterol as 59 mg/dL or less and 60 mg/dL or greater; and LDL cholesterol as 129 or less, 130–159, and 160 or more mg/dL.

In order to reduce the probability of obtaining statistically significant correlations with very little or no clinical relevance, the approach of analyzing the continuous variables as categoric ones was chosen. The limits of the classes of the variables were selected for their clinical relevance.

The relationships between bone mass (dichotomized as osteopenic and nonosteopenic) and the categorized lipid metabolism variables were tested using univariate analysis by means of ² test (and of the ² test for trend when more than two classes were present) and by multivariable analysis using multiple logistic regression according to a backward procedure. Variables with more than two classes were included in the model by using dummy variables in order to obtain contrast against a reference class. Goodness of fit of the model was checked by means of the Hosmer-Lemeshow test.⁵ In addition, tests for overdispersion⁶ and graphical residual analysis were done.

Comparison among the three classes of LDL cholesterol were done using one-way analysis of variance followed by the Tukey test, with the significance level of .05 adjusted for multiple comparisons.

	RESULTS

TOP ABSTRACT PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
All women who met the entry criteria (n = 1303) were studied. The mean age of enrolled women was 54.2 ± 4.3 years (median 54 years; range 45–65 years), and the mean time since the last menstrual period was 4.8 ± 4.3 years (median 3 years; range 1–26 years). Mean BMI was 24.1 ± 3.5 (median 24; range 16–42). Table 1 shows the descriptive statistics of the main characteristics of the two classes of women without or with osteopenia. According to described criteria, 1086 women (83.4%) were considered nonosteopenic and 217 (16.6%) osteopenic (only 15 had a BMD value slightly less than the mean minus two standard deviations, being formally osteoporotic).

Osteocalcin levels and alkaline phosphatase plasma levels were significantly lower (both P < .005) in nonosteopenic women (mean ± standard deviation 8.4 ± 7.4 ng/mL and 173.8 ± 51.1 U/L, respectively) than in osteopenic women (10.6 ± 8.1 ng/mL and 184.9 ± 51.2 U/L, respectively).

Table 2 shows the distribution (absolute frequencies and percentages) of the osteopenic and nonosteopenic groups with regard to age, time since last menstrual period, BMI, and lipid parameters.

A statistically significant association emerged between LDL cholesterol levels and osteopenic status (P = .015); more specifically, women with LDL cholesterol levels of 160 mg/dL or more had a higher probability of being osteopenic than women with lower LDL cholesterol levels (47.9% versus 21.2%, respectively, P = .004 for the trend) (Figure 1). Conversely, neither total nor HDL cholesterol levels were statistically associated with osteopenia.

View larger version (22K): [in this window] [in a new window]   Figure 1. Distribution of bone mineral density based diagnosis of osteopenia and nonosteopenia in three predefined low-density lipoprotein classes. Poli. LDL Cholesterol and BMD in Menopause. Obstet Gynecol 2003.

Table 3 shows the odds ratio (OR) of being osteopenic, together with pertinent 95% confidence intervals, adjusted for the age groups, time since last menstrual period, BMI, and LDL cholesterol values. Women with LDL cholesterol levels of at least 160 mg/dL have an OR of 1.74 (95% CI 1.18, 2.60) to be classified as osteopenic compared with women with LDL cholesterol less than 130 mg/dL.

Finally, mean BMD was statistically significantly higher in the women with elevated and moderately elevated LDL cholesterol values (>160 mg/dL and 130–159 mg/dL) than in the women with normal LDL cholesterol. No other significant differences among the classes of women with different plasma LDL cholesterol levels were found.

	DISCUSSION

TOP ABSTRACT PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
Our data show that postmenopausal women with increased plasma LDL cholesterol levels have a significantly greater probability of being osteopenic, based on a lumbar spine BMD determination, compared with women with normal LDL cholesterol. No significant association with HDL cholesterol levels was found.

Our data are in good accordance with those published by Koshiyama and colleagues ( Koshiyama H, et al. Arch Intern Med 2001;161:1678[Free Full Text]) obtained in diabetic women and by Yamaguchi and colleagues,² who identified a negative association between LDL cholesterol and BMD at the level of the radius and, as a nonsignificant trend, at the lumbar spine, together with a positive association between high-density lipoprotein (HDL) cholesterol and BMD, evaluated at the level of radius and of the lumbar spine. In our patients, on the other hand, the protective effect of HDL cholesterol on BMD could not be confirmed. Our data, conversely, are different from those published by Adami and colleagues¹ who, in healthy men, found an opposite pattern (ie, increased bone mass with increasing plasma LDL cholesterol values).

Our data can be explained by a direct negative effect of plasma LDL cholesterol on bone metabolism. Such an effect could be attributed, in accordance with Parhami et al,^7,8 to a negative effect of LDL cholesterol particles, possibly in their oxidized form, on bone cell activity, favoring bone catabolism rather than bone formation and accumulation. Because metabolic indexes of bone synthesis (osteocalcin and alkaline phosphatase) are also increased in women with high LDL cholesterol, compared with women with normal LDL cholesterol levels, we hypothesize a markedly increased bone turnover in these patients.

The association between increased LDL cholesterol levels and propensity to osteopenia could also be relevant to the discussion of the purported anti-osteoporotic effect of statins. The favorable effect of these drugs on bone formation, shown in some observational studies but not in others,^9–13 might not, or at least not only, be linked to a direct effect of statins on bone cells,¹⁴ but rather, or also, to their effect on plasma LDL cholesterol levels. This would be of practical relevance, because if it were true, not only lipophylic statins (which can cross cell membranes) but also hydrophylic ones (which cannot cross cell membranes and therefore cannot enter the osteoblastic cell) might favorably affect bone metabolism and, eventually, fracture risk.

Our study, being observational in design, cannot prove the existence of a causal relationship between plasma LDL cholesterol and BMD. The relationship could, theoretically, be not causal. For example, it is theoretically possible that women with high LDL cholesterol, who are aware of their plasma lipid condition, could have reduced their intake of milk and dairy products in the previous decades of their life, and that this reduced intake of foods rich in highly absorbable calcium may have contributed to their reduced bone mass. This interpretation is unlikely, however, because only rarely, at least in our country, do women with slightly increased plasma LDL cholesterol levels follow a hypolipidemic diet before menopause.

Our data show that postmenopausal women with increased plasma LDL cholesterol levels have a significantly greater probability, compared with women with normal LDL cholesterol, to be classified as osteopenic based on a lumbar spine BMD determination. These data may also be useful for properly interpreting the relationship between statin drug use and osteoporosis and in planning further clinical studies in this area.

	Footnotes

 
Fiorenza Bruschi died suddenly on March 13, 2002.

doi:10.1016/j.obstetgynecol.2003.07.004

Received February 4, 2003. Received in revised form June 30, 2003. Accepted July 31, 2003.

	REFERENCES

TOP ABSTRACT PATIENTS AND METHODS RESULTS DISCUSSION REFERENCES

 
1. Adami S, Braga V, Gatti D. Association between bone mineral density and serum lipids in men. JAMA 2001;286: 791–2.[Free Full Text]

2. Yamaguchi T, Sugimoto T, Yamauchi M, Sowa H, Chen Q, Chihara K. Plasma lipids and osteoporosis in postmenopausal women. Endocr J 2002;49:211–7.[Medline]

3. D’Amelio P, Pescarmona GP, Gariboldi A, Isaia GC. High density lipoproteins (HDL) in women with postmenopausal osteoporosis: A preliminary study. Menopause 2001;8:429–32.[Medline]

4. World Health Organization. Assessment of fracture risk and its application to screening for postmenopausal osteoporosis. Report of a WHO Study Group. WHO Tech Rep Series 1994;843:1–129.

5. Hosmer WD, Lemeshow S. Applied logistic regression. New York: Wiley & Sons, Inc., 1989.

6. Williams DA. Extra-binomial variation in logistic linear models. Appl Stat 1982;31:144–8.

7. Parhami F, Garfinkel A, Demer LL. Role of lipids in osteoporosis. Arterioscler Thromb Vasc Biol 2000;20: 2346–8.[Abstract/Free Full Text]

8. Parhami F, Jackson SM, Le V, Balucan JP, Territo M, et al. Atherogenic diet and minimally oxidized low density lipoprotein inhibit osteogenic and promote adipogenic differentiation of marrow stromal cells. J Bone Miner Res 1999;14:2067–78.[Medline]

9. Edwards CJ, Hart DJ, Spector TD. Oral statins and increased bone-mineral density in postmenopausal women. Lancet 2000;355:2218–9.[Medline]

10. Chan KA, Andrade SE, Boles M, Buist DS, Chase GA, Donahue JG, et al. Inhibitors of hydroxymethylglutaryl-coenzyme a reductase and risk of fracture among older women. Lancet 2000;355:2185–8.[Medline]

11. Meier CR, Schlienger RG, Kraenzlin ME, Schlegel B, Jick H. HMG-CoA reductase inhibitors and the risk of fractures. JAMA 2000;283:3205–10.[Abstract/Free Full Text]

12. Wang PS, Solomon DH, Mogun H, Avorn J. HMG-CoA reductase inhibitors and the risk of hip fractures in elderly patients. JAMA 2000;283:3211–6.[Abstract/Free Full Text]

13. Pasco JA, Kotowicz MA, Henry MJ, Sanders KM, Nicholson GC. Statin use, bone mineral density, and fracture risk: Geelong osteoporosis study. Arch Intern Med 2002; 162:537–40.[Abstract/Free Full Text]

14. Mundy G, Garrett R, Harris S, Chan J, Chen D, Rossini G, et al. Stimulation of bone formation in vitro and in rodents by statins. Science 1999;286:1946–9.[Abstract/Free Full Text]

This article has been cited by other articles:

				P. Hamel, E. Abed, L. Brissette, and R. Moreau Characterization of oxidized low-density lipoprotein-induced hormesis-like effects in osteoblastic cells Am J Physiol Cell Physiol, April 1, 2008; 294(4): C1021 - C1033. [Abstract] [Full Text] [PDF]

				E.M. Dennison, H.E. Syddall, A. Aihie Sayer, H.J. Martin, C. Cooper, and for the Hertfordshire Cohort Study Group Lipid profile, obesity and bone mineral density: the Hertfordshire Cohort Study QJM, May 1, 2007; 100(5): 297 - 303. [Abstract] [Full Text] [PDF]

				Y.-H. Hsu, S. A Venners, H. A Terwedow, Y. Feng, T. Niu, Z. Li, N. Laird, J. D Brain, S. R Cummings, M. L Bouxsein, et al. Relation of body composition, fat mass, and serum lipids to osteoporotic fractures and bone mineral density in Chinese men and women Am. J. Clinical Nutrition, January 1, 2006; 83(1): 146 - 154. [Abstract] [Full Text] [PDF]

				A. H. Eliassen, G. A. Colditz, B. Rosner, W. C. Willett, and S. E. Hankinson Serum Lipids, Lipid-Lowering Drugs, and the Risk of Breast Cancer Arch Intern Med, October 24, 2005; 165(19): 2264 - 2271. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Poli, A. Articles by Crosignani, P. G. Search for Related Content PubMed PubMed Citation Articles by Poli, A. Articles by Crosignani, P. G.