JOURNAL OF BONE AND MINERAL RESEARCH Volume 7, Number 1, 1992 Mary Ann Liebert, Inc., Publishers
Bone Mass and Body Composition in Normal Women ROBERT LINDSAY
FELICIA COSMAN,',' BRENDA S. HERRINGTON,' and SHARON HIMMELSTEIN'
ABSTRACT The interrelationships between measurements of bone mass and total-body bone mineral were examined in a cross-sectional study of normal healthy women aged 17-82 years. In addition we evaluated the relationship between measures of body composition, estimated by four independent techniques, and bone mass in the same population. Considering the group as a whole, bone mass at all sites correlated with each other and with total-body bone mineral (TBBM). Cancellous and cortical sites could predict TBBM equally well. As expected, all measurements of bone mass were significantly lower in postmenopausal women in comparison to premenopausal women. Declines in bone mass were only seen in premenopausal women in the femoral neck and Ward's triangle, not in lumbar spine, radius, or skeleton as a whole. In postmenopausal women bone mass correlated negatively with age and years from menopause equally at all sites. TBBM was significantly related to height and weight in both premenopausal and postmenopausal women. In premenopausal and postmenopausal women TBBM also correlated with fat mass, but TBBM was much better correlated with percentage body fat in premenopausal than postmenopausal women. TBBM was a constant proportion of lean body mass in premenopausal women, but the fraction of lean mass occupied by the skeleton declined with age in postmenopausal women. Correction of TBBM for lean mass did not change the relationship between TBBM and percentage fat in premenopausal women but eliminated the relationship in postmenopausal women. Regional measurements, which are at least partially corrected for body size by dividing mass by area, correlated less well with height and weight and with any index of obesity, especially in postmenopausal women.
INTRODUCTION
ometry (DPA) and the more recent development, dual-energy x-ray absorptiometry (DEXA), on the other hand, allow evaluation of both regional bone mass and totalbody bone mineral (TBBM) and are more widely available. Several groups have published cross-sectional data on TBBM in women using DPA.'6-8)Only one of these studies, however, had a sufficiently large sample of premenopausal women to suggest that TBBM was stable with age until ovarian Data on regional skeletal bone loss in women are conflicting, with some studies demonstrating significant loss at some sites before menopause, but not at others. (9-13) There are therefore discrepancies in the literature that have never been evaluated with measurements of both the entire skeleton and comprehensive standard regional measurements performed in the same individuals.
in bone mass with age among healthy individuals have involved measurements of regional areas of the skeleton, initially cortical peripheral bones, such as midshaft of the radius, and more recently the axial skeleton, especially the spine and femoral Only a few groups have evaluated total-body mineral content. (4-LI) Initially, total-body calcium measurements were obtained by neutron activation analysis and demonstrated a marked postmenopausal decline, ( 4 ) almost double that seen premenopausally, although only a few patients fell into the latter category. This technique, however, is associated with a high radiation dose and is not available for general clinical use.(s) Dual-photon absorpti-
M
OST STUDIES ON CHANGES
'Regional Bone Center, Helen Hayes Hospital, West Haverstraw, New York. 'Department of Medicine, College of Physicians and Surgeons, Columbia University, New York.
55
LINDSAY ET AL.
56
We have completed a detailed evalution of this issue in a cross-sectional study of bone mass (regional and total) in healthy adult pre- and postmenopausal women. We also evaluated the relationships among skeletal sites, TBBM, lean body mass (LBM), and body fat to evaluate the relative importance of body weight or fat mass in determining skeletal status in both pre- and postmenopausal women.
year, postmenopausal estrogens for more than 6 months, diuretics, steroids, and anticonvulsants, were all excluded. All subjects gave written informed consent before the study, which was approved by the institutional review boards of Helen Hayes Hospital and the New York State Department of Health.
Experimental methods
METHODS Experimental subjects Measurements of bone mass and body composition were made in 150 healthy volunteers solicited by telephone from the local community. All women completed a detailed medical questionnaire under supervision. Women with osteoporotic fractures, chronic medical disorders, or use of medications likely to interfere with skeletal or mineral homeostasis, including oral contraceptives for more than 1
TBBM and bone mass in the lumbar spine and femoral neck were measured by DPA using a Norland 2600 (Norland Corp., Fort Atkinson, WI) for TBBM and Lunar DP3 (Lunar Radiation Corp., Madison, WI) for regional measurements. Corrections were applied for alterations in source strength as already described.'") Bone mass in the heel (0s calcis) was assessed by isolating the heel regions of both feet from the total-body scan and calculating the area occupied by mineral. This measurement is an estimate of
CHARACTERISTICS~ TABLE1 . DESCRIPTIVE Premenopausal (n = 74)
Age, years Height, cm Weight, kg Menarche, years No. of pregnancies No. of births Duration of lactation, months
39.65 f 1.16 (17-54) 163.07 f 0.91 (1 37.16-1 85.42) 63.34 f 1.15 (43.18-89.55) 12.60 f 0.18 (9-17) 1.73 f 0.20 (0-6) 1.33 f 0.15 (0-4) 2.53 f 0.70 (0-36) -
Years postmenopause aMean
Postmenopausal (n = 76)
59.34 f 0.96 (36-82) 160.15 f 0.89 (45.45-102.73) 64.49 f 1.27 (45.45- 102.73) 12.55 f 0.18 (8-18) 3.05 f 0.21 (0-9) 2.67 f 0.18 (0-8) 2.79 f 0.79 (0-32) 9.67 f 1.16 ( 1-46)
Differences p < o.OOo1 NS
NS
NS p < o.OOo1
p < o.OOo1
NS -
* standard error of the mean (SEM), ranges are shown in parentheses.
TABLE2. BONEMASSMEASUREMENTS IN NORMAL WOMENa Premenopausal (n = 74)
TBBM, g Spine, g/cm2 Femoral neck, g/cm2 Ward's triangle, g/cm2 Trochanter, g/cm2 Distal radius, g Proximal radius, g 0 s calcis, g/cm2 aMean
SEM
2496.50 1.19 0.85 0.76 0.71 36.04 37.11 1.17
f 43.62 f 0.02
f 0.01 f f f f f
0.02 0.01 0.69 0.60 0.02
Postmenopausal (n Vi 76)
p Value
2213.90 f 47.67 1.05 f 0.02 0.76 f 0.01 0.63 f 0.02 0.65 f 0.01 29.55 f 1 . 1 1 30.68 f 1.08 1.08 f 0.02
Bone Mass and Body Composition in Normal Women ROBERT LINDSAY
FELICIA COSMAN,',' BRENDA S. HERRINGTON,' and SHARON HIMMELSTEIN'
ABSTRACT The interrelationships between measurements of bone mass and total-body bone mineral were examined in a cross-sectional study of normal healthy women aged 17-82 years. In addition we evaluated the relationship between measures of body composition, estimated by four independent techniques, and bone mass in the same population. Considering the group as a whole, bone mass at all sites correlated with each other and with total-body bone mineral (TBBM). Cancellous and cortical sites could predict TBBM equally well. As expected, all measurements of bone mass were significantly lower in postmenopausal women in comparison to premenopausal women. Declines in bone mass were only seen in premenopausal women in the femoral neck and Ward's triangle, not in lumbar spine, radius, or skeleton as a whole. In postmenopausal women bone mass correlated negatively with age and years from menopause equally at all sites. TBBM was significantly related to height and weight in both premenopausal and postmenopausal women. In premenopausal and postmenopausal women TBBM also correlated with fat mass, but TBBM was much better correlated with percentage body fat in premenopausal than postmenopausal women. TBBM was a constant proportion of lean body mass in premenopausal women, but the fraction of lean mass occupied by the skeleton declined with age in postmenopausal women. Correction of TBBM for lean mass did not change the relationship between TBBM and percentage fat in premenopausal women but eliminated the relationship in postmenopausal women. Regional measurements, which are at least partially corrected for body size by dividing mass by area, correlated less well with height and weight and with any index of obesity, especially in postmenopausal women.
INTRODUCTION
ometry (DPA) and the more recent development, dual-energy x-ray absorptiometry (DEXA), on the other hand, allow evaluation of both regional bone mass and totalbody bone mineral (TBBM) and are more widely available. Several groups have published cross-sectional data on TBBM in women using DPA.'6-8)Only one of these studies, however, had a sufficiently large sample of premenopausal women to suggest that TBBM was stable with age until ovarian Data on regional skeletal bone loss in women are conflicting, with some studies demonstrating significant loss at some sites before menopause, but not at others. (9-13) There are therefore discrepancies in the literature that have never been evaluated with measurements of both the entire skeleton and comprehensive standard regional measurements performed in the same individuals.
in bone mass with age among healthy individuals have involved measurements of regional areas of the skeleton, initially cortical peripheral bones, such as midshaft of the radius, and more recently the axial skeleton, especially the spine and femoral Only a few groups have evaluated total-body mineral content. (4-LI) Initially, total-body calcium measurements were obtained by neutron activation analysis and demonstrated a marked postmenopausal decline, ( 4 ) almost double that seen premenopausally, although only a few patients fell into the latter category. This technique, however, is associated with a high radiation dose and is not available for general clinical use.(s) Dual-photon absorpti-
M
OST STUDIES ON CHANGES
'Regional Bone Center, Helen Hayes Hospital, West Haverstraw, New York. 'Department of Medicine, College of Physicians and Surgeons, Columbia University, New York.
55
LINDSAY ET AL.
56
We have completed a detailed evalution of this issue in a cross-sectional study of bone mass (regional and total) in healthy adult pre- and postmenopausal women. We also evaluated the relationships among skeletal sites, TBBM, lean body mass (LBM), and body fat to evaluate the relative importance of body weight or fat mass in determining skeletal status in both pre- and postmenopausal women.
year, postmenopausal estrogens for more than 6 months, diuretics, steroids, and anticonvulsants, were all excluded. All subjects gave written informed consent before the study, which was approved by the institutional review boards of Helen Hayes Hospital and the New York State Department of Health.
Experimental methods
METHODS Experimental subjects Measurements of bone mass and body composition were made in 150 healthy volunteers solicited by telephone from the local community. All women completed a detailed medical questionnaire under supervision. Women with osteoporotic fractures, chronic medical disorders, or use of medications likely to interfere with skeletal or mineral homeostasis, including oral contraceptives for more than 1
TBBM and bone mass in the lumbar spine and femoral neck were measured by DPA using a Norland 2600 (Norland Corp., Fort Atkinson, WI) for TBBM and Lunar DP3 (Lunar Radiation Corp., Madison, WI) for regional measurements. Corrections were applied for alterations in source strength as already described.'") Bone mass in the heel (0s calcis) was assessed by isolating the heel regions of both feet from the total-body scan and calculating the area occupied by mineral. This measurement is an estimate of
CHARACTERISTICS~ TABLE1 . DESCRIPTIVE Premenopausal (n = 74)
Age, years Height, cm Weight, kg Menarche, years No. of pregnancies No. of births Duration of lactation, months
39.65 f 1.16 (17-54) 163.07 f 0.91 (1 37.16-1 85.42) 63.34 f 1.15 (43.18-89.55) 12.60 f 0.18 (9-17) 1.73 f 0.20 (0-6) 1.33 f 0.15 (0-4) 2.53 f 0.70 (0-36) -
Years postmenopause aMean
Postmenopausal (n = 76)
59.34 f 0.96 (36-82) 160.15 f 0.89 (45.45-102.73) 64.49 f 1.27 (45.45- 102.73) 12.55 f 0.18 (8-18) 3.05 f 0.21 (0-9) 2.67 f 0.18 (0-8) 2.79 f 0.79 (0-32) 9.67 f 1.16 ( 1-46)
Differences p < o.OOo1 NS
NS
NS p < o.OOo1
p < o.OOo1
NS -
* standard error of the mean (SEM), ranges are shown in parentheses.
TABLE2. BONEMASSMEASUREMENTS IN NORMAL WOMENa Premenopausal (n = 74)
TBBM, g Spine, g/cm2 Femoral neck, g/cm2 Ward's triangle, g/cm2 Trochanter, g/cm2 Distal radius, g Proximal radius, g 0 s calcis, g/cm2 aMean
SEM
2496.50 1.19 0.85 0.76 0.71 36.04 37.11 1.17
f 43.62 f 0.02
f 0.01 f f f f f
0.02 0.01 0.69 0.60 0.02
Postmenopausal (n Vi 76)
p Value
2213.90 f 47.67 1.05 f 0.02 0.76 f 0.01 0.63 f 0.02 0.65 f 0.01 29.55 f 1 . 1 1 30.68 f 1.08 1.08 f 0.02
