hfurutitas, 1 (1979) 175-181 0 Elsevier/North-Holland Biomedical Press
175
AGEATMENOPAUSEANDITSRELATIONTOOSTEOPOROSIS
OLOF LINDQUIST, CALLE BENGTSSON, TOMMY HANSSON and BENGT ROOS Department of Medicine II, Orthopedics Giiteborg, Sweden
(Received 25 July 1978, accepted
I and Radiophysics,
16 October
University of Gateborg,
1978)
As one phase of a comprehensive population study of women, bone density was determined in the third lumbar vertebra by dual photon absorptiometric technique. The method involves the use of two radionuclides, which both emit gamma radiation but with different energies (*41Am with 59.6 keV and t 3 ‘Cs with 662 keV). Women in three age strata were studied: 46,S4 and 62 yr. All women aged 46 were still menstruating, while the age strata 54 and 62 were subdivided into two groups: those who still menstruated or had been postmenopausal for a short time and those who had been postmenopausal for many years. The bone mineral content was higher in premenopausal or recently postmenopausal S4-yr-old women than in women of the same age who had been postmenopausal for a long time (F’< 0.01). A similar trend was found for women aged 62 (F’< 0.10). The differences could not be explained by differences in other factors studied, such as body weight, body height, smoking habits or physical activity. No significant differences were found when women with similar menopausal status in different age groups were compared. (Key words: Menopause, Menopausal age, Bone density, Osteoporosis,
Population study)
INTRODUCTION During recent years there has been much discussion about the cause and nature of osteoporosis. Osteoporosis has usually been regarded as a disease mainly affecting women over the age of 50. Albright et al. [l] attributed this finding to a low formation rate due to gonadal insufficiency. In more recent investigations, however, most authors support the hypothesis that the cause of bone loss in osteoporosis is an increased bone resorption in the presence of a basically normal rate of formation [2-61. The concept of bone loss with increasing age is well known and generally accepted. Controversy exists, however, about when it starts, the rate at which it proceeds and whether sex differences exist. In view of the controversy on this subject, we decided to determine bone density in women near menopausal age as one phase of a comprehensive population study of women.
Correspondence to: Olof Lindquist, Goteborg, Sweden.
M.D., Department
of Medicine, dstra
Sjukhuset,
S-416
85
176 TABLE I Numbers of participants and participation 1968-1969 and 1974-1975.
rates in the population
1968-1969
4%
1974-1975 n
(yrl
38 46 50 54 60 Total
studies of women in Goteborg in
372 431 398 180 81 1462
Participation rate (%I
Age (yrl
n
91.4 90.1 91.0 88.6 83.5 90.1
44 52 56 60 66 Total
336 387 351 163 65 1302
Participation rate (%I * 90.3 89.8 88.2 90.6 80.2 89.1
* Of those studied in 1968-1969.
SUBJECTS
A population study of women was carried out in 1968-1969 in Gateborg, Sweden 171. A random sample of women in five age strata (women born in 1908, 1914, 1918, 1922 and 1930) was studied. The same population sample of women was re-studied in 1974-1975 [8]. On this second occasion 1302 participated, corresponding to 80.2% of those initially sampled in 1968 and 89.1% of those who participated in 1968-1969 (Table I). Women in three of these age strata (women born in 1914, 1922 and 1930) participated later in an absorptiometric study, which was carried out during the first few months of 1976 (Table II). The participants in the absorptiometric study were at that time 46, 54 TABLE II Age and number of participants and non-participants 1976 in relation to menopausal status in 1974-1975. Group
I IIA IIB IIIA BIB Total
Age (yrl
46 54 54 62 62
Called for examination
30 50 30 33 30 170
in the absorptiometric
study carried out during
Participants PremenopausaI 20 38 0 0
Total Postmenopausal 14 yr
-
-
-
25 25 22
20’ 38 25 25 23 * 130
* One of these women had to be excluded because of a technical measurement
error.
Nonparticipants
Participation rate (%I
10 12 5 8 7 40
67 76 83 76 73 76
and 62 yr of age. The women born in 1930 (group I) who were selected for the absorptiometric study consisted of a systematic subsample of women who were still menstruating when examined in 1974-1975. Those born in 1922 consisted of one systematic subsample of women who were still menstruating in 1974-1975 (group IIA) and another group of women comprising all those who had been postmenopausal for 28 yr (group IIB). The women born in 1914 consisted of one group comprising all women who had been postmenopausal for G3 yr (group IIIA) and another group comprising all women who had been postmenopausal for >I4 yr (group IIIB). Numbers of participants and participation rates are presented in Table II. The overall participation rate in the sub. sample was 76%. One woman in group IA, four in group IIA and one in group IIIB were treated with oestrogens at the time of the study performed in 1974-1975. A comparison was made between participants and non-participants with respect to body height, body weight, calcium in serum, and smoking habits. No significant differences were found.
Fig. 1. Diagram showing measuring procedure. A = scintillation crystal, B = lead collimator for the detector, C = patient’s direction of movement, D = scale for demonstrating the successive measuring positions, E = polythene mattress, F = couch, G = ball bearing, H = container for the sources with collimation of emitted radiation beam, I = radiation sources. Distances are stated in mm.
METHODS
Information about menopausal status, number of pregnancies, smoking habits, and physical activity was obtained by interview and information about menopausal status and number of pregnancies was also obtained by means of a questionnaire. Body height and body weight were measured accurately to one decimal place. The bone mineral content was determined in the third lumbar vertebra by dual photon absorptiometry [9,10], The method involves the use of two radionuclides which emit gamma radiation with different energies (241Am, with 59.6 keV and 13’Cs with 662 keV). The radiation sources are so arranged that their gamma radiation passes the object to be measured in a common collimated radiation beam (Fig. 1). The transmitted gamma radiation is registered digitally with a scintillation detector both energies simultaneously. Transmission measurements are performed during preselected time intervals. Both photon energies are exponentially attenuated by the object, the higher energy mainly by the Compton effect and the lower by the Compton and photo-electric effect. In materials with higher atomic number than soft tissue, e.g., bone mineral, the photo-electric effect will dominate in the attenuation of the lower energy. This means that the lower energy will be relatively more attenuated than the higher when
Distance (cm) Fig. 2. Bone profile curve showing N as a function of measuring position X for a normal subject.
Ii'9
it passes a region containing bone. Application of the law of exponential attenuation of the two energies allows calculation of the amount of bone mineral at each point along the measuring path, irrespective of the amount of soft tissue. The bone mineral values are plotted as a function of position, thus giving a bone profile curve (Fig. 2) which can be integrated over a baseline between end-points on either side of the vertebra. The resulting area is proportional to the bone mineral content in g. cm-r. The reproducibility of the method is considered to be better than 6%. A more detailed description of the theory and experimental technique has been presented elsewhere [9,10]. RESULTS
Bone mineral content in relation to age and menopausal status In Table III bone mineral content is presented in relation to age and menopausal status. In the vertical columns of the table the menopausal status is kept as constant as possible. No signiticant age differences were found in bone mineral content when studied in this way. Between 46 and 54 yr of age the mean values indicate an average loss of bone mineral content per 8 yr of 0.06 g . cm-‘, compared to 0.18 g * cm-’ between 54 and 62 yr of age in women who were premenopausal or had late menopause (groups I, IIA and IIIA in Table III). The difference per 8 yr between women aged 54 and 62 who had had early menopause (groups IIB and IIIB in Table III) was found to be 0.15 g . cm-‘. In the horizontal columns a comparison is made between premenopausal and postmenopausal women of the same age and between women of the same age with late or early menopause. In the age-group 54 a statistically significant difference in bone mineral content was found when comparing women of different menopausal status (P < 0.02). A clear tendency in the same direction was found when the same distinction was made in the age-group 62 (P < 0.10). Correction of bone mineral content values within the different age groups by relating them to body surface area yielded no differences of the results noted above.
TABLE III Bone mineral content (g. cm-l) in relation to age and menopausal status in premenopausal (women aged 46 and 54) and postmenopausal women (women aged 54) and in women with late and early menopause (women aged 62). Age (yr)
46 54 62
Group
I DA IIIA
Premenopausal (late menopausal)
Group
n
Mean
SD
20 38 25
3.91 3.85 3.67
0.65 0.62 0.76
IIB IIIB
statistical significance
Postmenopausal (early menopausal) n
Mean
SD
25 22
3.46 3.31
0.49 0.40
P < 0.02 P < 0.10
180
Bone mineral content in relation to body height and body weight
When comparing women of different body height no obvious tendency. with respect to the values of bone mineral content was found either in the premenopausal or in the postmenopausal women. However, we found a reduction in bone mineral content in women within similar ranges of body height when postmenopausal. We also found a tendency towards higher values for bone mineral content in obese women, irrespective of menopausal status. Bone mineral content in relation to smoking habits, number of pregnancies, and physical acitivity
A faster bone loss among smokers has been reported previously [l]. In the present study we have not been able to demonstrate lower values of bone mineral content in female smokers compared to non-smokers of the same age and menopausal status. However, when comparing premenopausal women or women with late menopause with early menopausal women of the same age we found a more pronounced decrease in bone mineral content in those with early menopause compared to premenopausal women or women with late menopause and this difference was most obvious in smokers. In the women examined, we found no definite influences on the bone mineral content of number of pregnancies or parity. However, women with early menopause tended to have lower values of bone mineral content compared to premenopausal women or women with late menopause. When bone mineral content was related to physical activity, no significant differences were found. However, when comparing women of the same age and activity group but of differing menopausal status we found a tendency towards lower values for bone mineral content in postmenopausal women. DISCUSSION
When comparing women of the same age but of differing menopausal status we found significantly lower values in women with early menopause in each group studied. Our results thus support the hypothesis that there is an association between early menopause and increased bone loss. When the influence of body weight and body height on bone mineral content in this study is assessed, the most striking finding is the lower values for bone mineral content in the postmenopausal women compared to premenopausal women, without exception and irrespective of body weight and body height. From this study it also seems probable that menopausal status is of greater importance for development of osteoporosis than smoking habits, number of pregnancies, and the degree of physical activity. REFERENCES [l] Albright, F., Smith, P.H. and Richardsson, A.M. (1941) Postmenopausal osteoporosis: its clinical features. J. Am. Med. Assoc. 116,2465-2474. [ 21 Heany, R.P. and Whedon, G.D. (1958) Radiocalcium studies of bone formation rate in human metabolic bone disease. 1. Ciin. Endocrinol. 18,1246.
[ 3j Nordin, B.E.C. (1959) Investigation of bone metabolism with calcium4’. A preliminary report. Proc. Roy. Sot. Med. 52, 351-353. [4] Nordin, B.E.C. (1961) Pathogenesis of osteoporosis. Lancet 1, 1011-1015. [S] Frost, H.M. and Villanuea, A.R. (1961) Human osteoblastic activity. I: A comprehensive method of measurement with some results. Henry Ford Hosp. Med. Bull. 9,76-86. [6] Frost, H.M. (1966) Bone dynamics in osteoporosis and osteomalacia. CC. Thomas. Springfield, Ill., U.S.A., pp. 74-85. [7] Bengtsson, C., Blohme, G., Hallberg, L., Hallstrom, T., Jsaksson, B., Korsan-Bengtsen, K., Rybo, G., Tibblin, E., Tibblin, G. and Westerberg, H. (1973) The study of women in Gothenburg 19681969 - a population study. General design, purpose and sampling results. Acta Med. Stand. 193, 311-318. [ 81 Bengtsson, C., Hallberg, L., Hallstrom, T., Hultborn, A., Isaksson, B., Lennartsson, J., Lindquist, O., Lindstedt, S., Noppa, H., Redvall, L. and Samuelsson, S. (1978) The population study of women in Giiteborg 1974-1975 - the second phase of a longitudinal study. General design, purpose and sampling results. Stand. J. Sot. Med. 6, 49-54. [9] Roos, B.L. and Skoldborn, H. (1974) Dual photon absorptiometry in lumbar vertebrae. I: Theory and method. Acta Radio]. 3, 266-280. [ 10 J Roos, B.O. (1975) Dual photon absorptiometry in lumbar vertebrae. II: Precision and reproducibility. Acta Radiol. 3, 291-303. [ 111 Daniel], H.W. (1976) Osteoporosis of the slender smokers. Vertebral compression fractures and loss of metacarpal cortex in relation to postmenopausal cigarette smoking and lack of obesity. Arch. Int. Med. 136,298-304.
175
AGEATMENOPAUSEANDITSRELATIONTOOSTEOPOROSIS
OLOF LINDQUIST, CALLE BENGTSSON, TOMMY HANSSON and BENGT ROOS Department of Medicine II, Orthopedics Giiteborg, Sweden
(Received 25 July 1978, accepted
I and Radiophysics,
16 October
University of Gateborg,
1978)
As one phase of a comprehensive population study of women, bone density was determined in the third lumbar vertebra by dual photon absorptiometric technique. The method involves the use of two radionuclides, which both emit gamma radiation but with different energies (*41Am with 59.6 keV and t 3 ‘Cs with 662 keV). Women in three age strata were studied: 46,S4 and 62 yr. All women aged 46 were still menstruating, while the age strata 54 and 62 were subdivided into two groups: those who still menstruated or had been postmenopausal for a short time and those who had been postmenopausal for many years. The bone mineral content was higher in premenopausal or recently postmenopausal S4-yr-old women than in women of the same age who had been postmenopausal for a long time (F’< 0.01). A similar trend was found for women aged 62 (F’< 0.10). The differences could not be explained by differences in other factors studied, such as body weight, body height, smoking habits or physical activity. No significant differences were found when women with similar menopausal status in different age groups were compared. (Key words: Menopause, Menopausal age, Bone density, Osteoporosis,
Population study)
INTRODUCTION During recent years there has been much discussion about the cause and nature of osteoporosis. Osteoporosis has usually been regarded as a disease mainly affecting women over the age of 50. Albright et al. [l] attributed this finding to a low formation rate due to gonadal insufficiency. In more recent investigations, however, most authors support the hypothesis that the cause of bone loss in osteoporosis is an increased bone resorption in the presence of a basically normal rate of formation [2-61. The concept of bone loss with increasing age is well known and generally accepted. Controversy exists, however, about when it starts, the rate at which it proceeds and whether sex differences exist. In view of the controversy on this subject, we decided to determine bone density in women near menopausal age as one phase of a comprehensive population study of women.
Correspondence to: Olof Lindquist, Goteborg, Sweden.
M.D., Department
of Medicine, dstra
Sjukhuset,
S-416
85
176 TABLE I Numbers of participants and participation 1968-1969 and 1974-1975.
rates in the population
1968-1969
4%
1974-1975 n
(yrl
38 46 50 54 60 Total
studies of women in Goteborg in
372 431 398 180 81 1462
Participation rate (%I
Age (yrl
n
91.4 90.1 91.0 88.6 83.5 90.1
44 52 56 60 66 Total
336 387 351 163 65 1302
Participation rate (%I * 90.3 89.8 88.2 90.6 80.2 89.1
* Of those studied in 1968-1969.
SUBJECTS
A population study of women was carried out in 1968-1969 in Gateborg, Sweden 171. A random sample of women in five age strata (women born in 1908, 1914, 1918, 1922 and 1930) was studied. The same population sample of women was re-studied in 1974-1975 [8]. On this second occasion 1302 participated, corresponding to 80.2% of those initially sampled in 1968 and 89.1% of those who participated in 1968-1969 (Table I). Women in three of these age strata (women born in 1914, 1922 and 1930) participated later in an absorptiometric study, which was carried out during the first few months of 1976 (Table II). The participants in the absorptiometric study were at that time 46, 54 TABLE II Age and number of participants and non-participants 1976 in relation to menopausal status in 1974-1975. Group
I IIA IIB IIIA BIB Total
Age (yrl
46 54 54 62 62
Called for examination
30 50 30 33 30 170
in the absorptiometric
study carried out during
Participants PremenopausaI 20 38 0 0
Total Postmenopausal 14 yr
-
-
-
25 25 22
20’ 38 25 25 23 * 130
* One of these women had to be excluded because of a technical measurement
error.
Nonparticipants
Participation rate (%I
10 12 5 8 7 40
67 76 83 76 73 76
and 62 yr of age. The women born in 1930 (group I) who were selected for the absorptiometric study consisted of a systematic subsample of women who were still menstruating when examined in 1974-1975. Those born in 1922 consisted of one systematic subsample of women who were still menstruating in 1974-1975 (group IIA) and another group of women comprising all those who had been postmenopausal for 28 yr (group IIB). The women born in 1914 consisted of one group comprising all women who had been postmenopausal for G3 yr (group IIIA) and another group comprising all women who had been postmenopausal for >I4 yr (group IIIB). Numbers of participants and participation rates are presented in Table II. The overall participation rate in the sub. sample was 76%. One woman in group IA, four in group IIA and one in group IIIB were treated with oestrogens at the time of the study performed in 1974-1975. A comparison was made between participants and non-participants with respect to body height, body weight, calcium in serum, and smoking habits. No significant differences were found.
Fig. 1. Diagram showing measuring procedure. A = scintillation crystal, B = lead collimator for the detector, C = patient’s direction of movement, D = scale for demonstrating the successive measuring positions, E = polythene mattress, F = couch, G = ball bearing, H = container for the sources with collimation of emitted radiation beam, I = radiation sources. Distances are stated in mm.
METHODS
Information about menopausal status, number of pregnancies, smoking habits, and physical activity was obtained by interview and information about menopausal status and number of pregnancies was also obtained by means of a questionnaire. Body height and body weight were measured accurately to one decimal place. The bone mineral content was determined in the third lumbar vertebra by dual photon absorptiometry [9,10], The method involves the use of two radionuclides which emit gamma radiation with different energies (241Am, with 59.6 keV and 13’Cs with 662 keV). The radiation sources are so arranged that their gamma radiation passes the object to be measured in a common collimated radiation beam (Fig. 1). The transmitted gamma radiation is registered digitally with a scintillation detector both energies simultaneously. Transmission measurements are performed during preselected time intervals. Both photon energies are exponentially attenuated by the object, the higher energy mainly by the Compton effect and the lower by the Compton and photo-electric effect. In materials with higher atomic number than soft tissue, e.g., bone mineral, the photo-electric effect will dominate in the attenuation of the lower energy. This means that the lower energy will be relatively more attenuated than the higher when
Distance (cm) Fig. 2. Bone profile curve showing N as a function of measuring position X for a normal subject.
Ii'9
it passes a region containing bone. Application of the law of exponential attenuation of the two energies allows calculation of the amount of bone mineral at each point along the measuring path, irrespective of the amount of soft tissue. The bone mineral values are plotted as a function of position, thus giving a bone profile curve (Fig. 2) which can be integrated over a baseline between end-points on either side of the vertebra. The resulting area is proportional to the bone mineral content in g. cm-r. The reproducibility of the method is considered to be better than 6%. A more detailed description of the theory and experimental technique has been presented elsewhere [9,10]. RESULTS
Bone mineral content in relation to age and menopausal status In Table III bone mineral content is presented in relation to age and menopausal status. In the vertical columns of the table the menopausal status is kept as constant as possible. No signiticant age differences were found in bone mineral content when studied in this way. Between 46 and 54 yr of age the mean values indicate an average loss of bone mineral content per 8 yr of 0.06 g . cm-‘, compared to 0.18 g * cm-’ between 54 and 62 yr of age in women who were premenopausal or had late menopause (groups I, IIA and IIIA in Table III). The difference per 8 yr between women aged 54 and 62 who had had early menopause (groups IIB and IIIB in Table III) was found to be 0.15 g . cm-‘. In the horizontal columns a comparison is made between premenopausal and postmenopausal women of the same age and between women of the same age with late or early menopause. In the age-group 54 a statistically significant difference in bone mineral content was found when comparing women of different menopausal status (P < 0.02). A clear tendency in the same direction was found when the same distinction was made in the age-group 62 (P < 0.10). Correction of bone mineral content values within the different age groups by relating them to body surface area yielded no differences of the results noted above.
TABLE III Bone mineral content (g. cm-l) in relation to age and menopausal status in premenopausal (women aged 46 and 54) and postmenopausal women (women aged 54) and in women with late and early menopause (women aged 62). Age (yr)
46 54 62
Group
I DA IIIA
Premenopausal (late menopausal)
Group
n
Mean
SD
20 38 25
3.91 3.85 3.67
0.65 0.62 0.76
IIB IIIB
statistical significance
Postmenopausal (early menopausal) n
Mean
SD
25 22
3.46 3.31
0.49 0.40
P < 0.02 P < 0.10
180
Bone mineral content in relation to body height and body weight
When comparing women of different body height no obvious tendency. with respect to the values of bone mineral content was found either in the premenopausal or in the postmenopausal women. However, we found a reduction in bone mineral content in women within similar ranges of body height when postmenopausal. We also found a tendency towards higher values for bone mineral content in obese women, irrespective of menopausal status. Bone mineral content in relation to smoking habits, number of pregnancies, and physical acitivity
A faster bone loss among smokers has been reported previously [l]. In the present study we have not been able to demonstrate lower values of bone mineral content in female smokers compared to non-smokers of the same age and menopausal status. However, when comparing premenopausal women or women with late menopause with early menopausal women of the same age we found a more pronounced decrease in bone mineral content in those with early menopause compared to premenopausal women or women with late menopause and this difference was most obvious in smokers. In the women examined, we found no definite influences on the bone mineral content of number of pregnancies or parity. However, women with early menopause tended to have lower values of bone mineral content compared to premenopausal women or women with late menopause. When bone mineral content was related to physical activity, no significant differences were found. However, when comparing women of the same age and activity group but of differing menopausal status we found a tendency towards lower values for bone mineral content in postmenopausal women. DISCUSSION
When comparing women of the same age but of differing menopausal status we found significantly lower values in women with early menopause in each group studied. Our results thus support the hypothesis that there is an association between early menopause and increased bone loss. When the influence of body weight and body height on bone mineral content in this study is assessed, the most striking finding is the lower values for bone mineral content in the postmenopausal women compared to premenopausal women, without exception and irrespective of body weight and body height. From this study it also seems probable that menopausal status is of greater importance for development of osteoporosis than smoking habits, number of pregnancies, and the degree of physical activity. REFERENCES [l] Albright, F., Smith, P.H. and Richardsson, A.M. (1941) Postmenopausal osteoporosis: its clinical features. J. Am. Med. Assoc. 116,2465-2474. [ 21 Heany, R.P. and Whedon, G.D. (1958) Radiocalcium studies of bone formation rate in human metabolic bone disease. 1. Ciin. Endocrinol. 18,1246.
[ 3j Nordin, B.E.C. (1959) Investigation of bone metabolism with calcium4’. A preliminary report. Proc. Roy. Sot. Med. 52, 351-353. [4] Nordin, B.E.C. (1961) Pathogenesis of osteoporosis. Lancet 1, 1011-1015. [S] Frost, H.M. and Villanuea, A.R. (1961) Human osteoblastic activity. I: A comprehensive method of measurement with some results. Henry Ford Hosp. Med. Bull. 9,76-86. [6] Frost, H.M. (1966) Bone dynamics in osteoporosis and osteomalacia. CC. Thomas. Springfield, Ill., U.S.A., pp. 74-85. [7] Bengtsson, C., Blohme, G., Hallberg, L., Hallstrom, T., Jsaksson, B., Korsan-Bengtsen, K., Rybo, G., Tibblin, E., Tibblin, G. and Westerberg, H. (1973) The study of women in Gothenburg 19681969 - a population study. General design, purpose and sampling results. Acta Med. Stand. 193, 311-318. [ 81 Bengtsson, C., Hallberg, L., Hallstrom, T., Hultborn, A., Isaksson, B., Lennartsson, J., Lindquist, O., Lindstedt, S., Noppa, H., Redvall, L. and Samuelsson, S. (1978) The population study of women in Giiteborg 1974-1975 - the second phase of a longitudinal study. General design, purpose and sampling results. Stand. J. Sot. Med. 6, 49-54. [9] Roos, B.L. and Skoldborn, H. (1974) Dual photon absorptiometry in lumbar vertebrae. I: Theory and method. Acta Radio]. 3, 266-280. [ 10 J Roos, B.O. (1975) Dual photon absorptiometry in lumbar vertebrae. II: Precision and reproducibility. Acta Radiol. 3, 291-303. [ 111 Daniel], H.W. (1976) Osteoporosis of the slender smokers. Vertebral compression fractures and loss of metacarpal cortex in relation to postmenopausal cigarette smoking and lack of obesity. Arch. Int. Med. 136,298-304.
