- Email: [email protected]
Healthy elderly individuals do not inevitably lose bone density and weight as they age
Archives of Gerontology and Geriatrics 39 (2004) 283–290
Healthy elderly individuals do not inevitably lose bone density and weight as they age Hugh Taggart a , David Craig b,∗ , Karen McCoy a a b
Department of Health Care for the Elderly, Belfast City Hospital, Lisburn Road, Belfast BT9 6BA, UK Department of Geriatric Medicine, Queen’s University of Belfast, 97 Lisburn Road, Belfast BT9 7BL, UK Received 6 January 2004; received in revised form 29 April 2004; accepted 6 May 2004 Available online 28 July 2004
Abstract There have been relatively few extended longitudinal studies of bone mineral density (BMD) and anthropometric characteristics in the elderly. We recorded baseline measurements (height, weight, body mass index (BMI)) and BMD in 292 relatively healthy individuals (mean age: 75 years females, 76 years males) who were followed up over 8 years. At baseline, a significant inverse relationship was noted between increasing age and BMD at the neck of femur in both sexes. In female subjects, baseline age was inversely related to BMD at the total hip site. There was a significant positive relationship between baseline weight/BMI in both sexes and BMD at the total hip and neck of femur. Weight and BMI were positively related to BMD at the spine in female participants. After a mean follow-up period for males and females of 7.6 and 8.0 years, respectively, there was a significant fall in height for male and female subjects although weight and BMI did not significantly vary in either sex. Follow-up hip BMD fell marginally but significantly in female subjects only. A significant relationship existed between change in hip BMD and change in weight in females. Measurements of the neck of femur did not change significantly on follow-up in either sex. While cross-sectional measurements show weight and age are related to bone density in elderly people, longitudinal evaluation over 8 years failed to demonstrate consistent reductions in males and females. In addition, there was good maintenance of BMD at most sites. © 2004 Elsevier Ireland Ltd. All rights reserved. Keywords: Bone mineral density; Body weight; Body mass index; Aged; Longitudinal study
∗
Corresponding author. Tel.: +44 2890 272202; fax: +44 2890 325839. E-mail address: [email protected] (D. Craig).
0167-4943/$ – see front matter © 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.archger.2004.05.001
284
H. Taggart et al. / Archives of Gerontology and Geriatrics 39 (2004) 283–290
1. Introduction An estimated 3 million people in the UK suffer from osteoporosis which carries a lifetime risk of osteoporosis-related fracture above 40% (Cooper et al., 1992). The risk of osteoporotic fracture in elderly people is strongly related to bone density (Marshall et al., 1996; Schott et al., 1998; Fujiwara et al., 2003) which in turn is strongly related to body weight (Burger et al., 1998; Dargent-Molina et al., 2000). There have been several longitudinal studies of bone mineral density (BMD) in the elderly but these have been mainly <5 years duration (Greenspan et al., 1994; Jones et al., 1994; Ensrud et al., 1997; Burger et al., 1998; Schott et al., 1998; Hannan et al., 2000; Tromp et al., 2000; Trivedi and Khaw, 2001; Uusi-Rasi et al., 2001; Fujiwara et al., 2003). It is thought that BMD declines linearly over an extended period in the elderly but there is relatively little data to support this view. Prospective studies of weight change of the elderly are rare, often restricted to under 75 and of limited follow-up (Lehmann and Bassey, 1996; Rasmussen et al., 2003; Sheehan et al., 2003). These studies together with large cross-sectional reports suggest weight loss in older age is normal, albeit dependant on relative changes in fat mass and muscle bulk (Gillette-Guyonnet et al., 2003; Zamboni et al., 2003). We recorded baseline characteristics of 292 individuals who were followed up over 8 years in order to measure the change in BMD and its relationship to anthropometric characteristics such as weight, height and BMI.
2. Materials and methods 2.1. Study population Subjects were recruited initially as part of a study to measure normal ranges for BMD in an elderly UK population. They were excluded if they had a history of low trauma fracture or steroid use. All were ambulatory, mentally alert and free of serious disease. A cohort of 181 women and 111 men was recruited and written informed consent obtained. The baseline examination was carried out in 1991 and comprised of an interview followed by measurement of BMD. Information was obtained on age, sex, height and weight. After a mean period of 7.6 years (men) and 8.0 years (women) all participants were invited for follow-up assessments. 2.2. Measurement of BMD At the first visit, BMD of the lumbar spine (L2–L4), neck of femur and total hip were measured (g/cm3 ) by DXA using a Hologic 1000 bone densitometer (coefficient of variation: 0.38%). At follow-up visit these sites were rescanned using a Hologic 4500A bone densitometer (coefficient of variation: 0.34%). In order to ensure the comparability of the two scanners, 54 subjects had spine scans done on both machines. The correlation coefficient between the spine scans on each machine was 0.994.
H. Taggart et al. / Archives of Gerontology and Geriatrics 39 (2004) 283–290
285
2.3. Data analysis The percentage change in BMD, weight and height at follow-up was measured where available. Determination of fracture occurrence was obtained at interview and confirmed in cases of doubt through contact with the subject’s general practitioner. Men and women were analysed separately. Differences in baseline characteristics were compared using Student’s t-test for continuous variables and chi square for categorical variables. Pearson’s correlation coefficient was applied to relationship analysis. All calculations were done using the SPSS statistical package.
3. Results 3.1. Baseline measurements Mean age, height and weight at baseline are shown in Table 1. The relationships between baseline sex, age and weight/BMI and baseline BMD at each site scanned are shown in Table 2. A significant inverse relationship is seen between increasing age and measured BMD at the neck of femur in both sexes. In female subjects, baseline age was inversely related to BMD at the total hip site. There was a significant positive relationship between
Table 1 Baseline mean values for age, height, weight, BMI and BMD at each scanned site
Age (years) Height (cm) Weight (kg) BMI (kg/m2 ) Total hip BMD (g/cm2 ) Neck of femur BMD (g/cm2 ) Spinal BMD (g/cm2 )
Female (no. of subjects)
Male (no. of subjects)
74.9 (64–94) (n = 181) 157.5 (116.3–178.0) (n = 181) 60.5 (37.2–99.0) (n = 177) 24.4 (15.0–51.7) (n = 177) 0.701 (0.236–1.025) (n = 177) 0.608 (0.380–0.878) (n = 177) 0.825 (0.397–1.537) (n = 181)
75.9 (65–92) (n = 111) 170.7 (149.8–198.0) (n = 111) 73.7 (45.4–105.0) (n = 105) 25.2 (19.7 – 33.5) (n = 105) 0.917 (0.585–1.378) (n = 109) 0.769 (0.468–1.280) (n = 109) 1.015 (0.629–1.776) (n = 111)
Table 2 Relationship between baseline age and weight/BMI and baseline BMD at each site scanned Baseline characteristic
Neck of femur −0.399∗∗
Total hip −0.391∗∗
Lumbar spine
Female
Age Weight BMI
r= r = 0.353∗∗ r = 0.285∗∗
r= r = 0.399∗∗ r = 0.333∗∗
NS r = 0.367∗∗ r = 0.342∗∗
Male
Age Weight BMI
r = −0.189∗∗ r = 0.287∗∗ r = 0.263∗∗
NS r = 0.312∗∗ r = 0.280∗∗
NS NS NS
NS: non-significant correlation. ∗∗ Denotes correlation is significant at the 0.01 level (two-tailed).
286
H. Taggart et al. / Archives of Gerontology and Geriatrics 39 (2004) 283–290
baseline weight/BMI in both sexes and measured BMD at the hip and neck of femur. Increasing weight/BMI was related to increasing BMD at the spine in female participants. 3.2. Follow-up measurements Height, weight/BMI and BMD data for individuals available for follow-up is shown in Table 3. The mean follow-up period for males and females was 7.6 and 8.0 years, respectively. There was a significant fall in height in male and female subjects although a significant rise in spinal BMD was noted in all participants. Weight and BMI did not significantly change in either sex during the follow-up period. Measurements at the neck of femur did not change upon follow-up in either sex. Hip BMD fell significantly in female subjects only. A significant relationship existed in females between change in hip BMD and change in weight during the follow-up period (Fig. 1). Nineteen females and nine males alive at follow-up but unable to attend had mean baseline BMD measurements at the spine and total hip not significantly different to the follow-up population. 3.3. Mortality
BMD Hip Change During Follow-up (g/cm2)
Forty-nine of the 111 men (43%) died after a mean follow-up period of 7.6 years. Fifty-one of the 181 women (28%) died after a mean follow-up period of 8.0 years. This represents a significantly higher death rate (P < 0.02) in females. There was no significant difference between death rates (male or female) in our cohort and those estimated from the general population when analysed with reference to life expectancy tables for Northern Ireland.
.3
.2
.1
0.0
-.1
Rsq = 0.1732
-20
-10
0
10
Weight Change During Follow-up (kg) Fig. 1. Change in weight vs. BMD at the hip.
20
Variable
Baseline (mean)
At follow-up (mean)
% Change
% Annual change
% Subjects with an increase
% Subjects with a decrease
Females
Height (n = 97) Weight (n = 109) BMI (n = 95) Neck of femur (n = 94) Total hip (n = 94) Lumbar spine (n = 94)
158.2 61.7 25.0 0.637 0.745 0.840
155.4∗∗∗ 61.3 26.0 0.620 0.721∗∗∗ 0.870∗∗∗
−1.76 −0.57 4.04 −2.56 −3.17 +3.65
−0.22 −0.07 0.51 −0.32 −0.40 +0.46
13.4 51.4 66.0 28.7 35.1 70.2
75.5 47.7 34.0 55.3 59.6 21.3
Males
Height (n = 52) Weight (n = 52) BMI (n = 51) Neck of femur (n = 53) Total hip (n = 53) Lumbar spine (n = 53)
170.7 73.9 25.3 0.762 0.911 0.993
168.9∗∗∗ 73.6 25.7 0.774 0.914 1.039∗∗∗
−1.03 −0.47 +1.76 +1.55 +0.32 +4.59
−0.14 −0.06 +0.23 +0.20 +0.04 +0.60
13.5 50.0 62.7 54.7 52.8 69.8
76.9 46.2 37.3 32.1 39.6 26.4
∗∗∗
Denotes significant change for a paired samples t-test at the P < 0.001 level (two-tailed).
H. Taggart et al. / Archives of Gerontology and Geriatrics 39 (2004) 283–290
Table 3 Height, weight/BMI and BMD data for individuals available for follow-up with calculated variable and subject change
287
288
H. Taggart et al. / Archives of Gerontology and Geriatrics 39 (2004) 283–290
3.4. Fractures Thirty-four (19%) females, but only one male (1%) suffered at least one clinically apparent fracture during the follow-up period (P < 0.005). Six females had a hip fracture, three spinal fractures and 12 forearm fractures. In 17 women the fracture was at other sites.
4. Discussion This study of 292 elderly individuals surveyed at baseline and around 8 years later is one of the longest longitudinal studies reported. At initial assessment, BMD was lower in older individuals at the neck of femur in both sexes and at the hip in females. A similar relationship between weight/BMI and BMD was seen in both sexes. Upon follow-up, a significant fall in BMD at the hip but not femoral neck was observed in females. Males showed non-significant increases in BMD at hip and neck of femur after follow-up. Spinal BMD increased when measured at follow-up in both sexes. The observation of reduced BMD with age is well documented although few studies have restricted their analysis to the older old. The cross-sectional analysis of the OFELY cohort of over 1000 women described accelerated bone loss at most skeletal sites after the age of 75 years (Arlot et al., 1997). Our observed rate of bone loss at hip sites (−0.31 to −0.40% per year) is lower than the published rates of bone loss of −0.82% at the femoral neck and −1.30% at Wards triangle in a cross-sectional study of over 7000 elderly women (Steiger et al., 1992) and all site 3.4–4.8% per 4 year loss in women followed longitudinally in the Framingham Osteoporosis Study (Hannan et al., 2000). As observed in other studies, preservation of body weight significantly reduced measured bone loss and may explain up to a third of the decline in BMD with age (Steiger et al., 1992; Felson et al., 1993; May et al., 1994). The preservation of weight in this cohort is worth noting. Previous studies of weight change in the elderly report reductions in weight during the seventh decade in males and the eighth decade in females (Svanborg et al., 1981; Lehmann and Bassey, 1996; Gillette-Guyonnet et al., 2003) with falls in BMI apparent between 70 and 85 years (Rossman, 1977; Waaler, 1984). BMI was seen to fall with age in a recent large study of elderly Swedish participants especially those with a smoking history (Rasmussen et al., 2003). A smaller study of limited follow-up found instead significant increases in BMI in both men and women aged >68, with increases in percentage fat in women and decreases in percentage muscle mass in men (Zamboni et al., 2003). Males were seemingly able to preserve BMD with a trend towards increased BMD noted on follow-up at the hip and neck of femur. Large studies of bone density in males show lower BMD when compared to females in cross-section and over time with the decline in BMD per year approximately half the rate of similarly aged female subjects (Hannan et al., 2000; Berntsen et al., 2001). Spinal bone density rose significantly in both sexes during follow-up. Scoliosis, osteophytosis and bony fusion all serve to artefactually increase spinal BMD. As reported in the Rotterdam study, high measured BMD may obscure accelerated bone loss beyond the arthritic and disused joint (Burger et al., 1996).
H. Taggart et al. / Archives of Gerontology and Geriatrics 39 (2004) 283–290
289
These data demonstrate that healthy individuals do not inevitably lose bone mineral density nor suffer weight loss. Height loss is more common although this may be subject to the reliability of measurement.
References Arlot, M.E., Sornay-Rendu, E., Garnero, P., Vey-Marty, B., Delmas, P.D., 1997. Apparent pre- and post-menopausal bone loss evaluated by DXA at different skeletal sites in women: the OFELY cohort. J. Bone Miner. Res. 12, 683–690. Berntsen, G.K., Fonnebo, V., Tollan, A., Sogaard, A.J., Magnus, J.H., 2001. Forearm bone mineral density by age in 7,620 men and women: the Tromso study, a population-based study. Am. J. Epidemiol. 153, 465–473. Burger, H., Van Daele, P.L., Odding, E., 1996. Association of radiographically evident osteoarthritis with higher bone mineral density and increased bone loss with age. The Rotterdam Study. Arthritis Rheum. 39, 81–86. Burger, H., De Laet, C.E., Van Daele, P.L., 1998. Risk factors for increased bone loss in an elderly population: the Rotterdam Study. Am. J. Epidemiol. 147, 871–879. Cooper, C., Campion, G., Melton III, L.J., 1992. Hip fracture in the elderly: a worldwide projection. Osteoporos. Int. 2, 285–289. Dargent-Molina, P., Poitiers, F., Breart, G., EPIDOS Group., 2000. In elderly women weight is the best predictor of a very low bone mineral density: evidence from the EPIDOS study. Osteoporos. Int. 11, 881–888. Ensrud, K.E., Lipschutz, R.C., Cauley, J.A., Seeley, D., Nevitt, M.C., Scott, J., Orwoll, E.S., Genant, H.K., Cummings, S.R., 1997. Body size and hip fracture risk in older women: a prospective study. Study of Osteoporotic Fractures Research Group. Am. J. Med. 103, 274–280. Felson, D.T., Zhang, Y., Hannan, M.T., Anderson, J.J., 1993. Effects of weight and body mass index on bone mineral density in men and women: the Framingham study. J. Bone Miner. Res. 8, 567–573. Fujiwara, S., Kasagi, F., Masunari, N., Naito, K., Suzuki, G., Fukunaga, M., 2003. Fracture prediction from bone mineral density in Japanese men and women. J. Bone Miner. Res. 18, 1547–1553. Gillette-Guyonnet, S., Nourhashemi, F., Andrieu, S., Cantet, C., Albarede, J.L., Vellas, B., Grandjean, H., 2003. Body composition in French women +75 years of age: the EPIDOS study. Mech. Age. Dev. 124, 311–316. Greenspan, S.L., Maitland, L.A., Myers, E.R., Krasnow, M.B., Kido, T.H., 1994. Femoral bone loss progresses with age: a longitudinal study in women over age 65. J. Bone Miner. Res. 9, 1959–1965. Hannan, M.T., Felson, D.T., Dawson-Hughes, B., Tucker, K.L., Cupples, L.A., Wilson, P.W., Kiel, D.P., 2000. Risk factors for longitudinal bone loss in elderly men and women: the Framingham Osteoporosis Study. J. Bone Miner. Res. 15, 10–20. Jones, G., Nguyen, T., Sambrook, P., Kelly, P.J., Eisman, J.A., 1994. Progressive loss of bone in the femoral neck in elderly people: longitudinal findings from the Dubbo osteoporosis epidemiology study. Br. Med. J. 309, 691–695. Lehmann, A.B., Bassey, E.J., 1996. Longitudinal weight changes over four years and associated health factors in 629 men and women aged over 65. Eur. J. Clin. Nutr. 50, 6–11. Marshall, D., Johnell, O., Wedel, H., 1996. Meta-analysis of how well measures of bone mineral density predict occurrence of osteoporotic fractures. Br. Med. J. 312, 1254–1259. May, H., Murphy, S., Khaw, K.T., 1994. Age-associated bone loss in men and women and its relationship to weight. Age Ageing 23, 235–240. Rasmussen, F., Tynelius, P., Kark, M., 2003. Importance of smoking habits for longitudinal and age-matched changes in body mass index: a cohort study of Swedish men and women. Prev. Med. 37, 1–9. Rossman, I., 1977. Anatomic and body composition changes with aging. In: Finch, C.E., Hayflick, L. (Eds.). Handbook of the Biology of Aging. Van Nostrand Reinhold, New York, pp. 189–221. Schott, A.M., Cormier, C., Hans, D., Favier, F., Hausherr, E., Dargent-Molina, P., Delmas, P.D., Ribot, C., Sebert, J.L., Breart, G., Meunier, P.J., 1998. How hip and whole-body bone mineral density predict hip fracture in elderly women: the EPIDOS Prospective Study. Osteoporos. Int. 8, 247–254. Sheehan, T.J., DuBrava, S., DeChello, L.M., Fang, Z., 2003. Rates of weight change for black and white Americans over a twenty year period. Int. J. Obes. Relat. Metab. Disord. 27, 498–504.
290
H. Taggart et al. / Archives of Gerontology and Geriatrics 39 (2004) 283–290
Steiger, P., Cummings, S.R., Black, D.M., Spencer, N.E., Genant, H.K., 1992. Age-related decrements in bone mineral density in women over 65. J. Bone Miner. Res. 7, 625–632. Svanborg, A., Eden, S., Mellstrom, D., 1981. Metabolic changes in aging: predictors of disease. The Swedish Experience. In: Ingram, D.K., Bajer, G.T., Shock, N.W. (Eds.). The Potential for Nutritional Modulation of Aging. Food and Nutrition Press, Trumbull, CT, pp. 81–90. Trivedi, D.P., Khaw, K.T., 2001. Bone mineral density at the hip predicts mortality in elderly men. Osteoporos. Int. 12, 259–265. Tromp, A.M., Ooms, M.E., Popp-Snijders, C., Roos, J.C., Lips, P., 2000. Predictors of fractures in elderly women. Osteoporos. Int. 11, 134–140. Uusi-Rasi, K., Sievanen, H., Pasanen, M., Oja, P., Vuori, I., 2001. Maintenance of body weight, physical activity and calcium intake helps preserve bone mass in elderly women. Osteoporos. Int. 12, 373–379. Waaler, H.T., 1984. Hazard of obesity—the Norwegian experience. Acta Med. Scand. Suppl. 679, 1–56. Zamboni, M., Zoico, E., Scartezzini, T., Mazzali, G., Tosoni, P., Zivelonghi, A., Gallagher, D., De Pergola, G., Di Francesco, V., Bosello, O., 2003. Body composition changes in stable-weight elderly subjects: the effect of sex. Aging Clin. Exp. Res. 15, 321–327.
Healthy elderly individuals do not inevitably lose bone density and weight as they age Hugh Taggart a , David Craig b,∗ , Karen McCoy a a b
Department of Health Care for the Elderly, Belfast City Hospital, Lisburn Road, Belfast BT9 6BA, UK Department of Geriatric Medicine, Queen’s University of Belfast, 97 Lisburn Road, Belfast BT9 7BL, UK Received 6 January 2004; received in revised form 29 April 2004; accepted 6 May 2004 Available online 28 July 2004
Abstract There have been relatively few extended longitudinal studies of bone mineral density (BMD) and anthropometric characteristics in the elderly. We recorded baseline measurements (height, weight, body mass index (BMI)) and BMD in 292 relatively healthy individuals (mean age: 75 years females, 76 years males) who were followed up over 8 years. At baseline, a significant inverse relationship was noted between increasing age and BMD at the neck of femur in both sexes. In female subjects, baseline age was inversely related to BMD at the total hip site. There was a significant positive relationship between baseline weight/BMI in both sexes and BMD at the total hip and neck of femur. Weight and BMI were positively related to BMD at the spine in female participants. After a mean follow-up period for males and females of 7.6 and 8.0 years, respectively, there was a significant fall in height for male and female subjects although weight and BMI did not significantly vary in either sex. Follow-up hip BMD fell marginally but significantly in female subjects only. A significant relationship existed between change in hip BMD and change in weight in females. Measurements of the neck of femur did not change significantly on follow-up in either sex. While cross-sectional measurements show weight and age are related to bone density in elderly people, longitudinal evaluation over 8 years failed to demonstrate consistent reductions in males and females. In addition, there was good maintenance of BMD at most sites. © 2004 Elsevier Ireland Ltd. All rights reserved. Keywords: Bone mineral density; Body weight; Body mass index; Aged; Longitudinal study
∗
Corresponding author. Tel.: +44 2890 272202; fax: +44 2890 325839. E-mail address: [email protected] (D. Craig).
0167-4943/$ – see front matter © 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.archger.2004.05.001
284
H. Taggart et al. / Archives of Gerontology and Geriatrics 39 (2004) 283–290
1. Introduction An estimated 3 million people in the UK suffer from osteoporosis which carries a lifetime risk of osteoporosis-related fracture above 40% (Cooper et al., 1992). The risk of osteoporotic fracture in elderly people is strongly related to bone density (Marshall et al., 1996; Schott et al., 1998; Fujiwara et al., 2003) which in turn is strongly related to body weight (Burger et al., 1998; Dargent-Molina et al., 2000). There have been several longitudinal studies of bone mineral density (BMD) in the elderly but these have been mainly <5 years duration (Greenspan et al., 1994; Jones et al., 1994; Ensrud et al., 1997; Burger et al., 1998; Schott et al., 1998; Hannan et al., 2000; Tromp et al., 2000; Trivedi and Khaw, 2001; Uusi-Rasi et al., 2001; Fujiwara et al., 2003). It is thought that BMD declines linearly over an extended period in the elderly but there is relatively little data to support this view. Prospective studies of weight change of the elderly are rare, often restricted to under 75 and of limited follow-up (Lehmann and Bassey, 1996; Rasmussen et al., 2003; Sheehan et al., 2003). These studies together with large cross-sectional reports suggest weight loss in older age is normal, albeit dependant on relative changes in fat mass and muscle bulk (Gillette-Guyonnet et al., 2003; Zamboni et al., 2003). We recorded baseline characteristics of 292 individuals who were followed up over 8 years in order to measure the change in BMD and its relationship to anthropometric characteristics such as weight, height and BMI.
2. Materials and methods 2.1. Study population Subjects were recruited initially as part of a study to measure normal ranges for BMD in an elderly UK population. They were excluded if they had a history of low trauma fracture or steroid use. All were ambulatory, mentally alert and free of serious disease. A cohort of 181 women and 111 men was recruited and written informed consent obtained. The baseline examination was carried out in 1991 and comprised of an interview followed by measurement of BMD. Information was obtained on age, sex, height and weight. After a mean period of 7.6 years (men) and 8.0 years (women) all participants were invited for follow-up assessments. 2.2. Measurement of BMD At the first visit, BMD of the lumbar spine (L2–L4), neck of femur and total hip were measured (g/cm3 ) by DXA using a Hologic 1000 bone densitometer (coefficient of variation: 0.38%). At follow-up visit these sites were rescanned using a Hologic 4500A bone densitometer (coefficient of variation: 0.34%). In order to ensure the comparability of the two scanners, 54 subjects had spine scans done on both machines. The correlation coefficient between the spine scans on each machine was 0.994.
H. Taggart et al. / Archives of Gerontology and Geriatrics 39 (2004) 283–290
285
2.3. Data analysis The percentage change in BMD, weight and height at follow-up was measured where available. Determination of fracture occurrence was obtained at interview and confirmed in cases of doubt through contact with the subject’s general practitioner. Men and women were analysed separately. Differences in baseline characteristics were compared using Student’s t-test for continuous variables and chi square for categorical variables. Pearson’s correlation coefficient was applied to relationship analysis. All calculations were done using the SPSS statistical package.
3. Results 3.1. Baseline measurements Mean age, height and weight at baseline are shown in Table 1. The relationships between baseline sex, age and weight/BMI and baseline BMD at each site scanned are shown in Table 2. A significant inverse relationship is seen between increasing age and measured BMD at the neck of femur in both sexes. In female subjects, baseline age was inversely related to BMD at the total hip site. There was a significant positive relationship between
Table 1 Baseline mean values for age, height, weight, BMI and BMD at each scanned site
Age (years) Height (cm) Weight (kg) BMI (kg/m2 ) Total hip BMD (g/cm2 ) Neck of femur BMD (g/cm2 ) Spinal BMD (g/cm2 )
Female (no. of subjects)
Male (no. of subjects)
74.9 (64–94) (n = 181) 157.5 (116.3–178.0) (n = 181) 60.5 (37.2–99.0) (n = 177) 24.4 (15.0–51.7) (n = 177) 0.701 (0.236–1.025) (n = 177) 0.608 (0.380–0.878) (n = 177) 0.825 (0.397–1.537) (n = 181)
75.9 (65–92) (n = 111) 170.7 (149.8–198.0) (n = 111) 73.7 (45.4–105.0) (n = 105) 25.2 (19.7 – 33.5) (n = 105) 0.917 (0.585–1.378) (n = 109) 0.769 (0.468–1.280) (n = 109) 1.015 (0.629–1.776) (n = 111)
Table 2 Relationship between baseline age and weight/BMI and baseline BMD at each site scanned Baseline characteristic
Neck of femur −0.399∗∗
Total hip −0.391∗∗
Lumbar spine
Female
Age Weight BMI
r= r = 0.353∗∗ r = 0.285∗∗
r= r = 0.399∗∗ r = 0.333∗∗
NS r = 0.367∗∗ r = 0.342∗∗
Male
Age Weight BMI
r = −0.189∗∗ r = 0.287∗∗ r = 0.263∗∗
NS r = 0.312∗∗ r = 0.280∗∗
NS NS NS
NS: non-significant correlation. ∗∗ Denotes correlation is significant at the 0.01 level (two-tailed).
286
H. Taggart et al. / Archives of Gerontology and Geriatrics 39 (2004) 283–290
baseline weight/BMI in both sexes and measured BMD at the hip and neck of femur. Increasing weight/BMI was related to increasing BMD at the spine in female participants. 3.2. Follow-up measurements Height, weight/BMI and BMD data for individuals available for follow-up is shown in Table 3. The mean follow-up period for males and females was 7.6 and 8.0 years, respectively. There was a significant fall in height in male and female subjects although a significant rise in spinal BMD was noted in all participants. Weight and BMI did not significantly change in either sex during the follow-up period. Measurements at the neck of femur did not change upon follow-up in either sex. Hip BMD fell significantly in female subjects only. A significant relationship existed in females between change in hip BMD and change in weight during the follow-up period (Fig. 1). Nineteen females and nine males alive at follow-up but unable to attend had mean baseline BMD measurements at the spine and total hip not significantly different to the follow-up population. 3.3. Mortality
BMD Hip Change During Follow-up (g/cm2)
Forty-nine of the 111 men (43%) died after a mean follow-up period of 7.6 years. Fifty-one of the 181 women (28%) died after a mean follow-up period of 8.0 years. This represents a significantly higher death rate (P < 0.02) in females. There was no significant difference between death rates (male or female) in our cohort and those estimated from the general population when analysed with reference to life expectancy tables for Northern Ireland.
.3
.2
.1
0.0
-.1
Rsq = 0.1732
-20
-10
0
10
Weight Change During Follow-up (kg) Fig. 1. Change in weight vs. BMD at the hip.
20
Variable
Baseline (mean)
At follow-up (mean)
% Change
% Annual change
% Subjects with an increase
% Subjects with a decrease
Females
Height (n = 97) Weight (n = 109) BMI (n = 95) Neck of femur (n = 94) Total hip (n = 94) Lumbar spine (n = 94)
158.2 61.7 25.0 0.637 0.745 0.840
155.4∗∗∗ 61.3 26.0 0.620 0.721∗∗∗ 0.870∗∗∗
−1.76 −0.57 4.04 −2.56 −3.17 +3.65
−0.22 −0.07 0.51 −0.32 −0.40 +0.46
13.4 51.4 66.0 28.7 35.1 70.2
75.5 47.7 34.0 55.3 59.6 21.3
Males
Height (n = 52) Weight (n = 52) BMI (n = 51) Neck of femur (n = 53) Total hip (n = 53) Lumbar spine (n = 53)
170.7 73.9 25.3 0.762 0.911 0.993
168.9∗∗∗ 73.6 25.7 0.774 0.914 1.039∗∗∗
−1.03 −0.47 +1.76 +1.55 +0.32 +4.59
−0.14 −0.06 +0.23 +0.20 +0.04 +0.60
13.5 50.0 62.7 54.7 52.8 69.8
76.9 46.2 37.3 32.1 39.6 26.4
∗∗∗
Denotes significant change for a paired samples t-test at the P < 0.001 level (two-tailed).
H. Taggart et al. / Archives of Gerontology and Geriatrics 39 (2004) 283–290
Table 3 Height, weight/BMI and BMD data for individuals available for follow-up with calculated variable and subject change
287
288
H. Taggart et al. / Archives of Gerontology and Geriatrics 39 (2004) 283–290
3.4. Fractures Thirty-four (19%) females, but only one male (1%) suffered at least one clinically apparent fracture during the follow-up period (P < 0.005). Six females had a hip fracture, three spinal fractures and 12 forearm fractures. In 17 women the fracture was at other sites.
4. Discussion This study of 292 elderly individuals surveyed at baseline and around 8 years later is one of the longest longitudinal studies reported. At initial assessment, BMD was lower in older individuals at the neck of femur in both sexes and at the hip in females. A similar relationship between weight/BMI and BMD was seen in both sexes. Upon follow-up, a significant fall in BMD at the hip but not femoral neck was observed in females. Males showed non-significant increases in BMD at hip and neck of femur after follow-up. Spinal BMD increased when measured at follow-up in both sexes. The observation of reduced BMD with age is well documented although few studies have restricted their analysis to the older old. The cross-sectional analysis of the OFELY cohort of over 1000 women described accelerated bone loss at most skeletal sites after the age of 75 years (Arlot et al., 1997). Our observed rate of bone loss at hip sites (−0.31 to −0.40% per year) is lower than the published rates of bone loss of −0.82% at the femoral neck and −1.30% at Wards triangle in a cross-sectional study of over 7000 elderly women (Steiger et al., 1992) and all site 3.4–4.8% per 4 year loss in women followed longitudinally in the Framingham Osteoporosis Study (Hannan et al., 2000). As observed in other studies, preservation of body weight significantly reduced measured bone loss and may explain up to a third of the decline in BMD with age (Steiger et al., 1992; Felson et al., 1993; May et al., 1994). The preservation of weight in this cohort is worth noting. Previous studies of weight change in the elderly report reductions in weight during the seventh decade in males and the eighth decade in females (Svanborg et al., 1981; Lehmann and Bassey, 1996; Gillette-Guyonnet et al., 2003) with falls in BMI apparent between 70 and 85 years (Rossman, 1977; Waaler, 1984). BMI was seen to fall with age in a recent large study of elderly Swedish participants especially those with a smoking history (Rasmussen et al., 2003). A smaller study of limited follow-up found instead significant increases in BMI in both men and women aged >68, with increases in percentage fat in women and decreases in percentage muscle mass in men (Zamboni et al., 2003). Males were seemingly able to preserve BMD with a trend towards increased BMD noted on follow-up at the hip and neck of femur. Large studies of bone density in males show lower BMD when compared to females in cross-section and over time with the decline in BMD per year approximately half the rate of similarly aged female subjects (Hannan et al., 2000; Berntsen et al., 2001). Spinal bone density rose significantly in both sexes during follow-up. Scoliosis, osteophytosis and bony fusion all serve to artefactually increase spinal BMD. As reported in the Rotterdam study, high measured BMD may obscure accelerated bone loss beyond the arthritic and disused joint (Burger et al., 1996).
H. Taggart et al. / Archives of Gerontology and Geriatrics 39 (2004) 283–290
289
These data demonstrate that healthy individuals do not inevitably lose bone mineral density nor suffer weight loss. Height loss is more common although this may be subject to the reliability of measurement.
References Arlot, M.E., Sornay-Rendu, E., Garnero, P., Vey-Marty, B., Delmas, P.D., 1997. Apparent pre- and post-menopausal bone loss evaluated by DXA at different skeletal sites in women: the OFELY cohort. J. Bone Miner. Res. 12, 683–690. Berntsen, G.K., Fonnebo, V., Tollan, A., Sogaard, A.J., Magnus, J.H., 2001. Forearm bone mineral density by age in 7,620 men and women: the Tromso study, a population-based study. Am. J. Epidemiol. 153, 465–473. Burger, H., Van Daele, P.L., Odding, E., 1996. Association of radiographically evident osteoarthritis with higher bone mineral density and increased bone loss with age. The Rotterdam Study. Arthritis Rheum. 39, 81–86. Burger, H., De Laet, C.E., Van Daele, P.L., 1998. Risk factors for increased bone loss in an elderly population: the Rotterdam Study. Am. J. Epidemiol. 147, 871–879. Cooper, C., Campion, G., Melton III, L.J., 1992. Hip fracture in the elderly: a worldwide projection. Osteoporos. Int. 2, 285–289. Dargent-Molina, P., Poitiers, F., Breart, G., EPIDOS Group., 2000. In elderly women weight is the best predictor of a very low bone mineral density: evidence from the EPIDOS study. Osteoporos. Int. 11, 881–888. Ensrud, K.E., Lipschutz, R.C., Cauley, J.A., Seeley, D., Nevitt, M.C., Scott, J., Orwoll, E.S., Genant, H.K., Cummings, S.R., 1997. Body size and hip fracture risk in older women: a prospective study. Study of Osteoporotic Fractures Research Group. Am. J. Med. 103, 274–280. Felson, D.T., Zhang, Y., Hannan, M.T., Anderson, J.J., 1993. Effects of weight and body mass index on bone mineral density in men and women: the Framingham study. J. Bone Miner. Res. 8, 567–573. Fujiwara, S., Kasagi, F., Masunari, N., Naito, K., Suzuki, G., Fukunaga, M., 2003. Fracture prediction from bone mineral density in Japanese men and women. J. Bone Miner. Res. 18, 1547–1553. Gillette-Guyonnet, S., Nourhashemi, F., Andrieu, S., Cantet, C., Albarede, J.L., Vellas, B., Grandjean, H., 2003. Body composition in French women +75 years of age: the EPIDOS study. Mech. Age. Dev. 124, 311–316. Greenspan, S.L., Maitland, L.A., Myers, E.R., Krasnow, M.B., Kido, T.H., 1994. Femoral bone loss progresses with age: a longitudinal study in women over age 65. J. Bone Miner. Res. 9, 1959–1965. Hannan, M.T., Felson, D.T., Dawson-Hughes, B., Tucker, K.L., Cupples, L.A., Wilson, P.W., Kiel, D.P., 2000. Risk factors for longitudinal bone loss in elderly men and women: the Framingham Osteoporosis Study. J. Bone Miner. Res. 15, 10–20. Jones, G., Nguyen, T., Sambrook, P., Kelly, P.J., Eisman, J.A., 1994. Progressive loss of bone in the femoral neck in elderly people: longitudinal findings from the Dubbo osteoporosis epidemiology study. Br. Med. J. 309, 691–695. Lehmann, A.B., Bassey, E.J., 1996. Longitudinal weight changes over four years and associated health factors in 629 men and women aged over 65. Eur. J. Clin. Nutr. 50, 6–11. Marshall, D., Johnell, O., Wedel, H., 1996. Meta-analysis of how well measures of bone mineral density predict occurrence of osteoporotic fractures. Br. Med. J. 312, 1254–1259. May, H., Murphy, S., Khaw, K.T., 1994. Age-associated bone loss in men and women and its relationship to weight. Age Ageing 23, 235–240. Rasmussen, F., Tynelius, P., Kark, M., 2003. Importance of smoking habits for longitudinal and age-matched changes in body mass index: a cohort study of Swedish men and women. Prev. Med. 37, 1–9. Rossman, I., 1977. Anatomic and body composition changes with aging. In: Finch, C.E., Hayflick, L. (Eds.). Handbook of the Biology of Aging. Van Nostrand Reinhold, New York, pp. 189–221. Schott, A.M., Cormier, C., Hans, D., Favier, F., Hausherr, E., Dargent-Molina, P., Delmas, P.D., Ribot, C., Sebert, J.L., Breart, G., Meunier, P.J., 1998. How hip and whole-body bone mineral density predict hip fracture in elderly women: the EPIDOS Prospective Study. Osteoporos. Int. 8, 247–254. Sheehan, T.J., DuBrava, S., DeChello, L.M., Fang, Z., 2003. Rates of weight change for black and white Americans over a twenty year period. Int. J. Obes. Relat. Metab. Disord. 27, 498–504.
290
H. Taggart et al. / Archives of Gerontology and Geriatrics 39 (2004) 283–290
Steiger, P., Cummings, S.R., Black, D.M., Spencer, N.E., Genant, H.K., 1992. Age-related decrements in bone mineral density in women over 65. J. Bone Miner. Res. 7, 625–632. Svanborg, A., Eden, S., Mellstrom, D., 1981. Metabolic changes in aging: predictors of disease. The Swedish Experience. In: Ingram, D.K., Bajer, G.T., Shock, N.W. (Eds.). The Potential for Nutritional Modulation of Aging. Food and Nutrition Press, Trumbull, CT, pp. 81–90. Trivedi, D.P., Khaw, K.T., 2001. Bone mineral density at the hip predicts mortality in elderly men. Osteoporos. Int. 12, 259–265. Tromp, A.M., Ooms, M.E., Popp-Snijders, C., Roos, J.C., Lips, P., 2000. Predictors of fractures in elderly women. Osteoporos. Int. 11, 134–140. Uusi-Rasi, K., Sievanen, H., Pasanen, M., Oja, P., Vuori, I., 2001. Maintenance of body weight, physical activity and calcium intake helps preserve bone mass in elderly women. Osteoporos. Int. 12, 373–379. Waaler, H.T., 1984. Hazard of obesity—the Norwegian experience. Acta Med. Scand. Suppl. 679, 1–56. Zamboni, M., Zoico, E., Scartezzini, T., Mazzali, G., Tosoni, P., Zivelonghi, A., Gallagher, D., De Pergola, G., Di Francesco, V., Bosello, O., 2003. Body composition changes in stable-weight elderly subjects: the effect of sex. Aging Clin. Exp. Res. 15, 321–327.