Osteoporos Int (2005) 16: 1663–1668 DOI 10.1007/s00198-005-1899-z
O R I GI N A L A R T IC L E
Bone mineral density reference norms for Hong Kong Chinese H.S. Lynn Æ E.M.C. Lau Æ B. Au Æ P.C. Leung
Received: 14 October 2004 / Accepted: 7 March 2005 / Published online: 19 July 2005 Ó International Osteoporosis Foundation and National Osteoporosis Foundation 2005
Abstract The aim of this study was to establish bone mineral density (BMD) reference norms for Hong Kong Chinese using Hologic QDR 2000 and 4500 densitometers, and to estimate the prevalence of osteoporosis in the population. Altogether, 4,274 subjects (2,415 females and 1,859 males), aged 9–94 years old, were recruited using a combination of private solicitation and public advertising from schools, community centers, nursing homes, housing estates, and the general community in Hong Kong. Among females, BMD increased by 20% at the total hip and 48% at the lumbar spine between ages 10 and 20 but remained essentially constant between ages 20 and 40. Between ages 40 and 70, BMD declined by 17% at the total hip and 23% at the spine. Total hip BMD continued to drop after age 70 but little change in spine BMD was observed. Among males, BMD increased by 45% at the total hip and 77% at the spine between age 10 and 30. Between ages 30 and 80, total hip BMD decreased by 20%. Lumbar spine BMD decrease was milder, showing a loss of 4% between ages 30 to 50 and remaining relatively constant afterwards. The prevalence of
H.S. Lynn Department of Community and Family Medicine, Chinese University of Hong Kong, Hong Kong, China E.M.C. Lau Æ B. Au Æ P.C. Leung Jockey Club Centre for Osteoporosis Care, Chinese University of Hong Kong, Hong Kong, China P.C. Leung Department of Orthopaedics and Traumatology, Chinese University of Hong Kong, Hong Kong, China H.S. Lynn (&) Department of Community and Family Medicine, 4/F School of Public Health, Prince of Wales Hospital, Shatin, N.T.China E-mail:
[email protected] Tel.: +852-2252-8889 Fax: +852-2604-8091
osteoporosis was consistently overestimated when using Hologic-supplied Caucasian cutoffs as compared with local Chinese cutoffs. The prevalence of osteoporosis among Chinese women 50 years or older was 37% and 16% at the spine and total hip, respectively, while that among Chinese men 50 years or older was 7% and 6% at the spine and total hip, respectively. Prior studies have been limited by size or restricted to women. This study represents the largest sample of Hong Kong Chinese amassed to date, provides continuous BMD reference values from ages 10 to 85 for both women and men, and yields more reliable estimates of the prevalence of osteoporosis for the population. Keywords Bone mineral density Æ Chinese Æ Normal range
Introduction Osteoporosis is one of the major and growing health problems in Asia [1], and a crucial determinant of subsequent fracture risk [2]. It is therefore necessary to establish reliable bone mineral density (BMD) reference values in order to accurately diagnose osteoporosis and monitor changes in bone strength. The reference ranges supplied by manufacturers of bone densitometers are, however, unlikely to be applicable across all populations, since BMD measurements are affected by ethnic variations [3, 4, 5]. Moreover, manufacturer-supplied reference values are typically available only at selected ages (e.g., every 5 years). The current study represents the largest sample ( n =4,274) of Hong Kong Chinese amassed to date for establishing BMD reference ranges using the Hologic QDR 2000 and 4500 densitometers. Prior studies have focused on documenting BMD ranges in women [6, 7], while an earlier report provided population ranges of BMD based on 164 men aged 40–79 years and 436 women aged 20–89 [8]. The data reported here
1664
substantially extend the applicability of the prior studies, by providing continuous BMD reference values from ages 10 to 85 for both men and women, rendering a clear and informative description of BMD changes from puberty to old age among Chinese.
Methods Subjects Study participants, aged 9–94 years old, were recruited using a combination of private solicitation and public advertising from schools, social and community centers, nursing homes, housing estates, and the general community in Hong Kong. A total of 4,274 subjects, 2,415 females and 1,859 males, were included in the final analysis. Nine hundred and sixty-six subjects with the following characteristics were excluded from the study during recruitment: those with a history of metabolic bone disease or medical disorders related to calcium absorption; those receiving steroid hormones, heparin, thyroxine, calcium and/or vitamin D supplements, or anticonvulsants; and those suffering from any genetic skeletal disorders. Among women, those who were receiving estrogen hormones and those who were pregnant or had had ovariectomies were excluded. Elderly subjects who had had bilateral hip replacements or who could not walk independently were also excluded. Bone mineral density measurements Bone mineral density (BMD) was measured for the lumbar spine (L1–L4 in anteroposterior projection) and proximal femur using dual-energy X-ray absorptiometry (DXA) with Hologic QDR 2000 or QDR 4500 bone densitometers (Hologic, Waltham, MA, USA). Calibration was performed daily on a lumbar spine phantom, and the coefficients of variation for the spine, total hip, and total body BMD measurements were 0.7%, 0.8%, and 1.0%, respectively.
Fig. 1 Age distribution by gender
Data analysis Nonparametric additive models [9] were fitted to the BMD measurements at the spine, femoral neck, trochanter, and total hip. For each model, a smoothing spline [10] was used to model the BMD measurements as a function of age, and an iterative backfitting algorithm was used to generate the regression model estimates and predicted values. Curve-wise confidence intervals were estimated assuming a Bayesian model [11]. Osteoporosis was defined according to the WHO diagnostic criteria, i.e., having a BMD less than 2.5 standard deviations below the young adult mean [12]. Young adult data from Caucasians and from the current study were both used to define the reference cutoff values. The Caucasian cutoffs were based on the default peak BMD reference values at age 25 (for total hip) and age 30 (for lumbar spine) as supplied by Hologic, while the Chinese cutoffs used the predicted BMD at these ages obtained from the nonparametric additive spline model.
Results The age distribution for the study’s female and male subjects is shown in Fig. 1. The pattern of change in BMD from ages 10 through 85 is depicted in Figs. 2 and 3, while the predicted mean BMD from the spline models at each decade are presented in Tables 1 and 2 for females and males, respectively. From ages 10 to 20 in females, lumbar spine BMD increased at a rate of 0.061 g/cm2/year, compared with 0.036 g/cm2/year, 0.032 g/cm2/year and 0.027 g/cm2/year at the total hip, femoral neck, and trochanter, respectively. During this 10-year period, BMD gained 48% at the lumbar spine, 26% at the total hip, 23% at the femoral neck, and 20% at the trochanter. From ages 20 to 40, BMD levels essentially remained at a plateau at all sites. After age 40, BMD decreased at a rate of about 0.006 g/cm2/year, such that, by age 80, BMD was 25%, 28%, and 30% lower at the total hip, femoral neck, and trochanter, respectively. At the lumbar spine, BMD decreased at a rate of 0.010 g/cm2/year until age 70, when BMD started
1665 Fig. 2 Bone mineral density among females
leveling off. By age 70, BMD had decreased by 23% of its value attained at age 40. Among males, from ages 10 to 30, lumbar spine BMD increased at a rate of 0.028 g/cm2/year, compared with 0.021 g/cm2/year, 0.015 g/cm2/year and 0.014 g/ cm2/year at the total hip, femoral neck, and trochanter, respectively. During this 20-year period, BMD gained 77% at the lumbar spine, 45% at the total hip, 31% at the femoral neck, and 38% at the trochanter. Peak BMD was attained between age 25 and age 30. After age 30, BMD decreased at a rate of 0.004 g/cm2/year at the total hip and femoral neck, and at a rate of 0.003 g/cm2/ year at the trochanter, such that, by age 80, BMD was 20%, 24%, and 19% lower at the total hip, femoral
Fig. 3 Bone mineral density among males
neck, and trochanter, respectively. At the lumbar spine, however, there was only a mild decline in BMD of 4% from age 30 to age 50, and BMD levels remained fairly constant thereafter. The prevalence of osteoporosis was consistently overestimated when using the Hologic-supplied Caucasian cutoffs as compared with the local Chinese cutoffs (Tables 3 and 4). Using local Chinese normative values, the prevalence of osteoporosis for females at the lumbar spine increased from 9% during middle age (40–59 years old) to 45% after age 70, while that at the total hip increased from 4% during middle age to 25% after age 70. For males, the prevalence of osteoporosis at the lumbar spine increased from 2% during middle age to
1666 Table 1 Predicted bone mineral density for females from ages 10 to 80 ( CI confidence interval, SD standard deviation) Age
Lumbar spine Mean±SD
10 20 30 40 50 60 70 80
0.63±0.17 0.93±0.11 0.99±0.10 0.99±0.10 0.94±0.10 0.83±0.13 0.76±0.14 0.75±0.14
Femoral neck
95% CI 0.61 0.92 0.97 0.98 0.92 0.81 0.75 0.73
0.64 0.95 1.00 1.01 0.96 0.84 0.77 0.76
Mean±SD 0.62±0.11 0.76±0.10 0.76±0.10 0.75±0.09 0.73±0.09 0.66±0.10 0.59±0.10 0.54±0.09
Trochanter
95% CI 0.61 0.75 0.75 0.74 0.72 0.65 0.59 0.53
Mean±SD
0.64 0.77 0.77 0.77 0.74 0.67 0.60 0.55
9% after age 70, while that at the total hip was negligible during middle age but reached 8% after age 70.
Discussion Bone mineral density reference ranges for Caucasians and Japanese have been demonstrated to be different from BMD values measured from Chinese [8, 13, 14]. It is thus important to have a set of Chinese BMD reference norms for both men and women that are not only accurate but also encompass the lifespan from puberty to old age. Prior studies for establishing BMD reference norms have either been restricted to women [6, 7, 14, 15] or limited in size. For example, one study in Taiwan involved 430 healthy Chinese [13], while another recent report was based on 436 women and 164 men in Hong Kong [8], and all the men were above 40 years old, making it impossible to use the WHO criteria to estimate the prevalence of osteoporosis. The current study obtained BMD measurements on 4,274 subjects, the largest number of Hong Kong Chinese recruited so far. In particular, the total number of 1,859 males in this study represents the largest number of Asian men gathered for establishing BMD reference ranges. Manufacturer-supplied BMD reference values are typically available only at selected ages (e.g., every 5 years), while other studies generate reference norms by arbitrarily grouping values across different ages [6, 7, 8]. The current study avoids the arbitrary grouping of ages
0.54±0.10 0.65±0.09 0.66±0.09 0.66±0.08 0.63±0.09 0.57±0.09 0.51±0.09 0.46±0.08
Total hip
95% CI 0.52 0.64 0.65 0.64 0.62 0.56 0.51 0.45
0.55 0.66 0.67 0.67 0.64 0.57 0.52 0.47
Mean±SD 0.68±0.12 0.86±0.10 0.89±0.11 0.88±0.10 0.84±0.10 0.78±0.11 0.73±0.11 0.66±0.11
95% CI 0.66 0.85 0.88 0.87 0.83 0.77 0.72 0.64
0.69 0.88 0.91 0.90 0.86 0.79 0.73 0.67
and instead uses nonparametric additive regression models to estimate the mean BMD value from ages 10 to 85, thus providing a continuous range of BMD reference values for both men and women. The BMD reference curves produced in this study provide a clear picture of how BMD changes throughout life in Chinese people, and is consistent with observations reported in other studies. Bone accrual is quickest during the early teen years. Between ages 9 to 18 years, BMD in Caucasian girls increased 50% at the spine, 25% at the femoral neck, and 30% at the total hip, and similar or slightly greater gains were also observed in Caucasian boys [16]. The observed BMD plateaus in Chinese women from age 20 to age 40 may correspond to a period when bone mass is consolidated [17, 18]. In contrast, a steady decrease in BMD was observed in Chinese men after age 30, which is consistent with the reported linear decrease in bone mass during adulthood [19, 20], although others have suggested that a fall in bone mass is only detectable after age 50 [21, 22]. The difference in bone loss at the femur region compared with the lumbar spine is also characteristic of findings reported elsewhere [23, 24]. It is well known that an inappropriate young adult BMD reference group can lead to biased estimates of the prevalence of osteoporosis [15, 25]. Tenenhouse found that using Hologic reference norms overestimated the prevalence of osteoporosis at both the hip and spine when compared with using reference norms derived from a random sample of Canadians [26]. Lower
Table 2 Predicted bone mineral density for males from ages 10 to 80 ( CI confidence interval,SD standard deviation) Age
Lumbar spine Mean±SD
10 20 30 40 50 60 70 80
0.56±0.07 0.86±0.15 0.99±0.11 0.98±0.13 0.95±0.12 0.94±0.14 0.94±0.17 0.94±0.18
Femoral neck
95% CI 0.53 0.84 0.97 0.96 0.92 0.92 0.93 0.92
0.58 0.89 1.01 1.00 0.97 0.96 0.95 0.95
Mean±SD 0.65±0.09 0.81±0.15 0.85±0.13 0.81±0.11 0.76±0.11 0.72±0.12 0.69±0.11 0.65±0.11
Trochanter
95% CI 0.63 0.80 0.84 0.79 0.74 0.71 0.68 0.64
0.66 0.83 0.87 0.83 0.78 0.74 0.70 0.66
Mean±SD 0.53±0.08 0.68±0.14 0.73±0.11 0.71±0.10 0.68±0.11 0.66±0.12 0.63±0.11 0.59±0.11
Total hip
95% CI 0.51 0.67 0.71 0.69 0.67 0.65 0.62 0.58
0.54 0.70 0.75 0.72 0.70 0.68 0.64 0.60
Mean±SD 0.69±0.08 0.92±0.16 1.00±0.14 0.98±0.12 0.93±0.12 0.90±0.14 0.85±0.13 0.80±0.13
95% CI 0.67 0.90 0.98 0.96 0.91 0.88 0.85 0.79
0.70 0.94 1.02 0.99 0.95 0.91 0.86 0.81
1667 Table 3 Prevalence of osteoporosis for females, according to Caucasian and Chinese reference cutoffs (CI confidence interval) Age
Lumbar spine
n
Total hip
Caucasian Percentage 40–59 60–69 70–79 80+
321 720 624 87
19 52 60 56
Chinese
95% CI 15 48 56 46
23 56 63 67
Percentage 9 39 44 49
Caucasian
95% CI 6 35 40 39
Percentage
12 42 48 60
6 15 29 48
Chinese
95% CI 4 12 25 38
Percentage
9 17 32 59
4 11 23 43
95% CI 2 9 19 32
6 13 26 53
Table 4 Prevalence of osteoporosis for males, according to Caucasian and Chinese reference cutoffs (CI confidence interval) Age
Lumbar spine
n
Total hip
Caucasian Percentage 40–59 60–69 70–79 80+
129 402 839 95
15 20 27 22
Chinese
95% CI 9 16 24 14
21 24 30 31
Percentage 2 4 9 8
prevalences of osteoporosis were also noted in Chinese women when using local population reference norms as opposed to Hologic reference ranges for oriental women [14]. Using local young-adult BMD reference values, the prevalence of osteoporosis among Chinese women 50 years or older was 37% and 16% at the lumbar spine and total hip, respectively, similar to earlier reports [6, 7]. In contrast, the prevalence of osteoporosis among Chinese men 50 years or older was 7% and 6% at the lumbar spine and total hip, respectively. Interestingly, this difference in prevalence rates between men and women is very similar to that found among U.S. elderly, where there was a 16% prevalence of osteoporosis at the total hip for women compared with a 4% prevalence in men [27]. The findings from the current study are consistent with the conclusions reported in other prior studies among Chinese [7, 13, 14]. Nevertheless, it should be noted that this study was not a randomly selected sample of the Hong Kong Chinese population, but rather a community-based study of volunteers. The latter can be criticized for potential selection bias, while the former has to contend with non-response bias, and thus the generalizability of reference norms produced from either type of study design needs to be qualified. The use of DXA is also not without its limitations, as DXA may overestimate the true bone mass in larger individuals and underestimate bone mass in smaller individuals. This drawback is especially acute among children and growing adolescents. Ideally, quantitative computed tomography should be used to measure volumetric BMD directly, although its availability and radiation toxicity are still practical concerns. Alternatively, models can be used to estimate volumetric BMD
Caucasian 95% CI
0 2 7 3
Percentage
4 6 11 14
0 3 10 14
Chinese
95% CI
Percentage
-
0 2 7 12
1 8 7
5 12 21
95% CI 1 6 5
4 9 18
[28], but there is controversy concerning the validity of such models [29], and model-dependent reference norms may be less appealing in usage. With respect to ethnic variations, a recent study found that differences in estimated volumetric BMD between Hong Kong Chinese and US Caucasians were about half the size of the differences in areal BMD [30]. These differences were attenuated after adjusting for age, body weight and height, but certain components remained statistically significant, suggesting that it may be prudent to use ethnically specific reference norms in diagnosing osteoporosis. Moreover, Chinese have similar or lower hip fracture incidence rates compared with US Caucasians [31, 32], while vertebral fracture rates were slightly lower in Chinese woman compared with US Caucasian women [33]. Such evidence from fracture rates further demonstrates the importance of ethnically specific BMD reference norms and justifies the use of Chinese cutoffs in diagnosing osteoporosis in Chinese populations.
References 1. Lau EMC (1997) The epidemiology of osteoporosis in Asia. In: Lau EMC (ed) Osteoporosis in Asia: Crossing the frontiers. World Scientific Publishers pp 1–9 2. Cummings SR, Black DM, Nevitt MC (1993) Bone density at various sites for prediction of hip fractures. Lancet 341:72–75 3. Cundy T, Cornish J, Evans MC, Gamble G, Stapleton J, Reid IR (1995) Sources of interracial variation in bone mineral density. J Bone Miner Res 10:368–373 4. Russell-Aulet M, Wang J, Thornston JC, Cott EW, Pierson RN Jr (1993) Bone mineral density and mass in a cross-sectional study of white and Asian women. J Bone Miner Res 8:575–582
1668 5. Tobias JH, Cook DG, Chambers TJ, Dalzell N (1994) A comparison of bone mineral density between Caucasian, Asian and Afro-Caribbean women. Clin Sci 87:587–591 6. Ho S, Lau EMC, Woo J, Sham A, Chan KM, Lee S, Leung PC (1999) The prevalence of osteoporosis in the Hong Kong Chinese female population. Maturitas 32:171–178 7. Liao EY, Wu XP, Deng XG, Huang G, Zhu XP, Long ZF, Wang WB, Tang WL, Zhang H (2002) Age-related bone mineral density, accumulated bone loss rate and prevalence of osteoporosis at multiple skeletal sites in Chinese women. Osteoporos Int 13(8):669–676 8. Woo J, Li M, Lau E (2001) Population bone mineral density measurements for Chinese women and men in Hong Kong. Osteoporos Int 12:289–295 9. Stone CJ (1985) Additive regression and other nonparametric methods. Ann Stat 13:689–705 10. Eubank RL (1988) Spline smoothing and nonparametric regression. Dekker, New York 11. Wahba G (1983) Bayesian confidence intervals for the cross validated smoothing spline. J R Stat Soc Ser B 45:133–150 12. WHO study group (1994) Assessment of fracture risk and its application to screening for postmenopausal osteoporosis. WHO Technical Report Series 843. WHO, Geneva 13. Tsai SC, Kao CH, Wang SJ, Chang Lai SP, Yen RF, Chieng PU (1997) Normal data for lumbar spine bone mineral content in healthy Chinese men. Calcif Tissue Int 61:114–116 14. Liao EY, Wu XP, Luo XH, Zhang H, Dai RC, Huang G, Wang WB (2003) Establishment and evaluation of bone mineral density reference databases appropriate for diagnosis and evaluation of osteoporosis in Chinese women. J Bone Miner Metab 21(3):184–192 15. Xiaoge D, Eryuan L, Xianping W, Zhiguang Z, Gan H, Zaijing J, Xiaoli P, Hongzhuan T, Hanwen W (2000) Bone mineral density differences at the femoral neck and Ward’s triangle: a comparison study on the reference data between Chinese and Caucasian women. Calcif Tissue Int 67(3):195–198 16. Bachrach LK, Hastie T, Wang MC, Narasimhan B, Marcus R (1999) Bone mineral acquisition in healthy Asian, Hispanic, Black, and Caucasian youth: a longitudinal study. J Clin Endocrinol Metab 84:4702–4712 17. Kessenich CR, Rosen CJ (1996) The pathophysiology of osteoporosis. In: Osteoporosis: Diagnosis and Therapeutic Principles. CJ Rosen (ed) Humana, Totowa, NJ, USA, pp 47-63 18. Rosenthal DI, Mayo-Smith W, Hayes CW et al (1989) Age and bone mass in premenopausal women. J Bone Miner Res 4:533– 538 19. Hannan MT, Felson DT, Anderson JJ (1992) Bone mineral density in elderly men and women: results from the Framingham Osteoporosis Study. J Bone Miner Res 7:547–553 20. Meier DE, Orwoll ES, Jones JM (1984) Marked disparity between trabecular and cortical bone loss with age in healthy men: measurement by vertebral computed tomography and radial photon absorptiometry. Ann Intern Med 101:605–612
21. Mazess RB, Barden HS, Drinka PJ, Baumens SF, Orwoll ES, Bell NH (1990) Influence of age and body weight on spine and femur bone density in US white men. J Bone Miner Res 5:645– 652 22. Tobin JD, Fox KM, Cejku ML (1993) Bone density changes in normal men: a 4–19 year longitudinal study. J Bone Miner Res 8:102 23. Jones G, Nguyen T, Sambrook P, Kelly PJ, Eisman JA (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 24. Yoshimura N, Kinoshita H, Danjoh S, Takijiri T, Morioka S, Kasamatsu T, Sakata K, Hashimoto T (2002) Bone loss at the lumbar spine and the proximal femur in a rural Japanese community, 1990–2000: the Miyama study. Osteoporos Int 13(10):803–808 25. Gurlek A, Bayraktar M, Ariyurek M (2000) Inappropriate reference range for peak bone mineral density in dual-energy Xray absorptiometry: implications for the interpretation of Tscores. Osteoporos Int 11(9):809–813 26. Tenenhouse A, Joseph L, Kreiger N, Poliquin S, Murray TM, Blondeau L, Berger C, Hanley DA, Prior JC, CaMos Research Group (2000) Estimation of the prevalence of low bone density in Canadian women and men using a population-specific DXA reference standard: the Canadian Multicentre Osteoporosis Study (CaMos). Osteoporos Int 11(10):897–904 27. Looker A, Orwoll E, Johnston C, Lindsay R, Wahner H, Dunn W, Calvo M, Harris T, Heyse S (1997) Prevalence of low femoral bone density in older US Adults from NHANES III. J Bone Miner Res 12(11):1761–1768 28. Carter DR, Bouxsein ML, Marcus R (1992) New approaches for interpreting projected bone densitometry. J Bone Miner Res 7:137–145 29. Cummings SR, Marcus R, Palermo L, Ensrud KE, Genant HK (1994) Does estimating volumetric bone density of femoral neck improve the prediction of hip fracture? A prospective study. J Bone Miner Res 9:1429–1432 30. Lau EMC, Lynn H, Woo J, Melton LJ 3rd (2003) Areal and volumetric bone density in Hong Kong Chinese: a comparison with Caucasians living in the United States. Osteoporos Int 14:583–588 31. Lau EM, Lee JK, Suriwongpaisal P, Saw SM, Das De S, Khir A, Sambrook P (2001) The incidence of hip fracture in four Asian countries: the Asian Osteoporosis Study (AOS). Osteoporos Int 12(3):239–243 32. Chie WC, Yang RS, Liu JP, Tsai KS (2004) High incidence rate of hip fracture in Taiwan: estimated from a nationwide health insurance database. Osteoporos Int 15(12):998–1002 33. Ling X, Cummings SR, Mingwei Q, Xihe Z, Xioashu C, Nevitt M, Stone K (2000) Vertebral fractures in Beijing, China: the Beijing Osteoporosis Project. J Bone Miner Res 15(10):2019– 2025
