Screening for low bone mass with quantitative ultrasonography in a community without dual-energy X-ray absorptiometry: population-based survey

Osteoporosis is an epidemiologic disorder that frequently results in fractures, psychological problems, social consequences, functional limitations and poor quality of life [1, 2]. To reduce the incidence of osteoporotic fractures, individuals, especially postmenopausal women, who are at high risk for osteoporosis must be identified [3, 4]. Dual-energy x-ray absorptiometry (DXA) is the most accurate clinical method to identify low bone mineral density (BMD) [4, 5]. However, in some areas of the world, access to axial DXA is limited, or screening for low BMD with DXA is not cost effective [6]. Quantitative ultrasonography (QUS), which has been used to assess bone (especially calcaneal) status for almost 2 decades has proven to be widely and clinically useful [7‐9].

Kinmen, Taiwan, is a 176-km² island that is 154 miles west of Taiwan and 25 miles east of Mainland China. Two hospitals (one public and one military) and 41 physicians (as registered in 2000) serve the population. Because Kinmen is a military buffer district between Mainland China and Taiwan, Republic of China, (Figure 1), it has largely lacked medical resources. In Taiwan, more than 200 DXA machines available, but none are in Kinmen [10]. Although DXA is currently the criterion standard for diagnosing osteoporosis, referring high-risk individuals from Kinman for DXA is difficult because of travel costs. Therefore, inexpensive substitutes for DXA are needed. Bone QUS may be one such substitute.

The goals of our study were (1) to screen for low bone mass by using QUS in a community without DXA, (2) to conduct a reliability analysis and provide reference QUS values and (3) to estimate the prevalence of low QUS values.

Because populations in Taiwan and elsewhere are aging, interest in the diagnosis, treatment and costs of osteoporosis has been renewed. In Taiwan, including Kinmen, where 22 million (mostly Chinese) people live, 8.6% of the population are elderly (aged 65 year or older), and their life expectancy is rapidly increasing [15]. In the elderly population, osteoporosis and related fractures are a major health problem. From the viewpoint of community medicine, population screening for osteoporosis in the elderly seems to increase their awareness of falling, and such screening is associated with a reduced fracture rate [16].

The first step of any action plan to prevent osteoporotic fractures is raising awareness [17]. This was why we investigated the use of QUS measurements and associated health education in Kinmen, where DXA is unavailable. Because of different geographic and environmental factors, the epidemiology of osteoporosis in Kinmen may differ from that in the West or in other Chinese areas. Therefore, subsequent epidemiologic analysis is needed.

Because QUS is safe, portable and inexpensive, it is a well-accepted instrument for screening for low BMD [18]. Several studies have shown significant correlations between calcaneal QUS values and calcaneus, hip, or spine BMD values measured with DXA [19‐23]. A moderate and significant correlation was found, one also tested in our study; this can be applied as an alternative in populations other than those in the area we screened.

Although the diagnosis of osteoporosis by means of QUS remains controversial, the debate is due more to the limitations of present T-scores than to the technique [13, 14, 24‐27]. Previous community-based studies of Japanese, Taiwanese and Vietnamese populations revealed that specific T-scores (same as the WHO criterion applied to calcaneal QUS) is reasonable when the reference group was selected from the same population as the group being examined [11, 13, 14].

By 1999, the US Food and Drug Administration had approved five QUS devices for use in the routine diagnosis of osteoporosis, for the determination of fracture risk, and for the monitoring bone changes [28]. The QUS-II machine (Metra Biosystems) was one of those devices. More than 20 ultrasound instruments are now widely used throughout the world. BUA results of the gel-coupled QUS-II ultrasound system (Metra Biosystems) are significantly associated with age, body weight, level of physical activity and dietary calcium intake [29]. One epidemiologic investigation performed with this ultrasonometer showed that age, body height and BMI are significant determinants of BUA for each sex, with the three factors collectively accounting for 25% of the total variance in BUA [14].

The general trend of age-related change in QUS variables was negative for both sexes, but it was more pronounced in women than in men. Linear regression analysis of BUA and age showed a decreased β value (regression coefficient) for women. In our study using QUS-II machines (Metra Biosystems), results were β = -0.76 (p < 0.001) for women and β = -0.25 (p < 0.001) for men on multiple regression. In Sweden [30], results were β = -0.39 (p < 0.001) for women and β = -0.24 (p < 0.01) for men on multiple regression. In Poland [31], results were β = -0.4 (p < 0.001) for women and β = -0.2 (p < 0.05) for men on multiple regression. Investigators in Sweden and for Poland used Achilles QUS systems. For the Chinese population of Taiwan [13], we previously used a UBIS 3000 machine. Our results were β = -0.5 (p < 0.001) for women and β = -0.1 (p < 0.001) on multiple regression. In Chinese women in Hong-Kong [32], the result obtained by using a Sahara QUS system were β = -0.44 (p < 0.001) on simple regression.

Compared with peak bone mass in the lumbar spine BMD, the annual bone loss in women was 3-fold that in men (1.28–1.60% vs 0.4–0.64%) after the age of 50 years [33]. These data were reported in a population-based, routine health study of healthy Asian individuals. We observed the same difference between men and women (0.27% vs 0.83% loss per year, respectively), but the absolute percentage of annual loss was approximately half of that in the aforementioned study. In a study of similar QUS screening in Rotterdam, the change in BUA units per year were -0.15% for men and -0.43% for women [34]; in a study in Norfolk, the rates were and -0.13% and -0.70%, respectively [35].

Previous studies of Taiwanese populations demonstrated that use of a specific T-score (the same as the WHO criterion applied to the calcaneal QUS value) is reasonable when the reference group is selected from the same population as the group being examined [13]. With similar methods, the prevalence of T-scores ≤ -2.5 calculated from a nationwide calcaneal QUS screening of bone mass of healthy Japanese women in the sixth, seventh, or eighth-and-older decades were 5.2%, 18.7% and 43.6%, respectively [11]; the prevalence of low QUS values (T-scores ≤ = 2.5) in Taiwanese women were 7.9, 21.7 and 34.5%, respectively [13]. Our rates of low QUS values in the sixth, seventh and eighth-and-older decades were 16.5%, 41.0% and 61.2% in women, and 7.1%, 13.5% and 21.7% in men.

In comparison with DXA-based BMD measurements at the hip, spine or forearm yield prevalence of osteoporosis of 14.8% in the sixth decade, 21.6% in the seventh decade, and 38.5% in the eight-and-older decade among Western women [36]. Prevalence of osteoporosis based on BMD measurements at the spine or hip are 24–38% in Japanese women older than 50 years [37] and 16–24% in Taiwanese women older than 80 years [38]. Our 37.5% rate of low bone mass (cutoff T-score = -2.5) in women from Kinmen older than 50 years was closest to the result from the Japanese Population-Based Osteoporosis Study [37].

QUS has great potential for widespread use owing to its portability, low cost, and lack of ionizing radiation. The usefulness of QUS in evaluating hip, spinal or other fractures in both women and men [39] has been proved and it is similar to that of BMD [40, 41]. Cost-effectiveness analyses of QUS have shown a benefit in using ultrasonography as a selective population screening tool [42, 43]. Therefore, QUS is worthy of evaluation in sex-, age- and diagnosis-specific populations to identify individuals with low bone mass, especially in the primary healthcare system. QUS measurement of the calcaneus may be an effective, acceptable, and useful tool for epidemiologic screening of low bone mass at a primary care units, especially those in areas where DXA is not available.

QUS was reliable in this community survey. Age-related decreases in calcaneal ultrasonometric values indicated that low bone mass was more obvious in women than men. The WHO criteria can reasonably be applied to calcaneal QUS when the reference group is selected from the same population as that being screened.