In this cross-sectional study of community-dwelling adults age 60 and older, the prevalence of osteoporosis, sarcopenia and sarco-osteoporosis was higher in older individuals. The prevalence of osteoporosis and sarcopenia was similar, whereas sarco-osteoporosis was less common. Use of potentially more sensitive criteria (i.e., using tallest historical height instead of current height to define sarcopenia and including 1/3 radius BMD T-score in the definition of osteoporosis) increased the prevalence of sarcopenia, osteoporosis and sarco-osteoporosis. This is not surprising as both degenerative changes, which elevate DXA-measured BMD, and reported height loss were present in virtually all of these individuals.
To our knowledge, only very limited data currently exist on the prevalence of sarco-osteoporosis. Clearly, the prevalence of sarcopenia, osteoporosis and sarco-osteoporosis will vary depending on the population studied (i.e., prevalent hip fracture, included gender, age group and degree of frailty). As noted in the introduction, the prevalence of sarco-osteoporosis has been reported as high as 45 % in individuals who already suffered a hip fracture [31
]. Other authors report percentages closer to those found in this study [19
]. Similar to the results reported here, most studies that divided participants by age observed a higher prevalence in older age groups [19
Age is accepted as a major contributor to fracture risk [6
]. It is probable that there are many risk factors “hidden within age” contributing to this increased risk; evidence is increasing that sarcopenia is one such factor [19
]. Importantly, sarcopenia is closely linked to falls and increased risk of falling is a well-established contributor to fracture [10
]. In addition to increasing falls risk, sarcopenia might also decrease bone strength by reducing mechanical loading to the skeleton. Reduction of mechanical stimulation could result from decreased maximal force that weaker muscles produce and/or less time that the skeleton is loaded due to relative immobility [50
]. As such, the concept of “sarco-osteoporosis,” seems likely to facilitate fracture risk estimation [27
]. Moreover, simple approaches such as described here (radius BMD and use of historical height) with potential to enhance diagnostic sensitivity for sarco-osteoporosis, may have direct clinical applicability in improving fracture risk estimation.
Consideration of clinical risk factors for fracture in addition to BMD has significantly changed osteoporosis treatment decision-making. However, none of the currently available fracture risk models perfectly predict fragility fractures. Potential limitation of FRAX® (www.shef.ac.uk/frax
), relevant to this study is that only femoral neck (not spine or radius) BMD is considered. Furthermore, sarcopenia, muscle function and falls are not able to be included in this algorithm given limitations of the observational cohorts used to derive FRAX® [54
]. A second method to estimate fracture risk developed by the Garvan institute (www.garvan.org.au/bone-fracture-risk
) does include history and number of falls within the past year but no measure of muscle mass or function [10
It could be argued that measurement of BMD at additional skeletal sites, i.e., the forearm, is not necessary when BMD is measured at both the spine and femoral neck. However, lumbar spine BMD is elevated by degenerative changes that are present in virtually all older adults, 93 % in this study. Such a high prevalence is not unique to this cohort. For example, a study of 300 Australian men and women age 60 and above found X-ray evidence of osteophytes in 69 % of all males and females, disc narrowing in 67 % and apophyseal osteoarthritis (posterior element disease) in 99 % of participants [57
]. Another study of 600 Japanese women age 60 and above found osteoarthritic changes in 96 % of individuals on lumbar spine X-rays [58
]. As could be expected, spinal degenerative changes increase lumbar spine BMD measured by DXA; on average by 15–24 % [57
]. Additionally, femoral neck BMD measurement may also be elevated by degenerative changes [39
]. Other femoral neck BMD measurement confounders include internal artifacts, fat panniculi and anatomic variations [60
]. It seems likely that such confounders are one contributor to the less than ideal predictive capability of tools such as FRAX®. As such, evaluations of potential approaches to enhance fracture prediction capability, potentially including forearm BMD measurement, are appropriate.
Limitations of this study include relatively small sample size and cross-sectional nature. Furthermore, in this study sarcopenia could only be defined based on DXA measured ALM/ht2
ratio, as muscle function data were not available in all cohorts used for this analysis. Current consensus definitions for sarcopenia appropriately use both muscle mass and muscle function parameters as the combination of these measures improve the prediction of poor outcomes [20
]. Additionally, use of self-reported tallest height is inferior to serial height measurements over time as there is risk for recall errors and bias. Two systematic reviews have identified a tendency in adults to overestimate historical tallest height, leading to a greater height loss than if prospectively measured. [62
] Despite this, self-reported height loss has apparent clinical applicability and, importantly, predicts clinical outcomes including vertebral fractures, mortality and morbidity [62
]. Future study to further define the accuracy of height recall, and moreover to identify optimal approaches to determining tallest height, is needed. An additional limitation is that data regarding prior fragility fracture or other outcomes for sarco-osteoporosis are not available in this study. Future work is needed in longitudinal studies to evaluate the effect of sarco-osteoporosis diagnosis on risk for fragility fracture and also to explore the utility of using historical tallest height and radius BMD measurement on this diagnosis.