Osteoporosis is defined as a reduction of bone mass and an alteration of bone microarchitecture leading to an increased risk of fracture. The operational diagnosis of osteoporosis is established by measurement of bone mineral density (BMD) at the hip and spine using dual-energy X-ray absorptiometry (DXA). The World Health Organization (WHO) diagnostic criteria for osteoporosis is a BMD value that equals or is lower than 2.5 standard deviations below the average value in the reference population at one of these sites.1

Sarcopenia is a potentially reversible musculoskeletal disease, related to age (primary sarcopenia) or to other factors (secondary sarcopenia).2 It has been recently defined as “a loss of skeletal muscle mass and strength/function” and robust evidence about its association with falls, disability, and mortality is available.3 While awaiting the forthcoming operational definition from the newly formed Global Leadership Initiative on Sarcopenia (GLIS), the most widely acknowledged definition is the revised consensus on definition and diagnosis by the European Working Group on Sarcopenia in Older People (EWGSOP2), where sarcopenia is considered “probable” in presence of a low muscle strength, “confirmed” when associated with low muscle mass,4 and “severe” if the physical performance is altered.

When exploring the association between sarcopenia and its components and osteoporosis, some evidence is available, mostly from cross-sectional studies e.g., in the SarcoPhAge study, Locquet et al., 2019 found associations among sarcopenia components and osteoporosis, indicating that community-dwelling older people with muscle muscle impairment had poorer bone health.5 Tiftik et al., 2023 found that postmenopausal women with low grip strength had a 1.6-fold higher risk of osteoporosis.6 Taniguchi et al., 2019 reported a significant association between osteoporosis and muscle mass, but not with muscle strength, in community-dwelling Japanese older women.7,8 Nevertheless, these associations were explored mostly in cross-sectional studies, and data on this association over time from large longitudinal studies are scarce or provides inconclusive findings. The association between bone mineral density (BMD) loss and sarcopenia remains poorly explored. The most recent meta-analysis by Gao et al., 2025, assessed the risk factors for sarcopenia in community settings across the life course, compiling evidence from longitudinal studies to date, and could not identify the BMD loss as one of the risk factors associated to sarcopenia.9 Given the emerging body of evidence linking the pathophysiology of osteoporosis and sarcopenia, it is crucial to unravel the long-term associations between the loss of BMD, as a dynamic process over time, and the presence of sarcopenia.

The Fracture RIsk Brussels Epidemiological Enquiry (FRISBEE 1) study is a cohort study of community-dwelling postmenopausal women initiated in Brussels (Belgium), in 2007.10 In 2021, all participants were invited to join a follow-up study, FRISBEE 2, whose baseline data, combined with a bone mineral density (BMD) assessment conducted at FRISBEE 1 inclusion, are analyzed and presented in this manuscript. The objective of this study is to assess the associations between the presence of sarcopenia in community-dwelling older women and BMD loss over the preceding years.

The detailed flowchart of the study is depicted in Fig. 1. Nine-hundred seven community-dwelling older women were included in the FRISBEE 2 study and their data have been used for this analysis and manuscript.

Table 1 describes the participants’ characteristics in the FRISBEE 2 study at baseline. The median age was 77 (IQR 75–81) years and the median BMI was 26.2 (IQR 23.0–29.6)kg/m2. Out of 907 older women, 172 (19.0%) had EWGSOP2-defined probable sarcopenia, 76 (8.4%) had confirmed sarcopenia, and severe sarcopenia was present in 43 (4.8%) participants at baseline.

As expected, participants with confirmed sarcopenia were older, had a lower BMI, higher prevalence of osteoporosis, were more likely to present comorbidities, and reported a sedentary lifestyle more frequently than those participants without sarcopenia; those differences were statistically significant.

The time duration between the second DXA examination at FRISBEE 2 study baseline (T2) and the first DXA examination (T1) ranged from 6 to 16 years and its median was 11 (IQR 10–13) years. The difference (T1–T2) in BMD could not be assessed for 46 participants that had missing data for DXA: 17 individuals had missing data in T1 and 37 individuals (of which 8 are the same) had missing data in T2 (data not shown).

Overall, the mean annual BMD difference (T1–T2) in the past years was a loss of 6.12±6.25mg/cm2/year; this average annual BMD loss was higher in participants with confirmed sarcopenia (7.77mg/cm2/year) than in those without (5.93mg/cm2/year, p=0.018). A small proportion of the participants (68, 7.9%) gained more than 3% BMD between the two DXA examinations, but the majority of the participants (630, 73.2%) had experienced a BMD loss of more than 3% between the two DXA examinations. In between, 163 (18.9%) participants were considered to have stable BMD.

Table 2 shows the crude odds ratio obtained by logistic regression, estimating the association of EWGSOP2-defined probable, confirmed, and severe sarcopenia in the FRISBEE2 study baseline (T2) with a total hip bone mineral density (BMD) loss of more than 3% in the past 6–16 years. In these univariate models, a loss of more than 3% in total hip BMD over the past 6–16 years was associated with statistically significantly increased odds of EWGSOP2-defined sarcopenia (probable, confirmed, and severe).

All covariates associated with confirmed sarcopenia at a significance level of p<0.3 (Table 1) were included in the backward stepwise selection procedure. The set of retained covariates varied between models, reflecting differences in their respective contributions to the outcomes examined.

Table 3 presents the estimated odds ratios (ORs) for all covariates included in the multivariable logistic regression model. The association between probable sarcopenia and BMD loss remained statistically significant after adjusting for potential confounders. Specifically, individuals who experienced a BMD loss greater than 3% over the previous years had 2.23 times higher odds of presenting probable sarcopenia, even after controlling for age, sedentary lifestyle, comorbidities, history of major osteoporotic fracture, and follow-up duration. Each of these factors independently contributed to an increased likelihood of probable sarcopenia.

Table 4 shows the adjusted odds ratios from the logistic regression estimating the association between BMD loss and EWGSOP2-defined confirmed sarcopenia. After adjustment for relevant covariates, a BMD loss greater than 3% was not significantly associated with confirmed sarcopenia (p=0.358). Conversely, increasing age, a sedentary lifestyle, and the presence of comorbidities were again identified as contributing factors to a higher likelihood of confirmed sarcopenia. Notably, a higher BMI was associated with significantly lower odds (OR: 0.80; p<0.001). History of Major Osteoporotic Fractures and follow-up duration were not retained in this model.

Analysis for severe sarcopenia was not feasible due to the limited sample size of participants with this condition in the study sample (n=43).

The study assessed the associations between sarcopenia according to the EWGSOP2 and changes in the total hip BMD in the past 6 to 16 years, in a large population of community-dwelling older women. The presence of probable sarcopenia, i.e., low handgrip strength, was associated to a clinically significant BMD loss in the past years, also after adjustement for confounders. To the authors’ knowledge, this is the first study assessing the association between the dynamic process of the loss of bone mineral density in a long-term backwards timeline, and the presence of sarcopenia in community-dwelling older women.

Our results are in line with the findings about the cross-sectional association between the components of bone health, muscle health, and sarcopenia found in community-dwelling older people in the baseline of the SarcoPhAge study.5 In a longitudinal prospective study with a 5-year follow-up, the Baltimore Longitudinal Study of Aging (BLSA) and the National Institute for Longevity Sciences-Longitudinal Study of Aging (NILS-LSA) showed that bone loss and muscle loss were lost in parallel (co-occurred), and that the loss was higher in women than in men.19

Several considerations in relation to the findings and the adjusting variables should be discussed. The analysis used BMI as one of the counfounding factors, and a higher BMI was associated with significantly lower odds of confirmed sarcopenia. Lee et al., 2016, found an association between sarcopenia and osteoporosis, which was no longer significant after adjustment for body fat.20 These partially contradictory findings might be partly attributed to the inherent limitations of DXA in accurately assessing muscle mass, as it encompasses both water and fibrotic tissue alongside lean mass. The influence of BMI and body composition on musculoskeletal health remains controversial and futher longitudinal studies may be needed to clarify this association.

The study found an independent association between a self-reported sedentary lifestyle and confirmed sarcopenia, which is coherent with the evidence reporting the benefits of physical activity and reducing sedentary lifestyle. Physical activity has been found to play a key role in determining the relationship between low BMD and sarcopenia.20 Other authors found higher levels of physical activity in healthy women compared with osteoporotic women, and higher physical activity levels have shown to be associated with an increased cortical strength.21 Globally, there is robust evidence from longitudinal studies showing an inverse relationship between physical activity and bone loss.22

The presence of comorbidities was retained in the model for probable and confirmed sarcopenia, highlighting their association, which may reflect the presence and coexistance of both primary and secondary sarcopenia in the study population. Finally, the length of follow-up was also included in the statistical model, and it was found that the association was valid independently of the length of the follow-up and the timeframe between the two DXA examinations. This may be interesting to interpret the findings, showing robustness despite the large variety of duration in the timeframe between the DXA examinations. Yet the findings could not be generalized outside a 6–16 years follow-up duration.

Bone and muscle interact through a complex bidirectional crosstalk, with an emerging evidence linking the pathophysiological mechanisms of both sarcopenia and osteoporosis.23 Osteosarcopenia is a recently described geriatric syndrome, characterized by the concomitant presence of sarcopenia and osteoporosis. A recent meta-analysis showed that osteosarcopenia significantly increased the risk of falls, fracture, and mortality.24 A greater prevalence of osteosarcopenia in females compared to males, primarily attributed to postmenopausal hormonal changes affecting both bone and muscle metabolisms and to the tendency for women to have lower weight and BMD than men of the same age has been described.24 While the epidemiology of osteosarcopenia provides valuable information, the biggest potential of this newly described geriatric syndrome may lay in the possibility to develop synergistic therapeutic interventions, involving physical exercise and physical medicine and rehabilitation notably.25,26

Three strengths of our study could be highlighted. First, the study counts with the highest methodological quality, including a large sample size and a rigorous osteoporosis and sarcopenia assessment, with data obtained from systematic annual interviews, including validation of all past fractures in the radiological reports, over more than a decade. To authors’ knowledge, this is one of the largest studies including a rigourous osteoporosis and sarcopenia assessment available to date. Second, the study is highly innovative, as to the best of our knowledge, the relationship between BMD change over a decade and an objective evaluation of muscle strength and mass in a large group of postmenopausal women has not been reported before. Finally, the study is timely, as the study provides valuable data about the longitudinal relationship between the two diseases and may help to lay the grounds to unravel the crosstalk between bone and muscle.

Five limitations should be acknowledged. First, the retrospective study design does not allow to establish cause-effects relationships, but associations. Second, there is a selection bias, as the study was conducted only in those participants who were able to attend the evaluation centers on site, potentially leading to underrepresentation of frail women or those who had died in the previous years before study baseline. Moreover, the study mostly considered older white European women, which may limit the generability of the findings, and thus not be valid for men or other populations.27 Third, no validated assessment tools for the nutritional status and physical activity evaluation were used, which could also be considered as a limitation of the study. Fourth, inflammatory markers were not included, although they could be key confounders in the decline of muscle function and bone mass. Given the current evidence linking low-grade chronic inflammation to osteosarcopenia and the high potential of biomarkers, further studies incorporating inflammatory markers are warranted to better understand and treat musculoskeletal system during the aging process.28,29 Finally, there was an heterogeneity in the time between the two DXA examinations, which may vary from 6 to 16 years among participants, with 75% of the sample ranging from 10 to 13 years. Thus, the binary variable for BMD loss used in the analysis may have grouped older women with different profiles and characteristics, as their losing rate may vary widely.

In summary, we observed a two-fold higher odds for probable sarcopenia in the community-dwelling older women who lost more than 3% of their total hip BMD over the past years. Our findings also suggest that the systematic implementation of the inexpensive and easy-to-perform handgrip strength assessment into osteoporosis and comprehensive geriatric evaluations in clinical practice, particularly in Physical Medicine and Rehabilitation or Geriatric Medicine, could serve as a supportive tool in osteoporosis assessment. It may help identify individuals who have experienced a clinically relevant BMD loss in recent years and are therefore at increased risk of osteoporosis and fractures. Future studies should examine concomitant changes in sarcopenia and osteoporosis in a prospective manner, identifying modifications in their components over time to provide a more accurate perspective of the relationship between bone and muscle.

Ethical approval for this study was obtained from the Comité d’éthique du CHU Brugmann (Approval Number: CE 2021/91).

Written informed consent was obtained for anonymized patient information to be published in this article.

The study was supported by the Fondation Brugmann, CHU Brugmann, IRIS-Recherche, and Fondation Vésale.

Jean-Jacques Body reports financial support provided by UCB Inc. Dolores Sanchez-Rodriguez reports personal speaker fees from Nutricia, unrelated to this work. The other authors declare that they have no competing financial interests or personal relationships that could have influenced the work reported in this manuscript.