The major finding of the present study is that almost two weeks of disease-related immobility result in significant thigh muscle mass loss of 5.0% in a group of immobile older patients admitted to an acute care geriatric unit, while such an effect was not seen among mobile older patients. Notably, this substantial reduction occurred despite providing the medical support and training therapy offered in geriatric units. It can only be speculated how much muscle mass and strength would have been lost without this support. The detrimental impact of bed rest on muscle mass and strength has been already reported in previous experimental models of immobilization in healthy older adults. Studies using muscle mass disuse model have shown approximately 2-6% reduction of leg muscle mass following 5 to 14 days among this population [16, 18, 41].
However, only few published studies have investigated the actual muscle mass loss in a clinical setting. Namely, the effect of immobilization on muscle mass remains to be elucidated in acutely ill older hospitalized patients who are likely to experience a more pronounced loss of muscle mass due to their condition. Hospitalization is commonly accompanied by enforced bedrest or poor mobility induced by pain, surgical trauma, infections and mental stress, leading to changes in food intake and skeletal muscle catabolism [19, 42, 43]. Consequently, we hypothesized that even a short period of hospital stay may result in a significant loss of muscle mass and strength in patients with severe mobility limitation. In a recent observation study in older patients undergoing elective hip replacement, Kouw et al.  reported a significant loss of thigh muscle CSA by 4.2% ± 1.1% (0.6% per day) using CT scan during one-week hospitalization. Our findings were similar, although our patients were older and experienced 5.0% reduction in MRI-derived mid-thigh muscle CSA during the course of a 13-day hospitalization period (0.4% per day). Notably, disease-related immobilization in our cohort started up to three weeks before admission to our department and thus before the first MRI scan. In line with our findings and the aforementioned study, we assume that within the first days of immobilization, the rate of muscle mass loss may be even higher than in our study and may decrease over time, due to metabolic adaptations.
In another prospective study of 63 critically ill patients (mean age 54.7 years), Puthucheary et al.  has indicated the significant decline in the ultrasound-derived rectus femoris CSA by 17.7% at day 10 of hospitalization. However, the changes in muscle mass were greater compared to our findings. Although, direct comparison is difficult since that study used a different method for assessing muscle mass and patients were younger and suffered from multi-organ failure. Nevertheless, it has to be noted that in that study reduction of muscle mass was more severe among those with multi-organ failure compared with single organ failure. Indeed, despite the fact that immobilization is a factor in development of muscle atrophy, patients with multi-organ failure may develop more muscle mass loss as a result of greater metabolic changes [45, 46] and other deleterious factors associated with severe disease, rather than immobility alone . In the present study, despite the similar mid-thigh muscle CSA in both mobile and immobile older adults at baseline, the MRI scans, which provide a very sensitive and accurate measurement, clearly indicated the substantial decline in muscle mass only in immobile older patients during hospitalization. Moreover, the majority of our immobile patients were frail, probably sarcopenic and were at risk of malnutrition or malnourished at the time of admission. Therefore, the combination of these factors may have affected the extent of muscle mass loss in our study [19, 42, 43].
In addition, prior researches on the morphologic changes associated with immobility in older adults have commonly concentrated on muscle mass whereas changes in subcutaneous or intermuscular fat have not received a great deal of attention. However, this knowledge is important since alterations in adipose tissue are linked with dysfunction and metabolic changes in skeletal muscle [47, 48]. Indeed, interaction between adipose tissue and muscle mass is influenced by mobility and aging. Mobility limitation caused by aging leads to decline in muscle mass and function and alteration in body fat composition . With advancing age, intermuscular adipose tissue increases  and subcutaneous tissue decreases . These significant changes in fat composition may have a negative impact on health outcome in old age. Fatty infiltration of the skeletal muscle is a metabolically active component of muscle and affects muscle strength and muscle quality . It secretes inflammatory cytokines which negatively impact muscle cell proliferation and differentiation . The findings of the present study demonstrated a significant decline in mid-thigh subcutaneous fat area in immobile patients without changes in intermuscular fat. It has been previously shown that immobilization leads to increased intermuscular fat [51, 52]. For instance, intermuscular adipose tissue of thigh increased in healthy young patients during four weeks of immobilization  and in patients with spinal cord injury . However, this could not be shown in the period of 13 days of immobilization in our study. This discrepancy could be a result of differences in the study populations and in length of the follow-up period. Since the current sample included ill older adults with several risk factors such as malnutrition, frailty and severe disease, immobilization may specifically and differentially affect both adipose tissues in our population. From a metabolic point of view, our findings indicate that subcutaneous and intermuscular fat may have structural and functional differences in response to immobilization and are subject to distinct dysfunctional changes caused by disease, aging and lifestyle. Indeed, subcutaneous adipose tissue is metabolized during periods of immobilization and decreased nutritional intake and seems to be metabolically more active compared to intermuscular fat. Moreover, previous cross-sectional studies have demonstrated that a greater fat infiltration of the muscle is an independent risk factor for mobility limitations and is a potential contributor to decreasing muscle strength and muscle quality in older individuals [1, 23].
Our results indicate a significant decrease in isometric knee extension strength and nearly no change in handgrip strength during hospitalization of acutely immobile patients. Accordingly, it could be that even short periods of immobilization do not only influence muscle mass but may have also negative effects on muscle strength and physical functioning of lower extremity of older adults and are likely to contribute to impaired recovery, increased readmissions and a higher mortality rate after discharge . Interestingly, the distinct loss of lower and upper extremity strength demonstrates that disease-related immobility has a more pronounced effect on leg muscle strength compared to hand grip strength. Therefore, muscle strength of the leg seems to be the most relevant parameter for functional decline and can reflect mobility limitation whereas upper muscle strength is more related to general body composition. Hence, measurements of leg strength should receive more priority compare to hand grip strength, especially when studying older persons.
Some limitations of the study need to be addressed. Mobility status was defined according to walking ability as described by the BI, which may be imprecise. Nevertheless, a previous study in patients with stroke  has demonstrated that measurement of mobility as measured by the BI is reliable and agreement was generally high for total BI and walking ability. In addition, there was a shorter follow-up period of MRI scans for some patients during hospitalization, mostly due to organizational issues. However, this did not differ between both groups. Finally, due to the relatively small number of immobile patients (n=22), we were unable to reliably examine the individual contribution of risk factors such as disease severity, malnutrition, surgical trauma, inflammation and medication to the actual loss of muscle mass. Thus, this highlights the necessity to establish further studies to address the impact of individual risk factors on the extent of muscle mass loss in the clinical setting.