Background: The association between the quantity and composition of skeletal muscle and the decline in physical function in elderly is poorly understood. Therefore, the primary aim of this cross-over study was to investigate the association between thigh intermuscular adipose tissue (IMAT) infiltration, appendicular muscle mass, and risk of fall in postmenopausal osteoporotic elder women. Second, we examined the differences in muscle mass, IMAT, and risk of fall in the same sample of older subjects after being classified as sarcopenic or nonsarcopenic on the basis of the dual-energy X-ray absorptiometry (DXA)-based Appendicular Skeletal Muscle Mass Index (ASMMI). Methods: Twenty-nine subjects (age: 72.4 ± 6.8; BMI: 23.0 ± 3.3; and T-score: −2.7 ± 0.2) completed the following clinical evaluations: (1) whole-body DXA to assess the ASMMI; (2) magnetic resonance to determine the cross-sectional muscle area (CSA) and IMAT of thigh muscles, expressed both in absolute (IMAT<sub>abs</sub>) and relative (IMAT<sub>rel</sub>) values; and (3) risk of fall assessment through the OAK system (Khymeia, Noventa Padovana, Italy). The existence of a correlation between the risk of fall (OAK scores, an automated version of the Brief-BESTest) and the clinical parameters (ASMMI, CSA, IMAT<sub>rel</sub>, and IMAT<sub>abs</sub>) was tested by the Pearson’s correlation index while data homogeneity between sarcopenic and nonsarcopenic subjects was tested through unpaired Student t tests or with the Mann-Whitney rank test. Effect sizes (ES) were used to determine the magnitude of the effect for all significant outcomes. Results: Eleven subjects were classified as sarcopenic and 18 as nonsarcopenic based on their ASMMI (cutoff value: 5.5 kg/m<sup>2</sup>). A positive correlation between OAK and CSA was observed (r<sup>2</sup> = 0.19; p = 0.033), whereas a negative correlation between OAK and IMAT<sub>rel</sub> was detected (r<sup>2</sup> = 0.27; p = 0.009). No correlations were observed between OAK and ASMMI and between ASMMI and IMAT<sub>rel</sub>. Sarcopenic subjects showed significantly lower weight (p = 0.002; ES = 1.30, large), BMI (p = 0.0003; ES = 1.82, large), CSA (p = 0.010; ES = 1.17, moderate), and IMAT<sub>abs</sub> (p = 0.022; ES = 1.63, large) than nonsarcopenic individuals, whereas OAK scores and IMAT<sub>rel</sub> were similar between groups. Discussion/Conclusion: Increased IMAT and lower CSA in the thigh muscles are associated with higher risk of fall while ASMMI, a value of appendicular muscle mass, was not associated with physical performance in older adults.