Background <p>Patients undergoing hemodialysis have decreased skeletal muscle mass which is associated with higher mortality and lower physical functioning. Since these patients frequently experience hospitalizations, we hypothesized that hospital admissions could be a cause of loss of skeletal muscle mass.</p> Methods <p>This was a longitudinal observational study among patients on hemodialysis (<i>N</i> = 132). The outcome was the change in thigh muscle cross-sectional area (TMCSA) measured by computed tomography and standardized by squared height (cm²/m²) per year. The exposures were the presence or absence of hospital admissions, the total number of hospital admissions (times), and the total length of hospital stays (nights) between the TMCSA measurements. Association was analyzed by multivariable-adjusted linear regression analysis.</p> Results <p>During a median follow-up period of 3.95 years, 104 patients were hospitalized (hospitalization group), whereas 28 patients were not (non-hospitalization group). Both groups showed a significant decrease in TMCSA, and it was more rapid in the hospitalization group than the non-hospitalization group (<i>P</i> = 0.023). Multivariable-adjusted linear regression analysis showed that the total number of hospital admissions and the total length of hospital stays were the independent factors associated with the decrease in TMCSA.</p> Conclusion <p>This study showed for the first time that hospital admissions were independently associated with the longitudinal decrease in TMCSA in patients undergoing hemodialysis.</p>