The prevalence of sarcopenia on elderly maintenance hemodialysis (MHD) has been scarcely investigated. Objectives: To investigate the prevalence of decreased muscle mass and strength alone or combined (true sarcopenia) in elderly patients on MHD according to different methods and cutoff limits. Additionally, we evaluated the agreement between dual energy x-ray absorptiometry (DXA) and surrogate methods for the assessment of muscle mass. Design: Observational and cross-sectional study. Participants: Non-institutionalized 102 elderly (age > 60 years) patients on MHD. Measurements: Sarcopenia was considered when the patient fit one criteria for low muscle mass assessed by DXA, bioelectrical impedance (BIA), sum of skinfold thicknesses (SKF), calf circumference and mid-arm muscle circumference (MAMC) and one for low muscle strength evaluated by handgrip dynamometer. Results: Decreased muscle strength was found in 85% of the patients. The prevalence of decreased muscle mass varied from 4 to 73.5% and of sarcopenia (decreased muscle mass and strength combined) from 4 to 63%, depending on the method and cutoff limit applied. A small percentage of patients (2 to 15%) were classified as sarcopenic by more than one diagnostic criteria. The agreement between DXA and the surrogate methods to assess muscle mass showed better kappa coefficients with BIA (r=0.36; P<0.01) and SKF (r=0.40; P<0.01). Conclusion: A wide prevalence of sarcopenia is observed depending on the method and cutoff limit applied. This may limit extrapolate on to clinical practice. BIA and SKF were the surrogate methods to assess muscle mass with the best concordance with DXA in elderly MHD patients.