Sarcopenic obesity increases the risk of metabolic complications (MC), falls, fractures and even increased mortality. The identification of a low muscle mass (LMM) occurs through the muscle mass index (MMI). The height-adjusted MMI in the elderly is the most used for the diagnosis of sarcopenia (SARC). However, in the young obese subject there is data in the literature favoring the weight-adjusted MMI, due to a better identification of SARC and correlation with MC. Objective: To identify obese patients with low muscle mass (O-LMM) prior to bariatric surgery through three different MMIs regarding muscle function (hand grip (HG) and gait velocity (GV)), bone mineral density (BMD) and metabolic profile. Methods: The cross-sectional study, involving 59 obese women of a public hospital in Brazil, was conducted. The sample was divided into two groups according to the presence of O-LMM and obese patients with normal muscle mass (O-NMM) for each IMM. We considered O-LMM patients who were in the lowest quintile for each IMM and the patients who were in the other quintiles were considered as O-NMM. The body composition was evaluated by bio-impedance (inbody-370), multifrequency (5, 50, 250 HZ), with fasting of 12 hours and BMD by lunar densitometer prodigy advance. To estimate muscle mass, the following MMIs were used: height-adjusted MMI (MMI-height): appendicular muscle mass (AMM) / height2; MMI adjusted for weight (MMI-weight): AMM / weight x 100 and MMI adjusted for body mass index (MMI-BMI): MMA / BMI. The HG was evaluated by jamar dynamometer (3 measurements with interval of 30 seconds between them) and the physical performance by the GV for 6 minutes. Fasting glycemia, HbA1c, HOMA-IR, insulin, total cholesterol, HDL, LDL, triglycerides and high sensitivity quantitative C-reactive protein were assessed. Results: The participants had a mean age of 39.53 ± 8,99 years, weight: 108.60 ± 13.86 kg and BMI: 42.6 ± 4.64 kg / m2. O-LMM were identified at 30.5% (n = 18) by MMI-weight, and also at 20.33% by MMI-BMI and MMI-height. Patients in the lowest quintile for the MMI-weight had a lower HG (24.16 x 30.69 kg, p = 0.000), lower BMD in L1-L4 (1.17 x 1.27 g / cm2, p = 0.001), in the femoral neck (FN) ( 1.04 x 1.14 g / cm2, p = 0.012) and in the total femur (TF) (1.10 x 1.19 g / cm2, p = 0.039), respectively, but without significant difference for GV. For MMI-BMI, O-LMM presented lower GV (0.93 x 1.06 m / s, p = 0.017) and HG (23.19 x 30.22 kg, p = 0.000), however without significant difference for BMD. The O-LMM identified by the MMI-height had lower BMD in FN (1.01 x 1.13 g / cm2, p = 0.009) and TF (1.06 x 1.19 g / cm2, p = 0.012), with no significant difference for muscle function. Metabolic profile was not worse in the O-LMM by the three MMIs. Conclusion: The MMI-weight identifies more O-LMM patients than the other MMI. The O-LMM diagnosed by MMI-weight had lower BMD in all sites and lower HG than O-NMM, possibly being a better IMM to diagnose SARC in obese patients.