© 2018, European Geriatric Medicine Society. Purpose: To determine the relationship of the size of the rectus femoris muscle, assessed by ultrasonography, with parameters of muscle strength and body composition that are commonly used in the case-finding of sarcopenia in rehabilitation patients with chronic obstructive pulmonary disease (COPD). Methods: Cross-sectional pilot study of 18 men with severe COPD and 17 healthy controls. Main outcome variables: cross-sectional area, thickness, and width of the non-dominant rectus femoris muscle obtained by ultrasound, muscle strength determined by voluntary maximum isometric contraction of the quadriceps muscle, and fat-free mass assessed by bioimpedance analysis. Results: Ultrasounds detected differences in the size of the rectus femoris muscle: cross-sectional area was 4.3 (SD 1.05) cm 2 in patients, compared to 5.6 (SD 1.25) cm 2 in controls; patients also presented lower quadriceps strength, and fat-free mass index. Cross-sectional area of the rectus femoris muscle showed a moderate correlation with quadriceps strength (R = 0.497, p = 0.036) and fat-free mass (R = 0.584, p = 0.011). In a multivariate linear model adjusted for age, body mass index, fat-free mass and muscle size, muscle strength was 7.44 kg lower (p value = 0.014) in patients, compared to controls. Conclusions: A causal relationship was observed between the cross-sectional area of the rectus femoris muscle, assessed with ultrasonography, and maximum isometric strength of knee extension in COPD rehabilitation patients. Reduced cross-sectional area was also associated with loss of fat-free mass. Muscle ultrasound and bioimpedance analysis provide complementary and relevant information that could be useful in the case-finding of sarcopenia in COPD patients.