Study objective: To evaluate the contribution of body composition measurements to clinical assessment in patients on home nasal positive-pressure ventilation for chronic hypercapnic respiratory failure (CHRF), and their relationship to respiratory impairment. Methods: Patients with CHRF (restrictive lung disease (RLD), n = 37; chronic obstructive pulmonary disease (COPD), n = 19), during elective yearly evaluations underwent pulmonary function testing (forced expiratory volumes, arterial blood gases, maximal. inspiratory and expiratory pressure (PImax or PEmax)), and bioetectrical impedance analysis to determine fat-free mass (FFM) index (kg/mw(2)) and body fat mass index. Results: When compared with age- and sex-matched healthy controls, RLD patients (OR 5.5, CI 1.9-15.6, P < 0.002) and COPD (OR 5.2, CI 1.1-24.9, P = 0.04) were significantly more likely to have a tow FFM index. Roughly one-half of patients with RLD and one-third with COPI) had abnormally low FFM index. Estimation of nutritional status by body mass index (BMI) atone clearly underestimated the prevalence of FFM index depletion. Muscle mass assessed by FFM index explained 26% of variance Of PImax (P < 0.001) and 27% of that of PEmax (P < 0-001) Conclusion: BMI alone clearly underestimated FFM depletion, and presence of a very high body fat mass index. Indeed, normal or high BMI can be associated with FFM depletion. Because of its relationship to respiratory muscle strength, an assessment of FFM appears to be valuable in CHRF. (c) 2005 Elsevier Ltd. All rights reserved.