Studies on respiratory muscle resting by negative pressure ventilation (NPV) in patients with stable COPD have given conflicting results. Probable explanations lie in criteria of patients' selection, method of NPV application, and lack of supervision of respiratory muscle rest. Thirteen hypercapnic patients with COPD were, therefore, randomly assigned to either a NPV group or a control group. The NPV was applied by an airtight jacket (pneumosuit), 5 h a day, 5 consecutive days a week for 4 weeks. Both NPV group and control group performed in-hospital pulmonary rehabilitation program for a 4-week period. Arterial blood gases, spirometry, maximal inspiratory pressure (MIP) and maximal expiratory pressure (MEP), breathing pattern, and electromyogram (EMG) of the diaphragm and parasternal intercostal muscles were measured on the preintervention day, and at the end of the second and fourth weeks of treatment (days 13 and 27, respectively). The short-term effect of NPV on EMG suppression was also checked throughout the ventilatory sessions in three different days (1, 12, and 26, respectively). A 6-min walking test (WT) and level of dyspnea by a modified Borg scale were evaluated on the preintervention and the last days. Negative pressure ventilation resulted in a significant reduction in EMG activity of both diaphragm and parasternal muscles, associated with significant increase in MIP, tidal volume, and ventilation, and increase in PaO2 and decrease in PaCO2. A significant relationship between change in MIP and change in PaCO2 was observed (r = 0.72, p < 0.01). Improve-ment in 6-min WT and dyspnea sensation was also observed, both being the sole changes in the control group. These data seem to indicate a beneficial role of respiratory muscle rest in improving respiratory function. Adequate supervision by personnel familiar with the equipment is likely to contribute to successful treatment.