OBJECTIVE: Chronic obstructive pulmonary disease is currently the fourth leading cause of death in the world. Pulmonary rehabilitation is recommended for chronic obstructive pulmonary disease. MATERIAL AND METHODS: This study aimed to evaluate the effects of non-invasive ventilation, supplemental oxygen, and exercise training and supplemental oxygen during exercise training during pulmonary rehabilitation practice in comparison with only exercise training on lung functions, blood gases, lactate levels, respiratory muscle pressures, dyspnea, walking distances, quality of life, and depression in patients with severe chronic obstructive pulmonary disease. The main outcome measure is exercise capacity (6-minute walk test), and the secondary end-point included quality of life. RESULTS: Thirty-five patients (mean ± SD age, 65.4 ± 6.5 years) with a mean bronchodilator forced expiratory volume in the first second of expiration of 39.4 ± 7%, undergoing an 8-week outpatient pulmonary rehabilitation, were randomized to either non-invasive ventilation, supplemental oxygen, and exercise training, supplemental oxygen during exercise training, or exercise training groups. The improvements in respiratory muscle strength were higher in non-invasive ventilation, supplemental oxygen, and exercise training patients than the moderate improvements in the exercise training group. Both non-invasive ventilation, supplemental oxygen, and exercise training and supplemental oxygen during exercise training groups showed significant increases in the 6-minute walk test and incremental shuttle walk test. However, the increase in walking distance was better in non-invasive ventilation, supplemental oxygen, and exercise training group (69.8 ± 53.2 m in 6-minute walk test and 66.6 ± 65.2 m in incremental shuttle walk test, P = .001 and P = .005, respectively) compared to supplemental oxygen during exercise training group (42.5 + 55.5 m in 6-minute walk test and 53.5 + 70.2 m in incremental shuttle walk test, P = .01 each, respectively). The total St. George's Respiratory Questionnaire score was similar in all study groups after the intervention. Symptoms of depression significantly improved only in non-invasive ventilation, supplemental oxygen, and exercise training group (-2.8 + 2.8, P = .006). CONCLUSION: Non-invasive positive-pressure ventilation (NIPPV) added to supplemental oxygen during exercise training was associated with better physiological adaptations than other modalities. [ABSTRACT FROM AUTHOR]