Background: Lung volume reduction surgery improves quality of life, exercise capacity, and survival in selected patients but is accompanied by significant morbidity. Bronchoscopic approaches may provide similar benefits with less morbidity. Methods: In a randomized, sham procedure controlled, double-blind trial, 277 subjects were enrolled at 36 centers. Patients had emphysema, airflow obstruction, hyperinflation, and severe dyspnea. The primary effectiveness measure was a significant improvement in disease-related quality of life (St. George's Respiratory Questionnaire) and changes in lobar lung volumes. The primary safety measure was a comparison of serious adverse events. Results: There were 6/121 (5.0%) responders in the treatment group at 6 months, significantly >1/134 (0.7%) in the control group [Bayesian credible intervals (BCI), 0.05%, 9.21%]. Lobar volume changes were significantly different with an average decrease in the treated lobes of -224mL compared with -17mL for the control group (BCI, -272, -143). The proportion of responders in St. George's Respiratory Questionnaire was not greater in the treatment group. There were significantly more subjects with a serious adverse event in the treatment group (n=20 or 14.1%) compared with the control group (n=5 or 3.7%) (BCI, 4.0, 17.1), but most were neither procedure nor device related. Conclusions: This trial had technical and statistical success but partial-bilateral endobronchial valve occlusion did not obtain clinically meaningful results. Safety results were acceptable and compare favorably to lung volume reduction surgery and other bronchial valve studies. Further studies need to focus on improved patient selection and a different treatment algorithm.