Objective: To compare a new bronchial blocker, the Cohen blocker, with the Arndt blocker and a left double-lumen tube (DLT). Design: A prospective, randomized, controlled trial. Setting: University hospital. Participants: Forty-eight patients undergoing lung surgery. Intervention: Intubation with 1 of the 3 devices. Comparisons among groups included (1) time for initial positioning, (2) degree of lung collapse at pleura opening, and (3) number of intraoperative fiberoptic examinations. Measurements and Main Results: Positioning of the Cohen blocker (256 [166-341] seconds; median [interquartile range]) took no longer compared with the Arndt blocker (253 [184-305] seconds), and there was a trend toward difference between the 2 blockers and the DLT (137 [102-199] seconds) (p = 0.07). The time to place the Cohen blocker was longer in cases of left bronchus occlusion compared with a right one (340 [300-450] v 170 [124-259] seconds, p = 0.02); they were similar in the Arndt group. The degree of lung collapse was different among groups (p = 0.05), but the difference between any pair did not reach statistical significance. The number of patients who required at least 1 additional FOB examination was not statistically different (50% of patients in each blocker group v 19% in the DLT group). Conclusions: There was a trend toward a difference between times to place a bronchial blocker and the DLT. The Cohen blocker is more difficult to position in the left main bronchus than in the right one. (C) 2009 Elsevier Inc. All rights reserved