Objective: To compare paravertebral block under thoracoscopy with wound infiltration at an early stage after video-assisted thoracic lobectomy surgery. Design: A prospective, randomized, triple-blinded, placebo-controlled trial. Setting: A single-center university hospital. Participants: Patients scheduled for video-assisted thoracic lobectomy surgery between February 20, 2014 and June 1, 2014 randomly were allocated into paravertebral block (PVB) (n = 35) and infiltration (n = 35) groups. Interventions: In the PVB group, 0.5% ropivacaine was injected into the paravertebral space by the surgeon under direct vision with placebo infiltration of saline in the wounds. In the infiltration group, the wounds were infiltrated with 0.5% ropivacaine by the surgeon with a placebo paravertebral block. Subsequently, patient-controlled intravenous morphine analgesia and paracoxib were administered. Measurements and Main Results: The primary endpoints were visual analog scale (VAS) pain scores at rest and on cough 0, 2, 6, and 24 hours after surgery. The secondary endpoints were the total morphine during postoperative 0 hours to 24 hours, adverse events, and patient satisfaction with the analgesia. Sixty-one patients completed the study. VAS score on cough at each time point was significantly lower (p < 0.05) and median (25th, 75th) morphine consumption was lower in the PVB group than in the infiltration group (26 [10, 35] mg and 42 [29, 58] mg, p <0.001, respectively). There was no difference in VAS score at rest. Patients in the PVB group had higher satisfaction with analgesia than in the infiltration group (p = 0.003). Conclusions: As part of the multimodal postoperative analgesia, intraoperative paravertebral block provided better dynamic pain relief and reduced morphine consumption compared with local wound infiltration. (C) 2015 Elsevier Inc. All rights reserved.