Ｏｂｊｅｃｔｉｖｅ　To investigate the effect of different inhalation oxygen concentration on the lung collapse and postoperative recovery of thoracoscopic surgery.　Methods　Patients who underwent elective thoracoscopic radical resection of lung cancer were randomly divided into group A (n=29) and group B (n=30). Patents in the group A received a gas mixture of oxygen and air (FiO2=50%) whereas and patients in the group B received 100% oxygen(FiO2=100%) until the start of one lung ventilation (OLV). Then, the dependent lungs were ventilated with 50% oxygen during OLV in both groups. The surgeon evaluated the lung collapse using LCS score and Campos score for lung collapse. They were scaled at 1 min after the opening of the pleura(T0), 10 min after the start of OLV (T1), 20 min after the start of OLV (T2) and 30 min after the start of OLV (T3). During two lungs ventilation, arterial blood gases were measured before anesthesia induction, at 10, 20, 30 min after initiation of OLV. The incidence of postoperative atelectasis by chest CT at the third day after operation, hospital length of stay(LOS) and the incidence of patchy consolidation were recorded.　Results　At T1 and T2, the LCS scores and Campos scales of group B were higher than scores of group A (P<0.05), but the difference between the two groups was not obvious at T0 and T3(P＞0.05). The oxygen partial pressure of group B was higher than that of the group A before 20 min of OLV, but the difference were not significant after 20 min. The incidence of postoperative atelectasis, LOS and the incidence of patchy consolidation were not different (P＞0.05)　Conclusions　The inhalation of 100% oxygen before the OLV can make the pulmonary collapse effect better, and it has no effect on postoperative recovery.