BACKGROUND: One lung ventilation (OLV) offers special advantages such as good surgical exposure, prevention from contamination of intact down lung, prevention of bleeding from the lung to the noninvolved lung, etc. during many types of thoracic surgery including video-assisted thoracoscopic surgery (VATS). However, it has the disadvantage of possible hypoxemia due to an inevitable shunt. In general, the tidal volume (TV) has been known to have little influence on arterial oxygenation during OLV. The purpose of this study was to examine the changes in peak airway pressure (PAP) and arterial blood gas analysis (ABGA) following a variable TV and respiratory rate (RR) during OLV. METHODS: Twenty-one spontaneous pneumothorax patients scheduled for a VATS were selected randomly. Patients were anesthetized with O2-air-isoflurane (FiO2 60%)-vecuronium after 35F 37F double lumen endotracheal tube intubation. Patient data (PAP, ABGA) was checked after two lung ventilation (T1: 10 ml/kg TV and 12 f/min RR), OLV (T2: 10 ml/kg TV and 12 f/min RR), OLV (T3: 8 ml/kg TV and 12 f/min RR) and OLV (T4: 8 ml/kg TV and 14 f/min RR) in 20 minutes interval. Patient data between groups was compared and analyzed statistically. RESULTS: The PAP of T2 (31 +/- 4.3 cmH2O), T3 (27 +/- 4.2 cmH2O) and T4 (28 +/- 4.3 cmH2O) were significantly higher than T1 (20 +/- 2.9 cmH2O). PaO2 of T2 (132 +/- 26.3 mmHg), T3 (101 +/- 11.8 mmHg) and T4 (114 +/- 13.1 mmHg) were significantly lower than T1 (302 33.5 mmHg), T3 and T4 values were lower than T2 significantly. However, the SaO2 of all tests were over 98% during OLV. Only the PaCO2 of T3 (45 +/- 4.1 mmHg) was higher than T1 (38 +/- 3.8 mmHg) and T2 (40 +/- 5.5 mmHg) significantly. Changes of End-tidal CO2 (EtCO2) were insignificant between TLV and OLV. CONCLUSIONS: In terms of OLV, comparing 10 ml/kg TV to 8 ml/kg TV showed an advantage of decreasing PAP but disadvantages of decreasing PaO2 and increasing PaCO2. So, 8 ml/kg TV during OLV may need corrections of RR following a patient's status.