BACKGROUND: Respiratory function and pulmonary gas exchange are affected in thoracoscopic procedures where a pneumothorax is introduced using CO2. Carbon dioxide absorption into the blood during thoracoscopic surgery using intrathoracic carbon dioxide insufflation may lead to respiratory acidosis, increased ventilation requirements, and possible serious cardiovascular compromise. In the present study, the effects of propofol on pulmonary gas exchange were compared with those of enflurane in patients undergoing transthoracic endoscopic sympathectomy (TES) during standard tube ventilation with CO2 insufflation to the surgical side in a prospective randomised manner. METHODS: Sixteen patients with ASA physical status I were divided randomly into enflurane or propofol groups. After induction of anesthesia, patients were ventilated in the same mode in each group. Heart rate and mean arterial pressure were checked and blood gas analyses were performed at 3 time points: 10 min after induction of anesthesia (stage 1), 10 min after CO2 insufflation to the surgical side (stage 2), 10 min after CO2 desufflation from the surgical side (stage 3) under two lung ventilation during TES. RESULTS: Arterial oxygen tension (PaO2) did not differ significantly between enflurane and propofol groups and showed no difference among each stage. PaCO2 and end-tidal CO2 (PetCO2) increased significantly at stage 2 compared to stage 1 and then decreased significantly at stage 3 compared to stage 2 (P < 0.05). There were no significant changes in the mean arterial pressure throughout the procedure with CO2 insufflation in both groups, whereas heart rate was significantly lower in the propofol group than in the enflurane group (P < 0.05) at each stage. CONCLUSIONS: This study demonstrates that pulmonary gas exchange in patients with TES using standard tube ventilation with CO2 insufflation to the surgical side is not affected by choice of anesthesia.