Intrathoracic oxygen detects alveolar air leak following video-assisted thoracoscopic lung resection

The water submersion test (W-test) has the risk of overlooking air leaks during lung resection. We assessed the relationship between alveolar air leakage and intrathoracic gas concentrations as a novel intraoperative air leak detection method. We compared the W-test results with intrathoracic gas concentrations of desflurane, oxygen (O2), and carbon dioxide (CO2), prospectively measured using a gas analyser built-in into the anaesthesia machine before and after lung resection. Eighty-eight patients were included; 32 patients had a positive W-test, 31 of whom had elevated levels of all three gases. Eight of the 56 W-test negative patients had elevated levels of all three gases and were positive in a reconfirmation W-test. After repair, 27/31 patients demonstrated negative W-tests, but two patients with elevated levels of all three gases concentrations had postoperative air leaks. When intrathoracic O2 concentration increased, all patients showed an air leak. Conversely, no air leak was detected when intrathoracic concentrations without O2 were measured. We found a positive correlation (correlation coefficients: desflurane, r = 0.84; O2, r = 0.80; CO2. r = 0.77) between the number of bubbles observed in the W-test and gas concentrations. Intrathoracic gas concentration measurements could be a useful alternative or complementary method to the W-test for detecting intraoperative alveolar air leaks. Given that high concentration O2 is frequently use for reinflate the collapsed lungs, intrathoracic O2 measurement could be the most effective way to detect alveolar leakage.