The study was approved by the Ethics Committee of Harbin Medical University (No.201314), registered at clinical trials.gov (ChiCTR1800015993).
2.1 Screening
Patients who planned to undergo thoracic surgery under general anaesthesia were screened and randomized by the clinical anaesthesia service of our regional university hospital, the 1st Affiliated Hospital of Harbin Medical University, China, between October 2018 and April 2020.
2.2 Inclusion and exclusion criteria
Patients were enrolled if they met all of the following conditions: 1) they were scheduled to undergo lung resection, mediastinal tumour resection, or 3-incision oesophagectomy under general anaesthesia with double-lumen tube; 2) their one-lung ventilation time was expected to last more than 1 h; 3) ASA classification II or III. Patients were excluded if they met any of the following criteria: 1) their body mass index (BMI) was greater than 35 kg/m2; 2) they smoked during the last 2 weeks; 3) they had asthma, chronic obstructive pulmonary disease (COPD) or other chronic lung disease and cannot suffer high mechanical stress during OLV; 4) vital capacity or forced expiratory volume in 1 s <50% of the predicted values in the lung function, 5) numbers of leukocytes less than 4.0 × 109/L or more than 10.0 × 109/L); 6) they had experienced acute lung injury or acute respiratory distress syndrome before surgery; 7) they had received mechanical ventilation within the last 30 days; 8) they were receiving systemic corticosteroid therapy, or intractable shock was considered; or 9) they failed to maintain oxygen saturation of pulse oximeter (SpO2) greater than 95% during two-lung ventilation and greater than 85% during OLV.
2.3 Standard procedures
The patients were pre-medicated with an intravenous injection of midazolam 0.05 mg/kg. Before the patients underwent general anaesthesia, a radial arterial tube was inserted, and restrictive fluid therapy with crystalloids was administered at 2ml·kg−1·h−1to maintain diuresis >0.5ml·kg−1·h−1. A fluid bolus of 250ml of crystalloids was administered when diuresis was <0.5ml·kg−1·h−1 during surgery to stabilize haemodynamics [13]. All of the patients received routine anaesthesia according to protocol, including pre-oxygenated to attain FiO2 100%, intravenous sufentanil (0.25-0.5 μg/kg) and propofol (1-2 mg/kg) at induction 4 minutes after administration of cisatracurium (0.03 mg/kg), a double-lumen tube (Mallinckrodt, 35–37Ch; COVIDIEN, USA) was inserted, and the position of the endobronchial tube was confirmed using a fibre-optic bronchoscope (Olympus Asian, Japan). Thereafter, anaesthesia was maintained with sevoflurane to maintain a bispectral index (BIS) of 40-60; analgesia was provided with a single intravenous injection of sufentanil 5 µg. Cisatracurium was administered every 30 min, and the last administration occurred at least 30 min before the end of surgical suturing and FiO2 was maintained at approximately 0.5 during the whole anaesthesia procedure with tracheal intubation or increased to 100% when SpO2 was less than 90%. Routine intraoperative monitoring included invasive blood pressure, electrocardiography, pulse oximetry, end-tidal fraction of carbon dioxide (PETCO2), BIS and dynamic pressure-volume curve.
2.4 Ventilation protocol
The subjects were randomly divided into two groups: High-Volume group and Low- Volume group. All of the parameters were set by the same type of anaesthesia machine (Drager Fa-bius GS, Lubeck, Germany). Patients in both groups were subjected to two-lung ventilation with the following parameters: a VT of 8 ml/kg predicted body weight (PBW), a PEEP of 5 cmH2O, an inspiratory: expiratory (I:E) ratio of 1:2 and a fresh gas flow of 2 L/min. During OLV, the High-Volume group was maintained at a VT of 8 ml/kg PBW, a PEEP of 5 cmH2O, an I:E of 1:2, and a fresh gas flow of 2 L/min; the Low-Volume group had a VT of 5 ml/kg PBW, a PEEP of 5 cmH2O with recruitment [14] every 30 min, an I:E of 1:2, and a fresh gas flow of 2 L/min. We changed plates to maintain PETCO2 between 27 and 40 mmHg.
2.5 Breath sample and measurement methods
Breath sampling was performed at the following time points for ventilated lung: immediately following induction (T1), OLV 0 min (T2), OLV 15 min (T3), OLV 30 min (T4), OLV 1 h (T5) and immediately at re-ventilation (T6). Off-line sample of gas was sampled via a plastic sampling line placed just distal of the confluence of the ventilated limb of the double lumen tube into a hermetically sealed bag by Gas Sampler (Sunvou, SV-BSDM, Wuxi, China), which kept extracting the expiratory air at a flow rate of 300 ml min-1 when it detected a plateau. After nearly 5 min, the bag was filled and then immediately analysed using a Nano Coulomb Breath Analyzer (Sunvou, CA-2122, Wuxi, China). The analyser examined one gas sample 6 times to calculate an average value; after less than 1 min, ETNO was achieved.
2.6 Intraoperative observation
Blood pressure, heart rates, SpO2, FiO2 and BIS were obtained from a monitor (Datex-Ohmeda S/5, GE Healthcare, Finland, Europe) and were recorded at times entering into operating room (T0)-T6 and immediately at extubation (T7). When the FiO2 increased to 100% but SpO2 was still less than 90%, we defined the patients as hypoxemic. A D-lite sensor (GE Healthcare, Finland, Europe) was used to measure VT, respiratory rate (RR), PETCO2, plateau pressure (PLAT), peak pressure (PEAK) and respiratory system dynamic compliance (COMPL) at T1-T6.
2.7 Blood sample and analysis
The blood sampling time points were the same as breath sampling (T1-T6). Blood samples of approximately 10 ml were taken from arterial indwelling catheters using a sterile centrifuge tube, which had been heparinized. The plasma was collected and stored at -80°C via centrifugation at 3000 rpm for 10 min. The levels of NOS and inflammatory factors were determined using enzyme-linked immunosorbent assay kits (USCN Life Science Inc., Houston, USA).
2.8 Postoperative observation
The following observations were acquired postoperatively: duration of hospitalization, postoperative pneumonia, acute organ failure, primary postoperative intensive care unit (ICU) admission, and in-hospital death. All of the observations were diagnosed by the pneumology department with signs and exact auxiliary examination.
2.9 Statistical analysis
The normality of the data distribution was tested with the Kolmogorov–Smirnov test. The homogeneity of variance was tested by Levene’s test. The data are expressed as either the mean ± SD or median and interquartile range (25–75%), as appropriate. Comparisons of normally distributed variables were performed with two-tailed Student’s t test, whereas a non-parametric test was used for non-normally distributed variables, and a chi-square test was used for categorical data. Statistical significance was indicated at P<0.05. All of the statistical analyses were performed with SPSS13.0 software.