Study Population
We conducted a prospective randomized controlled study in obese subjects undergoing elective bariatric surgery in Huashan Hospital, Fudan University between January 2018 and March 2020. Inclusion criteria included adult (age between 18 year to 65 year), body mass index (BMI) more than 30 kg/m2, American society of anesthesiologists (ASA) physical status classification 1–2. Exclusion criteria were age less than 18 year or more than 65 year, BMI less than 30 kg/m2, ASA classification 3–4, surgical history of lung lobe resection, respiratory diseases (pneumonia, pneumatoceles, restrictive or obstructive respiratory dysfunction), cardiac diseases (ventricular arrhythmia, congestive heart failure, valvular stenosis or insufficiency), severe liver, kidney or cerebral dysfunction, or refusal to give written informed consent. The subjects were randomly enrolled in a 1:1 ratio to one of the two groups (PCV or PCV-VG mode) based on computer-generated random numbers.
Perioperative Management
All patients were given general anesthesia. The combinations of propofol (1–2 mg/kg), fentanyl (4–6 µg/kg) and rocuronium (0.9mg/kg) were administered for intravenous induction based on ideal body weight (IBW). IBW(kg) for males = 50 + 0.91 × (centimeters of height − 152.4); IBW for females = 45.5 + 0.91 × (centimeters of height-152.4) [6]. 2–3% sevoflurane were given to maintain anesthetic depth combined with fentanyl and rocuronium. Pneumoperitoneum with 12–14 mmHg and reverse Trendelenburg position (50–60 degrees) were utilized in this surgical procedure named laparoscopic sleeve gastroplasty. After surgery, all subjects were immediately transferred to the postoperative care unit (PACU). Neostigmine (0.05-0.07mg/kg) and neuromuscular monitoring (trains of four, TOF) were regularly used to prevent the residual neuromuscular blockade. The subjects’ tracheal tube was not removed until their spontaneous VT was greater than 6ml/kg and the TOF ratio was greater than 90%.
Study Protocol
Study Protocol
Individuals’ demographics and characteristics (e.g. age, gender, height, and weight) were collected by one research coordinator. The surgical and anesthetic characteristics were also obtained including surgical duration, anesthesia duration, and postoperative extubating time. The subjects were randomly ventilated with 8 ml/kg (IBW) plus PEEP 5 cmH2O in PCV or PCV-VG mode. Inspiration and expiration ratio (I/E) was set as 1:2. The gas flow rate was 2ml/min. Recruitment maneuvers, keeping airway pressure at 30 cmH2O for 30 seconds, were performed immediately after tracheal intubation. During surgery, the initial FIO2 was 50%. The FIO2 was increased until the subjects’ SpO2 was greater than 90%. The intraoperative ventilator settings, including VT, Ppeak, PEEP and FIO2, were recorded at T1 (10min after intubation), T2 (10min after pneumoperitonium), T3 (1 hour after pneunoperitonium + reverse trendelenburg position) and T4 (10min after non-pneumoperitonium). Blood gas analyses (PaO2 and PaCO2) were taken at T0 (before anesthesia induction), T1, T2, T3, T4, T5 (10min after tracheal extubation) and T6 (the 1st day after surgery). Heart rate (HR), mean blood pressure (MBP), SpO2 were also recorded from T0 to T5. Perioperative oxygenation index and respiratory dynamic compliance (Cdyn) during pneumoperitoneum were defined and calculated as follows: oxygenation index = PaO2 / FIO2; Cdyn = VT / (Ppeak-PEEP) [7]. On the 1st day after surgery, thoracic CT scanning was taken to identify the occurrence of abnormal radiographic changes (ARCs), such as infiltration, pleural effusion, ateclectasis and emphysema. The tracheal extubation time and the incidence of PPCs were also recorded including pneumonia, tracheal reintubation, respiratory failure, and death prior to discharge.
Statistical Analyses
In the current research, the subjects’ respiratory Cdyn during pneumoperitoneum was the primary study endpoint. According to our pilot study, the Cdyn value in PCV group was 32.04 ± 7.13 ml/cmH2O. We hypothesized that the PCV-VG mode could improve Cdyn from 32 ml/cmH2O to 37ml/cmH2O. Forty-three subjects per arm (86 total) should be recruited with 90% power to detect the difference of 5 ml/cmH2O in Cdyn between the two ventilation modes (PCV and PCV-VG) at a level of 0.05 significance. Equal variances of 5 ml/cmH2O were assumed in this study. Accounting for 20% loss to follow-up, at least 52 subjects (104 total) should be enrolled per group.
Skewness/Kurtosis test were used to check normal distribution for all continuous variables. Normal distributing variables were reported by means ± standard deviation (SD), and compared by Student t-test. Non-normal distributing variables were presented by median (interquartile ranges) and compared by Wilcoxon rank-sum test. The incidence of PPCs (binary variables) were reported as proportion and compared with Fisher’s exact test. The related factors of postoperative extubation time was identified by Spearman correlation coefficient test, and adjusted by multiple linear regression. The results were presented by coefficient with 95% confidence intervals (CI) and adjusted P values. The multivariable logistic regression analyses were used for the risk factors of postoperative pneumonia and ARCs. The results were presented by odds ratio with 95% CI and P values. The confounders, such as age, gender, BMI, surgical duration, were adjusted by logistic regression analyses. A P value of < 0.05 was considered significant. All analyses were carried out in STATA14.1 (STATA Corp, USA).