Ethics, consent and permission
The study was approved by the Ethics Committee of the University Medical Center of Freiburg (Engelbergstr. 21, 79106 Freiburg, Germany, Ethical Committee N° 179/18) on 29th March 2018 (Chairperson Prof. Dr. R. Korinthenberg) and registered at the German Register for Clinical Trials (DRKS00014925). Please note that this study adheres to CONSORT guidelines.
Study design and patient population
In order to cope with potential interindividual variability, the study was designed as a randomized controlled interventional cross-over trial. After obtaining written informed consent, we studied twenty-three obese patients with body mass index (BMI) ≥ 30 kg∙m-2. Patients eligible for enrolment were patients with physical status ASA ≤ III, undergoing elective bariatric surgery. Exclusion criteria were ASA physical status >III, age < 18 years, pregnancy, emergency procedure, cardiac pacemaker and other active implants, chronic obstructive pulmonary disease classified as GOLD stage > II and refusal of participation. The trial was conducted at the University Medical Center Freiburg, Germany. Participants were enrolled and assigned by a study related anesthesiologist. Data were collected at the University Medical Center of Freiburg, Germany.
After obtaining written informed consent, 23 patients were included in the study. After primary recruitment and preoperative evaluation, the patients received routine monitoring (electrocardiography, SpO2, noninvasive blood pressure measurement; Infinity Delta XL, Dräger Medical, Lübeck, Germany) and a 18-20-G intravenous catheter was established. After preoxygenation to an fraction of expired oxygen of 0.8, anesthesia was induced with 0.3-0.5 µg∙kg-1 predicted body weight  iv sufentanil (Janssen-Cilag, Neuss, Germany) and 2-3 mg∙kg-1 iv propofol 1% actual body weight (Fresenius Kabi, Bad Homburg vor der Höhe, Germany). Tracheal intubation was facilitated with 0.6 mg∙kg-1 predicted body weight iv rocuronium (Fresenius Kabi). If the patient required a rapid sequence induction, neuromuscular blockage was performed by the administration of 1.0 mg∙kg-1 predicted body weight iv rocuronium. Neuromuscular blockage was monitored with a mechanomyograph (TOFscan; Dräger Medical). For tracheal intubation, we used tracheal tubes with low pressure cuffs (internal diameter of 7.0-7.5 mm for women and 8.0 mm for men; Mallinckrodt Hallo-Contour; Covidien, Neustadt an der Donau, Germany). After adequate placement of the tracheal tube, propofol was administered continuously (between 110 and 150 µg∙kg-1∙min-1). Potential hypotension (defined as mean arterial pressure < 65 mmHg) was treated with a continuous infusion of iv noradrenaline (0.03-0.2 µg∙kg-1∙min-1). Perioperative volume requirements were addressed with a crystalloid solution (8 ml∙kg-1∙h-1, Jonosteril; Fresenius Kabi). According to our local standard, mechanical ventilation was started as volume-controlled baseline ventilation (Fabius Tiro, Dräger Medical) with a tidal volume of 7 ml∙kg-1 predicted body weight, inspiration-to-expiration ratio of 1:2, a positive end-expiratory-pressure (PEEP) of 9 cmH2O and ventilation frequency set to maintain an end-tidal carbon dioxide partial pressure between 4.7 and 5.1 kPa. These ventilation settings were based on our study protocol and in accordance with our clinical routine in obese patients. After 7 min of baseline ventilation, all patients were randomly allocated to one of two cross-over groups to receive ventilation sequences either VCV-FCV or FCV-VCV for 7 min per ventilation mode. To avoid irritations due to the surgical procedure (e.g. impaired respiratory mechanics by the capnoperitoneum and electrical irritations of the measurement of Electrical Impedance Tomography), our study was performed prior to the surgical intervention. For adequate allocation, a computer generated randomization in blocks was used. Disclosure of the randomization was requested right after induction of anesthesia. A study related anesthesiologist conducted the randomization in blocks, enrolled participants and assigned participants to the interventions. During the study protocol, ventilation variables were kept constant as set during the baseline measurements. To prevent from the risks of extubation and reintubation, FCV was performed by introducing the narrow-bore tracheal tube (Tribute, Ventinova Medical B.V.) into the standard tracheal tube. Blocking the cuff of the Tritube in the lumen of the tracheal tube provided a sufficient seal. By controlling both tube’s markings, placement of the Tritube’s tip exceeding that of the standard tracheal tube by 2-5 mm was ensured, and the potential risk of bronchial intubation was avoided. Respiratory data were collected from both ventilators via the respective serial communication interface and analyzed offline. Electrical impedance tomography (EIT) was performed with PulmoVista 500 (Dräger Medical) in all patients to measure regional ventilation, changes in relative thoracic electrical impedance during the different ventilation phases, relative end-expiratory lung volume (ΔEELV) and to compare the expiratory decrease in intrapulmonary air [10–12].
Ventilation settings during baseline measurements and VCV were identical. In each patient, baseline measurements were performed prior to the intervention. During FCV, patients were ventilated with a constant positive flow during inspiration and a constant negative flow during expiration (Fig. 1). To avoid intrinsic PEEP, the intratracheal pressure is monitored continuously via a dedicated pressure measurement lumen of the Tritube. During FCV, the operator is able to adjust the inspiratory flow rate, inspiration to expiration ratio, peak inspiratory pressure, end-expiratory pressure and the inspiratory concentration of oxygen. In this special ventilation mode, there is no direct way to control minute volume via tidal volumes and/or respiratory rate. However, the respiratory rate depends on the peak inspiratory pressure, the set (positive) end-expiratory pressure, the set inspiratory flow rate, the inspiration to expiration ratio and the patient’s lung compliance . The (end) expiratory pressure was kept constant in all conditions during the study procedure.
End points and data collection
ΔEELV was the primary endpoint of this study. EIT recordings were analyzed using software developed in Matlab (R2014, The MathWorks Inc.). We derived ΔEELV from adjusting end-expiratory impedance changes by tidal volume and tidal impedance changes as described before [7, 10]. As a first step, the lung area estimation method was applied to all EIT recordings to estimate the relevant lung area . Afterwards, global tidal impedance curves were calculated. These curves represent the sum of impedance of all pixels per frame over time. To scale the absolute impedance values to milliliters, the relation between tidal impedance change and tidal volume was used. Changes of the baseline of these tidal impedance curves were determined as estimates for changes of the end-expiratory lung volume. ΔEELV was then calculated as the difference of end-expiratory lung volume during the different ventilation phases . Secondary endpoints were the respiratory system variables: plateau pressure (PPlat), mean tracheal pressure (Pmean), peripheral oxygen saturation (SpO2), fraction of inspired oxygen (FiO2) and quasi-static respiratory system compliance (CRS). To calculate CRS during FCV, the plateau pressure was determined from a short (approximately 0.1 seconds) end-inspiratory pause. This pause is performed automatically by the Evone ventilator (Ventinova Medical B.V.) with every 10th breaths and used to calculate CRS. Non-invasively collected hemodynamic variables included mean systolic blood pressure, mean diastolic blood pressure, mean arterial pressure and heart rate. To compare relative intrapulmonary air distribution, baseline tidal impedance curves for ventral and dorsal lung areas were determined and compared as described before [7, 11]. The differences in mean lung volume (ΔMLV) between baseline ventilation and VCV and FCV were calculated, respectively. Further, the decrease in global thoracic electrical impedance during each ventilation mode was separated into four equal sections (ΔEI25, ΔEI50, ΔEI75 and ΔEI100), then matched with the correlating decrease in tidal volume and compared successively.
Pressure data from the Evone are based on direct tracheal pressure measurement via a dedicated lumen of the Tritube. To allow for comparability of pressure data from both ventilators and to calculate quasi-static compliance of the respiratory system, airway pressure data from the Dräger Fabius Tiro were generally converted into tracheal pressure data by calculating the flow dependent pressure drop across the respective tracheal tube and pointwise subtracting this value from airway pressure . Thus all pressure data in the following refer to the respective tracheal pressure.
The datasets used and analyzed during the current study are available from the corresponding author on request. Please note that EIT data files require large memory. A separate data transfer service will be used to transfer EIT data files.
Sample size calculation and statistical analysis
In regard to previous investigations on gas exchange during FCV in a porcine model of ARDS  the cross-over design (paired test conditions) we assumed a standardized effect size of the primary endpoint of 0.7 (being the quotient of differences in means and SD). To reach a test power of 0.8 and a desired level of significance of 0.05, 19 patients were required. To compensate for potential incomplete data sets, 23 patients were included in the study. Lilliefors tests were used to confirm that the assumed normal distribution cannot be rejected.
Values are presented as mean (standard deviation), unless indicated otherwise. Statistical analysis was done using Matlab (R2014, The MathWorks Inc., Natick, MA, USA). Linear mixed effects model analyses were performed to check for differences between respiratory variables and variables resulting from EIT measurements during the ventilation phases using R based software (jamovi project (2018), jamovi (Version 0.9.2.3), retrieved from https://www.jamovi.org). For each measured primary and secondary endpoint (dependent variable), the influence of the ventilation mode (baseline ventilation, VCV and FCV) and the ventilation sequence (baseline-VCV-FCV, baseline-FCV-VCV) (factors) was investigated. P < 0.05 was considered statistically significant.