Increased effort during partial ventilatory support is not associated with lung damage in experimental acute lung injury

Background An on-going debate exists as to whether partial ventilatory support is lung protective in an acute phase of ARDS. So far, the effects of different respiratory efforts on the development of ventilator-associated lung injury (VALI) have been poorly understood. To test the hypothesis whether respiratory effort itself promotes VALI, acute lung injury (ALI) was induced in 48 Sprague Dawley rats by hydrochloric acid aspiration model. Hemodynamics, gas-exchange, and respiratory mechanics were measured after 4 h of ventilation in pressure control (PC), assist-control (AC), or pressure support with 100% (PS100), 60% (PS60), or 20% (PS20) of the driving pressure during PC. VALI was assessed by histological analysis and biological markers. Results ALI was characterized by a decrease in PaO2/FiO2 from 447 ± 75 to 235 ± 90 mmHg (p < 0.001) and dynamic respiratory compliance from 0.53 ± 0.2 to 0.28 ± 0.1 ml/cmH2O (p < 0.001). There were no differences in hemodynamics or respiratory function among groups at baseline or after 4 h of ventilation. The reduction of mechanical pressure support was associated with a compensatory increase in an inspiratory effort such that peak inspiratory transpulmonary pressures were equal in all groups. The diffuse alveolar damage score showed significant lung injury but was similar among groups. Pro- and anti-inflammatory proteins in the bronchial fluid were comparable among groups. Conclusions In experimental ALI in rodents, the respiratory effort was increased by reducing the pressure support during partial ventilatory support. In the presence of a constant peak inspiratory transpulmonary pressure, an increased respiratory effort was not associated with worsening ventilator-associated lung injury measured by histologic score and biologic markers.

Sedation protocol: The animals were anesthetized with pentobarbital 55 mg/kg IP (Ceva Sante Animale, Montreal, QC, Canada) to provide surgical anesthesia for vessel cannulation. Once surgical anesthesia was confirmed, the animals were placed in supine position on the operation table, and the neck and femoral regions of the body were prepared for cut down and vessel cannulation. All animals received an intravenous infusion of remifentanil 4 µg/ml, diluted in saline (Ultiva®, Abbot Laboratories, QC, Canada) at a continuous rate of 0.4 µg /kg/min and intermittent doses of pentobarbital 0.25 mg IV every 30 min to maintain sedation and minimize pain and stress during mechanical ventilation. This dose was chosen based on physical assessments of the animals (toe pinch and tail flick), which had resulted in no responses from the animals to these stimuli in a previous methodological study (Ismaiel N, Chankalal R, Zhou J, Henzler D. Using remifentanil in mechanically ventilated rats to provide continuous analgosedation. J Am Assoc Lab Anim Sci. 2012;51(1):58-62). Hemodynamics, respiration and visual appearance were monitored continuously, while the level of sedation was additionally assessed every 30 min by the toe pinch and tail flick methods. In case of a reaction to stimuli or spontane-ous movement a bolus of 0.2-0.4 µg/kg was allowed and additional pentobarbital 0.5 mg IV as rescue medication. All animals received 3-5 ml/h of NaCl 0.9% (Hospira, Montreal, QC, Canada) including occasional boluses as needed to flush the lines. Animals were kept supine throughout the protocol and were not restrained to notice any fleeing tendency due to inadequate sedation.
Instrumentation: A 14G endotracheal tube was inserted through a tracheostomy and animals were connected to a standard neonatal ventilator (EVITA4, Draeger Medical Canada Inc, Richmond, ON). 20G catheters were inserted into the common carotid artery and external jugular vein for monitoring of blood pressure. The femoral artery was cannulated with a thermocouple probe (ADInstruments Inc., Colorado Springs, CO) for cardiac output measurements.
Blood samples (0.3 ml) were analyzed for gas pressures and acid-base status and for oxygenation using a species adjusted hemoxymeter (ABL510 & OSM3, Radiometer Copenhagen, Denmark).
A fluid filled 20G catheter with multiple perforations was inserted into the esophagus for measurement of esophageal pressure and attached to a physiologic pressure transducer. The catheter was advanced into the stomach and slowly retracted until cardiac pulsations were minimal. Before each measurement, the catheter was flushed and re-zeroed to ensure patency.

Experimental protocol
Lung injury was induced by intra-tracheal instillation of 0.2m HCl as described previously 13 . Due to the different lung volumes, 0.3 ml were used for the right side and 0.2 ml for the left side. For each instillation, animals were positioned on the respective side and ventilated with 10-12 ml/kg BW. After 2 min the remainder was suctioned and the animal placed supine. Lungs were instilled sequentially allowing the animal to recover between instillations. After the manoeuvre animals were closely observed for any evidence of developing intrinsic PEEP (PEEPi). The expiratory flow curve was continuously monitored on the LabView system to assure end-expiratory zero flow. If needed, tracheal secretions were suctioned. Physiologic measurements were performed at baseline (BL) and 1h after induction of lung injury (ALI-BL). Interim measurements were taken every hour. The final measurement (ALI-End) was planned to take place 4 h after ALI-BL. If animals died prematurely, the last set of measurements before death was taken as ALI-End. Animals were only entered into the study and used for analysis, if they had received at least 120 min of the study ventilation after establishment of ALI. The times to measurements for all groups are given in table S1.

Hemodynamic and respiratory measurements
Arterial pressures were directly transduced and recorded. Cardiac output was measured by transcardio-pulmonary temperature dilution of 0.5ml of saline solution (LabChart 6.0, ADInstruments). A data collection system (PowerLab, ADInstruments) was used for recording data and hemodynamic calculations Gas flow and airway pressure (Paw) were measured proximal to the endotracheal tube by a differential pressure transducer from a heated pneumotachograph (PTN 113252, Hans Rudolph Inc., Shawnee, KS). Pressure and flow were zeroed and calibrated before each experiment using a modified manometer and a 20ml calibration syringe. The tidal volume (V T ) was calculated from integrating the flow signal. The minute ventilation (V E ) was calculated by multiplying the V T with the respiratory rate (RR).
The dynamic compliance (C DYN ) was defined as tidal volume (V T ) divided by the pressure difference between end-inspiratory and end-expiratory airway pressure. The maximum difference between inspiratory airway pressure and esophageal pressure (P ES ) was defined as peak inspiratory transpulmonary pressure (P TPi ). The pressure-time product (PTP) was calculated as the area under the baseline from the P ES curve.

S4
The work of breathing (WOB) was calculated separately for total (WOB T ), ventilator (WOB V ) and respiratory muscles (WOB RM ) from the recorded flow and P ES tracings .
The work of breathing (WOB) was generally calculated as * where P is the driving pressure and f is the gas flow derived from application of that pressure. The WOBT was obtained by using transpulmonary pressure !" * WOB V was derived using the P AW applied by the ventilator !" * The work of breathing done by the subject was calculated by using P ES and the work needed to inflate the chest wall of a rat All calculations were done from automated extractions of data from the recorded curves that were entered manually into a spreadsheet (MS Excel). WOB are expressed in J/L of ventilation.

Histopathologic Examination
At the end of the experiment the animals were killed by a barbiturate overdose. The lungs were excised in inflated stated that was maintained with an inspiratory pressure of 10 cmH 2 O. The left lung was ligated and fixated in inflated state with 10% buffered formalin for histopathologic examination. The right middle lobe was ligated, cut and weighted. It was dried for 48 h at 37°C and weighted again. The ratio wet-to-dry ratio was calculated. Broncho-alveolar lavage fluid (BALF) was obtained from repeated lavage of the right lung (3ml), snap frozen and stored at -80° for subsequent analysis.

Tables
PC pressure-control; AC assist-control; PS pressure support ventilation with 100%(PS100), 60%(PS60) or 20%(PS20) of previous pressure control level. Figure S3: Plot of mean arterial pressure vs. work of breathing. The more of the work performed by the ventilator, the lower the arterial pressure (r=-0.389, p=0.007) Figure S4. Wet-to-dry ratio (W/D) of rat lung. Boxplot.