Lung Ultrasound, PEEP and Overdistension (LUPO)

Lung ultrasonography (LUS) is a rapid, easy and non-invasive bedside tool for the detection of specific ultrasound patterns, associated with several pulmonary and pleural disorders. The main limitation of studies assessing lung recruitability and lung monitoring during mechanical ventilation is the lack of detection of lung overinflation: the A pattern doesn't change if the lung tissue is completely re-aerated or is overinflated. Considering this, the investigators could not use LUS to guide lung recruitment due to the risk of lung damage. Speckle tracking echocardiography is a gray-scale based technique used for quantification of myocardial strain to provide a quantitative, objective measure of left ventricle (LV) systolic function.

The investigators suppose that lung sliding could be reduced in the same lung region moving from less ventilated to overinflated condition. This is supported by theoretical arguments by some authors but so far it has not been demonstrated.

The investigators suppose that speckle tracking is able to demonstrate a reduction or abolition in pleural sliding when lung tissue is overinflated by higher PEEP after lung recruitment maneuver. The overinflation is diagnosed by EIT and mechanical respiratory measurements (reduction in compliance as ratio between tidal volume over difference between plateau pressure and PEEP) and localized by EIT.

Primary endpoint: agreement between LUS and EIT to diagnose lung overdistension in Acute Hypoxaemic Respiratory failure patients during a PEEP trial.

Patients are ventilated in pressure controlled mode with a tidal volume of 6 mL/kg predicted body weight (PBW), respiratory rate (RR) is maintained as set per clinical management, I:E = 1:2, PEEP level is set at 5 cmH2O and inspiratory fraction of O2 (FiO2) at 100% before starting the protocol.

To check the presence of areas with reduced or absent pleural sliding not depending on overdistention, and to exclude absence of lung pulse due to pneumothorax, a basal LUS examination is performed before recruiting manoeuvre (RM): the ultrasound linear probe (12 megahertz, GE) is positioned parallel into the 4th-5th right intercostal space next to the sternal margin for the first acquisition; then other two acquisitions along the intercostal space from the medial part to the lateral side of the thorax are recorded; the procedure is repeated on the left side resulting in a final acquisition of 3 ultrasound clips of 10 seconds for each hemithorax. If the presence of pneumothorax is suspected or diagnosed (absence of lung pulse), the patient is excluded from the study according to exclusion criteria.

Then at the same level, marked with a dermographic pen, the EIT belt is positioned and a basal assessment of lung impedance is recorded.

After that a 40x40 RM is performed, setting a PEEP of 40 cmH2O for 40 seconds on continuous positive airway pressure mode mode. If hemodynamic instability or decrease in oxygen saturation occurs during the maneuver, RM is interrupted and previous ventilator settings restored until stability is obtained and the patient discharged from the protocol. If the presence of pneumothorax is suspected or diagnosed (absence of lung pulse), the patient is excluded from the study according to exclusion criteria and clinically managed.

Ventilation is then set again in volume controlled mode with a tidal volume of 6 ml/kg and a PEEP set to reach a maximum plateau pressure of 30 cmH20. I:E ratio and RR are left unchanged.

A decremental PEEP trial is performed reducing PEEP value of 2 cmH2O, static compliance is calculated with an inspiratory pause of 3 seconds just before PEEP modification, and a registration of EIT is acquired for each PEEP level. The PEEP trial starts with the maximum PEEP level (PEEPmax, which is 30 cmH2O minus the pressure generated by 6 ml/kg of tidal volume) and stops at 2 cmH2O less than the best PEEP (PEEPmin), identified according to the static compliance value.

At the same time ultrasound linear probe is transversely set 1 cm above or below the EIT belt and clips recorded at each PEEP level as described above (Fig.2). A total of 6 LUS clips for each level of PEEP is recorded.

Once best PEEP is identified, LUS clips are encoded stored on an external memory stick.

EIT images

Ultrasound setting pre-defined are fixed. Clips are analyzed using GE software and longitudinal speckle tracking applied for each pleural clip. A-lines are numbered per clip Measurements Maximal strain, strain rate and time to peak are calculated and stored by a second investigator blinded to respiratory maneuvers. Differences between parameters calculated on basal and on PEEP trial clips are analyzed.

EIT session is analyzed by dedicated software and lung impedance assessed; each LUS clip is matched with correspondent part of EIT image considered two Region of Interest on the upper part of each hemithorax.

Intra-observer variability on speckle-tracking is assessed by intra-class coefficient correlation; correlation between speckle tracking values and PEEP level is assessed by Spearman correlation test; receiving operating characteristic (ROC) curve (Youden's index) is calculated to predict the accuracy of speckle tracking for predicting hyperinflation. EIT and LUS clips are compared with Cohen's K.

A-lines are numbered at different level of PEEP to investigate if the number of A-lines could be related to lung overdistention