Pulmonary Overdistension Assessment With Electrical Impedance Tomography in Patients With Acute Respiratory Distress Syndrom in Prone Position.

Overview

Mechanical ventilation in COPD requires a good interaction Patient-Ventilator without asynchronies in order to reduce mortality. Dynamic hyperinflation with PEEPi is responsible for major asynchronies in COPD. It is supposed to be symmetrical between the 2 lungs with the same PEEPi. EIT can measure the distribution of tidal ventilation and interlung lung insufflation delay on the impedance time curve of each lung. Moreover, it is possible in intubated COPD patients to assess one lung PEEPi during low flow insufflation with pressure-EIT monitoring. This study aims to measure interlungs insufflation delay under mechanical ventilation in COPD patients

Scientific justification: Moderate or severe COPD with respiratory acidosis requires mechanical ventilation whether it is non-invasive or after intubation. There is a risk of various asynchronies due to dynamic hyperinflation, and it has been shown that a high level of asynchronies can be associated with mortality. EIT can measure the regional distribution of tidal ventilation, and its timing. It can also measure unilateral Pressure-EIT curves with low inflection point representing PEEPi. Strategy description: Patients with moderate or severe COPD treated with mechanical ventilation equipped with EIT will be included. EIT will be monitored during tidal ventilation and during PEEP titration by the attending physician as it is already done in our unit. In case of invasive ventilation (NIV failure), a classical low flow PV curve will be done in order to mesure one lung PEEPi. For that the acquisition of EIT data during the low flow PV curve will be analysed secondary with a dedicated software. Reconstruction of EIT derived PV curve will determine the regional PEEPi. Follow up description: * Measurement of EIT data and ventilator spirometry during NIV initiation and PEEP titration. * Blood gases under NIV as usual * Hemodynamic data

Conditions