In patients with severe acute respiratory distress syndrome, extracorporeal membrane oxygenation (ECMO), which also as known as extracorporeal life support, may be used. This technique helps the lungs by providing oxygenation to the blood via an external gas exchanger and thus participates partially or fully in gas exchange. The ECMO device includes a pump for draining and returning blood at a certain blood flow rate (ECMO blood flow). An ECMO rate that is adapted to the patient's cardiac output (CO) is essential for effective oxygenation for patients. The objective for clinicians is an ECMO blood flow to cardiac output ≥40%, which can go up to 100% as needed. In addition to the expected benefit in the management of the patient with ARDS, measuring CO is, therefore, all the more important in patients requiring ECMO.

Monitoring CO in a patient with ECMO is not only for determining the minimum ECMO blood flow rate but also for optimizing the functioning of the ECMO. However, the validity of techniques for measuring CO in patients with ECMO has been poorly studied. The reliability of the CO measurement by transpulmonary thermodilution is questioned since the extracorporeal circulation may influence the pathway of cold indicator injected into the patients' circulation and the thermodilution curve measured from the femoral arterial is thereby modified.

Acute respiratory distress syndrome (ARDS) is a severe and common condition with up to 45% mortality in ICU. Its respiratory management includes "protective" mechanical ventilation, the use of a positive end-expiratory pressure level, sometimes the performance of recruitment maneuvers or prone position, and the use of neuromuscular blockers. Due to heart-lung interactions, these different strategies can induce significant or even deleterious hemodynamic effects. Thus, hemodynamic monitoring in such patients is essential. It could assess the effect of ventilatory settings and optimize hemodynamic management. Optimization of arterial oxygen transport in hypoxemic patients may require an increase in the patient's cardiac output (CO). Fluid administration is one of the first-line strategies for achieving this objective. However, unnecessary fluid administration can lead to an increase in pulmonary edema and thus increase the duration of mechanical ventilation and hospitalization. Transpulmonary thermodilution is one of the advanced hemodynamic monitoring in ICU, which consists of the injection of three boluses of cold saline of 15 ml through a central venous catheter located in the superior vena cava, causing a drop in blood temperature which is detected by a femoral arterial catheter equipped with a thermistor. It was demonstrated to be a reliable and simple technique at the bedside for measuring CO. Combined with another technique, pulse contour analysis analysis, transpulmonary thermodilution allows to measure CO beat-to-beat. In addition, transpulmonary thermodilution can measure various hemodynamic variables of interest such as the global end-diastolic volume index (GEDVI, cardiac preload index), the cardiac function index (CFI, cardiac contractility index), extravascular lung water index (EVLWI, volume of fluid accumulated in the alveolar and interstitial lung sectors) and the pulmonary vascular permeability index (PVPI). The last two indices are particularly interesting in patients with ARDS, and were shown to be associated with mortality in patients with ARDS independently. In severe ARDS, ECMO may be another choice. This technique helps the lungs to rest by providing oxygenation to the blood via an external gas exchanger and thus participates partially or fully in gas exchange. The ECMO includes a central pump for draining and returning blood at a given blood flow rate (ECMO blood flow rate). The recommended oxygenation target for patients receiving venous-venous extracorporeal membrane oxygenation (VV-ECMO) assistance is when peripheral oxygen saturation (SpO2) is ≥80%. An ECMO blood flow rate adapted to the patient's CO is essential for effective extracorporeal oxygenation. The objective is an ECMO blood flow /CO ≥40%. It can also go up to 100% as needed (8). This rate can be modified by adjusting the central pump. Thus, it is obvious that measuring CO is important in patients requiring ECMO. However, the techniques measuring CO in patients with ECMO have been poorly studied. the reliability of the measurement of transpulmonary thermodilution was questioned since the analysis of the thermodilution curve can be modified by the extracorporeal circulation. This effect does not matter in a low flow extracorporeal deviation, such as that which is achieved for example during a continuous extrarenal purification . Some studies have shown that transpulmonary thermodilution measurements remain reliable when using another extracorporeal circuit, extracorporeal carbon dioxide removal, whose flow rate can go up to 2.4 L / min. It could be different from ECMO, the blood flow of which is higher. The only available study, conducted by Herner et al, demonstrated the reliability of measuring CO by transpulmonary thermodilution during ECMO. However, the ECMO blood flow/CO ratio was only around 40% in this study. For higher ECMO blood flow/CO ratio (> 40%), the problem remains unresolved.

Inclusion Criteria:

• ≥18yrs.
• With acute respiratory distress syndrome.
• Monitored by transpulmonary thermodilution device.
• Implanted by VV-ECMO.
• Affiliated to the French Health Insurance

Exclusion Criteria:

• Pregnancy
• Patient placed under judicial protection
• Patient on state medical aid (AME, Aide Medicale d'Etat)