In patients undergoing lung transplantation (LT), the investigators hypothesize that a "systematic" intraoperative ECMO strategy would reduce the need for invasive mechanical ventilation in the first 28 days without increasing adverse events, as compared to an "on-demand" intraoperative ECMO strategy.

To date, LT remains a highly hazardous procedure. Even if the surgical procedure is well established, the intraoperative support is not, and most intra-operative ECMO decisions rely on local protocols, anesthesiologists' habits, and surgeons' preference.

The efficacy of applying a "systematic" strategy on reducing the occurrence of severe primary graft dysfunction and thus mechanical ventilation in the 28 days following LT, without increasing mortality or morbidity, would support future guidelines on the use of ECMO in the intraoperative period of LT for obstructive and restrictive lung diseases.

Lung transplantation (LT) provides the prospect of improved survival and quality of life for patients with end stage lung and pulmonary vascular diseases. Its performance carries significant adverse effects, being either intra- or postoperative. The ventilation of a diseased lung for sometimes extended periods and the risk of reperfusion oedema and primary graft dysfunction is a challenge. Moreover, significant hemodynamic instabilities episodes might occur, because of pressure on, or displacement of the heart, clamping of the pulmonary arteries and ischemia-reperfusion syndrome. veno-arterial extracorporeal membrane oxygenation (VA-ECMO) has now replaced cardiopulmonary bypass for respiratory and hemodynamic intraoperative support, carrying less side effects, and an improved early survival.

Even though ECMO is a widely used technique, no precise guideline exists on the hemodynamic and respiratory indexes in LT settings to initiate intraoperative ECMO, but only experts' opinion. Besides, it has to be underlined that the rate of LT performed in the absence of any mechanical support is highly variable among centres, ranging from being exceptional up to 70%.

The investigators aim at evaluating two strategies of ECMO initiation in the pre- and intraoperative periods in patients with pulmonary disease requiring LT: an "on-demand" strategy, in which VA-ECMO will be initiated on high hemodynamic and respiratory needs thresholds and a "systematic" strategy in which VA-ECMO will be pre-emptively initiated.

The investigators hypothesize that a "systematic" strategy allows to reduce the risk of severe primary graft dysfunction and the need for mechanical ventilator in the 28 days following LT without increasing adverse events

Inclusion Criteria:

Age >18 years Assessed for bilateral sequential lung transplantation for obstructive or restrictive lung disease Affiliation to the French social security Written informed consent

Exclusion Criteria:

At listing :

• a pulmonary hypertension with a mPAP > 45 mmHg, including in the absence of haemodynamic collapse (MAP, LVEF, RV function all normal)

• a pulmonary hypertension with echocardiographic evidences of right heart dysfunction (paradoxical septum or RV dilatation or RVEF < 35%)

• a pre-capillary pulmonary hypertension at right heart catheterization with low cardiac output

  • LT for primary pulmonary hypertension
  • LT for cystic fibrosis and graft-vs-host disease
  • Re-do LT
  • Combined multi-organ transplantation
  • Active malignancy
  • Pregnancy, breastfeeding
  • Patients under guardianship (tutelle, curatelle, sauvegarde de justice)

Socondary exclusion criteria:

Patients without pulmonary hypertension or with pulmonary hypertension without right ventricular dilatation on an echocardiography in the last 6 months will be randomized; Patients meeting one of the following criteria will not be randomized and will be secondary excluded from the study :

• preoperative severe pulmonary hypertension with hemodynamic collapse on echocardiography defined by: paradoxical septum or dilatation of the right ventricle or RVEF < 20LT in a patient under ECMO as bridge-to-transplantation
• PreLT hypoxemia with PaO2/FiO2 < 80mmHg
• PreLT hypercapnia PaCO2 > 80 mmHg after induction

participant-level deidentified dataset

Up to two years after the last publication of the research results

IPD and any additional supporting information will be available on reasonable request from the corresponding author, for additional analyses such as -but not limited to- metaanalyses. The scientific advisory board of ECMOTOP study will assess the relevance of the request.