The presence of bilateral lung infiltrates on chest imaging is a mandatory criterion for the diagnosis of acute respiratory distress syndrome (ARDS). This retrospective single-center observational study assessed the agreement between chest radiograph interpreted by intensivists and chest computed tomography interpreted by radiologists for the diagnosis of ARDS in mechanically ventilated patients with hypoxemia. Chest CT was considered the reference standard. The study evaluated interobserver agreement and diagnostic performance of chest radiography in the overall population and in predefined subgroups.

Retrospective, single-center observational study conducted in the medical ICU of University Hospital of Poitiers, France, from January 1, 2015 to December 31, 2021. Included were intubated patients with PaO2/FiO2 ≤ 300 mmHg, PEEP ≥5 cmH2O, suspected ARDS except for imaging criterion, who underwent both chest CT and chest radiograph on the same day. Chest CT was interpreted by radiologists and used as reference for ARDS diagnosis. Chest radiographs were interpreted by intensivists. Agreement and diagnostic performance were assessed.

Inclusion Criteria:

• Adults admitted to intensive care unit
• Invasive mechanical ventilation
• Chest CT performed during ICU stay
• PaO2/FiO2 ≤ 300 mmHg with positive end-expiratory pressure ≥ 5 cmH2O the day of the CT-scan
• Chest radiograph performed within ± 24h of CT-scan

Exclusion Criteria:

• Cardiogenic pulmonary edema as the cause of hypoxemia
• Missing imaging data

Patients admitted to the medical intensive care unit of the University Hospital of Poitiers

Data will be available on reasonable written request to the principal investigator, one year after publication of the main manuscript.

Start date: one year after publication of the main manuscript End date: two years after publication of the main manuscript