Bedside lung ultrasound correlates to chest x-ray and histology in relevant sepsis+aspiration+VILI-induced triple-hit model of ARDS in swine

Acute respiratory distress syndrome (ARDS) is a severe form of acute lung injury that has a 40% mortality, with survivors often suffering with chronic lung disease. Early diagnosis is therefore critical for improving ARDS prognosis by employing early therapeutic measures. Here we aimed to test the utility of lung ultrasound (LUS) in detecting and tracking ARDS progression, using a swine model of human-relevant ARDS. We developed a 48 hour ARDS model of sepsis+ aspiration+ VILI-induced ARDS using Yorkshire swine (50 kg). We tracked LUS findings over time, and compared this to P/F ratio, chest x-ray (CXR) and histology findings. Sedated and intubated swine (n=3) were given a 1hr IV infusion of allogenic cecal slurry containing Enterococcus, E.coli and Enterobacter sp (8 x10e7 CFU). followed by autologous gastric fluid administration into the right and left lower lobes (25ml each). VILI induction followed with PIP of 35-45cm H2O (18-20 ml/kg VT), until lung compliance dropped below 30ml/cmH2O, then followed lung protective ventilation. Standard of care was followed for septic shock and ARDS. We found that LUS scores worsened over the first 8hrs in only aspiration-lobes to a max score of 3 (complete consolidation), improving to a score of 1-2 (>2 B-lines to coalescence) by 24hrs. LUS scores then worsened again during 24-48hrs period to a max score of 3, in both aspiration and non-aspiration lobes. CXR and P/F ratio corresponded with these changes. Histology (48hrs) confirmed alveolar proteinaceous fluid interstitial inflammation and microthrombi. LUS detected worsening 4 hrs prior to P/F and CXR changes.