Diagnostic approaches, ventilation and fluid management in patients with acute hypoxaemic respiratory failure

This thesis investigates key aspects of ventilation, fluid management, and diagnostic approaches in critically ill patients with acute hypoxemic respiratory failure, including those with COVID-19. It explores differences in ventilation strategies, aspects of fluid management, and the impact of advanced imaging techniques on diagnosing and managing acute respiratory distress syndrome (ARDS).
A central focus is the comparison of ventilation management between pre-pandemic ARDS and COVID-19 ARDS. The findings indicate that key ventilator settings differ significantly between these patient groups, with implications for clinical outcomes. Another key aspect is the comparison of high-flow nasal oxygen (HFNO) and invasive ventilation in COVID-19. Using the Berlin definition’s PaO2/FiO2 cutoffs resulted in three cohorts with varying mortality rates, highlighting differences in risk classification and patient outcomes.
Fluid management is also extensively analyzed, with particular attention to medication-related fluid creep. The findings suggest that a substantial portion of daily fluid intake in critically ill COVID-19 patients originates from intravenous medications, highlighting opportunities for fluid reduction through early oral transitions. Additionally, a higher cumulative fluid balance is associated with prolonged ventilation and worse outcomes.
Overall, this research provides insights into optimizing ventilation strategies, refining fluid management, and improving diagnostic approaches in critically ill patients, with potential implications for future clinical practice.
Finally, this thesis assesses ARDS classification accuracy, showing that an 8-grade confidence scale with lung ultrasound and CT improves consistency. It also examines prone positioning’s effects on lung aeration, inhomogeneity, and compliance in spontaneously breathing, invasively ventilated COVID-19 patients using electrical impedance tomography (EIT).