Automated and personalized mechanical ventilation in critically ill patients Closing the loop

This dissertation investigates lung ultrasound (LUS), automated ventilation, and personalized mechanical ventilation as strategies to predict and improve patient-centered outcomes in critically ill ventilated patients admitted to the Intensive Care Unit.
The ACTiVE trial compared automated closed-loop ventilation with conventional ventilation in 1,201 mechanically ventilated ICU patients. Automated ventilation did not increase ventilator-free days at day 28 or improve secondary outcomes. A post-hoc Bayesian analysis supported these findings, showing a low probability of clinically meaningful benefit across different prior assumptions. Together, these studies suggest that automated ventilation does not improve patient outcomes when applied broadly in a heterogeneous ICU population.
The PEGASUS study investigates personalized ventilation in patients with acute respiratory distress syndrome (ARDS), using LUS to distinguish focal from non-focal lung morphology and guide ventilation strategies. Its pilot phase demonstrated that LUS-based classification can be performed reliably by local investigators, with substantial agreement with an expert panel and an accuracy of 88%. Protocol adherence was generally good and safety limits were rarely exceeded, supporting the feasibility of this approach.
Finally, this dissertation explored the broader potential of LUS in critical care. A narrative review describes its established and emerging applications, including artificial intelligence-based image analysis. Furthermore, a post-hoc study showed that baseline LUS aeration scores had limited prognostic value. However, early deterioration in lung aeration was associated with increased mortality in patients without ARDS.