Abstract
BACKGROUND: Patient-ventilator asynchrony (PVA) is common in critically ill patients undergoing mechanical ventilation and may adversely affect clinical outcomes. Traditional bedside assessment methods are subjective and intermittent. We developed a real-time digital platform to continuously monitor ventilator waveforms and quantify overall asynchrony burden of severe pneumonia patients. METHODS: The study retrospectively analyzed mechanically ventilated coronavirus disease 2019 (COVID-19 patients admitted to the Department of Critical Care Medicine of Peking Union Medical College Hospital (PUMCH) from December 2022 to August 2023. Ventilator waveforms were continuously collected and processed using the remote ventilate view platform, which automatically identified eight PVA subtypes and calculated the Overall Asynchrony Index (OAI) across the full ventilation course. Respiratory mechanics were also extracted. Primary outcomes included intensive care unit (ICU) mortality and 28-day ventilator-free days (VFDs), while secondary outcomes included the length of ICU stay and duration of mechanical ventilation. The study used R, Jamovi, and Python for statistical analysis. RESULTS: Twenty-three mechanically ventilated COVID-19 patients admitted to the ICU at Peking Union Medical College Hospital were included in this study. No correlation was found between the index and ventilatory parameters, compliance, and disease severity. Patients with an OAI ≥10 % were more likely to have fewer 28-day VFDs (1.3 days vs. 11.4 days, P = 0.027) and were demonstrated to have a higher ICU mortality (66.7 % vs. 18.2 %, P = 0.036). Among eight types of PVA, flow insufficiency was found to be associated with prognosis (P = 0.012). OAI correlated with the prognosis of COVID-19 patients. Patients with an OAI ≥10 % were more likely to have fewer 28-day VFDs and higher ICU mortality. CONCLUSIONS: A higher OAI and increased flow insufficiency were associated with worse outcomes in COVID-19 patients receiving mechanical ventilation. This study demonstrates the feasibility and clinical potential of a real-time, platform-based approach for automated detection and longitudinal monitoring of PVA.