Early AI-driven detection of patient-ventilator asynchrony in critical care could help improve patient outcomes.

Patient-ventilator asynchrony occurs when there is a poor interaction between the patient and the ventilator. According to a study by Better Care, a biotech company that develops AI-based software solutions for continuous patient monitoring, this asynchrony is associated with poorer clinical outcomes. Therefore, early detection to allow for prompt clinical intervention is a key factor in improving patient prognosis. These were the main conclusions of a cohort study published in the journal *Critical Care Medicine*, which involved the Intensive Care Units of the Parc Taulí Hospital in Sabadell and the Althaia Foundation in Manresa. The study aimed to identify clusters of patient-ventilator asynchronies, specifically double triggering and ineffective inspiratory efforts in critically ill patients undergoing mechanical ventilation, and to investigate their association with mortality, length of ICU stay, and duration of mechanical ventilation using continuous monitoring software from Better Care. To identify asynchronies and determine their intensity and duration, the researchers processed and analyzed ventilator signals captured continuously. They determined that when these events occur frequently and in temporal clusters, they are associated with worse clinical outcomes. Similarly, they found an association between the duration and intensity of these clusters and the patient's length of stay in the ICU, the duration of mechanical ventilation, and mortality. "These findings represent a step forward toward precision mechanical ventilation and predictive medicine," explains Rudys Magrans, lead researcher of the study, PhD in biomedical engineering, and Director of Research and Development at Better Care, noting that "until now, the influence of asynchrony clusters on patient prognosis was unknown." To more accurately anticipate the prognosis of ICU patients, "Better Care is delving into predictive models that, in addition to mechanical ventilation, take into account other variables such as the patient's pathology or the critical criteria for ICU admission," adds Magrans. AI as a decision-support tool for clinicians "To identify these patient-ventilator asynchronies early, intervene in time, and adapt mechanical ventilation to the patient's needs, continuous and comprehensive monitoring throughout the course of mechanical ventilation is essential," explains Dr. Rafael Fernández, intensivist and co-author of the study. However, clinicians lack the capacity to understand and interpret in real-time the multiple signals and parameters generated by medical devices connected to a patient. For these reasons, Dr. Fernández adds, "it is essential to have appropriate technologies designed for this purpose. Therefore, this study supports the important role that artificial intelligence plays in the hospital environment as a tool for optimizing and processing information, facilitating agile clinical decision-making, and ultimately helping to improve the care of critically ill patients." Reference to the article in *Critical Care Medicine* Magrans R, Ferreira F, Sarlabous L, López-Aguilar J, Gomà G, Fernandez-Gonzalo S, Navarra-Ventura G, Fernández R, Montanyà J, Kacmarek R, Rué M, Forné C, Blanch L, de Haro C, Aquino-Esperanza J, for the ASYNICU group. The Effect of Clusters of Double Triggering and Ineffective Efforts in Critically Ill Patients. Crit Care Med. 2022 Feb 7. doi: 10.1097/CCM.0000000000005471. Online ahead of print.