Impact of the Display of Inspiratory Muscle Pressure Curves Estimated by Artificial Intelligence on the Ability of Health Care Professionals to Correctly Identify Patient-ventilator Asynchronies - Pmus Study

Patient-ventilator asynchronies can occur as a result of a mismatch between neural (patient) and ventilator inspiratory and expiratory phases. Sensitivity of this visual analysis, even when performed by experts in the field, is low, around 28% in one landmark publication. The impact of the display of Pmus together with the other ventilator waveforms on the ability of health-care professionals to identify asynchronies has not been tested so far. OBJECTIVES: To compare the sensitivity and specificity of the detection of patient-ventilator asynchrony by health professionals through visual inspection of the ventilator waveforms (conventional group) with the sensitivity and specificity of health professionals who have available, in addition to these ventilator waveforms, also the estimated inspiratory muscle pressure curve (Pmus group). METHODS: Participants will analyze 49 consecutive different scenarios of mechanical ventilation generated in a simulator. Intensive care unit physicians and respiratory therapist will be invited to participate and after the inclusion will be randomized to one of two groups: 1) the control group will inspect pressure and flow curves and 2) the Pmus group will inspect pressure, flow, and Pmus curves. Before the start of the study, all participants will have a 30-min training session to homogenize their concepts on the definitions of the different types of asynchrony. Subsequently, the participants will be randomized to the conventional group or Pmus group. Participants will be designated to watch different sessions, in groups of at most 20 individuals, according to their randomization. In these sessions, recorded ventilator waveforms will be projected to a large screen for 30 seconds. A still image containing a few ventilatory cycles will remain visible for another 30 seconds when participants will have to choose which asynchrony (if any) the participants can see on the screen. Sessions of the Pmus group will display, in addition to pressure and flow, the estimated muscle pressure curves. The main outcome is the asynchrony detection rate (sensitivity). It will be also compared specificity, positive and negative predictive values for asynchrony detection. Statistical significance will be set at an alpha level of 0.05. The sample size was estimated in 98 participants based on the expectation of a 10 percentage points difference in the sensitivity between groups.