The goal of this prospective interventional crossover randomized physiological study is to investigate the reliability of Pressure Muscle Index (PMI) - as an estimation of inspiratory effort - at different levels of expiratory cycling during pressure support ventilation. PMI will be compared with the esophageal pressure swing that is considered the gold standard technique.

This study aims to answer to the following questions:

• which is the optimal expiratory cycling threshold where PMI better correlates with the esophageal pressure swing?
• what is the optimal correlation between the occlusion pressure (Poc) estimated by an expiratory occlusion manoeuvre and P0.1 with PMI obtained at various degrees of expiratory cycling threshold?
• does airway resistance - evaluated by using esophageal pressure - correlate with the estimation of airway resistance on the pressure-time waveform by a high percentage of expiratory cycling mimicking the interrupter technique?

The concept of Patient-Self-Inflicted Lung Injury (P-SILI) defines the injury that a patient may cause to the lungs while exerting strong inspiratory efforts. During assisted spontaneous breathing, the alveolar distending pressure results from both the positive pressure delivered by the ventilator and the negative pressure generated by the patient. If the patient generates very low negative pleural pressure (Ppl), this may result in an elevated alveolar distending pressure (i.e. transpulmonary pressure). The assessment of esophageal pressure (Pes) - as a surrogate of Ppl - allows to estimate the transpulmonary pressure. However, oesophageal pressure monitoring is not commonly used across different Institutions and it is not often the standard of care to estimate the inspiratory effort and transpulmonary pressure. A manual inspiratory hold is a bedside available tool to potentially estimate a safe threshold of airway pressure, as it can reliably estimate plateau pressure (Pplat) during assisted mechanical ventilation and driving pressure (DP). The difference between peak (Ppeak) and plateau pressure (Pplat) during pressure support ventilation results in the pressure muscle index (PMI) which is considered a reliable measurement of the elastic contribution of patient's inspiratory effort. This index is a simple bedside tool able to uncover the "hidden pressure" that the patient generates during pressure support ventilation and it tightly correlates with the muscular pressure at end inspiration measured by a Pes catheter. Nowadays, in most ventilators, percentage of the peak flow can be adjusted from as low as 1% to as high as 80% resulting in longer or shorter inspiratory times, respectively. Therefore, Pplat can be equal or lower that Ppeak in the absence of inspiratory effort. In the presence of an inspiratory effort, after a manual inspiratory hold, the pressure generated by the patient will be released and the level of Pplat may be visible above Ppeak to an extent that may change based on the set expiratory cycling. We therefore aim to verifying whether pressure muscle index (PMI) - obtained by the pressure time waveform on the ventilator and used as an estimation of the inspiratory effort - is differently correlated with esophageal pressure swing (i.e. gold standard to describe the inspiratory effort) by changing expiratory cycling at different levels of pressure support. Furthermore, inspiratory effort estimated by PMI at different levels of pressure support and expiratory cycling will be compared with another estimator of patient's effort, the Pocc. Pocc is the pressure under assisted ventilation when the airway is briefly occluded during an expiratory manoeuvre. As last, during the steps at an early expiratory cycling (60%), the pressure-time waveform will be evaluated and airway resistance will be estimated by mimicking the interrupter technique. This will be compared to the estimation of the airway resistive component by using the esophageal catheter. Patients will be enrolled at least 6 hours after and within 72 hours since the switch to PSV from controlled mechanical ventilation modalities (CMV).

At this time, the following parameters will be recorded:

• Patient demographics (including age, gender, height, weight, predicted body weight)
• Date/time of informed consent
• Date and time of hospital and ICU admission
• Comorbidities
• Smoking history and status
• Admission diagnoses
• Date/time of onset of mechanical ventilation
• Determinants of Sequential Organ Failure Assessment (SOFA) score at ICU admission
• Ventilation parameters, level of sedation (RASS scale), sedative drugs and arterial blood gas analysis on the day before the switch to PSV and at the study enrolment
• Prior use of adjunctive therapies including neuromuscular blocking agents (NMBA) and prone position
• Components of the SOFA score at the enrolment. Esophageal pressure will be measured by using a specific nasogastic feeding tube with an air-filled balloon. Three levels of pressure support will be explored randomly: a clinical pressure support, and 4 cmH2O of PS above and below the clinical pressure support At each level of pressure support, four sets of data will be recorded by randomly varying the expiratory cycling threshold: 15%, 30%, 45% and 60 % of peak inspiratory flow. After switching from one level of expiratory cycling to the new one, a ten minutes interval will be required before the data collection, in order to allow the patient to adapt to the new ventilation setting.

For each degree of expiratory cycling, the following parameters will be recorded:

• Ventilation parameters: PEEP, PS, Inspiratory and expiratory trigger, Inspiratory rising time
• Ventilatory mechanics measurements (3 times each): RR, TV, Ppeak, Pplat, total PEEP, P0.1, occlusion pressure (Poc), end-inspiratory Pes and end-expiratory Pes after and inspiratory and expiratory occlusion maneuvers of 3 seconds, respectively
• Esophageal pressure swings evaluation - 20 breaths for each level of pressure support ventilation and each level of expiratory cycling at steady state considered at the end of the 10 minute step.

In addition, the following parameters will be reported:

• Maximal Inspiratory Pressure (MIP) and Pes during MIP (one single measurement)
• Level of sedation (RASS scale) and type of sedative drugs
• Arterial blood gas analysis
• Determinants of SOFA Score. Patients will be followed up until ICU and hospital discharge. Data regarding ICU and hospital mortality will be collected, as well as total duration of mechanical ventilation, total duration of CMV and PSV.

Inclusion Criteria:

• Invasive mechanical ventilation in PSV
• Presence of spontaneous breathing activity (ventilator triggering), since 6 hours and no longer than 72 hours after
• Weaning from mechanical ventilation
• Patient for full active management
• Subject ≥ 18 years
• Informed consent

Exclusion Criteria

• Age <18 years old
• Pregnancy
• Active air leaks
• Chronic Obstructive Pulmonary Disease and/or asthma
• Moribund state
• Neurological conditions potentially impairing the ventilatory drive (e.g. meningitis, encephalitis) and neuromuscular diseases impairing neuromuscular conduction (e.g. Guillain-Barre syndrome)
• Extracorporeal membrane oxygenation (ECMO)