Over the last two decades, non-invasive ventilation (NIV) has been widely reported as an effective method to avoid the need of endotracheal intubation (ETI) and improve survival in the acute care setting. Given the risks associated with either premature NIV discontinuation or delays in NIV interruption, evaluating readiness to weaning from NIV is a critical challenge in patients with Acute Respiratory Failure (ARF).

Up to date, bedside measurements used to predict NIV outcomes are extremely limited. NIV weaning as well as decision of ETI are mainly supported by clinical and physiologic parameters. More sophisticated techniques used to predict weaning outcome during spontaneous breathing trials have never achieved a bedside broad-spectrum use due to their invasiveness, the inconsistent results in demonstrating reproducible outcomes, the requirements of additional trainee personnel and complicated equipment, and the difficult application in awake and non-intubated patients.

Recently, ultrasound has been used for the rapid assessment of diaphragm function in acutely ill patients. The advantages of the ultrasound in detecting diaphragm dysfunction as compared with other techniques are the less invasiveness, the avoidance of radiation hazards and the bedside feasibility. Direct imaging of changes in diaphragm thickening (DT) during spontaneous breathing may provide the assessment of both the muscle and the respiratory pump functioning. Indeed, DT has been correlated with the diaphragm strength and the muscle shortening. The volume of diaphragm muscle mass remains constant while it contracts. Consequently, as the muscle shortens it thickens itself and measurements of changes in such a thickening (DT) are inversely related to changes in diaphragm length. Studies in patients with diaphragm paralysis have confirmed the absence of DT. Moreover, since the diaphragm is the major muscle of inspiration, the presence of diaphragm shortening and contraction may predict successful extubation in patients who are invasively ventilated.

The aim of the present study is to assess whether DT as measured by ultrasound may predict NIV outcome in patients with de-novo ARF admitted to the Emergency Department (ED).

All consecutive patients with de-novo ARF requiring NIV treatment while they will be admitted to the Emergency Department at Gemelli's Hospital, Rome-Italy will be included into the present study.

The present protocol will be approved by the local Ethics Committee and informed consent will be obtained by the all study participants or their next of kin.

All the patients will be ventilated in the ED over a maximum of 48 hours, by using 2 types of different ventilators (Evita XL, Drager or Infinity C500, Drager).

In all patients NIV will be delivered through either a facial mask with an inflatable soft cushion seal (Gibeck, Upplands, Sweden; Vitalsigns, Towota, NJ, USA) or to a clear, latex-free helmet (CaStar, Starmed, Mirandola, Italy), according to the clinical decision.

Pressure support ventilation with positive end-expiratory pressure (PEEP) will be used with all interfaces. Pressure support ventilation will be started at 10 cmH2O and increased with progressive stepwise increments of 2-3 cmH2O, to obtain an exhaled tidal volume of 6 mL/kg, a respiratory rate (RR) < 25 breaths/min, patient comfort and disappearance of both accessory muscle activity and/or paradoxical abdominal motion. PEEP will be increased with stepwise increments of 2-3 cmH2O up to 12 cmH2O to ensure peripheral oxygen saturation (SpO2) of ≥90% with the lowest possible FiO2. When the helmet will be used, part of the volume delivered to the system will be used to distend the helmet without reaching the patient. PEEP levels will be increased with stepwise increments of 2-3 cmH2O up to 15-18 cmH2O to ensure peripheral oxygen saturation (SpO2) of ≥90% with the lowest possible FiO2 during NIV delivered through the helmet. Ventilator settings will be then adjusted according to SpO2 and measurements of arterial blood gases. The flow trigger will be set at 3 L/s, checking out the absence of auto-triggering phenomena.

All study group will be managed according to the standard of care with respect of medical management of ARF (i.e, antibiotic, antiviral, or antifungal agents; bronchodilators; diuretics; frequent respiratory treatments and chest physiotherapy), timing of medical interventions (i.e placement of central and arterial catheters and frequency of blood withdrawals, and cultures) and other aspects of emergency support (fluid administration, correction of electrolytes abnormalities, nutrition).

All the patients will be kept with the head of the bed at 30-45 degrees.

Criteria for intubation and NIV discontinuation Pre-determined criteria for immediate ETI will include the inability to maintain a PaO2 /FIO2 > 140 during NIV, the onset of seizures or coma (Glasgow coma score < 8), hemodynamic instability (systolic blood pressure < 80 mmHg despite adequate fluid resuscitation and/or electrocardiographic signs of ischemia or arrhythmias), intolerance of the interface, inability to manage copious secretions, inability to alleviate dyspnea, or the need for an emergency surgical procedure. After intubation, all patients will be ventilated with the same ventilation protocol, according to the low-tidal-volume protective ventilatory strategy.

Criteria for NIV weaning NIV will be maintained over the next 48 hours until oxygenation and clinical status will improve. NIV will be progressively reduced in accordance with the degree of clinical improvement and discontinued if the patient will be able to maintain a respiratory rate < than 30 breaths/min and a PaO2 > 75 mmHg with a FiO2 of 0.5 without ventilatory support.

Definitions and Measurements The Simplified Acute Physiologic Score (SAPS II) will be calculated on admission to the study.

For all study group, the duration of mechanical ventilation and the ED stay, as well as the hospital outcome will be registered.

During the first 12 hours from the enrollment into the study, the arterial blood gas levels will be determined at baseline, at 1 hour, at 4 and 12 hours. Following this period, the parameters will be measured at 12 hour intervals until the 96th hour from the patient admission and/or his discharge.

Improvement in gas exchange will be defined as the ability to increase PaO2/FiO2 above 200 or an increase in this ratio of more than 100 from baseline.

Early NIV success will be defined as the improvement in clinical status and gas exchange within 1 hour of treatment.

NIV success will be defined as the improvement in clinical status and gas exchange within the first 48 hours of treatment.

Sustained NIV success will be defined as the ability to maintain the defined improvement in clinical status and gas exchange over the next 48 hours after the end of the time needed to define NIV success.

NIV failure will be defined as the need of ETI at any point of the study period and/or failure to reach an improvement in clinical status and gas exchange within 48 hours.

Patients will be monitored for the development of infections or other complications. Sepsis, severe sepsis, and septic shock will be defined according to recent consensus guidelines. Adult respiratory distress syndrome (ARDS) will be defined according to the Berlin definition.

Diaphragm thickness (DT) will be measured using a 7-10 MHz linear ultrasound probe set to B mode (Vividier, General Electrics). Both the right and left diaphragm will be imaged at the apposition point of the diaphragm and the rib cage on the midaxillary line, between the 8th and the 10th intercostal spaces. DT will be measured at either the end-expiration or the end-inspiration. The percent change in DT between end-expiration and end-inspiration (ΔDT%) will be calculated as [(DTend-inspiration-DTend-expiration/DTend-expiration)×100]. The ΔDT% for each patient will represent the mean of three to five breaths. During the first 12 hours from the enrollment into the study, the DTF measurements will be determined at baseline, at 1 hour, at 4 and 12 hours. Following this period, the measurements will be performed at 12 hour intervals until the 96th hour from the patient admission and/or his discharge.

Training the ultrasound operator to identify the diaphragm and measure its thickness will take from three to five sessions.

Measurements will be performed by 2 different and appropriately trained operators who are routinely involved in the management of the patients.

Inclusion Criteria:

• Respiratory rate > 35 breaths/min,
• PaO2/FiO2 < 200 while the patient is breathing oxygen through a Venturi mask and active contraction of the accessory muscles of respiration or paradoxical abdominal motion.

Exclusion Criteria:

• Age < 18 years,
• Pregnancy,
• Diaphragm paralysis,
• Neuromuscular disorders,
• Palliative NIV in patients with malignancy,
• Ineffective cough,
• Inability to protect airways.
• BMI > 35

All consecutive patients with de-novo ARF requiring NIV treatment while they will be admitted to the Emergency Department at Gemelli's Hospital, Rome-Italy.