Not approved for ethics

Hypoxemia during endotracheal intubation is one of this procedure's most frequent and severe complications, which can lead to cardiac arrest and other adverse outcomes. Although various studies have been conducted to ensure the safety of endotracheal intubation, most of them have focused on the physiological changes during the procedure in surgical settings, where general anesthesia is administered to healthy subjects. However, the physiological characteristics of critically ill patients who require endotracheal intubation in the intensive care unit (ICU) may differ from those of healthy subjects. When comparing the oxygen saturation-dissociation curve of healthy subjects and critically ill patients, the latter show a steeper decline in oxygen saturation (SpO2) below 90% when hypoxemia occurs. Clinically, it is difficult to maintain or recover oxygen saturation when it drops. Therefore, preoxygenation is also essential for critically ill patients with a relatively small lung reservoir, as it can help maintain higher and longer oxygen saturation during endotracheal intubation. For this reason, apneic oxygenation was proposed in the 1950s as a method of oxygen delivery during general anesthesia for surgery, and clinical randomized controlled trials (RCTs) applying it to critically ill patients in the ICU have been conducted since 20 years ago. In an observational study by Macamn et al., the adjusted odds ratio (OR) of a bag valve mask was 1, while that of a high-flow nasal cannula (HFNC) was 5.75 and that of non-invasive ventilation (NIV) was 0.10 when different preoxygenation methods were used. n FLORARI 2, a large-scale clinical RCT comparing NIV and HFNC with or without apneic oxygenation, NIV was significantly more effective than HFNC in reducing severe hypoxemia. Based on these previous studies, the investigators can infer that supplying oxygen during laryngoscopy is not very effective in preventing hypoxemia in critically ill patients. However, there is no guideline on which device to choose for preoxygenation in the ICU, and NIV or HFNC are selected according to the operator's preference. This is because few studies have examined the physiological effects of these devices on patients. Therefore, in this study, the investigators aim to compare the clinical outcomes and adverse events of NIV and HFNC as preoxygenation methods for endotracheal intubation in critically ill patients with acute respiratory failure and to evaluate their physiological effects using electrical impedance tomography.

Inclusion Criteria:

• Adult patients aged 18 or older
• Acute respiratory failure is defined as follows: Tachypnea over 30 breaths/minute / Symptoms or signs of respiratory failure / PaO2/FiO2 less than 300
• When there is an indication for endotracheal intubation for mechanical ventilation
• When sufficient information and a consent form for the study are obtained from the patient or the patient's guardian

Exclusion Criteria:

• Coma with Glasgow score < 8
• Endotracheal intubation during CPR
• Contraindications for NIV: poor patient cooperation, recurrent vomiting, GI bleeding, severe craniofacial injury, upper airway obstruction, etc.
• Contraindications for RSI: complete upper airway obstruction, severe trauma obscuring cervical structures, etc.
• Implanted electronic medical devices (defibrillator, pacemaker or spinal cord stimulator)
• Skin damage or inadequate electrode contact due to wound dressing at the electrode site
• BMI > 50
• Polydipsia
• Unmeasurable SpO2
• Pregnancy or lactation