Rationale: Critically ill patients are predisposed to oxyhaemoglobin desaturation during intubation. Objectives: To find out whether non invasive ventilation (NIV), as a preoxygenation method, is more effective at reducing arterial oxyhaemoglobin desaturation than usual preoxygenation during oro-tracheal intubation in hypoxemic critically ill patient. Methods: Prospective randomized study performed in two surgical-medical ICU's. Preoxygenation was performed, prior a rapid sequence intubation, for a 3-minute period using a nonrebreather bag-valve mask (control group) or pressure support ventilation delivered by an ICU ventilator through a face mask (NIV group) according to the randomization. Measurements and Main Results: The Control (n=26) and NIV (n=27) groups were similar in terms of age, disease severity, diagnosis at admission and pulse oxymetry values (SpO₂) prior preoxygenation. At the end of preoxygenation, SpO₂ was higher in NIV group as compared to control group, 98±2 % vs 93±6, p<0.001. During intubation procedure, the lower SpO₂ values were observed in the control group, 81±15% vs 93±8%, p<0.001. Twelve (46%) patients in the control group and 2 (7%) in the NIV group had a SpO₂ below 80%, p<0.01. Five minutes after intubation, SpO₂ values were still better in NIV group as compared to control group, 98±2 vs 94±6 %, p<0.01. Regurgitations, n=3 (6%), and new infiltrates on post procedure X-chest ray, n=4 (8%) were observed with no significant difference between groups. Conclusion: For the intubation of hypoxemic patients, preoxygenation using NIV is more effective at reducing arterial oxyhaemoglobin desaturation than usual method.