In the past, patients suffering from acute respiratory failure had to be treated with Oxygen and specific medication, or - if this had not been sufficient - had to be anesthesized and had to undergo endotracheal intubation for invasive ventilation. Nowadays, since emergency ventilators are available that can be used for non-invasive ventilation, the purpose of this study is to investigate if non-invasive ventilation is safe, feasible and effective for the treatment of acute respiratory failure and if less patients will need to be intubated out-of-hospital as well as in the further course.

Single center, prospective, randomized, intention-to-treat clinical trial. The study has been performed within a two tier EMS system of a major city. Prehospital emergency physicians are on scene in every patient enrolled to the study.

Patients with acute respiratory failure (SaO2 < 90%, respiratory rate > 20/min) were enrolled to the study using envelopes for randomisation for either "conventional" (O2 + medication) or "NIV." Vital signs have been documented as well as venous blood gases have been taken every 10 minutes after treatment has been started.

Primary endpoints are 28-day mortality, hospital-, ICU- and respirator-days, SAPS, SOFA II and complications. Secondary endpoints are out-of-hospital incidence of invasive ventilation, effect on SpO2, CO2, respiratory rate, heart rate and blood pressure.

Inclusion Criteria:

• acute respiratory failure
• at least 18 years old
• cooperative

Exclusion Criteria:

• cardiac arrest
• respiratory arrest
• unconscious
• uncontrollable agitation
• less than 18 years old