Summary Pre‐oxygenation is an essential part of rapid sequence induction of general anaesthesia for emergency surgery, in order to increase the oxygen reservoir in the lungs. We performed a randomised controlled trial of transnasal humidified rapid insufflation ventilatory exchange ( THRIVE ) pre‐oxygenation or facemask pre‐oxygenation in patients undergoing emergency surgery. Twenty patients were allocated to each group. No patient developed arterial oxygen saturation < 90% during attempted tracheal intubation. Arterial blood gases were sampled from an arterial catheter immediately after intubation. The mean ( SD ) PaO 2 was 43.7 (15.2) kPa in the THRIVE group vs. 41.9 (16.2) kPa in the facemask group (p = 0.722); PaCO 2 was 5.8 (1.1) kPa in the THRIVE group vs. 5.6 (1.0) kPa in the facemask group (p = 0.631); arterial pH was 7.36 (0.05) in the THRIVE group vs. 7.34 (0.06) in the facemask group (p = 0.447). No airway rescue manoeuvres were needed, and there were no differences in the number of laryngoscopy attempts between the groups. In spite of this, patients in the THRIVE group had a significantly longer apnoea time of 248 (71) s compared with 123 (55) s in the facemask group (p < 0.001). Transnasal humidified rapid insufflation ventilatory exchange is a practicable method for pre‐oxygenating patients during rapid sequence induction of general anaesthesia for emergency surgery; we found that it maintained an equivalent blood gas profile to facemask pre‐oxygenation, in spite of a significantly longer apnoea time.