Discomfort associated with underhumidified high-flow oxygen therapy in critically ill patients

Objective  To measure (1) the discomfort in non-intubated patients under high-flow oxygen therapy (HFOT) humidified with bubble (BH) or heated humidifiers (HH), and (2) the hygrometric properties of oxygen with a BH and an HH. Design and setting  This was a randomized cross-over study in critically ill patients during a 3-day period. The humidification device used at days 1 and 3 was changed for the other at day 2. (2) It was also an experimental bench study using the psychrometric method with five randomized flows (3, 6, 9, 12 and 15 l/min) and different humidification techniques. Methods  Discomfort, particularly dryness of the mouth and throat, was measured for two humidification conditions (BH and HH) using a 0–10 numerical rating scale (NRS) by patients requiring HFOT with a face mask at a flow ≥5 l/min, in a double-blinded condition. Results  (1) In this clinical study, 30 patients treated by HFOT at a median flow of 7.8 l/min (5.1–10.9) were included. The global incidence of moderate (NRS = 4–6) and severe discomfort (NRS = 7–10) was 25 and 29%, respectively. The median intensities of both mouth and throat dryness were significantly lower with the HH than with the BH [7.8 (5.0–9.4) vs. 5.0 (3.1–7.0), P = 0.001 and 5.8 (2.3–8.5) vs. 4.3 (2.0–5.0), P = 0.005, respectively]. (2) In the bench study, the mean absolute humidity measured at an ambient temperature of 26°C with the HH was two times greater than with the BH (30 ± 1 vs. 16 ± 2 mg/l, P < 0.05) regardless of the flow rate. Conclusions  Compared to bubble humidifiers, the use of a heated-humidifier in patients with high-flow oxygen therapy is associated with a decrease of dryness symptoms mediated by increased humidity delivered to the patient. Electronic supplementary material  The online version of this article (doi:) contains supplementary material, which is available to authorized users. This article is discussed in the editorial available at: doi:. The authors declare that Fisher & Paykel Healthcare France (Courtaboeuf, France) provided technical assistance for the bench test study, including a financial aid of 3,000 €, which was used to acquire technical equipment and to present results at scientific meetings. Respironics France (Carquefou, France) provided technical assistance for the bench test.