Comparison of Actual Performance in the Flow and Fraction of Inspired O₂ among Different High-Flow Nasal Cannula Devices: A Bench Study

Table 5

Characteristics of actual-FiO₂ in five devices.

Devices	≤40%	FiO2 deviation (%) 45%–50%	>50%	Effect of set-flow
AIRVO 2	−0.4 (−0.8, 0.1)	−0.6 (−0.9, −0.1)	−1.4 (−2, −0.8)	Best accuracy at 40 L/min^a
TNI softFlow 50	−1 (−1.2, −0.7)	−1.5 (−1.7, −1.3)	−2.8 (−4, −2.4)	No effect at 26%, 30%, and 50%^b; cannot reach higher FiO2 at high-flow levels^c
OH-70C	0.3 (−0.7, 1.5)	−1.7 (−2, −1.4)	−2.4 (−2.7, −2.2)	Best accuracy at 35%^d; effect increased at 60–70 L/min
Bellavista 1000	0.9 (0, 1.6)	2.5 (1.9, 3)	2.6 (2.1, 3.5)	Minor effect at 26%^e
HUMID-BH	−0.9 (−1.6, −0.5)	−3 (−4.1, −2.5)	−5.9 (−6.8, −5)	Varied effect with no obvious pattern

The data are presented as the median (IQR). The FiO₂ deviation equals actual-FiO₂ minus the corresponding set-FiO₂. ^aWhen the set-flow rate was 40 L/min and set-FiO₂ was 26%, 35%, 40%, 60%, and 70% (, 0.857, 0.322, 0.37, and 0.372, respectively), actual-FiO₂ was not significantly different from the corresponding set-FiO₂. ^bWhen set-FiO₂ was fixed at 26%, 30%, and 50%, actual-FiO₂ under different set-flows had no significant difference in TNI softFlow 50 (). ^cThe control system of TNI softFlow 50 stopped the rising process, triggering the “oxygen flow is too high” alarm. The alarm cannot be eliminated until the oxygen flow is lowered. ^dWhen set-FiO₂ was 35%, most of actual-FiO₂ at varied set-flows were not significantly different compared with set-FiO₂ (, and 0.37 for set-flow = 20, 25, 30, 45, and 60 L/min, respectively). ^eWhen set-FiO₂ was 26%, most of actual-FiO₂ at varied set-flows were not significantly different compared with set-FiO₂ (, and 0.713 for set-flow = 20, 25, 50, 60, and 70 L/min, respectively).