The Oxygenation Benefits Provided by High‐Flow Nasal Oxygen in Bronchoscopy May Be Underestimated

A recent article published in Respirology by Burton et al. [1] entitled “High Flow Nasal Oxygen and Low Flow Oxygen Are Equally Effective in Providing Oxygenation During Bronchoscopy Under Conscious Sedation: A Randomised Controlled Trial” presents design, data and results that warrant further clarification to ensure the accuracy of the study's conclusions. The design of this trial may not have fully considered the impact of specific breathing patterns, characterised by open-mouth position and intense cough during bronchoscopy, on the actual fraction of inspired oxygen (FiO2) in the HFNO group, which may lead to a significant deviation between the actual FiO2 and the initial set-FiO2, and resulting in biased conclusions. As we know, unlike low flow oxygen (LFO), HFNC delivers an airflow close to or higher than the patient's mean inspiratory flow, which ensures a higher and more reliable content of FiO2 with less dilution by the surrounding air. However, the actual FiO2 can be altered by the administered oxygen flow rate, the characteristics of the delivery device and the patient's respiratory pattern [2]. During bronchoscopy, whether via the nasal route or the oral route, the patient's mouth is usually opened rather than closed, especially when an oral route is adopted, a bite block is needed to protect the bronchoscope and keep the mouth fully open at all times. Previous studies have shown that only in mouth-closed position did the FiO2 set by the HFNO device coincide the actual FiO2. whereas in mouth-opened position, a significant deviation would occur [3]. Moreover, intense cough, which is quite common during bronchoscopy under conscious sedation, can further affect the actual FiO2 values. Notably, this trial did not provide detailed sedation methods, including the use of topical anaesthesia. In the discussion section it was only mentioned that patients had minimal cough and high comfort under light conscious sedation. The result appears inconsistent with our previous findings, where a conscious sedation regimen similar to this trial was adopted as the control group (0.4 mg/kg midazolam plus 1ug/kg fentanyl) [4]. Although we also superimposed topical anaesthesia, the cough response during bronchoscopy was still quite intense, and only 64% of patients expressed their willingness to return for a second bronchoscopy examination under the same sedation. Theoretically, the presence of an open-mouth position accompanied by intense cough can significantly affect the actual FiO2 value, though there is currently a lack of direct evidence or literature reports. However, indirect evidence indicates that the actual FiO2 during hyperpnea with an open mouth is much less than the initial set-FiO2. As demonstrated by Yaroshetskiy et al. [3] that in the case of hyperpnea with an open mouth, when HFNO was set to deliver 100% FiO2 at a flow rate of 30 L/min, the FiO2 measured in the hypopharynx was only 59.2% (32.8%–67.2%). Even if the flow rate was increased to 60 L/min or 80 L/min, the actual FiO2 could only reach 66.6% (34.1%–85.9%) or 64.3% (36.3%–82.3%), respectively. It is therefore reasonable to speculate that a similar deviation may occur in the breathing pattern characterised by intense cough with an open mouth during bronchoscopy under conscious sedation. In conclusion, the changes in breathing patterns and the deviation of actual FiO2 values during bronchoscopy will significantly affect the reliability and accuracy of the conclusions of this trial. Therefore, we hope the authors can provide more detailed data or explanations to help readers better understand or cite their results accurately. Xuefei Zhou: writing – original draft. Yunfei Cao: conceptualization, writing – review and editing. This work was supported by the Zhejiang Provincial Medical and Health Research Project (2024XY036). The authors declare no conflicts of interest.