High-flow nasal oxygenation during gastrointestinal endoscopy. Systematic review and meta-analysis

Abstract

Background: The use of high-flow nasal oxygen (HFNO) has the potential to improve patient safety by limiting hypoxaemia during gastrointestinal endoscopy. The degree of benefit is not adequately established.

Methods: English language literature searches of PubMed, Scopus, Web of Science, and Cochrane Library electronic databases were performed to identify randomised controlled trials comparing HFNO and conventional oxygen therapy (COT) for patients undergoing gastrointestinal endoscopy under deep sedation. The primary endpoint was the incidence of hypoxic events observed during endoscopic procedures. The secondary endpoints were the incidence of recourse to rescue manoeuvres, procedure interruption, and adverse events. A meta-analysis and a post hoc trial sequence analysis were performed.

Results: A total of 2867 patients from six randomised controlled trials were considered. Desaturation was observed in 5.2% and 27.2% of patients receiving HFNO and COT, respectively. Desaturation <90% was observed in 1.8% and 12.6% of the patients receiving HFNO and COT, respectively. In the subgroup analysis, desaturation occurrence was lower during HFNO than during COT in non-obese patients (2.2% vs 25.2%) and obese patients (22.9% vs 43.3%). Desaturation occurrence was lower during maximum (3.6% vs 26.9%) and minimum (15.9% vs 29.8%) HFNO therapy than during COT. HFNO showed a lower recurrence to rescue manoeuvres rate (4.7% vs 34.3%), a lower procedure interruption rate (0.4% vs 6.7%), and a lower adverse events rate (18.7% vs 21%) than COT. A high level of heterogeneity between the studies precluded confidence in drawing inference from the meta-analysis.

Conclusions: The evidence reviewed suggests that compared with COT, HFNO has fewer hypoxaemic events during gastrointestinal endoscopy, but this may not apply to all patients and clinical scenarios.

Keywords: complications; digestive system endoscopy; gastrointestinal endoscopy; hypoxia; oxygen inhalation therapy.

Fig 1

The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram of the study selection process. HFNO, high-flow nasal oxygen; RCT, randomized controlled trials.

Fig 2

Risk of Bias (RoB) 2 assessment (traffic light [a] and summary [b]) of included RCTs. The graphs of the Risk of Bias (RoB) 2 assessment show that the included studies were subject to an overall unclear risk of bias. Details are available in the Supplementary material (SMc2).