Table 2. Methodological quality assessment of the randomized trials.
| Randomized controlled trials evaluated using the revised Cochrane Risk of Bias (RoB 2.0) tool | ||||||
|---|---|---|---|---|---|---|
| Study | Allocation bias | Performance bias | Attrition bias | Measurement bias | Reporting bias | Overall bias |
| Ben-Menachem [13] | Low risk | Low riskd | Low risk | Low risk | Low risk | Low risk |
| Douglas [14] | Some concernsa | Low risk | Low risk | Low risk | Low risk | Some concerns |
| Irfan [15] | Some concernsb | Low risk | Low risk | Low risk | Low riske | Some concerns |
| Longhini [16] | Some concernsc | Low risk | Low risk | Low risk | Low risk | Some concerns |
| Lucangelo [17] | Low risk | Low risk | Low risk | Low risk | Low riske | Low risk |
aThe baseline of comorbidity of pulmonary carcinoma, procedural data, and the SpO2 following pre-oxygenation were imbalanced.
bThe baseline of diagnosis of metastatic cancer was imbalanced.
cThe baseline of age was not available.
dOne patient crossed over from the conventional oxygen group to the HFNC group in Ben-Menachem’s trial, but they performed adequate analysis on an intend-to-treat (ITT) basis. Hence, we did not raise the risk of bias.
eAlthough these two trials did not provide protocol registration, they investigated the same outcomes as other trials registered in ClinicalTrials.gov. Obvious selective reporting was not detected. Hence, we did not raise the risk of bias.