Extubation failure and subsequent reintubation in the intensive care unit (ICU) are associated with worse outcomes. Optimal postextubation respiratory support for patients at high risk of extubation failure remains uncertain. The HIGH-FLOW OXY systematic review and meta-analysis aimed to compare high-flow nasal cannula (HFNC) with noninvasive ventilation (NIV) for preventing short-term reintubation in critically ill adults identified as high risk for extubation failure.
The review systematically searched PubMed, Cochrane Central Register of Controlled Trials, Embase, and ClinicalTrials.gov from inception through the latest available date. Eligible studies were randomized controlled trials (RCTs) that compared HFNC with NIV in ICU patients classified as high risk for extubation failure. Fifteen RCTs comprising 2,073 participants were included in the pooled analyses.
Investigators extracted data using predefined criteria. Risk of bias (RoB) for individual trials was assessed using the Cochrane RoB tool. Certainty of evidence across outcomes was evaluated with the Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) approach. Random-effects meta-analytic models were used to pool effect estimates. Prespecified subgroup analyses targeted obese patients, very high-risk patients, elderly patients, and those with acute exacerbations of chronic obstructive pulmonary disease (COPD).
The primary outcome was short-term reintubation within 3 days after extubation. Across the 15 RCTs included, pooled results showed no statistically significant difference between HFNC and NIV for reintubation (odds ratio [OR], 1.19; 95% confidence interval [CI], 0.88–1.59). Thus, in the overall high-risk ICU population evaluated, HFNC did not demonstrate superiority over NIV for preventing early reintubation.
Secondary endpoints evaluated in the pooled analyses included ICU mortality, ICU length of stay, sepsis, and nosocomial pneumonia. Pooled estimates indicated no significant differences between HFNC and NIV for these outcomes: ICU mortality (OR, 0.74; 95% CI, 0.40–1.38), sepsis (OR, 1.29; 95% CI, 0.71–2.35), nosocomial pneumonia (OR, 1.01; 95% CI, 0.67–1.52), and ICU length of stay (mean difference [MD], −0.37 days; 95% CI, −1.42 to 0.68). These results suggest comparable performance of HFNC and NIV for the secondary outcomes assessed in the included trials.
Prespecified subgroup analyses identified potential heterogeneity in effect by patient risk strata and by obesity status. Specifically, subgroup results suggested a higher risk of reintubation with HFNC among very high-risk patients (OR, 1.63; 95% CI, 1.05–2.53) and among nonoperative obese patients (OR, 2.52; 95% CI, 1.45–4.38). The report also prespecified analyses for elderly patients and those with COPD exacerbations; the abstract does not report significant subgroup effects for these groups. These subgroup findings indicate that effectiveness may vary according to baseline risk and patient phenotype.
A sensitivity analysis that excluded trials adjudicated as having high risk of bias yielded an increased odds of reintubation with HFNC (OR, 1.32; 95% CI, 1.02–1.71). This finding suggests that trial quality influenced pooled estimates and that excluding lower-quality studies may reveal an association between HFNC and higher reintubation risk in this population.
Authors rated the overall certainty of evidence as low across outcomes. The abstract reports use of Cochrane RoB assessment and GRADE, but does not provide a full GRADE table or granular RoB breakdown within the abstract. Limitations implied by the low certainty rating include study heterogeneity, potential bias within some trials, and the need for larger, adequately powered, risk-stratified randomized trials to clarify the comparative effectiveness of HFNC and NIV in defined high-risk subgroups.
In this systematic review and meta-analysis of ICU patients at high risk of extubation failure, HFNC did not reduce short-term reintubation compared with NIV. Subgroup and sensitivity analyses signaled possible increased reintubation risk with HFNC among very high-risk patients and nonoperative obese patients, and when trials at high risk of bias were excluded. Overall certainty of evidence was low. The authors recommend further adequately powered, risk-stratified RCTs to establish optimal postextubation respiratory support strategies for high-risk critically ill patients.
Conflict of interest: the authors reported no potential conflicts of interest in the publication.