Non-invasive respiratory support in pediatric patients with acute hypoxemic respiratory failure: a systematic review and network meta-analysis of randomized controlled trials
Bibliographic record
Abstract
Background: Acute hypoxemic respiratory failure (AHRF) is globally one of the most common causes of admission to pediatric intensive care units (PICUs). There is no clear evidence on which non-invasive respiratory support (NRS) modalities are most effective. We aimed to summarize the existing evidence and compare different NRS modalities in children with AHRF. Methods: We searched Embase, PubMed, CENTRAL, CINAHL Complete, Web of Science, WHO ICTRP, ISRCTN and ClinicalTrials.gov up to July 26, 2025. Randomized controlled trials (RCTs) that evaluated the efficacy and safety of different NRS approaches compared with another NRS or standard oxygen therapy (SOT) for children with AHRF were included. Two investigators independently screened studies, extracted data, and assessed the risk of bias. We performed network meta-analysis (NMA) with frequentist random effects model and chose relative effects (risk ratios, RRs) and absolute effects (risk differences, RDs) or mean differences (MDs) with 95% confidence intervals (CIs) as measures of effect. In-hospital mortality, need of invasive mechanical ventilation (IMV) and treatment failure were identified as primary outcomes. Secondary outcomes were PICU and hospital length of stay, PICU admission, total duration of NRS and oxygen inhalation, ventilation intolerance rate, sedation use, serious adverse events (SAEs), pressure injuries, and abdominal distension. The protocol was registered in PROSPERO, CRD42024529804. Findings: 30 trials with 8163 participants were included. Continuous positive airway pressure (CPAP) reduced IMV risk compared with SOT (RR, 0.61; 95% CI, 0.38-0.97; high certainty evidence). Compared with SOT, both CPAP (RR, 0.52; 95% CI, 0.28-0.99; moderate certainty evidence) and high-flow nasal cannula (HFNC; RR, 0.52; 95% CI, 0.33-0.80; moderate certainty evidence) reduced treatment failure, but increased intolerance (CPAP: RR, 30.57; 95% CI, 4.71-198.63, HFNC: RR, 10.12; 95% CI, 1.79-57.31; moderate-certainty evidence). HFNC (RR, 1.29; 95% CI, 1.03-1.61) increased the risk of PICU admission and CPAP (MD, 6.4 h more; 95% CI, 1.7-11.1 more) prolonged NRS duration (both high certainty evidence). Compared with HFNC, CPAP (RR, 2.41; 95% CI, 1.30-4.47) increased the risk of pressure injury (moderate certainty evidence). Compared with SOT, HFNC (MD, 6.3 h fewer; 95% CI, 11.1-1.5 fewer) may reduce the duration of oxygen (low certainty evidence). Non-invasive ventilation (NIV) may reduce IMV relative to SOT (RR, 0.47; 95% CI, 0.23-0.96) but the certainty of evidence was very low. Subgroup analyses showed CPAP (versus SOT) reduced mortality/treatment failure risk in PICUs (ICEMAN: low credibility), whereas HFNC (versus SOT) lowered IMV risk in PICUs and low- and middle-income countries (ICEMAN: moderate credibility). No significant differences were found in in-hospital mortality, SAEs, hospital and PICU LOS, sedation use, and abdominal distension. Interpretation: In pediatric patients with AHRF, CPAP is likely more favorable NRS mode than HFNC. NRS appears beneficial particularly in resource-limited settings. Additionally, our findings highlight the need for more RCTs to confirm the present conclusions, especially for NIV. Funding: This study was funded by the Foundation of the First Hospital of Lanzhou University (ldyyyn2023-16), the Gansu Provincial Health Industry Scientific Research Project (GSWSKY2024-18), the Natural Science Foundation of Gansu Province (25JRRA776), the Lanzhou University Student Innovation and Entrepreneurship Action Plan Project (20230060047) and the Lanzhou University Student Innovation and Entrepreneurship Action Plan Project (20220060168).
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.056 | 0.048 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.154 | 0.017 |
| Bibliometrics | 0.003 | 0.005 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; both teacher heads agree on what is shown here.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".