Early Clinical and Physiological Predictors of High-Flow Nasal Cannula Failure in Infants With Acute Bronchiolitis: A Systematic Review

Acute bronchiolitis is a common cause of respiratory distress in infants, and high-flow nasal cannula (HFNC) therapy is frequently used to provide noninvasive respiratory support.However, treatment failure requiring escalation of respiratory support remains an important clinical concern.This systematic review aimed to identify clinical and physiological predictors of HFNC failure in infants and young children with acute bronchiolitis.This systematic review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines.PubMed/MEDLINE, Embase, Scopus, Web of Science, the Cochrane Library, and supplementary sources were searched through October 1, 2026.Eligible observational studies and relevant secondary analyses of clinical trials were included.Risk of bias was assessed using the Quality in Prognosis Studies (QUIPS) tool.Findings were synthesized narratively because differences in study characteristics, predictor assessment time points, statistical methods, and definitions of HFNC treatment failure precluded clinically meaningful quantitative pooling.A total of 1,358 records were identified, including 1,286 from electronic databases and 72 from supplementary sources.Thirteen studies met the inclusion criteria.Younger age, lower body weight, increased respiratory rate, tachycardia, hypercapnia, greater clinical severity, and persistent oxygen requirements were associated with HFNC failure, although findings varied across studies.The respiratory rate-oxygenation (ROX) index demonstrated limited discriminatory performance at baseline and one hour after HFNC initiation, with area under the receiver operating characteristic curve (AUC) values of 0.56 at both time points.At 12 hours after HFNC initiation, the ROX index demonstrated moderate discriminatory performance in predicting treatment failure, although differences in outcome definitions and study methodologies limited direct comparability.Younger age, lower body weight, greater clinical severity, hypercapnia, persistent respiratory distress, and increased oxygen requirements may help identify infants at risk of HFNC treatment failure.Serial clinical and physiological assessment may provide more useful prognostic information than isolated baseline measurements.Although the ROX index demonstrated moderate discrimination at 12 hours in two studies, differences in treatment failure definitions and study methodologies limit direct comparability and the establishment of universal prediction thresholds.The ROX index should not be used as the sole criterion for treatment escalation.Further prospective multicenter studies using standardized failure definitions and consistent assessment intervals are needed to improve clinical prediction and decision-making.