Validation of the ROX index to predict high flow nasal cannula therapy treatment failure in infants with bronchiolitis

Abstract

Abstract Purpose High-flow nasal cannula (HFNC) therapy is commonly used to provide an intermediate level of respiratory support for infants with bronchiolitis. The ratio of SpO2/FiO2 to respiratory rate (ROX index) has previously been shown to aid in prediction of HFNC treatment failure in adults and children with diverse respiratory diseases. We aimed to evaluate the utility of the ROX index in predicting HFNC treatment failure in infants with bronchiolitis. Methods Retrospective analysis of previously well infants (< 1 year) hospitalized for bronchiolitis and initiated on HFNC as their primary modality of respiratory support. Results Of 64 infants (median age 70 days), 5 (7.8%) required intubation within 6 hours of HFNC initiation (median time to intubation 225 minutes; interquartile range 125–290 minutes). No between-group differences were observed with respect to sex, age, weight, respiratory syncytial virus infection status, presumed bacterial pneumonia, hospital unit of HFNC initiation, or respiratory parameters at initiation and 1 hour following. Compared to infants who were successfully treated with HFNC, infants who required intubation were initiated earlier in the course of their illness (3 days vs 4 days; p = 0.02). The ROX index did not demonstrate discriminatory ability at time of HFNC initiation (AUROC 0.6; p = 0.5) or 1 hour after initiation (AUROC 0.6; p = 0.6). Conclusions The ROX index at HFNC initiation and 1 hour did not predict early treatment failure in infants with bronchiolitis. Examination of a larger cohort of infants and greater number of treatment failures is required.