Utility of ROX Index in Predicting Success of Weaning from HFNC: Prospective Observational Analytical Study

INTRODUCTION: High Flow Nasal Oxygen therapy (HFNOT) is widely used for managing acute hypoxemic respiratory failure (AHRF)(1), but optimal strategies for weaning patients from HFNC is unknown. Moreover, while ROX index can predict success of HFNOT in AHRF,(2–5) whether it can predict success of weaning from HFNOT is still unexplored. OBJECTIVES: The objective of the present study was to investigate the predictive role of ROX index for successful weaning from HFNOT to conventional oxygen therapy (COT) and to distinguish outcome between HFNOT success and failure. METHODS: This prospective observational analytical study was conducted over a duration of 18 months. Criteria for weaning from HFNOT were a) Age > 18 years; b) No signs of respiratory distress (increased work of breathing, use of accessory muscles of respiration leading to diaphoresis, anxiety or agitation); c) Arterial pH ≥ 7.35, SpO2 > 90% on FiO2 <= 0.6; d) Respiratory rate ≤ 25/min, Heart rate < 120/min and Systolic Blood Pressure ≥ 90mmHg. ROX indices were recorded at initiation of HFNC weaning (ROX_0), at subsequent intervals (6, 12, 24, 48, 72, and 96 hours), and at the time of HFNC withdrawal. During the study duration, 65 patients (who received only HFNOT for AHRF) were categorized into two groups based on clinical outcomes: A) HFNC success (weaning to COT without escalation to IMV or re-initiation of HFNC) and HFNC failure (escalation to IMV or re-initiation of HFNC within 48 hours of weaning of HFNC). Statistical analyses, including Mann Whitney U test and median tests, were used to compare ROX scores between groups and determine predictive thresholds. RESULTS: ROX index showed a clear distinction between success and failure groups across multiple time points. At baseline (ROX_0), the median score was 7.6 in the success group and 6.3 in the failure group. By 12 hours (ROX_12), the success group had a median ROX score of 9.6, compared to 8 in the failure group. At 24 hours (ROX_24), the median scores were 10.8 and 10.1 respectively in both groups; however, the trajectory of scores remained higher in the success group at later intervals, with marked differences at 48 hours (13.25 versus 11.10), although it was not statistically significant (p = 0.666). At the time of HFNC withdrawal, the ROX index demonstrated a marked difference between the two groups. Patients in the success group had a median ROX score of 14.45 (interquartile range 13.19–14.55), whereas those in failure group had a median score of 11.35 (interquartile range 9.60–12.80) and the difference was statistically significant (p < 0.05). CONCLUSION: The ROX index can be used to predict HFNC weaning success. The consistently rising and higher ROX scores in the success group, particularly at withdrawal, highlight its utility in identifying patients ready for sustained transition to COT. However, larger multi-centre studies are needed to validate the above findings and create ROX index scoring system which can be utilized while weaning patients of AHRF from HFNOT.