High-flow nasal cannula (HFNC) therapy is widely used for respiratory support in critically ill patients, offering benefits such as improved oxygenation and reduced respiratory rate. However, HFNC failure can lead to adverse outcomes, including increased mortality. This narrative review examines predictive factors and indices for HFNC failure, including respiratory rate, P/F and S/F ratios, the ROX index, HACOR score, and emerging indices, such as VOX and FOX. Among these, the ROX index and HACOR score currently provide the most robust predictive value, whereas newer tools such as VOX and FOX require further validation. The ROX index, combining oxygenation and respiratory rate, has shown significant predictive value, particularly in COVID-19 patients, though its thresholds and timing for assessment remain variable. Modified versions of the ROX index, incorporating heart rate and PaO2, have improved predictive accuracy. The HACOR score, initially developed for non-invasive ventilation, also predicts HFNC failure but may be less discriminative in emergency settings. Emerging indices such as VOX and FOX offer novel approaches but face clinical application challenges due to measurement complexities. Risk stratification models, scoring systems, ultrasound techniques, and machine learning methods show promise but require further validation. This review highlights the importance of integrating multiple predictive tools and tailoring assessments to individual patient conditions. Future strategies must also account for nursing quality variables to enhance prediction accuracy in real-world settings. Comprehensive training for healthcare professionals and future multicenter, large-scale studies is essential to refine these predictive strategies and improve patient care quality.
Muhetaer et al. (2026) studied this question.