ABSTRACT
Objective: This study aims to evaluate the respiratory rate-oxygenation index (ROX) and the ratio of pulse oximetry saturation (SpO2) to the fraction of inspired oxygen (FiO2) (SpO2/FiO2, [SF]) to determine whether these indices are predictive of outcome in dogs receiving high-flow nasal cannula oxygen therapy (HFNOT). Design: This is a prospective observational study. Setting: This study was carried out at two university teaching hospitals. Animals: In total, 88 dogs treated with HFNOT for hypoxemic respiratory failure due to various pulmonary diseases were selected. Measurements and main results: The ROX index was defined as the SF divided by the respiratory rate (RR). ROX and SF were calculated at baseline and for each hour of HFNOT. The overall success rate of HFNOT was 38% (N = 33/88). Variables predicting HFNOT success were determined using logistic regression, and the predictive power of each variable was assessed using the area under the receiver operating curve (AUC). ROX and SF were adequately predictive of HFNOT success when averaged over 0-16 h of treatment, with similar AUCs of 0.72 (95% confidence interval [CI] 0.60-0.83) and 0.77 (95% CI 0.66-0.87), respectively (p < 0.05). SF showed acceptable discriminatory power in predicting HFNOT outcome at 7 h, with an AUC of 0.77 (95% CI 0.61-0.93, p = 0.013), and the optimal cutoff for predicting HFNC failure at 7 h was SF ≤ 191 (sensitivity 83% and specificity 76%). Conclusion: These indices were easily obtained in dogs undergoing HFNOT. The results suggest that ROX and SF may have clinical utility in predicting the outcomes of dogs on HFNOT. Future studies are warranted to confirm these findings in a larger number of dogs in specific disease populations.