Validity of ROX index to predict High flow nasal cannula outcomes in COVID19 critically ill patients

<bold>Introduction:</bold> During the outbreak of COVID-19 disease, acute hypoxemic respiratory failure (AHRF) patients required high flow nasal cannula (HFNC). ROX index is a simple tool that can help to predict HFNC outcomes. There is a limited data for its validity in COVID-19 patients. <bold>Aim:</bold> to evaluate the validity of ROX index to predict HFNC outcomes in critically ill patients admitted for AHRF due to COVID-19. <bold>Methods:</bold> A retrospective study carried out in a medical ICU from September 19st, 2020 to December 31st, 2021 including all COVID-19 adult patients who are initially treated with HFNC. Data were collected by reviewing the medical patients’ charts. The receiver operating characteristic (ROC) curve was used to evaluate discrimination of ROX index for HFNC success. <bold>Results:</bold> In this study 134 (77.9%) received HFNC, the mean age was 50.97 ±13.86 years; 57 (42.5%) had at least one comorbidity. Median ROX index at 1H was 4.3[3.39-5.85]. HFNC failure was noted in 47 (35.1%) patients. Univariate analysis identified the following factors for HFNC success and failure: ROX index at 1H (5[4-6.31] vs 3.8[3-4.28], p=0.000); ROX index at 6H (6.1[5-8] vs 4.3[3.5-6], p=0.000); ROX index at 24H (7.7[6.9-9.56] vs 4[3.27-5], p=0.000) and ROX index at 48H (8[5.8-10] vs 4[3.2-4.8], p=0.000). The ROC curve analysis for ROX index at H 24 and H48 showed a significant area under the curve. A cut-off point >= 5.7 for H 24 and H48 (Se 0.8, Sp 0.83 and Se 0.85, Sp 0.89 respectively) was associated with HFNC success (Figure1). <bold>Conclusion:</bold> ROX index can be a useful tool to predict HFNC success in COVID-19 critically ill patient admitted for AHRF.