Performance of the Rox index to predict High Flow Nasal Cannula failure in COVID-19 patients.

Abstract Background While High Flow Nasal Cannula (HFNC) use increased in COVID-19 hypoxic patients, the optimal timing of mechanical ventilation in COVID-19 is uncertain. We sought to evaluate the ROX index ([SpO2/FIO2]/breathing frequency) capacity to predict weaning from high-flow nasal cannula (HFNC) in COVID-19. Methods We performed a single-center, retrospective, observational study of subjects with respiratory failure from COVID-19 managed with HFNC. The ROX index was calculated at H 12 from HNFC and applied to predict HFNC failure. Subjects managed with HFNC were divided into two groups. And HFNC failure was defined as endotracheal intubation and invasive ventilation necessity in the first five days from admission. Standard statistical comparisons and regression analyses were used to evaluate the ROX index performance in prediction of HFNC failure. Results 121 subjects with COVID-19 were managed with HFNC. 66 (55.5%) were successfully weaned from HFNC and 55 (45.5%) required intubation and mechanical ventilation. High mortality rates have been reported in the last group. Rox index < 2.8, a severe ARDS at admission and the occurrence of an acute cor pulmonale were independent risk factors of HFNC failure. The performance of a ROX index at H12 lower than 2,8 was significant (AUC = 0,621; 95% CI: 0,520-0,722) to predict a HFNC failure.: Conclusion In this retrospective review, HFNC success was associated to better clinical outcome. The ROX index was sensitive to predict HFNC failure and may allow better patient’s selection and earlier invasive mechanical ventilation initiation to optimize patient’s outcome.