ROX index to predict failure with low flow devices in COVID-19

Abstract

The ROX index (iROX) obtained from pulse oximetry saturation/inspired fraction of oxygen and respiratory rate, predicts success with high-flow nasal cannula (HFNC), however its performance for low-flow oxygenation devices (DOBF) is unknown. To determine suitability of iROX at 12 hours as a predictor of mechanical ventilation (VMI) in hospitalized patients with severe COVID-19 and DOBF. An historical cohort was performed. Adults with COVID-19, hospitalized, with supplemental oxygen supply are included, excluding patients with pathologies in chronic stages that could alter the results, calculating the iROX at 12 hours, obtaining the cut-off point using a ROC curve and Youden index, the risk of VMI is prolonged using relative risk (RR), with 95% confidence intervals (95%CI). Confounding variables were evaluated to determine the performance of the iROX. The dependent variable mechanical ventilation recorded as reported in the file and the independent iROX obtained in the same way as the previous one. 63 patients with a median age of 62 years were included. The best iROX cut-off point at 12 hours was 5.35. With this cut-off point, VMI was associated with a RR of 8.75 (95% CI 2.36-32.35). In the multivariate model with an OR of 9.26; (95% CI, 2.39 - 35.78), after initiation of DOBF was consistently associated with an increased risk of intubation. In hospitalized patients with severe COVID-19, an iROX < 5.35 at 12 hours appears to be a predictor for the onset of VMI.