Predicting Intubation in COVID-19 Patients by the ROX Index Method

Aim: To evaluate the diagnostic accuracy of the ROX index for predicting intubation in Coronavirus disease-2019 (COVID-19) patients receiving high nasal flow cannula (HNFC) and to underline the risk and association of intubation with pre-existing comorbidities in COVID-19 patients.
Materials and Methods: A multicenter prospective cohort study was conducted on 123 patients suffering from severe COVID-19 disease from March 2020 to December 2021. The ROX index was calculated at baseline, 2nd, 6th and 12th hour of HNFC and patients were followed for the primary outcome, i.e., invasive ventilation (IV). The diagnostic performance of ROX indices for primary outcome and risk of comorbidity were estimated.
Results: The primary outcome occurred in 49 (39%) patients, most of whom were elderly and suffering from underlying conditions. The ROX index was significantly associated with poor outcome, whereas the best predictive value (sensitivity; 91.8%, area under the curve: 0.905) for IV was found for 12th hour of HNFC. Furthermore, the risk of IV increases with comorbidity.
Conclusion: According to our findings, we speculate that the ROX index could predict the occurrence of adverse events. Moreover, we further suggest that the data regarding comorbidities are valuable in assessing the risk of IV in COVID-19 patients.