The following is the summary of “Prediction of noninvasive ventilation failure using the ROX index in patients with de novo acute respiratory failure” published in the  December 2022 issue of Intensive Care by Duan, et al.


This was a reanalysis of data from a prospective, multicenter observational research to revise risk categories. Patients with acute respiratory failure due to a new cause were included, but those who were hypercapnic were not. SpO2/FiO2 to respiratory rate (ROX) index was determined pre-treatment, post 1-2 h, post 12 h, and post 24 h Noninvasive ventilation (NIV). The predictive ability for NIV failure was compared using the ROX index, PaO2/FiO2, and PaO2/FiO2/respiratory rate.

A total of 1,286 people had recently developed acute respiratory failure. With that number, 568 (or 44%) had NIV failure. The ROX index was lower in patients who did not respond to NIV treatment. Patients with pre-NIV ROX indices of ≤ 2, 2-4, 4-6, 7-8, 10, and more than 10 were found to have 92.3%, 70.5%, 55.3%, 41.1%, 35.1%, and 29.5% rates of NIV failure, respectively. Analyses of the ROX index performed after 1, 2, 12, and 24 hours of NIV yielded comparable findings. When the ROX index was used to foretell NIV failure, the area under the receiver operating characteristics curve was 0.64 (95% CI 0.61-0.67). After 1-2 h NIV, it rose to 0.71 (95% CI 0.68-0.74). After 12 h NIV, it rose to 0.77 (0.74-0.80), and after 24 h NIV, it rose to 0.80. 

The ROX index and the PaO2/FiO2 had equivalent ability to foretell NIV failure. Similar to how there was no correlation between the ROX index and the PaO2 /FiO2 /respiratory rate outside of the 1-2 h NIV time point. In patients with de novo acute respiratory failure, the ROX index provides the moderate prognostic potential for NIV failure.

Source: annalsofintensivecare.springeropen.com/articles/10.1186/s13613-022-01085-7