The Use of the Rox Index to Guide High-Flow Nasal Cannula Therapy and the Need for Intubation Is Associated With...

Acute hypoxemic respiratory failure (AHRF) is a common and life-threatening condition frequently managed with high-flow nasal cannula (HFNC) oxygenation. The ROX index, the ratio of oxygen saturation to FIO₂ to respiratory rate, has been validated to predict HFNC oxygenation failure, but its impact on outcomes when used to guide decision-making in the timing of intubation remains unclear.

What is the impact on outcomes when the ROX index is used to guide decisions-making in the timing of intubation in patients receiving high-flow nasal cannula therapy?

A retrospective observational cohort study at an academic medical center emergency department (ED) was undertaken. Two hundred thirty-three adults were initiated on HFNC oxygenation for AHRF between January and December 2022. ROX indices were calculated retrospectively at 4, 6, and 12 hours after HFNC oxygenation initiation. Patients were categorized based on whether their care aligned with ROX predictions as ROX concordant or ROX discordant. The primary outcome was persistent organ dysfunction plus death (POD+D), a composite measure of death and ongoing need for mechanical ventilation, vasopressors, or new dialysis, assessed at day 28.

Eighty-two percent of patients (n = 191) received ROX-concordant care, whereas 18% of patients (n = 42) received ROX-discordant care. Despite similar baseline characteristics, patients receiving ROX-discordant care experienced significantly worse outcomes: higher POD+D (62% vs 16%; P < .001), 28-day mortality (38% vs 13.6%; P < .001), and in-hospital mortality (50% vs 15%; P < .001). Outcomes were driven by patients intubated despite ROX indices predicting HFNC oxygenation success. Multivariable regression showed that ROX-discordant care increased odds of POD+D 9-fold (OR, 9.2; 95% CI, 4.0-21.1; P < .001) and increased the odds of 28-day mortality by 4.76 times compared with receiving ROX-concordant care (OR, 4.76; 95% CI, 1.76-12.88; P < .001).

In patients treated with HFNC oxygenation in the ED, clinical decisions aligned with ROX index predictions were associated with significantly better outcomes. Discordant care, especially premature intubation, correlates with increased morbidity and mortality. These findings support prospective validation of improved outcomes with ROX-concordant actions.