Factors associated with intubation in patients with acute hypoxaemic respiratory failure treated with high-flow nasal cannula oxygen therapy: A prospective, observational study

Abstract

Background

High-flow nasal cannula (HFNC) oxygen is an alternative to conventional oxygen in acute hypoxaemic respiratory failure. Some patients require intubation, with a risk of delay; thus, early predictors may identify those requiring earlier intubation. The "ROX" index (ratio of pulse oximetry/fraction of inspired oxygen to respiratory rate) predicts intubation in patients with pneumonia treated with HFNC therapy, but this index has not been validated in non-pneumonia causes of acute hypoxaemic respiratory failure.

Aim/Objective

The aim of this study was to identify factors associated with intubation in a heterogeneous group of patients with acute hypoxaemic respiratory failure treated with HFNC oxygen.

Methods

This prospective observational study was undertaken in an Australian tertiary intensive care unit and included patients over 18 y of age with acute hypoxaemic respiratory failure who were treated with oxygen via HFNC. Vital signs and arterial blood gases were recorded prospectively at baseline and regular prespecified intervals for 48 h after HFNC initiation. Multivariate logistic regression was used to identify the factors associated with intubation.

Results

Forty-three patients were included (N = 43). The multivariate factors associated with intubation were admission Sequential Organ Failure Assessment score (odds ratio [OR]: 1.94 [95% confidence interval {CI}: 1.06–3.57]; p = 0.032) and Pneumonia Severity Index (OR: 0.95 [95% CI: 0.90–0.99]; p = 0.034). The ROX index was not independently associated with intubation when adjusted for Sequential Organ Failure Assessment score (OR: 0.71 [95% CI: 0.47–1.06]; p = 0.09). There was no difference in mortality between patients intubated early (<24 h) compared to those intubated late.

Conclusions

Intubation was associated with admission Sequential Organ Failure Assessment score and Pneumonia Severity Index. The ROX index was not associated with intubation when adjusted for admission Sequential Organ Failure Assessment score. Outcomes were similar irrespective of whether patients were intubated late rather than early.