Efficacy of High-Flow Nasal Cannula Oxygen Therapy in Patients with Mild Hypercapnia

Objectives To investigate the indications of high-flow nasal cannula (HFNC) oxygen therapy among patients with mild hypercapnia and to explore the predictors of intubation when HFNC fails. Methods This retrospective study was conducted based on the Medical Information Mart for Intensive Care IV (MIMIC-IV) database. Adult patients with mild hypercapnia (45 < PaCO2 <= 60 mmHg) received either HFNC or non-invasive ventilation (NIV) oxygen therapy. Propensity score matching (PSM) was implemented to increase between-group comparability. The Kaplan-Meier method was used to estimate overall survival and cumulative intubation rates, while 28-day mortality and 48-h and 28-day intubation rates were compared using the Chi-squared test. The predictive performances of HR/SpO(2) and the ROX index (the ratio of SpO(2)/FiO(2) to respiratory rate) at 4 h were assessed regarding HFNC failure, which was determined if intubation was given within 48 h after the initiation of oxygen therapy. The area under the receiver operating characteristic curve (AUC) for HR/SpO(2) and the ROX index were calculated and compared. Results A total of 524,520 inpatient hospitalization records were screened, 106 patients in HFNC group and 106 patients in NIV group were successfully matched. No significant difference in 48-h intubation rate between the HFNC group (the treatment group) and the NIV group (the control group) (14.2% vs. 8.5%, p = 0.278); patients receiving HFNC had higher 28-day intubation rate (26.4% vs. 14.2%, p = 0.029), higher 28-day mortality (17.9% vs. 8.5%, p = 0.043), and longer ICU length of stay (4.4 vs. 3.3 days, p = 0.019), compared to those of NIV group. The AUC of HR/SpO(2) at 4 h after the initiation of HFNC yielded around 0.660 for predicting 48-h intubation, greater than that of the ROX index with an AUC of 0.589 (p < 0.01). Conclusion Patients with impending respiratory failure had lower intubation rate, shorter ICU length of stay, and lower mortality when treated mild hypercapnia with NIV over HFNC. As opposed to the ROX index, a modest, yet improved predictive performance is demonstrated using HR/SpO(2) in predicting the failure of HFNC among these patients.