Objectives: To assess the diagnostic performance of HACOR scoring system using bedside variables and to predict failure of HFNO in patients with acute hypoxemic respiratory failure (AHRF).
Material and methods: 150 patients with AHRF who were receiving HFNO were enrolled in this study; to predict HFNO treatment failure. A scoring scale (HACOR score) consisted of Heart rate (beats/minute), acidosis (assessed by pH), consciousness (assessed by Glasgow coma score), oxygenation, and respiratory rate. Failure was defined as the need for intubation or death.
Results: Patients were analyzed according to the success or failure of HFNO. Total 150 patients, of which 100 (66.7%) had a successful treatment while 50 (33.3%) failed with such intervention. There was an improvement in HR and RR, and PaO₂/FiO₂ within the first hour (T1) in the success group and these parameters continued to improve even after 24 hours (T2) of HFNO treatment. Patients with HFNO failure had a higher HACOR score at initiation and after 1, 12, 24 and 48 hours. Before intubation, the highest value of the HACOR score was reached in the failure group. At 1h of HFNO assessment, the area under the receiver operating characteristic curve was 0.86, showing good predictive power for failure. We found that HACOR score at a cutoff point > 6 had 81.2% sensitivity and 91% specificity, 92.5% positive predictive value, and 71.4% negative predictive value with a diagnostic accuracy was 85%. Furthermore, the overall diagnostic accuracy exceeded 87% when the HACOR score was assessed at 1, 12, 24 or 48 h of HFNO.
Conclusions: The HACOR scale is a clinically useful bedside tool for the prediction of HFNO failure in hypoxemic patients. A HACOR score < 6 after 1 hour of HFNO highlights patients with < 85% risk of failure.
Keywords: critical care; hypoxemic respiratory failure.