Comparison of the clinimetric properties of the two versions of the HACOR scale for predicting noninvasive ventilation failure in Brazilian patients.

BACKGROUND

Scales that detect noninvasive ventilation (NIV) failure need to have adequate clinimetric properties to be reliable. This study aimed to compare the clinimetric properties of the Heart rate, Acidosis, Consciousness, Oxygenation, Respiratory rate (HACOR) and updated HACOR scales when applied to hypoxemic adult patients undergoing NIV.

METHODS

This prospective study applied the HACOR and updated HACOR scales to hypoxemic patients after one hour of NIV in an emergency department setting. A second application of the scales was performed after ten minutes to assess reliability (intraclass correlation coefficient), measurement error (standard error of measurement and minimum detectable difference), ceiling and floor effects, convergent validity by correlation (Pearson's r) with peripheral oximetry saturation (SpO2), and predictive validity (area under the receiver operating characteristic [ROC] curve) for the outcome of needing invasive mechanical ventilation.

RESULTS

Sixty patients were included in this study (59.45±17.48 years; Simplified Acute Physiology Score III, 56.1±13.95; 30% with respiratory disease and 25% with cardiovascular disease). After 1 hour of NIV, patients had a HACOR score of 3 (interquartile range [IQR], 1.0-5.0) and an updated HACOR score of 5 (IQR, 3.0-8.87). Clinimetric properties were adequate for both versions of the HACOR scale but were superior for the updated version, including predictive validity (ROC [95% CI], 0.78 [0.64-0.91] vs. 0.73 [0.57-0.89]) and the absence of the ceiling effect.

CONCLUSIONS

Both versions of the HACOR scale demonstrated adequate clinimetric properties for predicting NIV failure, with the updated HACOR version showing superior predictive validity and no ceiling effect compared with the original version.