Purpose: To evaluate the predictors of high flow nasal cannula (HFNC) failure in intensive care unit (ICU) patients with acute respiratory failure (ARF) Methods: The patients are collected from a medical center in Taiwan between June 13 2017 and December 31 2018 One hundred eighty-nine subjects were divided into a HFNC success group (48%) and a HFNC failure group (52%) The clinical characteristics and physiologic effects before HFNC used were retrospectively reviewed and compared between the groups Results: The study outcome of hospital mortality ICU length and hospital length were significantly higher in the HFNC failure group than in the HFNC success group The results of the multivariate analysis indicated that APACHE II and PaO2/FiO2 were significantly different between groups The support vector machine (SVM) classifier model shows that APACHE II + PaO2/FiO2 gives the highest accuracy of 72 1% to predict HFNC failure Conclusion: The baseline APACHE II and PaO2/FiO2 were independently associated with HFNC failure Key words: High flow nasal cannula failure SVM predictors
High flow nasal cannula to treat the acute respiratory failure
蓓怡, 羅. (Author). 2019
Student thesis: Doctoral Thesis