Early Influencing Factors On Successful Application Of High-flow Nasal Cannula Oxygen Therapy In Critically Ill Patients

Objective To analyze the early influencing factors of the successful application of high flow nasal cannula oxygen therapy（HFNC）in critically ill patients with the expectation of improving clinicians’ability to apply HFNC with accuracy and efficiency.Methods Clinical data of patients who received HFNC treatment from January2017 to December 2020 in the Emergency Intensive Care Unit（EICU）of Lanzhou University Second Hospital and met the inclusion criteria were retrospectively reviewed in this study.Baseline clinical data were collected:gender,age,primary illness,admission vital signs,blood gas markers measured within 24 hours of admission to the EICU,finger pulse oxygen saturation Sp O₂,ROX index（Sp O₂/Fi O₂/respiratory rate）,ICU-related complications,length of ICU stay,length of hospital stay,clinical outcomes,oxygen concentration at admission to the EICU Fi O₂,first laboratory findings after admission to the EICU,and Acute Physiology and Chronic Health Status Evaluation System II（APACHE II）score.The HFNC and mechanical ventilation groups were divided according to the need for higher respiratory support,and the factors influencing the success of HFNC use were analyzed.SPSS25.0 statistical software was applied to statistically analyze the above data.Firstly,the between-group analysis of single-factor data was performed,followed by multi-factor analysis using binary logistic regression models,different Cox proportional risk regression models were established,and the area under the curve（AUC）of subject work characteristics（ROC）was calculated to assess the influence of each independent influence on the successful application of transnasal high-flow oxygen therapy.P<0.05 indicates a statistically significant difference between groups.Results 1.According to the inclusion and exclusion criteria,376 critically ill patients were enrolled in our study,and the patients were divided into two groups according to the final mode of respiratory support,including 209 in the HFNC group and 167 in the mechanical ventilation group.The binary logistic regression results showed higher respiratory rate（RR）,high ROX index,high oxygenation index,and high Sp O₂as independent influences factors on successful application of HFNC.The higher ROX index had a higher specificity for predicting the successful application of HFNC,where the area under the curve（AUC）was 0.806（0.763,0.849）,sensitivity was 0.569,specificity was 0.934,Jorden index was 0.503,and intercept value was9.64.However,Higher RR had a negative effect on the successful prediction of HFNC.The timely application of more advanced respiratory support methods in the patients with higher RR may have better prognosis for patients.2.The patients in the HFNC group were then divided into survival group（n=177）and death group（n=32）according to clinical outcomes.The binary Logistic regression results showed that a high APACHE II score was an independent risk factor for death and the AUC was 0.856（0.798,0.914）,the sensitivity was 0.906,the specificity was 0.712,and the cut off value was 17 points.For the patients with APACHE II<15 points HFNC can be considered,and patients with APACHE II 15-25points should be used with caution.Patients with APACHE II>25 points should choose a higher level of respiratory support.Conclusions The choice of using HFNC for respiratory support in acutely ill patients suggested a higher likelihood of successful HFNC use if they were admitted with a high ROX index,respiratory rate<25 breaths/min,oxygenation index>140.73mm Hg,and Sp O2>88.35%;and APACHE II score<15 suggested a higher success rate when using HFNC.