The Predictive Index Of Therapeutic Effect In Patients With Hypoxic Respiratory Failure During The Period Of High-flow Nasal Cannula Oxygen Therapy

Objective:Respiratory failure is the most common acute and critical incident for clinicians.In addition to removing the cause of disease,it is necessary to choose appropriate oxygen therapy methods according to the type and the severity of respiratory failure for its treatment.High-flow nasal cannula oxygen therapy（HFNC）,as a new type of non-invasive oxygenation system,can be used for the treatment of acute and critical respiratory failure.The oxygenation index is the main indicator to assess the oxygenation status of patients in the state of oxygen therapy,but it requires repeated arterial blood draws which increases the patients’pain.This article analyzes the clinical data of the application of HFNC in the patients with hypoxic respiratory failure,in order to explore the non-invasive predictive factors of the clinical outcome of HFNC in the patients with hypoxic respiratory failure.The predictive factors can early identify the patients who require tracheal intubation and invasive mechanical ventilation（IMV）during HFNC treatment to avoid delays in treatment and get a better clinical prognosis.Methods:This study is an observational study.Patients who were hospitalized in the Respiratory Medicine department of Chengde Central Hospital,diagnosed with hypoxic respiratory failure and received HFNC treatment were selected as the research objects from October 2019 to October 2020.And a total of 108 patients were eventually included.Except HFNC,all patients were given adequate clinical comprehensive treatment according to underlying diseases.The clinical information were collected,inclouding basic patient information（age,gender,whether underlying chronic respiratory disease,the main cause of respiratory failure）,vital signs,blood routine,blood gas analysis,blood creatinine,ion,blood lactate,blood albumin,GCS score,APACHEⅡscore on admission.The patients’signs during HFNC treatment at 0h,2h,4h,6h,12h,18h and 24h were recorded,including respiratory rate（RR）,heart rate（HR）,and percutaneous oxygen saturation（SpO₂）.At the meantime,the parameter settings of HFNC at the corresponding time point were recorded,including the fraction of inspiration oxygen（FiO₂）.Then the SF ratio（SpO₂/FiO₂）and the ROX index（SpO₂/（FiO₂×RR））were calculatted and recorded.At last these information were recorded,including the duration of HFNC treatment and the clinical treatment outcome（succeed or failed during HFNC treatment）.SPSS 26.0statistical software was used to process and analyze the data.The Kolmogorov-Smirnov test was used to evaluate the normal distribution of measurement data:the data of the normal distribution was described by the mean±standard deviation（（?）±s）.The difference between groups was compared by independent sample t test or t’test.Non-normal distribution data is described by the median（interquartile range）,M（Q25,Q75）.The differences between groups were tested by Mann-Whitney U test.Count data was described by frequency（percentage）.The difference between groups was compared by chi-square test/Fisher’s exact test.The ROC curve and the area under the ROC curve were analysised.The Kaplan-Meier method were used to evaluate whether the predictive index was related to treatment failure and draw the KM curve.The Log-rank method was used to test the difference between groups.For other covariates that may cause confusion or interaction in the baseline data,the Cox proportional hazards model was used for multivariate analysis.The difference was statistically significant when P<0.05.Results:1.Comparison of baseline data in research objectsThe APACHEⅡscore and Pa CO₂ of the patients who failed in HFNC treatment were higher than the patients who succeeded.The difference was statistically significant（all P<0.05）.But there were no statistically significant difference in the age,gender,whether underlying chronic respiratory disease,the causes of respiratory failure,p H,PaO₂,HCO₃^-,blood lactic acid,hematocrit,blood albumin,and hemoglobin before applying HFNC between the patients who succeeded and the patients who failed in HFNC treatment（all P>0.05）.2.Comparison of RR,heart rate,SF ratio and ROX index between the patients who succeeded and the patients who failed in HFNC treatment at each time pointThe SF ratio and ROX index at each time point of the patients who failed after HFNC treatment was lower than that in patients who succeeded.The difference was statistically significant（all P<0.05）.There was no significant difference in respiratory rate and heart rate between the patients who succeeded and the patients who failed in HFNC treatment（all P>0.05）.3.Predictive indicators for evaluating the outcome of HFNC treatmentThe higher the APACHEⅡscore was,the greater the possibility of the failure and the need of tracheal intubation during HFNC treatment was（P<0.05,AUC=0.783,95%CI:0.662-0.904）.And the cut-off point was 14.50（sensitivity 0.762,specificity 0.747）.The RR and heart rate at each time point after HFNC treatment had no predictive significance for the failure and the need of tracheal intubation during HFNC treatment（all P>0.05）.The lower the SF ratio at each time point after HFNC treatment was,the greater the possibility of the failure and the need of tracheal intubation during HFNC treatment was（all P<0.05,all AUC>0.75）.The cut-off point of SF ratio at 2 hour was 185.00（sensitivity 0.900,specificity 0.597）.The lower the ROX index at 24h after HFNC treatment was,the greater the possibility of the failure and the need of tracheal intubation during HFNC treatment was（P<0.05,AUC=0.915,95%CI:0.819-1.000）.The cut-off point of ROX index at 24 hour was 7.31（sensitivity 1.000,specificity 0.724）.The prediction accuracy of ROX index at the rest of the time point was poor（all P<0.05,0.70<AUC<0.75）.4.Risk analysis between predictive indicators and the outcome of HFNC treatment4.1 SF ratioAfter HFNC treatment,the tracheal intubation rate of patients in the SF ratio at 2h≤185 group was significantly higer than the SF ratio at 2h>185group.The difference between the two groups was statistically significant（X²=15.546,P<0.05）.The tracheal intubation rate of patients in the SF ratio at 2h≤185 group was higher than the SF ratio at 2h>185 group（HR=10.521,95%CI:2.440-45.357）.Even adjusted by potential confounding factors,SF ratio at 2h≤185 after HFNC treatment was still independent predictor about the failure of HFNC treatment（all P<0.05）.4.2 ROX indexAfter HFNC treatment,the tracheal intubation rate of patients in the ROX index at 24h≤7.31 group was significantly higher than the ROX index at 24h>7.31group.The difference between the two groups was statistically significant（X²=10.144,P<0.05）.Conclusion:1.For patients with hypoxic respiratory failure treated with HFNC,APACHEⅡscore should be assessed to assist in evaluating the condition.2.For patients with hypoxic respiratory failure during HFNC treament,The SF ratio can be monitored to evaluate the efficacy of HFNC treament in patients with hypoxic respiratory failure and the risk of tracheal intubation.The SF ratio at 2h cann’t reach 185 after HFNC treatment indicates a higher risk of treatment failure.If necessary,tracheal intubation should be considered as soon as possible.3.The ROX index can be used to evaluate the efficacy of HFNC treatment in patients with hypoxic respiratory failure and the risk of early endotracheal intubation.