非心脏术后低氧性呼衰患者经鼻高流量氧疗失败危险因素及早期预警模型

Analysis of risk factors associated with failure of nasal high-flow oxygen therapy in hypoxic respiratory failure patients after non-cardiac surgery

目的分析非心脏术后低氧性呼衰患者接受经鼻高流量氧疗(HFNC)治疗失败的早期危险因素。方法回顾分析北京大学人民医院2016年1月至2019年12月接受HFNC治疗并符合标准的患者资料,探索HFNC失败的心脏及非心脏危险因素并建立早期预警模型。结果共纳入患者133例,根据HFNC失败标准分为HFNC失败组(n=40)和成功组(n=93)。多因素分析显示,治疗初期高急性生理与慢性健康状况评分Ⅱ(APACHⅡ)、发热、高心肌肌钙蛋白I(TNI)及治疗1 h后ROX(SpO_2/FiO_2/RR)<8是HFNC失败的独立危险因素;建立回归模型Y=-2.158+1.882×发热+0.283×APACHII-0.326×ROX+2.222×TNI,模型ROC曲线下面积(AUC)为0.915。结论非心脏术后低氧性呼衰患者应用HFNC,基础高APACHⅡ评分及TNI、发热,提示治疗失败风险较高,应谨慎选择,治疗1 h后ROX仍较低,应及时调整治疗策略。

Objective To investigate the risk factors associated with failure of nasal high-flow nasal oxygen therapy(HFNC)in hypoxic respiratory failure patients after non-cardiac surgery.Methods This retrospective study was conducted at Peking University People′s Hospital.The subjects included all patients who suffering hypoxic respiratory failure after non-cardiac surgery from January 2016 to December 2019.Clinical and laboratory parameters use statistical methods to explore the risk factors involved in HFNC failure and establish an early warning model.Results A total of 133 patients were included.According to HFNC failure criteria,patients were divided into HFNC failure group(n=40)and success group(n=93).Multivariate analysis showed that patient with temperature(T)>37.3℃,high acute physiology and chronic health status scoreⅡ(APACHⅡ)score,cardiac troponinⅠ(TNI)level before using HFNC,and low ROX(SpO2/FiO2/RR)<8 at 1 hour after treatment were independent risk factors for HFNC failure;a regression model was established:Y=-2.158+1.882×fever+0.283×APACHⅡ-0.326×ROX+2.222×TNI,the area under the ROC curve(AUC)of the model is 0.915(P<0.01).Conclusions Hypoxic respiratory failure patients after non-cardiac surgery fever,High APACHII score and TNI level at baseline are indicate a higher risk of HFNC failure.ROX is still lower after 1 h treatment an important indicator for further mechanical ventilation,the treatment strategy should be adjusted in time.