2022 年第 2 期 第 17 卷

体外膜氧合患者血红蛋白水平的监测与护理

Monitoring hemoglobin level and nursing in patients undergoing extracorporeal membrane oxygenation

作者：刘林田杜桂芳贾明侯晓彤王红

英文作者：Liu Lintian Du Guifang Jia Ming Hou Xiaotong Wang Hong

单位：首都医科大学附属北京安贞医院心脏外科危重症中心，北京100029

英文单位：Cardiac Surgery Critical Care Center Beijing Anzhen Hospital Capital Medical University Beijing 100029 China

关键词：体外膜氧合；血红蛋白；院内死亡；护理

英文关键词：Extracorporealmembraneoxygenation;Hemoglobin;Hospitalmortality;Nursing

• 摘要：

• 目的  分析静脉-动脉体外膜氧合（VA-ECMO）患者输血的血红蛋白阈值，为VA-ECMO患者血红蛋白水平监测和护理提供指导。方法  回顾性分析2018年1—12月首都医科大学附属北京安贞医院收治的64例心脏手术后因心源性休克应用VA-ECMO患者的临床资料。所有患者均行VA-ECMO辅助，根据患者出院存活状态，分为存活组（34例）和死亡组（30例）。比较2组术前、VA-ECMO第1天血红蛋白水平及血红蛋白下降率。记录2组临床结局（VA-ECMO成功撤机、连续性肾脏替代治疗（CRRT）、严重出血二次开胸探查、下肢严重缺血、脑卒中、缺血缺氧性脑病）。采用受试者工作特征（ROC）曲线评估VA-ECMO第1天血红蛋白水平及血红蛋白下降率对患者院内死亡的预测能力。采用Logistic回归方法分析VA-ECMO第1天血红蛋白水平及血红蛋白下降率与患者预后的关系。结果  死亡组男性比例、ECMO第1天血红蛋白水平均低于存活组，年龄、血红蛋白下降率均高于存活组，住院时间短于存活组，差异均有统计学意义（均P＜0.05）。死亡组VA-ECMO成功撤机比例低于存活组，接受CRRT比例高于存活组，差异均有统计学意义（均P＜0.05）。2组严重出血进行二次开胸探查、下肢严重缺血、脑卒中及缺血缺氧性脑病比例差异均无统计学意义（均P＞0.05）。ROC曲线分析结果显示，VA-ECMO第1天血红蛋白水平预测患者院内死亡的曲线下面积为0.87（95%置信区间：0.79~0.96）、最佳截断值为85 g/L，血红蛋白下降率的曲线下面积为0.85（95%置信区间：0.76~0.95）、最佳截断值为38.0%。Logistic回归分析结果显示，VA-ECMO第1天血红蛋白水平≤85 g/L（比值比=29.3，95%置信区间：5.9~145.6，P<0.001）、血红蛋白下降率≥40.0%（比值比=15.3，95%置信区间：4.5~52.0，P<0.001）是VA-ECMO患者院内死亡的独立危险因素。结论  VA-ECMO第1天血红蛋白≤85 g/L、血红蛋白下降率≥40.0%是VA-ECMO患者院内死亡的独立危险因素，ECMO护理中应密切监测患者血红蛋白水平及变化。

• Objective To analyze the transfusion threshold of hemoglobin in patients undergoing venoarterial extracorporeal membrane oxygenation (VA-ECMO), so as to provide guidance for monitoring hemoglobin level and nursing in the patients. Methods From January to December 2018, clinical data of 64 patients undergoing VA-ECMO due to cardiogenic shock after cardiac surgery in Beijing Anzhen Hospital, Capital Medical University were retrospectively analyzed. All patients underwent VA-ECMO adjuvant. According to survival state of patients at discharge, they were divided into survival group (34 cases) and death group (30 cases). The level of hemoglobin before surgery and on the 1st day of VA-ECMO, and decline rate of hemoglobin were compared between the two groups. The clinical outcomes of the two groups were recorded ［VA-ECMO successfully withdrawn, continuous renal replacement therapy (CRRT), secondary thoracotomy for severe bleeding, severe ischemia of lower limbs, stroke, and hypoxic encephalopathy］. Receiver operating characteristic (ROC) curve was used to evaluated predictive efficiency of hemoglobin level on the 1st day of VA-ECMO and decline rate of hemoglobin on hospital mortality. Logistic regression analysis was performed to assess the relationship between hemoglobin level on the 1st day of VA-ECMO and decline rate of hemoglobin with patients′ prognosis. Results The male rate and hemoglobin level on the 1st day of VA-ECMO in death group were lower than those in survival group, the age and decline rate of hemoglobin were higher than those in survival group, and the length of stay was shorten than that in survival group (all P<0.05). The rate of VA-ECMO successfully withdrawn in death group was lower than that in survival group, and rate of CRRT was higher than that in survival group (both P<0.05). There were no significant differences in rates of secondary thoracotomy for severe bleeding, severe ischemia of lower limbs, stroke, and hypoxic encephalopathy (all P>0.05). ROC curve analysis showed that the area under the curve (AUC) of hemoglobin level on the 1st day of VA-ECMO in predicting hospital mortality was 0.87 (95% confidence interval: 0.79-0.96) and optimal cut-off value was 85 g/L; the AUC of decline rate of hemoglobin was 0.85 (95 confidence interval: 0.76-0.95) and optimal cut-off value was 38.0%. Logistic regression analysis showed that hemoglobin level on the 1st day of VA-ECMO ≤85 g/L (odds ratio=29.3, 95% confidence interval: 5.9-145.6, P<0.001)and decline rate of hemoglobin ≥40.0% (odds ratio=15.3, 95% confidence interval: 4.5-52.0, P<0.001) were independent risk factors for hospital mortality in patients undergoing VA-ECMO. Conclusions The hemoglobin level on the 1st day of VA-ECMO ≤85 g/L and decline rate of hemoglobin ≥40.0% are independent risk factors for hospital mortality in patients undergoing VA-ECMO. The level and change of hemoglobin should be closely monitored in nursing care.