减重代谢手术病人麻醉苏醒期低氧血症预测模型的构建与验证  

Construction and validation of a predictive model for hypoxemia during anesthesia recovery in patients undergoing bariatric metabolic surgery

作　　者：邹旗 许孟楠 梁辉[1] 刘双源[1] 刘存明[1] 王玉婷 冯建萍[1,2] ZOU Qi;XU Mengnan;LIANG Hui;LIU Shuangyuan;LIU Cunming;WANG Yuting;FENG Jianping(The First Affiliated Hospital of Nanjing Medical University,Jiangsu 210029 China;不详)

机构地区：[1]南京医科大学第一附属医院,江苏210029 [2]南京医科大学护理学院

出　　处：《护理研究》2023年第21期3813-3819,共7页Chinese Nursing Research

基　　金：江苏省人民医院“临床能力提升工程”护理项目,编号:JSPH-NC-2021-23;南京医科大学研究生教育研究专项项目,编号:2021YJS-ZC003。

摘　　要：目的:探讨减重代谢手术病人麻醉苏醒期发生低氧血症的危险因素,构建低氧血症预测模型并验证,为临床预测减重代谢手术病人麻醉苏醒期发生低氧血症的风险提供参考。方法:选择江苏省某三级甲等医院2021年12月—2022年3月接受减重代谢手术的病人,应用Logistic回归筛选麻醉苏醒期低氧血症的独立危险因素,完成模型的初步建立,并以2022年4月—5月接受减重代谢手术的病人为验证组进行模型验证。结果:最终纳入建模组300例,验证组130例。Logistic回归显示,美国麻醉医师协会(ASA)分级、术前水肿、术中气道压、术后疼痛评分、睡眠呼吸暂停初筛量表(STOP-Bang)评分、麻醉苏醒期瑞芬太尼用量是减重代谢手术病人麻醉苏醒期发生低氧血症的独立危险因素(P<0.05)。根据上述危险因素建立减重代谢手术病人麻醉苏醒期低氧血症预测模型,模型通过Hosmer-Lemeshow检验拟合效果良好(P>0.05),建模组受试者工作特征(ROC)曲线下面积(AUC)为0.882,灵敏度为84.96%,特异度为78.07%,最佳临界值为0.630;验证组AUC为0.915,灵敏度为85.96%,特异度为90.41%,最佳临界值为0.764,校正曲线趋近于理想曲线,列线图预测模型预测低氧血症发生情况结果与实际情况相符(建模组C-index为0.882,验证组C-index为0.915)。结论:构建的减重代谢手术病人麻醉苏醒期低氧血症预测模型有较好的预测效能,对临床围术期麻醉护理有指导作用。Objective:To investigate the risk factors of hypoxemia in patients undergoing bariatric and metabolic surgery during anesthesia recovery,and to construct and verify a predictive model of hypoxemia,so as to provide a reference for clinical prediction of the risk of hypoxemia in patients undergoing bariatric and metabolic surgery during anesthesia recovery.Methods:A cross-sectional survey was conducted.Patients who underwent bariatric metabolic surgery in a tertiary hospital in Jiangsu Province from December 2021 to March 2022 were selected.Binary Logistic regression was used to screen the independent risk factors of hypoxemia during anesthesia recovery period,and the preliminary establishment of the model was completed.Patients who underwent bariatric metabolic surgery from April to May 2022 were used as validation group to verify the model.Results:There were 300 cases in the modeling group and 130 cases in the validation group.Binary Logistic regression showed that ASA grade,preoperative edema,intraoperative airway pressure,postoperative pain score,Stop-bang score and remifentanil dosage during anesthesia recovery were independent risk factors for hypoxemia during anesthesia recovery in patients undergoing Bariatric metabolic surgery(P<0.05).According to the above risk factors,a predictive model of hypoxemia in patients undergoing bariatric and metabolic surgery during anesthesia recovery was established.The model was tested by Hosmer-Lemeshow test(P>0.05).The area under ROC curve(AUC)of the modeling group was 0.882,the sensitivity was 84.96%,the specificity was 78.07%,and the optimal critical value was 0.630.The AUC of the validation group was 0.915.The sensitivity was 85.96%,the specificity was 90.41%,and the optimal critical value was 0.764.respectively,and the corrected curve was close to the ideal curve.The results of Nomogram prediction model in predicting the occurrence of hypoxemia were consistent with the actual situation(C-index of modeling group was 0.882,C-index of validation group was 0.915).Conclusi

关 键 词：减重代谢手术 麻醉苏醒期 低氧血症 列线图 麻醉护理 预测模型 

分 类 号：R473.6[医药卫生—护理学]