肝细胞癌伴低血糖症风险预测模型的构建与验证  

Construction and verification of risk prediction model for hypoglycemia with hepatocellular carcinoma

作　　者：郑小静 严红虹 李慧景 陈思涓 陈秀梅 ZHENG Xiao-jing;YAN Hong-hong;LI Hui-jing;CHEN Si-juan;CHEN Xiu-mei(Guangdong Provincial People’s Hospital(Guangdong Academy of Medical Sciences),Southern Medical University,Guangzhou 510080,China)

机构地区：[1]南方医科大学附属广东省人民医院(广东省医学科学院),广东广州510080

出　　处：《护理学报》2025年第4期53-58,共6页Journal of Nursing(China)

基　　金：广东省医学科研基金(B2021269)。

摘　　要：目的构建和验证原发性肝细胞癌患者低血糖风险预测模型。方法收集我科2020年4月—2021年6月原发性肝细胞癌患者共232例,将其是否发生低血糖分为低血糖组(n=67)和非低血糖组(n=165),回顾性分析危险因素并建立风险预测模型,采用ROC曲线验证预测模型的效能。结果经分析发现,Child-Pugh分级C级(OR=9.050,95%CI:2.911~28.130,P<0.001)是肝细胞癌患者发生低血糖的危险因素;糖化血红蛋白(OR=0.504,95%CI:0.325~0.780,P=0.002)是保护因素。回归模型为:Y=2.317+0.056×Child-Pugh分期(B)+2.203×Child-Pugh分期(C)-0.686×糖化血红蛋白;ROC曲线面积为0.737(95%CI:0.660~0.814),灵敏度为53.7%,特异度为87.3%,模型验证结果:ROC曲线面积为0.827(95%CI:0.727~0.927),灵敏度为55%,特异度为81.8%,提示本预测模型预测效果良好。结论肝细胞癌伴低血糖症风险预测模型有助于识别低血糖高危人群,为医护人员采取前瞻性护理干预提供科学依据。Objective To construct and validate a hypoglycemia risk prediction model in patients with primary hepatocellular carcinoma(HCC).Methods A total of 232 patients with primary HCC from April 2020 to June 2021 were involved and divided into hypoglycemia group(n=67)and non-hypoglycemia group(n=165)according to the occurrence of hypoglycemia.The risk factors were retrospectively analyzed and risk prediction model was established.ROC curve was used to verify the efficacy of the prediction model.Results The analysis revealed that Child-Pugh grade C(OR=9.050,95%CI 2.911-28.130,P<0.001)was a risk factor for hypoglycemia in patients with HCC;but glycated hemoglobin(OR=0.504,95%CI 0.325-0.780,P=0.002)was a protective factor.The regression model was:Y=2.317+0.056 Child-Pugh stage(B)+2.203 Child-Pugh stage(C)-0.686 HbA1c;The area of the ROC curve was 0.737(95%CI 0.660-0.814),with a sensitivity of 53.7%and specificity of 87.3%.Model validation showed that the area of the ROC curve was 0.827(95%CI 0.727-0.927),with a sensitivity of 55%and specificity of 81.8%,which suggests good prediction efficacy of the model.Conclusion The hypoglycemia risk prediction model in patients with HCC can help identify population with high risk of hypoglycemia and provide scientific basis for prospective nursing intervention for medical staff.

关 键 词：肝细胞癌 低血糖 风险预测模型 

分 类 号：R473.73[医药卫生—护理学]