基于血清Aβ1-42、BMI及Metrnl的列线图模型对2型糖尿病患者认知功能障碍的预测价值的研究  

The predictive value of serum Aβ1-42,BMI and Metrnl in cognitive impairment in patients with type 2 diabetes

作　　者：张静[1] 冀素晓 潘建秀 牛金龙 张丽莎 王海芳[1] ZHANG Jing;JI Suxiao;PAN Jianxiu;NIU Jinlong;ZHANG Lisha;WANG Haifang(Department of Endocrinology,Handan First Hospital,Handan 056000,China)

机构地区：[1]河北省邯郸市第一医院内分泌一科,056000 [2]河北省邯郸市第一医院骨科

出　　处：《中国煤炭工业医学杂志》2024年第6期604-609,共6页Chinese Journal of Coal Industry Medicine

基　　金：邯郸市科学技术研究与发展计划项目(编号:23422083249)

摘　　要：目的 构建血清β淀粉样蛋白1-42(Aβ1-42)、体质量指数(BMI)及镍纹蛋白样蛋白(Metrnl)的列线图模型,分析其对2型糖尿病患者认知功能障碍的预测价值。方法 随机选取2023年1月—2024年5月在邯郸市第一医院接受治疗的2型糖尿病患者76例作为研究对象,依据是否发生认知功能障碍分为未发生组(42例)和发生组(34例),收集两组患者临床资料,进行单因素分析,同时比较两组患者基线资料及血清Metrnl、BMI、Aβ1-42水平,运用多因素logistic回归分析对2型糖尿病患者认知功能障碍造成影响的相关因素进行分析,并构建列线图模型;运用Pearson相关性分析Metrnl、Aβ1-42、BMI与认知功能障碍发生的相关性,绘制标准曲线,分析三者列线图模型对认知功能障碍发生的预测价值。结果 与未发生组比较,发生组年龄、病程、BMI、家庭月收入≤5 000占比、糖化血红蛋白、低密度脂蛋白(LDL)、24h尿蛋白、Aβ1-42、同型半胱氨酸水平高,受教育年限及Metrnl水平低,差异有统计学意义(P<0.05)。BMI、Aβ1-42与认知功能障碍发生呈正相关(r=0.563、0.542,P<0.05),Metrnl与认知功能发生之间呈负相关(r=-0.492,P<0.05);多因素logistic分析结果显示,对认知功能障碍造成影响的因素包括糖化血红蛋白、BMI、年龄、24h尿蛋白、Metml、家庭月收入、糖尿病病程、同型半胱氨酸、LDL、Aβ1-42;ROC曲线结果显示,BMI的AUC值0.665,敏感度为61.76%,特异度为73.81%,准确度为68.72%,95%CI:0.542~0.788;Metrnl的AUC值0.646,敏感度为73.53%,特异度为66.67%,准确度为69.74%,95%CI:0.522~0.770;Aβ1-42的AUC值0.698,敏感度为67.65%,特异度为71.43%,准确度为69.74%,95%CI:0.578~0.818;联合检测的AUC值0.792,敏感度为91.18%,特异度为64.29%,准确度为76.32%,95%CI:0.688~0.896;联合检测敏感度及准确度明显高于单项检测,具有较高预测价值。结论 Aβ1-42、BMI及Metrnl是2型糖尿病患者发生认知功能障碍的标志物,基Objective To construct A nomogram model of serum amyloid beta 1-42(Aβ1-42),body mass index(BMI)and nickel striatin-like protein(Metrnl).To analyze its predictive value for cognitive dysfunction in patients with type 2 diabetes.Methods Seventy-Six patients with type 2 diabetes who received treatment in Handan First Hospital from January 2023 to May 2024 were selected as research objects.According to whether cognitive dysfunction occurred,42 cases were divided into non-occurrence group.Occurrence group 34 cases,Clinical data of the two groups were collected for univariate analysis.At the same time,serum Metrnl,BMI,Aβ1-42 levels were compared between the two groups.Multi-factor logistic was applied to analyze the related factors affecting cognitive dysfunction in type 2 diabetes patients.And construct the nomogram model.By using Pearson correlation analysis,Metrnl,Aβ1-42 and BMI were associated with cognitive dysfunction.Draw a standard curve,the predictive value of the three nomogram models to the occurrence of cognitive dysfunction was analyzed.Results Compared with the non-occurrence group,age,disease course,BMI,monthly household income≤5000,HBA1C,Lowdensity lipoprotein(LDL),24h urinary protein,Aβ1-42 and homocysteine levels were high in the occurrence group.The years of education and Metrnl level are low,and the difference was statistically significant(P<0.05).BMI and Aβ1-42 were positively correlated with cognitive dysfunction(r=0.563,0.542,P<0.05).Metrnl was negatively correlated with cognitive function(r=-0.492,P<0.05).The results of multi-factor logistic analysis show that factors affecting cognitive dysfunction include glycated hemoglobin,BMI,age,24h urinary protein,Metml,monthly household income,diabetes duration,homocysteine,LDL,Aβ1-42.ROC curve showed the AUC of BMI=0.665,sensitivity was 61.76%,specificity was 73.81%,accuracy was 68.72%,95%CI:0.542-0.788.The AUC of Metrnl=0.646,sensitivity was 73.53%,specificity was 66.67%,accuracy was 69.74%,95%CI:0.522-0.770.The AUC of Aβ1-42=0.698,sensitivity

关 键 词：血清β淀粉样蛋白1-42 体质量指数 镍纹样蛋白 列线图模型 2型糖尿病 认知功能障碍 

分 类 号：R741[医药卫生—神经病学与精神病学]