含动态血糖分析的2型糖尿病慢性并发症风险预测模型及效能评价  

Risk prediction model and efficacy evaluation of chronic complications of type 2 diabetes mellitus with dynamic blood glucose analysis

作　　者：杨洁[1] 丁文萃[1] 王海英[1] 史亚男 Yang Jie;Ding Wencui;Wang Haiying;Shi Yanan(The First Department of Endocrinology and Diabetes,Cangzhou Central Hospital of Hebei Province,Hebei Province,Cangzhou 061000,China)

机构地区：[1]河北省沧州市中心医院内分泌糖尿病一科,061000

出　　处：《疑难病杂志》2022年第8期822-827,共6页Chinese Journal of Difficult and Complicated Cases

基　　金：沧州市重点研发计划指导项目(204106137)。

摘　　要：目的探讨含动态血糖分析的2型糖尿病(T2DM)慢性并发症风险预测模型及效能评价。方法选取2017年1月-2019年1月河北省沧州市中心医院内分泌糖尿病一科收治T2DM患者265例为研究对象,均口服降糖药治疗,随访观察2年,根据患者是否出现慢性并发症分为并发症组和无并发症组,观察2组一般资料、实验室指标及动态血糖水平。使用多因素Logistic回归模型分析糖尿病慢性并发症危险因素,使用R软件建立列线图预测模型,并行一致性检验。结果随访2年,排除失访4例,最终纳入261例,其中出现慢性并发症125例(47.89%)为并发症组,未出现并发症136例(52.11%)为无并发症组;并发症组动态血糖指标最大血糖波动幅度(LACE)、日内血糖平均波动幅度(MAGE)、血糖标准差(BG-SD)、日间血糖平均绝对差(MODD)、空腹血糖变异系数(FPG-CV)、平均餐后血糖波动幅度(MPPGE)、HbA_(1c)标准差(HbA_(1c)-SD)、HbA_(1c)变异系数(HbA_(1c)-CV)、血糖波动系数(BGFC)、低血糖指数(LBGI)、平均日风险范围(ADRR)均高于无并发症组,差异具有统计学意义(t/P=5.551/<0.001,7.713/<0.001,4.242/<0.001,10.396/<0.001,7.454/<0.001,8.514/<0.001,8.357/<0.001,9.909/<0.001,6.791/<0.001,4.565/<0.001,6.965/<0.001);并发症组低密度脂蛋白胆固醇(LDL-C)、颈动脉内膜中层厚度(IMT)高于无并发症组,差异具有统计学意义(t/P=2.110/0.036,2.179/0.030)。多因素Logistic回归分析显示,MAGE、MODD、FPG-CV、MPPGE、HbA_(1c)-SD、HbA_(1c)-CV、BGFC、LBGI、ADRR、LDL-C、IMT升高为预测T2DM慢性并发症发生的危险因素[OR(95%CI)=13.353(2.064~86.407)、5.160(1.268~20.991)、9.266(1.202~71.398)、12.532(2.446~64.213)、9.832(1.392~69.424)、11.452(1.592~82.374)、5.360(1.359~21.146)、3.098(1.253~7.656)、4.067(1.069~15.472)、28.419(1.414~571.109)、8.001(1.657~38.645)];以多因素Logistic回归分析结果建立列线图模型,其预测T2DM慢性并发症发生因素的一致性指数(C-index)为0.994。结�Objective To explore the risk prediction model and efficacy evaluation of chronic complications of type 2 diabetes mellitus(T2DM)with dynamic blood glucose analysis.Method A total of 265 patients with T2DM admitted to the Department of Endocrinology and Diabetes,Cangzhou Central Hospital,Hebei Province from January 2017 to January 2019 were selected as the research subjects,all of whom were treated with oral hypoglycemic drugs,and were followed up for 2 years.In the complication group and the non-complication group,the general data,laboratory indexes and dynamic blood glucose levels of the two groups were observed.The multivariate Logistic regression model was used to analyze the risk factors of chronic complications of diabetes,and the R software was used to establish a nomogram prediction model,and the parallel consistency test was performed.Results After 2 years of follow-up,4 cases were excluded and 261 cases were finally included.Among them,125 cases(47.89%)with chronic complications were in the complication group,and 136 cases(52.11%)without complications were in the non-complication group;the complications group Ambulatory blood glucose indicators:maximum blood glucose fluctuation range(LACE),intraday average blood glucose fluctuation range(MAGE),blood glucose standard deviation(BG-SD),inter-day blood glucose mean absolute difference(MODD),fasting blood glucose coefficient of variation(FPG-CV),average Postprandial Glucose Fluctuation Range(MPPGE),HbA_(1c)Standard Deviation(HbA_(1c)-SD),HbA_(1c)Coefficient of Variation(HbA_(1c)-CV),Glycemic Fluctuation Coefficient(BGFC),Low Glycemic Index(LBGI),Average Daily Risk Range(ADRR)were higher than the uncomplicated group,and the difference was statistically significant(t/P=5.551/<0.001,7.713/<0.001,4.242/<0.001,10.396/<0.001,7.454/<0.001,8.514/<0.001,8.357/<0.001,9.909/<0.001,6.791/<0.001,4.565/<0.001,6.965/<0.001);low density lipoprotein cholesterol(LDL-C)and carotid intima-media thickness(IMT)in the complication group were higher than those in the uncomplicated

关 键 词：糖尿病 2型 糖尿病慢性并发症 动态血糖 糖化血红蛋白变异性 风险预测 

分 类 号：R587.1[医药卫生—内分泌]