使用胰岛素治疗的2型糖尿病住院患者代谢相关脂肪性肝病与血糖波动相关性的研究

Correlation analysis of metabolically associated fatty liver disease and blood glucose fluctuation among hospitalized patients with type 2 diabetes mellitus

目的 探讨住院使用Ins治疗的T2DM患者代谢相关脂肪性肝病(MAFLD)与血糖波动的相关性。方法 选取2012年1月至2017年8月于北京大学人民医院内分泌科住院使用Ins治疗的T2DM患者1373例,计算患者每日血糖变异系数(CV)和住院第2~3天、4~5天、6~7天CV均值(CV_(2-3)、CV_(4-5)、CV_(6-7))。根据CV中位数(27.8%、26.0%、24.8%)分为CV_(2-3)<27.8%组(n=686)、CV_(2-3)≥27.8%组(n=687);CV_(4-5)<26.0%组(n=687)、CV_(4-5)组≥26.0%组(n=686);CV_(6-7)<24.8%组(n=686)、CV_(6-7)≥24.8%组(n=687)。根据是否合并MAFLD分为T2DM合并MAFLD组(MAFLD,n=793)和单纯T2DM组(n=580),其中MAFLD组按照中位数(26.2%、24.4%、23.4%)分为CV_(2-3)<26.2%亚组(n=396)、CV_(2-3)≥26.2%亚组(n=397);CV_(4-5)<24.4%亚组(n=397)、CV_(4-5)≥24.4%亚组(n=396);CV_(6-7)<23.4%亚组(n=396)、CV_(6-7)≥23.4%亚组(n=397),再根据MAFLD纤维化评分(NFS)分为NFS<-1.455(无进展性肝纤维化亚组,n=115)、-1.455≤NFS≤0.676(可疑肝纤维化亚组,n=514)、NFS>0.676(进展性肝纤维化亚组,n=164)。Logistic回归分析不同阶段CV的影响因素。结果 与CV_(2-3)<27.8%组比较,CV_(2-3)≥27.8%组BMI、FC-P、MAFLD患病率降低(P<0.01)。与CV_(4-5)<26.0%组比较,CV_(4-5)≥26.0%组BMI、FC-P、MAFLD患病率降低(P<0.01)。与CV_(6-7)<24.8%组比较,CV_(6-7)≥24.8%组BMI、FC-P、MAFLD患病率降低(P<0.01)。Logistic回归分析显示,BMI、FC-P、Hb A1c是CV_(2-3)≥27.8%的影响因素,性别、年龄、BMI、FC-P、MAFLD是CV_(4-5)≥26.0%的影响因素,BMI、FC-P、MAFLD是CV_(6-7)≥24.8%的影响因素。与T2DM组比较,MAFLD组CV_(2-3)、CV_(4-5)、CV_(6-7)降低(P<0.01),FC-P升高(P<0.01)。与CV_(2-3)<26.2%亚组比较,CV_(2-3)≥26.2%亚组BMI、FC-P降低(P<0.01)。与CV_(4-5)<24.4%亚组比较,CV_(4-5)≥24.4%亚组BMI、FC-P降低(P<0.01)。与CV_(6-7)<23.4%亚组比较,CV_(6-7)≥23.4%亚组BMI、FC-P降低(P<0.01)。进展性肝纤维化亚组CV_(2-3)高于可疑肝纤维化亚组(P<0.05)。Logistic回归分析显示,FC-P、Hb A1c是CV_(2-3)≥26.2%的影响因素,FC-P是CV_(4-5)≥24.4%的影响因素,DM病程、FC-P是CV_(6-7)≥23.4%的影响因素。结论 BMI、FC-P是Ins治疗的T2DM患者血糖波动的影响因素,无MAFLD患者血糖波动大,FC-P是T2DM合并MAFLD患者血糖波动的影响因素。

Objective e To investigate the relationship between metabolically associated fatty liver disease(MAFLD)and blood glucose fluctuation in hospitalized type 2 diabetes mellitus(T2DM)patients treated with insulin.Methods A total of 1373 T2DM inpatients treated with insulin were collected.The mean values of the coefficient of variation(CV)of daily blood glucose on days 2~3,4~5,and 6~7 were calculated respectively.According to the median of CV_(2-3),CV_(4-5) and CV_(6-7)groups(27.8%,26.0%,24.8%,respectively),subjects were divided into subgroups:CV_(2-3)<27.8%(n=686),CV_(2-3)≥27.8%(n=687);CV_(4-5)<26.0%(n=687),CV_(4-5)≥26.0%(n=686);CV_(6-7)<24.8%(n=686),CV_(6-7)≥24.8%(n=687).Subjects were divided into T2DM combined with MAFLD group(MAFLD,n=793)and T2DM group(n=580).According to the median CV(26.2%,24.4%,23.4%),MAFLD patients were divided into subgroups:CV_(2-3)<26.2%(n=396),CV_(2-3)≥26.2%(n=397);CV_(4-5)<24.4%(n=397),CV_(4-5)≥24.4%(n=396);CV_(6-7)<23.4%(n=396),CV_(6-7)≥23.4%(n=397).According to the MAFLD fibrosis score(NFS),subjects were divided into NFS<-1.455(non advanced liver fibrosis subgroup,n=115),-1.455<NFS<0.676(suspected liver fibrosis subgroup,n=514),NFS>0.676(advanced liver fibrosis subgroup).Logistic regression analysis was used to identify risk factors associated with CV.Results Compared with CV_(2-3)<27.8%subgroup,BMI,FC-P and prevalence of MAFLD in CV_(2-3)≥>27.8%subgroup were lower(P<0.01).Compared with CV_(4-5)<26.0%subgroup,BMI,FC-P and prevalence of MAFLD in CV_(4-5)≥26.0%subgroup were lower(P<0.01).Compared with CV_(6-7)<24.8%subgroup,BMI,FC-P and prevalence of MAFLD in CV_(6-7)≥24.8%subgroup were lower(P<0.01).Logistic regression analysis showed that BMI,FC-P and HbAic were the influencing factors of CV_(2-3)≥27.8%;gender,age,BMI,FC-P and MAFLD were the influencing factors of CV_(4-5)≥26.0%;BMI,FC-P and MAFLD were the influencing factors of CV_(6-7)≥24.8%.Compared with T2DM group,CV_(2-3),CV_(4-5)s and CV_(6-7) in MAFLD group were decreased(P<0.01);FC-P was increased(P<0.01).Compared with CV_(2-3)<26.2%subgroup,BMI and FC-P in CV_(2-3)≥26.2%subgroup were lower(P<0.01).Compared with CV_(4-5)<24.4%subgroup,BMI and FC-P in CV_(4-5)≥24.4%subgroup were lower(P<0.01).Compared with CV_(6-7)<23.4%subgroup,BMI and FC-P in CV_(6-7)≥23.4%subgroup were lower(P<0.01).CV_(2-3) in advanced liver fibrosis subgroup was higher than that in suspect liver fibrosis subgroup(P<0.05).Logistic regression analysis showed that FC-P and HbAic were the influencing factors of CV_(2-3)≥26.2%;FC-P was the influencing factors of CV_(4-5)≥>24.4%;the course of DM and FC-P were the influencing factors of CV_(6-7)≥23.4%.Conclusions BMI and FC-P are the influencing factors of blood glucose fluctuation in T2DM patients treated with insulin.Blood glucose fluctuation in patients without MAFLD is large.FC-P is the influencing factor of blood glucose fluctuation in T2DM patients combined with MAFLD。