2型糖尿病合并非酒精性脂肪性肝病、进展性肝纤维化的危险因素及其预测效能

Risk factors for non-alcoholic fatty liver disease and progressive liver fibrosis in combination with type 2 diabetes and their predictive efficacy

目的分析2型糖尿病(T2DM)合并非酒精性脂肪性肝病(NAFLD)、进展性肝纤维化的危险因素及其预测效能。方法431例T2DM患者根据腹部超声结果分为T2DM合并NAFLD 316例(合并组)、单纯T2DM 115例(单纯组),316例T2DM合并NAFLD患者再根据NAFLD纤维化评分(NFS)分为进展性肝纤维化39例(进展组)、可疑肝纤维化122例(可疑组)、除外肝纤维化155例(除外组),采用单因素分析法和多因素Logistic回归分析法分析T2DM合并NAFLD、进展性肝纤维化的影响因素和危险因素,采用ROC曲线评估相关危险因素对T2DM合并NAFLD、进展性肝纤维化的预测效能。结果合并组和单纯组年龄、BMI、WBC、LY、AST、ALT、GGT、Alb、SUA、TC、TG、HDL-C、UHR、FIns、FC-P、HOMA-IR比较,P均<0.05;多因素Logistic回归分析结果显示,BMI、AST、Alb、TG、UHR、FC-P是T2DM合并NAFLD的独立危险因素;当BMI预测临界值为24.87 kg/m^(2)时,其诊断T2DM合并NAFLD的灵敏度为77.5%,特异度为72.2%;当AST预测临界值为19.5 U/L时,其诊断T2DM合并NAFLD的灵敏度为56.7%,特异度为78.3%;当Alb预测临界值为4.42 g/dL时,其诊断T2DM合并NAFLD的灵敏度为57.0%,特异度为58.3%;当TG预测临界值为1.45 mmol/L时,其诊断T2DM合并NAFLD的灵敏度为73.7%,特异度为65.2%;当UHR预测临界值为308.5时,其诊断T2DM合并NAFLD的灵敏度为66.5%,特异度为81.7%;当FC-P预测临界值为1.44 ng/mL时,其诊断T2DM合并NAFLD的灵敏度为67.1%,特异度为63.5%。进展组及可疑组和除外组性别、年龄、高血压史、BMI、PLT、AST/ALT、GGT、Alb、TBil、Scr、Cys-C、SUA、TC、TG、HDL、LDL、UHR、FPG、FIns、FC-P、HOMA-IR比较,P均<0.05;多因素Logistic回归分析结果显示,UHR、FC-P、HOMA-IR是肝纤维化进展的危险因素;当UHR预测临界值为313.98时,其诊断T2DM合并NAFLD进展性肝纤维化的灵敏度为84.6%,特异度为54.9%;当HOMA-IR预测临界值为5.56时,其诊断T2DM合并NAFLD进展性肝纤维化的灵敏度为69.2%,特异度为72.6%;当FC-P预测临界值为2.64 ng/mL时,其诊断T2DM合并NAFLD进展性肝纤维化的灵敏度为53.8%,特异度为82.3%。结论BMI、AST、Alb、TG、UHR、FC-P是T2DM合并NAFLD的危险因素,UHR、FC-P、HOMA-IR是T2DM合并NAFLD进展性肝纤维化的危险因素,各指标对于T2DM合并NAFLD及T2DM合并NAFLD进展性肝纤维化有良好的预测效能。

Objective To analyse the risk factors for type 2 diabetes mellitus(T2DM)combined with non-alcoholic fatty liver disease(NAFLD)and progressive liver fibrosis and their predictive efficacy.Methods Based on the results of abdominal ultrasound,431 patients with T2DM were divided into 316 patients with T2DM combined with NAFLD(com⁃bined group)and 115 patients with T2DM(simple group);316 patients with T2DM combined with NAFLD were further di⁃vided into 39 patients with progressive liver fibrosis(progressive group),122 patients with suspected liver fibrosis(suspi⁃cious group),and 155 patients without liver fibrosis(excluded group)according to the NAFLD fibrosis score(NFS).Uni⁃variate and multifactorial Logistic regression was used to analyze of influencing factors and risk factors for T2DM combined with NAFLD and progressive liver fibrosis.The ROC curve was used to assess the predictive efficiency of the risk factors on NAFLD and progressive liver fibrosis in T2DM combined with NAFLD.Results There were significant differences in age,body mass index(BMI),white blood cell(WBC),lymphocyte(LY),aspartate aminotransferase(AST),alanine aminotransferase(ALT),γ-glutamyl transpeptidase(GGT),albumin(Alb),serum urid acid(SUA),total cholesterol(TC),triglyceride(TG),high-density lipoprotein cholesterol(HDL-C),UHR,fasting insulin(FIns),fasting C-peptide(FC-P),and homeostatic model assessment of insulin resistance(HOMA-IR)between the combined group and simple group(all P<0.05);multifactorial Logistic regression analysis showed that BMI,AST,Alb,TG,UHR,and FC-P were risk factors for T2DM in combination with NAFLD.When the predicted critical value of BMI was 24.87 kg/m^(2),the sensi⁃tivity of T2DM combined with NAFLD was 77.5%and the specificity was 72.2%;when the predicted critical value of AST was 19.5 U/L,the sensitivity of T2DM combined with NAFLD was 56.7%and the specificity was 78.3%;when the pre⁃dicted critical value of Alb was 4.42 g/dL,its sensitivity was 57.0%and specificity was 58.3%;when the predicted criti⁃cal value of TG was 1.45 mmol/L,its sensitivity was 73.7%and specificity was 65.2%;when the predicted critical value of UHR was 308.5,its sensitivity was 66.5%and specificity was 81.7%;when the FC-P predicted critical value was 1.44 ng/mL,its sensitivity was 67.1%and specificity was 63.5%for the diagnosis of T2DM combined with NAFLD.There were significant differences in gender,age,history of hypertension,BMI,platelet(PLT),AST/ALT,GGT,Alb,total bilirubin(TBil),serum creatinine(Scr),cystatin C(Cys-C),SUA,TC,TG,HDL,low-density lipoprotein choles⁃terol(LDL),UHR,fasting plasma glucose(FPG),FIns,FC-P,and HOMA-IR between the progressive group,suspi⁃cious group and the excluded group(all P<0.05);multifactorial Logistic regression analysis showed that UHR,FC-P and HOMA-IR were risk factors for progressive liver fibrosis;when the predicted critical value of UHR was 313.98,the sensi⁃tivity of the diagnosis of progressive liver fibrosis in T2DM combined with NAFLD was 84.6%and the specificity was 54.9%;when the predicted critical value of HOMA-IR was 5.56,the sensitivity of the diagnosis of progressive liver fibro⁃sis in T2DM combined with NAFLD was 69.2%and the specificity was 72.6%;when the the predicted critical value of FC-P was 2.64 ng/mL,the sensitivity of the diagnosis of progressive liver fibrosis in T2DM combined with NAFLD was 53.8%,and the specificity was 82.3%.Conclusions BMI,AST,Alb,TG,UHR,and FC-P are risk factors for NAFLD in combination with T2DM;UHR,FC-P,and HOMA-IR are risk factors for progressive liver fibrosis in T2DM combined with NAFLD.Each index has good predictive efficacy for T2DM combined with NAFLD and for progressive liver fibrosis in T2DM combined with NAFLD.