| Backgroud:Hemoglobin A1c(HbA1c)is viewed as the gold standard which assess the glycemic condition in the past 3 months.However,HbAlc could not reflect glycemic fluctuation.In addition,numerous factors would influence the accuracy of HbAlc to reflect glycemia in patients with diabetic kidney disease(DKD).Continuous glucose monitoring(CGM)provide lots of index,which can capture transient hypoglycemia and hyperglycemia.CGM derived glucose management indicator(GMI)could complement to HbA1c effectively in patients with DKD.Morever,the discordance between HbAlc and GMI was reported to be correlated with hypoglycemia or hyperglycemia.This study was conducted to understand the glycemic characteristcs of patients with DKD stratified by renal function,investigate the relationship between HbAlc and CGM index and to explore the application of combination of HbAlc-GMI discordance for the glycemic management of DKD patients.Methods:According to the inclusion and exclusion criteria,this study enrolled a total of 156 DKD patients in Nanfang Hospital of Southern Medical University from October 2018 to October 2022.Demographic information,laboratory data and CGM index were retrospective collected to be analyzed.Included patients were grouped into stage 1-2(estimated glomerular filtration rate,eGFR≥60 ml/min/1.73m2,n=64)、stage 3(eGFR≥60 ml/min/1.73m2,n=56)and stage 4-5(eGFR≥60 ml/min/1.73m2,n=36).Glycemic profiles were characterized and the relationship of HbAlc and CGM index were described by Pearson correlation analysis.Linear regression was performed to explore the association of CGM inex and renal function.At the same time,enrolled patients were divided into training set and validation set by simple random sampling with 6:4 ratio.Based on biochemical data,models to calculate GMI were constructed.In the training set,univariate linear regression analysis was performed for each clinical variable,and univariate significant variables(P<0.05)were selected into multivariate stepwise regression analysis.For model validation,the estimated GMI was compared with real GMI by Pair’s t test.Results:1.Comparision of the glycemia profile in different stages with DKD revealed that fasting blood glucose and CGM derived mean blood glucose did not differ across groups(P>0.05).However,compared with stage G1-2,the level of HbAlc in stage G3 and G4-5 decreased(P=0.025).Morever,the risk of hypoglycemia increased in stage G4-5 compared to stage G1-2 and G3(P<0.001).the level of hemoglobin(Hb)was lower in stage G3 and G4-5 compared with G1-2(P<0.01).2.Compared with stage G1-2,the correlation of HbAlc and MBG was weaker in stage G3(r=0.576,P<0.01;r=0.266,P=0.047).There was no significant correlation between HbAlc and MBG in stage G4-5(r=0.296;P=0.079).Morever,multiple linear analysis showed that Hb might be a influential factor of HbAlc in DKD patients.3.Multiple linear regression analysis showed significant association of TBR and eGFR level(P<0.05).However,when anemia status was taken into the regression model,no significant correlation was found between TBR and eGFR level(P=0.16).4.With the decrease of eGFR level,HbAlc level reduced but GMI did not differ across subgroups.The gap between HbAlc and GMI get reduced as the level of eGFR decreased(P=0.009).Multiple linear regression analysis demonstrated significant association between HbAlc-GMI and renal function(P=0.015).5、Based on renal function,HbAlc,anemia status,fasting blood glucose and serum albumin(Alb),linear regression equation was built for GMI estimation to better assess glycemic control.Conclusion:1,Impaired renal function was accompanied with an increased risk of hypoglycemia.Application of TBR would be a supplement to HbA1c to assess hypoglycemic condition.2.HbA1c was less accurately to reflect glycemia in DKD patients with eGFR<60ml/min/1.73m2.3.The combination of HbA1c and GMI help better understand the glycemic profile of DKD patients.4.A linear equation based on clinial parameters would estimate GMI,which could better assess glycemic status for DKD. |
PDF Full Text Request