Clinical Value Of Serum And Lens Advanced Glycation End Products For Diabetes And Its Complications In Chinese Population

Part Ⅰ: Association between serum advanced glycation end products peptide and diabetes mellitus and its complications in Chinese populationBackground and Aims: Advanced glycation end products(AGEs),formed in a multistep process by glycation and oxidation of proteins,are reported to underpin the development of diabetes and its complications.Several studies have shown that serum AGEs levels in diabetes mellitus(DM)patients are higher than healthy controls.Our previous studies have shown that advanced glycation end products peptides(AGEs-P),detected by the flow injection assay,is effective in the screening of DM;however,its association with DM in the general population has not yet been evaluated.Studies have shown that in patients with type 2 diabetes mellitus(T2DM),serum AGEs levels are related to the severity of diabetic retinopathy(DR)and diabetic kidney disease(DKD).However,most of these studies used enzyme-linked immunosorbent assay(ELISA)to detect AGEs,and were focused on T2 DM patients at high risk for cardiovascular events.Given these,the primary aim of this study is to investigate the association of AGEs-P with diabetes in the general Chinese populations and to assess its value as a screening or predicting tool for diabetes mellitus(DM).The secondary aim of this study is to evaluate the clinical value of AGEs-P in DR and DKD.Methods: 1.A cross-sectional survey was conducted from July 2018 to December 2018 among permanent residents aged 20-70 in 10 regions from 8 provinces across China.Data on demographic characteristics,physical examination,lifestyle and laboratory test results were collected.Nondiabetic participants were selected as an observational cohort.The mean follow-up time was 1.5 years and DM was assessed as the endpoint.2.Fasting plasma glucose(FPG),2 hour-plasma glucose(2h-PG)after a 75-g oral glucose tolerance test(OGTT)and hemoglobin A1c(Hb A1c)were assessed.Diabetes was determined by World Health Organization(WHO)1999 diagnostic criteria,self-reported history,and/or use of antidiabetic drugs.3.The Early Treatment Diabetic Retinopathy Study(ETDRS)scale was used to diagnose DR.Diabeties-specific retinopathy was defined as ETDRS ≥ 20.Very mild nonproliferative diabetic retinopathy(NPDR)was defined as ETDRS ≥ 20 and < 31,and mild and worse NPDR was defined as ETDRS ≥ 31.4.DKD was defined according to urinary albumin-to-creatinine ratio(UACR)or estimated glomerular filtration rate(e GFR).Participants were grouped by UACR: Group A1: UACR < 30 mg/g;Group A2: UACR 30-300 mg/g;Group A3: UACR > 300 mg/g.5.Serum AGEs-P was detected though flow injection assay.Each serum specimen was tested three times.AGEs-P concentrations were calculated based on the standard curve plotted according to the peak area of the standard solution.6.Associations between AGEs-P and DM,DR,and DKD were analyzed using Student’s t-test,one-way ANOVA,correlation analyses and logistic regression analyses.The receiver operating characteristic(ROC)curve was used to evaluate the efficacy of AGEs-P alone or combined with FPG and Hb A1 c for screening and predicting diabetes.Results: 1.A total of 3,756 subjects were included in the cross-sectional observation.Subjects were divided into normal glucose tolerance(NGT)(n = 2,330),prediabetes mellitus(pre DM)(n = 824),and DM(n = 602),according to glucose status.The overall mean value of AGEs-P was 21.76 U/L,and sequentially increased in NGT(21.63 ± 1.69 U/L),pre DM(21.80 ± 1.66 U/L)and DM groups(22.09 ± 2.47 U/L)(p < 0.05).Correlation analyses showed that AGEs-P was positively correlated with FPG,2h-PG,and Hb A1 c in the overall population(p < 0.05).After stratification by glucose metabolism status,AGEs-P only remained positively correlated with FPG in the pre DM group(r = 0.09,p < 0.01).2.A total of 3,154 subjects without DM were selected.After an average follow-up of 1.5 years,2,188 subjects were included in the analysis,of which 161 subjects progressed to diabetes.AGEs-P levels in subjects who progressed to DM was higher than those who did not(22.16 ± 2.67 vs 21.64 ± 1.65,p < 0.05).3.Regression analyses showed that AGEs-P was an independent risk factor for DM(odds ratio [OR] 1.06,95% confidence interval [CI] 1.02 ~ 1.11,p < 0.01),and was independently correlated with the risk of incident DM(OR 1.12,95% CI 1.03 ~ 1.22,p < 0.01)after adjusting for the multivariate model.Subgroup analyses showed that the association between AGEs-P and the risk of DM was affected by e GFR and Hb A1 c.4.The ROC results showed that the area under curve(AUC)of AGEs-P for screening DM was 0.550.The sensitivity and specificity of AGEs-P were 47.18% and 62.73% respectively,which were lower than that of FPG and Hb A1 c.The AUC,sensitivity and specificity of AGEs-P for prediting DM were 0.559,46.58% and 64.63% respectively,which were also inferior to that of FPG and Hb A1 c.The screening and prediction efficiency of AGEs-P was improved after combining with FPG and Hb A1 c.5.Participants were classified in the following groups: no DR(n = 496),very mild nonproliferative DR(NPDR)(n = 50),mild and worse NPDR(n = 47).The levels of AGEs-P in the very mild NPDR group(22.45 ± 3.46 U/L)and the mild and worse NPDR groups(22.13 ± 2.78 U/L)were higher than those in the non-DR group(22.05 ± 2.34 U/L)but the difference was not of statistical significance(p > 0.05).6.In regard to DKD,participants were divided into Group A1: UACR < 30 mg/g(n = 378),Group A2: UACR 30-300 mg/g(n = 170),and Group A3: UACR > 300 mg/g(n = 22).The level of AGEs-P increased with UACR(Group A1: 21.99 ± 2.42 U/L;Group A2: 22.08 ± 2.10 U/L;Group A3: 22.94 ± 4.85 U/L),but there was no significant difference(p > 0.05).7.Multivariable regression analyses showed that there was no significant correlation between AGEs-P and DR or DKD(p > 0.05).Conclusions: 1.AGEs-P was positively correlated with the prevalence and incidence of diabetes in general Chinese populations.2.The efficiency of AGEs-P in screening diabetes in the general Chinese populations was inferior to FPG and Hb A1 c,which can be improved after combining with FPG and Hb A1 c.3.Based on the general Chinese population,this study found the level of AGEs-P in the DR or DKD group was higher than the non-DR or non-DKD group.There was no significant correlation between AGEs-P and the DR or DKD after multivariable-adjustment.Part Ⅱ: Influencing factors of lens autofluorescence ratio in the general Chinese populationBackground and Aims: AGEs are commonly measured in serum by ELISA and highperformance liquid chromatography.In addition,recent studies suggest that AGEs can also be measured in the skin or lens by noninvasive fluorescence spectroscopy.Although serum AGEsP is positively correlated with blood glucose levels and the risk of DM,its efficacy in screening DM in Chinese populations is inconsistent,suggesting that there may be factors other than glucose that affect AGEs levels.Studies have reported that diets,smoking,and renal function may affect serum AGEs levels,while skin AGEs levels are related to gender,body mass index(BMI),blood pressure,blood lipids,and e GFR.However,most studies focused on serum or skin AGEs.The detection method of serum AGEs has the advantages of high sensitivity and specificity,yet procedures from these approaches are both cumbersome and expensive;skin AGEs is noninvasive,fast and easily performed,however,results are often affected by the skin reflection,the application of creams,and the extreme vasodilatation and vasoconstriction at the measurement site.It is reported that lens AGEs can be reflected by lens fluorescence ratio(LFR),which is a relatively novel non-invasive AGEs detection method with the advantages of simplicity and rapidity.Nowadays,few studies have explored the influencing factors of LFR in general populations.This study aims to explore the influencing factors of LFR in the general Chinese population,and explore the factors that may affect the relationship between Hb A1 c and LFR.Methods: 1.This cross-sectional study was conducted in the general population in 8 different provinces across China between May 2020 and January 2021.Details are mentioned in part 1.2.LFR was measured via a scanning confocal lens fluorescence biomicroscope optical system(Clearpath DS-120).The average ratio of the lens autofluorescence to scattered light(that is,LFR)in the central portion of the lens is then calculated.3.Participants were classified according to LFR tertiles(the lowest LFR group: < 17.30%;the middle LFR group: 17.30%-21.94%;the highest LFR group:≥ 21.94%).Association between LFR and the risk factors of interest was analyzed by one-way ANOVA and the generalized linear model.Subgroup analysis was used to analyze the factors that may affect the relationship between Hb A1 c and LFR.Results: 1.LFR was tested in 5056 participants.FPG,2h-PG,Hb A1 c,age,proportion of women,total cholesterol(TC),triglycerides(TG),high-density lipoprotein cholesterol(HDL-C)and lowdensity lipoprotein cholesterol(LDL-C)increased with LFR(p < 0.05).BMI,serum creatinine,left eye vision,right eye vision,lens thickness and the proportion of smoking decreased with LFR(p < 0.05).Females had higher LFR levels than males(20.49 ± 5.56 vs 19.30 ± 5.31,p < 0.001).After age stratification,gender difference were still observed in participants over the age of 45(p < 0.001).2.The results of the multivariate generalized linear model showed that Hb A1c(B 0.87,95% CI 0.32 ~ 1.42),female(B 3.39,95% CI 1.54 ~ 5.24)and lens thickness(B-1.01,95% CI-2.00 ~-0.02)were independently correlated with LFR after adjusting for blood glucose,age,gender,BMI,blood pressure,blood lipid,serum creatinine,smoking,and visual acuity.3.Subgroup analysis showed that pupil diameter may affect the association between Hb A1 c and LFR(p < 0.05).Conclusions: 1.A higher LFR in women was observed in general Chinese adults.2.In the multivariable model,LFR was only related to Hb A1 c,gender,and lens thickness.3.Subgroup analyses showed that pupil diameter may affect the relationship between Hb A1 c and LFR.Part Ⅲ: The value of lens autofluorescence ratio in screening for diabetes in Chinese populationBackground and Aims: Both serum and skin AGEs are positively correlated with blood glucose levels and the risk of DM;the second part of this study found that Hb A1 c was independently positively correlated with LFR,suggesting that LFR may be associated with DM.Previous studies have confirmed that LFR in DM patients are higher than those in non-DM subjects,and is positively correlated with blood glucose levels.However,interpretations of these observations are largely limited by the small sample sizes and possible selection bias.It remains unclear whether LFR is related to DM in the general Chinese population,and whether it can be used as a screening tool for DM.Thus,we conducted this study to investigate the association between LFR and DM and assess whether it could be used as an indicator for screening DM in the general Chinese population.Methods: 1.Research subjects and diagnostis criteria of DM are the same as the second part.The New Chinese Diabetes Risk Score(NCDRS)based on age,sex,BMI,waist circumference,systolic blood pressure,and family history of diabetes,was calculated for each participants.2.LFR was determined by the biomicroscope(Clear Path DS-120).3.Participants were divided into quartiles according to LFR level(1st quartile: < 16.06%;2nd quartile: 16.06%-19.30%;3rd quartile: 19.30%-23.26%;4th quantile group: > 23.26%).Correlation and logistic regression analyses were performed to examine the association of LFR with DM.The relationship between LFR and DM was explored by mediation analysis.ROC was used to evaluate the efficacy of LFR in screening DM.The LFR and NCDRS combined diagnosis model was obtained by logistic regression and ROC curve.The optimal cut-off point and corresponding sensitivity and specificity was determined by the highest Youden index(Youden index = sensitivity + specificity-1).Results: 1.A total of 4,705 participants were included in the final data analysis.NGT group included 2,907 participants,pre DM group included 1,129 participants,and DM group included 669 participants.LFR increased sequentially among NGT group(19.17 ± 5.01 %),pre DM group(20.16 ± 5.17 %)and DM group(23.27 ± 6.51 %)(p < 0.001).2.LFR was positively correlated with FPG(r = 0.30,p < 0.001),2h-PG(r = 0.12,p < 0.001),and Hb A1c(r = 0.29,p < 0.001)in the overall population.In the DM group,LFR was significantly correlated with FPG(r = 0.36,p < 0.001),Hb A1c(r = 0.36,p < 0.001),but not with 2h-PG(p = 0.606).3.Risk of DM increased by 6% for every 1% increase in LFR(OR 1.06,95% CI 1.04 ~ 1.08,p < 0.001)after adjusting for multivariable.Participants in the highest quartile showed a higher risk of DM compared with those in the lowest quartile after multivariable-adjustment(OR 1.83,95% CI 1.33 ~ 2.52,p < 0.001).4.Mediation analysis showed that triglyceride-glucose index(Ty G),may underline the relationship between high LFR and risk of DM.Its mediating effects accounted for 27.7% of the total effect.5.The AUC,sensitivity and specificity of LFR were 0.673,63.38% and 62.81% respectively in the overall population,which were similar to NCDRS(p < 0.001).The AUC increased to 0.756 after combining with NCDRS,with the sensitivity and specificity of 69.51%,and 67.91% respectively.For participants younger than 45 years old,the AUC,sensitivity and specificity of LFR for screening DM were 0.749,61.29% and 82.17% respectively.The specificity of LFR screening for DM was higher than the NCDRS score(82.17% vs 58.65%)in people under 45 years of age.After combining with NCDRS,the AUC increased to 0.873 and the specificity increased to 81.64%.Conclusions: 1.LFR was positively associated with glycemia and the risk of diabetes in the general Chinese population.2.Ty G index was a mediator variable between LFR and risk of DM.3.LFR showed well performance in screening DM for people younger than 45 years old.The screening efficiency of LFR can be improved after combing with NCDRS.Part Ⅳ: Association between lens autofluorescence ratio and diabetes retinopathy in Chinese populationBackground and Aims: AGEs have been confirmed to have a clear pathogenic role in DR.The level of LFR increases with the severity of DR and is an independent risk factor for DR in T2 DM patients.However,most extant literature is based on single-center studies with small sample sizes,and the association between LFR and DR as well as its screening value for DR have not been reported.The purpose of this study was to investigate the association between LFR and DR in a large,multi-center general population,and explore its role in the screening of DR in Chinese population.Methods: 1.A cross-sectional study design was conducted in 8 provinces in China between May 2020 and January 2021,and enrolled 590 diabetic patients aged 20-70 years.Details are shown in part 1.2.Lens autofluorescence biomicroscope(Clear Path DS 120)was used to measure the LFR of the eye.3.Grading of fundus images were based on the ETDRS score.DR was defined as ETDRS ≥ 20.4.Groups were divided according to LFR tertiles(lowest group: LFR < 20.20%;middle group: LFR 20.20%-25.89%;highest group: LFR ≥ 25.89%).Cochran-Armitage trend test was used to analyze differences in the prevalence of DR in LFR groups.Logistic regression analyses were used to examine the association of LFR and DR.ROC curve was used to evaluate the screening efficacy of LFR alone and combined with Hb A1 c in screening DR.Results: 1.LFR in the DR group was higher than that in the NDR group(28.29 ± 6.63 vs 22.43 ± 6.06,p < 0.001).2.Cochran-Armitage trend test showed that,LFR was positively correlated with prevalence of DR(p < 0.001).Based on ETDRS ≥ 20,risk of DR increased 1.93-fold for per standard deviation(SD)increase of LFR after adjustment for confounders including age,gender,BMI,systolic blood pressure,diastolic blood pressure,Hb A1 c,FPG,2h-PG,HDL-C,TG,e GFR,and smoking history(OR 1.93,95% CI 1.51 ~ 2.47,p < 0.001).Based on ETDRS ≥ 31,risk of DR increased 2.34-fold for per SD increase of LFR after multivariable-adjustment(OR 2.34,95% CI 1.68 ~ 3.26,p < 0.001).3.Based on ETDRS ≥ 31,the AUC of LFR was 0.783,the sensitivity and specificity was 64.15%,and 84.36%,respectively.The AUC,sensitivity and specificity of Hb A1 c for screening DR were 0.816,71.70%,and 80.26%,respectively.The AUC increased to 0.865 and the sensitivity increased to 94.34% after combining LFR with Hb A1 c.Conclusions: 1.LFR was significantly elevated in the DR patients and was positively associated with the risk of DR.2.LFR showed high AUC and specificity for DR screening in Chinese population.The screening efficiency were improved after combining with Hb A1 c.