Metabolomics In Diabetic Retinopathy

Objectives:We study the changes of plasma metabolite relevant to diabetic retinopathy in order to identify of the specific markers to assist in the diagnosis.Methods:We enrolled hospitalized patients in Ophthalmology Department and Endocrinology Department of First Hospital of Jilin University from January 2017 to October 2018.All enrolled patients were diagnosed with type 2 diabetes.Among them,66 cases were females and 107 cases were male.According to the results of fundus examination,the patients were divided into the non-diabetic retinopathy group and the diabetic retinopathy group.We analyzed the general data and laboratory examination data of the patients,and collected the morning fasting plasma for metabolite analysis.All the data were analyzed by SPSS22.0 software.Result:1、This study included a total of 173 patients,124 cases in diabetic retinopathy group and 49 cases in control group.Of diabetic retinopathy group,73 cases were males(58.9%)and 51 cases were females(41.1%);the median age was 55.5(48.5,63)years;the course of diabetes was 11.9(6.0,17.0)years.Among them,68 cases were in non-proliferative phase(NPDR)and 56 cases were in proliferative phase(PDR).Of control group,34 cases were males(69.4%)and 15 cases were females(30.6%);the median age was 55.5(50.5,62)years and the course of diabetes was 10.9(6.0,14.5)years.2、There were no significant differences in the proportion of men and women,age,diabetes duration,BMI,blood glucose,glycosylated hemoglobin,2-hour postprandial glucose,high-sensitivity c-reactive protein,high-density lipoprotein,low-density lipoprotein,triglyceride,cholesterol,serum creatinine,urea nitrogen,serum uric acid,fasting c-peptide,smoking and drinking situation between DR group and control group(P >0.05).3、The retinol binding protein,24-hour urinary protein and 24-hour urinary microalbumin increased gradually in the control group,non-proliferation group and proliferation group(P<0.05).4、89 metabolites were detected.Logistic regression analysis showed that 10 of them were independently related to DR.The increase of Ala,Glu,Gly,Phe/Tyr and C4/C3 were related to the occurrence of DR,while the decrease of Gln,Lys,C16-0H,,C3/C16,C5-OH/C8 were related to the occurrence of DR.5、ROC curve indicated that 13 metabolites were of diagnostic value for DR.The optimal cutoff values were Ala > 96.436 umol/L,Glu > 165.65 umol/L,Gly > 165.65 umol/L,Gly > 135.5 umol/L,Phe/Tyr > 0.989,C4/C3 > 0.1335;Gln < 8.0005 umol/L,Lys < 98.96 umol/L,C16-0H < 0.0195umol/L,C3/C16 < 0.0345,C3/C16 < 0.747 226,.Conclusion:1、Metabonomics plays an important role in the occurrence and development of DR.Ala,Glu,Gly,Phe/Tyr,C4/C3,Gln,Lys,C16-0H,C3/C0,C3/C16,C5/C0,C5-OH/C8 and C5-OH/C0 can be used as markers of DR.2、Metabolic markers are valuable in the diagnosis of diabetic retinopathy.The diagnostic values from high to low are C4/C3、Gly、C3/C0、C5-OH/C0、Glu、Phe/Tyr、Ala、C16-OH、Gln、C3/C16、Lys、C5/C0、C5-OH/C8.When Ala > 96.436umol/L 、Glu >165.65umol/L、Gly >135.5umol/L、Phe/Tyr >0.989、C4/C3 >0.1335;Gln <8.0005umol/L、Lys <98.96 umol/L、C16-0H <0.0195umol/L、C3/C16 <0.747、C5-OH/C8 <3.226,it can be used as a diagnostic index of DR.