Analysis of risk factors for macular edema in patients with type 2 diabetes and construction of a nomogram model

Chu Hai; Hu Fengyun; Ma Kai; Chen Fenfen; Yang Yu; Zheng Chun; Li Shasha

doi:10.11986/j.issn.1673-873X.2025.10.13

Abstract：Objective To analyze the risk factors for macular edema(ME)in patients with type 2 diabetes(T2DM)and con-struct a nomogram model to provide a reference for clinical decision-making.Methods A total of 208 T2DM patients admitted to our hospital from March 2023 to February 2025 were selected as study participants.Based on the presence or absence of ME,they were assigned to the ME group(n=76)and the non-ME group(n=132).The risk factors for ME in T2DM pa-tients were analyzed using the multivariate logistic stepwise regression analysis.A nomogram model was constructed to predict the risk of ME in T2DM patients,and the prediction value was evaluated using receiver operating characteristic(ROC)curves.Results In the ME group,patients were older,had a higher proportion with a history of hypertension,a T2DM duration of≥10 years,and a higher proportion using insulin compared to the non-ME group,with statistically significant differences(P<0.05).Additionally,fasting blood glucose,hemoglobin A1c(HbA1c),total cholesterol,low-density lipoprotein cholesterol(LDL-C),and blood uric acid levels were higher,and estimated glomerular filtration rate(eGFR)was lower in the ME group compared to the non-ME group,with statistically significant differences(P<0.05).Multivariate analysis showed that a histo-ry of hypertension(OR=2.243,95%CI:1.552-3.243),duration of T2DM≥10 years(OR=2.593,95%CI:1.708-3.937),high HbA1c(OR=3.025,95%CI:1.943-4.711),high LDL-C levels(OR=2.370,95%CI:1.611-3.487),and relatively high eGFR(OR=0.475,95%CI:0.333-0.679)were influence factors for the development of ME in T2DM pa-tients(P<0.05).The concordance index(C-index)of the nomogram model was 0.895(95%CI:0.844-0.946),with the o-riginal curve close to the ideal curve.Decision curve analysis showed that when the threshold probability>0.05,the net bene-fit rate of the model for predicting ME in T2DM patients was>0.The area under the curve(AUC)of the nomogram model for predicting ME in T2DM patients was 0.901(95%CI:0.850-0.952),with a specificity of 68.96%and a sensitivity of 92.55%.Conclusion The influence factors for ME in T2DM patients include a history of hypertension,duration of T2DM,HbA1c,LDL-C,and eGFR.The nomogram model constructed based on these factors has good predictive performance and can serve as an important tool to guide clinical treatment.