糖尿病黄斑水肿患者基线血清学指标与玻璃体内抗VEGF治疗未达标的关系及其预测研究OA

Objective To investigate the relationship between baseline serological parameters and suboptimal re-sponse to intravitreal anti-vascular endothelial growth factor(VEGF)therapy,and to explore their predictive value in pa-tients with diabetic macular edema(DME).Methods Totally 100 DME patients undergoing intravitreal anti-VEGF therapy at our hospital between June 2022 and June 2024 were enrolled and divided into the suboptimal-response group and the opti-mal-response group based on treatment response.Clinical data,including baseline characteristics and serological parame-ters,were collected.Indicators with significant differences between the two groups were included in Logistic regression analysis to identify risk factors affecting the therapeutic response to intravitreal anti-VEGF injection.A nomogram prediction model was constructed based on the regression results.The predictive performance of the model for suboptimal response was assessed using receiver operating characteristic curves.Internal validation was performed via the Bootstrap method to evaluate the discrimination,calibration and clinical applicability of the model.Results There were statistically significant differences in age,duration of diabetes,baseline central macular thickness(CMT),C-reactive protein(CRP),interleukin-6(IL-6),glycated hemoglobin(HbA1c),and total bilirubin(TBil)between the two groups(all P＜0.05).Stepwise regres-sion analysis revealed that CRP,IL-6,HbA1c and TBil were influencing factors for suboptimal response to intravitreal anti-VEGF therapy in DME patients.The area under the curve(AUC)of the nomogram prediction model was 0.955,with a sen-sitivity of 0.857 and a specificity of 0.973.Its predictive performance was significantly superior to that of individual predic-tors(AUCs of CRP,HbA1c,IL-6 and TBil were 0.714,0.801,0.858 and 0.689,respectively).After Bootstrap correction,the AU Cwas0.942(95％CI:0.903-0.994).The calibration curve of the model was nearly consistent with the ideal curve,and the decision curve demonstrated a positive net benefit across the threshold probability range of 0.03-1.00.Conclusion The nomogram constructed based on CRP,HbA1c,IL-6 and TBil demonstrates preliminary predictive potential for treatment outcomes of anti-VEGF therapy in DME patients,yet further multicenter and large-sample studies are required for valida-tion.