ObjectiveTo explore the related risk factors of vitreous macular interface abnormalities (VMIA) in eyes with diabetic macular edema (DME) after receiving anti-vascular endothelial growth factor (VEGF) drug treatment, and to evaluate the influence of VMIA on the best corrected visual acuity (BCVA) and central retinal thickness (CRT) of the affected eyes. MethodsA retrospective cohort study. From January 2021 to January 2023, 285 DME patients (285 eyes) who received anti-VEGF drug treatment at Tianjin Medical University Eye Hospital and had no VMIA at baseline were included in the study. All affected eyes underwent BCVA examination, and CRT was measured by optical coherence tomography. The treatment plan was a monthly stress therapy for the initial three months, followed by treatment as needed. According to whether VMIA was formed 12 months after treatment, the affected eyes were divided into the VMIA formation group and the non-VMIA formation group, that was further subdivided based on the VMIA classification. Logistic regression model was used to analyze the risk factors for VMIA formation and its influence on BCVA and CRT 12 months after treatment. ResultsTwelve months after treatment, among 285 eyes, 111 eyes (38.9%) developed VMIA (VMIA formation group), and 174 eyes (61.1%) did not develop VMIA (non-VMIA formation group). Logistic regression analysis showed that a thinner baseline CRT [odds ratio (OR=0.99, 95% confidence interval (CI) 0.98-0.99, P=0.04] was associated with a higher number of anti-VEGF drug injections (OR=1.12, 95%CI 1.02-1.23, P=0.02) was a risk factor for the formation of VMIA. However, the formation status of VMIA itself was not an influencing factor for BCVA (OR=1.89, 95%CI 0.98-3.67, P=0.06) or CRT (OR=1.34, 95%CI 0.30-0.83, P=0.40) at 12 months after treatment. The intergroup comparison showed that at 12 months after treatment, the improvement degrees of BCVA and CRT in the VMIA formation group were both worse than those in the non-VMIA formation group (t=2.99, 2.07; P＜0.00, 0.05). Furthermore, in the VMIA subtype analysis, the improvement degree of CRT in the affected eyes with epiretinal membrane (ERM) was significantly lower than that in the affected eyes without ERM (t=4.31, P＜0.001). ConclusionsThinner baseline CRT and more injection times are associated with the occurrence of VMIA; compared with the eyes without VMIA formation, the improvement of BCVA and CRT in the eyes with VMIA formation is less during the 12-month follow-up period after treatment. The formation of VMIA has no significant effect on BCVA or CRT at 12 months after treatment. The improvement effect of CRT is the poorest in patients with ERM.