Baseline microperimetry and metabolic status predict functional outcomes in diabetic macular oedema: a prospective cohort study of anti-VEGF therapy.

AIMS: Diabetic macular oedema (DME) is a leading cause of visual impairment with anti-vascular endothelial growth factor (anti-VEGF) therapy representing first-line treatment. However, best-corrected visual acuity (BCVA) cannot fully capture subtle retinal functional improvements. As a precise functional examination technique, microperimetry can quantitatively evaluate retinal sensitivity and fixation stability, exhibiting higher sensitivity for detecting early retinal functional changes. This study aimed to compare the efficacy of microperimetry versus BCVA in detecting early functional changes following anti-VEGF therapy and to evaluate the predictive value of baseline microperimetry parameters and metabolic factors for treatment functional outcomes.

MATERIALS AND

METHODS

This prospective, single-centre study enrolled 58 patients with DME who received intravitreal Conbercept injection. Comprehensive assessments, including BCVA, swept-source optical coherence tomography (SS-OCT), OCT angiography (OCTA), and microperimetry (Microperimeter-3), were performed at baseline, 1 day, and 1 month post-injection. Cohen's d effect sizes were calculated to evaluate treatment sensitivity. Correlation analyses were conducted to explore the associations between post-treatment improvements in microperimetry parameters and baseline metrics, including the Triglyceride-Glucose (TyG) Index.

RESULTS

Microperimetry demonstrated significantly larger effect sizes (Cohen's d ranging from 0.51 to 1.82) compared to BCVA (d=-0.61). Microperimetry detected significant improvements in retinal sensitivity as early as 1 day after intravitreal injection. Among patients with unchanged BCVA (n = 26), microperimetry revealed obvious retinal functional benefits at the 1-month follow-up (e.g. foveal sensitivity, p < 0.0001). Baseline bivariate contour ellipse area (BCEA) values were markedly elevated in BCVA non-responders (e.g. BCEA 68%: 4.68 vs. 1.18 deg2, p < 0.0001), identifying BCEA as a sensitive predictor for treatment response. The TyG index showed a strong negative correlation with the improvement in foveal sensitivity (r=-0.483, p = 0.0001). Exploratory threshold analysis identified a TyG index cutoff value of 9.78 for distinguishing functional treatment responses, which warrants further large-sample validation.

CONCLUSIONS

Microperimetry is more sensitive than BCVA in assessing functional outcomes in DME. Baseline retinal functional and metabolic parameters serve as key predictors for treatment efficacy, which provides a novel theoretical basis for personalized and precise clinical management of DME.