Impact of Anti-VEGF treatment on retinal anatomy and electrophysiological function in diabetic macular edema

The objective of this study is to evaluate changes in the foveal avascular zone (FAZ), vascular density (VD), and electrophysiological parameters in patients with diabetic macular edema (DME) following intravitreal bevacizumab therapy.

The study comprised 27 eyes of 22 patients with treatment-naive DME who received a loading phase of three consecutive monthly intravitreal bevacizumab injections, together with 30 eyes of 15 healthy controls. All participants underwent optical coherence tomography (OCT), OCT angiography (OCTA), pattern electroretinography (PERG) and visual evoked potentials (VEP) at baseline and at one and three months. As both eyes of some participants were included, all comparisons used generalized estimating equations clustered on subject, with cluster-robust variance estimation.

Central retinal thickness decreased significantly after treatment, from 409.2 ± 105.9 µm at baseline to 368.4 ± 97.8 µm at 1 month (p = 0.007) and 352.6 ± 93.6 µm at 3 months (p < 0.001). Best-corrected visual acuity improved from 0.69 ± 0.47 to 0.57 ± 0.44 logMAR, but this change did not reach statistical significance (p = 0.166). FAZ measurements showed no significant change at either visit (all p > 0.57) and superficial VD remained stable (p = 0.404 and p = 0.185), whereas measured deep VD increased significantly at both 1 month (+ 1.02%, p = 0.045) and 3 months (+ 1.85%, p = 0.002), a change that may reflect improved OCTA signal penetration after edema resolution rather than true reperfusion. The 100% contrast response magnitude increased at 3 months (p = 0.040) but not at 1 month (p = 0.293), and did not retain significance after adjustment for multiple comparisons. Lc and Hc amplitudes increased significantly at 3 months (p = 0.021 and p < 0.001), whereas latencies remained prolonged relative to controls at baseline (Lc, p = 0.001; Hc, p = 0.004).

Intravitreal bevacizumab therapy in treatment-naive DME patients produced significant anatomical improvement, accompanied by changes in selected electrophysiological parameters. The stability of FAZ measurements, the apparent increase in deep VD, and the improved VEP amplitudes, in the absence of concurrent normalisation of latency, suggest that structural, microvascular and functional recovery may follow different time courses. This indicates that anatomical response alone may overestimate functional benefit. The electrophysiological changes observed were modest and require confirmation in larger cohorts.