Retinal Microvascular and Structural Findings in Children with Peripheral Arteriovenous Malformations Receiving Systemic Bevacizumab: An OCTA-Based Case–Control Study

Background: Systemic bevacizumab is increasingly used in selected pediatric patients with peripheral arteriovenous malformations (AVMs), particularly when interventional or surgical treatment is unsuitable. Although vascular endothelial growth factor (VEGF) contributes to physiological vascular integrity, the effects of systemic VEGF inhibition on the normal retinal microvasculature in children remain poorly understood. Optical coherence tomography angiography (OCTA) enables noninvasive assessment of retinal microvascular and structural changes. Aim: To evaluate retinal microvascular and thickness parameters using OCTA in pediatric patients with peripheral AVMs receiving systemic bevacizumab and to assess potential subclinical ocular effects. Methods: This cross-sectional case–control study included 60 eyes of 30 participants: 15 pediatric patients with peripheral AVMs treated with systemic bevacizumab for at least 6 months and 15 age- and sex-matched controls. All participants underwent comprehensive ophthalmologic examination and OCTA imaging. Superficial and deep capillary plexus vessel densities, radial peripapillary capillary plexus vessel density, foveal avascular zone parameters, fractal dimension, outer retinal and choriocapillaris flow areas, and retinal thickness were analyzed using mixed-design analysis of variance. Results: No significant differences were observed between groups in vessel densities, foveal avascular zone parameters, fractal dimension, flow areas, or total and outer retinal thickness (p > 0.05). Inner retinal foveal thickness was significantly lower in the bevacizumab group in both eyes (right: p = 0.044; left: p = 0.046). Conclusions: The lower inner retinal foveal thickness observed in the bevacizumab group represents a cross-sectional structural difference in uncertain clinical significance. Although experimental studies provide a biological rationale for investigating retinal changes during VEGF inhibition, they do not establish that the difference observed in our cohort was treatment-related. Pre-existing differences, underlying disease characteristics, residual confounding, and chance in the context of multiple comparisons remain possible explanations.