Number of Intravitreal Injections Predicts Neovascularization, Nonperfusion on Ultra-widefield FA

The above image from the study shows an example of ultra-widefield FA overlayed with segmentations of areas of nonperfusion (blue), neovascularization (yellow) and the foveal avascular zone (green). While intravitreal treatment was significantly associated with reduced neovascularization and nonperfusion on ultra-widefield fluorescein angiography, neither steroid nor anti-VEGF injection counts were predictive of changes in the foveal avascular zone. This aligns with prior research showing that FAZ area remains statistically unchanged after treatment. Photo: Reddy K, et al. Adv Ophthalmol Pract Res. October 25, 2025. Click image to enlarge. Current treatments for diabetic retinopathy (DR), such as anti-VEGF and corticosteroid injections, exhibit inconsistent effects on the disease's pathological features, particularly neovascularization (NV) and nonperfusion (NP), according to existing studies. Since ultra-widefield fluorescein angiography (UWF-FA) enhances visualization of peripheral changes in DR, a new study used this modern technology to evaluate the connection between retinal perfusion, NV and foveal avascular zone (FAZ) dimensions. The goal of the research was to assess how these imaging biomarkers correlate with treatment levels and disease advancement in DR. The findings, published last weekend in Advances in Ophthalmology Practice and Research, highlighted a significant association between intravitreal treatment and reductions in NV and NP areas, while also concluding that the number of injections, whether of anti-VEGF or steroids, did not predict changes in FAZ size.A total of 705 eyes from patients diagnosed with type 1 or type 2 diabetes were included in the retrospective cohort study, with clinical data derived from the University of Michigan Kellogg Eye Center between 2009 and 2018. UWF-FA images of all participants were analyzed to assess the extent of NV and NP by analyzing changes in the FAZ, NV and NP areas. The mean age of participants was 59.2 years, and 56.3% were male.The eyes received an average of 5.6 ±7.7 anti-VEGF injections and 0.63 ±1.94 intravitreal steroid injections. The results indicated a significant association between the number of anti-VEGF and steroid injections and the reductions in NV and NP areas. Specifically, for every additional anti-VEGF injection administered, there was a corresponding decrease of 0.24mm² in neovascularization area and a reduction of 2.54mm² in nonperfusion area. The use of steroids was associated with an estimated reduction of 10.79mm² in NP areas and a decrease of 0.91mm² in total NV area. Furthermore, each additional steroid injection was associated with a further reduction of 13.68mm² in NP and 0.68mm² in NV.Of note, neither treatment significantly impacted the FAZ area, although the findings did suggest a link between diabetic macular edema (DME) on presentation and a smaller FAZ. “Similarly,” the researchers wrote in their paper, “sex, DR severity, vitreous hemorrhage development and need for anti-VEGF treatment were also predictive of the area of non-perfusion, while increasing age, DR severity and development of DME were predictive of neovascularization.”Summarizing their research, the authors wrote, “These findings suggest that total retinal NV and NP, as characterized on UWF-FA, may have the potential to serve as biomarkers for assessing intravitreal treatment response.” They remarked, “This has immense potential for long-term disease monitoring in diabetic retinopathy by using a greater degree of the retina to capture responses to current treatment strategies, and can inform future clinical management and research.”Click here for the journal source. Reddy K, Deng C, Purt B, et al. Predicting intravitreal treatment response using ultrawide-field angiographic biomarkers in diabetic retinopathy. Adv Ophthalmol Pract Res. October 25, 2025. [Epub ahead of print].  This article was developed by the editorial staff in conjunction with experts in the field. In the process, AI may have been among the editorial tools used to meet the goals of human editors, who approved all content.