Speakers in Thursday afternoon’s Euretina session on diabetic retinopathy (DR) and diabetic macular oedema (DMO) covered a range of topics that included screening and early detection of DR, novel methods for treatment delivery and applications for imaging techniques.
Beginning the programme, Sumit Randhir Singh, MD (India) discussed artificial intelligence (AI)-based eye care and screening in low-income countries, explaining that such an approach offers a scalable, cost-effective path to handling the growing need for diabetic eye care in these nations. He reviewed systems in use and shared real-life examples to highlight image acquisition methods, automated detection and grading of DR, implementation in resource-limited settings, validation, workflow integration, and challenges relating to cost, infrastructure, and equity.
In summary, Dr. Singh said that AI can help bridge the gap between the growing burden of diabetes mellitus and limited eye-care resources in low-income countries. Successful implementation, however, depends on multiple factors, including use of appropriate technology, local validation, availability of trained personnel, and sustainable health-system financing.
A talk by Francesco Bandello, MD (Italy) also related to screening. Dr. Bandello presented a study showing how autofluorescence-based macular pigment optical density (AF-MPOD) assessment, optical coherence tomography (OCT), and OCT angiography (OCT-A) might be combined to characterise early retinal involvement in patients with diabetes mellitus without overt retinopathy.
The study, which had an observational, cross-sectional design, was based on the assumption that metabolic dysfunction may precede overt morphological damage in the retina and that advanced non-invasive imaging techniques may help reveal subtle metabolic, structural and microvascular alterations in diabetic eyes without retinopathy that are not detectable during routine clinical examinations. The results showed strong correlations between MPOD and both HbA1c and body mass index. In addition, AF-MPOD reduction was associated with both OCT detected progressive neuroretinal thinning and OCT-A detected progressive perfusion impairment. Dr. Bandello suggested that the findings support the need for further prospective studies to investigate the AF-MPOD in patients at risk of DR onset, which could determine its prognostic value and a potential role in clinical screening pathways.
Talks relating to management included a presentation by Andrew Chang, MD, PhD (Australia) who discussed novel drug delivery pathways beyond conventional intravitreal therapy for addressing unmet needs for managing DR and DMO. In his presentation, Dr. Chang examined the rationale and considerations for various alternative drug delivery routes and technologies designed to improve retinal drug exposure, extend treatment durability, and potentially reduce the burden of intravitreal injections. He discussed suprachoroidal injection, topical therapies, and systemic therapies and briefly reviewed clinical trials investigating these novel systems. Concluding that unmet needs in diabetic eye care are driving therapeutic innovation, Dr. Chang said that non-invasive delivery approaches beyond the vitreous may allow for early preventative treatment of nonproliferative DR and that systemic approaches could provide the advantage of bilateral DR treatment.
Matias Iglicki, MD, PhD (Argentina) spoke about the value of using intraoperative OCT for guidance during vitreoretinal surgery. “In complex diabetic retinopathy, the decisive dimension is depth. Intraoperative OCT gives it back,” he said.
Presenting situations where intraoperative OCT was used, he described four moments to implement scanning: before the first cut to find the safest entry point; while dissecting to confirm depth and detect residual adhesion, when encountering anatomy surprises, and before closing to assess for residual traciton, retinal apposition, retained subretinal perfluorocarbon liquid. However, he also identified that iOCT has some limitations, and importantly, although there is evidence that use of iOCT changes surgical decisions, it also must be proven that it changes vision.
Nadia Waheed, MD, MPH (United States) discussed diabetic macular ischaemia (DMI), noting that it is an unexplored frontier in DR but important to understand because it is a new frontier in drug development for DR. She discussed assessment of DMI with different imaging techniques, noting that fluorescein angiography is not ideal while OCT-A appears to be “a match made in heaven” as it could be used to evaluate foveal avascular zone area and morphology, vessel perfusion and density, and quantitate nonperfusion.
Dr. Waheed concluded that defining meaningful and reproducible endpoints for DMI remains a major challenge. “With a multidimensional approach involving OCT, OCT-A, functional testing and possibly fluorescein angiography when needed, however, we may well be on our way,” she said.
In the last talk of the session, Rafael Simó, MD (Spain) presented research that extends and reinforces the idea that neurodegeneration is an early event in DR, occurring before overt microvascular abnormalities can be detected on ophthalmoscopy. The study he discussed asked the key question of whether HbA1c levels that are associated with thinning of neuroretinal layers may be lower than the thresholds currently used for diagnosing prediabetes and diabetes and that have been associated with DR development in terms of microvascular disease. The results supported the hypothesis, showing that progressive retinal thinning of both the inner nuclear layer (INL) and ganglion cell-inner plexiform layer (GC-IPL) were present at HbA1c values lower than the currently accepted threshold for diagnosing prediabetes and diabetes. Furthermore, the breakpoint towards GC-IPL and INL thinning occurred in diabetic patients without overt DR in fundoscopic examination.
Dr. Simó suggested that the findings might help to build a foundation for revisiting the current diagnosis of diabetes based on HbA1c values.
