New tools are changing how we detect and treat diabetic nerve damage

Diabetic peripheral neuropathy (DPN) affects nearly one-third of people with diabetes and is a leading cause of foot ulcers and amputation — yet up to 90% of cases go undiagnosed in the U.S. Researchers at the EASD 2026 Annual Meeting unveiled three promising advances that could change that.

High-frequency (10 kHz) spinal cord stimulation (SCS) combined with conventional medical management delivered dramatic pain relief — about 80% of SCS patients hit the primary pain response endpoint at 6 months vs. just 4% on standard care alone. Benefits held strong at 12 months, and 92% of the control group crossed over to SCS after seeing the results. Separately, a deep learning model trained on routine retinal fundus images identified symptomatic DPN with 90% sensitivity and 83% overall accuracy — a potential "zero-burden add-on" to existing diabetic eye screening programs. A third study introduced a six-variable risk model that predicted 4-year incident neuropathy risk with moderate accuracy, performing well across both type 1 and type 2 diabetes patients and multiple ethnicities.

By the Numbers:

• ~80% of SCS patients achieved ≥50% pain relief at 6 months vs. 4% with standard care
• Sensory improvement: 55% (SCS) vs. 25% (standard care)
• Retinal AI model: AUC 0.85, sensitivity 0.90, overall accuracy 83%
• 4-year DSPN risk model: AUC 0.725, predicted risk 27.5% vs. observed 27.9%

Why it matters: These tools offer clinicians earlier detection and more effective treatment options for one of diabetes's most debilitating complications — potentially reducing amputations, improving quality of life, and integrating seamlessly into existing care workflows.