A multi-parameter MRI approach, incorporating diffusion tensor imaging (DTI) and T2 mapping, proves more accurate at detecting diabetic peripheral neuropathy than relying on any single parameter alone, according to findings unveiled on December 3 at RSNA. The researchers suggest this method could significantly improve noninvasive assessment of nerve health in people with diabetes.
The study, led by Eunsun Oh, MD, PhD, from Soonchunhyang University Seoul Hospital in South Korea, enrolled 17 diabetic patients with clinically confirmed peripheral neuropathy, 16 diabetic patients without neuropathy, and 12 healthy controls. Each participant underwent knee MRI that included axial T2-weighted Dixon imaging, DTI, and T2 mapping. Measurements focused on tibial nerve properties at the calf level, specifically fractional anisotropy (FA), axial diffusivity (AD), radial diffusivity (RD), mean diffusivity (MD), and T2 relaxation time. Diagnostic performance was evaluated using the area under the receiver operating characteristic curve (AUC).
Key findings included:
- FA values in the tibial nerve were notably reduced in both diabetic groups (with and without neuropathy) relative to healthy controls (p < 0.01).
- T2 relaxation times were significantly prolonged in both diabetic groups compared with controls (p < 0.01).
- Among single-parameter models, FA demonstrated the strongest capability to identify diabetic peripheral neuropathy (AUC = 0.85).
- A multiparametric model that combined FA and T2 relaxation time yielded higher diagnostic accuracy (AUC = 0.91), surpassing all single-parameter models.
- AD, RD, and MD yielded lower AUC values and were less effective as standalone indicators.
OH emphasized that FA reductions appear early in the disease process, signaling microstructural axonal loss even in individuals who are asymptomatic or in the subclinical stage.
The core takeaway is clear: integrating structural (FA) and pathological (T2) markers produces the most accurate diagnosis. This combined approach may enhance the precision of noninvasive nerve assessment in diabetic populations.
For full RSNA 2025 coverage, see ourRADCast at https://www.auntminnie.com/resources/conference/rsna/2025.
