Noninvasive MRI biomarkers in diabetic neuropathy: correlation with electrophysiology, clinical scores, and nerve pathology.

backgroundDiabetic peripheral neuropathy (DPN) is a common and disabling complication of diabetes, and reliable noninvasive biomarkers for its detection and monitoring remain limited. Magnetic resonance imaging (MRI) has emerged as a promising tool for evaluating peripheral nerve structural and microstructural alterations.

methodsPubMed, EMBASE, and Web of Science were systematically searched for studies assessing MRI findings in patients with DPN. Studies comparing diabetic patients with and without neuropathy to healthy controls were included. Data extraction was performed using a structured framework, and quantitative synthesis was conducted using random-effects models. Outcomes included diffusion tensor imaging metrics such as fractional anisotropy (FA) and apparent diffusion coefficient (ADC), as well as associations with clinical severity scores and electrophysiological parameters.

resultsA total of 25 studies comprising 1,048 patients with diabetes and 462 healthy controls were included. Meta-analysis demonstrated significantly reduced FA in patients with DPN compared with controls in the tibial nerve (MD −0.10, 95% CI −0.12 to −0.08), sciatic nerve (MD −0.12, 95% CI −0.19 to −0.06), and common peroneal nerve (MD −0.09, 95% CI −0.11 to −0.07). ADC values were significantly higher in DPN for the tibial nerve (MD 0.20, 95% CI 0.15–0.24) and common peroneal nerve (MD 0.18, 95% CI 0.13–0.22). Across studies, diffusion metrics, T2 mapping, and MR neurography findings were frequently associated with clinical severity scores and nerve conduction parameters. However, included studies were generally small, heterogeneous, and predominantly cross-sectional.

conclusionMRI techniques can detect structural and microstructural alterations in peripheral nerves in DPN and show consistent associations with clinical and electrophysiological measures. However, the current evidence base is limited by heterogeneity and study design. Larger, longitudinal studies with standardized imaging protocols are required to establish the clinical utility of MRI for diagnosis and disease monitoring in DPN. Diabetic peripheral neuropathy (DPN) is a common and disabling complication of diabetes, and reliable noninvasive biomarkers for detecting and monitoring disease remain limited. Magnetic resonance imaging (MRI) has been increasingly investigated as a tool for assessing peripheral nerve structure and microstructure. This systematic review and meta-analysis evaluated MRI findings in patients with diabetic neuropathy and their associations with clinical, electrophysiological, and pathological measures. A total of 25 studies including 1,048 patients with diabetes and 462 healthy controls were analyzed. Meta-analysis demonstrated significantly reduced fractional anisotropy (FA) in patients with DPN compared with controls in the tibial nerve (MD − 0.10, 95% CI − 0.12 to − 0.08), sciatic nerve (MD − 0.12, 95% CI − 0.19 to −igher in DPN for the tibial nerve (MD 0.20, 95% CI 0.15–0.24) and common peroneal nerve (MD 0.18, 95% CI 0.13–0.22). Across included studies, diffusion metrics, T2 mapping, and MR neurography findings were frequently associated with clinical severity scores and nerve conduction parameters. However, most studies were small, heterogeneous, and predominantly cross-sectional. Current evidence therefore suggests that MRI techniques can detect structural and microstructural differences in peripheral nerves in DPN, but further large, longitudinal studies with standardized imaging protocols are required to clarify their potential role in clinical assessment and disease monitoring. 0.06), and common peroneal nerve (MD − 0.09, 95% CI − 0.11 to − 0.07). Apparent diffusion coefficient (ADC) values were significantly h