A study of diffusion tensor imaging of median nerve in diabetic peripheral neuropathy

Abstract

Objective: To evaluate the role of diffusion tensor imaging (DTI) in the evaluation of diabetic peripheral neuropathy (DPN) compared to clinical scores and nerve conduction studies (NCS). Patients and methods: We included 30 patients with diabetes mellitus complaining of neuropathy symptoms and 15 healthy volunteers. All subjects underwent evaluation using 1.5-T DTI of median nerves and NCS. Patients underwent clinical evaluation using the Neuropathy Deficit Score (NDS), Neuropathy Impairment Score in the Lower Limbs (NIS-LL), and Diabetic Neuropathy Examination (DNE) score. Results: We found statistically significant differences between healthy volunteers and patients in fractional anisotropy (FA) of the distal segment (P = 0.016) and whole median nerve (P = 0.024), apparent diffusion coefficient (ADC) of proximal (P = 0.027) and distal (P < 0.001) segments, and whole median nerve (P = 0.019). Distal segment FA was significantly correlated with NDS (P = 0.003), DNEs (P = 0.003), sensory amplitude (P = 0.048), and motor CV (P = 0.020). Distal segment ADC was significantly correlated with NDS (P = 0.007), NIS-LL (P = 0.003), DNEs (P = 0.01), and sensory amplitude (P = 0.032). The best cut-off value of distal segment for FA was 0.45 (87% sensitivity, 80% specificity) and of distal segment ADC was 1.217 (80% sensitivity and specificity). Conclusions: Our results suggest that 1.5-T DTI examination of the median nerve can provide useful non-invasive information in patients with DPN. Trial registration: ClinicalTrials.gov, NCT03934970. Registered on May 1, 2019.