Characterization and quantification of alcohol-related polyneuropathy by magnetic resonance neurography

Background We characterized and quantified peripheral nerve damage in alcohol-dependent patients (ADP) by magnetic resonance neurography (MRN) in correlation with clinical and electrophysiologic findings. Methods Thirty-one adult patients with a history of excessive alcohol consumption and age-/sex-matched healthy controls were prospectively examined. After detailed neurologic and electrophysiologic testing, the patient group was subdivided into ADP with alcohol-related polyneuropathy (ALN) and without ALN (Non-ALN). 3T MRN with anatomical coverage from the proximal thigh down to the tibiotalar joint was performed using dual-echo 2-dimensional relaxometry sequences with spectral fat saturation. Detailed quantification of nerve injury by morphometric (cross-sectional area [CSA]) and microstructural MRN markers (proton spin density [rho], apparent T2-relaxation-time [T2(app)]) was conducted in all study participants. Results MRN detected nerve damage in ADP with and without ALN. A proximal-to-distal gradient was identified for nerve T2-weighted (T2w)-signal and T2(app) in ADP, indicating a proximal predominance of nerve lesions. While all MRN markers differentiated significantly between ADP and controls, microstructural markers were able to additionally differentiate between subgroups: tibial nerve rho at thigh level was increased in ALN (p < 0.0001) and in Non-ALN (p = 0.0052) versus controls, and T2(app) was higher in ALN versus controls (p < 0.0001) and also in ALN versus Non-ALN (p = 0.0214). T2w-signal and CSA were only higher in ALN versus controls. Conclusions MRN detects and quantifies peripheral nerve damage in ADP in vivo even in the absence of clinically overt ALN. Microstructural markers (T2(app), rho) are most suitable for differentiating between ADP with and without manifest ALN, and may help to elucidate the underlying pathomechanism in ALN.