Diffusion tensor imaging can be used to detect lesions in peripheral nerves in patients with chronic inflammatory demyelinating polyneuropathy treated with subcutaneous immunoglobulin

Abstract

Introduction: Magnetic resonance imaging (MRI) with diffusion tensor imaging (DTI) has shown that fractional anisotropy (FA) is lower in peripheral nerves in chronic inflammatory demyelinating polyneuropathy (CIDP). We examined whether DTI correlates to muscle strength or impairment.

Methods: MRI of sciatic and tibial nerves was performed on 3-T MR scanner by obtaining T2- and DTI-weighted sequences with fat saturation. On each slice of T2-weighted (T2w) and DTI, the tibial and sciatic nerves were segmented and served for calculation of signal intensity. On DTI images, pixel-by-pixel calculation of FA and apparent diffusion coefficient (ADC) was done. Muscle strength at knee and ankle was determined by isokinetic dynamometry and severity of CIDP by neuropathy impairment score (NIS).

Results: Fourteen CIDP patients treated with subcutaneous immunoglobulin were compared to gender- and age-matched controls. T2w values expressed as a nerve/muscle ratio (nT2w) were unchanged in CIDP versus controls 0.93 ± 0.21 versus 1.02 ± 0.21 (P = 0.10). FA values were lower in CIDP compared to controls 0.38 ± 0.07 versus 0.45 ± 0.05 (P < 0.0001), and ADC values were higher in CIDP versus controls 1735 ± 232 versus 1593 ± 116 × 10(-6) mm(2)/s (P = 0.005). In CIDP, FA values correlated to clinical impairment (NIS) (r = -0.57, P = 0.03), but not to muscle strength. FA value in the sciatic nerve distinguishes CIDP from controls with a sensitivity and a specificity of 92.9 %.

Conclusion: CIDP patients have unchanged nT2w values, lower FA values, and higher ADC values of sciatic and tibial nerves compared to controls. FA values correlated to NIS but were unrelated to muscle strength. DTI of sciatic nerves seems promising to differentiate CIDP from controls.

Keywords: CIDP; Diffusion tensor imaging; MRI; Muscle strength; Subcutaneous immunoglobulin.