MR Neurography in Children and Adolescents

Objectives: MR neurography (MRN) is a modern technique for visualizing peripheral nerves and quantifying microstructural pathology, yet its use in pediatric populations remains largely unexplored. This study evaluates the applicability and diagnostic performance of MRN in children and adolescents with genetically confirmed long-chain 3-hydroxyacyl-CoA dehydrogenase deficiency (LCHADD) and mitochondrial trifunctional protein deficiency (MTPD), in which peripheral neuropathy is a known long-term complication. Materials and Methods: In a prospective cross-sectional study, 15 patients (LCHADD n = 6; MTPD n = 9) and 14 age-matched controls underwent high-resolution mid-thigh MRN of the sciatic nerve to assess (1) T2-based lesion burden for tibial (SN Tib ) and peroneal (SN Per ) fascicles, (2) functional nerve integrity of the tibial fascicles using diffusion tensor metrics, including fractional anisotropy (FA) and radial diffusivity (RD), and (3) tibial fascicle-based T2 relaxometry parameters. In addition, clinical and electrophysiological data were obtained. Age-adjusted linear regression, ROC analyses, and linear discriminant analyses (LDA) quantified group effects and classification performance. Results: Overall, patients showed higher T2 lesion burden compared with controls (SN Tib : +2.75%, P = 0.001; SN Per : +1.94%, P = 0.001), reduced tibial fascicle FA (Δ: -0.098, P = 0.001), and increased tibial fascicle RD (Δ: +147.4×10 -6 mm 2 /s, P = 0.011). Subgroup comparisons between LCHADD and MTPD revealed no significant differences. Of the 15 patients, 7 exhibited signs of clinical neuropathy. Neuropathic individuals showed pronounced abnormalities (SN Tib : +4.22%, P Per : +2.29%, P = 0.002; ΔFA: -0.138, P Per lesion burden (+1.64%; P = 0.018) and reduced tibial fascicle FA (Δ: -0.062, P = 0.03), compared with controls, indicating subclinical involvement. SN Tib lesion burden showed excellent discrimination (AUC: 95.2%), and FA performed well (AUC: 81.2%). Multiparametric LDA achieved 93% apparent in-sample accuracy for patients versus controls, 86% for LCHADD versus MTPD, and 90% for classifying neuropathic, non-neuropathic, and control groups. Conclusions: MRN can be readily applied in children and adolescents and sensitively detects both clinically manifest and subclinical peripheral nerve involvement in long-chain fatty acid oxidation disorders. Extending this capability, exploratory LDA suggests that combining multiple MRN metrics may provide complementary diagnostic and phenotypic information beyond individual parameters.