Related Experiment Video
Updated: Jan 8, 2026

Diffusion Tensor Magnetic Resonance Imaging in Chronic Spinal Cord Compression
Published on: May 7, 2019
Diagnostic Utility of Fractional Anisotropy and the Apparent Diffusion Coefficient in the Median Nerve: A Nerve
Tsutomu Inaoka1, Rui Iwata1, Minori Kasai1
1Department of Radiology, Toho University Sakura Medical Center, Sakura, JPN.
Objectives:
This study aimed to evaluate the clinical utility of fractional anisotropy (FA) and the apparent diffusion coefficient (ADC) values obtained by diffusion tensor imaging (DTI) of the median nerve at the wrist, in human subjects stratified by nerve conduction velocity (NCV) findings.
Materials And Methods:
Wrist DTI was performed on a 3T scanner using a dedicated flex coil. A total of 85 wrists from 43 NCV-normal subjects and 17 wrists from 10 NCV-abnormal subjects were included. The normal subjects were asymptomatic, with no upper-extremity pathology. The abnormal subjects had suspected or subclinical median nerve impairment confirmed by abnormal NCV findings. FA and ADC values were extracted, and the statistical analyses included the correlation with age and sex (in NCV-normal subjects), group comparisons, and a receiver operating characteristic (ROC) analysis to determine the diagnostic utility of FA and the ADC.
Results:
In the NCV-normal subjects, the mean FA was 0.535 ± 0.072, and the mean ADC was 0.756 ± 0.126 × 10-3 mm2/s. FA showed a mild negative correlation with age (r = -0.35, p=0.001), and the ADC showed a mild positive correlation with age (r = 0.29, p=0.008). No significant effects of age group, sex, or their interaction were observed for either metric (p > 0.05). Compared to the NCV-normal group, the abnormal group exhibited significantly lower FA (p = 0.027) and significantly higher ADC (p < 0.001) values. The ROC analysis demonstrated higher diagnostic performance for the ADC (area under the curve (AUC): 0.774; Youden index: 0.443) (p < 0.001) than for FA (AUC: 0.640; Youden index: 0.282) (p = 0.065). Combining both FA and ADC metrics and using logistic regression did not improve the performance over ADC alone (AUC: 0.775; Youden index: 0.443).
Conclusion:
FA and ADC values both reflected median nerve impairment confirmed by the NCV. The ADC showed a trend toward superior diagnostic performance in distinguishing NCV-abnormal cases compared to the FA. These findings suggest the potential value of ADC as a clinically relevant imaging biomarker for peripheral nerve evaluations.

