Application of magnetic resonance diffusion tensor imaging for the treatment of hemifacial spasm.

OBJECTIVES

To evaluate diffusion tensor imaging (DTI) for assessing facial nerve root exit zone (REZ) microstructure in hemifacial spasm (HFS) and predicting microvascular decompression (MVD) outcomes.

METHODS

Data from 60 patients with primary unilateral HFS and 30 healthy volunteers were retrospectively examined. High-resolution 3.0T DTI was performed on all participants. DTI data (b=1000 s/mm2, 32 directions) were processed using FMRIB Software Library. The fractional anisotropy (FA), mean diffusivity (MD), axial diffusivity (AD), and radial diffusivity (RD) were derived from the facial nerve REZ. Group comparisons and receiver operating characteristic curve analysis assessed diagnostic performance. Preoperative DTI parameters were also evaluated for predicting MVD outcomes.

RESULTS

In HFS, FA on the symptomatic side (0.35±0.06) was significantly lower than that on the contralateral side (0.48±0.05) and controls (0.49±0.04, P<0.001). MD and RD were significantly elevated (both P<0.001), while AD remained stable. The area under the curve (AUC) for FA in diagnosing HFS was 0.968 (sensitivity 86.00%, specificity 98.00%). Multivariate logistic regression analysis showed that the preoperative FA value was an independent predictor of postoperative outcomes (odds ratio=0.752, P=0.005), and the optimal cutoff value for FA was ≤ 0.378 (AUC=0.780).

CONCLUSIONS

DTI noninvasively quantifies facial nerve REZ microstructural damage in HFS, characterized by reduced FA and elevated RD. As diagnostic and prognostic biomarkers, FA differentiates pathological compression from asymptomatic contact, supporting clinical decisions.