Assessment of white matter microstructure integrity in subacute postconcussive vestibular dysfunction using NODDI

Abstract

Abstract Vestibular symptoms, such as dizziness and balance impairment, are frequently reported following mild traumatic brain injury (mTBI) and are associated with a protracted recovery, yet the underlying neuroanatomical substrates remain unclear. The present study utilized advanced diffusion MRI (dMRI) techniques including both conventional diffusion tensor imaging (DTI) and neurite orientation dispersion and density imaging (NODDI) to investigate microstructural white matter integrity in individuals with postconcussive vestibular dysfunction (PCVD) within the subacute injury period (median of 35 days from injury; IQR of 23). Study participants included 23 individuals with subacute PCVD and 37 healthy control subjects who underwent imaging and comprehensive clinical vestibular testing. Between-group voxelwise analysis of differences in white matter revealed areas of higher intra-neurite volume fraction (VIn) and isotropic volume fraction (VIso) within PCVD subjects compared to controls, which involved overlapping regions within the left hemisphere of the brain. Affected areas of higher VIn and VIso included the superior longitudinal fasciculus (SLF) and superior and posterior corona radiata (SCR and PCR, respectively). We examined the relationship between clinical vestibular measures and diffusion metrics including DTI (fractional anisotropy [FA], mean diffusivity [MD], radial diffusivity [RD] and axial diffusivity [AD]) and NODDI (intraneurite volume fraction [VIn], isotropic volume fraction [VIso], dispersion anisotropy [DA], orientation dispersion indexTotal/Primary/Secondary [ODIT/P/S]) within 32 regions-of-interest. Clinical vestibular measures included self-reported measures, including the Dizziness Handicap Inventory, Visual Vertigo Analog Scale, and Vestibular/Ocular-Motor Screening, as well as objective vestibular testing using the sensory organization test. Significant correlations were found with clinical measures across all diffusion maps (except DA), within various regions of interest (ROIs), including SLF, SCR, and PCR. These results implicate several important association bundles that may potentiate sensory processing dysfunction related to PCVD. Whether these neuroanatomical differences found within the subacute phase of PCVD are in response to injury or represent preexisting structural variations that increase vulnerability to sensory processing dysfunction is unclear and remains an active area of study.