Abstract
Paroxysmal sympathetic hyperactivity (PSH) is a clinically important manifestation of dysautonomia following traumatic brain injury (TBI). While it is thought to arise from central autonomic network disconnection, supporting evidence is limited. Here, we integrate clinically obtained magnetic resonance imaging (MRI) lesion data with human connectome data to identify specific white matter tract disconnections and gray matter parcel damage associated with PSH. Our sample included 117 patients who underwent susceptibility weighted imaging and 3D T1 MRI sequences as part of clinical care while admitted at our institution between January 1, 2016 and July 1, 2018. Susceptibility lesion masks were manually created and registered to standard template space. High quality registrations were obtained in 96 patients (50% with PSH), who were included in the study. Using the Matlab Lesion Quantification Toolkit, we assessed white matter tract disconnection severity and gray matter parcel damage for each patient. We compared results according to a binary PSH clinical diagnosis using Wilcoxon rank sum tests and a standard ordinal PSH diagnostic likelihood score (with 0-11 range) using Pearson correlations, Bonferroni-corrected for multiple comparisons. PSH diagnosis was associated with greater disconnection severity in nine tracts, two of which also correlated with higher diagnosis likelihood: the right uncinate fasciculus and the anterior corpus callosum. Damaged parcels associated with PSH included left prefrontal regions of the default mode network and the ventral salience network. In summary, our work implicates disconnection of fronto-limbic components of the central autonomic network in the pathophysiology of TBI-related PSH.