Fig 1.
Magnetic resonance imaging (MRI) scans depicting ROIs and longitudinal scans of all groups across time points.
(Top) Regions of interest (ROIs) were defined using the T2-weighted datasets of the ipsilateral (right) and contralateral (left) side of the brain for the following areas: cerebral cortex (outlined in red), hippocampus (yellow), external capsule (green) and corpus callosum (blue). (Bottom) T2-weighted images from a representative sham, moderate TBI, and severe TBI animal imaged longitudinally at 7 different time points. Progressive cerebrostructural changes were observed in several different areas across time points corresponding to the temporal evolution of the injury as indicated by hyperintense signal (white region) in the ipsilateral side. Moderately and severely injured animals developed enlarged lateral ventricles in the injured hemisphere where the cortical lesion was present relative to the contralateral non-injured hemisphere. Chronic images of severe TBI animals also revealed thinning of cortical tissue and shrinkage of ipsilateral hippocampus.
Fig 2.
Apparent diffusion coefficient (ADC) was significantly altered in all ipsilateral ROIs.
Within the cortex, significance was observed at a 4-month time point between sham and moderate animals. ADC within the ipsilateral hippocampus was significant at all post-injury time points between sham and injured groups (see S1 Table for values and groups exhibiting significance). ADC in the corpus callosum was primarily statistically significant between sham and severe groups; however, at 8 months, significant differences were also observed between moderate and severe animals. The external capsule exhibited statistically different ADC at all, but a 4-month time point between both shams versus moderate and sham versus severe groups. *p ≤ 0.05; **p ≤ 0.01.
Fig 3.
Fractional anisotropy (FA) values were significantly reduced in the ipsilateral hippocampus, corpus callosum, and external capsule.
FA in the cortex initially dropped and then increased by 1 month, however, these changes were not significantly different from sham. FA values in the hippocampus were aberrantly reduced after severe injury at chronic 8- and 12-month time points. Chronic FA differences were also evident in the corpus callosum, but only between sham and moderate groups. In the external capsule, FA was reduced at several acute and chronic time points between sham and injury groups and between moderate and severe groups at 8 months post-TBI. *p ≤ 0.05; **p ≤ 0.01; ***p≤ 0.001; ****p≤ 0.0001.
Fig 4.
Radial diffusivity (RD) values were significantly elevated in the ipsilateral hippocampus, corpus callosum, and external capsule.
Similar to what was observed in FA, RD values recorded in ipsilateral ROIs exhibited statistically significant increases compared to control groups except for the cortex. Within the hippocampus, RD was pathologically elevated at every time point between sham and injury groups except for at an acute phase 3 days post TBI. In the corpus callosum, there was statistically abnormal heightened RD in the majority of time points between sham and injured animals, while imaging at 8 months revealed a significant increase between moderate and severe RD values. The external capsule had elevated RD from 1 month until the 12-month endpoint between sham and TBI groups. *p ≤ 0.05; **p ≤ 0.01; ***p ≤ 0.001; ****p≤ 0.0001.
Fig 5.
After moderate and severe TBI, PWI revealed CBF was significantly decreased in the ipsilateral cortex and hippocampus at several time points relative to sham animals.
In the cortex, depressed CBF was observed at 4 and 6 months between sham versus moderate and sham versus severe groups, while the hippocampus showed regional CBF decreases at an acute 3-day time point, and chronically from 4 months out until 12 months post moderate and severe TBI relative to sham uninjured controls. *p ≤ 0.05; **p ≤ 0.01.
Fig 6.
Significant histopathology was observed 12 months after moderate and severe TBI.
(Top) Volumetric analyses revealed that both moderate and severely injured animals had significant atrophy in all ipsilateral ROIs compared to sham uninjured control animals at a 12-month chronic time point. There were no significant differences in atrophy between moderate and severe injuries. *p ≤ 0.05; **p ≤ 0.01; ***p ≤ 0.001; **** P ≤ 0.000. (Bottom) Double-stained hematoxylin-eosin and Luxol fast blue sections 1 year after moderate TBI or sham procedure. (A) Representative sham-operated animal appeared unremarkable. (B) Representative moderate TBI animal showed gross atrophy with a marked expansion of the ipsilateral lateral ventricle. (C) Higher magnification of external capsule in sham animal appeared normal. (D) Higher magnification of external capsule thinning after moderate TBI (arrows). (E) Representative sham animal showed normally stained white matter fibers. (F) Representative moderate TBI animal showed loss of white matter staining in the corpus callosum, indicative of demyelination (*). (G) Representative sham animal had an unremarkable cerebral peduncle. (H) In contrast, moderate TBI animal demonstrated atrophic changes within the ipsilateral cerebral peduncle (arrows).
Fig 7.
DTI and PWI modalities as prognostic indicators after moderate and severe TBI.
(Top row) At one month after severe TBI, there was a significant correlation between ADC and PWI in the ipsilateral cortex. Lower, more normalized ADC values correlated to greater CBF perfusion. Six months after severe injury, reduced CBF perfusion in the ipsilateral cortex significantly correlated with worsened degree of chronic cortical atrophy at 12 months. (Middle row) Correlation analyses showed that at 6 and 8 months after severe TBI, there was a positive correlation between FA values and PWI CBF. The larger the FA value, more closely resembling sham uninjured levels, the greater the degree of CBF perfusion in the ipsilateral hippocampus. (Bottom row) At an acute 3-day time point, ADC and RD were noticeably reduced in the ipsilateral corpus callosum of moderate TBI animals. We observed a significant positive correlation in which the greater the value of ADC and RD, the worse the extent of atrophic histopathology after moderate TBI. These findings illustrated that the degree of DTI reduction at 3 days after moderate injury was significantly correlative of 12-month cytoprotection. Statistical details for correlation analyses can be found in the Materials and Methods section.