Abstracts

Penetrating head injury leads to the development of epilepsy in [sim]50% of patients. Non-invasive detection of surrogate markers that predict development of post-traumatic epilepsy would be of value for predicting prognosis of traumatic brain injury (TBI) and for development of antiepileptogenic treatments. We applied quantitative MRI to follow progression of tissue alterations in a rat model of TBI-induced epileptogenesis. At the end of the study, MRI will be correlated with the development of epilepsy. Here we show an interim analysis of the study that will be completed in June. TBI was induced in Spraque Dawley rats (n=14) by fluid percussion [1]. MRI data were acquired 3h, 3d, 9d, 23d, 2mo, 3mo, and 6mo after TBI at 4.7T. Volumetric changes were detected using T[sub]2[/sub]-wt SE imaging (TE=70ms, TR=3s, 128*256pts, FOV 3*3cm², thk=0.75mm, 19 slices). T[sub]2[/sub], T[sub]1[rho][/sub], and trace of the diffusion tensor (D[sub]av[/sub]) were quantified from a single slice using a fast SE sequence (TR=3.0s, 128*256pts, FOV 3*3cm², thk=1.5mm; T[sub]2[/sub]: TE=20-80ms; T[sub]1[/sub][sub][rho][/sub]: spin-lock times=18-78ms, B[sub]1[/sub]=0.8G; D[sub]av[/sub]: b-val=90-1014s/mm[sup2]). T[sub]2[/sub]^* was measured using a GE sequence (TE=5-15ms, TR 0.7s). At 6mo rats will undergo 2wk of video-EEG monitoring for seizure detection followed by perfusion for histology. T[sub]2[/sub]* images showed signal void areas due to hemorrhage in 11 animals. Three animals developed severe damage at 2mo, characterized by a large cortical lesion ([gt]40mm³) and elevated relaxation times and D[sub]av[/sub] in the ipsilateral cortex (T[sub]2[/sub][gt]300ms, T[sub]1[rho][/sub][gt]400ms and D[sub]av[/sub][gt]2.0*10^-3mm²/s vs. T[sub]2[/sub]=56.4[plusmn]0.3ms, T[sub]1[rho][/sub]=82.0[plusmn]1.2ms and D[sub]av[/sub] =0.70[plusmn]0.01 *10^-3 mm²/s in controls). Four animals had no MRI changes at 2mo, in spite of initial increases in relaxation times (by 6-10%) and decreased D[sub]av[/sub] ([sim]4%). The rest of the animals (n=7) displayed decreased D[sub]av[/sub] in the cortical lesion in the initial phase (typical for cytotoxic oedema) followed by an increase in D[sub]av[/sub] after day 9. T[sub]2[/sub] was initially increased by [sim]26% (p[lt]0.01) at day 3, but this declined to an [sim]8% (p[lt]0.01) increase at 2mo. Interestingly, T[sub]1[rho][/sub] showed a similar initial increase but no recovery. In the ipsilateral hippocampus, relaxation times were slightly increased (3-7%, p[lt]0.01) 3d after induction of TBI after which they recovered. Interestingly, D[sub]av[/sub] showed an increase of [sim]5% (p[lt]0.01) at 2mo probably due to delayed tissue damage in the hippocampus. Progression of the damage was also evident from increased ipsilateral ventricle and hippocampal atrophy. Our interim analysis indicates that a subpopulation of animals develops cortical and/or hippocampal damage that continues to progress during epileptogenic phase, several weeks after TBI. Association of these alterations with epileptogenesis will indicate which of the markers best associate with tissue pathology and predict epileptogenesis.
[1] Kharatisvili et al. [italic]Epilepsia[/italic], 44, suppl. 8, 2003 (Supported by Academy of Finland, Sigrid Juselius Found.)