Electrophysiological basis of functional MRI signals in rat models of permanent and transient cerebral ischemia

Background and Aims: Functional MRI (fMRI) is increasingly applied as a tool to study changes in activation patterns after stroke. However, the electrophysiological basis of fMRI-detected changes in activation responses are unresolved. Here, we measured somatosensory-evoked potentials (SEPs) in the forelimb region of the primary somatosensory cortex (S1fl) at different time-points after permanent and transient stroke in rats. Based on previous fMRI findings1,2, we expected loss of SEPs in ipsilesional S1fl and increased contralesional electrical activity when the affected forelimb is stimulated acutely after stroke. Methods: Stroke was induced by permanent or transient (60-90minutes) right middle cerebral artery-occlusion (MCA-O) in rats.3 SEP recordings were performed at 2 and 3h (n=5) and at 24h (n=3) after permanent MCA-O, and at 24h after transient MCA-O (n=2). SEPs were simultaneously recorded epidurally from left and right S1fl with increasing and alternating electrical stimulation (0, 0.5, 1, 2, 4 and 8mA) of left and right forelimb under propofol anesthesia. After SEP recordings, animals were sacrificed for TTC staining to assess size and location of the cerebral ischemic lesion. SEP data were digitally low-pass filtered and analysis was done on the averaged SEP for each S1fl out of two sessions for each rat. We calculated the Rate Dispersion Factor (RDF) for the 5-30ms latency range, which provides an index of the overall shape of the SEP waveform.4 Results: Normal SEPs were observed in contralesional S1fl when the unaffected forelimb was stimulated. At 2 and 3h after permanent MCA-O, although reduced, significant SEPs were measured in right, ipsilesional S1fl when the left, affected forelimb was stimulated. At 24 hpost-stroke, ipsilesional SEPs were largely absent. RDF values at 2mA stimulation were 54.0±15.8%, 49.6±16.4% and 8.9±4.3% (percentage of SEP in left, contralesional S1fl during right, unaffected forelimb stimulation; mean±SEM), at 2, 3 and 24h after permanent MCA-O, respectively. At 24h after transient MCA-O, RDF in ipsilesional S1fl was 4.9±3.2%. Electric activity in contralesional S1fl was also detected when the affected forelimb was stimulated. At 2 and 3h post-stroke, contralesional (left) RDF values at 2mA stimulation of the affected (left) forelimb were 15.6±6.9% and 11.5±3.5%, respectively. These RDF values were diminished at 24h post-stroke (permanent stroke: 5.5±1.8%, transient stroke: 5.9±2.5%). Conclusions: Our study demonstrates that electrical activity in ipsilesional S1fl is partially preserved at 2-3h after permanent MCA-O, but has largely disappeared at 24h after transient or permanent MCA-O. The latter is in agreement with lack of fMRI activity in rat S1fl during the first days after stroke.1,2 Furthermore, contralesional activation responses upon stimulation of the affected forelimb, as observed in this electrophysiology study, have also been described in fMRI studies in rat stroke models.1,2 Thus, stroke-induced changes in fMRI-detected brain activation, which are based on measurement of hemodynamic responses, correlate with changes in electrophysiological activity.