Validation of T2* Weight Signal Change of Oxygen Challenge as a Potential better Penumbra Estimation

INTRODUCTION Mismatch of diffusion/perfusion by MRI has been used as an estimate of the ischemic penumbra, but there are large parts of the mismatch region appear not to at risk, even though they may contribute to functional impairment (1). It was also reported that some of the apparent diffusion coefficient (ADC) reduction area can be salvaged by early reperfusion. It is essential to find a method to more acutely distinct salvageable tissue from irreversibly damaged tissue. In a permanent middle cerebral artery occlusion (MCAO) model, we demonstrated that T2* weighted (T2*W) signal change induced by 100% oxygen challenge (OC) showed the potential to reflect the metabolic status of tissues (2, 3). It was found that the diffusion/perfusion mismatch region showed higher than normal T2*W signal increase during OC and some ADC reduction area showed lower than normal but significantly higher than ischemic core positive response. We hypothesized that those tissues are more amendable to treatment and T2*W MRI of OC has the potential to better approximate penumbra. To validate this hypothesis, T2*W MRI associated with OC was used to study a group of transient (45-min) MCAO rats. Standard perfusion and diffusion MRI was also performed to identify perfusion-diffusion mismatch. Final lesion determined by T2 weighted imaging was correlated with OC responses. METHODS Male Sprague Dawley rats (250-300g, n=5) were subjected to 45-min MCAO (4). Rats breathed spontaneously under ~1.2% isoflurane in air. Body temperature and respiration rate were continuously monitored and maintained within normal ranges. OC experimental paradigm was: 1 min OFF, 2 mins ON, 5 mins OFF, 2 mins ON and 1 min OFF, 720 repetitions in total. OC response percent change maps were calculated. MRI was performed on a Bruker 7T/30cm scanner. A surface coil (2.3-cm ID) with active decoupling was used for brain imaging and a neck coil for perfusion labeling. CBF (cerebral blood flow) was measured using cASL gradient-echo EPI. ADC (apparent diffusion coefficient) was measured using spin-echo EPI. MRI parameters were: single shot, matrix = 96x96, FOV = 25.6 x 25.6mm, seven 1.5mm thick slices, TR=3s, TE=10 ms for CBF and 30ms for ADC, 90 flip angel. OC T2*W MRI was acquired before and after reperfusion using similar parameters as CBF measure except TR=1s, TE=26ms, 60 flip angle. T2 weighted imaging was acquired using RARE sequence at 24hrs post-occlusion to determine the final lesion. Four ROIs were analyzed: LH (left hemisphere), MM (perfusion-diffusion mismatch), IC (ischemic core) and BZ (border zone of abnormal ADC). ADC and CBF abnormal thresholds of 0.53x10 mm/s and 0.3 ml/gram/min (5), respectively, were used to define normal, mismatch and core tissue types.