Functional CT Perfusion Imaging in Predicting the Extent of Cerebral Infarction from a 3-Hour Middle Cerebral Arterial Occlusion in a Primate Stroke Model
Leena M. Hamberga,b,
George J. Huntera,b,
Kenneth I. Maynardc,
Chris Owenc,
Pearse P. Morrisb,
Christopher M. Putmanb,
Christopher Ogilvyc and
R. Gilberto Gonzálezb
a MGH Perfusion and Physiology Analysis Laboratory, the Massachusetts General Hospital and Harvard Medical School, Boston
b Department of Neuroradiology, the Massachusetts General Hospital and Harvard Medical School, Boston
c Department of Neurosurgery, the Massachusetts General Hospital and Harvard Medical School, Boston
View larger version (36K):
[in a new window]
|
FIG 1. Acutely hypoperfused lesion sizes, as measured with perfusion CT, for animals 1, 3, 4, and 5 at 30- and 150-minute time points after the start of endovascular occlusion. At each time point, the leftmost column corresponds to lesion size (in square millimeters) measured on the CBF map; middle column, lesion size measured on the CBV map; and rightmost column, lesion size measured on the MTT map. The dashed line represents outcome infarct size, as determined on the ex vivo T2-weighted MR image. Data for animal 5 were not available at 150 minutes after occlusion because of technical factors.
| |
View larger version (20K):
[in a new window]
|
FIG 2. Regression statistics between the CT-determined CBF-, CBV-, and MTT-based lesion sizes and the 48-hour outcome lesion sizes, as measured at ex vivo MR imaging. Each point represents a single measurement by an observer. Scatter represents inter- and intra-observer variations. These variations were not statistically significant, as determined by using ANOVA. Data from animal 2 were not included in this particular analysis because the occluding balloon leaked, with resultant early reperfusion and subsequent reduction in the size of the infarcted region (Fig 6).
| |
View larger version (82K):
[in a new window]
|
FIG 3. Pre- and postocclusion maps for animals 3 and 5. The first row and third row from the top present the CBV, MTT, and CBF maps from the control preocclusion study in animals 3 and 5, respectively. The second row and fourth row from the top present the maps obtained 30 minutes after the onset of MCA occlusion. In animal 3, right MCA territorial hypoperfusion is present, with involvement of the putamen and subtle involvement of the anterolateral thalamus. In animal 5, left MCA territorial hypoperfusion is clearly visible after occlusion, but the basal ganglia are spared. Note that each image is individually windowed to facilitate visualization of the lesions.
| |
View larger version (113K):
[in a new window]
|
FIG 4. DSA images obtained immediately after MCA occlusion (left) and immediately after reperfusion (right) in animal 5. During occlusion, the left MCA vessels are absent (arrows), but they are clearly seen after reperfusion.
| |
View larger version (98K):
[in a new window]
|
FIG 5. Images in animal 4. Top row, CBV, MTT, and CBF maps from the control preocclusion study. Middle row, Results from the study obtained 30 minutes after the start of occlusion. The right MCA territorial hypoperfusion is clearly visible after occlusion. Bottom row, Examples of ROIs representing the lesion on the maps. The lesion is larger on the MTT map than on the CBV and CBF maps. This is likely due to the presence of collateral circulation that results in normal perfusion on the CBV and CBF maps but prolonged transit time. Each image is individually windowed to facilitate presentation of the lesion.
| |
View larger version (28K):
[in a new window]
|
FIG 6. In animal 2, which did not have any neurologic deficit, the balloon leaked between the 30- and 150-minute postocclusion study points. This early reperfusion reduced the outcome size of the lesion. This is an example of a desirable situation in which partial reperfusion of an ischemic territory occurs sufficiently early to substantially reduce the outcome infarct size.
| |
