Figure 1.
Representative DWI (A and C) and corresponding permeability maps (B and D) of coronal sections of rat brains.
Images shown are from two different groups: A and B from group 1 (scanned at 3 h, n = 8) and C and D from group 2 (studied at 48 h, n = 8). DWI images were used to calculate ADC maps and generate tissue signature maps shown in Fig. 5A and 5B. Areas of the subcortex and cerebral cortex are demarcated in A and C.
Figure 2.
Cumulative spatiotemporal distribution of BBB permeability for cerebral cortex and subcortex (Panels A and B).
Statistical analysis of raw data presented in A and B was performed by calculating the area of leakage for each brain region with Ki values higher than 0.001 ml/g-min. As shown in panel C and D, BBB permeability changes at 48 h were significantly higher than at 3 h. The vertical axis shows the frequency of pixels with different Ki values (shown in the horizontal axis).
Figure 3.
A. Changes in the area with hypointensive ADC at 3 and 48 h of reperfusion. Significant increases in areas with ADC abnormalities were found between 3 and 48 h in the cerebral cortex. Brain damage observed at 48 h in ADC maps (B) was confirmed histologically using TTC staining (C). Areas of the subcortex and cerebral cortex are demarcated in panels B and C.
Figure 4.
Scatterplots and fit lines showing the correlations between ADC and Ki for the ipsilateral side of cerebral cortex (A and B) and subcortex (C and D) at 3 and 48 h of reperfusion following 2 h of MCAO in the rat.
Figure 5.
Analysis of correlations between BBB permeability and the ADC values in cerebral cortex and subcortex.
Color-coded Ki-ADC maps (A and B) show the three main populations of pixels with different Ki and ADC values for both time points. Quantitative changes of each area over time are presented in panels C and D. Based on the abnormality of ADC values and the BBB permeability; we have recognized three different areas in the ipsilateral side. These areas are labeled as tissue signatures corresponding to different pathophysiology in the evolution of lesion. Each tissue signature is represented by a different color at two different time points, 3 and 48 h of reperfusion. *p<0.05 and **p<0.01 with respect to 3 h.
Figure 6.
Proposed model showing possible transitions between different states of tissue injury during the progression of ischemic stroke.
Cytotoxic edema is characterized by low ADC values with preserved BBB function (normal Ki), while vasogenic edema is identified by high Ki values. The core of the infarct display high Ki and low ADC values (a mixture of cytotoxic and vasogenic edema). Based on data from the merged ADC+Ki maps (Fig. 5), the main transition occurs from areas with cytotoxic edema to areas with both cytotoxic and vasogenic edema (core of the infarct), which is depicted with a thicker arrow in the schematic.