Figure 1.
Quantitative measurement of perfusion rate based on spatiotemporal ICG dynamics.
(A) Contrast NIR fluorescence angiography. (B) Temporal sequence of ICG fluorescence in the hindlimb ischemia model. The number indicates the time after ICG injection. (C) Temporal dynamics in the regions indicated with boxes in D are plotted. Relative fluorescence units (R.F.U.) were normalized to the maximal intensity. (D) Perfusion map reconstructed based on the time-series ICG images shown in B. (E) Histogram of perfusion rates in each limb for the regions indicated by the dotted box in D. The arrows indicate two perfusion rate peaks in the normal limb. (F) Photographic image of a normal hindlimb after skin excision to expose femoral vessels (right). Perfusion map of the limb (left). (H) Histogram of perfusion rates of the total hindlimb is indicated by the blue line, and the red line depicts the perfusion rates of the ROIs marked by the red line on femoral vessels.
Figure 2.
Necrosis predictability of ICG dynamic perfusion imaging.
(A) Average limb perfusion rates of the normal and ischemic hindlimbs at POD 0 (postoperative day 0; Post-Op.). The gray lines indicate the mean of each group. The P value is shown (Student's t-test, n = 24). (B) Correlation between average limb perfusion rates in the ischemic limbs at POD 0 (black boxes and whiskers) and limb necrosis levels at POD 7. Bonferroni post hoc test applied to the significant effect of groups (ANOVA F2,44 = 58.358, P<0.001). *, P<0.001. **, P = 0.054. The y-axis is in log scale. The gray boxes and whiskers show a relationship between the ischemic/normal laser Doppler blood flow ratio at POD 0 and limb necrosis levels at POD 7. (ANOVA F2,16 = 0.465, P = 0.636). (C) Sigmoidal relationship between regional perfusion rates of the ischemic hindlimbs at POD 0 and necrosis probability of the corresponding region at POD 7. Graphs were drawn from data of 70,714 regions from 20 mice (Boltzmann sigmoidal fit, R2 = 0.998, P<0.001). The x-axis is in log scale. (D) Comparison of diagnostic predictability between LDI and ICG perfusion imaging. For three representative mice with different prognoses, LDI images, perfusion maps, and necrosis maps at POD 0 are shown along with pictures of the ischemic limbs at POD 7. The region for the necrosis map from the corresponding area in the perfusion map is indicated by the white dotted boxes.
Figure 3.
Synergistic proangiogenic effects of VEGF and cAng1.
(A) Correlations between necrosis probability and regional perfusion rates were determined. (B) Differences in perfusion rates according to the time period are indicated for each group. ANOVA and Bonferroni post hoc test applied to the significant effect of groups on Δ (POD 3 - POD 7), (ANOVA F3,33 = 4.890, P = 0.006). ††, P = 0.004 vs. BSA control. Paired t-tests were performed for each intragroup comparison. (C) Representative examples of BSA control and combined treatment groups.
Figure 4.
Therapeutic effects of VEGF and cAng1 on tissue perfusion.
Representative cases of BSA control and combined treatment groups.
Figure 5.
Arteriogenic effects of VEGF and cAng1.
(A) Confocal micrographs of calf muscle sections from ischemic limbs treated with VEGF/cAng1 and BSA. Magnification, 200×. Bar = 100 µm. (B) Macro- and microvessel densities were evaluated in the calf muscles. The number of vessels was expressed as vessels/mm2. Data from six fields of the four samples were averaged. Student's t-test was applied. (C) Micro-CT angiography. The arrows indicate macrovessels in the calf.