Indocyanine green angiography: A new method to quantify collateral flow in mice

Objectives: Therapeutic augmentation of collateral artery growth (ie, arteriogenesis) is of particular clinical interest for improving blood flow in vascular occlusive disease. Quantification of collateralization in small animal models is difficult, however, and the commonly used technique of laser Doppler perfusion imaging (LDPI) has always been criticized. Therefore, a new method, termed indocyanine green angiography (ICGA), was established for in vivo imaging of arteriogenesis in mice and compared with LDPI. Method: Using the accepted model of ligation of the left femoral artery of 45 C57BL6 mice, we determined arteriogenesis both by LDPI and ICGA, which were applied before and periodically after ligation of the left femoral artery (each group n = 7). Collateral artery growth within the hind limb was additionally verified by histologic workup. Results: Determination of flow by ICGA, as represented by maximal pixel intensity (ratio of left/right hind limb) demonstrated a drop from 0.97 +/- 0.06 before ligation to 0.11 +/- 0.12 directly after ligation, which recovered to 0.48 +/- 0.22 after 1 week, to 0.65 +/- 0.11 after 2 weeks, and to 0.59 +/- 0.22 after 3 weeks (n = 7, P < .05). Similarly, flow determined as the perfusion index (slope of pixel intensity, ratio left/right) dropped from 1.18 +/- 0.4 before ligation to 0.02 +/- 0.03 immediately after ligation but recovered to 0.08 +/- 0.01 after 1 week, to 0.17 +/- 0.01 after 2 weeks, and to 0.17 +/- 0.06 after 3 weeks (n = 7, P < .05). Quantification by LDPI demonstrated a drop from 1.06 +/- 0.06 (left/right ratio) before ligation to 0.37 +/- 0.03 immediately after ligation. In contrast to ICGA, perfusion recuperated completely within 1 week to 1.01 +/- 0.14 and tended to be even higher in the ligated than in the unligated hind limb after 2 (1.09 +/- 0.25) and 3 weeks (1.20 +/- 0.29), pointing towards limitations of this technique. Histologic analysis confirmed the significant increase in the number of collaterals. The intraindividual ratio increased from 1.0 +/- 0.05 before ligation to 1.35 +/- 0.10 at 2 weeks and 1.41 +/- 0.08 at 3 weeks after ligation (P < .05). Conclusion: Our data demonstrate that ICGA represents a potent tool for the quantification of collateral flow in small animal models. The current standard of LDPI seems to rather represent blood movements within the superficial skin but not of the entire hind limb. (J Vasc Surg 2008;48:1315-21.)