Blood Flow Assessment of Arteriovenous Malformations Using Intraoperative Indocyanine Green Videoangiography

Abstract

Intraoperative indocyanine green (ICG) videoangiography is widely used in patients undergoing neurosurgery. FLOW800 is a recently developed analytical tool for ICG videoangiography to assess semi-quantitative flow dynamics; however, its efficacy is unknown. In this study, we evaluated its functionality in the assessment of flow dynamics of arteriovenous malformation (AVM) through ICG videoangiography under clinical settings. ICG videoangiography was performed in the exposed AVM in eight patients undergoing surgery. FLOW800 analysis was applied directly, and gray-scale and color-coded maps of the surgical field were obtained. After surgery, a region of interest was placed on the individual vessels to obtain time-intensity curves. Parameters of flow dynamics, including the maximum intensity, transit time, and cerebral blood flow index, were calculated using the curves. The color-coded maps provided high-resolution images; however, reconstruction of colored images was restricted by the depth, approach angle, and brain swelling. Semi-quantitative parameters were similar among the feeders, niduses, and drainers. However, a higher cerebral blood flow index was observed in the feeders of large AVM (>3 cm) than in those of small AVM (P < 0.05). Similarly, the cerebral blood flow index values were positively correlated with the nidus volume (P < 0.01). FLOW800 enabled visualization of the AVM structure and safer resection, except in case of deep-seated AVM. Moreover, semi-quantitative values in the individual vessels through using ICG intensity diagram showed different patterns according to size of the AVM. ICG videoangiography showed good performance in evaluating flow dynamics of the AVM in patients undergoing surgery.

Figure 1

Example of ROI placement in an individual vessel of the AVM. Diagram showing the time-intensity curve of ICG in the individual vessel with ROI placement.

Figure 2

Comparison of parameters according to vessel types. Maximum intensity (a), delay time (b), and CBFi (c). Values showed absence of significant differences among the vessel types. The CBFi value of the feeders was significantly higher in the niduses of > 3-cm diameter (d). In addition, the CBFi of the feeders was correlated with the cubage of the nidus (e). Representative time-intensity diagrams presenting three different patterns according to the nidus volume (f). CBFi: cerebral blood flow index.

Figure 3

Case 3. DSA revealed the presence of AVM in the right parasagittal postcentral region (a, b). Intraoperative photograph at after dural opening demonstrating restricted exposure of the AVM (c). Delay map using FLOW800 showed poor performance in demonstrating the accurate structure of the entire AVM (d).

Figure 4

Case 5. Preoperative DSA showed an AVM in the right occipital lobe (a, b). Intraoperative photograph at after dural opening demonstrating surface appearance of the AVM (c). Delay map showing the feeders and accurate location of the nidus (d).