gms | German Medical Science

Objective: CT angiography (CTA) has been reported to yield a high sensitivity, specificity, and accuracy in detection of (i.) no vasospasm or (ii.) severe cerebral vasospasm in proximal cerebral arteries in SAH patients. However, it has been shown to be less accurate for detecting mild and moderate vasospasms. In order to provide an accurate method for computer-based grading of vasospasm in small animal models we have developed a method for volumetric vessel segmentation from CTA data using Amira software (FEI). The present study was made to evaluate the accuracy of the virtual reconstruction of the vascular tree by comparing optical evaluation of anatomical cast preparations of murine brains with an evaluation of the virtual vascular tree reconstructed from CT data.

Method: 9 mice, 5 of these with subarachnoid hemorrhage, were subjected to PFA perfusion and casting with a radiopaque agent. Arteries of the skull base were photographed using a reflecting microscope. Volume datasets were obtained using an industrial micro-CT (Yxlon Y.Fox CT scanner) at a resolution of 15 µm. Amira software, which offers the possibility of volumetric analysis of whole vessel segments, was used to analyze vessel diameter at thirteen anatomically defined points. The vessel diameters were determined from the photographs. Finally, the ratios of CT- to optical diameter were determined.

Results: The overall ratio (mean ± 1standard deviation) of CT- to optical diameter was 1.03 ± 0.05 (n=117), with comparable ratios for the single vessel segments evaluated in the basilar artery, internal carotid artery, and middle cerebral artery (1.05 ± 0.04, 1.04 ± 0.05 µm, and 0.98 ± 0.03, respectively).

Conclusions: Amira software provides anatomically accurate reconstructions of the vascular tree from micro-CT datasets in mice with SAH-related vasospasm. Thus, the evaluated approach seems to be well suited to serve a as method for quantification of treatment effects in animals as small as mice.