Sequential Visualization of Brain and Fiber Tract Deformation during Intracranial Surgery with Three-dimensional Ultrasound: An Approach to Evaluate the Effect of Brain Shift

Abstract

Abstract OBJECTIVE: We present a technique that allows intraoperative display of brain shift and its effects on fiber tracts. METHODS: Three patients had intracranial lesions (one malignant glioma, one metastasis, and one cavernoma) in contact with either the corticospinal or the geniculostriate tract that were removed microneurosurgically. Preoperatively, magnetic resonance diffusion-weighted imaging (DWI) was performed to visualize the fiber tract at risk. DWI data were fused with those obtained from anatomic T1-weighted magnetic resonance imaging. A single-rack three-dimensional ultrasound neuronavigation system, which simultaneously displays the MRI scan and the corresponding ultrasound image, was used intraoperatively for 1) navigation; 2) definition of fixed and potentially shifting ultrasound landmarks near the fiber tract; and 3) sequential image updating at different steps of resection. The result was time-dependent brain deformation data. With a standard personal computer equipped with standard image software, the brain shift-associated fiber tract deformation was assessed by use of sequential landmark registration. After surgery, DWI was performed to confirm the predicted fiber tract deformation. RESULTS: The lesions were removed without morbidity. Comparison of three-dimensional ultrasound with DWI and T1-weighted magnetic resonance imaging data allowed us to define fixed and potentially shifting landmarks close to the respective fiber tract. Postoperative DWI confirmed that the actual fiber tract position at the conclusion of surgery corresponded to the sonographically predicted fiber tract position. CONCLUSION: By definition and sequential intraoperative registration of ultrasound landmarks near the fiber tract, brain shift-associated deformation of a tract that is not visible sonographically can be assessed correctly. This approach seems to help identify and avoid eloquent brain areas during intracranial surgery.