Brain Tumor Surgery With 3-Dimensional Surface Navigation

BACKGROUND: Precise lesion localization is necessary for neurosurgical procedures not only during the operative approach, but also during the preoperative planning phase. OBJECTIVE: To evaluate the advantages of 3-dimensional (3-D) brain surface visualization over conventional 2-dimensional (2-D) magnetic resonance images for surgical planning and intraoperative guidance in brain tumor surgery. METHODS: Preoperative 3-D brain surface visualization was performed with neurosurgical planning software in 77 cases (58 gliomas, 7 cavernomas, 6 meningiomas, and 6 metastasis). Direct intraoperative navigation on the 3-D brain surface was additionally performed in the last 20 cases with a neurosurgical navigation system. For brain surface reconstruction, patient-specific anatomy was obtained from MR imaging and brain volume was extracted with skull stripping or watershed algorithms, respectively. Three-dimensional visualization was performed by direct volume rendering in both systems. To assess the value of 3-D brain surface visualization for topographic lesion localization, a multiple-choice test was developed. To assess accuracy and reliability of 3-D brain surface visualization for intraoperative orientation, we topographically correlated superficial vessels and gyral anatomy on 3-D brain models with intraoperative images. RESULTS: The rate of correct lesion localization with 3-D was significantly higher (P = .001, chi(2)), while being significantly less time consuming (P < .001, chi(2)) compared with 2-D images. Intraoperatively, visual correlation was found between the 3-D images, superficial vessels, and gyral anatomy. CONCLUSION: The proposed method of 3-D brain surface visualization is fast, clinically reliable for preoperative anatomic lesion localization and patient-specific planning, and, together with navigation, improves intraoperative orientation in brain tumor surgery and is relatively independent of brain shift.