Pathological analysis of the surgical margins of resected glioblastomas excised using photodynamic visualization with both 5-aminolevulinic acid and fluorescein sodium.

During glioma resection, 5-aminolevulinic acid (5-ALA) and fluorescein sodium (Fl-Na) are used for photodynamic tumor visualization. The objective of this study was to evaluate the pathological findings of the boundary zone between the tumor and adjacent normal brain in glioblastoma patients undergoing simultaneous double staining with 5-ALA and Fl-Na during surgery. Eight patients received 5-ALA (20 mg/kg orally) before the induction of general anesthesia, and Fl-Na (20 mg/kg) was administered intravenously before the dural incision was performed. The tumor bulk was removed under the guidance of Fl-Na staining alone using conventional white light. Subsequently, residual tumor was removed under the guidance of both fluorescent agents within functionally safe limits until both were visibly undetectable. Twenty specimens exhibiting different staining intensities of both agents were obtained. The vessel index (VI) was calculated from CD31 immunohistochemistry (IHC) samples. Boundary zone tumor cells were detected by IHC for olig2, and were expressed as the olig2 index (OLI). The VI was significantly higher in Fl-Na-positive areas than in Fl-Na-negative areas (p = 0.0005). In contrast, the OLI was significantly higher in 5-ALA-positive areas than in 5-ALA-negative areas (p = 0.0149). 5-ALA-positive/Fl-Na negative areas were observed in 7 patients. These findings indicate that Fl-Na accumulates in areas with a disrupted blood-brain barrier, and that 5-ALA fluorescence is dependent on tumor cell protoporphyrin IX metabolism. In conclusion, 5-ALA was better for detecting tumor cells in the boundary zone than was Fl-Na. Of note, tumor cells existed outside the fluorescence-stained boundaries of both agents.