The utilization of fluorescein-guided biopsies has recently been discussed to improve and expedite operative techniques in the detection of tumor-positive tissue, as well as to avoid making sampling errors. In this study, we aimed to report our experience with fluorescein-guided biopsies and elucidate distribution patterns in different histopathological diagnoses in order to develop strategies to increase the efficiency and accuracy of this technique. We report on 45 fluorescence-guided stereotactic biopsies in 44 patients (15 female, 29 male) at our institution from March 2016 to March 2021, including 25 frame-based stereotactic biopsies and 20 frameless image-guided biopsies using VarioGuide (R). A total number of 347 biopsy samples with a median of 8 samples (range: 4-18) per patient were evaluated for intraoperative fluorescein uptake and correlated to definitive histopathology. The median age at surgery was 63 years (range: 18-87). Of the acquired specimens, 63% were fluorescein positive. Final histopathology included glioblastoma (n = 16), B-cell non-Hodgkin lymphoma (n = 10), astrocytoma, IDH-mutant WHO grade III (n = 6), astrocytoma, IDH-mutant WHO grade II (n = 1), oligodendroglioma, IDH-mutant and 1p/19q-codeleted WHO grade II (n = 2), reactive CNS tissue/inflammation (n = 4), post-transplantation lymphoproliferative disorder (PTLD; n = 2), ependymoma (n = 1), infection (toxoplasmosis; n = 1), multiple sclerosis (n = 1), and metastasis (n = 1). The sensitivity for high-grade gliomas was 85%, and the specificity was 70%. For contrast-enhancing lesions, the specificity of fluorescein was 84%. The number needed to sample for contrast-enhancing lesions was three, and the overall number needed to sample for final histopathological diagnosis was five. Interestingly, in the astrocytoma, IDH-mutant WHO grade III group, 22/46 (48%) demonstrated fluorescein uptake despite no evidence for gadolinium uptake, and 73% of these were tumor-positive. In our patient series, fluorescein-guided stereotactic biopsy increases the likelihood of definitive neuropathological diagnosis, and the number needed to sample can be reduced by 50% in contrast-enhancing lesions.
