Abstract

INTRODUCTION: In patients with medulloblastomas, the combination of CSF cytology and MR imaging is believed to be superior to either tool alone in detecting leptomeningeal dissemination (LMD). However, MRI has drastically improved since the initial assessment of this approach. METHODS: A retrospective review of 117 unique patients with medulloblastomas from a single institution was conducted. CSF samples and MR images obtained within 36 days of each other were included. A 3-tier system (negative, atypical/suspicious, positive) was utilized to account for equivocal findings. A total of 253 CSF-MRI pairs (range: 1-12 pairs per patient) obtained between 2000-2021 were analyzed. RESULTS: The mean (SD) age of this cohort was 14.1 (13.5) years (62.4% were male). Gross total resection with surgery was achieved in 64.5%, 83.7% received radiotherapy, 81.5% received chemotherapy. 16.2% were deceased, with the mean (SD) time till death being 4.6 (4.7) years following the first operation. The molecular subtypes were: SHH (17.0%), WNT (3.6%), non-WNT/non-SHH (46.6%), unknown (32.8%). There was a 60.5% concordance between CSF cytology and MRI, with no significant differences by molecular subtype. In 57.3% of cases, CSF cytology and MRI were both negative. In 1.2% of cases, CSF cytology and MRI were both equivocal. 26.9% of cases were positive on CSF cytology and/or MRI. MRI had an 86.8% sensitivity and CSF cytology had a 20.6% sensitivity for detecting LMD that was positive in CSF, MRI, or both. In 10 cases (4.0%), MRI was negative when CSF was atypical or positive. In 80 cases (31.6%), CSF was negative when MRI was positive or suspicious. In 94.5% of cases, MRI demonstrated greater or equal ability to CSF cytology in detecting LMD. CONCLUSIONS: MRI is superior to CSF cytology in detecting LMD but misses a small percentage of cases.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call