Find en forsker – Københavns Universitet

Videresend til en ven Resize Print Bookmark and Share

Find en forsker

Evaluation of MRI tumor characteristics and quantitative FDG-PET assessments of cerebro-cerebellar diaschisis: Pathophysiologic implications for gliomas

Publikation: Forskning - peer reviewKonferenceabstrakt i tidsskrift

E. A. Segtnan, J. Holm, J. H. Decker, C. Constantinescu, A. Gjedde, P. F. Hoilund-Carlsen, J. Ivanidze

Purpose: Using FDG-PET-based THGr methodology and MRI-based volume segmentation of key lesion characteristics, we sought to improve understanding of the implications of cerebral and cerebellar diaschisis in the diagnosis and management of supratentorial gliomas.

Methods: A prospective cohort of 14 glioma patients (5 men, 9 women; mean age 63 years, range 35–77) underwent baseline PET-CT and MRI. Tumor, edema, and necrosis volume were obtained based on volume segmentation of gadolinium-enhanced T1-weighted images using 3D Slicer Software, Version 4.5 (http://www.slicer.org). We obtained total hemispheric glucose metabolic ratios (THGr) by dividing total hemispheric FDG uptake in each hemisphere with the expected diaschisis site, i.e., the ipsilateral cerebral hemisphere (THGr(Ce)) and the contralateral cerebellar hemisphere (THGr(Cb)), to its respective contralateral side. THGr values were compared with qualitative assessment of diaschisis. Linear regression analysis was performed using GraphPad Prism 6 software.

Results: Volumetric segmentation yielded the following volumes in mL (mean followed by (standard deviation)): (enhancing) tumor 30.25 (23.50); edema 94.83 (62.65); necrosis 5.51 (5.23). Using the same notation, quantitative PET analysis yielded the following THGr values: THGr(Ce), 0.72 (0.24); THGr(Cb), 0.83 (0.22). Logistic regression analysis demonstrated the following R-square values (with p values in parentheses): THGr(Ce) versus edema volume, 0.2 (0.10); THGr(Ce) versus enhancing lesion volume, 0.03 (0.5); THGr(Ce) versus necrosis volume, 0.001 (0.9). THGr(Cb) versus edema volume, 0.47 (0.0065); THGr(Cb) versus enhancing lesion volume, 0.11 (0.2579); THGr(Cb) versus necrosis volume, 0.009 (0.74).

Conclusions: The THGr(Cb) and MRI segmentation analyses demonstrated an inverse correlation of cerebellar diaschisis with edema volumes. In contrast, there was no correlation of diaschisis to lesion volume or necrosis. Given the prognostic value of diaschisis in glioma assessment, our findings have important pathophysiologic and clinical implications, laying the foundation for future studies evaluating diaschisis and tumor-associated edema in a larger cohort
OriginalsprogEngelsk
TidsskriftJournal of Cerebral Blood Flow and Metabolism
Vol/bind37
TidsskriftsnummerS1
Sider (fra-til)245-245
ISSN0271-678X
DOI
StatusUdgivet - apr. 2017

ID: 182544515