Vera Catharina Keil MD, Presenter: Nothing to Disclose

Gerrit H. Gielen, Abstract Co-Author: Nothing to Disclose

Bogdan Pintea, Abstract Co-Author: Nothing to Disclose

Hans H. Schild MD, Abstract Co-Author: Nothing to Disclose

Susanne Greschus MD, Abstract Co-Author: Nothing to Disclose

Dariusch Reza Hadizadeh Kharrazi MD, Abstract Co-Author: Nothing to Disclose

Matthias Simon, Abstract Co-Author: Nothing to Disclose

Juergen Gieseke DSc, Abstract Co-Author: Employee, Koninklijke Philips NV

To assess whether the quality of brain biopsies defined by permeability maps based on T1-dynamic contrast-enhanced (DCE) MRI differs from biopsy site selection determined by a standard mapping method (T1+gadolinium maps).

In 12 brain tumor patients (5 glioblastoma (GBM), 1 anaplastic astrocytoma (AA), 1 oligoastrocytoma (OA), 5 meningioma), preoperative MRI were performed to guide neurosurgical biopsies; the examination encompassed a T1-DCE permeability sequence at 3.0 Tesla (flip angle, 8°; dynamic scans, 50; voxel size, 1.57 x 1.6 x 3.0 mm, ap x rl x cc) as well as a 3D T1 sequence after iv gadobutrol (Gadovist) administration (0.1 mmol/kg body weight). Regions of interest (ROI) for tumor biopsies were defined either by choice of a neurosurgeon based on T1-weighted gadolinium-enhanced images (T1+Gd), or by a neuroradiologist based on permeability (Ktrans) and extravascular extracellular volume fraction (ve) hot-spots. Subsequently, biopsies at these locations of the tumor were retrieved using a dedicated software (iPlanNet/BrainLab). A blinded neuropathologist compared biopsy histology to a reference histology based on separate large biopsies.

A total of 21 glioma (14 GBM, 5 AA, 2 OA) and 10 meningioma navigated biopsies were taken. Of these, 9 GBM biopsies were exclusively selected by Ktrans maps (neuroradiologist), whereas 5 were chosen based on T1+Gd maps (neurosurgeon, standard method). 4/9 (44.4%) of Ktrans defined GBM biopsy histologies, but only 1/5 (20.0%) of T1+Gd based biopsy histologies matched the reference histological diagnosis which based on large biopsy samples. In retrospective analysis, ROI based on T1+Gd showed lower average Ktrans permeability and ve than those selected by permeability maps (617.56 vs. 43.15 10-3/min; ve 2.57 vs. 0.16).

ROI selection based on Ktrans permeability maps may improve the diagnostic quality of glioma biopsies.

Unfavourable choice of CNS biopsy location may lead to doubtful diagnoses and false treatment. T1-DCE MRI planned biopsy may improve the diagnostic quality of biopsy samples over standard T1+Gd based biopsy location.

Keil, V, Gielen, G, Pintea, B, Schild, H, Greschus, S, Hadizadeh Kharrazi, D, Simon, M, Gieseke, J, T1-DCE MRI for Targeting Navigated Biopsies in Intracranial Neoplasms.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14045571.html