Jan Gralla MD, Presenter: Nothing to Disclose

Caspar Brekenfeld, Abstract Co-Author: Nothing to Disclose

Peter Kalus, Abstract Co-Author: Nothing to Disclose

Raphael Guzman, Abstract Co-Author: Nothing to Disclose

Claus Kiefer, Abstract Co-Author: Nothing to Disclose

Johannes Slotboom, Abstract Co-Author: Nothing to Disclose

Functional magnetic resonance imaging (fMRI) is a non-invasive widely available technique to locate eloquent cortex areas. Compared to the sensorimotor function the localization of speech related areas shows a lateral dominance and a high inter-individual variation. Our objective was the development of reliable speech paradigm in combination with a steady setup for fMRI applicable in preoperative routine imaging. In a prospective study 18 patients with left frontal lesions or right frontal lesions and history of supposed speech deficits have been investigated. The value of the fMRI activation patterns on the surgical plan was evaluated.

The T2*-weighted sequence consists of 68 measurements (TE 60ms, TR 6000ms)covering the whole brain with 30 slices (4mm)performed on the Sonata 1.5T scanner (Siemens, Germany). The speech paradigm (6:54 min) is composed of 8 blocks (per 24 sec) with semantic comparison of two sentences (activation) alternating nonsense words (baseline). The paradigm is patient triggered, by replying "yes" (button 1 = same semantic meaning) or "no" (button 2) a new task is presented.

In all 18 patients (mean 39.1 years), fMRI data acquisition was performed during routine preoperative imaging and revealed a high activation of Wernicke's and Broca's area. In the group of patients with right frontal lesions (9 pat.) one left-handed patient had speech activation on the same side. In one right-handed patient of the subgroup with left frontal lesions (9 pat.) fMRI revealed unexpected speech activation in the right frontal lobe, confirmed by intraoperative cortical stimulation. In 6 of the remaining 8 cases extent of resection was modified according to the results of fMRI. Bilateral co-activation occurred in 4 cases.

The setup including the new paradigm is applicable for routine imaging and the patient-triggered design allows an individual and maximum number of tasks. This leads to high activation levels providing good delineation of speech related eloquent cortex areas. The activation patterns reveal unexpected findings and have an influence on the extent of resection. The introduced setup favors speech-activating fMRI for preoperative routine imaging.

Gralla, J, Brekenfeld, C, Kalus, P, Guzman, R, Kiefer, C, Slotboom, J, Preoperative fMRI in Routine Imaging and Clinical Value of Speech Activation Patterns.  Radiological Society of North America 2004 Scientific Assembly and Annual Meeting, November 28 - December 3, 2004 ,Chicago IL. http://archive.rsna.org/2004/4413238.html