[The rolandic line: a simple baseline for the identification of the central region. An MRI study and functional validation].

INTRODUCTION: The identification of the central region--i.e. the central sulcus, the pre- and post-central gyri, the paracentral lobule--on MRI and angiographic images may be necessary prior to stereotactic procedures such as biopsies or resection of centrally located tumors, depth electrode recordings for presurgical evaluation of drug-resistant epilepsies, or radiosurgery of arteriovenous malformations. Stereotactic methods, such as the Talairach's proportional grid based on the bicommissural system, demonstrated the statistical position of the central sulcus according to the Ac-Pc, Vac and Vpc baselines. However, the course and the spatial position of this sulcus have remarkable individual differences that sometimes make the sulcus difficult to identify on serial sagittal MRI or lateral angiographic images. In order to facilitate this identification, the authors propose a new oblique baseline, the rolandic (R) line.

MATERIAL AND METHODS: The stereotactic MRI and angiography of 22 patients were reviewed for this study. Eleven of these patients had stereotactic biopsies for a low-grade tumor located in the central region, while eleven others had multiple intracerebral electrodes implantation and depth EEG recording (SEEG: stereoelectroencephalography) in the presurgical evaluation of drug-resistant partial epilepsy, prior to epileptogenic cortex resection. The Ac-Pc, Vac, Vpc baselines and segments of the central sulcus were drawn from the mid-sagittal and lateral T1-weighted MRI images and reported on an individual graph. Surface and deep margins as well as axis of the central sulcus were also reported along with corpus callosum baselines as defined by Olivier et al.: horizontal plane, anterior and posterior callosal planes. The rolandic line was then traced from the graph:it joined the intersection point between the anterior callosal plane and an orthogonal line passing through the floor of the temporal fossa, and the intersection point between posterior callosal plane and an orthogonal line passing through the top of the hemisphere. The rolandic line was then superimposed on any sagittal MRI image or lateral stereotactic angiographic film. Finally, the spatial position of electrode contracts through which electrical stimulations elicited motor and/or sensory responses, either from central electrode implanted for motor fibers identification prior to stereotactic biopsies or from SEEG electrodes implanted for epileptogenic zone identification and cortical mapping, was reported on the individual graph. Angular and linear measurements were taken from the graph, between the rolandic line, the central sulcus axis, the Ac-Pc and callosal baselines, and the central sulcus limits (top, bottom, anterior and posterior margins).

RESULTS: Graph measurements indicated that the rolandic line was significantly closer to the inferior part of the central sulcus than to its superior part (average distance between the line and the inferior point of the sulcus: 1.86 +/- 1.87 mm; average distance between the line and the superior point of the sulcus: 4.5 +/- 2.3 mm; p < 0.001-t test); similarly, the rolandic line was closer to the deep margin of the sulcus rather than to its superficial border (average distance between the line and the most anterior point of the sulcus: 11.43 +/- 3.16 mm; average distance between the line and the most posterior point of the sulcus: 7.95 +/- 4.14 mm; p < 0.01-t test). In 90% of the cases, the rolandic line followed the deep or middle part of the sulcus, with an average angle of 4.18 degrees +/- 2.53 degrees between the line and the sulcus axis. The spatial position of the electrode contacts that elicited motor/or sensory responses to stimulations correlated topographically well in all cases with the position of the motor and sensory fibers defined according to the central sulcus, baselines and reference to stereotactic atlases. (ABSTRACT TRUNCATED)