Identification of premotor white matter tracts involved in language before and after resection of tumors in the left hemisphere.

The mouth and hand are two important effector systems (speaking, writing) to
express yourself in language. Input to the involved motor cortex is channeled
via premotor cortex areas. Dysfunction in language is classically caused by
disturbances in the left inferior frontal gyrus (Broca's area, expressive
aphasia) or the left posterior temporal area (Wernicke's area, receptive
aphasia). Conduction-aphasias are caused by damage to the arcuate fascicle, the
white matter tract connecting the two language areas. In practice however,
aphasias are a heterogenous group, with both Broca- and Wernicke-like symptoms.
Dysfunction of language can also be caused by damage to the supplementary motor
area (SMA), which varies from less spontaneous speech to mutism. Both for
tumors in the SMA and perisylvian tumors it is difficult to predict
postoperative deficit preoperatively. It is important to realize that a
function is not isolated in a specific cortical area, but is represented by a
broader network of interconnected cortical areas. Aside from disturbance of
these connections caused by the tumor, the neurosurgical approach to the tumor
may also be a cause of disturbing these connections. The effects of a tumor on
language-related connections to the premotor cortex are evaluated in this study
using DTI.

The SMA is located in the dorsomedial frontal cortex, anterior of the
representation of the legs in M1, and can be regarded as the medial extension
of the premotor cortex. The SMA is somatotopically organised. Orofacial
movements are evoked with anterior stimulation of the SMA, whereas more
posterior stimulation evokes arm and leg movements. A substantial part of the
cells in the piramidal tract originates directly from the SMA. Apart from
direct output to the piramidal tract, the SMA is also a secondary motor area.
Activation of the SMA is associated with starting movements, the performance of
more complex movements such as bilateral movements or the selection of
movements and motor learning. There are strong interhemispheral connections
between the two SMAs. Neuronal activity in the SMA is associated with both
contra- and ipsilateral movements. The SMA is connected with the piramidal
tract, primary motor cortex, other premotor areas and primary and secondary
sensory areas.

A lesion in the SMA leads to dysfunction in the performance of bimanual
movements in monkeys. In humans, surgical excision of the SMA can lead tot
typical postoperative deficit, called the SMA syndrome, with many different
symptoms. Most of the time there is contralateral motor dysfunction, varying
from a complete hemiparesis to less spontaneous motor output, also called motor
neglect. The motor deficit is unpredictable, but seems to be more prevalent
when the resection includes more dorsal parts of the SMA, close to the motor
strip. Language dysfunction can also be part of the SMA syndrome (less
spontaneous speech, mutism). The broad clinical picture may suggest a
heterogenous group. However, the resemblance is that all postoperative deficits
recover almost completely within weeks to months. With fMRI is has become clear
that postoperative deficit is associated with a stronger activation of premotor
areas (SMA and laterale premotor cortex) in the healthy hemisphere.
Furthermore, there was more activation in the premotor cortex of the healthy
hemisphere, which was related to the extent of resection of preoperative SMA
activation. This suggests that there is plasticity.

Our hypothesis is that a left hemisphere tumor causes language dysfunction by
mass effect on or infiltration of white matter tracts connecting Wernicke's
area with Broca's area and between Broca's area and the SMA. The neurosurgical
operation can have influence on these connections (by a decrease in mass effect
or by resection of infiltrated areas during which also functional white matter
tracts are resected). Functional recovery postoperatively can be caused by
recovery of existing connection and/or functional reorganization, where other
connections take over the affected functions. Language dysfunction will be
related to the disturbance of white matter tracts to gain more insight in the
nature of aphasias. The involved white matter tracts will be quantified using
diffusion tensor imaging (DTI). Also, the relation between disturbance of SMA
connections and motor dysfunction (especially bilateral tasks) is examined.

DTI images in 60 directions will be made in subjects with a intraparenchymal
tumor in the left hemisphere. During a period of two years patients will be
included in this observational study.
Standard neurological examination will be performed preoperatively to document
preoperative dysfunction. This is a baseline for later examinations. Phase and
antiphase movements will be tested as a specific parameter for SMA function.
Language functions and more general neuropsychological examinations will be
quantified by the participating neuropsychologist with standard tests. Also,
handedness is assessed preoperatively with the Edinburgh Handedness Inventory.
The neurological examination en testing of language functions will be repeated
postoperatively and at discharge to document possible deficit. These pre- and
postoperative neuropsychological and neurological tests are part of the
standard work-up in this patient population. Within 48 hours after operation a
MRI with diffusion-weighted imaging (DWI) will be made to assess the
postoperative picture. This is also part of the standard care in these patients.
After a mean of 4,5 months another DTI scan will be made to compare the white
matter tracts with the preoperative situation. At that time the standard
postoperative neurological and neuropsychological examination will be repeated.
General patient characteristics such as date of birth, gender, time to recovery
of deficit and pathology of the tumor will be documented.

First the ventral premotor cortex, SMA and language areas of Broca and Wernicke
will be identified as a base for seed-based tractography to identify white
matter tracts and displacement by tumor. Using seed-based tractography we track
the healthy contralateral hemisphere to see if there are stronger connections
in the homologous equivalents of the mentioned areas. The areas of Broca and
Wernicke in the right hemisphere will be defined on the same coordinates as the
left hemisphere where we mirror left and right. At first we investigate if
there is a relation between the disruption of white matter tracts and
preoperative symptoms or postoperative deficit. We compare preoperative and
late postoperative DTI images to assess if there is irreversible damage or that
there is a decreased functional resolution caused by edema. A difference in
white matter tracts is characterized with probabilistic tractography in FMRIB
Software Library (FSL).

Extra investigations in these subjects:
• A preoperative DTI investigation in the Neuro-Imaging Center.
• Within a mean of 4,5 months postoperatively an extra DTI investigation should
take place in the Neuro-Imaging Center.

The use of MRI is considered very safe, especially when there is good attention
for contra-indications. The risk is negligible and the burden is small so we
think this research is justifiable.