Page 220 - ILAE_Lectures_2015

Decline of memory function following anterior temporal lobe resection, particularly of verbal
memory after left-sided ATLR, is a major concern. Several studies have investigated the role
of fMRI in predicting the effects of ATLR on memory88-93. Most focused on the prediction of
verbal memory decline; only a few investigated visual memory decline after non-dominant
ATLR. Patients with greater ipsilateral than contralateral medial temporal lobe activation on
fMRI memory activation studies had greater memory decline following temporal lobe
resection for both verbal memory decline following dominant temporal lobe resection, and
for non-verbal memory decline following non-dominant temporal lobe resection. This
investigation suggests that preoperative memory fMRI may be a useful non-invasive
predictor of postoperative memory change following temporal lobe resection93.

The choice of paradigm is important in determining activation patterns. Greater left anterior
hippocampal activation on word encoding was predictive of greater post-operative decline in
verbal memory after left-sided resection, and greater right anterior hippocampal activation on
face encoding predicted greater decline in design learning after right-sided resection94. Also,
greater left than right posterior hippocampal activation correlated with better postoperative
verbal memory outcome and greater right than left posterior hippocampal activation
correlated with better visual memory outcome. This suggests that reorganisation of function
to the posterior hippocampus is associated with better preservation of memory following
anterior resection188.

MAGNETIC RESONANCE SPECTROSCOPY

The metabolites which are detectable using proton spectroscopy (1H MRS) depend on the
conditions used for the acquisition. In epilepsy studies in vivo, the principal signals of interest
have been those from N-acetyl aspartate (NAA), creatine + phosphocreatine (Cr), choline-
containing compounds (Cho), and lactate (Lac). There is evidence that NAA is located
primarily within neurons and precursor cells and a reduction of NAA signal is usually
regarded as indicating loss or dysfunction of neurons. Cr and Cho are found in both neurons
and in glia.

MRS in temporal lobe epilepsy

In TLE caused by HS, MRS showed reduction of NAA and increases of choline-containing
compounds, creatine + phosphocreatine, reflecting neuronal loss or dysfunction and
astrocytosis95. Analysis of individual patients showed a reduced NAA/choline + creatine ratio
on the side of the focus in 88%, with 40% having bilateral abnormalities. Quantitative short
echo time MRS showed the association of HS with low NAA and raised myoinositol, and
also an elevation of glutamate and glutamine in epileptic hippocampi that were structurally
normal96. The implication from these data was that there is neuronal loss or dysfunction and
astrocytosis in the temporal lobes of patients with TLE. Abnormalities of metabolite profiles
may be found in temporal lobes with normal MRI97-99 and bilateral abnormalities have been
noted in up to 50% of patients with apparently unilateral structural abnormality100, indicating
that MRS may be more sensitive for detecting pathology. The role of MRS in predicting
outcome is not clear: in one study of patients with TLE and normal MRI, a lower
NAA/choline + creatine was found in those who did not become seizure free101. NAA was
not reduced in the hippocampi of patients with neocortical epilepsy, either ipsilateral or
contralateral to the focus102, suggesting that hippocampal dysfunction is not a feature of
neocortical epilepsy. Mesial TLE was associated with reductions of NAA in frontal grey and
white matter, which is consistent with other data suggesting more widespread involvement103.