Epilepsy surgery around language cortex

Background. Both epilepsy surgery and surgery close to functionally
significant cortical areas have challenged neurosurgeons in the
last two decades. With this work we wanted to illustrate the
current status of epilepsy surgery close to language cortex in our
clinic (Neurosurgery Clinic, University of Munich) and to evaluate
our tactic of repeated intra-operative language mapping after
initial extra-operative language mapping in cases, where language
areas lie very close to or overlaps with the epileptogenic zone.
First part of this work describes the process of decision making in
epilepsy surgery – patient admission criteria, gradual
investigational process from non-invasive to invasive. The main
emphasis is put to the analysis of invasive language mapping
(extra- and intra-operative) as this is the current gold standard
of language localization in neurosurgery. Here the historical
development of language mapping, together with its physical and
physiological concerns is discussed. The next part of this work is
devoted to the analysis of two different invasive language mapping
tactics – extra-operative versus combined extra- and
intra-operative mapping. Methods. Group of retrospective (19) and
prospective (3) patients, operated in our clinic in time period
from 1997 to 2007, was gathered. Among these 22 patients were 11
male and 11 women with a mean age of 31,9 years and mean epilepsy
duration of 16,3 years. Only those patients, by whom either by
extra-, intra-operative or both stimulation methods a language
cortex close to or overlapping with epileptogenic zone was found,
were included in our study. The patients were divided in 2 groups,
basing on the language mapping tactic, used during the
investigation. Only extra-operative language mapping was used in
cases, where rather safe distance (more than 10 mm) between
language sites and epileptogenic zone was seen (Ex-M group). The
necessity for additional intra-operative language mapping was seen
in cases, where rather small (less than 10 mm) distance between
language sites and epileptogene cortex or overlapping of both zones
was seen (Co-M group). Results. Only extra-operative language
mapping was used for 8 patients and the combination of both
language mapping techniques was used in 12 cases. In 1 case
language was mapped by functional magnetic resonance and in 1 case
– only intra-operatively. All patients underwent resective
operations. Immediate post-operative language deterioration was
seen only in 10 (45,4 %) cases (6 (75%) cases in Ex-M sub-group and
4 (33,3%) in Co-M sub-group) out of the whole group. In 2 cases (1
in each group) the language deterioration was permanent (detectable
also 6 months after surgery). The patient in the Co-M sub-group had
permanent language deterioration already pre-operatively. Thus the
only new permanent post-operative language deterioration was seen
in 1 case of Ex-M sub-group, where rather safe distance between
language and epileptogenic zone was thought pre-operatively.
Regarding seizure outcome, patients were evaluated for at least 2
years (mean follow up 46,6 months). The results were gathered from
18 patients (only retrospective patients) and were as follows:
Engel I – 9 cases (50%), Engel II – none, Engel III – 2 (11,1%)
cases, Engel IV – 7 (38,9%) cases. In 9 unfavourable seizure
outcome cases (combination of Engel class III and IV cases)
apparently no full resection of the epileptogene zone was achieved.
In 5 cases this was known already intra-operatively, in the
remaining 4 cases it was noted during the follow up period. In 8 of
these cases the reason for incomplete resection of the epileptogene
zone was its close relationship or overlapping with speech cortex
and/or difficult localization of the epileptogenic zone. In 1 case
complete resection could not be done due to intra-operative
complications. In the Co-M sub-group (n=9) the results were
following: Engel I – 3 (33,3%) cases, Engel II – none, Engel III –
1 (11,1,%) case and Engel IV – 5 (55,6%) cases. In the Ex-M
sub-group (n=7), the results were following: Engel I – 4 (57,1%),
Engel II-none, Engel III – 1 (14,2%), Engel IV – 2 (28,7%) cases.
No statistically significant differences were observed between both
groups regarding immediate post-operative language deterioration,
new persistent language deterioration and Engel class I outcome.