Epilepsy

J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp-2012-304083 on 18 April 2013. Downloaded from http://jnnp.bmj.com/ on June 2, 2023 at Cote d'Ivoire:BMJ-PG Spon.
RESEARCH PAPER

Long-term outcome of surgical disconnection of the

epileptic zone as an alternative to resection for
nonlesional mesial temporal epilepsy
Nicolas Massager,1,2 Patrick Tugendhaft,2,3 Chantal Depondt,2,3 Thomas Coppens,2,3
Landry Drogba,1 Nadir Benmebarek,1 Olivier De Witte,1 Patrick Van Bogaert,2,4
Benjamin Legros2,3
1
Department of Neurosurgery, ABSTRACT Spencer, transcortical-transventricular temporal
Clinic of Stereotactic and Background Pharmacoresistant epilepsy can be treated resection by Niemeyer and transsylvian selective
Functional Neurosurgery,
ULB-Hôpital Erasme, Brussels, by either resection of the epileptic focus or functional amygdalohippocampectomy by Yasargil and Wieser.
Belgium isolation of the epileptic focus through complete All these surgical procedures include resection of
2
Reference Center for the disconnection of the pathways of propagation of the the amygdalohippocampal complex; their differ-
Treatment of Refractory epileptic activity. ences are related to various surgical approaches of
Epilepsy, ULB-Hôpital Erasme,
Objective To evaluate long-term seizure outcome and mesial temporal structures and to the amount of
Brussels, Belgium
3
Department of Neurology, complications of temporal lobe disconnection (TLD) lateral temporal and parahippocampal gyrus
ULB-Hôpital Erasme, Brussels, without resection for mesial temporal lobe epilepsy removals associated with the amygdalohippocampal
Belgium
4
(MTLE). resection. The results of these surgical techniques
Department of Methods Data of 45 patients operated on for are broadly similar with 60%–90% of patients
Neuropediatrics, ULB-Hôpital
Erasme, Brussels, Belgium
intractable MTLE using a functional disconnection becoming seizure-free.2 These operative procedures
procedure have been studied. Indication of TLD surgery create a large empty volume that can be the cause

Protected by copyright.
Correspondence to was retained after a standard preoperative evaluation of of complications such as bleeding, brain shifts and
Dr Nicolas Massager, refractory epilepsy and using the same criteria as for subdural collections.
Department of Neurosurgery,
standard temporal resection. In the present article, we report our experience
Clinic of Stereotactic and
Functional Neurosurgery, Results Mean follow-up duration was 3.7 years. At the with the use of an alternative surgical technique for
University Hospital Erasme, last follow-up, 30 patients (67%) were completely MTLE characterised by absence of brain tissue
Route de Lennik 808, Brussels seizure-free (Engel-Ia/International League Against resection. Actually, our technique includes only
B-1070, Belgium; Epilepsy class 1) and 39 patients (87%) remained disconnections around the temporal structures
[email protected]
significantly improved (Engel-I or -II) by surgery. Actuarial involved in the epileptogenic activity. The rationale
Received 4 September 2012 outcome displays a 77.7% probability of being seizure- of this approach was based on different concerns:
Revised 17 January 2013 free and an 85.4% probability of being significantly ▸ Removal of brain tissue is not required in nonle-
Accepted 19 March 2013 improved at 5 years. No patient died after surgery and sional MTLE as there is no risk of progression
Published Online First
18 April 2013
no subdural haematoma or hygroma occurred. of a mass lesion.
Permanent morbidity included hemiparesis, hemianopia ▸ Some disconnection procedures for the treat-
and oculomotor paresis found in three, five and one ment of intractable epilepsy, like functional
patient, respectively, after TLD. hemispherotomy, are well known and effective.
Conclusions TLD is acceptable alternative surgical ▸ Disconnection instead of resection of mesial
technique for patients with intractable MTLE. The results temporal structures could potentially reduce the
of TLD are in the range of morbidity and long-term risks of extensive microsurgical dissection in
seizure outcome rates after standard surgical resection. critical areas.
We observed a slightly higher rate of complications after ▸ The lack of brain tissue removal could avoid
TLD in comparison with usual rates of morbidity of complications related to the large parenchymal
resection procedures. TLD may be used as an alternative defect created in temporal lobectomy
to resection and could reduce operating time and the procedures.
risks of subdural collections. ▸ The size of the craniotomy and the operative
time could be reduced in comparison with the
standard temporal lobe resection.
This study analysed the long-term results of a
INTRODUCTION large series of patients operated on for MTLE
Mesial temporal lobe epilepsy (MTLE) is the most using a surgical technique of temporal lobe discon-
▸ http://dx.doi.org/10.1136/
jnnp-2013-304963 frequent form of refractory focal epilepsy in adults. nection (TLD).
In most of the patients, seizures become pharma-
coresistant after several years. Surgery has proven
To cite: Massager N,
to be very efficient in MTLE and superior to METHODS AND MATERIALS
Tugendhaft P, Depondt C, optimal medical therapy in a randomised trial.1 Study population
et al. J Neurol Neurosurg Different surgical procedures have been developed We reviewed the charts of all patients who under-
Psychiatry 2013;84: including large anterior temporal resection by went TLD in our epilepsy surgery centre. Since
1378–1383. Falconer, modified anterior temporal lobectomy by April 2002, we proposed this surgical technique to

1378 Massager N, et al. J Neurol Neurosurg Psychiatry 2013;84:1378–1383. doi:10.1136/jnnp-2012-304083

 Epilepsy

J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp-2012-304083 on 18 April 2013. Downloaded from http://jnnp.bmj.com/ on June 2, 2023 at Cote d'Ivoire:BMJ-PG Spon.
patients who are candidates for epilepsy surgery for MTLE, through the posterior end of the hippocampus. The fourth tra-
either patients with normal brain MRI or patients with mesial jectory (figure 1, blue line) begins at the anterior limit of the
temporal lobe sclerosis, as an alternative to temporal resection. temporal horn of the lateral ventricle and goes forward and
Patients with lesional temporal lobe epilepsy underwent a stand- superior to reach the deep part of the sylvian fissure.
ard temporal resection or lesionectomy to obtain histology of On the day of surgery, the patient is placed under general
resected tissue. anaesthesia in a semilateral position with the head tilted on
the contralateral side and fixed on a Mayfield clamp.
Preoperative evaluation and patient selection Correspondence between the neuronavigation system, micro-
All patients underwent a standard pre-surgical evaluation of refrac- scope and the patient’s head is performed with a pointer moni-
tory epilepsy, including clinical and neurological evaluations, tored by the camera of the tracking system. Under
neuropsychological assessment, 5-day continuous video-EEG neuronavigation, the skin is incised and the temporal lobe is
monitoring with ictal recording, structural brain MRI and positron exposed through a craniotomy rounded about 6 cm in diameter
emission tomography with [18F]-fluorodeoxyglucose.2 Wada test and centred on the superior temporal gyrus. The dura is opened
was performed in selected patients. Some patients also underwent in a C-shaped hinged anteriorly. Under microscope, we use the
invasive preoperative video-EEG monitoring using intracranial first neuronavigation trajectory to reach the anterior limit of
electrodes, either with subdural electrodes or with depth stereotac- temporal horn of the lateral ventricle through the superior tem-
tic EEG electrodes. Indication for surgery was discussed at multi- poral gyrus at a distance of 2.5 cm from the temporal tip using
disciplinary epilepsy surgery meetings for all patients at the end spatula, coagulation and section. The lateral part of the tem-
of the preoperative evaluation. All patients were extensively poral lobe is sectioned from the temporal cortex down to the
informed about the surgical procedure by their epileptologist and base and then to the tentorium; this section represents our pos-
neurosurgeon. terolateral disconnection. We cut only the arachnoid at the con-
vexity; that is, on the base and on medial structures—we let the
Surgical procedure arachnoid in place and dissect only subpially—which provides a
A neuronavigation planning is performed the day before surgery security layer to the perimesencephalic vessels and the brain-
on the basis of a recent volumetric three-dimensional MRI in a stem. A microcotton is placed at the medial end of this section.
dedicated computer station. Four trajectories of neuronavigation Into the lateral ventricle, we follow the second neuronavigation
are defined (figure 1). The first trajectory (figure 1, yellow line) trajectory from the anterior recess to the posterior part of the

Protected by copyright.
enters the superior temporal gyrus at a distance of about 25 mm temporal horn of lateral ventricle, after having placed a retractor
from the temporal tip and targets the anterior limit of the tem- upwards to recline the choroid plexus and protect the anterior
poral horn of the lateral ventricle. The second trajectory choroidal artery. At the posterior limit of the hippocampus, we
(figure 1, green line) starts from the anterior limit of the tem- dissect the white matter of the temporo-occipital junction using
poral horn of the lateral ventricle and follows the horn poster- spatula and suction, along the third neuronavigation trajectory,
iorly to the height of the posterior limit of the hippocampus. until the arachnoid of the perimesencephalic cistern through
The third trajectory (figure 1, pink line) begins at the target of which the brainstem parenchyma is visible. This section is pro-
the previous trajectory and targets the perimesencephalic cistern longed upwards until the entire posterior end of the

Figure 1 Preoperative axial MRI-T1 showing a neuronavigation planning of right temporal lobe disconnection procedure. Line 1 represents the first
trajectory, line 2 represents the second trajectory, line 3 represents the third trajectory and line 4 represents the fourth trajectory. Access the article
online to view this figure in colour.

Massager N, et al. J Neurol Neurosurg Psychiatry 2013;84:1378–1383. doi:10.1136/jnnp-2012-304083 1379

 Epilepsy

J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp-2012-304083 on 18 April 2013. Downloaded from http://jnnp.bmj.com/ on June 2, 2023 at Cote d'Ivoire:BMJ-PG Spon.
hippocampus is sectioned and then forward to reach the cotton presented in table 1. The age of the patients ranged from 7 to
previously inserted at the limit of our posterolateral disconnec- 60 years (median age 34 years, IQR 28–44); four patients were
tion. We place pieces of gelfoam along our posterior section of less than 18 years at the time of surgery. There were 25 women
the temporal lobe. Then, the fourth neuronavigation trajectory and 20 men. Surgery involved the left temporal lobe for 28
is followed to perform a section in the frontotemporal white patients and the right temporal lobe for 17 patients. Thirty-nine
matter from the anterior recess of temporal horn of the lateral patients (86.7%) had unilateral mesial temporal sclerosis on
ventricle to the deepest part of the sylvian fissure where a micro- MRI. Nine patients (20%) underwent invasive preoperative
cotton is placed. From the cortex, we dissect subpially the video-EEG monitoring using intracranial electrodes: subdural
superior temporal gyrus along the sylvian fissure until we reach electrodes for six patients and depth stereotactic EEG electrodes
the microcotton let in place previously so that the anterior part for three patients.
of the amygdalohippocampal formation is entirely disconnected.
At this time, the temporal lobe has no more neural connections
with the rest of the brain. We verify haemostasis, close the dura, Seizure outcome
fix the craniotomy piece and suture the temporal muscle, sub- All patients were followed up for a minimum period of 1 year
cutaneous tissue and skin. after surgery. The mean follow-up duration was 3.7 years and
ranged from 1 to 9.5 years. Data on the epileptic outcome are
presented in table 2. At 1 year post-surgery, 40 out of 45
Patient follow-up
patients (89%) were seizure-free. At last follow-up, using the
After surgery, patients were followed up by the neurological
Engel classification, 67% of the patients (30/45) were in Engel
team of our centre. A control MRI was performed in the follow-
class Ia, and 80% of the patients (36/45) were in Engel class
ing days after surgery to verify the completeness and to display
I. According to the ILAE classification, 67% of the patients were
the limits of the disconnection (figure 2). Since pieces of
in ILAE class 1.
gelfoam have been placed along the trajectories of resection, it
Figure 3 shows the actuarial curves for seizure outcome of
is usually easy to see the sections on the MRI in order to
our population. The actuarial epilepsy outcome showed that at
confirm the completeness of the disconnection. Postoperative
2 years after surgery, the probability of being completely seizure-
complications and permanent morbidity were recorded for all
free (ILAE 1 and Engel Ia) was 82.9% and of being significantly
patients. For all patients with surgery in the speech-dominant
improved was 91.1%. At 5 years after surgery, the probability of
temporal lobe, transient speech fluency difficulty was analysed
being completely seizure-free (ILAE 1 and Engel Ia) was 77.7%

Protected by copyright.
by a neurologist after surgery and confirmed by a logopedic
and of being significantly improved was 85.4%.
testing. We categorised the seizure outcome according to Engel
and International League Against Epilepsy (ILAE) classifica-
tions.3 The actuarial epilepsy outcome was evaluated with the Postoperative complications and permanent morbidity
Kaplan–Meier analysis. No neurological sign of significant infarction or swelling of the
disconnected tissue occurred postoperatively, except for three
RESULTS patients who complained of increasing headache without any
Forty-five patients with nonlesional intractable epilepsy under- neurological deficit 2–4 days after surgery and received corticos-
went TLD. The main demographic data of our population are teroids for few days.

Figure 2 Brain axial MRI-T1 3 months after a left temporal lobe disconnection procedure. The white arrows show areas of disconnection into the
left temporal lobe: anterior and lateral limits of disconnection (left image), posterior limits of disconnection (right image).

1380 Massager N, et al. J Neurol Neurosurg Psychiatry 2013;84:1378–1383. doi:10.1136/jnnp-2012-304083

 Epilepsy

J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp-2012-304083 on 18 April 2013. Downloaded from http://jnnp.bmj.com/ on June 2, 2023 at Cote d'Ivoire:BMJ-PG Spon.
Table 1 Demographical data of patients included in this study
Demographical data Value

Number of patients 45
Age
Median (SD) 34 years (12.4)
Range 7–60 years
<18 years 4
Sex
Male 20
Female 25
Side
Left 28
Right 17
MTS on MRI
Yes 39
No 6 Figure 3 Graph showing Kaplan–Meier estimates of percentage of
Follow-up duration probability for epilepsy control with time. Solid line represents patients in
Mean 3.7 years
International League Against Epilepsy class 1 (ILAE 1) (seizure-free since
surgery), dashed line represents patients in classes ILAE 1–4 (significantly
Range 1–9.5 years
improved). Access the article online to view this figure in colour.
MTS, mesial temporal sclerosis.

DISCUSSION
Table 3 shows the postoperative complications encountered in Disconnection of the temporal lobe as alternative
our series. No patient died in the postoperative course after to resection
TLD. No subdural haematoma developed after surgery. We The concept of isolating the area of the brain that includes the

Protected by copyright.
observed two postoperative intraparenchymal haematomas; one epileptic focus instead of resecting it is well known. Some dis-
of them required surgical drainage with subsequent full patient connective procedures for the treatment of intractable epilepsy,
recovery. No infection developed after surgery. Transient speech such as hemispherotomy or corpus callosotomy, are well estab-
fluency problems, diplopia or hemiparesis occurred in six, five lished and their results are well known.4 5 Disconnection techni-
and one patient, respectively, after surgery and fully recovered ques could possibly reduce some postoperative complications
within several days or weeks. Five patients complained of a loss related to the resection of a large area of brain and to the
of visual field related to permanent postoperative hemianopia, nearby location of the critical structures. Here, we applied this
with complete amputation of superior quadrant and inferior concept to the surgical treatment of MTLE. Isolation of mesial
quadrant in some extend. An everlasting hemiparesis occurred temporal structures from the rest of the brain is expected to
in three patients. This motor deficit improved significantly produce equivalent results in terms of seizure control compared
within days in three patients but did not recover completely. with removal of these structures.
One patient presented a permanent diplopia. This patient suf- Since 2002, we applied a surgical procedure of disconnection
fered preoperatively from a congenital strabismus without diplo- without resection of temporal lobe structures to patients with
pia due to a premature birth. pharmacoresistant MTLE. Our protocols of standard preopera-
tive evaluation, indication criteria, postoperative follow-up and
medication withdrawal, used for temporal resection in MTLE
were applied unchanged for patients operated with the tech-
nique of TLD. Few other epilepsy surgery teams have developed
Table 2 Epilepsy outcome following temporal lobe disconnection
a similar approach.6–9 In 2004, Smith et al6 published the
Seizure outcome Value results of a surgical procedure of temporal disconnection for
Number of patients 45
MTLE on 10 patients with extensive lobar or hemispheric
Outcome at last follow-up
Engel classification, n (%)
Engel Ia 30 (67)
Engel I (a+b+c+d) 36 (80) Table 3 Complications after temporal lobe disconnection
Engel II 3 (7) Complications Value
Engel III 2 (4)
Engel IV 4 (9) Number of patients 45
ILAE classification, n (%) Mortality
ILAE 1 30 (67) Postoperative complications 0
ILAE 2 0 (0) Subdural haematoma 0
ILAE 3 8 (18) Intraparenchymal haematoma (one required surgical drainage) 2
ILAE 4 3 (7) Permanent morbidity
ILAE 5 4 (9) Hemiparesis 3
ILAE 6 0 (0) Oculomotor paresis 1
Hemianopia 5
ILAE, International League Against Epilepsy.

Massager N, et al. J Neurol Neurosurg Psychiatry 2013;84:1378–1383. doi:10.1136/jnnp-2012-304083 1381

 Epilepsy

J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp-2012-304083 on 18 April 2013. Downloaded from http://jnnp.bmj.com/ on June 2, 2023 at Cote d'Ivoire:BMJ-PG Spon.
atrophy. They report a good clinical outcome (70% of the epilepsy, Cohen-Gadol et al12 found a Kaplan–Meier based
patients were seizure-free and 80% had more than 90% probability of Engel class I outcome of 76% at 2 years and 72%
decrease in seizures) and a low morbidity (one patient with at 10 years after surgery. From a series of 325 patients who
speech fluency deficit). Their follow-up period was limited to underwent anterior temporal lobectomy for epilepsy, McIntosh
2–6 years after surgery. Their surgical technique differs from et al13 reported a probability of complete seizure freedom of
our TLD procedure by several aspects. First of all, they did not 55.3% at 2 years and 41% at 10 years after surgery. Recently, de
use neuronavigation during surgery. They proposed awake Tisi et al14 related the long-term outcome of surgery for epi-
surgery for procedures performed on the language-dominant lepsy in a series of 615 patients including 497 mesial temporal
side. The limits for posterior transection margin differed resections and found an actuarial seizure-free outcome of 52%
between language-dominant and non-language-dominant lobot- at 5 years and 47% at 10 years after surgery.
omy. They removed the superior temporal gyrus and superior Complications of TLD were similar to the one observed after
temporal polar cortex; the posterior and superior amygdala as resection of the temporal lobe and included hemiparesis, oculo-
well as the anterior uncus were aspirated. So, the surgical tech- motor paresis, hemianopia and speech deficit.6 8 Previous
nique proposed by Smith et al is not a fully disconnection pro- studies have reported diplopia from third or fourth cranial
cedure but rather a mix between disconnection and resection of nerve palsy after temporal lobe surgery. Rydenhag et al15 docu-
some structures. mented oculomotor nerves palsy in 4 out of 247 after mesial
The epilepsy surgery team of Grenoble in France also devel- temporal lobe resection, that is, Behren’s et al16 in 4 out of 279
oped a surgical technique of the temporal lobe.7 They reported patients and Slanova et al17 in 7 out of 215 patients. Trochlear
in 2008, the results of a series of 47 patients with intractable paresis occurred in 19% of the cases in the series of the Mayo
MTLE operated using a disconnection technique.8 At 2 years Clinic.18 Contralateral hemiparesis, mainly attributed to injury
after surgery, 85% of the patients had an epilepsy outcome clas- to the anterior choroidal artery, occurred in 2%–5% of the
sified in Engel I, including 58% in Engel Ia. They did not report cases after surgery and improved to some extent in all cases.19
long-term follow-up. The morbidity is low: one patient with a We observed a postoperative hemiparesis in three patients,
hemiparesis, one patient with a facial paresis and one patient which is higher than usually reported in large series of temporal
with a hemianopia. Their surgical technique is very similar to resection. All these patients have actually a very limited residual
our procedure, except that the hippocampus is aspirated with deficit. We postulate that a limited ischaemic event has occurred
Cavitron and not only disconnected as we do. Actually, our dis- during the superior disconnection of the hippocampus, which

Protected by copyright.
connection procedure is the first reported in literature with a represents the last step of the procedure when the brain shift
fully disconnection technique and no resection of any part of can be significant and the neuronavigation could therefore be of
the temporal lobe, including mesial temporal structures. reduced fiability. Visual field defects after resection surgery for
In a rare case of a patient with Dandy–Walker malformation MTLE are often limited to the upper contralateral quadrant,
and parahippocampal herniation associated with an intractable and patients are unaware of their deficit.19 However, 4%–12%
MTLE, Ng et al9 performed an anterolateral temporal lobec- of the patients suffer a more extensive hemianopia that may be
tomy with hippocampal disconnection. In this case, the unusual related to individual variability in the course of the optic radia-
anatomy created a technical difficulty for resection of mesial tions as they encompass the temporal lobe.19 20 In patients of
temporal structures and justified disconnection instead of resec- our series who developed hemianopia after surgery, we found
tion. The patient was free of seizures in the following that the posterior limit of hippocampal disconnection was 0.5–
12 months. 1 cm more posterior than planned, maybe due to brain shift
during surgery. To reduce the risk of significant visual field
Results of TLD deficit after surgery in future, we intend to use tractography of
Clinical seizures are related to the propagation of epileptic dis- the optic pathways in the systematic preoperative workup of
charges from the epileptic zone to the rest of the brain. In TLD surgery21 and take into account the potential anteroposter-
MTLE, isolation without resection of mesial temporal structures ior brain shift to define the posterior limit of hippocampal dis-
would therefore theoretically have comparable seizure control connection. The main advantage of TLD compared with
rates with temporal resection. Widespread results of a surgical resection surgery could be the reduced risk of subdural haema-
procedure of TLD for epilepsy, including the long-term seizure toma or hygroma. In our series as in others,6–8 no postoperative
control rate, are unknown. In our series, we found that 89% of subdural collection has been found after TLD, which has been
the patients were seizure-free at 1 year post-surgery. In the long the case after different temporal lobe resection techniques and
term, more than 80% of the patients remained significantly sometimes required surgical drainage.12 19 However, the rate of
improved and 67% were seizure-free at last follow-up. The actu- this complication is very low: less than 1% in many series.12 19
arial probabilities of being completely seizure-free or has signifi- It is likely that the absence of subdural haematoma and
cantly improved at 5 years after surgery are 77.7% and 85.4%, hygroma after TLD is due to the avoidance of brain shift from
respectively. These results are similar with those of other series the cavity that occurred after lobectomy. This benefit was also
of disconnection for MTLE.6–8 Our series provide information reported after functional hemispherotomy instead of hemispher-
on the long-term results after TLD. ectomy.8 On the other hand, surgical disconnection does not
The epileptic outcome of our patients is comparable to the allow histological analysis of the epileptic tissue. Yet, samples
one of patients treated by resection techniques. Elwes et al10 could be taken during TLD procedures for pathological analysis.
reported 62% of the patients being seizure-free at 2 years after On the basis of a comparison between the last five surgical
‘en bloc’ anterior temporal lobectomy and 55% remaining procedures of temporal disconnection and the last five surgical
seizure-free after a median follow-up of 41.5 months. Wieser procedures of mesial temporal resection for epilepsy performed
et al11 recorded a rate of seizure-free patients of 57.1% after by our team, the surgical time was reduced by 30–45 min with
selective amygdalohippocampectomy in a series of 201 patients temporal disconnection. This is in agreement with the experi-
with nonlesional mesial temporal epilepsy. In 399 patients oper- ence of Benabid et al8 and of Smith et al6 who report a gain in
ated in the Mayo Clinic for nonlesional mesial temporal operation time of about 30 min.

1382 Massager N, et al. J Neurol Neurosurg Psychiatry 2013;84:1378–1383. doi:10.1136/jnnp-2012-304083

 Epilepsy

J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp-2012-304083 on 18 April 2013. Downloaded from http://jnnp.bmj.com/ on June 2, 2023 at Cote d'Ivoire:BMJ-PG Spon.
CONCLUSION 6 Smith JR, VanderGriff A, Fountas K. Temporal lobotomy in the surgical management
Functional disconnection of the temporal lobe instead of resec- of epilepsy: technical report. Neurosurgery 2004;54:1531–6.
7 Benabid AL, Chabardès S, Seigneuret E. Surgical Disconnections of the
tion is a safe and effective surgical treatment of pharmacoresis- epileptic zone as an alternative to lobectomy in pharmacoresistant epilepsy. In:
tant non-tumoural MTLE. The seizure outcome of this surgical Lüders HO. Textbook of epilepsy surgery. London: Informa Healthcare Ltd, 2008:
procedure is good, very comparable to the one observed after 1155–62.
resection, even in the long term. The complications of temporal 8 Chabardès S, Minotti L, Hamelin S, et al. Déconnexion du lobe temporal dans les
épilepsies temporalis pharmacorésistantes: techniques, complications et résultats.
disconnection are the one found after temporal resection. The
[article in French] Neurochirurgie 2008;54:297–302.
rate of hemianopia and hemiparesis was slightly higher than 9 Ng WH, Valiante T. Lateral temporal lobectomy with hippocampal disconnection as
usually reported after resection. The TLD technique could an alternative surgical technique for temporal lobe epilepsy. J Clin Neurosci
reduce the duration of surgery and avoid subdural collections. 2010;17:634–5.
Randomised prospective trials directly comparing TLD versus 10 Elwes RDC, Dunn G, Binnie CD, et al. Outcome following resective surgery for
temporal lobe epilepsy: a prospective follow-up study of 102 consecutive cases.
resective surgery are warranted to better define the role of a J Neurol Neurosurg Psychiatr 1991;54:949–52.
functional disconnection procedure in the treatment of refrac- 11 Wieser HG, Ortega M, Friedman A, et al. Long-term seizure outcomes following
tory nonlesional MTLE. amygdalohippocampectomy. J Neurosurg 2003;98:751–63.
12 Cohen-Gadol AA, Wilhelmi BG, Collignon F, et al. Long-term outcome of epilepsy
Acknowledgements The authors thank Frédéric Schoovaerts, Rachid Kamouni, surgery among 399 patients woth non-lesional seizure foci including mesial
Thierry Leloup and Marc Levivier for their active involvement in the surgical temporal lobe sclerosis. J Neurosurg 2006;104:513–24.
procedures of our patients and their contribution to the present manuscript. 13 McIntosh AM, Kalnins RM, Mitchell LA, et al. Temporal lobectomy: long-term
seizure outcome, late recurrence ans risks for seizure recurrence. Brain
Contributors The following persons have contributed to the present manuscript as
2004;127:2018–30.
authors according to the ICMJE guidelines for authorship: NM, PT, CD, TC, LD, NB,
14 de Tisi J, Bell GS, Peacock JL, et al. the long-term outcome of adult epilepsy
ODW, PVB, BL.
surgery, patterns of seizure remission, and relapse: a cohort study. Lancet
Competing interests None. 2011;378:1388–95.
Patient consent Obtained. 15 Rydenhag B, Silander HC. Complications of epilepsy surgery after 654 procedures in
Sweden, September 1990–1995: a multi-center study based on the Swedish
Ethics approval The present study was approved by the local ethical committee of National Epilepsy Surgery Register. Neurosurgery 2001;49:51–6.
the ULB-Hospital Erasme. 16 Behrens E, Schramm J, Zentner J, et al. Surgical and neurological complications in a
Provenance and peer review Not commissioned; externally peer reviewed. series of 708 epilepsy surgery procedures. Neurosurgery 1997;41:1–9, discussion
9-10.
17 Salanova V, Markland O, Worth R. Temporal lobe epilepsy surgery:

Protected by copyright.
REFERENCES outcome, complications, and late mortality rate in 215 patients. Epilepsia
1 Wiebe S, Blume WT, Girvin GP, et al. A randomized, control trial of surgery for 2002;43:170–4.
temporal-lobe epilepsy. N Engl J Med 2001;345:311–18. 18 Cohen-Gadol AA, Leavitt JA, Lynch JJ, et al. Prospective analysis of diplopia after
2 Massager N, Tugendhaft P, Brotchi J. Evaluation du traitement chirurgical de anterior temporal lobectomy for mesial temporal lobe sclerosis. J Neurosurg
l’épilepsie temporale réfractaire. [abstract in French] Acta Neurol Belgica 2003;99:496–9.
1996;96:169–70. 19 Sasaki-Adams D, Hadar EJ. Temporal lobe epilepsy surgery: surgical complications.
3 Wieser HG, Blume WT, Fish D, et al. Proposal of a new classification of outcome In: Lüders HO. Textbook of epilepsy surgery. London: Informa Healthcare Ltd, 2008:
with respect to epileptic seizures following epilepsy surgery. Epilepsia 1288–99.
2001;42:282–6. 20 Tecoma ES, Laxer KD, Barbaro NM, et al. Frequency and characteristics of
4 Delalande O, Bulteau C, Dellatolas G, et al. Vertical parasagittal hemispherotomy: visual field deficits after surgery for mesial temporal sclerosis. Neurology
surgical procedures and clinical long-term outcomes in a population of 83 children. 1993;43:1235–8.
Neurosurgery 2007;60:19–32. 21 Winston GP, Mancini L, Stretton J, et al. Diffusion tensor imaging tractography of
5 Tanriverdi T, Olivier A, Poulin N, et al. Long-term outcome after corpus callosotomy: the optic radiation for epilepsy surgical planning: a comparison of two methods.
a retrospective analysis of 95 patients. J Neurosurg 2009;110:332–42. Epilepsy Res 2011;97:124–32.

Massager N, et al. J Neurol Neurosurg Psychiatry 2013;84:1378–1383. doi:10.1136/jnnp-2012-304083 1383