American Epilepsy Society Plenary II

1
American Epilepsy Society Plenary II
Treatment of Epilepsy Improving Outcomes with
Innovative Surgical Techniques
2
Plenary II

• The Overview
• The theme of the symposium is innovative surgical
  techniques for epilepsy. Three topics
  (hemispheric syndromes, temporal lobe epilepsy,
  and tuberous sclerosis) will each include two
  subtopics addressed by a neurologist and a
  neurosurgeon. The talks will address 1)
  selection of candidates for hemispherectomy, 2)
  experience related to seizure outcome and
  complications of hemispherectomies performed in
  different clinical settings, and the strategy for
  choosing the best surgical approach, 3)
  challenges related to offering temporal
  resection, 4) current available data from
  surgical trials that may have bearing on this
  issue, 5) the available data concerning the roles
  of conventional MRI, scalp EEG, PET (FDG,
  flumazenil, and AMT), ictal SPECT, and
  diffusion-weighted MRI in identifying the
  epileptogenic tuber(s) for resection, and 6) the
  overall surgical strategy for patients with
  tuberous sclerosis.

3
Plenary II

• Learning Objectives
• At the conclusion of this activity, participants
  should be able to
• Identify favorable candidates for hemispherectomy
  for refractory epilepsy, and understand the
  issues related to selection of operative
  technique
• Recognize the issues related to surgical planning
  in temporal lobe epilepsy, to obtain the best
  seizure outcome with the least risk for
  postoperative memory deficit
• Appreciate strategies for developing a surgical
  plan for patients with focal epilepsy due to
  tuberous sclerosis.

4
Plenary II

• It is the policy of the American Epilepsy Society
  that all faculty participating in continuing
  medical education activities are expected to
  disclose to the audience any real or apparent
  conflicts of interest related to the content of
  the activity. Faculty disclosure has been made
  in writing and can be found in the Annual Meeting
  program book.
• Each speaker has been asked to disclose to you if
  and when unapproved products and/or unapproved
  indications will occur in his or her
  presentation.

5
Plenary II

• AES now has the Medical Education Evaluator, an
  online system to obtain CME and Pharmacy credit.
  The Evaluator will be used to verify course
  attendance, enter pre- tests and post-tests and
  complete course and activity evaluations. You use
  the pre-assigned numbers printed on your badge to
  gain access to the activity and course
  evaluations. A booklet containing paper copies of
  test and evaluation forms are available in the
  AES conference bag and will allow for note taking
  before completing the tests and evaluations
  online. 

6
Plenary II

• After logging in at AESnet.org with your badge
  number, you must enter the time spent in each
  Annual Meeting educational session, complete any
  required pre/post-tests and then complete an
  evaluation on each session you attend. The system
  will allow you to enter your information while at
  the Annual Meeting, through the Cyber Cafes or
  with a personal laptop through the wireless
  network, or at your leisure back at your office.
  When complete, the system will format a PDF
  certificate for you to download and print.

7
Evolving ConceptsIn Selection Of Hemispherectomy
Candidates Plenary II SessionAmerican Epilepsy
Society MeetingDecember 7, 2004

• Ajay Gupta, M.D.
• Cleveland Clinic Foundation
• Cleveland Ohio

8
What have we learnt so far from the published
literature on selection of candidates for
hemispherectomy?
9
50-70 Patients Who Had Hemispherectomy Are
Seizure Free

• Intractable (catastrophic) partial epilepsy
• Unilateral focal/multifocal EEG discharges
• Unilateral hemispheric abnormality
• Cortical dysplasia, Hemimegalencephaly, Stroke,
  Rasmussens encephalitis
• Low risk of new post-operative deficits
• Pre-existing hemiparesis, visual field deficit,
  cognitive delay

Kossoff EH et al. Neurology 2003 Jonas R et al.
Neurology 2004 Devlin AM et al. Brain 2003
Boongird A et al. J Neurosurg 2005 (abstract)
10
Evolving Concepts Selection of Hemispherectomy
Candidates

• Could more children benefit from hemispherectomy?

11
Children With Exclusively Generalized Scalp EEG
Discharges And Hemispheric/Focal Brain
Abnormalities Could They Be Surgical Candidates?

• Ajay Gupta, MD Elaine Wyllie, MD
• Deepak Lachhwani, MD Prakash Kotagal, MD
• William Bingaman, MD
• Cleveland Clinic Foundation
• Cleveland OH

Platform presentation at the American Academy of
Neurology, 2004 (Abstract)
12
Limitations To Localize Epileptogenic Zone In
Children

• Semiology is of limited use under 2 years
• Acharya et al, Neurology, 1997 Nordli et al,
  Epilepsia, 1997
• Scalp EEG could be non-localizing
• Infants with hypsarrhythmia and non-localizable
  ictal scalp EEG during spasms became seizure free
  after resection of congenital lesion seen on PET
  and MRI
• Chugani et al., Annals of Neurology, 1990

13
The Question

• If infants with generalized EEG pattern like
  hypsarrhythmia and a brain lesion could be
  surgical candidates,
• then why may not children with generalized
  epileptic discharges and a brain lesion be
  candidates for surgery?

14
Hypotheses

• Developmental evolution of scalp EEG patterns in
  the presence of a focal lesion in a developing
  brain is unknown
• Generalized scalp EEG discharges may represent
  progressive abnormal networking in the presence
  of a lesion acquired during perinatal or early
  postnatal life

15
Methods Pediatric Series Cleveland Clinic
2001-04

• 8 children
• Exclusively generalized or multiregional scalp
  epileptiform discharges
• No predominant epileptogenic region defined on
  neurophysiology
• Unilateral hemispheric or focal brain lesion on
  brain MRI
• Pre-surgical evaluation and surgery at Cleveland
  Clinic

16
Methods Patient Population

• Sick children with dead end
• Multiple daily seizures, status epilepticus
• Not candidates for surgery at other centers or at
  CCF during previous evaluations
• Consideration for no new post-operative deficits
• Parental understanding, informed consent
• Approved by patient management committee
• Independent evaluation by bio-ethicist

17
Katie Lennox Gestaut Syndrome

• 8 years-old, onset at 18 months
• Seizure types (no focal semiology)
• Abrupt drop attacks with injury, 10-20/day
• Slumping over, bobbing, clonic jerking of the
  head and upper body for 10-30 sec, 50-100/day
• Exam
• Severe cognitive impairment (20-40 words), left
  hemiparesis
• Failed medications
• FBM, VPA, CBZ, LTG, TPM, ZON, LEV, ESM, MSM,
  Clonazepam, Chlorazepate, B6, Ketogenic diet, VNS

18
(No Transcript)
19
Interictal EEG
20
Seizure Head nod, Fall
21
Seizure, continued
22
Katies Brain MRI
23
Katie Now 10 Years Old

• Right functional hemispherectomy (Nov 2002)
• No seizures since surgery
• On reduced antiepileptic medications
• EEG 6 months after surgery
• Diminished right hemisphere background
• Left Hemiparesis unchanged
• Parents report brighter affect

24
Results

• Age of children 2 - 24 (mean 8) years
• Seizure onset 1 day 4 (mean 2) years
• Mean sz. (loss of awareness/fall) 32/day
• Current medications 2 - 5 (mean 3.5)
• Failed meds 4 - 9 (mean 5), 4 had VNS
• Neurological Exam
• Cognitive delay All
• Hemiparesis 4
• Hemianopsia 3
• Spastic quadriparesis (asymmetric) 1

25
Results - Pre-surgical Evaluation

• Semiology Lateralization in 3 (focal motor
  component)
• EEG
• Exclusively generalized interictal and ictal
• Slow spike and wave pattern in 5
• Brain PET
• Correlated with MRI 6
• Poorly localized in 1
• SPECT Not done

26
Results

• Surgery
• Hemispherectomy 4
• Temporo-occipital/parietal resections 3
• Frontal lobe resection 1
• Etiology
• Malformations of Cortical Development 4
• Trauma (Perinatal, non-accidental) 2
• Encephalomalacia after febrile status 1
• Perinatal ischemic stroke 1

27
Results (N8)

• Mean follow up after surgery
• 16 months (11 22 months)
• Seizure outcome
• No seizures 7
• 1 patient had 2 surgeries (within 6 weeks with
  extension of posterior temporal resection)
• gt75 reduction in 1 (most disabling type)

28
Results

• Complications
• No deaths
• No new post-operative motor deficits
• Chemical meningitis and VP shunt 1
• Cognition
• Remain delayed
• Parents report developmental gains
• Improved affect and behavior

29
How could a focal or hemispheric epileptogenic
lesion give rise to exclusively generalized
epileptic discharges or slow spike and wave
patterns?

• Progressive phenomena
• Early focal EEG ictal onset Generalized
  scalp EEG patterns
• This has been shown in infants where very early
  seizures with focal EEG onset are later replaced
  by hypsarrhythmia (Chugani et al, Epilepsia 1993
  Wyllie et al, Ann Neurol 1998)

30
Discussion cont.

• Could this be an extralesional phenomenon or a
  form of Secondary epileptogenesis in a developing
  brain?
• Hypothalamic Hamartoma
• A model of progressive symptomatic generalized
  epilepsy with focal lesion where not all seizures
  originate from the hamartoma (invasive studies)
  yet patients may become seizure free after
  surgery (Freeman JL et al., Neurology 2003)

31
Conditions with exclusively generalized
progressive EEG findings in the presence of focal
lesions where seizure freedom is possible after
surgery

• Infants with hypsarrhythmia
• Hypothalamic hamartoma
• ..? Some older children with brain lesions
  (extensive) that are congenital or acquired
  during early brain development

32
Conclusions

• Our observational series suggests that even older
  children (beyond infancy) with focal or
  hemispheric brain lesions may be surgical
  candidates in the presence of exclusively
  generalized interictal and ictal epileptiform
  discharges on scalp EEG.

33
Conclusions

• Careful consideration to several complicating
  factors is required in surgical decision and is
  best done at a specialized and experienced
  Pediatric epilepsy center
• More data are required before the results of this
  small observational series are validated

34
Thank You
35
Tailoring the Hemispherectomy Improving Surgical
Outcomes

• William E. Bingaman, MD
• Cleveland Clinic Comprehensive Epilepsy Center
• Cleveland, Ohio

36
Theory

• Theoretical chance of seizure freedom for
  unilateral hemispheric epilepsy after
  hemispherectomy is 100.

37
Seizure-Free Outcomes All Etiologies

• UCLA series (2004) 71 at 2 years.
• Baltimore series (2003) 65 at 4.5 years.
• Bonn series (2002) 63 at 26 months.
• Multicenter study (1995) 70 seizure free at 6
  months.
• Personal series (2004) 76 at 33 months.

38
Reasons for Failure

• Bilateral disease (patient selection)
• Malformations of cortical development.
• Rasmussen's encephalitis.
• Peri-natal infarction (esp. IVH).
• Incomplete disconnection
• Likely most important cause of failure.
• Technical issues.
• Pathology related?

39
Hemispherectomy Techniques

• Surgical terminology is confusing.
• Basic categories include anatomic removal of
  hemisphere, functional hemispherectomy,
  hemispherotomy and variants, hemidecortication.
• Trend to smaller craniotomies and less cortical
  resection to improve peri-operative course and
  avoid long-term complications.

40
Are techniques equal?

• Anatomic
• Functional hemispherectomy
• Hemispherotomy
• Delalande
• Villemure Mascott
• Schramm
• Shimizu
• Mathern
• Decortication

41
Outcome

• Efficacy of all hemispherectomy variants reported
  to be similar (Rasmussen, UCLA, Schramm,
  Kossoff).
• Decision regarding which technique used appears
  to be surgeons training, intraoperative
  concerns, and avoidance of stormy post-operative
  course.
• No prospective, randomized comparisons exist.

42
UCLA Series

• N115
• Anatomic, Rasmussen functional hemispherectomy,
  modified lateral hemispherotomy (resect
  BG/thalamus).
• 48 cases MCD 40 received modified lateral
  hemispherotomy.
• No statistical difference in outcome based on
  technique or substrate.

• Cook S., et al. J Neurosurg 100 125-141, 2004.

43
Multicenter Review (Holthausen)

• 333 patients retrospectively studied from ten
  centers.
• Surgical technique matters with respect to
  outcome Adams modification and hemispherotomy
  associated with better outcomes over functional,
  anatomic, hemidecortication.
• 56 patients with MCD seizure free versus 82 of
  SWD.
• Seizure free outcome after hemispherotomy for
  SWD, Rasmussens encephalitis, perinatal infarct
  was 95.
• Problems with study retrospective, multiple
  centers, human factor not considered.

Holthausen et al. Seizures Post Hemispherectomy.
In Tuxhorn I, Holthausen H, Boenigk H (eds)
Pediatric Epilepsy Syndromes and Their Surgical
Treatment. 1997, pp 749-773.
44
Are there differences in outcome according to
etiology?

• Dysplasia hemimegalencephaly with worst outcome
• Holthausen, Multicenter review, 1997
• Schramm, Neurosurgery Clin NA 37 113-133, 2002.
• Carreno et al, Neurology 2001 57 331-333.
• Jonas et al., Neurology 2004 62 1712-1721.
• Kossoff et al, Neurology 2003 61 887-890.
• Maehara et al, No To Hattatsu 32 395-400, 2000.
• Dysplasia hemimegalencephaly with better
  outcomes
• Shimizu et al, Neurosurgery 47 367-373, 2000.
  (14 mcd,12 hm) 67 seizure free with
  hemispherotomy.
• DiRocco et al, Pediatr Neurosurg 2000 33
  198-207. 14/15 HM seizure free (11 anatomic
  technique).

45
Malformations of Cortical Development

• Variable presentation of dysplasia from HM to
  regional MCD.
• Each malformation individual but share certain
  surgical issues
• Involvement of subcortical tissue with absence of
  normal white matter tracts.
• Enlarged hemisphere.
• Ventricular abnormalities.
• Possibility of imaging invisible disease.

46
Hemimegalencephaly

• Enlarged hemisphere, hemiparesis, epilepsy,
  severe psychomotor delay.
• Cortical dysplasia.
• Worst developmental outcome of all children
  undergoing hemispherectomy.
• Suspected to be secondary to abnormal
  contralateral hemisphere (Rintahaka et al,
  Pediatric Neurol 921-8, 1993).
• Jahan (1997) Bilateral neuropathology in HM.
• Robain (1988) 2/4 bilateral neuropathologic
  changes.

47
Posterior frontal lobe deep subcortical
involvement
T2 hyperintensity gross architectural
distortion with relative sparing of
parietal/occipital region
Hemimegalencephaly
JM
48
Disordered Cortical/Subcortical Anatomy with HM
Patient DW
Patient JC
Patient JM
49
Hemispheric Dysplasia

• Mild or no enlargement of affected brain.
• Minimal T2 signal change.
• Minimal (no) mass effect on other hemisphere.
• Multi-lobar or holohemispheric.
• Sometimes subtle on neuroimaging.
• Less developmental delay.

50
Hemispheric Dysplasia
Patient EF
Patient ZN
Patient AH
51
Personal Experience

• 102 patients undergoing hemispherectomy January,
  1997 through October, 2004.
• Technique used initially Rasmussen functional
  hemispherectomy.
• Anatomic hemispherectomy applied later.
• Finally, tailored combinations of anatomic
  removal of most affected tissue with
  disconnection of less affected tissue.

n102, overall seizure free rate 76 at mean 33
month f/u.
52
Outcome versus Technique

• Functional 53/75 (71) seizure free.
• Anatomic 22/27 (82) seizure free.
• Majority of all patients are dramatically
  improved (Engel I,II) 86/102 (84).

53
Outcome versus Etiology

• Hemispheric dysplasia 22/30 seizure free (73).
• Hemimegalencephaly 14/17 seizure free (82).
• Rasmussens encephalitis 6/9 seizure free (67).
• Sturge Weber disease 4/4 seizure free (100).
• Perinatal Stroke 29/40 seizure free (73).
• Miscellaneous 0/2 seizure free.

54
Hemimegalencephaly

• First four cases treated with FH all with
  improved but persistent epilepsy and residual
  frontal tissue (One converted to AH with
  cessation of seizures).
• 9/13 subsequent cases treated with modified
  anatomic approach and 4 with functional technique
  (One FH underwent reop to AH with cessation of
  seizures).
• 11/17 treated with anatomic technique and all are
  seizure free. 6/17 treated with FH 50 sz free.
• Re-operated FH cases converted to AH 2/2 seizure
  free.
• Overall, 14/17 (82) seizure free with 3 off
  AEDs and 7 on monotherapy.

55
Hemimegalencephaly
Sylvian fissure
Dural opening with exposure cerebral hemisphere
56
Post resection cavity
57
Hemimegalencephaly
58
My First Hemispherectomy
DW Pre-operative MRI. R hemimegalencephaly.
59
DW Post-op MRI after functional hemispherectomy
with residual dysplastic frontal tissue.
Underwent VPS and redo x2.
60
DW Post-operative anatomic hemispherectomy. VPS
removed at last surgery.
61
JC Pre-operative MRI demonstrating R
hemimegalencephaly. Note involvement of entire
hemisphere including subcortical structures.
62
JC Post-operative Anatomic Hemispherectomy
63
Hemispheric Dysplasia

• Diffuse or multi-lobar dysplasia with pre-op
  hemiparesis, /- hemianopsia.
• 22/30 (73) seizure free overall.
• 12/15 (80) seizure free with functional
  technique.
• 10/15 (67) seizure free with anatomic technique.
• 5 re-ops with only 1 seizure free.

64
Perinatal Infarction

• All treated with functional technique.
• 29/40 (73) seizure free.
• No re-ops.
• Of 11 failures, 6 with b/l MRI findings and 4
  with b/l ictal onset.
• Patient selection and informed consent important
  in this population.

65
Rasmussens Encephalitis

• All nine treated with functional technique.
• One converted to anatomic after seizure
  recurrence (became seizure free).
• 6/9 (67) seizure free.
• 4/9 (44) with residual FL 2 with persistent
  seizures.
• 2/9 (22) with residual insula both seizure
  free.
• 2/3 failures with contralateral EEG onset
  post-op (only one known pre-op).

66
Re-operations

• Nine patients (8.8) re-operated for persistent
  seizures and residual tissue (8 FH, 1 AH).
• 7/9 re-ops in MCD/HM group
• Two HMs converted from functional to anatomic
  (residual tissue) with relief of seizures.
• Five patients with hemispheric dysplasia only 1
  seizure free from re-operation.

67
Re-operations

• One patient with Rasmussens encephalitis
  converted from functional to anatomic with
  cessation of seizures (additional insula
  resected).
• One patient with SWD and incomplete callosal
  sectioning (immediate post op recurrence of
  seizures) seizure free after completed
  callosotomy.

68
Historical Re-operation Rate

• Greatest with FH technique and dysplastic
  etiology
• UCLA (2004) 14 re-ops (12) highest in FH and
  dysplasia. None in anatomic group.
• Germany (2002) too anteriorly placed
  disconnection lines were seen.
• Mittal S. et al (2001) failure of FH due to
  residual FL dysplastic tissue. After re-op,
  patient seizure free.
• Carreno M. et al (2001) 3/13 re-ops for residual
  tissue in MCD population.
• Shimizu H. (2000) 3/34 patients (26 MCD) with
  re-op following hemispherotomy.
• Kossoff E. (2003) 5/105 re-operated for residual
  tissue. Four others with persistent seizures and
  residual tissue.

69
Success of Reoperation

• UCLA (2004) 7/14 (50) re-ops seizure free
• Cleveland (2004) 5/9 (56) re-op's seizure free
• Baltimore (2003) 1/5 (20) re-ops seizure free
• Montreal (2001) case report with MCD seizure
  free after basal FL resected.

70
Examples of Re-operations

• Splenial remnant
• Value of post-op MRI scan.
• Small amount of callosal tissue intact can lead
  to persistence of seizures.
• Frontal lobe remnant (easy to identify unclear
  significance?)

71
Incomplete Disconnection
Re-operation with splenial section seizure free
x 7 years.
72
Disconnective Hemispherectomy Frontal Lobe
Surgical landmarks during disconnection crucial.
Steep learning curve to become familiar with this
area, especially via limited exposure technique.
73
Residual FL Tissue MRI Analysis

• Arbitrarily defined as greater than 1cm of tissue
  anterior to carotid bifurcation on sagittal MRI.
• Identify whether MRI abnormality present.

74
Residual Dysplastic FL in HM
75
Tailored Hemispherectomy without Residual FL
76
Residual FL

• 18/47 (38) of MCD/HM group with gt1cm residual
  FL.
• 5/18 (28) persist with seizures 4/5 had pre-op
  dysplasia in area of residual tissue.
• 13/18 were seizure free 11/13 had no pre-op
  dysplasia in area of residual tissue.

77
Residual FL without dysplasia
78
Morbidity and Mortality

• No mortality
• Infection 3
• Hydrocephalus
• Pre-op shunts 12
• Post-op shunts (new) 9 (8.8)
• Post-op ventriculomegaly- 1
• Hemorrhagic 34 coagulopathic, 1 post-op
  hemorrhage requiring evacuation, 1
  intraventricular hemorrhage treated medically.
• Ischemic 1 return to OR for temporal lobectomy
• Aseptic meningitis 25

79
Anatomical

• Blood loss
• UCLA 688 cc average
• CCF 559 cc average
• Italy not reported
• VP shunts
• UCLA 29/37 (78)
• CCF 5/27 (19)
• Italy 4/11 (36)
• Reoperation for seizure control
• UCLA 0/37
• CCF 1/27 (4)
• Italy 0/11

80
Functional hemispherectomy

• Blood loss
• UCLA 547 cc average
• CCF 330 cc average
• VP shunts
• UCLA 3/32 (9.4)
• CCF 4/75 (5)
• Reoperation for seizure control
• UCLA 8/32 (25)
• CCF 8/75 (11)

81
Hemispherotomy

• Blood loss
• UCLA ave. 288cc
• Japan ave. 359cc (79 transfused).
• Germany 3/20 transfused
• VP shunts
• UCLA 22 new shunt rate
• Japan 15 new shunt rate
• Germany no new shunts
• Reoperation for seizure control
• UCLA 4/46 (9).
• Japan 3/34 (9)
• Germany 0/20

82
Factors in Tailoring Surgical Treatment

• Anatomy of hemisphere to be disconnected
• Presence of gray-white anomalies (distortion of
  normal anatomy)
• Hemispheric size
• Ventricular size
• Posterior basal frontal lobe dysplasia
• Corpus Callosum
• Etiology
• MCD/HM most challenging
• Body weight/age of patient

83
Modified Anatomic Hemispherectomy
Frontal/parietal dysplasia greater than posterior
quadrant, enlarged hemisphere, small ventricles,
MCD/HM etiology
Removal of Frontal-parietal-temporal lobes with
disconnection of posterior parietal-occipital
lobes
84
Summary

• Surgeon experience with steep learning curve.
• Pay attention to anatomy
• Small ventricular size, subcortical dysplasia,
  loss of white matter tracts
• Tailor the operation to accomplish the goal
• More tissue removal for HM, MCD
• Remove insula?
• Pay special attention to posterior basal frontal
  lobe particularly when dysplastic!

85
Pediatric Epilepsy Surgery Team

• William Bingaman, MD
• Atthaporn Boongird, MD
• Joann Palmer, RN
• Ann Warbel, RN

86
Pediatric Epilepsy Team

• Adina Chirla, RN
• Ajay Gupta, MD
• Prakash Kotagal, MD
• Deepak Lachhwani, MD
• Kathy Powaski, RN
• Elaine Wyllie, MD

87
Used with parental permission
88
Striking a Balance Between Memory and Seizure
Outcome

• Frank Gilliam, MD, MPH
• Professor of Neurology
• Columbia University
• New York, NY

89
(No Transcript)
90
Memory vs Seizure Outcomes

• Measures for Evaluation?
• Objective memory change
• Subjective memory change
• Function (employment, education achievement)
• Criteria for Clinical Decisions?
• Risk for objective memory change
• Risk for subjective memory change
• Risk for functional change
• Patient preference

91
Memory vs Seizure Outcomes

• Classic Problem of Statistical versus Clinical
  Significance

92
Lateral Temporal EEG and Memory
Ojemann et al. Nat Neurosci. 564-7, 2002
93
Predictors of Postsurgical Memory

• Lower Baseline Memory Scores
• MRI- Mesial Temporal Sclerosis
• WADA Memory Scores
• FDG-PET Hypometabolism

Better Memory Outcome
94
Martin et al, Arch Neurol. 591895-901, 2002.
95
Cognition in Temporal Lobe Epilepsy
Helmstaedter et al. Ann Neurol. 54425-32, 2003
96
Outcomes Research
1) The study of the results of health
services that takes patients experiences,
preferences, and values into account. 2)
Outcomes research is intended to provide
scientific evidence relating to decisions made by
all who participate in health care.
-Clancy and Eisenberg, Science, 1998
97
Subjective vs Objective Memory
Sawrie et al. Neurology. 1999 Oct
2253(7)1511-7.
98
Subjective vs Objective Memory
Table 2. Base rates of significant change on
measures of subjective and objective memory
Sawrie et al. Neurology. 531511-7, 1999.
99
Patient Preference
Memory Risk
Probability of Seizure Freedom
100
Memory vs Seizure Outcomes
Memory Change
Seizure Control
101
Standard vs. Selective Resectionvs.
RadiosurgeryWhat have we learned from
comparative trials?

• Nicholas M. Barbaro, MD
• University of California at San Francisco

102
Three measures of outcome

• Seizure outcome
• Neuropsychological outcome
• Language
• Memory
• Quality of life

103
Goals of Epilepsy Surgery

• Seizure-free
• Remove sufficient amount of functionally-irrelevan
  t tissue
• No reduction in neurological function
• Memory
• Language
• Motor, sensory, visual
• Improve QOL
• Correlates directly with seizure-free state

104
Techniques of Temporal Resection

• Temporal lobectomy
• Anatomical (standard, en bloc)
• Tailored (ECoG with or without speech mapping)
• Selective medial resection
• Amygdalohippocampectomy
• Transcortical, trans-sylvian

105
A Randomized, Controlled Trial of Surgery for
Temporal-Lobe EpilepsyNEJM 345311-318August
2, 2001 Number 5Wiebe S, Blume WT, John P.
Girvin JP, Eliasziw M Temporal Lobe Epilepsy
Study Group for the Effectiveness and Efficiency
of Surgery
106
Determinants of outcome in temporal resection

• Mesial temporal sclerosis
• MRI is best predictor
• PET
• MRS
• Normal MRI
• Best case for tailored approach

107
Outcome for Temporal Lobectomy

• Seizure-free outcome 60-90
• Improved performance IQ for non-dominant
• Significant (gt30) reduction in verbal memory
  (Boston naming) for dominant resections
• Unproven benefit of speech mapping
• Possible benefit of sparing superior temporal
  gyrus
• Smaller reduction in oral fluency (5)
• Quality of life determined by seizure outcome

108
Results of Selective Temporal Resections

• Seizure-free rates 60-85
• Disagreement on neuropsychological advantage
  (verbal memory)
• Most large series fail to show difference between
  selective and standard resection
• Possible disadvantage in pediatric cases
• Vasospasm increased in trans-sylvian approach
• QOL varies with seizure outcome

109
(No Transcript)
110
Why radiosurgery?

• Selective temporal resections are effective
• Morbidity is low, but not zero
• Infection
• Neuropsychological change
• Blood loss (intra-operative, post-operative)
• Other focal neurological deficits
• Medical contraindications of open surgery
• Some patients are afraid of surgery

111
Indirect evidence for efficacy

• Tumors
• Radiation and radiosurgery reduces seizures
• Hypothalamic hamartomas
• AVMs
• Standard AVM
• Cavernous malformations (Regis et al)

112
Pre-clinical Evidence

• Kainic acid University of Pittsburgh
  (Kondziolka)
• Spontaneous Limbic Epilepsy University of
  Virginia (Lee)

113
Direct evidence for efficacy

• Barcia Salorio et al 1994
• 11 patients treated with 10-20 Gy
• 4/11 Seizure free, 5/11 reduced, 2/11 no response
• Radiosurgical treatment in Marseille (Regis)
• Radiosurgical treatment of patients with temporal
  lobe epilepsy (24 Gy)
• European Multi-center Trial
• 65 seizure-free

114
U.S. Multi-center Trial

• Pilot Trial designed to gain preliminary evidence
  that Gamma Knife radiosurgery is safe and
  effective in reducing or eliminating seizures in
  patients with temporal lobe epilepsy
• Dose comparison in advance of Phase 3
  multi-center trial

115
U.S. Multi-center Trial

• Study Center UCSF
• Centers with at least one patient treated
• UCSF (Laxer, Garcia)
• University of Pittsburgh (Kondziolka)
• University of Virginia (Quigg)
• Columbia University (Goodman)
• Indiana University (Witt)
• University of Southern California (Heck)
• SUNY Syracuse (Beach)

• NINDS PSMB
• Dennis O. Dixon, Ph.D, chair
• Margaret Jacobs
• Ruben Kuzniecky, MD
• Quynh-Thu Xuan Le, MD
• Allen R. Wyler, MD
• Neuropathology
• Yale (de Lanerolle

116
Treatment Protocol

• Select patients with medically refractory
  temporal lobe epilepsy (best surgical treatment
  group)
• Randomize patients into 2 groups
• High dose 24Gy to 50 isodose line
• Low dose 20Gy to 50 isodose line
• Medical follow-up for 3 years

117
(No Transcript)
118
Radiosurgical Treatment

• Conformal radiation directed at temporal portion
  of the amygdala, the anterior 2cm of the
  hippocampus and adjacent parahippocampal gyrus
• Total volume within 50 isodose line between 5.5
  and 7.5cc
• Treatment isocenters 2-6

119
Radiosurgical Treatment

• Brainstem dose below 10Gy
• Optic nerve dose below 8Gy
• Shielding techniques used to achieve optimum fit
  and remain below safety limits
• All treatment plans monitored by Study Center on
  day of treatment

120
(No Transcript)
121
Preliminary Results

• Patients treated 30
• Patients at gt18 month follow-up 18
• Patients at gt12 months 29 (1 lost to f/u)

122
Typical Clinical Response

• Initial increase in auras with simultaneous
  decrease in complex partial seizures
• Headaches
• Radiological changes

123
Pre- and post-treatment(1 year) FLAIR
124
One year post radiosurgery
125
Two years post radiosurgery
126
(No Transcript)
127
(No Transcript)
128
(No Transcript)
129
(No Transcript)
130
Radiosurgery Adverse Events
131
Preliminary Neuropsychological Analysis

• Performance IQ
• improvement in non
• dominant hemisphere
• cases

132
Preliminary Neuropsychological Analysis

• No verbal memory
• decline in dominant
• hemisphere patients

133
Preliminary Neuropsychological Analysis

• No short-term affective changes
• Improved quality of life in seizure-free patients

134
Lessons from Temporal Lobe Surgery Trials

• Its the seizures, stupid
• Patients will accept mild reduction in function
  in exchange for a seizure-free state
• Verbal memory deficits are likely caused by
  surgical access route
• Radiosurgical treatment results in reduction in
  seizures in well-defined cases

135
Well-designed surgical trials are needed

• Is there a neuropsychological advantage of open
  selective resections?
• Is radiosurgery a reasonable alternative for some
  patients with temporal lobe epilepsy?
• Is there an advantage in the dominant temporal
  lobe?

136
Surgical trials continued

• Radiosurgery Phase 3 trial
• Direct comparison with temporal resection
• Attention to neuropsychological aspects
• Verbal function
• Post-operative mood changes
• Investigators meeting
• 430-600pm December 7th Sheraton Hotel
• Grand Chenier room

137
Tuberous Sclerosisidentifying epileptogenic
tubers by noninvasive techniques

• J Helen Cross
• Institute of Child Health Great Ormond Street
  Hospital for Children NHS Trust, London, UK

138
Tuberous sclerosis

• Epilepsy
• 78 in population based study
• Age dependence, first seizure
• lt12m 69 IS in one third
• 1-16 27
• gt16 4
• Cognitive outcome dependent on
• h/o seizures
• age of onset
• seizure type (IS)

Webb, Fryer Osborne 1996
139
Tuberous sclerosis

• Surgical management
• Evidence stereotyped seizure, localised focus
  from likely tuber
• Guerreiro et al Neurology 1998 51 1263-1269
• 12/18 patients focal resection 5 SF, 2 gt75
  improved
• F/up 1m-47 yrs
• Koh et al Epilepsia 2000 41 1206-1213
• 13/21 patients focal resection 9/13 seizure
  free, 1gt75 improved F/up 6-82m
• Karenfort et al Neuropediatrics 2002 33255-261
• 8/9 patients, 2 hemispherectomy, 2 SF, 5gt75
  improved
• F/up 6-52m

140
Questions

1. Is there evidence of seizures arising from a
   single focus?
2. Can a single tuber be determined to be
   responsible?
3. Can the tuber be removed without a functional
   deficit?

141
Noninvasive evaluation

• Localisation of area responsible for seizure
  onset
• Clinical evaluation
• Interictal EEG
• Ictal EEG
• how many seizures?
• MRI including FLAIR
• ?CT
• Other SPECT, PET,

142
(No Transcript)
143
(No Transcript)
144
F8-F4
F4-Fz
Fz-F3
F3-F7
100 µV
0.5 sec
T10-C6
C6-C4
C4-Cz
Cz-C3
C3-C5
C5-T9
P8-P4
P4-Pz
Pz-P3
P3-P7
ECG
145
(No Transcript)
146

147
Other techniques
SPECT

• Koh et al 1999
• 15 patients TSC, age 3m-15yrs
• EEG
  localised 10 5 focal ictal SPECT unlocalised 5
  1 focal ictal SPECT

148
Other techniques
AMT PET
Chugani et al 1998 9 children, 1-9yrs,
increase AMT uptake in 1-4 tubers in 8/9
children
Fedi et al 2003 8 patients TSC, increase uptake
in epileptogenic tuber in 4, multifocal 2
correlation with interictal spike activity
149
Other techniques
150
The role of invasive monitoring?
151
Noninvasive evaluation

• Discussion about outcome imperative
• What is the likely longterm outcome for
• Seizure control?
• Cognitive outcome?
• Behaviour?
• What are the outcome aims for the family?

152
Outcome
Lachhwani et al 2004 17 patients, 1981-2003,
f/up 12m-15 yr Localised EEG concordant MRI
(9) 8 SF (89) Localised EEG multituber MRI
(6) 2SF Non-localised EEG predominant tuber
(2) 1SF
Jarrar et al 2004 22 patients, 1986-2002, f/up
1-14 yrs 12m 5yrs Seizure
free 13/22 (59) 9/21 (42) III/IV 9/22
(41) 12/21 (57)
153
GOSH
11 children 12m 15yrs, 12m-10 yrs follow-up 8
noninvasive evaluation all EEG, MRI, 4 ictal
SPECT Outcome 12m 5/8 SF 3yrs 0/7 SF 4gt75
improved 3 invasive evaluation (for proximity to
motor cortex) Outcome 12m 2/3 SF 3 yrs 1/3
SF
154
Tuberous sclerosisnoninvasive evaluation

• Noninvasive evaluation is possible in children
  with TS
• Caution accumulation of evidence for single
  seizure type
• Invasive evaluation may still be required where
  functional concern, or uncertainty of responsible
  tuber
• Outcome aims for child will be individual and
  family should be counselled accordingly

155
Surgery in Tuberous Sclerosisthe role of
invasive monitoring in the identification of
epileptogenic tubers

• Howard L. Weiner, MD, FACS, FAAP
• Associate Professor of Neurosurgery Pediatrics
• NYU Comprehensive Epilepsy Center
• Division of Pediatric Neurosurgery
• NYU School of Medicine

156
NYU Comprehensive Epilepsy Center

• Neurology
• Orrin Devinsky, MD, Director
• Ruben Kuzniecky, MD, Co-Director
• Daniel Miles, MD
• Josiane LaJoie, MD
• Blanca Vazquez, MD
• Steven Pacia, MD
• Daniel Luciano, MD
• Catherine Schevon, MD, PhD
• Anuradha Singh, MD
• Rolando Sousa, MD
• Nandor Ludvig, MD, PhD
• Siddhu Nadkarni, MD
• Sathyashanka Subbana, MD
• Ravi Tikoo, MD

• Neurosurgery
• Werner Doyle, MD
• Neuropsychology
• Charles Zaroff, PhD
• William Barr, PhD
• Chris Morrison, PhD
• Neuro-Psychiatry
• Kenneth Alper, MD
• Melanie Shulman, MD
• Oliver Sacks, MD
• Collaborators
• Peter Crino, MD, PhD (UPenn)
• Guy McKhann II, MD (Columbia)

157
Surgery is effective in Tuberous Sclerosis

• What is the traditional approach ?
• The NYU experience with invasive
• monitoring

158
1966

• The first report of epilepsy surgery for Tuberous
  Sclerosis
• Montreal Neurological Institute
• Perot P, Weir B, Rasmussen T Arch Neurol

159

• What can we learn from the literature about the
  traditional surgical approach in Tuberous
  Sclerosis ?

160
Epilepsy Surgery for TSThe Literature

• Individual studies have shown surgery to be safe
  and effective in TS patients
• Older children/adolescents
• Single tuber/epileptogenic region
• Little reliance on invasive monitoring

161
Epilepsy Surgery for TSThe Literature

• Does the literature have an inherent bias ?
• Outcome is good if single primary focus can be
  identified and resected
• Outcome is not good if this is not the case
• (ie. any pt without single primary focus offered
  palliation, such as callosotomy)

Can one consider a different approach for TS
pts with more than one primary seizure focus ?
162
The traditional approach in TS (1)

• Operate approximately at age 10 (based on 81
  patients reported in 9 papers since 1989)

163
The traditional approach in TS (1)

• Operate approximately at age 10 (based on 81
  patients reported in 9 papers since 1989)
• Limitation Much of development occurs before
  this age
• ARE WE UNDERESTIMATING THE POTENTIAL BENEFIT OF
  SURGERY IF IT IS NOT PERFORMED AT AN EARLY AGE ?

164
The traditional approach in TS (1)

• Take Home Point 1 Identify potential surgical
  candidates at an early age

165
The traditional approach in TS (2)

• Identify single epileptogenic focus based on
  correlation between EEG and imaging

166
The traditional approach in TS (2)

• Identify single epileptogenic focus based on
  correlation between EEG and imaging
• Limitation this is often very difficult in TS
  patients
• DO WE EXCLUDE CHILDREN WHO DO NOT MEET THIS
  CRITERIA ?

167
Multiple bilateral tubers
168
The NYU approach in TS (2)

• Take Home Point 2 identify potential surgical
  candidates with bilateral electrode survey

169
The traditional approach in TS (3)

• Resect primary tuber plus surrounding
  epileptogenic zone

Pre-op
Post-op
170
The traditional approach in TS (3)

• Resect primary tuber plus surrounding
  epileptogenic zone
• Limitation defining the margins of the
  epileptogenic zone is often difficult in TS
  patients
• WILL SURGERY FAIL DUE TO AN ADJACENT OR DISTANT
  TUBER/EPILEPTOGENIC ZONE ?

171
The Challenge of Localization in Tuberous
Sclerosis
172
The NYU approach in TS (3)

• Take Home Point 3 Identify unrecognized adjacent
  or distant tuber/epileptogenic zones with
  multi-stage surgery

173
The Challenge of Localization in Tuberous
Sclerosis
174
Epilepsy Surgery for Children with Tuberous
Sclerosis

• We have utilized a novel surgical approach using
  invasive monitoring, which is frequently
  multi-staged and bilateral

175
What is the traditional epilepsy surgery
approach ?

• Operation 1 Grid/strip/depth electrode
  placement
• Operation 2 Resection of seizure focus/lesion

176
Multi-stage approach

• Operation 1 Grid/strip/depth electrode
  placement
• Operation 2 Resection of seizure
  focus/lesion/ REPLACE grid/strip
• Operation 3 Resection of seizure focus/lesion

177
Multi-stage Epilepsy Surgery

• Multi-stage invasive monitoring can detect
  residual adjacent or distal epileptogenesis, and
  is especially useful when pre-surgical data
  suggest eloquent cortex or bi-hemispheric
  involvement
• Bauman, Feoli, Romanelli, Doyle, Devinsky,
  Weiner, in press, Neurosurgery

178
5 year old boy with TS

• Onset of seizures age 2 months
• Infantile spasms age 9 months
• Multiple seizures daily- CP, T, GTC
• Significant developmental delay with subsequent
  regression

179
Previous treatment and evaluation

• Failed 10 AEDs, VNS and ketogenic diet
• MRI, EEG and VEEG multi-focal
• Not considered a surgical candidate
• Developmental delay with regression
• Poor quality of life for child and family

• NYU Plan a bilateral strip survey to identify
  one or two predominant seizure foci that could be
  further studied with grid/strip coverage.

180
Multiple bilateral tubers
181
Bilateral Strip StudyVertex Craniotomy
182
Stage I
183
(No Transcript)
184
(No Transcript)
185
(No Transcript)
186
(No Transcript)
187
Stage I
8 days of monitoring
Interictal
Seizure Onset
Seizure Spread

• Seizures with right hemisphere onset 77

• Seizures with left hemisphere onset 17

188
Resection of right frontal tuber/seizure focus
189
Replacement of electrodes after resection at 2nd
stage
190
MRI scan after 2nd stage
Pre-op
191
Stage II
Interictal
Seizure Onset
Seizure Spread
192
Surgical Outcome over 1 Year
The surgery has changed our lives. Thank
you!!! Glad we could make your day. You've made
our lives!(And if things start up again, we
know we have options...) Is he benefiting from
surgery????? His life--our lives--have completely
changed!! Every single day we think of you and
your team and think how thankful and lucky we
are.Since his surgery, he's had a handful of
very mild seizures--nothing like he had
before--when he was having an average of 6 huge
seizures a day. (At times he would have stretches
of 14 strong seizures a day. Diastat daily. The
works.) Never in his life has he had such relief!
193
Surgical Outcome

• I just wanted to let you know that he is
  still doing really well.
• No seizures and he's happy, happy. As you can
  see in this picture.

194
NYU TS Epilepsy Surgery in Children

• 21 children
• 7 boys
• 14 girls
• Mean age 3.6 yrs
• range 7 months - 16 years
• 66 total operations
• 3-Stage18 pts
• 2-Stage 3 pts
• 3-Stage at re-op 2 pts
• Mean follow-up 24 months
• range 3 months - 6 years

195
NYU TS Epilepsy Surgery

• 6 children prior surgery
• 1 resection
• 5 VNS
• 12 children had bilateral electrode studies
• 5 bilateral strips
• 7 gridcontralateral strips

196
NYU TS Epilepsy Surgery
2 pts not improved at initial follow-up?
underwent re-operation one year later and now
Engel I outcomes
197
NYU TS Epilepsy Surgery

• 5 pts whose seizures could not be lateralized
  were found to have unilateral primary focus on
  bilateral strip survey
• 4 are seizure free
• 18/21 pts underwent resections of 2 or more
  tubers
• 15/21 pts underwent multi-lobar resections
• 2 pts underwent bilateral resections

198
Re-operation in 2 children

• 5-year-old girl seizure-free x 3 months
• 4-year-old girl not improved after surgery
• Both are now seizure free after extension of
  original resections

199
Complications

• 1 pt infection (2-stage)
• 5 pts transient hemiparesis
• recovery within 2 months time
• 3 partial bone flap resorption
• 2 required repair

200
What are we learning about surgery for TS ?

• Take Home Point 1 Identify potential surgical
  candidates at an early age
• Take Home Point 2 Identify potential surgical
  candidates with bilateral electrode survey
• Take Home Point 3 Identify unrecognized adjacent
  or distant tuber/epileptogenic zones with
  multi-stage surgery

201
Surgery in Tuberous Sclerosis

• Multiple or bilateral epileptogenic foci are not
  necessarily a contraindication to surgery in
  selected patients.
• The multi-stage surgical approach has been useful
  in identifying both primary and secondary
  epileptogenic zones in TS patients with multiple
  tubers.
• Invasive EEG may reveal epileptogenic zones that
  are more extensive than the radiographic lesions.
• Selected children with TS may benefit from a more
  aggressive surgical approach.
• Long-term follow-up will determine whether this
  approach has durable effects.
• Better methods for identifying the epileptogenic
  zone, both non-invasive and invasive, are needed.

202
Surgery in Tuberous Sclerosis

• How does removal of the primary seizure focus
  alter the remaining network ?
• What can our observations about surgery in TS
  teach us about epilepsy surgery in general ?
• Is TS really a model system for lesional epilepsy
  ?

203
Special Thanks to

• Dr. Chad Carlson
• Fellow
• NYU Comprehensive Epilepsy Center