Pediatric Neurology 51 (2014) 800e805
Contents lists available at ScienceDirect
Pediatric Neurology journal homepage: www.elsevier.com/locate/pnu
Original Article
Corpus Callosotomy for Childhood-Onset Drug-Resistant Epilepsy Unresponsive to Vagus Nerve Stimulation Ravindra Arya MD, DM a, *, Hansel M. Greiner MD a, Paul S. Horn PhD a, b, Michele Turner APRN a, Katherine D. Holland MD, PhD a, Francesco T. Mangano DO c a
Comprehensive Epilepsy Center, Division of Neurology, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio Division of Epidemiology and Biostatistics, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio c Division of Pediatric Neurosurgery, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio b
abstract PURPOSE: Corpus callosotomy and vagus nerve stimulation are common palliative options for people with drugresistant epilepsy when resective epilepsy surgery is not feasible. Because most of the published corpus callosotomy experience comes from a period before vagus nerve stimulation was approved and widely used, there is a paucity of data about efficacy of corpus callosotomy in patients with inadequate response to vagus nerve stimulation. METHODS: We report seven patients who had complete corpus callosotomy after an inadequate response to vagus nerve stimulation. At the time of surgery, these patients had failed a median of six antiseizure medications, three patients also had failed a trial of ketogenic diet, and all the patients had a vagus nerve stimulation implanted for a mean duration of 2.5 years with maximal tolerated settings. RESULTS: There was a decrease in total daily seizure frequency of 34.7% (94.7; median, 71.4%; interquartile range, 55.3) after corpus callosotomy at a mean follow-up of 2.6 years (1.4). One patient achieved complete seizure freedom and five patients had 50% reduction in seizure frequency. Six patients continued to have partial-onset seizures though the frequency was decreased. Drop attacks and tonic seizures stopped in all the patients. CONCLUSION: Seizure outcomes after corpus callosotomy in our series are most likely a result of complex dynamic interaction between the natural history of epilepsy, the effect of the surgery, ongoing vagus nerve stimulation modulation, and modification in antiseizure drugs. Our study supports the clinical decision to try corpus callosotomy in patients having nonlateralizing drugresistant epilepsy with inadequate response to vagus nerve stimulation. Keywords: corpus callosotomy, vagus nerve stimulation, drug-resistant epilepsy, seizure outcomes
Pediatr Neurol 2014; 51: 800-805 Ó 2014 Elsevier Inc. All rights reserved.
Introduction Conflicts of Interest: The authors declare no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. Ethical Approval: The study was approved by Institutional Review Board of Cincinnati Children’s Hospital Medical Center (IRB: 2008-1006). The study was conceived, and data were extracted by R.A. and H.M.G. K.D.H. created and maintains the epilepsy surgery database. P.S.H. helped with data management and analysis. All surgeries were performed by F.T.M. and follow-up evaluations were performed by M.T. The article draft was prepared by R.A., which was critically reviewed and approved by all authors.
Article History: Received June 20, 2014; Accepted in final form September 12, 2014 * Communications should be addressed to: Dr. Arya; Comprehensive Epilepsy Center; Division of Neurology; Cincinnati Children’s Hospital Medical Center; E4-145, ML 2015, 3333 Burnet Avenue; Cincinnati, Ohio 45229. E-mail address: [email protected] 0887-8994/$ - see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.pediatrneurol.2014.09.008
In people with drug-resistant epilepsy (DRE), when it is not possible to identify a resectable core epileptic network, it becomes pertinent to offer palliative surgical options. These options have conventionally included vagus nerve stimulation (VNS) and corpus callosotomy. However, there is a lack of well-defined patient selection criteria for either technique because of lack of comparative studies generating data about differential efficacy or adverse effects.1 Usually, VNS is tried first in relatively older children with multiple seizure types whereas corpus callosotomy is recommended for relatively younger children with predominant drop attacks. We present a series of patients with nonlateralizing DRE with multiple seizure types who had complete corpus
R. Arya et al. / Pediatric Neurology 51 (2014) 800e805
callosotomy after failing multiple antiseizure medications and VNS. Most of the larger corpus callosotomy series were reported before VNS was approved by the Food and Drug Administration and widely used.2 Hence, there is a paucity of experience with corpus callosotomy in patients with inadequate response to VNS. This case series is likely to contribute towards the evidence for efficacy of corpus callosotomy in such patients. Methods The patients were identified by a retrospective chart review using an institutional epilepsy surgery electronic database. All patients with DRE who had a corpus callosotomy between January 2007 and December 2012 were screened for those with a history of prior VNS implantation. The usual noninvasive presurgical evaluation at our institution includes a detailed history, physical and neurological examination, video electroencephalograph (EEG) monitoring to capture sufficient number of habitual seizures, magnetic resonance imaging (MRI) of the brain including three-dimensional volumetric acquisition, interictal flurodeoxyglucose positron emission tomography, ictal and interictal single-photon emission computed tomography (SPECT), magnetoencephalography, and neuropsychologic evaluation. The positron emission tomography and SPECT data are routinely postprocessed using statistical parametric mapping and subtraction ictal SPECT coregistration to MRI (SISCOM) respectively. The information about each patient is synthesized and reviewed in a multi-disciplinary conference (Table 1). None of the patients included in this report had any prior resective epilepsy surgery or partial callosotomy, and all underwent complete corpus callosotomy. As VNS is a palliative procedure, we did not use conventional criteria for complete seizure freedom to define treatment failure.3 In cognizance of the fact that 50% reduction in seizure frequency is a common end point in clinical trials on DRE,4 we defined inadequate response to VNS as failure to achieve 50% reduction in preimplant seizure frequency in spite of maximum tolerated settings. The primary outcome measure was the reduction in total daily seizure frequency from presurgery to last available follow-up visit. Seizure frequency was determined based on parental reporting only. Secondary outcomes included reduction in the frequency of individual seizure types and number of seizure-related admissions during followup. Seizure outcome was also classified on the International League Against Epilepsy outcome scale based on seizure days. Data were extracted on to a Microsoft Excel (Microsoft Corporation, Redmond, WA) spreadsheet, and descriptive statistics were calculated. The study was approved by the Institutional Review Board of Cincinnati Children’s Hospital Medical Center.
Results
During the study duration, 29 patients underwent corpus callosotomy at our institution. Of these, seven patients (four males) were identified who had an inadequate response to a previous VNS (Table 1). Five patients (71.4%) had seizure onset in infancy (median, 6 months; interquartile range [IQR], 9). All patients had two seizure types including atonic, tonic, myoclonic, and generalized tonicclonic seizures (Table 1). The seizure frequency varied widely from one to two seizures/week to hundreds/day. All the patients had moderate to severe intellectual disability. At the time of surgical referral, the patients had failed a median of 6 (IQR, 5.5) antiseizure medications, with 3 also having failed a trial of ketogenic diet. Video-EEG revealed bilateral multifocal and/or diffuse seizure onsets in all the patients. A specific etiology could be established in two patients with one having a right frontal cortical dysplasia
801
(with discordant data from other modalities) and the other having bilateral encephalomalacia as a consequence of herpes encephalitis. One patient had a static encephalopathy with an interstitial duplication of 857 kb at 17q25.1 of unknown clinical significance. MRI was able to detect abnormalities in two other patients, although their etiologic significance was uncertain. The functional imaging modalities revealed nonlateralizing discordant data in all patients (Table 1). The mean time interval between VNS implantation and corpus callosotomy was 2.5 years (1.43). The maximal output current varied from 1.75 to 3 mA and six of seven patients were on conventional cycling (Table 2). Seizure outcomes
After a mean follow-up of 2.6 years (1.4), there was a mean reduction of 34.7% (94.7; median, 71.4%; IQR, 55.3) in the total daily seizure frequency (one-sided paired t test, P ¼ 0.0834). One patient was completely seizure free at 2.8 years of follow-up and four other patients had a 50%90% decrease in seizure burden. One patient experienced only mild reduction in seizure frequency (15.5%) and another, increase in seizure frequency (Fig 1). Six of the seven patients continued to have partial-onset seizures with or without generalization, although the frequency was decreased as described previously. Drop attacks and tonic seizures were abolished altogether in the respective patients (n ¼ 2 each). Episodes of staring and behavioral arrest ceased in one of the two patients; however, myoclonic seizures continued in the patient who had them at baseline (Fig 2). Most of the patients had either no (n ¼ 4) or only one (n ¼ 2) seizure-related admission to emergency or inpatient services, but one patient had six such admissions during 3.6 years of follow-up because of persistence of generalized seizures, in spite of substantial reduction in atonic drop attacks and tonic seizures. Because it is infrequent to achieve seizure freedom with corpus callosotomy, we have described the seizure-free patients below. Seizure-free patient
This 18-year-old girl with onset of seizures at 3 years of age had multiple seizure types including events with tremor-like movements of upper extremity and eyelid fluttering occurring in five to six daily clusters consisting of tens-to-hundreds of brief seizures. Sometimes, longer clusters would evolve into a generalized clonic seizure with eyes rolled back and postictal pallor and/or cyanosis. Other events consisted of staring, sometimes with upward gaze deviation, and pallor. She had failed nine antiseizure medications and had a VNS implanted at 11 years of age with 1year trial at maximally tolerated settings. She functioned cognitively at a 5-year-old level and had multiple behavioral issues. Her EEG seizure onsets were characterized by diffuse bilateral low-voltage fast activity (Table 1). Two weeks after complete corpus callosotomy, she experienced a decrease in seizure duration and frequency and reported improvement in behavior. After 8 weeks, seizures reduced to about 1-2 clusters/week, lasting 5-10 minutes, vs 15-40 minutes at baseline. Ten months postcallosotomy, she had a generalized seizure requiring an emergency visit. After that, seizure frequency further decreased, and she has remained seizure
802
R. Arya et al. / Pediatric Neurology 51 (2014) 800e805
TABLE 1. Outcomes and Demographic Details of Patients With Corpus Callosotomy After Failure of Vagus Nerve Stimulation
Patient
Sex
Age at Seizures Onset, mo
Seizure Types
Average Seizure Frequency
Etiology
MRI
1 2 3 4 5 6
M M F F F M
6 18 6 6 36 5
Atonic, tonic, myoclonic, GTCS Tonic progressing to GTCS Tonic or dystonic, þ/ GTCS Head and eye version, þ/ GTC Clonic, þ/ GTCS Tonic, staring
100/day 20/mo 8-10/wk 1-2/wk 100/day 6-8/wk
7
M
12
Atonic, tonic, GTCS staring
10-12/day
Unknown Unknown Unknown Cortical dysplasia Unknown Postherpes encephalitis Static encephalopathy*
Normal Nonspecific WM signal changes Dysgenesis of CC Loss of gray-white differentiation Normal Bilateral encephalomalacia Normal
Abbreviations: AED ¼ Antiepileptic drug MEG ¼ Magnetoencephalography CC ¼ Corpus Callosum MRI ¼ Magnetic resonance imaging F ¼ Female PET ¼ Positron emission tomography GTCS ¼ Generalized tonic-clonic seizures RF ¼ Right Frontal ILAE ¼ International League Against Epilepsy RT ¼ Right Temporal KD ¼ Ketogenic Diet SISCOM ¼ Subtraction Ictal SPECT Co-registered to MRI LH ¼ Left Hemisphere SPECT ¼ Single-photon emission computed tomography LT ¼ Left Temporal SPM ¼ Statistical parametric mapping LP ¼ Left Parietal WM ¼ White Matter M ¼ Male * Interstitial duplication of 857 kb of DNA at 17q25.1 detected by microarray of unknown clinical significance.
free for last 15 months on valproate, topiramate, and clonazepam. Disconnection syndromes
Three patients experienced transient postoperative deficits consistent with supplementary motor area (SMA) syndrome: reduced spontaneous speech and lack of interest in surrounding environment. Additionally, two patients respectively had weakness in nondominant leg and nondominant hand automatisms characterized by repetitive grasping movements. The manifestations completely resolved by 2 weeks in two patients and 5 weeks in another patient. Discussion
This study supports the efficacy of corpus callosotomy in patients with DRE who have failed VNS. Five of seven patients in this series had 50% reduction in seizure frequency, and one patient achieved complete seizure freedom (Table 1). These data agree with published experience, which has indicated the seizure freedom rates after corpus callosotomy to vary from 0% to 10%.5-8 In the second worldwide survey of epilepsy surgery programs, corpus
callosotomy was reported to have seizure-free outcome in 7.6% of the patients and an improvement in overall seizure burden in 60.9%.9 There was complete cessation of drop attacks and tonic seizures in two patients each. Six of the seven patients continued to have partial-onset seizures, although the frequency and duration were reduced. One of two patients experienced a remission of seizures with staring and behavioral arrest, whereas myoclonic seizures continued in the only patient that had those (Fig 2). This agrees with prior experience that drop attacks resulting from both tonic and atonic seizures are abolished by corpus callosotomy.6,7,10 The incidence of seizure-free outcome for drop attacks has varied from 51% to 90% across studies, depending primarily on patient selection.1,11 Additional studies have reported 90% reduction in seizure frequency in 85%-92% of patients.7,12 The reports of success with other seizure types have been more variable. Initially myoclonic seizures were believed to be poorly responsive to corpus callosotomy based on Yale experience, but later reports indicated better outcomes.7,13,14 Generalized tonic-clonic and atypical absence seizures were reported to significantly improve in 55% and 73% patients, respectively, in one study.15 The older smaller studies with 10 patients found higher rates of improvement but probably represent a biased cohort.2 Among the partial-onset seizures,
TABLE 2. Maximal Vagus Nerve Stimulator Settings and Interval Between Stimulator Implantation and Callosotomy
VNS Settings
Patient #1
Patient #2
Patient #3
Patient #4
Patient #5
Patient #6
Patient #7
Output current, mA Frequency, Hz Pulse width, mseconds Time on, seconds Time off, min Magnet stimulation, mA Magnet time, seconds Duration between VNS and callosotomy, yr
2.25 20 250 20 3 2.5 60 4.75
1.75 30 250 30 5 2 60 1.5
1.75 30 250 30 1.8 2 60 2.25
2 20 250 30 1.1 2.25 60 1.42
2.25 20 250 30 1.1 2.5 60 1.42
3 20 250 7 0.3 3.25 60 1.83
2 20 250 30 1.8 2.25 30 4.33
Abbreviation: VNS ¼ Vagus nerve stimulation
R. Arya et al. / Pediatric Neurology 51 (2014) 800e805
803
TABLE 1. (Continued)
PET/SPM
SPECT/SISCOM
MEG
Normal Normal LH LF, LT, LP Nonlateralizing LH
Nonlateralizing Nonlateralizing LF Nonlateralizing Bilateral multiple
Bilateral sources RF, RT LH Bilateral sources LH
-
-
-
#-AED Failed KD 3 6 8 3 9 3 20
Follow-up, % Reduction in Seizure ILAE Outcome Seizure-related yr Frequency Admissions
Failed Failed
3.61 3.90 1.20 3.59 2.79 2.89
50.0 71.4 15.5 170.0 100.0 86.0
4 4 4 4 1 4
6 1 0 0 1 0
Failed 0.33
90.0
4
0
callosal section is believed to prevent rapid bilateral synchrony of seizures originating in frontal lobe, whereas temporal lobe seizures are probably not affected by anterior callosotomy.5,16 All the patients in this series had complete corpus callosotomy. In certain patients, anterior callosotomy sparing the splenium is preferred, to preserve some interhemispheric transfer of sensory information and decrease the risk of nondominant apraxia and agnosia.10 However, there is evidence of relatively lower efficacy of anterior
callosotomy for seizure reduction compared with complete corpus callosotomy, particularly for patients with multiple seizure types in addition to drop attacks.5,17 Complete cessation of atonic and tonic drop attacks was found in 91% of pediatric patients by total callosotomy but in only 67% by partial callosotomy in one study.18 In patients with preexisting moderate-to-severe cognitive impairment, like in present series, the benefit of greater seizure reduction outweighs the risk of disconnection syndromes.5 The mechanism of seizure reduction after callosal section is not
FIGURE 1. Seizure frequency at each time point expressed as a percentage of baseline seizure frequency (taken as 100%). (The color version of this figure is available in the online edition.)
804
R. Arya et al. / Pediatric Neurology 51 (2014) 800e805
FIGURE 2. Breakup of patients according to individual seizure types at baseline and at last available follow-up. (The color version of this figure is available in the online edition.)
fully known. It has been hypothesized that achieving a critical degree of synaptoreduction with decreased number of neuronal connections may be responsible for the decrease in seizure frequency and altered seizure pattern.15 The persistence of most generalized seizures to some extent after corpus callosotomy suggests a role for other interhemispheric pathways for spread of ictal activity.5 The seizure outcomes were reported in our study at a mean follow-up duration of 2.6 years (1.4). It is known that seizure frequency after corpus callosotomy varies as a function of the length of follow-up. A meta-analysis found that in the long term (5 years), only 35% (95% confidence interval, 26-44) of the patients continued to remain free of drop attacks.19 The age at surgery varied from 6-17 years in our patients, except for one patient who was 25 years old at the time of corpus callosotomy. This patient experienced 90% reduction in seizure burden in short term. There is some evidence that younger age at surgery (18 years) is an independent predictor of improvement in neuropsychologic and quality-oflife outcomes.12,20 However, the association is equivocal for seizure outcome. Some studies have not found age at surgery to have prognostic significance; others have reported similar seizure outcomes in children compared with adults.12,21-23 Three patients in our series experienced transient postoperative deficits resembling the SMA syndrome. This confirms to the observation that SMA syndrome is more prominent and prolonged after 1-stage total callosal section.24 In general, however, the incidence of acute disconnection syndromes and long-term adverse effects are reported to be lower in children, especially those 12 years of age.10,25 All the patients included in this report had moderateto-severe cognitive impairment, which precluded testing for other relatively subtle chronic disconnection syndromes.24 Seizure outcomes after corpus callosotomy in our series are most likely a synergistic effect of the surgery itself, ongoing VNS modulation and modification in antiseizure drugs. Regarding efficacy of combining nonpharmacologic
therapies, a previous study has indicated 50%-responder rate of 70% in 30 children after 3 months of combined use of VNS and dietary therapy.26 Whereas VNS is known to provide accruing seizure reduction with increased duration,27 corpus callosotomy often makes an immediate impact on seizure frequency. Three of the five responders in our series had a clinically relevant reduction in seizure burden in first few months (Fig 1). This is valuable in the management of childhood epileptic encephalopathy, where corpus callosotomy may prevent the need for escalating antiseizure medication requirements and frequent emergency department visits. We believe that it is the short-term reduction in seizure burden that represents the true callosotomy effect, whereas the long-term outcomes are more representative of a complex interaction between the natural history of epilepsy and various treatment measures. Overall, this study supports the clinical decision to try corpus callosotomy in patients with nonlateralizing DRE who have had inadequate response to VNS. Although an important limitation of this study is the lack of neuropsychologic and quality-of-life outcomes, it emphasizes the need for studying comparative efficacy and safety of palliative treatments for patients with DRE who are not suitable for resective epilepsy surgery. R.A. receives research support from NIH (NINDS 2U01 NS045911) and American Epilepsy Society and Epilepsy Foundation of America Infrastructure award (pSERG); K.D.H. receives research support from NIH (NINDS R01 NS062756 [principal investigator]).
References 1. Rosenfeld WE, Roberts DW. Tonic and atonic seizures: what’s nexteVNS or callosotomy? Epilepsia. 2009;50(Suppl 8):25-30. 2. Spencer DD, Spencer SS. Corpus callosotomy in the treatment of medically intractable secondarily generalized seizures of children. Cleve Clin J Med. 1989;56(Suppl Pt 1):S69-78. discussion S9e83.
R. Arya et al. / Pediatric Neurology 51 (2014) 800e805 3. Kwan P, Arzimanoglou A, Berg AT, Brodie MJ, Allen Hauser W, Mathern G, et al. Definition of drug resistant epilepsy: consensus proposal by the ad hoc Task Force of the ILAE Commission on Therapeutic Strategies. Epilepsia. 2010;51:1069-1077. 4. Arya R, Glauser TA. Pharmacotherapy of focal epilepsy in children: a systematic review of approved agents. CNS Drugs. 2013;27:273-286. 5. Asadi-Pooya AA, Sharan A, Nei M, Sperling MR. Corpus callosotomy. Epilepsy Behav. 2008;13:271-278. 6. Shimizu H, Maehara T. Neuronal disconnection for the surgical treatment of pediatric epilepsy. Epilepsia. 2000;41(Suppl 9):28-30. 7. Cukiert A, Burattini JA, Mariani PP, Camara RB, Seda L, Baldauf CM, et al. Extended, one-stage callosal section for treatment of refractory secondarily generalized epilepsy in patients with Lennox-Gastaut and Lennox-like syndromes. Epilepsia. 2006;47:371-374. 8. Hanson RR, Risinger M, Maxwell R. The ictal EEG as a predictive factor for outcome following corpus callosum section in adults. Epilepsy Res. 2002;49:89-97. 9. Engel Jr J. Surgery for seizures. N Engl J Med. 1996;334:647-652. 10. Wong TT, Kwan SY, Chang KP, Hsiu-Mei W, Yang TF, Chen YS, et al. Corpus callosotomy in children. Childs Nerv Syst. 2006;22:999-1011. 11. Sunaga S, Shimizu H, Sugano H. Long-term follow-up of seizure outcomes after corpus callosotomy. Seizure. 2009;18:124-128. 12. Maehara T, Shimizu H. Surgical outcome of corpus callosotomy in patients with drop attacks. Epilepsia. 2001;42:67-71. 13. Spencer SS. Corpus callosum section and other disconnection procedures for medically intractable epilepsy. Epilepsia. 1988;29(Suppl 2):S85-99. 14. Spencer SS, Spencer DD, Sass K, Westerveld M, Katz A, Mattson R. Anterior, total, and two-stage corpus callosum section: differential and incremental seizure responses. Epilepsia. 1993;34:561-567. 15. Madsen JR, Carmant L, Holmes GL, Black PM. Corpus callosotomy in children. Neurosurg Clin N Am. 1995;6:541-548. 16. Purves SJ, Wada JA, Woodhurst WB. Corpus callosum section for complex partial seizures. In: Reeves AG, Roberts DW, eds. Epilepsy and the Corpus Callosum 2. Springer; 1995:175-182.
805
17. Spencer SS, Spencer DD, Williamson PD, Sass K, Novelly RA, Mattson RH. Corpus callosotomy for epilepsy. I. Seizure effects. Neurology. 1988;38:19-24. 18. Shimizu H. Our experience with pediatric epilepsy surgery focusing on corpus callosotomy and hemispherotomy. Epilepsia. 2005; 46(Suppl 1):30-31. 19. Tellez-Zenteno JF, Dhar R, Wiebe S. Long-term seizure outcomes following epilepsy surgery: a systematic review and meta-analysis. Brain. 2005;128:1188-1198. 20. Turanli G, Yalnizoglu D, Genc-Acikgoz D, Akalan N, Topcu M. Outcome and long term follow-up after corpus callosotomy in childhood onset intractable epilepsy. Childs Nerv Syst. 2006;22: 1322-1327. 21. Kim DS, Yang KH, Kim TG, Chang JH, Chang JW, Choi JU, et al. The surgical effect of callosotomy in the treatment of intractable seizure. Yonsei Med J. 2004;45:233-240. 22. Mamelak AN, Barbaro NM, Walker JA, Laxer KD. Corpus callosotomy: a quantitative study of the extent of resection, seizure control, and neuropsychological outcome. J Neurosurg. 1993;79: 688-695. 23. Rougier A, Claverie B, Pedespan JM, Marchal C, Loiseau P. Callosotomy for intractable epilepsy: overall outcome. J Neurosurg Sci. 1997;41:51-57. 24. Jea A, Vachhrajani S, Widjaja E, Nilsson D, Raybaud C, Shroff M, et al. Corpus callosotomy in children and the disconnection syndromes: a review. Childs Nerv Syst. 2008;24:685-692. 25. Reutens DC, Bye AM, Hopkins IJ, Danks A, Somerville E, Walsh J, et al. Corpus callosotomy for intractable epilepsy: seizure outcome and prognostic factors. Epilepsia. 1993;34:904-909. 26. Kossoff EH, Pyzik PL, Rubenstein JE, Bergqvist AG, Buchhalter JR, Donner EJ, et al. Combined ketogenic diet and vagus nerve stimulation: rational polytherapy? Epilepsia. 2007;48:77-81. 27. Morris 3rd GL, Mueller WM. Long-term treatment with vagus nerve stimulation in patients with refractory epilepsy. The Vagus Nerve Stimulation Study Group E01-E05. Neurology. 1999;53:1731-1735.
Contents lists available at ScienceDirect
Pediatric Neurology journal homepage: www.elsevier.com/locate/pnu
Original Article
Corpus Callosotomy for Childhood-Onset Drug-Resistant Epilepsy Unresponsive to Vagus Nerve Stimulation Ravindra Arya MD, DM a, *, Hansel M. Greiner MD a, Paul S. Horn PhD a, b, Michele Turner APRN a, Katherine D. Holland MD, PhD a, Francesco T. Mangano DO c a
Comprehensive Epilepsy Center, Division of Neurology, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio Division of Epidemiology and Biostatistics, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio c Division of Pediatric Neurosurgery, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio b
abstract PURPOSE: Corpus callosotomy and vagus nerve stimulation are common palliative options for people with drugresistant epilepsy when resective epilepsy surgery is not feasible. Because most of the published corpus callosotomy experience comes from a period before vagus nerve stimulation was approved and widely used, there is a paucity of data about efficacy of corpus callosotomy in patients with inadequate response to vagus nerve stimulation. METHODS: We report seven patients who had complete corpus callosotomy after an inadequate response to vagus nerve stimulation. At the time of surgery, these patients had failed a median of six antiseizure medications, three patients also had failed a trial of ketogenic diet, and all the patients had a vagus nerve stimulation implanted for a mean duration of 2.5 years with maximal tolerated settings. RESULTS: There was a decrease in total daily seizure frequency of 34.7% (94.7; median, 71.4%; interquartile range, 55.3) after corpus callosotomy at a mean follow-up of 2.6 years (1.4). One patient achieved complete seizure freedom and five patients had 50% reduction in seizure frequency. Six patients continued to have partial-onset seizures though the frequency was decreased. Drop attacks and tonic seizures stopped in all the patients. CONCLUSION: Seizure outcomes after corpus callosotomy in our series are most likely a result of complex dynamic interaction between the natural history of epilepsy, the effect of the surgery, ongoing vagus nerve stimulation modulation, and modification in antiseizure drugs. Our study supports the clinical decision to try corpus callosotomy in patients having nonlateralizing drugresistant epilepsy with inadequate response to vagus nerve stimulation. Keywords: corpus callosotomy, vagus nerve stimulation, drug-resistant epilepsy, seizure outcomes
Pediatr Neurol 2014; 51: 800-805 Ó 2014 Elsevier Inc. All rights reserved.
Introduction Conflicts of Interest: The authors declare no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. Ethical Approval: The study was approved by Institutional Review Board of Cincinnati Children’s Hospital Medical Center (IRB: 2008-1006). The study was conceived, and data were extracted by R.A. and H.M.G. K.D.H. created and maintains the epilepsy surgery database. P.S.H. helped with data management and analysis. All surgeries were performed by F.T.M. and follow-up evaluations were performed by M.T. The article draft was prepared by R.A., which was critically reviewed and approved by all authors.
Article History: Received June 20, 2014; Accepted in final form September 12, 2014 * Communications should be addressed to: Dr. Arya; Comprehensive Epilepsy Center; Division of Neurology; Cincinnati Children’s Hospital Medical Center; E4-145, ML 2015, 3333 Burnet Avenue; Cincinnati, Ohio 45229. E-mail address: [email protected] 0887-8994/$ - see front matter Ó 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.pediatrneurol.2014.09.008
In people with drug-resistant epilepsy (DRE), when it is not possible to identify a resectable core epileptic network, it becomes pertinent to offer palliative surgical options. These options have conventionally included vagus nerve stimulation (VNS) and corpus callosotomy. However, there is a lack of well-defined patient selection criteria for either technique because of lack of comparative studies generating data about differential efficacy or adverse effects.1 Usually, VNS is tried first in relatively older children with multiple seizure types whereas corpus callosotomy is recommended for relatively younger children with predominant drop attacks. We present a series of patients with nonlateralizing DRE with multiple seizure types who had complete corpus
R. Arya et al. / Pediatric Neurology 51 (2014) 800e805
callosotomy after failing multiple antiseizure medications and VNS. Most of the larger corpus callosotomy series were reported before VNS was approved by the Food and Drug Administration and widely used.2 Hence, there is a paucity of experience with corpus callosotomy in patients with inadequate response to VNS. This case series is likely to contribute towards the evidence for efficacy of corpus callosotomy in such patients. Methods The patients were identified by a retrospective chart review using an institutional epilepsy surgery electronic database. All patients with DRE who had a corpus callosotomy between January 2007 and December 2012 were screened for those with a history of prior VNS implantation. The usual noninvasive presurgical evaluation at our institution includes a detailed history, physical and neurological examination, video electroencephalograph (EEG) monitoring to capture sufficient number of habitual seizures, magnetic resonance imaging (MRI) of the brain including three-dimensional volumetric acquisition, interictal flurodeoxyglucose positron emission tomography, ictal and interictal single-photon emission computed tomography (SPECT), magnetoencephalography, and neuropsychologic evaluation. The positron emission tomography and SPECT data are routinely postprocessed using statistical parametric mapping and subtraction ictal SPECT coregistration to MRI (SISCOM) respectively. The information about each patient is synthesized and reviewed in a multi-disciplinary conference (Table 1). None of the patients included in this report had any prior resective epilepsy surgery or partial callosotomy, and all underwent complete corpus callosotomy. As VNS is a palliative procedure, we did not use conventional criteria for complete seizure freedom to define treatment failure.3 In cognizance of the fact that 50% reduction in seizure frequency is a common end point in clinical trials on DRE,4 we defined inadequate response to VNS as failure to achieve 50% reduction in preimplant seizure frequency in spite of maximum tolerated settings. The primary outcome measure was the reduction in total daily seizure frequency from presurgery to last available follow-up visit. Seizure frequency was determined based on parental reporting only. Secondary outcomes included reduction in the frequency of individual seizure types and number of seizure-related admissions during followup. Seizure outcome was also classified on the International League Against Epilepsy outcome scale based on seizure days. Data were extracted on to a Microsoft Excel (Microsoft Corporation, Redmond, WA) spreadsheet, and descriptive statistics were calculated. The study was approved by the Institutional Review Board of Cincinnati Children’s Hospital Medical Center.
Results
During the study duration, 29 patients underwent corpus callosotomy at our institution. Of these, seven patients (four males) were identified who had an inadequate response to a previous VNS (Table 1). Five patients (71.4%) had seizure onset in infancy (median, 6 months; interquartile range [IQR], 9). All patients had two seizure types including atonic, tonic, myoclonic, and generalized tonicclonic seizures (Table 1). The seizure frequency varied widely from one to two seizures/week to hundreds/day. All the patients had moderate to severe intellectual disability. At the time of surgical referral, the patients had failed a median of 6 (IQR, 5.5) antiseizure medications, with 3 also having failed a trial of ketogenic diet. Video-EEG revealed bilateral multifocal and/or diffuse seizure onsets in all the patients. A specific etiology could be established in two patients with one having a right frontal cortical dysplasia
801
(with discordant data from other modalities) and the other having bilateral encephalomalacia as a consequence of herpes encephalitis. One patient had a static encephalopathy with an interstitial duplication of 857 kb at 17q25.1 of unknown clinical significance. MRI was able to detect abnormalities in two other patients, although their etiologic significance was uncertain. The functional imaging modalities revealed nonlateralizing discordant data in all patients (Table 1). The mean time interval between VNS implantation and corpus callosotomy was 2.5 years (1.43). The maximal output current varied from 1.75 to 3 mA and six of seven patients were on conventional cycling (Table 2). Seizure outcomes
After a mean follow-up of 2.6 years (1.4), there was a mean reduction of 34.7% (94.7; median, 71.4%; IQR, 55.3) in the total daily seizure frequency (one-sided paired t test, P ¼ 0.0834). One patient was completely seizure free at 2.8 years of follow-up and four other patients had a 50%90% decrease in seizure burden. One patient experienced only mild reduction in seizure frequency (15.5%) and another, increase in seizure frequency (Fig 1). Six of the seven patients continued to have partial-onset seizures with or without generalization, although the frequency was decreased as described previously. Drop attacks and tonic seizures were abolished altogether in the respective patients (n ¼ 2 each). Episodes of staring and behavioral arrest ceased in one of the two patients; however, myoclonic seizures continued in the patient who had them at baseline (Fig 2). Most of the patients had either no (n ¼ 4) or only one (n ¼ 2) seizure-related admission to emergency or inpatient services, but one patient had six such admissions during 3.6 years of follow-up because of persistence of generalized seizures, in spite of substantial reduction in atonic drop attacks and tonic seizures. Because it is infrequent to achieve seizure freedom with corpus callosotomy, we have described the seizure-free patients below. Seizure-free patient
This 18-year-old girl with onset of seizures at 3 years of age had multiple seizure types including events with tremor-like movements of upper extremity and eyelid fluttering occurring in five to six daily clusters consisting of tens-to-hundreds of brief seizures. Sometimes, longer clusters would evolve into a generalized clonic seizure with eyes rolled back and postictal pallor and/or cyanosis. Other events consisted of staring, sometimes with upward gaze deviation, and pallor. She had failed nine antiseizure medications and had a VNS implanted at 11 years of age with 1year trial at maximally tolerated settings. She functioned cognitively at a 5-year-old level and had multiple behavioral issues. Her EEG seizure onsets were characterized by diffuse bilateral low-voltage fast activity (Table 1). Two weeks after complete corpus callosotomy, she experienced a decrease in seizure duration and frequency and reported improvement in behavior. After 8 weeks, seizures reduced to about 1-2 clusters/week, lasting 5-10 minutes, vs 15-40 minutes at baseline. Ten months postcallosotomy, she had a generalized seizure requiring an emergency visit. After that, seizure frequency further decreased, and she has remained seizure
802
R. Arya et al. / Pediatric Neurology 51 (2014) 800e805
TABLE 1. Outcomes and Demographic Details of Patients With Corpus Callosotomy After Failure of Vagus Nerve Stimulation
Patient
Sex
Age at Seizures Onset, mo
Seizure Types
Average Seizure Frequency
Etiology
MRI
1 2 3 4 5 6
M M F F F M
6 18 6 6 36 5
Atonic, tonic, myoclonic, GTCS Tonic progressing to GTCS Tonic or dystonic, þ/ GTCS Head and eye version, þ/ GTC Clonic, þ/ GTCS Tonic, staring
100/day 20/mo 8-10/wk 1-2/wk 100/day 6-8/wk
7
M
12
Atonic, tonic, GTCS staring
10-12/day
Unknown Unknown Unknown Cortical dysplasia Unknown Postherpes encephalitis Static encephalopathy*
Normal Nonspecific WM signal changes Dysgenesis of CC Loss of gray-white differentiation Normal Bilateral encephalomalacia Normal
Abbreviations: AED ¼ Antiepileptic drug MEG ¼ Magnetoencephalography CC ¼ Corpus Callosum MRI ¼ Magnetic resonance imaging F ¼ Female PET ¼ Positron emission tomography GTCS ¼ Generalized tonic-clonic seizures RF ¼ Right Frontal ILAE ¼ International League Against Epilepsy RT ¼ Right Temporal KD ¼ Ketogenic Diet SISCOM ¼ Subtraction Ictal SPECT Co-registered to MRI LH ¼ Left Hemisphere SPECT ¼ Single-photon emission computed tomography LT ¼ Left Temporal SPM ¼ Statistical parametric mapping LP ¼ Left Parietal WM ¼ White Matter M ¼ Male * Interstitial duplication of 857 kb of DNA at 17q25.1 detected by microarray of unknown clinical significance.
free for last 15 months on valproate, topiramate, and clonazepam. Disconnection syndromes
Three patients experienced transient postoperative deficits consistent with supplementary motor area (SMA) syndrome: reduced spontaneous speech and lack of interest in surrounding environment. Additionally, two patients respectively had weakness in nondominant leg and nondominant hand automatisms characterized by repetitive grasping movements. The manifestations completely resolved by 2 weeks in two patients and 5 weeks in another patient. Discussion
This study supports the efficacy of corpus callosotomy in patients with DRE who have failed VNS. Five of seven patients in this series had 50% reduction in seizure frequency, and one patient achieved complete seizure freedom (Table 1). These data agree with published experience, which has indicated the seizure freedom rates after corpus callosotomy to vary from 0% to 10%.5-8 In the second worldwide survey of epilepsy surgery programs, corpus
callosotomy was reported to have seizure-free outcome in 7.6% of the patients and an improvement in overall seizure burden in 60.9%.9 There was complete cessation of drop attacks and tonic seizures in two patients each. Six of the seven patients continued to have partial-onset seizures, although the frequency and duration were reduced. One of two patients experienced a remission of seizures with staring and behavioral arrest, whereas myoclonic seizures continued in the only patient that had those (Fig 2). This agrees with prior experience that drop attacks resulting from both tonic and atonic seizures are abolished by corpus callosotomy.6,7,10 The incidence of seizure-free outcome for drop attacks has varied from 51% to 90% across studies, depending primarily on patient selection.1,11 Additional studies have reported 90% reduction in seizure frequency in 85%-92% of patients.7,12 The reports of success with other seizure types have been more variable. Initially myoclonic seizures were believed to be poorly responsive to corpus callosotomy based on Yale experience, but later reports indicated better outcomes.7,13,14 Generalized tonic-clonic and atypical absence seizures were reported to significantly improve in 55% and 73% patients, respectively, in one study.15 The older smaller studies with 10 patients found higher rates of improvement but probably represent a biased cohort.2 Among the partial-onset seizures,
TABLE 2. Maximal Vagus Nerve Stimulator Settings and Interval Between Stimulator Implantation and Callosotomy
VNS Settings
Patient #1
Patient #2
Patient #3
Patient #4
Patient #5
Patient #6
Patient #7
Output current, mA Frequency, Hz Pulse width, mseconds Time on, seconds Time off, min Magnet stimulation, mA Magnet time, seconds Duration between VNS and callosotomy, yr
2.25 20 250 20 3 2.5 60 4.75
1.75 30 250 30 5 2 60 1.5
1.75 30 250 30 1.8 2 60 2.25
2 20 250 30 1.1 2.25 60 1.42
2.25 20 250 30 1.1 2.5 60 1.42
3 20 250 7 0.3 3.25 60 1.83
2 20 250 30 1.8 2.25 30 4.33
Abbreviation: VNS ¼ Vagus nerve stimulation
R. Arya et al. / Pediatric Neurology 51 (2014) 800e805
803
TABLE 1. (Continued)
PET/SPM
SPECT/SISCOM
MEG
Normal Normal LH LF, LT, LP Nonlateralizing LH
Nonlateralizing Nonlateralizing LF Nonlateralizing Bilateral multiple
Bilateral sources RF, RT LH Bilateral sources LH
-
-
-
#-AED Failed KD 3 6 8 3 9 3 20
Follow-up, % Reduction in Seizure ILAE Outcome Seizure-related yr Frequency Admissions
Failed Failed
3.61 3.90 1.20 3.59 2.79 2.89
50.0 71.4 15.5 170.0 100.0 86.0
4 4 4 4 1 4
6 1 0 0 1 0
Failed 0.33
90.0
4
0
callosal section is believed to prevent rapid bilateral synchrony of seizures originating in frontal lobe, whereas temporal lobe seizures are probably not affected by anterior callosotomy.5,16 All the patients in this series had complete corpus callosotomy. In certain patients, anterior callosotomy sparing the splenium is preferred, to preserve some interhemispheric transfer of sensory information and decrease the risk of nondominant apraxia and agnosia.10 However, there is evidence of relatively lower efficacy of anterior
callosotomy for seizure reduction compared with complete corpus callosotomy, particularly for patients with multiple seizure types in addition to drop attacks.5,17 Complete cessation of atonic and tonic drop attacks was found in 91% of pediatric patients by total callosotomy but in only 67% by partial callosotomy in one study.18 In patients with preexisting moderate-to-severe cognitive impairment, like in present series, the benefit of greater seizure reduction outweighs the risk of disconnection syndromes.5 The mechanism of seizure reduction after callosal section is not
FIGURE 1. Seizure frequency at each time point expressed as a percentage of baseline seizure frequency (taken as 100%). (The color version of this figure is available in the online edition.)
804
R. Arya et al. / Pediatric Neurology 51 (2014) 800e805
FIGURE 2. Breakup of patients according to individual seizure types at baseline and at last available follow-up. (The color version of this figure is available in the online edition.)
fully known. It has been hypothesized that achieving a critical degree of synaptoreduction with decreased number of neuronal connections may be responsible for the decrease in seizure frequency and altered seizure pattern.15 The persistence of most generalized seizures to some extent after corpus callosotomy suggests a role for other interhemispheric pathways for spread of ictal activity.5 The seizure outcomes were reported in our study at a mean follow-up duration of 2.6 years (1.4). It is known that seizure frequency after corpus callosotomy varies as a function of the length of follow-up. A meta-analysis found that in the long term (5 years), only 35% (95% confidence interval, 26-44) of the patients continued to remain free of drop attacks.19 The age at surgery varied from 6-17 years in our patients, except for one patient who was 25 years old at the time of corpus callosotomy. This patient experienced 90% reduction in seizure burden in short term. There is some evidence that younger age at surgery (18 years) is an independent predictor of improvement in neuropsychologic and quality-oflife outcomes.12,20 However, the association is equivocal for seizure outcome. Some studies have not found age at surgery to have prognostic significance; others have reported similar seizure outcomes in children compared with adults.12,21-23 Three patients in our series experienced transient postoperative deficits resembling the SMA syndrome. This confirms to the observation that SMA syndrome is more prominent and prolonged after 1-stage total callosal section.24 In general, however, the incidence of acute disconnection syndromes and long-term adverse effects are reported to be lower in children, especially those 12 years of age.10,25 All the patients included in this report had moderateto-severe cognitive impairment, which precluded testing for other relatively subtle chronic disconnection syndromes.24 Seizure outcomes after corpus callosotomy in our series are most likely a synergistic effect of the surgery itself, ongoing VNS modulation and modification in antiseizure drugs. Regarding efficacy of combining nonpharmacologic
therapies, a previous study has indicated 50%-responder rate of 70% in 30 children after 3 months of combined use of VNS and dietary therapy.26 Whereas VNS is known to provide accruing seizure reduction with increased duration,27 corpus callosotomy often makes an immediate impact on seizure frequency. Three of the five responders in our series had a clinically relevant reduction in seizure burden in first few months (Fig 1). This is valuable in the management of childhood epileptic encephalopathy, where corpus callosotomy may prevent the need for escalating antiseizure medication requirements and frequent emergency department visits. We believe that it is the short-term reduction in seizure burden that represents the true callosotomy effect, whereas the long-term outcomes are more representative of a complex interaction between the natural history of epilepsy and various treatment measures. Overall, this study supports the clinical decision to try corpus callosotomy in patients with nonlateralizing DRE who have had inadequate response to VNS. Although an important limitation of this study is the lack of neuropsychologic and quality-of-life outcomes, it emphasizes the need for studying comparative efficacy and safety of palliative treatments for patients with DRE who are not suitable for resective epilepsy surgery. R.A. receives research support from NIH (NINDS 2U01 NS045911) and American Epilepsy Society and Epilepsy Foundation of America Infrastructure award (pSERG); K.D.H. receives research support from NIH (NINDS R01 NS062756 [principal investigator]).
References 1. Rosenfeld WE, Roberts DW. Tonic and atonic seizures: what’s nexteVNS or callosotomy? Epilepsia. 2009;50(Suppl 8):25-30. 2. Spencer DD, Spencer SS. Corpus callosotomy in the treatment of medically intractable secondarily generalized seizures of children. Cleve Clin J Med. 1989;56(Suppl Pt 1):S69-78. discussion S9e83.
R. Arya et al. / Pediatric Neurology 51 (2014) 800e805 3. Kwan P, Arzimanoglou A, Berg AT, Brodie MJ, Allen Hauser W, Mathern G, et al. Definition of drug resistant epilepsy: consensus proposal by the ad hoc Task Force of the ILAE Commission on Therapeutic Strategies. Epilepsia. 2010;51:1069-1077. 4. Arya R, Glauser TA. Pharmacotherapy of focal epilepsy in children: a systematic review of approved agents. CNS Drugs. 2013;27:273-286. 5. Asadi-Pooya AA, Sharan A, Nei M, Sperling MR. Corpus callosotomy. Epilepsy Behav. 2008;13:271-278. 6. Shimizu H, Maehara T. Neuronal disconnection for the surgical treatment of pediatric epilepsy. Epilepsia. 2000;41(Suppl 9):28-30. 7. Cukiert A, Burattini JA, Mariani PP, Camara RB, Seda L, Baldauf CM, et al. Extended, one-stage callosal section for treatment of refractory secondarily generalized epilepsy in patients with Lennox-Gastaut and Lennox-like syndromes. Epilepsia. 2006;47:371-374. 8. Hanson RR, Risinger M, Maxwell R. The ictal EEG as a predictive factor for outcome following corpus callosum section in adults. Epilepsy Res. 2002;49:89-97. 9. Engel Jr J. Surgery for seizures. N Engl J Med. 1996;334:647-652. 10. Wong TT, Kwan SY, Chang KP, Hsiu-Mei W, Yang TF, Chen YS, et al. Corpus callosotomy in children. Childs Nerv Syst. 2006;22:999-1011. 11. Sunaga S, Shimizu H, Sugano H. Long-term follow-up of seizure outcomes after corpus callosotomy. Seizure. 2009;18:124-128. 12. Maehara T, Shimizu H. Surgical outcome of corpus callosotomy in patients with drop attacks. Epilepsia. 2001;42:67-71. 13. Spencer SS. Corpus callosum section and other disconnection procedures for medically intractable epilepsy. Epilepsia. 1988;29(Suppl 2):S85-99. 14. Spencer SS, Spencer DD, Sass K, Westerveld M, Katz A, Mattson R. Anterior, total, and two-stage corpus callosum section: differential and incremental seizure responses. Epilepsia. 1993;34:561-567. 15. Madsen JR, Carmant L, Holmes GL, Black PM. Corpus callosotomy in children. Neurosurg Clin N Am. 1995;6:541-548. 16. Purves SJ, Wada JA, Woodhurst WB. Corpus callosum section for complex partial seizures. In: Reeves AG, Roberts DW, eds. Epilepsy and the Corpus Callosum 2. Springer; 1995:175-182.
805
17. Spencer SS, Spencer DD, Williamson PD, Sass K, Novelly RA, Mattson RH. Corpus callosotomy for epilepsy. I. Seizure effects. Neurology. 1988;38:19-24. 18. Shimizu H. Our experience with pediatric epilepsy surgery focusing on corpus callosotomy and hemispherotomy. Epilepsia. 2005; 46(Suppl 1):30-31. 19. Tellez-Zenteno JF, Dhar R, Wiebe S. Long-term seizure outcomes following epilepsy surgery: a systematic review and meta-analysis. Brain. 2005;128:1188-1198. 20. Turanli G, Yalnizoglu D, Genc-Acikgoz D, Akalan N, Topcu M. Outcome and long term follow-up after corpus callosotomy in childhood onset intractable epilepsy. Childs Nerv Syst. 2006;22: 1322-1327. 21. Kim DS, Yang KH, Kim TG, Chang JH, Chang JW, Choi JU, et al. The surgical effect of callosotomy in the treatment of intractable seizure. Yonsei Med J. 2004;45:233-240. 22. Mamelak AN, Barbaro NM, Walker JA, Laxer KD. Corpus callosotomy: a quantitative study of the extent of resection, seizure control, and neuropsychological outcome. J Neurosurg. 1993;79: 688-695. 23. Rougier A, Claverie B, Pedespan JM, Marchal C, Loiseau P. Callosotomy for intractable epilepsy: overall outcome. J Neurosurg Sci. 1997;41:51-57. 24. Jea A, Vachhrajani S, Widjaja E, Nilsson D, Raybaud C, Shroff M, et al. Corpus callosotomy in children and the disconnection syndromes: a review. Childs Nerv Syst. 2008;24:685-692. 25. Reutens DC, Bye AM, Hopkins IJ, Danks A, Somerville E, Walsh J, et al. Corpus callosotomy for intractable epilepsy: seizure outcome and prognostic factors. Epilepsia. 1993;34:904-909. 26. Kossoff EH, Pyzik PL, Rubenstein JE, Bergqvist AG, Buchhalter JR, Donner EJ, et al. Combined ketogenic diet and vagus nerve stimulation: rational polytherapy? Epilepsia. 2007;48:77-81. 27. Morris 3rd GL, Mueller WM. Long-term treatment with vagus nerve stimulation in patients with refractory epilepsy. The Vagus Nerve Stimulation Study Group E01-E05. Neurology. 1999;53:1731-1735.
