Vagus Nerve Stimulation in children: A focus on intellectual disability

e u r o p e a n j o u r n a l o f p a e d i a t r i c n e u r o l o g y x x x ( 2 0 1 7 ) 1 e1 4

Official Journal of the European Paediatric Neurology Society

Review article

Vagus Nerve Stimulation in children: A focus on intellectual disability Jo Sourbron a, Sylvia Klinkenberg b, Alfons Kessels c, Helenius Jurgen Schelhaas d, Lieven Lagae e, Marian Majoie b,d,* a

Laboratory for Molecular Biodiscovery, Department of Pharmaceutical and Pharmacological Sciences, KU Leuven, Leuven, Belgium b Department of Neurology, Maastricht University Medical Center, Maastricht, The Netherlands c Department of Clinical Epidemiology and Medical Technology Assessment, Maastricht, The Netherlands d Department of Neurology, Epilepsy Center Kempenhaeghe, The Netherlands e Department of Development and Regeneration, Section Pediatric Neurology, University Hospitals KU Leuven, Leuven, Belgium

article info

abstract

Article history:

Introduction: Vagus Nerve Stimulation (VNS) can be an efficacious add-on treatment in

Received 16 March 2016

patients with drug-resistant epilepsy, who are not eligible for surgery. Evidence of VNS

Received in revised form

efficacy in children with intellectual disability (ID) is scarce.

26 October 2016

Objectives: The purpose of this study was to review all available VNS data in the pediatric

Accepted 23 January 2017

population (
18 years old) and focus on the subpopulation with ID since appropriate treatment of these children is often challenging and complex.

Keywords:

Methods: Cochrane, EMBASE, PubMed and MEDLINE were used to collect all research

Drug-resistant epilepsy

associated to VNS and ID (or synonyms) leading to a total of 37 studies. Seven studies

Alternative treatment

showed the results of patients with ID and those without separately; thereby only these

VNS

studies were included in the VNS meta-analysis.

Neurostimulation

Results: Ourmeta-analysis showedthat VNS wasless effective in pediatric epilepsy patientswith

Meta-analysis

IDcomparedtothose without ID (Mantel-Haenszel meta-analysis;p ¼ 0.028, OR 0.18 (CI 95% 0.039

Quality of life

e0.84)). However, there were no prospective controlled studies. Numerous studies reported quality of life (QoL) improvements in this subpopulation. The most common adverse events were transient and well tolerated. Side effects on cognition and behavior were not reported. Discussion: These results might be a reason to consider VNS early on in the treatment of this subgroup. The significantly greater amount of retrospective studies, differences in followup (FU), lack of control data, heterogeneous series and limited number of patients could have biased the outcome measurements. Hence, current data do not exclude VNS for children with drug-resistant epilepsy and ID but should be interpreted with caution. © 2017 European Paediatric Neurology Society. Published by Elsevier Ltd. All rights reserved.

* Corresponding author. Department of Neurology, Epilepsy Center Kempenhaeghe, Sterkselseweg 65, P.O. Box 61, NL-5590 AB, Heeze, The Netherlands. Fax: þ31 40 2265691. E-mail address: [email protected] (M. Majoie). http://dx.doi.org/10.1016/j.ejpn.2017.01.011 1090-3798/© 2017 European Paediatric Neurology Society. Published by Elsevier Ltd. All rights reserved. Please cite this article in press as: Sourbron J, et al., Vagus Nerve Stimulation in children: A focus on intellectual disability, European Journal of Paediatric Neurology (2017), http://dx.doi.org/10.1016/j.ejpn.2017.01.011

2

e u r o p e a n j o u r n a l o f p a e d i a t r i c n e u r o l o g y x x x ( 2 0 1 7 ) 1 e1 4

Contents 1. 2.

3.

4.

1.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. Search strategy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. Selection criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3. Data analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Results and discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2. Patient characteristics, including epilepsy syndrome and seizure type . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3. Duration of treatment and concomitant AED treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4. VNS parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.5. Efficacy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.6. Quality of life (QoL) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.6.1. QoL assessment methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.6.2. QoL meta-analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.6.3. QoL and ID . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.6.4. QoL improvements and seizure control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.6.5. Limitations of QoL assessments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.6.6. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.7. Safety . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conflicts of interest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Supplementary data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Introduction

Epilepsy is characterized by recurrent, unpredictable and unprovoked epileptic seizures that interfere with normal brain function.1 Around one out of three children with epilepsy shows cognitive and/or behavioral impairments2 and treatment of seizures in these patients is often challenging and complex.3e6 Since 30% including a considerable amount of children with ID does not respond to current anti-epileptic drugs (AEDs), other treatment options like ketogenic diet and VNS can be promising.7 VNS is indicated in patients with drug-resistant epilepsy for whom epilepsy surgery is not possible.8e13 For the adult population it has been proven to be efficacious and well tolerated.14 Moreover, QoL improvements have been reported15 and VNS can improve mood significantly which highlights the clinical application of VNS for the treatment of severe, drug-resistant depression.16 Less is known about the pediatric population. In 1996 the first randomized controlled trials (RCTs) (E01-E05) were conducted but only one included some children 12 years of age, suffering from drug-resistant localized epilepsy.17 The first RCT with solely children with drug-resistant epilepsy, also below the age of 12, was 16 years later.12 Hence, the minority of the herein reviewed studies was large and most evidence of VNS efficacy in children is retrieved from case series, either retro- or prospective. Children, with drugresistant epilepsy, have a significantly higher chance for

00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

intellectual and mental deterioration, and therefore might benefit from VNS0 potential contribution to cognition and behavior.2 In conclusion, data are scarce regarding the efficacy of VNS and related QoL improvements in children. Our aim was to review all available data about VNS efficacy and safety in children with ID.

2.

Methods

2.1.

Search strategy

Cochrane, EMBASE, PubMed and MEDLINE served to obtain all available literature until May 2015. The subsequent terms were used: “VNS”, “Vagus Nerve Stimulation” or “nervous vagus stimulation” in combination with “mental retardation”, “mental retarded”, “low IQ”, “developmental disabled”, “developmental disabilities”, “intellectual disabled”, “intellectual disabilities”, “quality of life”, “QoL” (Species: humans).

2.2.

Selection criteria

A review of the full text of the obtained articles was performed to exclude articles: (1) without notion of intellectual or developmental status, (2) with insufficient

Please cite this article in press as: Sourbron J, et al., Vagus Nerve Stimulation in children: A focus on intellectual disability, European Journal of Paediatric Neurology (2017), http://dx.doi.org/10.1016/j.ejpn.2017.01.011

e u r o p e a n j o u r n a l o f p a e d i a t r i c n e u r o l o g y x x x ( 2 0 1 7 ) 1 e1 4

information to distinguish data of patients
18 of those >18 years of age (at follow-up (FU)) and (3) studies that consisted of insufficient data to evaluate VNS efficacy (i.e. responder rate, defined as a 50% seizure frequency reduction). Outcomes of interest were seizure frequency, seizure severity, QoL and safety. To determine the presence of intellectual disabilities the definition proposed by the American Association on Intellectual and Developmental Disabilities (AAIDD) was used.18 These selection criteria led to 12 articles of which the references were analyzed to discover other valuable articles. Finally, this led to a total of 37 hits.

2.3.

Data analysis

To examine the effect of VNS in children with drug-resistant epilepsy and ID, an overview of the patient demographics and VNS parameters was made. Details were summarized with the evaluation of seizures (frequency and severity), functional status changes (alertness, attention, behavior, language, memory, mood, seizure clustering, postictal period, sleep, task performance) and safety. Responder rate was defined as a 50% seizure frequency reduction (SFR). Side effects were categorized as minor (mild, transient) or major (requiring additional medication and/or surgery). In addition, changes in the electroencephalography (EEG) pattern and AED regimen were examined. Some included studies reported data of children and adults. If sufficient detailed patient information was available, data of the children were taken into account and calculations were only performed on this subgroup. Furthermore, the responder rate was determined for only those patients with ID if sufficient individual patient data were available. This was done for seven studies, which are indicated by a ¶ in Table 1.19e25 If both children and adults were included in the study, calculations were solely performed on the subgroup
18 years of age and only if sufficient patient information was available (Table AeD Supplementary data). Aforementioned recalculations were done for 13 studies.20,23e34 The difference between the responder rate in retro-versus prospective studies was evaluated by a Student's t-test after Welch's correction (equal variances could not be assumed), using GraphPad Prism 5 software (GraphPad Software, Inc.). Values were presented as mean ± standard deviation (SD). The meta-analysis was conducted on all data in children with metan in STATA using a random effect model to study VNS efficacy (responder 50% SFR). Statistical heterogeneity was assessed by calculating I2 and p-value. VNS efficacy was assessed in the subgroup of patients with ID (compared to those without ID) by a Mantel-Haenszel meta-analysis to calculate Odds ratios (OR) with 95% confidence intervals (95% CIs). The number of patients with a QoL improvement (compared to those without) was analyzed for all data in children that were supported by statistical tests, using contingency tables and a Chi-square test by GraphPad Prism 5 software (GraphPad Software, Inc.). The amount of patients was presented as percentages. Statistically significant differences are represented by one (p < 0.05) or two stars (p < 0.01).

3.

Results and discussion

3.1.

Studies

3

All articles concerning VNS in children with ID published between 1998 and 2015 were included in this review, leading to total of 37 articles (Fig. 1). The majority of the included studies were retrospective (75.7%) which could have affected the outcome parameters and might have impeded an accurate evaluation of VNS efficacy.35 Our results showed that retrospective studies reported a significantly greater effect on seizure reduction, compared to prospective studies (p < 0.01, Fig. 2). Therefore, we acknowledge a potential biased higher responder rate in these retrospective studies with pronounced differences in follow-up (FU) periods. In contrast, prospective studies are more reliable and could offer a higher level of evidence in the future for the recommendation of VNS in children. Nearly all research articles concerning VNS in children with ID published between 1998 and 2015 are single-armed studies (i.e. a single intervention (VNS) is given to all subjects). This study design inherently leads to the absence of adequate control data. If these studies would consequently be excluded, the amount of data would be very limited. A long FU period could overcome this problem since a preserved responder rate over time is more likely due to the intervention (VNS), rather than to the natural course of devastating, drugresistant epilepsies. This assumption has already been validated in children with tuberous sclerosis complex (TSC) and could be valid for other drug-resistant epilepsies as well.21,36 In addition, prospective studies were not always blinded, neither placebo-controlled and the small number of patients made it difficult to extrapolate the results to the entire population and perform statistical analysis.37 In our review the number of patients included in a study ranged between six and 347, with an average of 28. This low average arises from only two studies with more than 100 patients.38,39 Furthermore, these studies reported an apparent difference in responder rate (i.e. 64.8%38 vs. 37.6%39). Hence, to provide more robust data on the effects of VNS in children, future large-scale, multicenter, RCTs are needed.28

3.2. Patient characteristics, including epilepsy syndrome and seizure type Since drug-resistant epilepsies are often accompanied by ID, it is worthwhile to examine the effects of VNS specifically in this subpopulation. The patient characteristics are summarized in Table 1. The complexity of multiple seizure types and the lack of individual, follow-up patient data did not allow any statistical test to correlate VNS efficacy to a specific seizure type. However, Cukiert et al. found that VNS was more effective in the control of atypical absences, generalized tonic-clonic seizures and myoclonic seizures, if compared to tonic and atonic seizures.40 In contrast, most studies are in favor of VNS in controlling focal seizures, although a recent meta-analysis did not associate the seizure frequency reduction (SFR) by VNS to a certain seizure type.41 This is in line with other studies which did not show a correlation between VNS efficacy and a distinct seizure type.24,25,36 In addition, most of the research,

Please cite this article in press as: Sourbron J, et al., Vagus Nerve Stimulation in children: A focus on intellectual disability, European Journal of Paediatric Neurology (2017), http://dx.doi.org/10.1016/j.ejpn.2017.01.011

4

Year

Type of study

Pediatric patients

Form of epilepsy

Age (y)

FU (m)

Patients with ID (%)

Patients with 50% seizure reduction at FU (%)

Percentage seizure reduction at FU (%)

Alexopoulos Andriola Arhan Arthur Blount Buoni Chen Cukiert Elliot Elliot Frost

2006 2001 2010 2006 2006 2004 2012 2013 2011 2009 2001

R R R R R R R R R R R

40* (37¶) 5* 20* (12¶) 5 6 9* 8 20 128 9* (8¶) 20

Various Various Various MD Various Various Various LGS and LGSl Various TSC LGS

2.3e17.9 3.0e15 6.4e17.9 (13.6) <12.0 2.5e6.4 6.0e18.0 4.0e17.0 <18.0 1.3e18.0 2.0e18.0 (10.9) <12.0

24.0 19.2 (6.0e30.0) 34.5 (6.0e100.0) ND (12.0e48.0) 21.7 (6.0e58.0) 27.2 (9.0e36.0) 20.1 (9.0e33.0) 12.0 12.0 38.1 (8.5e108.0) 6.0

92.5 (100.0¶) 100.0 60.0 (100.0¶) 100.0 100.0 100.0 100.0 100.0 79.4 88.9 (100.0¶) >50.0

60.0 (59.5¶) 60.0 80.0 (75.0¶) 0.0 83.3 33.3 62.5 46.6 64.8 77.8 (75.0¶) ND

€o €k Hallbo Hosain Khurana Kim Klinkenberg Lundgren MajkowskaZwolinska Majoie Major Mikati Nagarajan Orosz Parain Park (a, LKS) Park (b, ASD) Parker Rychlicki Shahwan Wakai Wilfong You You Zamponi Zamponi (DS) Zamponi (TSC) Zamponi Zamponi

2005 2000 2007 2015 2012 1998 2012

P P R R P P P

15 8* 26 4* 41 15* (13¶) 56

Various LGS Various LGS Various Various Various

4.0e17.0 4.0e16.0 3.0e17.0 13.8e17.1 3.0e17.0 4.0e19.0 <18.0

9.0 6.0 12.0 70.8 (42.0e97.2) 9.0 11.0 (10.0e12.0) 24.0

93.3 100.0 77.0 100.0 92.7 87.5 (100.0¶) 65.0

40.0 50.0 54.0 50.0 16.0 40.0 (38.5¶) 55.6

83.0 ND ND ND ND 23.3 ND ND 58.9 ND 37.0 (1 m), 48.0 (3 m), 79.0 (6 m) 63.0 (9 m) 53.3 ND ND ND 65.9 (66.1¶) ND

2001 2008 2009 2002 2015 2001 2003 2003 2001 2006 2009 2001 2006 2008 2007 2011 2011 2010 2008 2002

P R R R R R R R P R R R R R P R R R R P

16 10* 9* (8¶) 16 347 8* 6 59 16 36 26 4* (3¶) 7 10 28 31* 7* 7* (6¶) 6 13

LGS and LGSl TSC Various Various Various TSC Various Various Various Various Various Various Rett LGS Various Various DS TSC Various Various

6.0e17.0 2.2e15.4 2.2e13.8 (8.6) 3.0e17.0 0.1e14.0 5.0e18.0 4.0e7.8 0.0e10.4 5.0e16.0 1.5e18.0 5.0e16.0 3.8e16.8 1.0e14.0 10.7 2.4e17.8 2.2e17.1 6.0e11.0 3.6e15.3 (9.5) 0.7e2.6 1.4e17.0

6.0 52.8 (6.0e84.0) 9.2 (4.8e16.8) 24.9 (6.0e47.0) 12.0 17.5 (6.0e42.0) 6.0 12.0 12.0 24.0 18.0 7.0 (3.0e10.0) 12.0 33.0 (12.0e83.0) 12.0 12.0 12.0 12.0 41.6 (14.3e81.3) 3.0

100.0 100.0 90.0 (100.0¶) 100.0 >70.5 100.0 100.0 100.0 100.0 86.1 88.5 75.0 (100.0¶) 100.0 100.0 >50.0 100.0 100.0 85.7 (100.0¶) 100.0 100.0

25.0 60.0 33.3 (25.0¶) 62.5 37.6 87.5 50.0 58.0 25.0 71.0 54.0 75.0 (66.7¶) 85.7 90.0 53.6 48.4 57.1 85.7 (66.7¶) 100.0 76.9

26.9 ND 30.4 ND 42.9 ND ND ND ND 49.0 (6 m) ND 57.5 ND ND ND 38.6 (12 m) 30.6 55.3 ND ND

ID ¼ intellectual disability; FU ¼ follow-up; y ¼ year(s); m ¼ month(s); w ¼ weeks; R ¼ retrospective; P ¼ prospective; * ¼ included amount of pediatric patients
18 years old; ¶ ¼ amount of patients with ID; ND ¼ not determined; MD ¼ Mitochondrial disease; LGS ¼ Lennox-Gastaut syndrome; LGSl ¼ Lennox-Gastaut-like syndrome; TSC ¼ tuberous sclerosis complex; LKS ¼ LandaueKleffner syndrome; ASD ¼ autism spectrum disorder; DS ¼ Dravet syndrome.

e u r o p e a n j o u r n a l o f p a e d i a t r i c n e u r o l o g y x x x ( 2 0 1 7 ) 1 e1 4

Please cite this article in press as: Sourbron J, et al., Vagus Nerve Stimulation in children: A focus on intellectual disability, European Journal of Paediatric Neurology (2017), http://dx.doi.org/10.1016/j.ejpn.2017.01.011

Table 1 e Demographics and evaluation of seizures. First author

e u r o p e a n j o u r n a l o f p a e d i a t r i c n e u r o l o g y x x x ( 2 0 1 7 ) 1 e1 4

5

Fig. 1 e Citation flowchart of search strategy.

including those that focused on a specific genetic syndrome (Dravet (DS)), Rett, TSC or a specific epilepsy syndrome (e.g. Doose or Lennox-Gastaut (LGS)), was unable to correlate VNS efficacy with a certain genetic etiology or epilepsy syndrome.42,43 However, from the available data we get the impression that TSC patients can benefit from VNS. Almost 5.0% of all included patients (n ¼ 51/1097) had TSC and VNS efficacy seemed to be insignificantly higher (SFR>50% in 33/51 patients at FU, i.e. 64.7%) compared to the total population (57.0%, Fig. 1).

Fig. 2 e Percentage of children with ≥50% seizure reduction (i.e. responder rate). Significant difference between VNS efficacy reported in retrospective (R) studies, compared to prospective (P) studies (**p < 0.01; unpaired Student t-test with Welch's correction (equal variances could not be assumed)).

Based on our analysis VNS appeared to be efficacious, safe and well tolerated in the pediatric subpopulation (see below: 3.5 Efficacy; 3.6 Quality of life; 3.7 Safety). Alexopoulus et al. even suggested that children below the age of 12 had a greater seizure reduction, if compared to older children.19 Even younger patients (below five years of age) with lifethreatening epilepsy had a relatively satisfactory SFR, although VNS did not ameliorate the progression of the underlying condition and larger studies are needed to support the use of VNS in very young children.44 In line with these findings, recent research correlated higher SFR rates to the relatively younger age of VNS implantation.36 In addition, newer smaller VNS devices are available which allows the VNS implantation at younger ages.36 Moreover, it is expected that decreasing seizure burden during the child's development will offer important future, functional advantages.38 Consistently, Zamponi et al. reported that the subgroup of patients below six years of age had better cognitive and neuropsychological outcomes when compared to older patients.25 In line with these findings, some studies suggested a higher plasticity of the brain in younger children45 and/or earlier use of the device as determinants that could lead to a more pronounced, favorable effect of VNS.19 This review thereby offers some evidence that VNS might be effective in the pediatric population, regardless of epilepsy syndrome and seizure type. Furthermore, it has been suggested that VNS outcomes (SFR19,36 and QoL25) could be better if treatment is started at younger ages. Nonetheless, VNS

Please cite this article in press as: Sourbron J, et al., Vagus Nerve Stimulation in children: A focus on intellectual disability, European Journal of Paediatric Neurology (2017), http://dx.doi.org/10.1016/j.ejpn.2017.01.011

6

e u r o p e a n j o u r n a l o f p a e d i a t r i c n e u r o l o g y x x x ( 2 0 1 7 ) 1 e1 4

efficacy depends on multiple factors and not solely on age of implantation.36

3.3. Duration of treatment and concomitant AED treatment In this review, the range of FU duration of the selected studies varied between three months and six years. Zamponi et al. documented that a minimum period of three months could predict the long-term outcome.46 Even two months would be sufficient, as suggested by Buoni and colleagues.26 Consistently, You et al. observed that the treatment duration did not coincide with the effects.47,48 Nevertheless, long-term FU was recommended to determine the maximal efficacy49 and fully examine behavioral improvements.50 Furthermore, most studies suggested that the efficacy of VNS is more pronounced after longer treatment duration, and thereby were in favor of a >12 months FU.20,37 As a result, different FU periods could explain the described differences in improvements.42,51 More importantly, there is a reduction of sample size at longer FU in the plethora of studies. Therefore, we critically reviewed the data at different FU time points, e.g. for the study by Orosz et al. 12 months was chosen as FU period because at longer FU (24 months) the amount of included patients was significantly lower (346 patients at 12 months FU vs. 208 patients at 24 months FU).39 Since VNS is used as add-on treatment of drug-resistant epilepsies, it was important to examine the AED regimen. No significant increases were noted in nearly all studies, thus the observed effects were not due to a higher number or higher dose of AEDs. Moreover, a significant decrease in the number of AEDs has been reported in a few studies at FU.27,42 This observation could underline the efficacy of VNS, since less AEDs were needed to obtain adequate seizure control. Additionally, AED treatment is associated with important functional side effects; e.g. cognitive, behavioral, or psychiatric adverse reactions. VNS on the other hand does not severely affect and can even improve the functional status on different levels (see below: 3.6 Quality of life). AEDs are nearly always continued so that the reported beneficial effects after initiation of VNS can be due to this continued best drug treatment (BDT).52 Additionally, the observed positive outcomes can be due to the natural course of the epileptic syndrome (not related to VNS), despite the unfavorable progress of most drug-resistant epilepsies.34 These aforementioned effects should be considered when interpreting VNS studies.

3.4.

VNS parameters

VNS settings of the included studies are summarized in Table 2. Most studies reported an average range between 1.75 and 2.0 mA (maximum 3.0 mA), signal frequency of 20 or 30 Hz, 30 s time “on”, 3 or 5 min time “off” and a pulse width of 500 ms (standard cycling). Even as no significant differences of optimal VNS settings have been found between the included studies a distinct patient intervariability was clearly present. Unexpectedly, a recent RCT12 noted a reduction in seizures in the low-stimulation group (active control group; 0.25 mA, 1 Hz, 14 s time “on”, 60 min time “off” and a pulse width of 0.1 s), while the intensity and pulse width were too low in this group to evoke an action potential, as demonstrated by Koo

et al. and Heck et al..53,54 A recent open-label study by Orosz et al. reported an increased response rate in the subgroup receiving a higher total charge by VNS per day. This post-hoc analysis showed a doseeresponse correlation, although no exact effect sizes were published and the effect was not continued after longer FU (i.e. 24 months).39 Even though a faster intermittent pulse stimulation (rapid cycling) could improve seizure control and patient comfort27 no significant improvements were reported in nearly all patients on rapid cycling. Moreover, standard cycling has already been validated for over 14 years32 and a few studies involving pediatric patients showed relatively higher SFR rates in patients on this cycling mode.26,47,48,55 The number of patients on rapid cycling, however, is relatively low (i.e. less than four patients) and the comparison between standard and rapid cycling has never been a primary goal of any study. As a result, appropriate statistical analyses to compare these patients to the ones on standard cycling mode are not possible and no definite conclusions can be drawn from this metaanalysis. Special caution is also needed in cognitively impaired children since aggressive ramping could worsen preexisting dysphagia.56 Additionally, we identified the successful use of the magnet in some studies.57,58 This approach underlines the active stimulating of the vagus nerve during or in the early onset of a seizure. Hence, self-dependent use of the magnet and even automatic stimulation of the vagus nerve could lead to better seizure control, less side effects and QoL improvements in adults.59 These latter observations of closed-loop neurostimulation still need to be confirmed in the pediatric population by future prospective studies. In brief, dosing of VNS is as variable as AED dosing and implies thorough patient screening and follow-up. Furthermore, VNS programming can add to the high variability in responder rates reported in VNS-related research. This stated interpatient variability in parameter settings and stimulation pattern of the vagus nerve underlines the unacquaintance of VNS0 exact mechanism of action.

3.5.

Efficacy

With respect to efficacy of VNS, high variability has been documented. Hence, we expected the broad range of 0e100% of patients reaching 50% SFR, i.e. responder rate (57.0 ± 22.1%, average ± SD) (Table 1, Figs. 2 and 3). This overall percentage is consistent with the reported efficacy of VNS in 470 children by the American Academy of Neurology evidence-based methodology (55.0%, average).60 Additionally, seven studies published sufficient data to distinguish VNS outcomes of patients with ID, compared to those with a normal intellectual and developmental status.19e25 The responder rates appeared to be relatively lower in patients with ID (Mantel-Haenszel meta-analysis; p ¼ 0.029, OR 0.18 (CI 95% 0.039e0.84)). This enforces the hypothesis that ID can negatively influence VNS efficacy (significantly lower SFR, compared to VNS SFR in pediatric patients with normal ID).37,50,61e64 However, difficult to treat seizures are more prominent in epilepsy patients with ID and seem to be more treatment-resistant when the degree of ID is more pronounced.65 Thus, the lower VNS efficacy in ID patients

Please cite this article in press as: Sourbron J, et al., Vagus Nerve Stimulation in children: A focus on intellectual disability, European Journal of Paediatric Neurology (2017), http://dx.doi.org/10.1016/j.ejpn.2017.01.011

7

e u r o p e a n j o u r n a l o f p a e d i a t r i c n e u r o l o g y x x x ( 2 0 1 7 ) 1 e1 4

Table 2 e VNS parameters. First author

Year

VNS Stimulus (mA)

Signal frequency (Hz)

Time “on” (s)

Time “off” (min)

Signal pulse width (ms)

Alexopoulos Andriola Arhan Arthur Blount Buoni Chen Cukiert Elliot Elliot Frost €o €k Hallbo Hosain Khurana Kim Klinkenberg Lundgren Majkowska-Zwolinska Majoie Major Mikati Nagarajan Orosz Parain Park (a, LKS) Park (b, ASD) Parker Rychlicki Shahwan Wakai Wilfong You You Zamponi Zamponi (DS) Zamponi (TSC) Zamponi Zamponi

2006 2001 2010 2006 2006 2004 2012 2013 2011 2009 2001 2005 2000 2007 2015 2012 1998 2012 2001 2008 2009 2002 2015 2001 2003 2003 2001 2006 2009 2001 2006 2008 2007 2011 2011 2010 2008 2002

2.5 1.5e2.5 ND 1.25e2.0 (1.9) ND 1.5e2.5 1.5 (max) 3.0 (max) ND ND 1.25 1.0e1.5 0.5e1.75 1.5e2.0 ND 1.75e2.25 1.25e2.0 0.75e2.5 1.5e2.0 0.75e3.0 2.0 1.5e2.75 1.8 1.25e3.5 ND ND 1.5e2.0 ND 1.5e2.0 ND 1.5 (12 m) 2.0e2.5 3.5 (max) 1.0e2.0 2.0 2.0 2.0 (max) 2.0

30 20e30 ND 30 ND 30 30 30 ND ND 30 30 30 ND ND 30 30 20e30 30 30 30 30 30 30 ND ND ND ND 30 ND 30 30 30 30 30 30 30 30

30 30 ND 30 ND 30 30 30 ND ND 30 30 30 30 ND 30 30 30 30 7e30 30 30 30 30 ND ND 30 ND 30 ND 30 30 30 30 30 30 30 30

3 5 ND 1.1e3 ND 5 5 5 ND ND 5 5 5 5 ND 5 5 5 3 0.2e3 5 3e5 5 5 ND ND 5 ND 3 ND 5 5 5 5 5 5 5 5

250 500 ND 250e500 ND 500 500 500 ND ND 500 500 ND 250 ND 500 500 500 500 ND 250 500 500 500 ND ND 500 ND ND ND 500 500 500 500 500 500 ND 500

re; Hz ¼ Herz; s ¼ seconds; min ¼ minutes; ms ¼ microseconds. ND ¼ not determined; mA ¼ milli Ampe

compared to those without ID can be due to the higher prevalence of difficult to treat seizures in this subpopulation (compared to the subpopulation without ID). Nearly all studies measured VNS efficacy by this outcome (SFR), although the seizure burden and seizure severity should be considered as well.12,42 Many caregivers mentioned that QoL improvements (e.g. less seizure burden) were more important than SFR, although this often lacks proper statistical analyses (see below: 3.6 Quality of life).51 In addition, the epilepsy of VNS-treated patients is drug-resistant thus one should not expect to obtain a spectacular SFR or seizure freedom due to VNS treatment.23 The seizure severity can be examined by using the NHS-3 scale, which is the National Hospital Seizure Severity Scale. This is a further improvement of the Chalfont Seizure Severity Scale described by O'Donoghue and colleagues.66 Nonetheless, other scales for seizure severity have been published.67 Our results showed a seizure severity reduction of 12.6e50.0%, although limited data were available for the needed

calculations.12,22,29,32,55,63 Not only a decrease in seizure severity, but an association with milder, shorter seizures and quicker recovery were reported in most responders.42 EEG is a potential method to record a reduction in epileptic brain activity by anti-epileptic treatments, however there is no clear correlation between treatment efficacy and EEG recordings. Consistently, Cukiert et al. showed that there were no changes in EEG, while clinical improvements did occur.40 Furthermore, there was no concordance in changes in EEG during a 24 h registration (degree of spike reduction) and seizure reduction, severity and QoL.66 Nevertheless, this study demonstrated a significant reduction in the frequency of electrographic seizures after three and nine months of VNS treatment. In addition, Elliot et al. reported a 54.6e86.2% seizure reduction of focal, multifocal, diffuse and multifocal & diffuse EEG.38 Another study showed a disappearance of diffuse slow spike-wave complexes, paroxysmal fast activities and finally multifocal spikes. However, this reduction can be associated to the long-term FU and higher age, that both can

Please cite this article in press as: Sourbron J, et al., Vagus Nerve Stimulation in children: A focus on intellectual disability, European Journal of Paediatric Neurology (2017), http://dx.doi.org/10.1016/j.ejpn.2017.01.011

8

e u r o p e a n j o u r n a l o f p a e d i a t r i c n e u r o l o g y x x x ( 2 0 1 7 ) 1 e1 4

Fig. 3 e Forest plot of VNS effect size on seizure frequency reduction. ES ¼ effect size, CI ¼ 95% Confidence Interval, I2 ¼ statistical heterogeneity. Pooled analysis of responder rate of individual studies and overall effect size of VNS (≥50% seizure frequency reduction) 58% (CI 50%e65%). I2 ¼ 83.6%, p < 0.001. Since Arthur et al. (2006) had no responders and Zamponi et al. (2008) had only responders, we estimated in these cases the proportion and 95% CI using the inverse of the cumulative binomial distribution.

affect the EEG outcome.66 To conclude, EEG recordings were only done in three studies. Because there is a link between cognition and interictal epileptiform discharges, EEG recordings can be of significant value. Nonetheless, EEG recordings are not crucial to evaluate VNS efficacy12 since no

significant correlation with clinical improvements can be made and other developmental processes can bias the EEG outcome. The VNS guidelines confirm VNS as an alternative treatment option for drug-resistant epilepsy in addition to

Please cite this article in press as: Sourbron J, et al., Vagus Nerve Stimulation in children: A focus on intellectual disability, European Journal of Paediatric Neurology (2017), http://dx.doi.org/10.1016/j.ejpn.2017.01.011

e u r o p e a n j o u r n a l o f p a e d i a t r i c n e u r o l o g y x x x ( 2 0 1 7 ) 1 e1 4

conventional AED therapy, although the evidence is based on studies with important limitations.68 Three studies showed that having an intellectual disability was a negative prognostic factor in the seizure reducing effect of VNS.37,50,63 This conclusion was made since a better anti-seizure effect was observed in children with relatively less severe impairments. Accordingly, less VNS efficacy was noted in children with severe encephalopathy. Whereas Shahwan et al. noted that children with the most severe degree of ID were more likely to obtain a 50% SFR.42 Overall, the drug-resistant nature and the potential cognitive improvements (see below: 3.6 Quality of life) by VNS support this treatment option for children with epilepsy and intellectual disabilities.

3.6.

Quality of life (QoL)

VNS appeared to affect the mind, behavior and mood in a favorable manner (Table 3). These benefits were further underlined by the effective use of VNS in treatment-resistant depression.69 Although this review represents improvements on different QoL outcomes, it should be noted that this only reached statistical significance in less than one third of all studies (11/37). Of all these studies only 21 reported the methodology of QoL assessments (Table 3). As a result, it is more accurate to state that 52.4% (11/21) of the included VNS studies reported statistics on QoL changes, which were used to perform a meta-analysis.

3.6.1.

QoL assessment methods

The variety of methods for assessing the patient's QoL (Table 3) prevented to make a general conclusion. Sometimes it was even impossible to carry out specific assessments; e.g. cognitive tests70 or correct measurements of the children's discomforts, due to the severity of the intellectual impairments.47,71 Thus, uniform methods and questionnaires to assess QoL can ameliorate the comparison between several studies. Subsequently, this can facilitate future research to draw general conclusions. Moreover, the developmental status should be evaluated before the start of VNS to examine cognitive functioning changes after VNS implantation. We believe that this should be determined by specific validated measurements, depending on the child's level of functioning50 and age.33 Our review proposes the Bayley Scales of Infant Development (BSID, age: 1 monthe2 years), StanfordeBinet (SB, age: 2e6 years), Wechsler Preschool and Primary Scale of Intelligence (WPPSI-R, age: 2.5e7.5 years), Wechlser Intelligence Scales for Children (WISC, age: 6e16 years) and Wechsler Adult Intelligence Scale (WAIS, age: older than 16 years). Since drug-resistant epilepsies in children with ID can be accompanied by autistic features (e.g. TSC25), clinicians can also consider to assess this by using standardized diagnostic tools, such as the Autism Diagnostic Interview (ADI) and the Autism Diagnostic Observation Schedule (ADOS).

3.6.2.

QoL meta-analysis

Our meta-analysis only included those studies with statistically significant results and well described QoL assessment techniques (11/37 studies). Overall, the amount of patients with one or more function status improvements was greater

9

than those without any improvement. This difference appeared to be statistical significant (Chi-square test, p < 0.01; Fig. 4), which underlines the potential of VNS to improve the QoL.

3.6.3.

QoL and ID

Although we were able to statistically compare VNS efficacy (SFR, see above: 3.5 Efficacy) in patients with ID to those without, insufficient individual patient data were available to perform a similar meta-analysis of QoL between these two subgroups. Even though nearly all patients are intellectual disabled, the reported QoL improvements and the duration of treatment (mean FU of 12 months) suggest that VNS therapy is well tolerated despite the presence of ID or other unusual sensitivities (e.g. ASD72). Overall, these results accentuate that adequate management of epilepsy implies more than achieving a SFR. This is especially applicable for developing children who have additional psychosocial needs, compared to adults.

3.6.4.

QoL improvements and seizure control

The exact mechanism of action of VNS is still unknown. Consequently, the QoL improvements can be due to better seizure control and/or the intrinsic effect of VNS.31 Fifteen studies did not correlate the seizure reduction to QoL imKlinkenberg provements.12,22,63,40,21,24,34,25,56,46,49,38,26,27,23 et al. even documented a significant decrease in the Profile of Mood States (POMS) subscale on tension and depression in the non-responder group (instead of the responder group). Hence, these studies are in favor of a direct effect of VNS on some QoL parameters (irrespective of seizure control), which is further underlined by the use of VNS in treatmentresistant depression and Alzheimer.42,73 On the contrary, there is some evidence that QoL improvements are associated with better seizure control, related to less seizure burden.23,35,38

3.6.5.

Limitations of QoL assessments

The reported QoL measurements should be interpreted with caution due to some limitations. First, QoL improvements were often obtained by unpublished measures of which the quality is not validated and often subjective. In addition, the observed differences were mostly not statistically significant and QoL outcomes are often no primary endpoint. In spite of these limitations we summarized all QoL data and clearly highlighted improvements in attention, alertness and communicative skills. Due to the limited number of patients in a lot of the studies, statistical analyses were often not possible. Nonetheless, some studies did obtain sufficient data but did not perform statistical analyses on the QoL outcomes.22,35,47,51,55,74 QoL was usually reported by multiple caregivers and/or the child's parents (or guardians). Self-reports are very limited due to the age and the developmental level of these children. We believe that the group of Zamponi obtained QoL assessments most objectively by two psychologists in a single-blinded fashion, i.e. at the two FU sessions by a different investigator, not aware of the baseline assessments.25,33,34,46,75 In summary, most of these QoL assessments were done in a subjective and not clearly quantifiable manner, often by

Please cite this article in press as: Sourbron J, et al., Vagus Nerve Stimulation in children: A focus on intellectual disability, European Journal of Paediatric Neurology (2017), http://dx.doi.org/10.1016/j.ejpn.2017.01.011

10

First author

Year

Assessment timing

Alexopoulos Andriola Arhan Arthur Blount

2006 2001 2010 2006 2006

ND Pre- (baseline); post- (1e3 m interval) ND ND ND

ND Changes in a subset of QoL parameters: NQ ND ND ND

Assessment technique

ND Improved (NS) (Alertness, Tasks)* ND ND ND

Buoni Chen Cukiert Elliot Elliot Frost

2004 2012 2013 2011 2009 2001

ND Pre- (baseline); post- (ND) Pre- (baseline); post- (ND) ND ND Pre- (baseline); post- (1, 3, and/or 6 m)

€o €k Hallbo

2005

Pre- (baseline); post- (3 and 9 m)

Improved (Alertness) Improved (Language, Mood, Tasks)* Improved (NS) (Attention)* ND ND Improved (NS) (Alertness, Seizure clustering, Tasks)* Improved

Hosain Khurana Kim Klinkenberg

2000 2007 2015 2012 2013

Lundgren MajkowskaZwolinska Majoie

1998 2012

ND ND ND Pre- (baseline, 3 m before); post- (5 and 10 m, respectively end of blinded and end of open-label phase) Pre- (baseline); post- (ND) ND

Changes in a subset of QoL parameters: NQ BSID or WPPSI-R (developmental status) QOLIE-31 (QoL); SNAP-IV (attention level, verbal fluency) ND ND Changes in a subset of QoL parameters: 5-point rating scale (simple, unvalidated) BSID, WPPSI-R or WISC (developmental status); VAS (QoL); CBCL (behavior); Dodrill Mood Analogue Scale and DSRS (mood) ND ND ND PPVTeIIIeNL and Beery VMI (cognition, performance-based); POMS (mood and depression), HARCES (epilepsy-based restrictions); PARS-III (psychosocial adjustments) VAS (QoL) ND

2001

Pre- (baseline); post- (ND)

Major Mikati Nagarajan

2008 2009 2002

Pre- (baseline); post- (ND) Pre- (baseline); post- (ND) Pre- (baseline); post- (3e6 m interval)

Orosz

2015

Pre- (baseline); post- (12 and 24 m)

Parain Park (a, LKS)

2001 2003

Pre- (baseline); post- (ND) Pre- (baseline); post- (3 and 6 m)

Changes in a subset of QoL parameters: 3-point rating scale; CGI-I (health outcomes) ND Changes in a subset of QoL parameters: 3-point rating scale

Park (b, ASD)

2003

Pre- (baseline); post- (3 and 12 m)

Changes in a subset of QoL parameters: 3-point rating scale

Parker Rychlicki Shahwan Wakai

2001 2006 2009 2001

ND ND ND Changes in a subset of QoL parameters: NQ

Wilfong You You

2006 2008 2007

ND ND ND Pre- (baseline, 3 m before); post- (1 m and 3e11 m interval) Pre- (baseline); post- (12 m) ND Pre- (baseline); post- (12 m or >12 m)

BSID, MSCA, WISC (developmental status), performance-based); SRZ score, SGZ score, AVZ-R score, TVZ score (QoL, questionnairebased) ND WISC and DDST (developmental status); CEQ-P II, QOLCE (QoL) Changes in a subset of QoL parameters: 3-point rating scale

Changes in a subset of QoL parameters: NQ ND K-QOLCE (QoL)

QoL outcome

ND ND ND Improved (Mood)

Improved ND Improved (NS) (Memory, Mood)*

Improved (NS) (Behavior, Memory)* Improved (NS) Improved (NS) (Alertness, Behavior, Language, Tasks)* Improved (Alertness, Behavior, Language, Memory, Mood, Tasks) Improved (NS) (Alertness)* Improved (Alertness, Tasks, Postictal period, Seizure clustering) Improved (Alertness, Mood, Tasks, Postictal period, Seizure clustering) ND ND ND Improved (NS) (Postictal period, Memory)* Improved (Alertness) ND Improved (Alertness, Behavior, Language, Memory, Mood, Tasks)

e u r o p e a n j o u r n a l o f p a e d i a t r i c n e u r o l o g y x x x ( 2 0 1 7 ) 1 e1 4

Please cite this article in press as: Sourbron J, et al., Vagus Nerve Stimulation in children: A focus on intellectual disability, European Journal of Paediatric Neurology (2017), http://dx.doi.org/10.1016/j.ejpn.2017.01.011

Table 3 e Quality of life (QoL).

QoL ¼ quality of life; ND ¼ not determined; m ¼ month(s); NQ ¼ not clearly quantifiable; NS ¼ not statistically significant; * ¼ reports of improvement; BSID ¼ Bayley Scales of Infant Development; WPPSI-R ¼ Wechsler Preschool and Primary Scale of Intelligence Revised; ID ¼ intellectual disability; QOLIE-31 ¼ Quality of Life in Epilepsy Inventory (31 items-questionnaire); SNAP ¼ Swanson, Nolan and Pelham questionnaire; WISC ¼ Wechlser Intelligence Scales for Children; VAS ¼ Visual Analogue Scale; CBCL ¼ Child Behavior Checklist; DSRS ¼ Birleson Depression Self-Rating Scale; PPVTeIIIe NL ¼ Peabody Picture Vocabulary test III(NL ¼ dutch); Beery VMI ¼ Beery Visual Motor Integration test; POMS ¼ Profile of Mood States; HARCES ¼ Hague Restrictions in Childhood Epilepsy Scale; PARSIII ¼ Personal Adjustment and Role Skills Scale III; MSCA ¼ McCarthy Scales of Children's Abilities; SRZ ¼ sociale redzaamheidsschaal voor zwakzinnigen (dutch); SGZ ¼ schaal voor zwakzinnigen (dutch); AVZR ¼ Autisme-en verwante Stoornissenschaal Revisie (dutch); TVZ ¼ temperamentschaal voor zwakzinnigen (dutch); DDST ¼ Denver Developmental Screening Test; CEQ-P II ¼ Child Epilepsy Questionnaire Parental Form II, QOLCE ¼ Quality of Life in Childhood Epilepsy Questionnaire; CGI-I ¼ Clinical Global Impression of Improvement; K-QOLCE ¼ Korean version of the Quality of Life in Childhood Epilepsy questionnaire; SB ¼ StanfordeBinet scale; WAIS ¼ Wechsler Adult Intelligence Scale; VABS ¼ Vineland Adaptive Behavior Scales; DS ¼ Dravet syndrome; TSC ¼ tuberous sclerosis complex; OAS ¼ Overt Aggression Scale; ADI ¼ Autism Diagnostic Interview; ADOS ¼ Autism Diagnostic Observation Schedule; FU ¼ follow-up.

2010 2008 2002 Zamponi (TSC) Zamponi Zamponi

Pre- (baseline); post- (12 m) Pre- (baseline); post- (>12 m, i.e. at FU) Pre- (baseline); post- (12 m)

Eq. Zamponi et al. 2011; OAS (QoL); ADI, ADOS (autistic symptoms) VABS and analogical scale of parental satisfaction (QoL) VABS (QoL)

Improved (NS) (Alertness, Behavior, Tasks) * Improved (Alertness, Behavior, Tasks) Improved (NS)* Improved 2011 Zamponi (DS)

Pre- (baseline); post- (12 m)

2011 Zamponi

Pre- (baseline); post- (12 and 36 m)

SB, WISC or WAIS (developmental status); VABS and analogical scale of parental satisfaction (QoL) Eq. Zamponi et al. 2011

Improved (NS)*

e u r o p e a n j o u r n a l o f p a e d i a t r i c n e u r o l o g y x x x ( 2 0 1 7 ) 1 e1 4

11

unverified scales51,57 which reflects the significant shortage of validated QoL measurement techniques (Table 3).

3.6.6.

Conclusion

The reported QoL improvements by VNS are from added value since the most common, first-line treatment of epilepsy (AED) is known to possibly cause adverse cognitive, behavioral and/ or mood related effects. Frost et al. even concluded that improvements in QoL appeared to be more life affecting than the reduction of seizures.51 Although the QoL assessment techniques in this review are heterogeneous, the clear absence of a QoL decline might be an advantage over standard AED therapy.33,68 It is worth noting that clinical relevant improvements are not always statistically significant which can underestimate VNS0 effects on QoL (e.g. one out of three patients had a clinically relevant improvement in adaptive behavior that could not be demonstrated by statistical tests33).

3.7.

Safety

Side effects were categorized into two groups, i.e. minor (mild and transient) and major (requiring additional medication and/or surgery). Minor side effects were documented in 30 out of 37 studies (total of 974 patients) and major side effects were documented in 31 out of 37 studies (total of 992 patients). Overall, less than 24.0% of all patients experienced minor side effects (n ¼ 230/974 patients). These side effects occur often at the beginning of the treatment and dissolve spontaneous over time, or diminish or disappear by adjusting the VNS parameters: hoarseness, cough and local discomfort were mostly reported. Only 5.0% of all patients (n ¼ 30/992 patients) experienced major side effects, mostly an infection. Serious complications were very limited but need to be considered as implantation risk.76,77 The current review highlights VNS as a safe and well tolerated treatment option for drug-resistant epilepsy patients who are not eligible for surgery. Additionally, VNS leads to significant less severe side effects on cognition and consciousness, when compared to several commonly used AEDs. Although the number of AEDs has increased greatly the past decade, a relative amount of epilepsy patients is not able to control their seizures and the AED side effect profile has not improved significantly.11 An important advantage of VNS is the absence of pharmacokinetic and pharmacodynamic interactions.11,78 Besides, VNS might even slightly improve the patient's cognition and function, which underlines the importance to consider this treatment option in epilepsy patients with ID.31 Numerous studies mentioned a postoperative two-week lag before starting VNS treatment to minimize adverse events.76 Thereafter the stimulation is usually step-wise increased by 0.25 mA to avoid adverse events. Although the surgery and start of VNS therapy are conducted with caution, it is recommended to obtain a detailed evaluation of the VNS candidate in advance. The constellation of breathing problems, severe swallowing problems, gastro-intestinal diseases and obstructive sleep apneas could limit the use of VNS.51,76 Some drug-resistant epilepsy syndromes, e.g. Rett syndrome, are characterized by breathing irregularities and

Please cite this article in press as: Sourbron J, et al., Vagus Nerve Stimulation in children: A focus on intellectual disability, European Journal of Paediatric Neurology (2017), http://dx.doi.org/10.1016/j.ejpn.2017.01.011

12

e u r o p e a n j o u r n a l o f p a e d i a t r i c n e u r o l o g y x x x ( 2 0 1 7 ) 1 e1 4

Fig. 4 e Percentage of children with a significant QoL improvement. A significant higher amount of patients with QoL improvement, compared to those without any improvement (**p < 0.01; Chi-square test).

oropharyngeal dysfunction. In addition, severe intellectual retardation can lead to swallowing difficulties. Consequently, closely monitoring of children with these particular risk factors is advised. As a result, some studies even suggested to imply special screening methods before considering VNS, i.e. detailed sleep questionnaires and even a baseline polysomnography (PSG) in patients with a history of snoring.58 In conclusion, the abovementioned risk factors did not make VNS therapy impossible. For example, swallowing problems during meals could be reduced by switching off the stimulator (applying magnet over device).30 Better control of adverse events is even possible by controlling the current output, as suggested by Chen and colleagues.68 It is worth noting that the presence of intellectual disabilities can hinder a proper assessment of the child's discomforts.68 Finally, the most frequently encountered adverse events were well tolerated, transient and unlikely to cause discontinuation of treatment.

patients in each study. Hence, future large-scale, multicenter, RCTs with a special focus on this subpopulation, are recommended.

Conflicts of interest None.

Acknowledgements This project was carried out with the support of Zogenix and the Agency for Innovation by Science and Technology (Grant number: 131179, IWT, Flanders).

Appendix A. Supplementary data 4.

Conclusions

To our knowledge, this is the first review of VNS that exclusively presents the outcome data of pediatric patients (
18 years of age) and focuses on the patients with intellectual disabilities. We demonstrated a significant difference of VNS effectiveness (regarding seizure frequency reduction (SFR)) in favor of children with normal cognitive abilities compared to children with ID. Nonetheless, the absence of side effects and drug interactions highlight some of the advantages of VNS, compared to adding an extra AED. In addition, some studies showed relevant QoL improvements, irrespective of seizure control, which should encourage pediatric neurologists to consider VNS early on in treating drug-resistant epilepsy, even accompanied by ID. Regardless, this meta-analysis shows that VNS data in the pediatric population are scarce and need to be interpreted carefully since the majority of studies are retrospective, which can imply the shortage of consistent follow-up and a high degree of subjectivity in the patient records. Furthermore, these studies include heterogeneous series of patients with different types of epilepsy and often a limited number of

Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.ejpn.2017.01.011.

references

1. Iliescu C, Craiu D. Diagnostic approach of epilepsy in childhood and adolescence. Maedica 2013;8:195e9. 2. Engel J. Seizures and epilepsy. Oxford University Press; 2013. 3. Tuchman R, Moshe SL, Rapin I. Convulsing toward the pathophysiology of autism. Brain Dev 2009;31:95e103. 4. Devinsky O, Asato M, Camfield P, et al. Delivery of epilepsy care to adults with intellectual and developmental disabilities. Neurology 2015;85:1512e21. 5. Jackson CF, Makin SM, Marson AG, et al. Pharmacological interventions for epilepsy in people with intellectual disabilities. Cochrane Database Syst Rev 2015;9. CD005399eCD005399. 6. Alvarez N, Besag F, Iivanainen M. Use of antiepileptic drugs in the treatment of epilepsy in people with intellectual disability. J Intellect Disabil Res JIDR 1998;42(Suppl. 1):1e15. 7. Neal EG, Chaffe H, Schwartz RH, et al. The ketogenic diet for the treatment of childhood epilepsy: a randomised controlled trial. Lancet Neurol 2008;7:500e6.

Please cite this article in press as: Sourbron J, et al., Vagus Nerve Stimulation in children: A focus on intellectual disability, European Journal of Paediatric Neurology (2017), http://dx.doi.org/10.1016/j.ejpn.2017.01.011

e u r o p e a n j o u r n a l o f p a e d i a t r i c n e u r o l o g y x x x ( 2 0 1 7 ) 1 e1 4

8. Handforth A, DeGiorgio CM, Schachter SC, et al. Vagus nerve stimulation therapy for partial-onset seizures: a randomized active-control trial. Neurology 1998;51:48e55. 9. Labar D, Murphy J, Tecoma E. Vagus nerve stimulation for medication-resistant generalized epilepsy. E04 VNS Study Group. Neurology 1999;52:1510e2. 10. No authors listed. A randomized controlled trial of chronic vagus nerve stimulation for treatment of medically intractable seizures. The Vagus Nerve Stimulation Study Group. Neurology 1995;45:224e30. 11. Cross JH, Kluger G, Lagae L. Advancing the management of childhood epilepsies. Eur J Paediatr Neurol EJPN Off J Eur Paediatr Neurol Soc 2013;17:334e47. 12. Klinkenberg S, Aalbers MW, Vles JSH, et al. Vagus nerve stimulation in children with intractable epilepsy: a randomized controlled trial. Dev Med Child Neurol 2012;54:855e61. 13. Cross JH, Jayakar P, Nordli D, et al. Proposed criteria for referral and evaluation of children for epilepsy surgery: recommendations of the subcommission for pediatric epilepsy surgery. Epilepsia 2006;47:952e9. 14. Connor Jr DE, Nixon M, Nanda A, et al. Vagal nerve stimulation for the treatment of medically refractory epilepsy: a review of the current literature. Neurosurg Focus 2012;32. E12eE12. 15. Terry R. Vagus Nerve Stimulation: a proven therapy for treatment of epilepsy strives to improve efficacy and expand applications. In: Conference proceedings: annual International Conference of the IEEE engineering in medicine and biology society; 2009. p. 4631e4. IEEE Engineering in Medicine and Biology Society. Annual Conference 2009. 16. Fitzgerald PB. Non-pharmacological biological treatment approaches to difficult-to-treat depression. Med J Aust 2013;199:S48e51. 17. 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;53:1731e5.  MJ, Luckasson R, Nygren M. AAIDD proposed 18. Tasse recommendations for ICD-11 and the condition previously known as mental retardation. Intellect Dev Disabil 2013;51:127e31. 19. Alexopoulos AV, Kotagal P, Loddenkemper T, et al. Long-term results with vagus nerve stimulation in children with pharmacoresistant epilepsy. Seizure 2006;15:491e503. 20. Arhan E, Serdaroglu A, Kurt G, et al. The efficacy of vagal nerve stimulation in children with pharmacoresistant epilepsy: practical experience at a Turkish tertiary referral center. Eur J Paediatr Neurol 2010;14:334e9. 21. Elliott RE, Carlson C, Kalhorn SP, et al. Refractory epilepsy in tuberous sclerosis: vagus nerve stimulation with or without subsequent resective surgery. Epilepsy Behav 2009;16:454e60. 22. Lundgren J, Amark P, Blennow G, et al. Vagus nerve stimulation in 16 children with refractory epilepsy. Epilepsia 1998;39:809e13. 23. Mikati MA, Ataya NF, El-ferezli JC, et al. Quality of life after vagal nerve stimulator insertion. Epileptic Disord 2009;11:67e74. 24. Wakai S, Kotagal P. Vagus nerve stimulation for children and adolescents with intractable epilepsies. Pediatr Int 2001;43:61e5. 25. Zamponi N, Petrelli C, Passamonti C, et al. Vagus nerve stimulation for refractory epilepsy in tuberous sclerosis. Pediatr Neurol 2010;43:29e34. 26. Buoni S, Mariottini A, Pieri S, et al. Vagus nerve stimulation for drug-resistant epilepsy in children and young adults. Brain Dev 2004;26:158e63.

13

27. Andriola MR, Vitale SA. Vagus nerve stimulation in the developmentally disabled. Epilepsy Behav 2001;2:129e34. 28. Hosain S, Nikalov B, Harden C, et al. Vagus nerve stimulation treatment for Lennox-Gastaut syndrome. J Child Neurol 2000;15:509e12. 29. Kim HJ, Kim HD, Lee JS, et al. Long-term prognosis of patients with LennoxeGastaut syndrome in recent decades. Epilepsy Res 2015;110:10e9. 30. Lundgren J, Ekberg O, Olsson R. Aspiration: a potential complication to vagus nerve stimulation. Epilepsia 1998;39:998e1000. 31. Major P, Thiele EA. Vagus nerve stimulation for intractable epilepsy in tuberous sclerosis complex. Epilepsy Behav 2008;13:357e60. 32. Parain D, Penniello MJ, Berquen P, et al. Vagal nerve stimulation in tuberous sclerosis complex patients. Pediatr Neurol 2001;25:213e6. 33. Zamponi N, Passamonti C, Cesaroni E, et al. Effectiveness of vagal nerve stimulation (VNS) in patients with drop-attacks and different epileptic syndromes. Seizure 2011;20:468e74. 34. Zamponi N, Passamonti C, Cappanera S, et al. Clinical course of young patients with Dravet syndrome after vagal nerve stimulation. Eur J Paediatr Neurol EJPN Off J Eur Paediatr Neurol Soc 2011;15:8e14. 35. Nagarajan L, Walsh P, Gregory P, et al. VNS therapy in clinical practice in children with refractory epilepsy. Acta Neurol Scand 2002;105:13e7. 36. Lagae L, Verstrepen A, Nada A, et al. Vagus nerve stimulation in children with drug-resistant epilepsy: age at implantation and shorter duration of epilepsy as predictors of better efficacy? Epileptic Disord Int Epilepsy J Videotape 2015;17:308e14. 37. Parker AP, Polkey CE, Robinson RO. Vagal nerve stimulation in the epileptic encephalopathies: 3-year follow-up. Pediatrics 2001;108. 221e221. 38. Elliott RE, Rodgers SD, Bassani L, et al. Vagus nerve stimulation for children with treatment-resistant epilepsy: a consecutive series of 141 cases. J Neurosurg Pediatr 2011;7:491e500. 39. Orosz I, McCormick D, Zamponi N, et al. Vagus nerve stimulation for drug-resistant epilepsy: a European long-term study up to 24 months in 347 children. Epilepsia 2014;55:1576e84. 40. Cukiert A, Cukiert CM, Burattini JA, et al. Long-term outcome after callosotomy or vagus nerve stimulation in consecutive prospective cohorts of children with Lennox-Gastaut or Lennox-like syndrome and non-specific MRI findings. Seizure 2013;22:396e400. 41. Englot DJ, Chang EF, Auguste KI. Vagus nerve stimulation for epilepsy: a meta-analysis of efficacy and predictors of response. J Neurosurg 2011;115:1248e55. 42. Shahwan A, Bailey C, Maxiner W, et al. Vagus nerve stimulation for refractory epilepsy in children: more to VNS than seizure frequency reduction. Epilepsia 2009;50:1220e8. 43. Luders HO, Amina S, Baumgartner C, et al. Modern technology calls for a modern approach to classification of epileptic seizures and the epilepsies. Epilepsia 2012;53:405e11. 44. Blount JP, Tubbs RS, Kankirawatana P, et al. Vagus nerve stimulation in children less than 5 years old. Child's Nerv Syst 2006;22:1167e9. 45. Mahncke HW, Bronstone A, Merzenich MM. Brain plasticity and functional losses in the aged: scientific bases for a novel intervention. Prog Brain Res 2006;157:81e109. 46. Zamponi N, Rychlicki F, Cardinali C, et al. Intermittent vagal nerve stimulation in paediatric patients: 1-year follow-up. Child's Nerv Syst ChNS Off J Int Soc Pediatr Neurosurg 2002;18:61e6.

Please cite this article in press as: Sourbron J, et al., Vagus Nerve Stimulation in children: A focus on intellectual disability, European Journal of Paediatric Neurology (2017), http://dx.doi.org/10.1016/j.ejpn.2017.01.011

14

e u r o p e a n j o u r n a l o f p a e d i a t r i c n e u r o l o g y x x x ( 2 0 1 7 ) 1 e1 4

47. You SJ, Kang HC, Kim HD, et al. Vagus nerve stimulation in intractable childhood epilepsy: a Korean multicenter experience. J Korean Med Sci 2007;22:442e5. 48. You SJ, Kang HC, Ko TS, et al. Comparison of corpus callosotomy and vagus nerve stimulation in children with Lennox-Gastaut syndrome. Brain Dev 2008;30:195e9.  ska B, Zwolin  ski P, Roszkowski M, et al. 49. Majkowska-Zwolin Long-term results of vagus nerve stimulation in children and adolescents with drug-resistant epilepsy. Child's Nerv Syst 2012;28:621e8. €o € k T, Lundgren J, Stjernqvist K, et al. Vagus nerve 50. Hallbo stimulation in 15 children with therapy resistant epilepsy; its impact on cognition, quality of life, behaviour and mood. Seizure 2005;14:504e13. 51. Frost M, Gates J, Helmers SL. Vagus nerve stimulation in children with refractory seizures associated with LennoxeGastaut syndrome. Epilepsia 2001;42:1148e52. 52. Hoppe C, Wagner L, Hoffmann JM, et al. Comprehensive longterm outcome of best drug treatment with or without add-on vagus nerve stimulation for epilepsy: a retrospective matched pairs case-control study. Seizure 2013;22:109e15. 53. Koo B, Ham SD, Sood S, et al. Human vagus nerve electrophysiology: a guide to vagus nerve stimulation parameters. J Clin Neurophysiol 2001;18:429e33. 54. Heck C, Helmers SL, DeGiorgio CM. Vagus nerve stimulation therapy, epilepsy, and device parameters: scientific basis and recommendations for use. Neurology 2002;59:S31e7. 55. Helmers SL, Wheless JW, Frost M, et al. Vagus nerve stimulation therapy in pediatric patients with refractory epilepsy: retrospective study. J Child Neurol 2001;16:843e8. 56. Rossignol E, Lortie A, Thomas T, et al. Vagus nerve stimulation in pediatric epileptic syndromes. Seizure 2009;18:34e7. 57. Wilfong AA, Schultz RJ. Vagus nerve stimulation for treatment of epilepsy in Rett syndrome. Dev Med Child 2006;48:683e6. 58. Khurana DS, Reumann M, Hobdell EF, et al. Vagus nerve stimulation in children with refractory epilepsy: unusual complications and relationship to sleep-disordered breathing. Child's Nerv Syst ChNS Off J Int Soc Pediatr Neurosurg 2007;23:1309e12. 59. Sun FT, Morrell MJ. Closed-loop neurostimulation: the clinical experience. Neurotherapeutics 2014;11:553e63. 60. Morris 3rd GL, Gloss D, Buchhalter J, et al. Evidence-based guideline update: vagus nerve stimulation for the treatment of epilepsy: report of the guideline development subcommittee of the American Academy of Neurology. Neurology 2013;81:1453e9. 61. Panebianco M, Rigby A, Weston JMA, et al. Vagus nerve stimulation for partial seizures. Cochrane Libr 2015;4. CD002896eCD002896.  simo RO, Bartuluchi M, De Los Santos C, et al. Vagus 62. Cerso nerve stimulation: effectiveness and tolerability in patients

63.

64.

65.

66.

67. 68. 69.

70.

71.

72.

73.

74.

75.

76.

77.

78.

with epileptic encephalopathies. Child's Nerv Syst ChNS Off J Int Soc Pediatr Neurosurg 2011;27:787e92. Majoie HJM, Berfelo MW, Aldenkamp AP, et al. Vagus nerve stimulation in children with therapy-resistant epilepsy diagnosed as Lennox-Gastaut syndrome. J Clin Neurophysiol 2001;18:419e28. Rychlicki F, Zamponi N, Trignani R, et al. Vagus nerve stimulation: clinical experience in drug-resistant pediatric epileptic patients. Seizure 2006;15:483e90. Kerr M, Bowley C. Multidisciplinary and multiagency contributions to care for those with learning disability who have epilepsy. Epilepsia 2001;42:55e6. Discussion 57e58. €o € k T, Lundgren J, Blennow G, et al. Long term effects on Hallbo epileptiform activity with vagus nerve stimulation in children. Seizure 2005;14:527e33. Fisher RS, Nune G, Roberts SE, et al. The personal impact of epilepsy scale (PIES). Epilepsy Behav 2015;42:140e6. Dugan P, Devinsky O. Epilepsy: guidelines on vagus nerve stimulation for epilepsy. Nat Rev Neurol 2013;9:611e2. Shah A, Carreno FR, Frazer A. Therapeutic modalities for treatment resistant depression: focus on vagal nerve stimulation and ketamine. Clin Psychopharmacol Neurosci Off Sci J Korean Coll Neuropsychopharmacol 2014;12:83e93. Klinkenberg S, Van den Bosch CN, Majoie HJ, et al. Behavioural and cognitive effects during vagus nerve stimulation in children with intractable epilepsy e a randomized controlled trial. Eur J Paediatr Neurol 2013;17:82e90. Chen C-Y, Lee H-T, Chen C-C, et al. Short-term results of vagus nerve stimulation in pediatric patients with refractory epilepsy. Pediatr Neonatol 2012;53:184e7. Park YD. The effects of vagus nerve stimulation therapy on patients with intractable seizures and either Landau-Kleffner syndrome or autism. Epilepsy Behav 2003;4:286e90. Vonck K, Raedt R, Naulaerts J, et al. Vagus nerve stimulation… 25 years later! What do we know about the effects on cognition? Neurosci Biobehav Rev 2014;45:63e71. Hornig GW, Murphy JV, Schallert G, et al. Left vagus nerve stimulation in children with refractory epilepsy: an update. South Med J 1997;90:484e8. Zamponi N, Rychlicki F, Corpaci L, et al. Vagus nerve stimulation (VNS) is effective in treating catastrophic 1 epilepsy in very young children. Neurosurg Rev 2008;31:291e7. Rychlicki F, Zamponi N, Cesaroni E, et al. Complications of vagal nerve stimulation for epilepsy in children. Neurosurg Rev 2006;29:103e7. Revesz D, Rydenhag B, Ben-Menachem E. Complications and safety of vagus nerve stimulation: 25 years of experience at a single center. J Neurosurg Pediatr 2016;18:97e104. http:// dx.doi.org/10.3171/2016.3171.PEDS15534. Epub 12016 Mar 15525. Tanaka E. Clinically significant pharmacokinetic drug interactions between antiepileptic drugs. J Clin Pharm Ther 1999;24:87e92.

Please cite this article in press as: Sourbron J, et al., Vagus Nerve Stimulation in children: A focus on intellectual disability, European Journal of Paediatric Neurology (2017), http://dx.doi.org/10.1016/j.ejpn.2017.01.011