Is functional electrical stimulation an alternative for orthotics in patients with cerebral palsy? A literature review

Transcription

1 european journal of paediatric neurology 22 (2018) 7e16 Official Journal of the European Paediatric Neurology Society Review article Is functional electrical stimulation an alternative for orthotics in patients with cerebral palsy? A literature review Sam Khamis a,*, Talia Herman b, Sima Krimus a, Barry Danino a,c a The Gait and Motion Analysis Laboratory, Department of Pediatric Orthopaedics, Dana Children's Hospital, Tel Aviv Sourasky Medical Center, Tel Aviv, Israel b Center for the Study of Movement, Cognition, and Mobility (CMCM), Department of Neurology, Tel-Aviv Sourasky Medical Center, Tel Aviv, Israel c Department of Pediatric Orthopaedics, Dana Children's Hospital, Tel Aviv Sourasky Medical Center, Tel Aviv, Israel article info Article history: Received 22 December 2016 Received in revised form 9 August 2017 Accepted 8 October 2017 Keywords: Cerebral palsy Functional electrical stimulation Ankle foot orthoses abstract Background: Functional electrical stimulation (FES) is a well-known intervention used during walking to improve walking abilities and correct gait deviations by facilitating the proper muscle group at the appropriate timing in the gait cycle. Our aim was to study the types of surface FES currently used in a cerebral palsy (CP) population and examine the evidence of its ability to improve gait deviations, functional ability and therapeutic effects. Methods: A computerized database search was conducted from inception until 6/2016. Included were all clinical trials performing gait analysis of children with CP applying surface FES to any lower leg muscles evaluating the efficiency of the stimulation and any carry-over effect. Results: Fifteen studies met the inclusion criteria. The most common FES stimulated the dorsi flexors muscles with a positive orthotic effect, improved dorsi flexion during the swing phase and enhanced the foot contact pattern. A smaller positive effect was found for knee extensors stimulation facilitating knee extension during the stance phase and for hip abductors stimulation improving frontal plane knee alignment. No evidence was found to support the use of plantar flexors stimulation in correcting gait deviations. There is scarce evidence of any retention effect. Conclusion: We encourage the clinician to evaluate the use of FES on a case to case basis. Controlled investigations with larger numbers of participants are warranted to determine the orthotic and therapeutic efficacy of FES European Paediatric Neurology Society. Published by Elsevier Ltd. All rights reserved. * Corresponding author. The Gait and Motion Analysis Laboratory, Department of Pediatric Orthopaedics, Dana Children's Hospital, Tel Aviv Sourasky Medical Center, 6 Weizman St., Tel Aviv, Israel. Fax: þ address: khamisam@gmail.com (S. Khamis) / 2017 European Paediatric Neurology Society. Published by Elsevier Ltd. All rights reserved.

2 8 european journal of paediatric neurology 22 (2018) 7e16 Contents 1. Introduction Methods Inclusions and exclusion criteria of the reviewed studies Inclusion Exclusion Data analysis Results Study design Patient selection Muscle groups stimulation Protocol Common inclusion criteria for studies using FES Main outcomes Discussion Types of surface FES presently applied Does surface FES improve gait deviations (an orthotic effect)? Were there any functional advantages to using FES? Are there criterions for fitting FES? Does the use of an FES device produce a carry-over/retention effect or only an orthotic effect? Conclusions Funding source Conflicts of interest Acknowledgements References Introduction Cerebral palsy (CP), occurring in the early stages of development, is a permanent motor disorder attributable to a nonprogressive defect or brain lesion (4th International Congress of the Study Group on Child Neurology and Cerebral Palsy, Oxford 1964) encompassing distinct symptoms (i.e movement and postural disorders) rather than a specific diagnosis. 1 Neuromuscular impairments in CP include a number of disturbances which may compromise gait function i.e., spasticity and muscle tendon contractures, bony deformation, weakness, diminished motor control and impaired proprioception. Diverse treatment options such as physical and occupational therapy may promote psychomotor development and help prevent secondary deformities. Braces stabilizing the skeleton in an anatomical position during weight-bearing may produce the same effect. Additionally, orthopedic surgery is often performed to correct different types of gait problems and deviations. The outcomes of surgical interventions may vary and are sometimes difficult to predict. 1,2 Ankle foot orthoses (AFOs) are typically used to correct flexible ankles or subtalar malposition deformities as well as controlling proximal joints. AFOs can prevent excessive motions such as hinged AFOs which enable dorsiflexion to occur during the stance phase of the gait cycle, while preventing plantar flexion during the swing phase. They can restrict range of motion (ROM) and prevent abnormal movement i.e. fixed AFOs, preventing plantarflexion and dorsiflexion movement which consequently supports the extensor moment and thus, prevents excessive dorsiflexion and knee flexion during the stance phase. AFOs can support weak muscles, i.e. the posterior leaf spring orthotic, thereby, facilitating support during the stance phase and provides a plantar flexion assist at toe off. Nevertheless, when prescribing an orthotic intervention for a patient, it should be clear as to the type and location of the deviation in conjunction with the timing of its occurrence during the gait cycle. Teenaged children with CP, especially those with mild pathology involvement, usually find braces cumbersome and unappealing and may request a more discrete device. Functional electrical stimulation system (FES), a neuroprosthesis device, is a well-known intervention utilized for many years to support muscle groups during walking. It delivers electrical stimulations to a motor nerve which stimulates a muscle group to overcome functional obstacle during gait - for example a stimulation to the common peronealnerveduringswingphaseofgait,causestheankle to dorsiflex and thus preventing foot drop. 9 It enables improvement in walking ability and corrects gait deviations by recruiting the proper muscle groups at the appropriate timing in the gait cycle. One of the potential benefits of the FESdeviceoverbracesisitssmallsize,easyuseandhightech that might be more appealing to the patients. FES endeavors to improve walking appearance, ability and correct gait deviations, thus the intensity setting of the stimulation

3 european journal of paediatric neurology 22 (2018) 7e16 9 is determined by achieving the required motor action. Surface FES is the preferable type used due to its feasibility and comfortability. In general, FES attempts to increase muscle strength, reduce muscle spasticity and improve movement control/movement pattern. The efficacy can be demonstrated by the improvement in temporal parameters, 9 however limited data is available to support the efficacy of FES in improving muscle strength and reducing spasticity. It may also serve as an orthotic device used while performing routine daily activities including walking. At present, FES is widely used to control dorsi flexors muscles and prevent drop foot during the swing phase. 3,4 Since data in the literature relating to FES is sporadic and various forms of FES are delivered to the lower extremities for children with CP, a review of the literature was conducted. Questions raised in the present review were: (1) what types of surface FES are currently used in CP patients; (2) does surface FES improve gait deviations; (3) is there any functional advantage in using FES; (4) what are the criterions for fitting FES; and (5) is there a carry-over or therapeutic effect using FES or only an orthotic effect? 2. Methods An exhaustive search of a computerized database consisting of Pubmed, Cochrane Database of Systematic Review, EMBASE, Web of Science, Pedro and CINAHL was conducted from inception until June Key search words included: cerebral palsy, gait analysis, functional electrical stimulation, gait and walking. Titles and abstracts were screened by one reviewer to test their suitability for this review. In addition, relevant articles listed as references in the retrieved articles were also included. The search was restricted to published clinical trials written in English. Full paper copies of all references were retrieved and then reviewed by a second assessor Inclusions and exclusion criteria of the reviewed studies Inclusion All clinical trials performing gait analysis evaluation in children <18 years old with CP, who had surface FES applied to any lower leg muscle during walking, were included. Studies comparing gait with and without FES in order to evaluate the efficiency of the stimulation, any carry-over effects and any studies with secondary therapeutic aims i.e., improving range of motion or decreasing spasticity, were also included Exclusion Studies evaluating the effect of percutaneous electrical stimulation, studies of subjects with any additional neurological conditions other than CP and studies evaluating the simultaneous effect of a combination of more than one muscle group, were excluded Data analysis The articles included in this review describing the study design, participants, FES stimulator parameters and protocols used for each muscle group simulation, are summarized in Table 1. The clinical evaluations chosen as inclusion criteria for the study participants were documented in order to evaluate their influence on the findings. We presented our results as: 1. the orthotic effect, which is the FES effect on gait deviations, spatiotemporal or any other gait parameters during the actual use of FES; and 2. the retention effect, i.e., a FES carryover effect after the immediate withdrawal of the stimulation or after a predefined period of disuse. The results section below incorporates changes in gait parameters, clinical features and functional evaluations. 3. Results The main findings are presented in Table Study design Fifteen studies met the inclusion criteria. Since a relatively small number of studies using FES in children with CP were found, we decided to also include case reports evaluating the clinical aspects of FES. Only one RCT, 5 five single subject designs, 6e10 one cross-over design, 11 two none-randomized control trials, 12,13 three case series, 4,14,15 two case studies 16,17 and one exploratory design 18 were found. Six studies included a control group, 5,10e13,18 with five matched control studies 5,10e13,18 and three healthy control groups. 11e Patient selection FES was used in a total of 151 children diagnosed with CP. Studies reporting on the subject's functional level were scored I or II on the Gross Motor Function Classification System (GMFCS), implying that FES is mostly used in children with a high level of functional ability. The ages ranged from 5 to 18 years, most were >8 years old. Six studies reported using FES on children diagnosed with CP hemiplegia and diplegia, 7,8,10,11,15,18 five studies on children with CP hemiplegia 4-4e6,9,17 and four studies on children with CP diplegia. 12,13,16, Muscle groups stimulation Overall, the search revealed the use of FES on 4 muscle groups in children with CP. The most common stimulation was using FES on dorsi flexors (FesDF). 4e9,14,15,17,18 FES on knee extensors (FesKE) was used infrequently (3 children in two studies). 16,18 One group used FES to stimulate hip abductors (Fes- HABD), 12,13 reporting on 38 children with CP diplegia. Plantar flexors stimulation (FesPF) was used on 15 children diagnosed with CP diplegia and hemiplegia. 10, Protocol We found inconsistencies in the literature amongst studies with no accepted methodology or preferred protocol. While screening the protocol design for daily usage of dorsi flexors, we found that the period of stimulation time varied from 20 min up to 6.2 h daily. Two out of the ten studies used a

4 10 european journal of paediatric neurology 22 (2018) 7e16 protocol of stimulation up to 1 h daily 2,3 and six studies used FES for <5.6 h a day. 4,5,9,14,15,18 The period of use was either for a short adjustment before the trial 8,17 ; 8 weeks 5e7,18 ; 12 weeks, 9,15 4 months 14 and within a period of one year. 4 Knee extensor stimulation was used in one case-control study, 25e30 min daily for a period of up to 6 months, 16 however van der Linden chose to stimulate knee extensors all day for 8 weeks in two subjects. 18 Hip abductor stimulation was used for a short adjustment time prior to assessment and after 15 min, three times daily for a week. 12,13 Plantar flexors stimulation was assessed immediately after adjustment 11 or up to 15 min a day, 3 days a week for 4 weeks Common inclusion criteria for studies using FES FesDF: presence of insufficient dorsiflexion during the swing phase or the absence of heel initial contact 9,15,18 and an adequate passive dorsiflexion ROM (0e10 dorsiflexion). 4e6,9,14,17 Only one study evaluated children with a( 10 ) dorsiflexion. 18 FesKE: crouch gait pattern, passive knee extension >10, weak extensor moment. 16,18 FesHABD: hip adductors spasticity of 2 and crossed knee during walking. 12,13 FesPF: equines gait pattern, gastroc-soleus spasticity of at least grade 1 on the Modified Ashworth Scale. 10, Main outcomes Gait parameters including spatiotemporal and kinematics were reported as outcomes in all studies. Some authors described clinical parameter outcomes, i.e., ROM and spasticity 5,6,12,13 while others reported functional outcomes such as the ability to ascend and descend stairs and frequency of falls. 5,6,16 However, all studies evaluated the kinematic change that was targeted as a result of the stimulation. Main outcome measurements for dorsiflexion stimulation were: maximum dorsiflexion during the swing phase, dorsiflexion at initial contact and a foot contact pattern 4,5,7e9,14,15,17,18 thus, revealing that FesDF achieved improvement in maximum dorsiflexion during the swing phase, hence, reducing the frequency of falls. 4,5,7e9,14,15,17,18 Seven studies found an improved foot contact pattern. 5,7e9,14,17,18 Two studies described a reduction in selfreported toe drags and falls while walking with FesDF for 8 weeks. 5,6 The retention effect of the stimulation on the tibialis anterior cross-sectional area 15 was evaluated in one study revealing an increase in muscle thickness, however, this was not correlated with dorsiflexion improvement during the swing phase after 6 months of FesPF usage. An additional retention effect for dorsiflexor stimulation was reduced plantar flexors spasticity found immediately after 8 weeks of FesDF 5,6 and after 6 weeks of discontinuing FesDF. 6 Dorsiflexion during the swing phase and foot contact pattern was maintained in three studies after the immediate discontinuation of FesDF and up to 12 weeks. 7,9,15 However, one study combined a botulinum injection intervention to the gastrocsoleus muscles. 7 For knee extensor stimulation, 3 children improved their knee extension during the stance phase 16,18 and 1 child achieved a step through pattern using stairs immediately after starting the stimulation. 16 In one case, an immediate carryover effect of knee extension and step through stairs strategy was achieved after prolonged use of 6 months. 16 As for hip abductors stimulation, one study showed improvement in knee position immediately upon walking with the stimulation and a reduction in hip adductors spasticity after a one-time trial. However, prolonged use of FesHABD utilizing a one week home management protocol, had a significant effect on reducing hip adductors spasticity. 12,13 Plantar flexors stimulation increased in power generation while using FesPF. 11 Another study, using the, d.bgs>gross Motor Function Measure (GMFM) found an improvement in ankle kinetics and functional ability immediately after discontinuation of FesPF for four weeks and post-two weeks follow-up Discussion In the current work, different types of FES devices were reviewed comparing their specific indications, long-term therapeutic effects and potential advantages of stimulation Types of surface FES presently applied FES was conducted on 151 children with CP. The most investigated and used FES device was the common fibular nerve stimulation for improving dorsi flexors activity. We examined data of 95 patients with CP using FesDF, 4e9,14,15,17,18 in 10 studies, two studies investigating FesHABD in 38 patients 12,13 and two others investigating the gastroc-soleus complex facilitation in 15 patients. 10,11 Quadriceps stimulation was investigated in a case study 16 and in another small study. 18 FES, mainly used to control dorsi flexors muscles and prevent drop foot during the swing phase, is widely used due to its feasibility, simplicity and synchronization with gait Does surface FES improve gait deviations (an orthotic effect)? All studies evaluating FesDF found a positive orthotic effect. Dorsi flexors stimulation improved dorsiflexion during the swing phase and enhanced the foot contact pattern. 4,5,8,9,14,15,17,18 However, a prerequisite for the success of FesDF is an adequate dorsiflexion ROM of at least 0, 4e6,9,14,17 therefore, if the treatment goal is to improve dorsiflexion during the swing phase of the gait cycle and there is no significant dorsiflexion ROM limitation, there is sufficient evidence for choosing FesDF as an orthotic device. Regarding knee extensors stimulation, there is very limited evidence supporting the use of FesKE, with only 3 reported cases, although some positive effect was found with an increase in knee extension during the stance phase. 16,18 If knee extensors stimulation is to be considered, knee flexion contracture should be as minimal as possible, in order to achieve knee extension. According to the studies presented here, further research is warranted to evaluate the orthotic effect of FesKE.

5 european journal of paediatric neurology 22 (2018) 7e16 11 Table 1 e Characteristics of the included studies. Authors Design Participants, mean age (Years), CTRL group (participants and age), classification & GMFCS FES Parameters Protocol Inclusion criteria Dorsi flexors stimulation 4 RCT N ¼ 16 Age: years CTRL: 16 Classification: hemiplegia GMFCS I, II PW: 25e100 ms Frequency: 33 Hz 2 Multiple single subject design N ¼ 12 PW: 300 ms Age: 5e16 years Frequency: 33 Hz Classification: hemiplegia GMFCS I, II 5 Case series N ¼ 14 Age: 13.1 years Classification: 13 hemiplegia, diplegia 1 GMFCS I, II 6 Case series N ¼ 19 Age: years Classification: ND GMFCS I, II 3 Single subject design N ¼ 8 Age: 7.9 years Classification: 2 diplegia, 6 hemiplegia PW: 25 or 50 ms Frequency: 25 Hz On ramp: 0.2e0.5 s PW: 25e300 ms Frequency: 16.7e33 Hz PW: 300ms Frequency: 30 Hz Max intensity: 20e40 ma 7 Case study N ¼ 1 Age: 11 years PW: 300ms Frequency: 30 Hz Ramp: 0 ms Classification: hemiplegia Max intensity: 20 ma GFMCS I 8 Multiple single subject design N ¼ 10 PW: 3e350 ms Age: 9.5 years Frequency: 40 Hz Ramp: 0e4 ms Classification: hemiplegic Intensity: 15e100 ma 9 Case series N ¼ 5 Age: 16.5 years Classification: 5 hemiplegic patients 4 CP, 1 Diffuse pontine glioma GMFCS I, II 18 Exploratory design N ¼ 5 Age: 8 years CTRL n ¼ 5, aged 9, classification matched Classification: 1 diplegia, 3 hemiplegia and 1 monoplegia PW: 300ms Frequency: 33 Hz PW: 3e350 ms Frequency: 40 Hz Intensity: 20e70 ma 6.2 H/D 6 D/W 8W 6 W follow up 1 H/D 6 D/W 8W 6 W follow up 6 H/D 3 months followed by 4.2 H/D 3 months 5.6 H/D 4 months 20e30 Min/D 4 W with FES 4 W no FES 4 W with FES 4 W follow up 30 min prior to data collection 12 W during daily walking A- baseline A-12 W follow up B-12 W with FES A-12 W follow up 1 year during daily walking Passive dorsiflexion >5 Full passive knee extension bilaterally Dynamic popliteal angle <45 Passive dorsiflexion >5 Full passive knee extension bilaterally Dynamic popliteal angle <45 Foot drop (determined by clinical observation and gait analysis) >0 dorsiflexion with knee extended No Dorsiflexion >0 Dorsiflexion MMT >4- Plantar flexors MAS ¼ 1 Toe gait pattern No ankle plantar flexion contracture Hemiplegic type gait 1 or 2 Passive dorsiflexion >0 Dorsiflexion >10 8 W during daily walking Decreased dorsiflexion during swing phase Dorsiflexion with knee extended >( 10 ) (continued on next page)

6 12 european journal of paediatric neurology 22 (2018) 7e16 Table 1 e (continued) Authors Design Participants, mean age (Years), CTRL group (participants and age), classification & GMFCS FES Parameters Protocol Inclusion criteria 10 Single subject design N ¼ 5 Age: 12.6 years Classification: 4 diplegia, 1 hemiplegia Knee extensors stimulation 11 Case study N ¼ 1 Age: 18 years Classification: diplegia GMFCS II 18 Exploratory design N ¼ 2 Age: 8 years CTRL 2 n, age 10.5, matched classification Classification: 2 diplegia Hip abductors stimulation 12 Non randomized control trial 13 Non randomized control trial Plantar flexors stimulation 14 Cross-over study design 15 Single subject control design N ¼ 17 Age: 11 years CTRL 15 diplegic þ17 healthy aged 10 children Classification: diplegia N ¼ 21 Age: 7.4 years CTRL 20 healthy aged 7.7 Years þ 10 diplegic aged 8.3 years Classification: diplegia N ¼ 9 Age: 7.55 years CTRL 6 healthy, aged 8.16 Classification: 5 diplegia, 4 hemiplegia GMFCS I N ¼ 6 Age: 6.3 years CTRL 6, aged 6.5 years, 5 diplegic, 1 hemiplegic Classification: 2 diplegia and 4 hemiplegia PW: 300 ms Frequency: 33 Hz or 50 Hz Ramp time: 0.1 e 0.2 s. PW 300 ms Frequency: 40 Hz Intensity: 40 ma PW 3e350 ms Frequency: 40 Hz Intensity: 20e70 Ma PW: 50 ms Frequency: 20 Hz Intensity until isometric contraction was seen PW 50 ms Frequency: 20 Hz PW: 300 ms Frequency: 32 Hz Intensity: 10e40 ma Ramp time: 0.2s Frequency: 30e35 Hz 15 min FES and treadmill or 15 min treadmill only (CTRL) 10 consecutive walks during each phase A- no FES B- with FES A- no FES B- with FES 2 months 25 Min/D 7 D/W Followed by: 4 months 30 Min/D 7 D/W 8 W during daily walking 15 min/d 3 Times a day 1W 15 min/d 3 day 1W 15 trials with FES 15 trial without FES 15 Min/D 3 D/W 4W 2 W- baseline 4 W-with FES 2 W- follow up No Crouch gait Passive knee extension >10 Weak extensor moment Increased knee flexion at initial contact and during stance phase SLR >40 Hip flexion contracture <( 15 ) Crouched gait with crossed or touched knees during walking as a result of spastic hip adductors Hip adductor spasticity Gastrocsoleus MAS <3 No heel strike at initial foot contact Gastrocsoleus MAS >1 Equinus gait pattern Dorsiflexion >10 FESefunctional electrical stimulation; GMFCSeGross Motor Function Classification; CTRLecontrol group; GMFMeGross Motor Function Measure; NDenot described; RCTeRandomized controlled trial; Min/Deminutes per day; D/Wedays per week; Weweeks; PWePulse Width; MASeModified Ashworth Scale. Hip abductors stimulation showed an improvement in knee alignment among 17 children during the stance phase while walking with the device. 13 These are preliminary data, however, this encourages the use of FesHABD for improving hip and knee frontal plane alignment and decreasing hip spasticity. The effect of plantar flexion stimulation on gait deviations was not evaluated. One study evaluated the kinetic and spatio-temporal effect of FesPF on gait. 11 Their goal was to achieve the appropriate timed force production of the gastrocsoleus during gait, which is abnormal due to muscle weakness. The authors found an increase in power generation at

7 european journal of paediatric neurology 22 (2018) 7e16 13 Table 2 e Outcome measurements: orthotic and therapeutic results. Study Outcome measures Effects with FES Retention effects (without FES) Dorsi flexors stimulation 4 Ankle kinematics Passive and dynamic DF ROM Spatio-temporal parameters Gastroc-soleus spasticity Community mobility balance skills Self-reported toe drag and falls [ DF at initial contact [ Max dorsiflexion during swing phase Normalized time during stance phase Normalized step length Y Self-reported toe drag Y Self-reported falls Immediate effect: Normalized time during stance phase [ Community mobility balance scores [ Dynamic DF ROM NC passive DF NC popliteal angle Y Gastroc-soleus spasticity NC ankle kinematics 2 Ankle ROM Selective motor control Dorsiflexion and plantar flexion strength Gastroc-soleus spasticity Single-limb balance Observational Gait Scale (OGS) score Self-reported toe drag and falls YSelf-reported toe drag YSelf-reported falls Follow up: [ Community mobility balance scores NC passive DF NC dynamic DF ROM NC gastroc-soleus spasticity Immediate effect: [ Dorsiflexion ROM Selective motor control [ Dorsiflexion strength YGastroc-soleus spasticity YSelf-reported toe drag YSelf-reported falls NC in OGS 5 Ankle kinematics Temporal Parameters Tibialis anterior cross- sectional area and thickness [ Max dorsiflexion during swing phase NC gait velocity Follow up: [ Dorsiflexion ROM Selective motor control [ Dorsiflexion strength YGastroc-soleus spasticity Selective motor control NC in OGS Immediate effect post 3 months: NC Max dorsiflexion during swing phase [Tibialis anterior cross-sectional area and muscle thickness NC gait velocity 6 Ankle kinematics Spatio-temporal parameters 3 Ankle kinematics Foot contact pattern [ Max dorsiflexion during swing phase [ Dorsiflexion at initial contact Partial preservation of plantar flexion during toe off NC spatio-temporal No Immediate effect post 6 months: YMax dorsiflexion during swing phase [Tibialis anterior cross- sectional area and muscle thickness NC gait velocity Immediate Effect: Foot contact pattern 7 Ankle kinematics Spatio-temporal parameters Ankle power [ Dorsiflexion at initial contact [ Max dorsiflexion during swing phase [ Mean dorsiflexion during swing phase Y Velocity and cadence NC in step length Ankle absorption work Y Ankle generation work Follow up: [ Dorsiflexion at the end of the swing phase NC foot contact pattern (continued on next page)

8 14 european journal of paediatric neurology 22 (2018) 7e16 Table 2 e (continued) Study Outcome measures Effects with FES Retention effects (without FES) 8 Spatio-temporal parameters Foot contact pattern Acceptability of the intervention from a user perspective 9 Gait Profile Score (GPS) Gait Deviation Index (GDI) Gillette Gait Index (GGI) Temporal parameters 18 Knee and ankle kinematics Spatio-temporal parameters Foot contact pattern Gillette Gait Index 10 Statistical outcomes: Lower limb kinematic data Spatio-temporal parameters Foot contact pattern Clinical outcomes: Dorsiflexion at initial contact Minimum dorsiflexion angle during the swing phase. Knee extensors stimulation 11 Spatio-temporal Lower limb kinematics Stairs strategy 18 Knee and ankle kinematics Foot contact pattern Walking speed Gillette Gait Index Hip Abductors Stimulation 12 Spatio-temporal parameters Hip adductors MAS Knee position 13 Spatio-temporal parameters Hip adductor MAS Reduction in double stance time Mean swing and stance times were close to symmetry Heel-toe contact pattern and symmetry GPS and GDI GGI (but no statistical significance) High satisfaction [ Max dorsiflexion during swing phase [ Foot-floor angle Y Walking speed NC in knee kinematics GGI Immediate results per patients: [ Dorsiflexion in mid swing (F) Improved mode of IC (F) [ Dorsiflexion during swing phase (H) Initial contact pattern (H) [ Dorsiflexion during stance phase (H) Stance phase kinematics (B, F, H) [ Heel contact duration (B,F) Dorsiflexion during stance phase (B,F) NC in walking speed and step length (with the exception of child E). [ Walking speed due to increased step length and cadence [ Knee extension during stance phase Ascend and descend stairs step through [ Knee extension during stance phase NC in foot floor angle NC in knee extension during initial contact GGI Immediate: Step length and walking speed Knee position Post 1 W: Knee position Immediate: Walking velocity Step length Stride length NC in step width Follow up Mean swing and stance times were close to symmetry Foot contact pattern, though smaller improvement compared to the FES usage Immediate effect: No significant change in GGI No significant change in spatio temporal parameters No significant change in knee ankle kinematics Immediate: Ascend and descend stairs step through [ Knee extension during stance phase Immediate: No significant change in GGI No significant change in spatio temporal parameters No significant change in knee ankle kinematics Immediate (one time trial): Y Adductors MAS post 1 W: Y Adductors MAS Immediate: NC in adductors MAS Post 1 W: Y Adductors MAS Plantar Flexors Stimulation 14 Spatio-temporal parameters Power generation 15 Ankle kinetics GMFM Post 1 W Walking velocity Step length Stride length [Power generation NC in stride length NC in stride frequency Immediate and follow up effect: GMFM Trend of immediate improvement in ankle moment and ankle power FES e functional electrical stimulation; GMFCS e Gross Motor Function Classification; CTRL e control group; GMFM e Gross Motor Function Measure; DF e Dorsiflexion; ND e not described; W e weeks; MAS e Modified Ashworth Scale; ROM e range of motion; NC e no change; IC e initial contact; e improvement.

9 european journal of paediatric neurology 22 (2018) 7e16 15 push off, with no change occurring in stride length or stride frequency, thus, the basis of FesPK in altering gait deviations is scarce, however, there is a potential for improving gastrocsoleus power generation at push off Were there any functional advantages to using FES? No standardized tool was used to estimate functional improvement in the included studies. Temporal parameters as well as gait indices were used to calculate the extent by which a subject's gait deviated from the average typical gait. Functional improvement occurred in some studies when walking with FesDF and FesHABD. 4e6,9,12,13 However, spatiotemporal parameters did not show a consistent improvement during or after the use of FesDF. Patients using FesDF showed a significant reduction in self-reported toe drag and falls in the community. 5,6 In summary, there is some evidence that FES can improve spatiotemporal parameters and prevent falls Are there criterions for fitting FES? No absolute or uniform criteria were found for using any type of the FES devices. Our analysis indicates that the typical patient using the FES device, is an individual with spastic hemiplegic CP or diplegia scoring GMFCS levels of I or II implying that the group with relatively higher functional abilities as opposed to GMFCS III or IV levels, might have more sufficient cognitive and communication skills in order to comply with the FES apparatus and its electrical stimulation. It is obvious that the FES type was adjusted according to the targeted gait deviation. FesDF was adapted for children presenting with excessive plantar flexion during the swing phase or an absence of heel initial contact and a prerequisite for its use is the ability to maintain adequate passive dorsiflexion ROM. FesKE was adapted for children presenting with excessive knee flexion during the stance phase and FesHABD for excessive hip adduction Does the use of an FES device produce a carry-over/ retention effect or only an orthotic effect? There is scarce evidence in the current literature as to the above question: the longest follow-up study was 1 year, which might be too premature to determine any therapeutic effects. A few studies have reported a short term follow-up post FesDF and found an increased cross-sectional area muscle thickness of the tibialis anterior 15 and reduced gastrocnemius spasticity. 5,6 The retention effect of dorsiflexion during the swing phase and initial contact pattern were shown in three studies, 7,9,15 after the discontinuation of FES for 3e4 months. However, these groups used FES for a relatively long period of time, approximately 2e6 months. In addition, in FesHABD, a retention effect was achieved only a week after stimulation in decreasing hip adductors spasticity amongst 38 children with CP diplegia. 12,13 As for knee extensors stimulation, one case report found an immediate retention effect of increased knee extension during the stance phase and the ability to climb up and down stairs with a step through pattern. 16 FesPF, in one study of 6 children diagnosed with CP diplegia, improved their functional level, GMFM and ankle kinetics after two weeks of discontinuing a four-week FES program. Therefore, it seems that the retention effect is temporary and dependent on the continuous use of FES. This effect might fade away as the stimulation stops. There is no evidence as yet of a prolonged therapeutic effect of FES or its ability to achieve independency from orthotics in children with CP. 5. Conclusions We performed a systematic review examining the effects of several FES modalities. Due to a limited number of studies, we were unable to employ a formal meta-analysis approach to examine the differences in outcomes of treatment with different treatment modalities. Our literature review was restricted since different studies presented with various sampling designs and altered study protocols. The quality of most current studies was poor, with only one RCT found in addition to a few comparative studies. Most included a small number of children with CP, therefore, controlled investigations with larger numbers of participants are needed to fully determine the efficacy of FES and establish how to achieve a longer-lasting benefit using this relatively new modality in this population. Studies should also aim at identifying which CP patients are most likely to benefit from neuro-prosthesis and identify factors that may impede or improve the likelihood of a response, such as an adequate ROM or spasticity control. Another important justification for future studies is a longer follow up investigating patient compliance. On the basis of this review, we encourage the clinician to recommend and evaluate the use of FES on a case to case basis. Funding source This research did not receive any specific grant from funding agencies in the public, commercial or not-for-profit sectors. Conflicts of interest None. Acknowledgements The authors thank Mrs. Phyllis Curchack Kornspan for her editorial services. references 1. Rodda JM, Graham HK, Carson L, Galea MP, Wolfe R. Sagittal gait patterns in spastic diplegia. J Bone Jt Surg Br 2004;86:251e8. 2. Arnold AS, Blemker SS, Delp SL. Evaluation of a deformable musculoskeletal model for estimating muscle-tendon lengths during crouch gait. Ann Biomed Eng 2001;29:263e74.

10 16 european journal of paediatric neurology 22 (2018) 7e16 3. Laufer Y, Ring H, Sprecher E, Hausdorff JM. Gait in individuals with chronic hemiparesis: one-year follow-up of the effects of a neuroprosthesis that ameliorates foot drop. J Neurol Phys Ther 2009;33:104e Danino B, Khamis S, Hemo Y, et al. The efficacy of neuroprosthesis in young hemiplegic patients, measured by three different gait indices: early results. J Child Orthop 2013;7:537e Pool D, Valentine J, Bear N, et al. The orthotic and therapeutic effects following daily community applied functional electrical stimulation in children with unilateral spastic cerebral palsy: a randomised controlled trial. BMC Pediatr 2015;15: Pool D, Blackmore AM, Bear N, Valentine J. Effects of shortterm daily community walk aide use on children with unilateral spastic cerebral palsy. Pediatr Phys Ther 2014;26:308e Galen S, Wiggins L, McWilliam R, Granat M. A combination of Botulinum Toxin A therapy and functional electrical stimulation in children with cerebral palsy-a pilot study. Technol Health Care 2012;20:1e9. 8. Postans NJ, Granat MH. Effect of functional electrical stimulation, applied during walking, on gait in spastic cerebral palsy. Dev Med Child Neurol 2005;47:46e Durham S, Eve L, Stevens C, Ewins D. Effect of functional electrical stimulation on asymmetries in gait of children with hemiplegic cerebral palsy. Physiotherapy 2004;90:82e Chan NN, Smith AW, Lo SK. Efficacy of neuromuscular electrical stimulation in improving ankle kinetics during walking in children with cerebral palsy. Hong Kong Physiother J 2004;22:50e Ho CL, Holt KG, Saltzman E, Wagenaar RC. Functional electrical stimulation changes dynamic resources in children with spastic cerebral palsy. Phys Ther 2006;86:987e Al-Abdulwahab SS, Al-Khatrawi WM. Neuromuscular electrical stimulation of the gluteus medius improves the gait of children with cerebral palsy. NeuroRehabilitation 2009;24:209e Al-Abdulwahab SS. Electrical stimulation improves gait in children with spastic diplegic cerebral palsy. NeuroRehabilitation 2011;29:37e Prosser LA, Curatalo LA, Alter KE, Damiano DL. Acceptability and potential effectiveness of a foot drop stimulator in children and adolescents with cerebral palsy. Dev Med Child Neurol 2012;54:1044e Damiano DL, Prosser LA, Curatalo LA, Alter KE. Muscle plasticity and ankle control after repetitive use of a functional electrical stimulation device for foot drop in cerebral palsy. Neurorehabil Neural Repair 2013;27:200e Khamis S, Martikaro R, Wientroub S, Hemo Y, Hayek S. A functional electrical stimulation system improves knee control in crouch gait. J Child Orthop 2015;9:137e Pierce SR, Orlin MN, Lauer RT, Johnston TE, Smith BT, McCarthy JJ. Comparison of percutaneous and surface functional electrical stimulation during gait in a child with hemiplegic cerebral palsy. Am J Phys Med Rehabil 2004;83:798e van der Linden ML, Hazlewood ME, Hillman SJ, Robb JE. Functional electrical stimulation to the dorsiflexors and quadriceps in children with cerebral palsy. Pediatr Phys Ther 2008;20:23e9.