Medical Engineering & Physics 33 (2011) 56–61
Contents lists available at ScienceDirect
Medical Engineering & Physics journal homepage: www.elsevier.com/locate/medengphy
Patient tolerance of neuromuscular electrical stimulation (NMES) in the presence of orthopaedic implants Barry J. Broderick a,b,∗ , Cian Kennedy c , Paul P. Breen a,b , Stephen R. Kearns c , Gearóid ÓLaighin a,b a b c
School of Engineering & Informatics, Electrical & Electronic Engineering, NUI Galway, University Road, Galway, Ireland Bioelectronics, National Centre for Biomedical Engineering Science, NUI Galway, University Road, Galway, Ireland Department of Orthopaedics, Galway University Hospitals, Newcastle Road, Galway, Ireland
a r t i c l e
i n f o
Article history: Received 23 April 2010 Received in revised form 18 August 2010 Accepted 7 September 2010 Keywords: Neuromuscular electrical stimulation NMES Deep vein thrombosis Orthopaedic surgery
a b s t r a c t Neuromuscular electrical stimulation (NMES) may help reduce the incidence of deep vein thrombosis (DVT) in the postoperative total hip and knee arthroplasty patient. However, discomfort associated with stimulus may reduce patient acceptance of NMES as therapy. The aim of this study was to determine if patient comfort and tolerance of NMES was affected by applying stimulation in proximity to an orthopaedic implant. There was a concern that this may cause a concentration of current around the metal which could result in hypersensitivity of NMES and reduce its effectiveness. Twenty patients took part in this study, 10 total hip and 10 total knee arthroplasty patients. Each patient was at least 3 weeks post surgery. NMES was applied to the calf muscles of each leg using skin surface electrodes. Four excitatory levels were recorded, which were: sensory threshold, motor threshold, pain threshold and pain tolerance. Following this, patients underwent a 5 min stimulation session and indicated their overall comfort level on a visual analogue scale. Measurements of peak venous velocity, mean velocity and volume flow were recorded by duplex scanning from the popliteal vein at rest and in response to NMES elicited contractions during this session. Finally, patients completed a short verbal interview detailing their experience with the NMES treatment. The blood flow results showed increases in peak venous velocities, mean velocities and volume flow produced by NMES of 200%, 60% and 60% respectively when compared to resting blood flow. Comfort assessment indicated that the presence of a metallic implant did not give rise to hypersensitivity due to NMES. Patients found the application of calf muscle NMES comfortable and acceptable as a treatment. We conclude that the use of NMES on postoperative orthopaedic patients can be safely administered as a DVT prevention method. © 2010 IPEM. Published by Elsevier Ltd. All rights reserved.
1. Introduction Neuromuscular electrical stimulation (NMES) is a potential mechanical deep vein thrombosis (DVT) preventative method that is often overlooked. Despite early reports indicating the effectiveness of NMES at preventing DVT in general surgical patients, the technique has not gained the same widespread acceptance as other mechanical techniques such as intermittent pneumatic compression or graduated compression stockings [1–4]. Early implementations of NMES devices were so uncomfortable that they could only be used when the patient was under anaesthetic [1]. Furthermore, poor quality electrodes combined with monophasic stimulation waveforms often resulted in skin irritation and burns.
∗ Corresponding author at: School of Engineering & Informatics, Electrical & Electronic Engineering, NUI Galway, University Road, Galway, Ireland. Tel.: +353 91 493126. E-mail address:
[email protected] (B.J. Broderick).
In fact, Pambianco et al. [5] had to discontinue the NMES portion of a study investigating the effects of heparin, IPC and NMES on DVT rates in stroke rehabilitation patients due to discomfort and blister formation. More recently, the inclusion of microcontrollers in the design of NMES devices has allowed the implementation of precisely controlled stimulation waveforms and novel algorithms [6]. These waveforms and algorithms have significantly increased the comfort and tolerance of NMES in users. Moreover, the use of biphasic pulses, either symmetric or asymmetric, has minimized ion redistribution and the subsequence risk of skin irritation and burns. Subsequently, many researchers are revisiting the idea of using NMES as a DVT prevention device. Patients undergoing major orthopaedic surgery including total hip arthroplasty (THA) and total knee arthroplasty (TKA) are of particularly high risk of DVT [7]. The efficacy of NMES on DVT development rates during the recovery period post-THA/TKA has not been evaluated as of yet. Before such a therapy can be evaluated postoperatively, patient comfort and tolerance of NMES must
1350-4533/$ – see front matter © 2010 IPEM. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.medengphy.2010.09.003
B.J. Broderick et al. / Medical Engineering & Physics 33 (2011) 56–61
57
Fig. 1. Patients lay on a physiotherapy plinth for the duration of the study in a trunk elevated position. NMES was applied to the calf muscles using the Duo-STIM NMES system.
be investigated. There was a concern that applying NMES in close proximity to a metal implant may cause a concentration of current around the metal. According to Baker et al. [8], “very small currents can be measured virtually anywhere on or in the body during any form of surface electrical stimulation.” This could result in a reduction of NMES effectiveness and cause “hypersensitivity” or increased discomfort associated with the applied stimulus. Overall, this would limit the postoperative use of NMES to patient groups whose implants are far enough away from the stimulus site so as not to cause problems. Thus, the primary outcome of this study was to assess patient tolerance of NMES in the presence of metallic implants. The secondary outcome was to measure venous outflow from the legs associated with a tolerable NMES intensity in these patients.
2.2. Equipment setup
2. Methods
Prior to conducting the experiment, patients were asked to lie comfortably on a plinth in a trunk elevated position (Fig. 1). Comfort was assessed using two metrics. The first involved gradually increasing the stimulation intensity and recording 4 excitatory levels: initial perception of the stimulus sensation (sensory threshold), when a muscle contraction is observed (motor threshold), the onset of discomfort (pain threshold) and the intensity at which the discomfort becomes unbearable (pain tolerance). All four excitatory levels were recorded for each leg. The first leg was chosen randomly. The second comfort metric was a visual analogue scale (VAS), where patients were asked to mark the level of discomfort on a 100 mm non-hatched scale ranging from no pain whatsoever to severe pain [11]. It was pre-determined that VAS pain scores of 30 mm or less would be categorised as mild pain, between 31 and 69 mm as moderate pain and scores of 70 mm or greater as severe pain [12]. The minimum clinically significant difference (MCSD) is the smallest change in two VAS scores for them to be considered different from each other. The MCSD was calculated as an increase in scores between test stages of 12 mm [13]. The first VAS was taken before the application of NMES. Once the excitatory levels were recorded the patients underwent a 5 min stimulation protocol. The second VAS was taken during this protocol. During this period the stimulation intensity was set below the pain threshold. A 20 s rest period followed each contraction.
2.1. Recruitment For convenience, patients who had undergone orthopaedic surgery in Galway University Hospitals and were scheduled for a follow-up clinical visit were recruited. As NMES may cause hypersensitivity at the wound site itself, perhaps through limb movement originating from ankle displacement caused by the calf contraction, patients who were at least 3 weeks post surgery were recruited. The concern was that had we directly approached patients during the post-operative period and they expressed discomfort or low tolerance of NMES, we would not be able to identify the origin of the discomfort reported. Furthermore, the use of pain management in the early postoperative period may mask these effects. Therefore, the inclusion criteria were THA or TKA and at least 3 weeks post surgery. The exclusion criteria included a history of heart problems, bilateral orthopaedic implants and a current involvement in another study. A total of 20 patients took part in the trial, 10 in the THA group and 10 in the TKA group. No patient had more than one orthopaedic implant and no patient was on any major analgesia or other pain relieving therapy. Each patient gave their written informed consent to take part and the study was approved by the Clinical Research Ethics Committee.
Four 5 cm × 5 cm PALS UltraStim self-adhesive, hypo-allergenic electrodes were used to facilitate NMES delivery (Axelgaard Manufacturing Co. Ltd., CA, USA). Electrodes were placed over the calf muscles of both legs. NMES was applied to both the operated limb and the contralateral limb, the latter acting as the control limb. The Duo-STIM system was used to apply the NMES [9]. The Duo-STIM was programmed with a pulse width of 350 s, a frequency of 36 Hz 500 ms ramp-up time, 1 s ON time, 500 ms ramp-down time. These stimulation parameters were chosen to maximize expelled blood flow while maintaining patient comfort [8,10].
2.3. Comfort evaluation procedure
58
B.J. Broderick et al. / Medical Engineering & Physics 33 (2011) 56–61
a
Control Limb Operated Limb
60
Stimulation Intensity (V)
b
70
70 60
50
50
40
40
30
30
20
20
10
10
0
0 Sensory
Motor
Pain
PTO
Sensory
Motor
Pain
PTO
Fig. 2. The mean stimulation intensity in volts corresponding to sensory threshold, motor threshold, pain threshold and pain tolerance (PTO) for the THA group (a) and the TKA group (b). Error bars indicate standard deviation. No differences were observed between the operated limb and control limb in either group. The voltage corresponding to each consecutive threshold was significantly different than the previous (p > 0.001).
For the remainder of the study, stimulation intensity was set just below the pain threshold. At the end of study, the patients were asked to complete a short verbal questionnaire. They were asked: • To give a verbal categorical rating of the NMES treatment, as ‘very comfortable’, ‘comfortable’, ‘bearable’ or ‘unbearable’, • If they noticed a difference in the sensation due to stimulation between the operated limb and the contralateral limb, • If they would consider NMES as an acceptable treatment for DVT prevention post orthopaedic surgery. 2.4. Doppler ultrasound scanning Ultrasound sound measurements were taken using a 4–8 MHz linear transducer (LOGIQ e, GE Medical Systems) during the 5 min stimulation protocol. Doppler ultrasound superimposed over realtime B-mode was used to measure the blood flow resulting from the NMES elicited calf muscle contractions of both legs. Again, the first leg was chosen at random. Blood flow was measured from the
popliteal vein, imaged from the popliteal fossa, below the saphenopopliteal junction. Peak venous velocity, mean velocity flow and vein cross-sectional area were recorded. Mean velocity flow was determined by tracing the average of the flow velocity signal over a 4 s window. Venous volume was calculated by multiplying the mean velocity flow by the cross-sectional area of the popliteal vein. Resting and stimulated blood flow were recorded. Three measurements were taken per parameter and the average used for analysis. 2.5. Statistical analysis A repeated measures analysis of variance (ANOVA) was used to determine differences between the stimulus intensities for each of the excitatory levels in the operated and control limb. The Wilcoxon signed-rank test was used to determine the following intra-group differences in blood flow measurements: stimulated blood flow versus resting, operated limb versus control limb. The Mann–Whitney U-test was used to make inter-group blood flow comparisons. A p-value of <0.05 was considered statistically significant in all cases.
Table 1 VAS scores, pain category and the difference in scores before (initial) and during (final) NMES application. Patient no. Total hip replacement 1 2 3 4 5 6 7 8 9 10 Total knee replacement 11 12 13 14 15 16 17 18 19 20
VAS scores (mm) Initial
Pain category
VAS scores (mm) final
Pain category
Difference in VAS scores (mm)
0 7 2 4 7 0 14 6 2 26
Mild Mild Mild Mild Mild Mild Mild Mild Mild Mild
69 7 40 6 32 81 15 11 3 24
Moderate Mild Moderate Mild Moderate Severe Mild Mild Mild Mild
69 0 38 2 25 81 1 5 1 −2
1 3 2 3 5 3 17 5 48 4
Mild Mild Mild Mild Mild Mild Mild Mild Moderate Mild
8 17 34 51 33 52 28 5 49 52
Mild Mild Moderate Moderate Moderate Moderate Mild Mild Moderate Moderate
7 14 32 48 28 49 11 0 1 48
>MCSD >MCSD >MCSD >MCSD
>MCSD >MCSD >MCSD >MCSD >MCSD
>MCSD
B.J. Broderick et al. / Medical Engineering & Physics 33 (2011) 56–61
300
Stimulated contraction Resting
40
30
20
10
*
*
*
*
Venous volume flow (mil/min)
Peak Venous Velocity (cm/s)
60
50
59
250
Stimulated contraction Resting
200 150
* *
*
*
100 50 0 Control Operated
0 Control
Operated
Control
(a)
(a)
Operated
(b)
Fig. 3. Peak venous blood flow velocity measurements due to the NMES elicited calf muscle contraction versus resting in the THA group (a) and the TKA group (b). *p < 0.001 compared to resting peak velocities.
3. Results Fig. 2 shows the stimulus intensities for each of the excitatory levels in the operated limb of the THA group and TKA group compared to the respective control limb. The repeated measures ANOVA revealed no differences between the operated limb and the control limb for either patient group (p = 0.48). Each threshold was significantly different from the previous threshold in both the operated limb and control limb and for each patient group (p < 0.001). VAS scores compared comfort levels of patients before and during the application of NMES. Table 1 summarizes the scores of each patient in both groups. Initially, categorical scores indicated mild pain for 19 patients and moderate pain for 1. During the application of NMES, 6 patients in the THA group indicated no change in their comfort level, whereas 3 indicated moderate pain and 1 severe pain. All 4 elevations in comfort level were greater than the MCSD. Four patients in the TKA group indicated no change in their comfort level during NMES, whereas 6 indicated moderate discomfort. Again, all 6 elevations in comfort level were greater than the MCSD. Blood flow measurements were taken from all n = 10 patients in the THA group and n = 8 patients in the TKA group. Blood flow results could not be taken from 2 patients in the TKR group as the
Control Operated
(b)
Fig. 5. Volume flow measurements due to the elicited calf muscle contraction versus resting in the total hip replacement group (a) and the total knee replacement group (b). *p < 0.001 compared to volume.
patients found it too difficult to keep their leg in the desired position long enough for the operator to take the measurement. No differences were observed between the control limb and operated limb in terms of peak venous velocity, mean velocity flow and volume flow in both groups (p = 0.647, 0.983, 0.744 respectively). There were no differences in peak venous velocity between the THA and TKA groups (p = 0.44). Both groups experienced an almost 200% increase in peak venous velocity compared to resting, p < 0.001 (Fig. 3). There were no differences in mean velocity flow between the THA and TKA groups, p = 0.54 (Fig. 4). Stimulated mean velocity flow was almost 60% higher than resting in both groups (p < 0.001). Furthermore, there were no differences in venous volume flow between groups, p = 0.67 (Fig. 5). Stimulated volume flow of both groups was over 60% higher than resting (p < 0.001). Five patients described the use of NMES as ‘very comfortable’, 13 as ‘comfortable’ and 2 as ‘bearable’. Of the two patients that indicated a verbal rating of ‘bearable’, one patient was from the THA group and one from the TKA group. No patient reported a difference in sensation between the operated and contralateral limb during the application of NMES and all patients stated that they would find the use of NMES during post-operative recovery for DVT prevention acceptable. 4. Discussion
Mean flow velocity (cm/s)
8
Stimulated contraction Resting
6
*
*
4
*
*
2
0 Control Operated
(a)
Control Operated
(b)
Fig. 4. Time average mean velocity measurements due to the NMES elicited calf muscle contraction versus resting in the total hip replacement group (a) and the total knee replacement group (b). *p < 0.001 compared to resting mean velocities.
There are essentially two situations where calf muscle NMES could be used to prevent DVT associated with orthopaedic surgery: perioperatively and postoperatively. Perioperative NMES has been previously investigated on patients during general elective surgery while under anesthesia, with mostly positive results [1,3,4]. While there has been a demonstrated reduction in DVT rates in general surgery associated with the use of neuromuscular electrical stimulation (NMES); the clinical use of calf muscle NMES as DVT prevention methodology has yet to be established [3,14,15]. The present study focused on the comfort assessment of patients using NMES with orthopaedic implants in the context of DVT prevention. Patients who have undergone orthopaedic procedures such as total hip and knee arthroplasties remain at high risk of DVT in the days following surgery albeit several weeks following surgery [16]. The use of NMES during this period may reduce DVT development rates in this patient group. This study showed that both THA and TKA patients had similar sensory and motor thresholds in both the operated and contralateral limb. Moreover, both groups showed a similar pain threshold and pain tolerance level between
60
B.J. Broderick et al. / Medical Engineering & Physics 33 (2011) 56–61
the operated and contralateral limb. This suggests that the presence of a metallic implant did not affect the expected operation of the applied NMES whatsoever. This finding was reinforced by the patients verbally reporting no difference between the stimulus sensation of the operated and control limb. Previous studies have examined patient perception of NMES. Alon et al. [17] examined the effect of 4 different electrode sizes (2.25, 9, 20.25 and 40.3 cm2 ) on excitatory responses (sensory, motor, pain and pain tolerance) on healthy participants. The authors observed that NMES comfort increases with increasing electrode size. They also observed that increasing electrode sizes decrease the peak voltage corresponding to the excitatory levels. Interestingly, with the exception of maximal pain tolerance, the most comfortable excitation levels the authors recorded were very similar to those recorded in the present study. This suggests that patients who have a metallic implant have a similar response to calf NMES as healthy individuals. The lower maximal pain tolerance recorded in the present study may be due to an older study group who had recently undergone major surgery. Lyons et al. [18] investigated the influence of two electrode sizes (19.93 and 38.48 cm2 ) and position on perceived pain and discomfort during NMES of the gastrocnemius muscles on healthy participants. The authors assessed comfort by recording excitatory responses to increases in stimulus amplitude. Their results indicated that both electrode placement and electrode size were important considerations for NMES comfort. They concluded that the smaller of the two electrode sizes used was more comfortable for calf stimulation. The stimulation parameters used in the present study were similar to those used by Lyons et al. [18]. Clarke Moloney et al. [11] assessed the comfort associated with and without NMES use on patients with chronic venous disease using a VAS. The authors found that the comfort categorical rating remained unchanged in all but one patient. This indicated that patients found NMES to be an acceptable therapy which could be significant for future studies involving NMES treatments for venous leg ulcer healing. Broderick et al. [10] previously conducted a study involving healthy participants who underwent a 4-h stimulation protocol. The NMES parameters were the same as those used in this study and VAS was also used to measure comfort levels. Three participants indicated moderate discomfort and 7 only mild discomfort when NMES was started. By the end of the study, 2 participant’s categorical score increased to moderate while 2 other participant’s scores decreased to mild and the rest remained the same. NMES altered categorical scores of 4 THA patients and 6 TKA patients in the present study (9 to moderate, 1 to severe). The patients in the present study definitely found NMES less comfortable than the healthy participants in the previous study. This may also be associated with the difference in study groups, i.e. a young healthy cohort versus an older patient cohort. Moreover, participants in the previous study had a 4-h stimulation session which may have allowed them time to grow accustomed to the sensation of NMES. Kaplan et al. [19] asked each healthy participant who received either electrical stimulation of the calf or foot muscles to complete a questionnaire regarding the acceptance of the electrical stimulation. Both groups found NMES comfortable and strongly felt that they would use an NMES device if directed by their doctor. Discomfort has been identified as a limiting factor in the use of surface NMES. The activation of skin surface sensory receptors is essentially unavoidable but can be minimized through the use of advanced skin surface electrodes, correct electrode placement and stimulation waveforms known to be comfortable [20,21]. The VAS scores in this study were used to assess the overall comfort of orthopaedic patients with and without NMES. The VAS scores were taken after patients had indicated their pain tolerance levels, when the Duo-STIM intensity was set to the level each patient indicated as their pain threshold level. This intensity level resulted in a
very strong calf muscle contraction. Discomfort at this intensity level was indicative of rapid muscle fatigue bordering on muscle cramping rather than sensory discomfort and, therefore, is an unpractical stimulation level for a continuous NMES therapy. The results indicated that patients were able to tolerate this level of stimulus intensity; however, the majority of patients indicated an increase in their comfort scores. A more practical NMES DVT prevention therapy should ideally use a lower stimulation intensity, at least initially, until the patients calf muscles become more accustomed to the increased activity. The VAS of one patient in the THA group indicated a categorical score of ‘severe’ pain; however the same patient also verbally described the NMES therapy as being ‘very comfortable’. This could be attributed to a misunderstanding of how the VAS score worked. The blood flow results in this study showed increases in venous velocities and volume flow produced by the application of NMES: 200% peak venous velocity, 60% mean velocity, 60% volume flow increases when compared to resting blood flow. This shows that the application of NMES on post hip and knee surgery patients is capable of eliciting large increases in lower limb venous blood flow over resting. This could significantly reduce venous stasis in this patient group and therefore, reduce the risk of DVT development. The lack of a significant difference in blood flow between the operated limb and non-operated limb by NMES suggested the blood flow in the operated limb was not affected by the presence of the implant in either group. 5. Conclusion The presence of a metallic implant did not give rise to hypersensitivity due to NMES. Patients who were post-THA/TKA find the application of calf muscle NMES comfortable and acceptable as a treatment. We conclude that the use of NMES on postoperative orthopaedic patients can be safely considered as a DVT prevention method. The results of this study show a beneficial hemodynamic response to NMES in post orthopaedic surgery patients. Conflict of interest There are no conflicts of interest to the authors’ knowledge. References [1] Browse NL, Negus D. Prevention of postoperative leg vein thrombosis by electrical muscle stimulation. An evaluation with 125 I-labelled fibrinogen. Br Med J 1970;3:615–8. [2] Dejode LR, Khurshid M, Walther WW. The influence of electrical stimulation of the leg during surgical operations on the subsequent development of deep-vein thrombosis. Br J Surg 1973;90:31–2. [3] Doran FSA, White HM. A demonstration that the risk of postoperative deep venous thrombosis is reduced by stimulating the calf muscles electrically during the operation. Br J Surg 1967;54:686–9. [4] Nicolaides AN, Kakkar VV, Field ES, Fish P. Optimal electrical stimulus for prevention of deep vein thrombosis. Br Med J 1972;3:756–8. [5] Pambianco G, Orchard T, Landau P. Deep vein thrombosis: prevention in stroke patients during rehabilitation. Arch Phys Med Rehabil 1995;76:324–30. [6] Broderick B, Breen P, ÓLaighin G. Electronic stimulators for surface neural prosthesis. J Autom Control 2008;18:25–33. [7] Geerts WH, Bergqvist D, Pineo GF, Heit JA, Samama CM, Lassen MR, Colwell CW. Prevention of venous thromboembolism. Chest 2008;133:381S–453S. [8] Baker LL, McNeal DR, Benton LA, Bowman BR, Waters RL.Neuromuscular electrical stimulation – a practical guide. 3rd ed. Downey California: Rancho Los Amigos Research and Education Institute; 1993. [9] Breen PP, Corley GJ, O’Keeffe DT, Conway R, ÓLaighin G. A programmable and portable NMES device for drop foot correction and blood flow assist applications. Med Eng Phys 2009;31:400–8. [10] Broderick BJ, O’Briain DE, Breen PP, Kearns SR, ÓLaighin G. A pilot evaluation of a neuromuscular electrical stimulation (NMES) based methodology for the prevention of venous stasis during bed rest. Med Eng Phys 2010;32:349–55. [11] Clarke Moloney M, Lyons GM, Breen P, Burke PE, Grace PA. Haemodynamic study examining the response of venous blood flow to electrical stimulation of the gastrocnemius muscle in patients with chronic venous disease. Eur J Vasc Endovasc Surg 2006;31:300–5.
B.J. Broderick et al. / Medical Engineering & Physics 33 (2011) 56–61 [12] Collins S, Moore R, McQuay H. The visual analogue pain intensity scale: what is moderate pain in millimetres? Pain 1997;72:95–7. [13] Kelly A-M. The minimum clinically significant difference in visual analogue scale pain score does not differ with severity of pain. Emerg Med J 2001;18:205–7. [14] Nicolaides A, Kakkar V, Field E, Fish P. Optimal electrical stimulus for prevention of deep vein thrombosis. Br Med J 1972;3:756. [15] Merli GJ, Herbison GJ, Ditunno JF, Weitz HH, Henzes JH, Park CH, Jaweed MM. Deep vein thrombosis: prophylaxis in acute spinal cord injured patients. Arch Phys Med Rehabil 1988;69:661–4. [16] Westrich GH, Menezes A, Sharrock N, Sculco TP. Thromboembolic disease prophylaxis in total knee arthroplasty using intraoperative heparin and postoperative pneumatic foot compression. J Arthroplasty 1999;14:651–6.
61
[17] Alon G, Kantor G, Ho HS. Effects of electrode size on basic excitatory responses and on selected stimulus parameters. J Orthop Sports Phys Ther 1994;20:29–35. [18] Lyons G, Leane G, Clarke-Moloney M, O’Brien J, Grace P. An investigation of the effect of electrode size and electrode location on comfort during stimulation of the gastrocnemius muscle. Med Eng Phys 2004;26:873–8. [19] Kaplan RE, Czyrny JJ, Fung TS, Unsworth JD, Hirsh J. Electrical foot stimulation and implications for the prevention of venous thromboembolic disease. Thromb Haemost 2002;88:200–4. [20] Lyons GM, Sinkjær T, Burridge JH, Wilcox DJ. A review of portable FES-based neural orthoses for the correction of drop foot. IEEE Trans Neural Syst Rehabil Eng 2002;10:260–79. [21] Keller T, Kuhn A. Electrodes for transcutaneous (surface) electrical stimulation. J Autom Control 2008;18:35–45.