Functional electrical stimulation of the ankle dorsiflexor muscles effectively reduces foot drop in stroke patients, but also is associated with limitations. Now researchers are looking to address those limitations by adding plantar flexor stimulation to the mix.
By Emily Delzell
Functional electrical stimulation of the ankle dorsiflexors is a commonly used method of addressing poststroke gait deficits, including foot drop. Although this modality can correct swing-phase foot drop for many patients, stimulating the dorsiflexors alone fails to address other gait deficits. Consequently, researchers are exploring the efficacy of alternative therapeutic approaches, including FES to both the ankle dorsiflexor and plantar flexor muscles.
Gait deficits are common after stroke, even after individuals undergo rehabilitation, and recovering walking ability is often a major focus among survivors. Foot drop, which affects up to 20% of stroke patients1 and is caused by total or partial paralysis of the dorsiflexor muscles, can cause affected individuals to develop compensatory motor patterns such as circumduction, which occurs during swing phase and involves hip abduction and external rotation, and hip hiking on the paretic side. These patterns can result in other issues, including increased energy costs, reduced endurance and walking speed, and falls.2
“There is a need to develop interventions that can address multiple gait problems at multiple joints, including important stance and swing-phase poststroke gait deficits at the hip and knee,” explained Trisha M. Kesar, PT, PhD, a research scientist in the University of Delaware’s Department of Physical Therapy.
Kesar’s research group is currently conducting a large National Institutes of Health-funded multidisciplinary project aimed at developing new intervention strategies that combine FES with fast treadmill training to maximize improvements in poststroke walking function.
Supplementing dorsiflexor FES with plantar flexor stimulation offers a number of potential advantages. Along with foot drop, poststroke gait deficits also encompass decreased propulsive force generation by the plantar flexors, which is related to deficits during both swing and stance phases as well as to slow walking speeds.3 In addition, Kesar’s group recently documented unexpected effects with dorsiflexor FES—specifically, a decrease in swing-phase knee flexion during walking compared with walking without FES2—that could theoretically be addressed with plantar flexor stimulation.
The Physical Therapy study2 in which Kesar et al noted this phenomenon was aimed at investigating swing-phase knee and ankle motion kinematics during dorsiflexor FES using two different types of frequency trains, or patterns of pulses, for stimulation. The constant-frequency trains (CFTs) typically used with FES were compared to variable-frequency trains (VFTs), which have been shown to enhance isometric and nonisometric muscle performance.4-7
Investigators worked with 13 hemiparetic individuals (nine men, four women; aged 46-72 years) who were at least six months poststroke, were able to walk for five minutes at self-selected speeds, and had sufficient passive dorsiflexion range of motion in the paretic ankle joint to achieve at least 5° of plantar flexion with the knee flexed.
Investigators attached electrodes to participants’ dorsiflexor muscles and set stimulation amplitude by gradually increasing the amplitude of a 300-ms-long, 30-Hz train with a pulse duration of 300 microseconds until participants achieved a neutral ankle joint position (0°) or at least 5° of plantar flexion in patients with range of motion deficits.2
To control timing during gait, two footswitches were attached bilaterally to the soles of each shoe, one under the forefoot and the other under the hindfoot. FES was delivered to paretic dorsiflexors during swing phase as detected by the footswitches (i.e., from when the forefoot footswitch left the ground to when the hindfoot footswitch touched the ground).
For FES using CFTs investigators delivered a 30-hz constant frequency train. For FES with VFTs they delivered a 200-Hz three-pulse burst followed by lower frequency (30-Hz) CFT. All parameters for stimulation with VFTs and CFTs were the same except for the 200-Hz three-pulse burst used at initiation of the VFTs.
Investigators compared three walking conditions: walking without FES, walking with dorsiflexor FES using CFTs, and walking with dorsiflexor FES using VFTs. All conditions involved 20- to 40-second periods of walking at self-selected speeds.
Both CFT and VFT FES improved ankle dorsiflexion angles during walking compared with no FES. The VFT FES walking condition significantly enhanced gait performance compared with CFT FES, generating greater swing-phase ankle dorsiflexion.
Investigators, however, were surprised to find that the dorsiflexor FES decreased ankle plantar flexion at toe-off as well as swing-phase knee flexion; the use of VFTs compared with CFTs enhanced these effects. Because hemiparetic poststroke individuals are already more likely to demonstrate swing-phase knee flexion deficits in the paretic leg, the finding suggests that dorsiflexor FES actually exacerbates this type of gait impairment and in turn may adversely affect foot clearance during swing phase.
The decreased plantar flexion angles could lead to reduced push-off forces at the ankle during the transition from swing to stance, the authors suggested.
Timing of the electrical stimulation, which is crucial to the effectiveness of FES, could have contributed to the surprising findings, according to Kesar et al. Initiation of dorsiflexor FES began when the forefoot footswitch left the ground, which was just prior to actual toe-off. So at toe-off, the net plantar flexor moment could have been reduced by forces generated by the dorsiflexor muscles, which could have led to the observed reduced plantar flexion angle at toe-off and reduced knee flexion during swing phase.
The authors hypothesized that delivering stimulation to the plantar flexors could potentially attenuate this issue by increasing force generation during push-off.
“By stimulating both the plantar and dorsiflexors, we can produce improvements not only in foot drop, but also in knee flexion during swing and in push-off forces, walking speed, and other deficits,” Kesar said. “In addition, unlike neuroprostheses, the other commonly used intervention for foot drop, FES functions as a motor learning or gait training tool that patients can use to take advantage of the brain’s plasticity to ‘relearn’ correct walking patterns.”
Kesar and colleagues tested this hypothesis in a 2009 study published in Stroke.3 This study utilized the same participant group as the Physical Therapy study.
Investigators used a customized FES system (no current commercially available systems can provide FES to both dorsi- and plantar flexors to deliver VFT stimulation during gait). They compared three walking conditions: walking without FES, walking with VFT FES to the dorsiflexor muscles during the swing phase, and walking with VFT FES delivered to both the dorsiflexor and plantar flexor muscles.
The timing of electrical stimulation is critical, Kesar explained.
“During normal walking, dorsiflexor and plantar flexor muscles are on during different parts of the gait cycle. Thus, the FES system needs to enable accurate timing of each muscle being stimulated at the appropriate phase of the cycle,” she said. “The system used in our study delivers stimulation to the dorsiflexors when the foot is in the air—swing—and to the plantar flexors when the foot is about to transition from stance to swing—i.e., during push-off.”
During walking without FES, the average ankle plantar flexion angle at paretic toe-off was -9.2±1.2°. Investigators noted a significant reduction in ankle plantar flexion at toe-off during dorsiflexor stimulation (-3.1±1.5°) compared with the other two walking conditions (no FES and combined FES). There were no differences in ankle angle at toe-off between no FES and combined FES (-8.3±1.3°, P=0.41).
They also found significant reduction in peak swing-phase knee flexion angles during dorsiflexor FES (40.8±4.2°) compared with no FES and combined FES, but no difference in peak swing-phase knee flexion between no FES (44.1±4.2°) and combined FES (44.3±4.6°, P=0.81).
Compared with walking without FES the peak anterior ground reaction forces (AGRF) for the paretic leg showed an 18% increase during walking with FES to both the dorsi- and plantar flexors. Data showed no difference in peak AGRF between dorsiflexor-alone FES and combined dorsi- and plantar flexor FES.3
Combined FES produced significantly greater peak swing-phase ankle dorsiflexion than no FES but significantly less than with dorsiflexor-alone FES, suggesting that dorsiflexor stimulation intensity may need to be increased during periods when both dorsi- and plantar flexors are being stimulated simultaneously.3
“So far, the combined approach appears promising, and our ongoing research shows improvements in walking speed, endurance, activity, and participation with combined FES when used in conjunction with fast treadmill training, a modality we’re currently evaluating,” Kesar said. “We have also shown the immediate effects of combined FES are greater than those with dorsiflexor-alone FES. However, we don’t know whether the long-term effects of combined FES will be greater. But based on the importance of the plantar flexor muscles during walking and on previous research that has shown the importance of push-off, we think that the combined approach may be better at generating therapeutic effects than the dorsiflexor FES alone.”
Using combined dorsi- and plantar flexor FES to retrain the brain for efficient walking is also the focus of research by a group of investigators at the MultiCare Good Samaritan Physical Medicine and Rehabilitation Center in Puyallup, WA.
A recent randomized crossover trial involving 28 adults with chronic hemiplegia who were 4.9±3.8 years poststroke looked at the effect of combined FES on walking as measured by a six-minute walk test (6MWT), Emory Functional Ambulatory Profile (EFAP), and the Stroke Impact Scale (SIS).8
Investigators randomized participants to two interventions that each lasted three months, at which point participants crossed over to the other intervention. In intervention A participants wore a customized FES system (now patented by the MultiCare Good Samaritan group as the Gait MyoElectric Stimulator; see figures 1 and 2) for six to eight hours a day, seven days a week, that activated automatically during walking. During intervention A participants also walked with the system one hour a day, six days a week, at a self-selected speed during sessions of at least 15 minutes. In intervention B, participants walked one hour a day for six days a week without FES. Group A-B had 15 individuals, while group B-A had 13.
The customized FES system was designed with a heel and forefoot switches and was placed in the shoe of the nonparetic leg to stimulate the paretic limb. The system stimulated paretic dorsiflexors during loading response, preswing, and swing, and paretic plantar flexors during midstance and terminal stance. Participants were able to put on the FES system themselves or with the help of a caregiver and wore it daily, which allowed them to use it in their natural environment.
Importantly, said lead author David G. Embrey, PhD, PT, none of the study pre- or postintervention assessments were done while participants were wearing the stimulator.
“This allowed us to test whether the system was actually creating an environment that enabled the brain to help patients relearn to walk,” said Embrey, who is research program coordinator of the Children’s Therapy Unit in the Good Samaritan Movement Laboratory. “I’m not aware of any other studies that have assessed the effects of combined dorsi- and plantar flexor FES during real-world walking conditions after stimulation is removed. In addition, research documents that most of propulsion during walking comes from the plantar flexors. If you’re helping the foot to lift with just dorsiflexors and not propelling the body forward, you’re not creating natural steps. So if you can propel your body forward with the plantar flexors and lift during the dorsiflexor phases, then you’re creating a more natural walking pattern.”
At study midpoint, the A intervention produced statistically significant improvements on two primary study outcome measures compared with intervention B. The 6MWT improved (P=0.02; mean ± SD scores of 47.7±40.3m for intervention A compared with 18.4±16.5m for intervention B). At three months the SIS for the A group showed statistically significant improvements (P=0.03; mean SD scores 23.5±26.6 points compared with intervention B; mean ± SD scores 3.6±28).
At three months, the difference in time to complete the EFAP between the A and B interventions approached statistical significance (P=0.08; mean 23.7±23.9 seconds in the A intervention compared with 9.8±8.9 seconds in the B intervention.)
At six months, overall gains in the 6MWT were 57.1±35.7m for the A-B group and 32.3±20.5m for the B-A groups. EFAP times decreased 30.9±24.4 seconds and 21.6±11.4 seconds for the A-B and B-A groups, respectively, while SIS scores increased by 45.4±47.0 points in the A-B group and 33.4±30.7 points in the B-A group. All changes were significantly improved (P=0.01) at the end of the study compared with baseline assessments.
The study’s crossover design also allowed investigators to look at carryover effects in the A-B group, who stopped using the FES system at three months but continued to walk an hour a day, six days a week.
“Participants not only maintained their improvements, they continued to make small gains,” Embrey said.
During the B intervention, the A-B group walked an additional 9.3±36.4m on the 6MWT, decreased EFAP time by another 7.2±5.3 seconds, and gained 24.8±48.7 points on the SIS.
“It’s interesting that participants who received stimulation in the first three months of the study did better than those who received it after the walking intervention,” Embrey said. “One explanation is that the people who received the stimulation first were walking better at crossover, felt more confident about their ability, and, in the second three months, built on the skill they had already developed.”
Appropriate timing of stimulation is critical to creating a normal gait pattern, and has not been on target in a number of studies of FES, Embrey said.
“Most studies of dorsiflexor FES for drop foot have applied stimulation only during the swing phase, yet during normal gait firing, dorsiflexors activate more during loading response than at any time during that gait cycle,” he said. “And EMG recordings of normal walking show that plantar flexors fire at the middle part of the stance phase, not at the end of stance, which is when many other researchers have timed plantar flexor stimulation.”
One of those “other researchers,” Kesar, defended her team’s choices.
“Our FES system stimulates the dorsiflexors during swing phase to correct foot drop. Loading response is the phase that immediately follows the end of swing phase. There is residual force in the muscle even after the FES switches off, due to which we observe that the FES is able to produce mild effects during loading response as well. That is, the foot shows a more controlled plantar flexion and does not ‘slap’ during loading response,” she said.
“With regard to timing of plantar flexion, the plantar flexors play a role during early/mid-stance to control forward progression of the tibia, or eccentric contraction. They also play a role during end-stance to generate the push-off force needed for propulsion. Due to the recent studies by Bowden et al9 and Balasubramanian et al10 showing the importance of propulsion in stroke gait, we have chosen to focus on this push-off role of the plantar flexors in our FES paradigm instead of the role played by these muscles in early or mid-stance phases.”
In clinical rehabilitation settings physical therapists work with patients who exhibit a wide range of variability in their gait, stance, and control deficits as well as in their habitual compensatory patterns, said Alison Lichy, PT, DPT, NCS, who owns and practices with the neurological physical therapy group at NeuroPT in Falls Church, VA.
“We use FES in a number of different ways in the clinic, combining it with gait and other training depending on the individuals’ needs and concerns,” Lichy said. “Some patients, for example, may be more concerned about their walking speed, while others may be more concerned about balance or foot clearance.”
The concepts behind the current research on combined FES make sense and are a worthwhile avenue of investigation, but practically speaking, it may work better for some patients than others, she said.
“In the big clinical picture, patients’ most common complaint is that their foot is dragging, not that they need more assistance to propel forward,” Lichy said. “Clinically, a lot of these patients who are having trouble with foot drag don’t step, step, step, at the same pace all the time. They have many biomechanical gait issues, and will step and hesitate, step and hesitate, etc., so there’s a lot of other training that has to occur once you start stimulating muscles in their ankle.”
Bracing with ankle foot orthoses can play a large role in poststroke gait performance, she said.
“Braces can provide much-needed support, but must be individualized and appropriate to the patient’s gait pattern and control. But for individuals who need more support than FES can provide, it’s important to evaluate bracing,” Lichy said.
Lichy uses dorsiflexor-only FES in her clinic, but some of her patients have participated in studies of combined FES systems. The technology, she notes, doesn’t work for all her patients.
“Some respond well, while others dislike the systems or find them ineffective. It’s hard to predict who is going to do well. I’ve worked with very low functioning individuals with a lot of spasticity and applied the FES without seeing a lot of contraction, yet these patients say they ‘just feel better’ when walking,” she said. “Patients with neurological issues report FES helps make them aware of the affected leg. It all boils down to giving patients the support they need to walk as efficiently as they can.”
1. Burridge JH, Taylor PN, Hagan SA, et al. The effects of common peroneal stimulation on the effort and speed of walking: a randomized controlled trial with chronic hemiplegic patients. Clin Rehabil 1997;(3):201-210.
2. Kesar TM, Perumal R, Jancosko A, et al. Novel patterns of functional electrical stimulation have an immediate effect on dorsiflexor muscle function during gait for people poststroke. Phys Ther 2010;90(1):55-66.
3. Kesar TM, Perumal R, Reisman DS, et al. Functional electrical stimulation of ankle plantarflexor and dorsiflexor muscles: effects on poststroke gait. Stroke 2009;40(12):3821-3827
4. Binder-Macleod S, Kesar T. Catchlike property of skeletal muscle: recent findings and clinical implications. Muscle Nerve 2005;31(6):681-693.
5. Lee SC, Becker CN, Binder-Macleod SA. Catchlike-inducing train activation of human muscle during isotonic contractions: burst modulation. J Appl Physiol 1999;87:1758-1767.
6. Lee SC, Binder-Macleod SA. Effects of activation frequency on dynamic performance of human fresh and fatigued muscles. J Appl Physiol 2000;88(6):2166-2175.
7. Maladen RD, Perumal R, Wexler AS, Binder-Macleod SA. Effects of activation pattern on nonisometric human skeletal muscle performance. J Appl Physiol 2007;102(5):1985-1991.
8. Embrey DG, Holtz SL, Alon G, et al. Functional electrical stimulation to dorsiflexors and plantar flexors during gait to improve walking in adults with chronic hemiplegia. Arch Phys Med Rehabil 2010;91(5):687-696.
9. Bowden MG, Balasubramanian CK, Neptune RR, Kautz SA. Anterior-posterior ground reaction forces as a measure of paretic leg contribution in hemiparetic walking. Stroke 2006;37(3):872-876.
10. Balasubramanian CK, Bowden, MG, Neptune, RR, Kautz, SA. Relationship between step length asymmetry and walking performance in subjects with chronic hemiparesis. Arch Phys Med Rehabil 2007;88(1):43-49.