Functional Neuromuscular Electrical Stimulation
*Medicare has a policy.
NOTE: This policy does not address commercially available exercyles that use electrical muscle stimulation technology for rehabilitation during clinic based physical therapy. Functional Electrical Stimulator (FES) devices, such as the RT 300 (Restorative Therapies, Inc.) motorized FES ergometer cycle (when used independently in the home), should not be confused with functional neuromuscular electrical stimulation (NMES) neuroprosthesis devices.
Although the devices are similar in name, the Plan considers FES devices to be exercise equipment. Most contract plans exclude coverage of exercise equipment for use in the home. Code E0764 is considered a contract exclusion and is therefore not covered. Please refer to the contract language for details.
Sacral Nerve Modulation/Stimulation
Functional neuromuscular electrical stimulation (NMES) is a method being developed to restore function to patients with damaged or destroyed nerve pathways (e.g., stroke, spinal cord injury, multiple sclerosis [MS], cerebral palsy) through use of an orthotic device with microprocessor controlled electrical neuromuscular stimulation (neuroprosthesis). NMES are used to prevent or retard disuse atrophy, relax muscle spasm, increase blood circulation, maintain or increase range of motion, and re-educate muscles.
Neural prosthetic devices consist of an orthotic and a microprocessor-based electronic stimulator with one or more channels for delivery of individual pulses through surface or implanted electrodes connected to the neuromuscular system. Microprocessor programs activate the channels sequentially or in unison to stimulate peripheral nerves and trigger muscle contractions to produce functionally useful movements that allow patients to sit, stand, walk, and grasp. Functional neuromuscular stimulators are closed-loop systems, which provide feedback information on muscle force and joint position, thus allowing constant modification of stimulation parameters which are required for complex activities such as walking. These are contrasted with open-loop systems, which are used for simple tasks such as muscle strengthening alone, and typically in healthy individuals with intact neural control.
One application of functional NMES is to restore upper extremity functions such as grasp-release, forearm pronation, and elbow extension in patients with stroke, or C5 and C6 tetraplegia (quadriplegia). The Neurocontrol Freehand® system is an implantable upper extremity neuroprosthesis intended to improve a patient's ability to grasp, hold, and release objects and is indicated for use in patients who are tetraplegic due to C5 or C6 spinal cord injury. The implantable Freehand System is no longer marketed in the United States, though the company provides maintenance for devices already implanted. The NESS H200® (previously known as The Handmaster NMS I system) is another device that uses surface electrodes and is purported to provide hand active range of motion and function for patients with stroke or C5 tetraplegia.
Other neural prosthetic devices have been developed for functional NMES in patients with footdrop. Footdrop is weakness of the foot and ankle that causes reduced dorsiflexion and difficulty with ambulation. It can have various causes such as cerebral palsy, stroke or MS. Functional electrical stimulation of the peroneal nerve has been suggested for these patients as an aid in raising the toes during the swing phase of ambulation. In these devices, a pressure sensor detects heel off and initial contact during walking. A signal is then sent to the stimulation cuff, initiating or pausing the stimulation of the peroneal nerve, which activates the foot dorsiflexors. Examples of such devices used for treatment of footdrop are the Innovative Neurotronics’ (formerly NeuroMotion Inc.) WalkAide®, Bioness’ radiofrequency controlled NESS L300™, and the Odstock Foot Drop Stimulator. An implantable peroneal nerve stimulator system (ActiGait) is being developed in Europe.
Another application of functional electrical stimulation is to provide spinal cord -injured patients with the ability to stand and walk. Generally, only spinal cord injury patients with lesions from T4 to T12 are considered candidates for ambulation systems. Lesions at T1 to T3 are associated with poor trunk stability, while lumbar lesions imply lower -extremity nerve damage. Using percutaneous stimulation, the device delivers trains of electrical pulses to trigger action potentials at selected nerves at the quadriceps (for knee extension), the common peroneal nerve (for hip flexion), and the paraspinals and gluteals (for trunk stability). Patients use a walker or elbow-support crutches for further support. The electrical impulses are controlled by a computer microchip attached to the patient’s belt that synchronizes and distributes the signals. In addition, there is a finger-controlled switch that permits patient activation of the stepping.
Other devices include a reciprocating gait orthosis with electrical stimulation. The orthosis used is a cumbersome hip-knee-ankle-foot device linked together with a cable at the hip joint. The use of this device may be limited by the difficulties in putting the device on and taking it off.
Neuromuscular stimulation is also proposed for motor restoration in hemiplegia and treatment of secondary dysfunction (e.g., muscle atrophy and alterations in cardiovascular function and bone density) associated with damage to motor nerve pathways. These applications are not addressed in this policy. (See Benefit Application section.)
The Neurocontrol Freehand system received approval from FDA in 1997 through the pre-market approval (PMA) process.
The Handmaster NMS I system (now named NESS H200) was originally cleared for use in maintaining or improving range of motion, reducing muscle spasm, preventing or retarding muscle atrophy, providing muscle re-education, and improving circulation; in 2001, its 510(k) marketing clearance was expanded to include provision of hand active range of motion and function for patients with C5 tetraplegia. FDA product code: GZC.
The WalkAide device first received 510(k) marketing clearance from FDA in the 1990s; the current version of the WalkAide device received 510(k) marketing clearance in September 2005.
The Odstock Foot Drop Stimulator received 510(k) marketing clearance in 2005.
The Bioness NESS L300 received 510(k) marketing clearance in July 2006. FDA summaries for the devices state that they are intended to be used in patients with drop foot by assisting with ankle dorsiflexion during the swing phase of gait. FDA product code: GZI
To date, the Parastep® Ambulation System is the only noninvasive functional walking neuromuscular stimulation device to receive PMA from FDA. The Parastep device is approved to “enable appropriately selected skeletally mature spinal cord injured patients (level C6-T12) to stand and attain limited ambulation and/or take steps, with assistance if required, following a prescribed period of physical therapy training in conjunction with rehabilitation management of spinal cord injury.” FDA product code: MKD
Medical policies are systematically developed guidelines that serve as a resource for Company staff when determining coverage for specific medical procedures, drugs or devices. Coverage for medical services is subject to the limits and conditions of the member benefit plan. Members and their providers should consult the member benefit booklet or contact a customer service representative to determine whether there are any benefit limitations applicable to this service or supply. This medical policy does not apply to Medicare Advantage.
Note: Functional Electrical Stimulator (FES) devices, such as the RT 300 motorized FES ergometer cycle, should not be confused with functional neuromuscular electrical stimulation (NMES) neuroprosthesis devices. Although the device classifications are similar in name, the Company considers FES devices to be exercise equipment. Most contract plans exclude coverage of exercise equipment for use in the home. Please refer to the member’s contract language for details.
Assessment of efficacy for therapeutic interventions involves a determination of whether the intervention improves health outcomes. The optimal study design for a therapeutic intervention is a randomized controlled trial (RCT) that includes clinically relevant measures of health outcomes. Intermediate outcome measures, also known as surrogate outcome measures, may also be adequate if there is an established link between the intermediate outcome and true health outcomes. Nonrandomized comparative studies and uncontrolled studies can sometimes provide useful information on health outcomes but are prone to biases such as noncomparability of treatment groups, the placebo effect, and variable natural history of the condition.
Ambulation in Patients with Spinal Cord Injury
The clinical impact of the Parastep® device rests on identification of clinically important outcomes. The primary outcome of the Parastep® device, and the main purpose of its design, is to provide a degree of ambulation that improves the patient’s ability to complete the activities of daily living (ADLs), or positively affect the patient’s quality of life. Physiologic outcomes (i.e., conditioning, oxygen uptake) have also been reported, but these are intermediate, short-term outcomes, and it is not known whether similar or improved results could be attained with other training methods. In addition, the results are reported for mean peak values, which may or may not be a consistent result over time. The effect of the Parastep® on physical self-concept and depression are secondary outcomes and similar to the physiologic outcomes; interpretation is limited due to lack of comparison with other forms of training.
The largest study was conducted by Chaplin, who reported on the ambulation outcomes using the Parastep® 1 in 91 patients. (1) Of these 91 patients, 84 (92%) were able to take steps and 31 (34%) were able to eventually ambulate without assistance from another person. Duration of use was not reported. Other studies on the Parastep® device include a series of 5 studies from the same group of investigators, which focused on different outcomes in the same group of 13 to 16 patients.(2-6) In a 1997 study, Guest et al reported on the ambulation performance of 13 men and 3 women with thoracic motor complete spinal injury. (5) All patients underwent 32 training sessions before measuring ambulation. The group’s mean peak distance walked was 334 meters, but there was wide variability, as evidenced by a standard deviation of 402 meters. The mean peak duration of walking was 56 minutes, again with wide variability, evidenced by a standard deviation of 46 minutes. It should be noted that peak measures reflect the best outcome over the period evaluated; peak measures may be an inconsistent, one-time occurrence for the individual patient. The participants also underwent anthropomorphic measurements of various anatomic locations. Increases in thigh and calf girth, thigh cross-sectional area, and calculated lean tissue were all statistically significant. The authors emphasize that the device is not intended to be an alternative to a wheelchair, and thus other factors such as improved physical and mental well-being should be considered when deciding whether or not to use the system. The same limitations were noted in a review article by Graupe and Kohn, who state that the goal for ambulation is for patients to get out of the wheelchair at-will, stretch, and take a few steps every day. (7)
Jacobs et al reported on physiologic responses related to use of the Parastep® device. (3) There was a 25% increase in time to fatigue and a 15% increase in peak values of oxygen uptake, consistent with an exercise training effect. There were no significant effects on arm strength. Needham-Shropshire et al reported no relationship between use of the Parastep® device and bone mineral density, although the time interval between measurements (12 weeks) and the precision of the testing device may have limited the ability to detect a difference. (4) Nash et al reported that use of the Parastep device was associated with an increase in arterial inflow volume to the common femoral artery, perhaps related to the overall conditioning response to the Parastep®. (6) Also, Guest et al reported significant improvements in physical self-concept and decreases in depression scores. (5) Finally, it should be noted that evaluations of the Parastep® device were performed immediately following initial training or during limited study period durations. (1, 8-10) There are no data regarding whether patients remained compliant and committed with long-term use.
Brissot et al reported independent ambulation was achieved in 13 of 15 patients, with 2 patients withdrawing from the study. (8) In the home setting, 5 of the 13 patients continued using the device for physical fitness, but none used it for ambulation. Sykes et al found low use of a reciprocating gait orthosis device with or without stimulation over an 18-month period. (10) In addition, the more recent Davis et al study of a surgically implanted neuroprosthesis for standing and transfers after spinal cord injury showed mixed usability/preference scale results for ambulation with device assistance versus conventional transfers in 12 patients followed up for a 12-month period postdischarge.(9) Therefore, the advantage of using device assistance could not be evaluated.
The effect of a surgically implanted neuroprosthesis on exercise, standing, transfers, and quality of life was reported in 2012. (11, 12) This study was supported by the U.S. Department of Veterans Affairs, the Office of Orphan Product Development of the Food and Drug Administration, the New York State Department of Health, and the National Center for Research Resources of the National Institutes of Health. The device is not commercially available at this time.
As stated by various authors, the Parastep system is not designed to be an alternative to a wheelchair and offers, at best, limited, short-term ambulation. Final health outcomes, such as ability to perform ADLs or quality of life, have not been reported.
Functional Neuromuscular Electrical Stimulation of the Upper Extremity
Spinal Cord Injury
Most of the early published evidence for upper -extremity devices to restore function in patients with spinal cord injuries report experience with the Freehand System, an implantable device that is no longer marketed in the United States. (13-15) The device is controlled through a joystick on the shoulder or wrist. A disadvantage of this system is that additional surgery is required to repair hardware failures. The published studies, all case series with fewer than 10 subjects, suggest that the device may give patients the ability to grasp and release objects and independence or greater independence in such ADLs as using a fork or the telephone in the study setting. User satisfaction was generally high, and most subjects reported continued use of the device at home, although details of specific activities or frequency of use at home are not provided. In a review of the role of electrical stimulation for rehabilitation and regeneration after spinal cord injury, Hamid and Hayek report that the company which marketed the Freehand System in the United States no longer manufactures new devices. (16)
Use of the Handmaster NMS I (NESS L200) was reported in a series of 10 patients with cervical spinal cord injuries. (17) After 2 months of training, performance on a defined set of tasks and one or more tasks chosen by the patient was evaluated. In six patients, a stimulated grasp and release with either one or both grasp modes (key- and palmar pinch) of the Handmaster was possible. Four patients could perform the set of tasks using the Handmaster, while they were not able to do so without the Handmaster. Eventually, one patient continued using the Handmaster during ADLs at home. In another study using the Handmaster device, 7 subjects with C5 or C6 spinal cord injury practiced using the device daily on one of their paralyzed hands to regain the ability to grasp, hold, and release objects. (18) They were observed 2 to 3 times weekly for 3 weeks, and their ability to pick up a telephone, eat food with a fork, and perform an individually selected ADL task plus 2 grasp, hold, and release tasks was evaluated. At the end of the study, all 7 subjects were successful at using the device in the studied ADLs and grasp, hold, and release tasks. Improvements occurred in secondary measures of grip strength, finger linear motion, and Fugl-Meyer (developed to assess sensory-motor recovery after stroke) scores.
Hamid notes that, with either device, there is a time delay of 1 to 2 seconds between command generation and execution of grasp function that interferes with the speed with which the patient can grasp and release objects.
Alon et al., reporting on a case series of 29 patients, investigated whether the Handmaster system (NESS L200) could improve selected hand function in persons with chronic upper extremity paresis following stroke. (19) The main outcome measures were 3 ADL tasks: lifting a 2-handled pot, holding a bag while standing with a cane, and another ADL chosen by the patient. Secondary measures included lifting a 600 -gram weight, grip strength, electrically induced finger motion, Fugl-Meyer spherical grasp, and perceived pain scale. At the end of the 3-week study period, the percent of successful trials compared with baseline were: lifting pot, 93% versus 0%, lifting 600 -gram weight, 100% versus 14%, and lifting bag, 93% versus 17%—all respectively. All subjects performed their selected ADL successfully and improved their Fugl-Meyer scores using the neuroprosthesis.
Interpretation of the evidence for upper -extremity neuroprostheses for patients with spinal cord injuries or post-stroke is limited by the small number of subjects and lack of data demonstrating its utility outside the study setting. The available evidence is insufficient to conclude that NMES improves outcomes by providing some upper -extremity function.
Functional NMES for Chronic Foot-drop
Stroke and Spinal Cord Injury
Functional NMES with a foot-drop stimulator (WalkAide) was compared with an ankle-foot orthosis in an industry-affiliated multicenter RCT (NCT01087957) that included 495 Medicare-eligible individuals who were at least 6 months poststroke. (20) A total of 399 individuals completed the 6-month study. Primary outcome measures were the 10-Meter Walk Test (10MWT), a composite measure of daily function, and device-related serious adverse event rates (SAEs). There were 7 secondary outcome measures that assessed function and quality of life. Intention-to-treat analysis found that both groups improved walking performance over the 6 months of the study, and the NMES device was non-inferior to the ankle-foot orthosis on the primary outcome measures. Only the WalkAide group showed significant improvements from baseline to 6 months on several secondary outcome measures, but there were no significant between-group differences for any of the outcomes.
FASTEST (NCT01138995) is an industry-sponsored single-blinded multicenter trial that randomized 197 patients to 30 weeks of a foot-drop stimulator (NESS L300) or a conventional ankle-foot orthosis (AFO). (21) The AFO group received transcutaneous electrical nerve stimulation at each physical therapy visit during the first 2 weeks to provide a sensory control for stimulation of the peroneal nerve in the NESS L300 group. Evaluation by physical therapists who were blinded to group assignment found that both groups improved gait speed and other secondary outcome measures over time, with similar improvement in the 2 groups. There were no between-group differences in the number of steps per day at home, which were measured by an activity monitor over a week. User satisfaction was higher with the foot-drop stimulator.
Secondary analysis of data from this study was reported in 2014. (22) Comfortable gait speed was assessed in the 99 individuals from the NESS L300 group at 6, 12, 30, 36, and 42 weeks, with and without use of the foot drop stimulator. A responder was defined as achieving a minimal clinically important difference (MCID) of 0.1 m/sec on the 10MWT or advancing by at least 1 Perry Ambulation Category. Non-completers were classified as non-responders. Seventy percent of participants completed the assessments at 42 weeks, and 67% of participants were classified as responders. Of the 32 participants who were classified as non-responders, 2 were non-responders and 30 were non-completers. The percentage of patients in the conventional AFO group who were classified as responders at 30 weeks was not reported. There were 160 adverse events (AEs), of which 92% were classified as mild. Fifty percent of the AEs were related to reversible skin issues and 27% were falls.
Prospective Crossover Trials
A multicenter within-subject crossover trial of the WalkAid footdrop stimulator versus conventional AFO was published in 2013. (23) Patients who had a stroke within the previous12 months and residual foot-drop but no prior experience with an orthotic device were randomly assigned to WalkAid followed by AFO (6 weeks each, n=38), AFO followed by WalkAid (n=31), or AFO for 12 weeks (n=24). Walking tests were performed both with and without a device at 0, 3, 6, 9, and 12 weeks. The orthotic effect of the device is considered to be the immediate effect of NMES measured at any of the time points with the stimulator on compared with off. The therapeutic effect is the improvement over time (improvement in neuromuscular function) measured under the same conditions (i.e., stimulator on vs. on or stimulator off vs. off) at different time points. The Physiological Cost Index (PCI), which is an indication of the amount of effort in walking, is assessed by the difference between resting heart rate and heart rate during walking, divided by the average walking speed. Both devices had significant orthotic (on-off difference) and therapeutic (changes over time when off) effects. The AFO had a greater orthotic effect on walking speed (figure 8 and 10MWT), while the WalkAid tended to have a greater therapeutic effect. The orthotic effect on PCI was significantly higher with an AFO than the WalkAid. Users felt equally safe with the 2 devices. Seventy percent preferred to keep the WalkAid after the 12-week study.
Van Swigchem et al. published a within-subject comparison of a functional NMES device (NESS L300) and AFO in 26 patients with chronic (>6 months) post-stroke foot-drop in 2010. (24) Baseline walking speed on a 10-meter walkway was assessed with the patient’s custom-made AFO; physical activity at home was measured with a pedometer and averaged over 7 days, and satisfaction with the device was assessed with a “purpose-designed” 5-point questionnaire. After a 2-week period of adaptation to the NESS L300, walking speed was assessed with both the AFO and the NMES devices. For the next 6 weeks, patients increased use of the NMES device to the whole day, using the AFO 1 hour a day to maintain familiarity of walking with this device. At the end of the study, walking speed was assessed with both the AFO and the NMES devices, while activity at home and satisfaction were assessed for the NMES device. Two patients dropped out of the study due to discomfort from the electrical stimulation (n=1) and skin reaction to the electrodes (n=1). The remaining 24 patients provided an average satisfaction rating of 3.0 (neutral) for the AFO and 4.0 (satisfied) for the NMES device regarding comfort to wear, appearance, quality of gait, walking distance, effort of walking, and stability during gait. The objective measures of walking speed (1.02 for the AFO, 1.03 for NMES) and steps per day (5541 for the AFO, 5733 for NMES) were not significantly different for the 2 devices.
Uncontrolled Case Series
In 1999, Taylor et al. reported a retrospective study on the clinical use of the Odstock dropped foot stimulator in 151 patients with chronic foot drop resulting from an upper motor lesion. (25) This retrospective study included 27 age-matched able-bodied controls and 140 patients (93%) who used the device for at least 4 1/2 months (111 patients with chronic foot drop due to stroke, 21 patients with multiple sclerosis [MS, described next], 8 patients with incomplete spinal cord injury). The average time since stroke was 5.4 years. Walking speed was assessed on a 10-meter course. In stroke patients, the immediate (orthotic) effect of the stimulation was an increase in walking speed of 12% and a decrease in PCI of 18%. An improvement over time was also observed, with an increase in walking speed of 14% and a reduction of PCI of 19%, suggesting a therapeutic, as well as orthotic effect for this group.
In 2010, Stein et al. reported improvements in both the orthotic and therapeutic effects of NMES in 41 patients with chronic nonprogressive foot drop 26 stroke, nine spinal cord injury, three surgical complication, two head injury, 1 cerebral palsy) and 32 patients with progressive foot drop (described in more detail following) after one, two, three, six, nine, and eleven months of use.(26) With the stimulator on compared with off (orthotic effect), walking speed improved by 5% for a figure 8 (0.59 vs 0.56 m/s) and 6% for a 10MWT (0.80 vs 0.76 m/s). With the stimulator off, walking speed at 3 months had improved by 17% for a figure 8 (0.56 vs 0.48 m/s) and 12% for a 10-meter test (0.76 vs 0.68 m/s, all respectively) compared with baseline. The combined (orthotic and therapeutic) improvement in walking speed over the 3 months was 23% for the figure 8 (0.59 vs 0.48 m/s) and 18% for the 10MWT (0.80 vs 0.68 m/s, both respectively).
The 1999 study by Taylor et al. described earlier included 21 patients with MS. This group showed a 7% decrease in walking speed and a 16% increase in PCI over the course of the study when not using the Odstock dropped foot stimulator (absence of a therapeutic effect), while use of the stimulator (orthotic effect) resulted in an increase in walking speed of 16% and a decrease in PCI of 24%.
In 2009, an RCT of functional NMES to improve walking performance in patients with MS was published by Barrett et al. (27) Fifty-three patients with secondary progressive MS and unilateral dropped foot were randomized to an 18-week program of either NMES of the common peroneal nerve using a single channel Odstock Dropped Foot Stimulator or a home exercise program, and assessed at six, twelve, and eighteen weeks. Patients in the stimulator group were encouraged to wear the device most of the day, switching it on initially for short walks and increasing daily for two weeks, after which they could use the device without restriction. Subjects in the control group were taught a series of exercises tailored to the individual to be done twice daily. The primary outcome measure was walking speed over a ten meter distance. Two secondary outcome measures were energy efficiency based on increase in heart rate during walking and walking distance in three minutes. Six subjects in the NMES group and three in the exercise group dropped out very early in the study, leaving twenty in the NMES group and 24 in the exercise group. In the NMES group, mean changes between baseline and eighteen week measures were non-significant for all three outcome measures, both with and without stimulation. However, within the NMES group, when mean values for walking speed and distance walked were compared with and without stimulation, outcomes were significantly better with stimulation. In the exercise group, increases in walking speed over ten meters and distance walked in three minutes were highly significant (P=0.001 and P=0.005, respectively). At eighteen weeks, the exercise group walked significantly faster than the NMES group (P=0.028). The authors note a number of limitations of their study: power calculations were based on the ten meter walking speed measure only and indicated that 25 subjects would be required in each group, patients were highly selected, clinical assessors also provided treatment (issues with blinding), and the validity and reliability of the three minute walk test have not been confirmed (fatigue prevented use of the validated 6-minute test). In addition, subjects in the exercise group were told they would receive a stimulator at the end of the trial, which may have impacted adherence to the exercise regimen, as well as retention in the trial. The authors concluded that “while a simple program of home exercise therapy appears to significantly increase walking speed and endurance over an eighteen week intervention period, single channel common peroneal stimulation does not. However, it does appear to have a significant orthotic benefit, resulting in significantly increased walking speed and endurance when performance without stimulation is compared to performance with stimulation.”
A 2010 publication by the same group of investigators reported the impact of eighteen weeks of physiotherapy exercises or the Odstock Dropped Foot Stimulator on ADL. (28) Results of 53 patients from the trial previously described were reported, using the Canadian Occupational Performance Measure (COPM). The COPM is a validated semi-structured interview that was originally designed to assist the design of occupational therapy interventions. The interviews at baseline identified 265 problems of which 260 activities were related to walking and mobility. Subjective evaluation at eighteen weeks showed greater improvements in performance and satisfaction scores in the NMES group (35% of problems had an increased score of 2 or more) than the exercise group (17% of problems had an increased score of 2 or more). The median satisfaction rating improved from 2.2 to 4.0 in the NMES group and remained stable (2.6 to 2.4) in the exercise group. The median number of falls recorded per patient over the eighteen week study period was five in the NMES group and 18 in the exercise group. About 70% of the falls occurred while not using the NMES device or an AFO device.
In a preliminary study, Sheffler et al. compared functional ambulation tasks under conditions of no device or peroneal nerve stimulator. (29) Eleven subjects with MS, dorsiflexion weakness, and prior usage of an AFO were evaluated on the timed 25-foot walk component of the MS Functional Composite and the Floor, Carpet, Timed Up and Go, Obstacle, and Stair components of the Modified Emory Function Ambulation Profile. Performance on Stair and Obstacle components was enhanced in the stimulator condition versus no device (P=0.05 and P=0.09, respectively), and there were no significant differences between no device and stimulator conditions on other measures. The authors concluded that “the neuroprosthetic effect of the peroneal nerve stimulator is modest relative to no device in the performance of specific functional tasks of ambulation in MS gait. A longitudinal, controlled trial is needed to show effectiveness.”
The study by Stein et al. previously described also assessed the orthotic and therapeutic effects of NMES in 32 patients with progressive foot drop (31 MS and one familial spastic paresis).(26) With the stimulator on compared with off (orthotic effect), walking speed improved by 2% for a figure -8 test and 4% for a 10MWT. With the stimulator off (therapeutic effect), walking speed at 3 months had improved by 9% for a figure -8 test and 5% for a 10MWT when compared with baseline. The combined improvement in walking speed over the three months was 13% for the figure 8 (0.61 vs 0.53 m/s) and 13% for the 10MWT (0.88 vs 0.78 m/s, both respectively). The 20 subjects (63%) who returned for testing at eleven months did not show continued improvement when compared with 3-month test results, with a combined (orthotic and therapeutic) improvement of 13% on the figure 8 (0.62 vs 0.55 m/s) and 10% on the 10MWT (0.86 vs 0.78 m/s, both respectively) compared with baseline. The PCI was not significantly improved (0.73 vs 0.78 beats/min, respectively). Subjects with nonprogressive foot drop used the device for an average 85% of days, 9.2 h/d, and walked about 2 km/d.
Cauraugh et al. conducted a 2010 meta-analysis of seventeen studies on NMES and gait in children with cerebral palsy. (30) Fourteen of the studies used a pretest-post-test, within-subjects design. A total of 238 participants had NMES. Included were studies on acute NMES, functional NMES and therapeutic NMES (continuous subthreshold stimulation). Five of the studies examined functional NMES, and 1 of these studies examined percutaneous NMES. There were 3 outcome measures for impairment; range of motion, torque/movement, and strength/force. There were 6 different outcome measures for activity limitations; gross motor functions, gait parameters, hopping on one foot, six minute walk, Leg Ability Index, and Gillette gait index. Moderate effect sizes were found for impairment (0.616) and activity limitations (0.635). The systematic review is limited by a lack of blinding in the included studies and the heterogeneity of outcome measures. The review did not describe if any of the included studies used a commercially available device.
A 2012 report examined the acceptability and effectiveness of a commercially available foot drop stimulator in 21 children who had mild gait impairments and unilateral foot drop. (31) Three children did not experience an improvement in walking and did not complete the study. Gait analysis in the remaining 18 showed improved dorsiflexion when compared with baseline. There was no significant change in other gait parameters, including walking speed. The average daily use was 5.6 hours (range, 1.5-9.4) over the 3 months of the study, although the participants had been instructed to use the device for at least 6 hours per day. Eighteen children (86%) chose to keep using the device after the 3-month trial period. Data from this period were collected but not reported.
In 2013, Meilahn assessed the tolerability and efficacy of a commercially available neuroprosthesis in ten children (age, 7-12 years) with hemiparetic cerebral palsy who typically wore an AFO for correction of foot drop. (32) All of the children tolerated the fitting and wore the device for the first six weeks. The mean wear time was 8.4 hours per day in the first 3 weeks and 5.8 hours per day in the next 3 weeks. Seven children (70%) wore the device for the three month study period, with average use of 2.3 hours daily (range, 1.0-6.3 hours/day). Six children (60%) continued to use the neuroprosthesis after study completion. Gait analysis was performed, but quantitative results were not included in the report. Although it was reported that half of the subjects improved gait velocity, mean velocity was relatively unchanged with the neuroprosthesis.
Two recent within-subject studies have evaluated tolerability and efficacy of a commercially available neuroprosthesis in children with cerebral palsy. Both of the studies, which should be considered preliminary, show no improvement in walking speed with the device. In addition, daily use decreased over the course of one trial. Study in a larger number of subjects over a longer duration is needed to permit conclusions concerning the effect of the technology on health outcomes.
Ongoing and Unpublished Clinical Trials
A search of online site www.clinicaltrials.gov in December 2014 identified the following studies with a neuroprosthesis:
Also identified were a number of studies on functional NMES for treatment of patients with acute and chronic stroke conditions. These trials primarily focus on rehabilitation and strengthening.
Summary of Evidence
Evidence for neuromuscular stimulation to provide functional movement in patients with spinal cord injury is limited by the small number of subjects studied to date. For chronic post-stroke foot drop, 2 large randomized controlled trials and a crossover study of NMES versus a standard ankle-foot orthosis showed improved satisfaction with NMES but no significant difference between groups in objective measures of walking. A small randomized trial examining neuromuscular stimulation for foot drop in patients with multiple sclerosis showed a reduction in falls and improvement in satisfaction when compared with a program of exercise, but did not demonstrate a clinically significant benefit in walking speed. The literature on NMES in children with cerebral palsy includes a systematic review of small studies with within-subject designs; additional study in a larger number of subjects is needed. Due to insufficient evidence for some indications, and a lack of improvement for others, functional NMES remains investigational.
Practice Guidelines and Position Statements
In January 2009, the National Institute for Health and Clinical Excellence (NICE) published guidance stating that the current evidence on functional electrical stimulation (FES) for drop foot of neurologic origin appears adequate to support its use, provided that normal arrangements are in place for clinical governance, consent, and audit. (33) They noted that patient selection should involve a multidisciplinary team. NICE advises that further publication on efficacy of FES would be useful, specifically including patient-reported outcomes, such as quality of life and ADL, and these outcomes should be examined in different ethnic and socioeconomic groups.
U.S. Preventive Services Task Force Recommendations
Medicare National Coverage
In 2002, Medicare issued a national coverage policy recommending coverage for NMES for ambulation in spinal cord injury patients consistent with the Food and Drug Administration labeling for the Parastep device, effective April 1, 2003. (34) The Medicare decision memorandum indicates that Medicare considered the same data as those discussed here in their decision-making process. The decision memorandum notes that the available studies are flawed but concluded that the limited ambulation provided by the Parastep device supported its clinical effectiveness and thus its coverage eligibility. The inclusion and exclusion criteria outlined by Medicare are as follows:
Functional neuromuscular stimulation, transcutaneous stimulation of sequential muscles groups of ambulation with computer control, used for walking by spinal cord injured, entire system after completion of training program
Functional electrical stimulator, transcutaneous stimulation of nerve and/or muscle groups, any type, complete system, not otherwise specified
Type of Service
Physical Therapy, Durable Medical Equipment
Place of Service
Add to Therapy Section - New Policy
Replace Policy - Policy revised to focus on ambulation.
Replace Policy - Literature review update; added to Rationale/Source section; No change in policy statement.
Replace Policy - Policy updated; no change in policy statement.
Replace Policy - Policy reviewed with literature search; no new clinical trials found. Policy statement unchanged.
Codes updated - No other changes.
Update Scope and Disclaimer - No other changes.
Replace Policy - Policy updated with literature review; policy statement clarified to include: “ambulation in patients with spinal cord injury and post-stroke” as investigational. References added.
Replace Policy - Policy updated with literature search. Policy statements modified to add a second policy statement that use of these devices in post-stroke patients is considered investigational. References added.
Replace Policy - Policy extensively updated with literature search. Additional applications added to policy statement (hand and foot). Title updated to Functional neuromuscular electrical stimulation. References updated.
Replace Policy - Policy updated with literature review; references added and reordered. Policy statement remains unchanged.
Replace policy. Policy updated with literature review through December 2011; reference 25 added; policy statement unchanged.
Update Coding Section – ICD-10 codes are now effective 10/01/2014.
Replace policy. Policy updated with literature review through January 16, 2013; references 11-12 and 29-31 added; cerebral palsy added to investigational policy statement.
Update Related Policies. Change title for 7.01.69 to “Sacral Nerve Neuromodulation/Stimulation”.
Clarification note added. This policy does not apply to specialized exercise equipment, such as the RT 300 Exercycle, that is used in the rehabilitation setting under the supervision of a physical therapist or other rehab specialist. Please refer to medical policy 8.03.502.
Update Related Policies. Remove 1.01.19 as it was archived.
Annual Review. Policy updated with literature review January 7, 2014. References 20 and 21 added; others renumbered/removed. Policy statement unchanged. All codes removed from policy with the exception of HCPCS codes; these are the only code utilized for adjudication.
Update Related Policies. Change title to 1.01.17.
Annual Review. Policy updated with literature review through January 16, 2015; references 20 and 22 added; policy statement unchanged. Clarification notes in policy statements retained.
Disclaimer: This medical policy is a guide in evaluating the medical necessity of a particular service or treatment. The Company adopts policies after careful review of published peer-reviewed scientific literature, national guidelines and local standards of practice. Since medical technology is constantly changing, the Company reserves the right to review and update policies as appropriate. Member contracts differ in their benefits. Always consult the member benefit booklet or contact a member service representative to determine coverage for a specific medical service or supply. CPT codes, descriptions and materials are copyrighted by the American Medical Association (AMA).