Ankle-Foot Orthosis Selection to Facilitate Gait Recovery in Adults After Stroke: A Case Series : JPO: Journal of Prosthetics and Orthotics

Secondary Logo

Journal Logo

Case Report

Ankle-Foot Orthosis Selection to Facilitate Gait Recovery in Adults After Stroke

A Case Series

McCain, Karen J. PT, DPT, NCS; Smith, Patricia S. PT, PhD, NCS; Querry, Ross PT, PhD

Author Information
JPO Journal of Prosthetics and Orthotics: July 2012 - Volume 24 - Issue 3 - p 111-121
doi: 10.1097/JPO.0b013e31825f860d
  • Free

Gait dysfunction after stroke is common1 and is related to impairments such as weakness, altered reflexes, and impaired sensation.2–4 Gait after stroke is perhaps best distinguished by asymmetry with reduced stride and step length, wide base of support, and altered stance and swing phase periods.4–10 Weakness of the lower leg muscles contributes to the asymmetry by interfering with foot clearance, knee control, and propulsion.11,12

Most clinicians would agree that the goal of gait rehabilitation after stroke is to promote the most functional and adaptable, that is, normative and symmetrical, gait possible.10,13 Temporal muscle activity in the lower limbs during gait has been evaluated in persons after stroke and been found to be different from activity observed in healthy persons.14 Buurke and colleagues15 tracked changes in electromyographic (EMG) patterns during gait over a period of months after stroke in an effort to evaluate motor recovery. They determined that gait performance did improve, but without concomitant changes in timing of muscle activation in the paretic limb, suggesting improvements in gait performance were primarily compensatory in nature.15

Ankle-foot orthoses (AFOs) are often prescribed to improve gait after stroke. Specifically, they are used to improve mediolateral ankle stability, improve foot position, increase walking speed, and decrease energy expenditure.2,16,17 In addition, AFOs have been used to promote gait symmetry after stroke.11,18 Although a variety of designs are available, many significantly restrict range of motion, particularly plantarflexion, which interferes with the heel rocker and forefoot rocker, both of which are critical for typical gait.12,19 A 2002 review of the impact of AFOs on gait in adults with hemiplegia concluded that the devices may improve velocity, stride length, gait pattern, and walking efficiency, but the effects on symmetry were highly variable.3 The review identified only four studies that examined muscle activation with the devices. The authors determined that the findings were not adequate to draw conclusions about muscle activity in the AFOs.3 Since the time of the review, only a few additional studies have attempted to evaluate the effects of AFOs on muscle activity12,17,20 and symmetry11,21 in persons after stroke. In each of the studies, different orthosis designs were evaluated, making generalizations about the findings challenging.

Despite the widespread use of AFOs, little is reported about the impact of different orthosis designs on the motor recovery of persons with stroke. In fact, there is often reluctance to use AFOs for fear of interfering with functional recovery by promoting dependence on the devices3 or of producing unwanted activity in the plantarflexors.16,18 Current understanding of neuroplasticity after neurologic injury22,23 would suggest that orthosis design has the potential to significantly impact motor recovery of gait.

The purpose of this case report was to describe the muscle activation patterns (EMG) and gait characteristics of three persons recovering from stroke who wore an AFO designed to facilitate typical gait mechanics and who received early standardized task-specific training (ESTT).24 The AFO that was used was custom-fabricated thermoplastic with a metal double action joint (DAJ) and metal upright. The joints (medial and lateral) were composed of two channels, one anterior and one posterior to the metal upright. The configuration used in this report consisted of a stiff spring in each of the posterior channels and a pin in each of the anterior compartments. The springs served a dual purpose, providing a dorsi assist during swing phase and resistance to plantarflexion during loading response. The pins in the front channels allowed controlled tibial advancement during stance (into dorsiflexion) (Fig.1).

Figure 1:
Double adjustable joint ankle-foot orthosis.



Participants (n = 3; one woman, two men) were recruited from a convenience sample of persons who received ESTT24 at Baylor Institute for Rehabilitation, Dallas, TX, USA. Table 1 outlines the characteristics of the individuals, including age, lesion details, comorbidities, Stroke Rehabilitation Assessment of Movement (STREAM) scores, duration of AFO use, length of rehabilitation stay, and time of testing. All participants had spasticity in the plantarflexors to varying degrees and each had functional passive ankle range of motion (at least 5° of dorsiflexion and 20° of plantarflexion). Participants wore a DAJ AFO when they initiated the ESTT program in the acute phase of their recovery and during subsequent overground training. They were all discharged from inpatient rehabilitation (IP) with a custom DAJ AFO fabricated by the same orthotist and all used the device while walking at home and in the community. Subject 1 wore the brace for 3 months after stroke onset, whereas subjects 2 and 3 used the orthosis for 2 months. Each person discontinued use of the AFO during outpatient therapy under the direction of a physical therapist. The study was approved by the institutional review board, and each person signed a consent form before testing.

Table 1:
Patient characteristics


Each participant was admitted to IP after first-time stroke and was trained using the ESTT protocol, which has been described in detail elsewhere.24 In summary, task-specific gait training using a treadmill with partial body weight support was initiated before overground gait training. Thirty minutes of each individual’s daily scheduled 3-hr therapy program was allocated to gait training implemented by a physical therapist trained in the protocol. During all gait training, a DAJ AFO was used. The AFOs were all initially set to allow 3° to 5° of dorsiflexion and no more than 8° to 10° of plantarflexion.


All participants completed a battery of tests. The initial testing was completed while the persons were in IP. The final testing was done after IP in the Crowley Research and Rehabilitation laboratory (LAB) in the School of Health Professions at UT Southwestern Medical Center, Dallas, TX, USA. Participants 1 and 2 were tested on one date in the LAB. Participant 3 was tested on two dates in the LAB. Times for LAB testing ranged from 80 to 919 days after stroke.


The 6-Minute Walk Test (6MWT) has been used successfully to measure gait endurance in subjects with stroke25 and was completed at the time of discharge from IP and during LAB testing. The participants walked as far as possible (self-selected walking velocity [SSWV]) indoors on level surfaces in a 6-minute time period. No verbal encouragement was provided, and distance was recorded with a measuring wheel. Each person used the AFO and a single-tip cane (STC) for the 6MWT at the IP testing. Participant 3 used the AFO (no cane) for one trial during the initial LAB assessment. All participants completed the final LAB 6MWT with no AFO or cane.


The STREAM test of motor recovery has been deemed valid and reliable and was used to assess lower limb motor control.26 Each section of the STREAM can be scored individually, with a score of 100 representing no impairment. The test was administered at time of admission to IP, at discharge from IP, and during LAB testing (Table 1).


The GAITRite™ computerized gait analysis system (CIR Systems, Havertown, PA, USA) was used to measure temporal and spatial gait parameters. This system has been demonstrated to be reliable and valid in this population.27,28 The system consists of a 16-ft pressure-sensitive mat with an area of activation measuring 24 in (61 cm) wide and 192 in (487 cm) long with a total of 18,432 sensors. Data were sampled at 1000 Hz. Participants walked over the walkway a total of four times per walking condition, first at an SSWV and then at a fast walking velocity (FWV). Participants 1 and 2 completed both test conditions without an AFO or STC. Participant 3 used his AFO during one of two testing conditions at the initial LAB testing but not during the second test or during the final LAB testing. Depending on individual step length, after editing, there was an average of 28.3 ± 5.1 steps for each subject for the SSWV condition and an average of 25.3 ± 6.1 steps for the FWV condition. Data collected included velocity, cadence, step length, percentage swing time, and percentage stance time.


With the use of data obtained from the GAITRite testing, symmetry was calculated using the ratio of paretic to nonparetic limb for step length, swing time, and stance time.


The EMG activity of the bilateral vastus lateralis (VL), anterior tibialis (AT), and medial gastrocnemius (GAS) muscles were recorded using MyoPac/DataPac fiber-optic hardware (Run Technologies, Pasadena, CA, USA). Electromyographic signal was recorded at a sampling rate of 1 KHz using bipolar paired Ag-Cl surface electrodes (Blue Sensor, Ambu, Denmark). Electrode diameter was 10 mm, with an interelectrode distance of 20 mm (center to center). Electrode sites were prepared by shaving the skin and mildly abrading the skin with prep paper and alcohol, with skin impedance of the electrodes maintained below 5 Kohm resistance. A reference ground electrode was placed over the patellar surface. Pressure-sensitive foot switches were used to determine stance and swing phases of gait. In addition, a sync pulse transmitted from the GAITRite system at the initial contact of each walking trial was recorded concurrently with the EMG data. Four individual walks were recorded for each of the SSWV and FWV trials.


Electromyographic signals were full-wave rectified and smoothed (root-mean squared, 30 milliseconds), and linear envelopes were calculated using a FIR Bandpass filter with a low-pass cutoff frequency of 10 Hz and a high-pass cutoff frequency of 450 Hz. The extraction of individual gait cycles was done automatically using the threshold method. The threshold level was set to 1 V of the maximum 5-V signal from the electronic pressure-sensitive foot switches (initial contact heel sensor voltage was 3 V). Confirmation of the accurate phases of the gait cycle was evaluated with synchronized GAITRite footfall time data. Outliers (the steps with incorrect foot contacts or obvious artifacts) were identified by visual analysis of both systems and were excluded from further analysis. After the outliers were removed, a minimum of 20 gait cycles were available for each subject and trial (four individual walks at each velocity). Events were defined as the stance phase (foot contact), and the interevent interval was defined as the swing phase of gait. All events for a given walking velocity were then signal averaged and normalized to 100% of the gait cycle. For graphical comparison of individual muscle activity and velocities, the highest amplitude of EMG signal (FWV in all cases) was set as 100% of scale for each side independently.


Tables 2 and 3 detail the results of the gait assessments. Symmetry was calculated as the ratio of paretic to nonparetic limb for step length, swing time, and stance time. Values were inverted, when necessary, to be greater than 1 for clarity of analysis.29Figures 2, 3, and 4 depict the muscle activity of the GAS, AT, and VL for all participants at the time of final LAB testing without AFOs or devices. The SSWV and FWV for the affected limb (left) are presented with the EMG profile for the right limb (nonhemiparetic limb) during the FWV trial as a reference. The FWV trial was chosen as the reference because the fast velocity created the greatest demand for motor activity. Figures 5 (SSWV) and 6 (FWV) depict the activity of subject 3 at the initial LAB testing with and without the AFO.

Table 2:
Gait parameters, subjects 1 and 2
Table 3:
Gait parameters, subject 3
Figure 2:
Subject 1 signal averaged electromyographic muscle activity during SSWV and FWV normalized to 100% of gait cycle (time 0, initial contact). A–C, left hemiparetic SSWV. D–F, left hemiparetic FWV. G–I, right nonhemiparetic FWV for comparison. Scaling for left hemiparetic muscle activity uses signal amplitude of FWV for 100% scale. Time since initial injury is 234 days. SSWV, self-selected walking velocity; FWV, fast walking velocity.
Figure 3:
Subject 2 signal averaged electromyographic muscle activity during SSWV and FWV normalized to 100% of gait cycle (time 0, initial contact). A–C, left hemiparetic SSWV. D–F, left hemiparetic FWV. G–I, right nonhemiparetic FWV for comparison. Scaling for left hemiparetic muscle activity uses signal amplitude of FWV for 100% scale. Time since initial injury is 919 days. SSWV, self-selected walking velocity; FWV, fast walking velocity.
Figure 4:
Subject 3 signal averaged electromyographic muscle activity during SSWV and FWV normalized to 100% of gait cycle (time 0, initial contact). A–C, left hemiparetic SSWV. D–F, left hemiparetic FWV. G–I, right nonhemiparetic FWV for comparison. Scaling for left hemiparetic muscle activity uses signal amplitude of FWV for 100% scale. Time since initial injury is 543. SSWV, self-selected walking velocity; FWV, fast walking velocity.
Figure 5:
Subject 3 signal averaged electromyographic muscle activity during SSWV normalized to 100% of gait cycle (time 0, initial contact). B–D, left hemiparetic SSWV without the use of left AFO. F–H, left hemiparetic SSWV with the use of left AFO. A and E, alteration of right nonhemiparetic vastus lateralis without AFO that was significantly attenuated with the use of left AFO during SSWV walking. I and J, right nonhemiparetic FWV for comparison. Scaling for left hemiparetic muscle activity uses signal amplitude of FWV for 100% scale. Time since initial injury is 80 days (48 days after inpatient discharge). SSWV, self-selected walking velocity; AFO, ankle-foot orthosis; FWV, fast walking velocity.
Figure 6:
Subject 3 signal averaged electromyographic muscle activity during FWV normalized to 100% of gait cycle (time 0, initial contact). B–D, left hemiparetic FWV without the use of left AFO. F–H, left hemiparetic FWV with the use of left AFO. A and E, alteration of right nonhemiparetic vastus lateralis without AFO that was significantly attenuated with the use of left AFO during FWV walking. I and J, right nonhemiparetic FWV for comparison. Scaling for left hemiparetic muscle activity uses signal amplitude of FWV for 100% scale. Time since initial injury is 80 days (48 days after inpatient discharge). FWV, fast walking velocity; AFO, ankle-foot orthosis.


The 6MWT was used as a measure of endurance and was used to calculate gait velocity. Baseline testing was done at discharge from IP and repeated during LAB testing. Values increased between initial and final testing points by 147% for subject 1, 429% for subject 2, and 108% for subject 3.


All participants exhibited motor impairment at the time of initial testing, as demonstrated in the STREAM test results, and all showed improvement at subsequent testing (Table 1).


General symmetry improved from the time of IP testing to the time of final LAB testing. At IP testing, step length symmetry ranged from 1.09 to 2.14. Swing symmetry ranged from 1.20 to 2.45, and stance symmetry, from 1.05 to 1.58. At the time of final LAB testing, step length symmetry ranged from 1.0 to 1.08; swing symmetry, from 1.0 to 1.05; and stance symmetry, from 1.0 to 1.03.


Figures 2, 3, and 4 represent the EMG profiles of each participant at the time of LAB testing without the AFO or STC. The graphs depict a consistent picture of general temporal symmetry across testing conditions. Timing of muscle activity across the gait cycle was consistent with expected patterns, that is, activation of the AT at initial contact, VL activation early in the gait cycle, and peak activation of GAS late in stance.30,31 In addition, amplitude increases were seen appropriately in relation to increases in velocity. Likewise, similar peak muscle activation was seen in the nonparetic limb across muscle groups. Figures 5 and 6 depict SSWV and FWV with and without the AFO early in the recovery for subject 3 (80 days after stroke). At this time, he could walk without the AFO but was routinely using it in the community. For this reason, the reference (right) EMG is depicted with the participant wearing the AFO. As with later testing, general firing patterns fit expected profiles in all muscle groups. Interestingly, during both SSWV and FWV conditions, the VL activity on the nonparetic limb was unusually high without the AFO. With the AFO, there was a dampening of this atypical activity. In addition, temporally appropriate muscle activity of the GAS and AT was evident with and without the AFO. In fact, GAS activity was greatest in the FWV condition with the AFO.


The purpose of this case series was to report the muscle activation patterns and gait characteristics of three adults who used a DAJ AFO during early standardized gait training after stroke. The resulting gait was remarkably fast and symmetrical, with muscle activation patterns characteristic of normative gait.

There is some evidence in the literature that AFO design can impact gait outcomes. Miyazaki et al.32 suggested that brace design could significantly impact gait across the gait cycle and that AFOs should not be considered simply as “mechanical substitutes” for insufficient muscle forces. They theorized that the active muscle forces of persons with hemiparesis are not determined solely by the underlying central nervous system lesion and that they can change substantially with the peripheral mechanical input afforded by an AFO. Few studies have attempted to consider the long-term implications of orthosis use after stroke.3 Historically, there has been a general distrust of AFOs for fear that they would negatively impact muscle activity after stroke. Indeed, some studies have demonstrated decreased muscle activity in AFOs, which was likely related to the rigidity of the devices.3 Mulroy and colleagues17 did not find decreased AT activity in the AFO, which was similar in design to the DAJ AFO. Lairamore et al.20 suggested that the DAJ design that allows plantarflexion at loading results in activation of the AT. In the present report, subject 3 demonstrated appropriate AT activity while wearing the AFO, and all participants had appropriate AT activity without the AFO at final testing. This suggests that the AFO did not decrease muscle activity but facilitated recovery of functional strength by allowing eccentric activation at loading.

There is evidence that prolonged activity of the quadriceps and hamstrings during stance on the paretic and nonparetic limb is common with and without AFOs.14,18 Den Otter and colleagues14 suggested that this may represent a strategy to supply additional support in the presence of weak calf muscles. Interestingly, subject 3 demonstrated a similar tendency on the nonparetic limb without the AFO at both FWV and SSWV. However, with the AFO, the increased activity in the contralateral VL was absent, suggesting a more stable condition.

Previous studies have shown variable effects on calf muscle firing with AFOs. None showed premature firing of the plantarflexors,3 whereas two suggested that the biomechanical features of the AFOs have considerable effect on the power generation of calf muscles.32,33 Using the AFO that was similar to our DAJ AFO, Mulroy et al.17 concluded that it did provide some control for the advancing tibia in the presence of gastrocnemius weakness. In the present report, the DAJ AFO was chosen, in part, to provide a supplement to plantarflexion.32 In light of the fact that recovery of calf strength is typically limited to about 37% after stroke34 and considering its role in knee stability in normative gait,31 the ability of the AFO to provide the opportunity for calf strengthening becomes key. The DAJ AFO did so by allowing tibial advancement beyond 90° with support from the anterior pin, theoretically stimulating eccentric firing of the posterior calf muscles. The EMG data from the present report suggest that this facilitation of typical joint kinematics may have contributed to functional recovery of the posterior compartment strength. When the calf functions sufficiently to control tibial advancement, excessive quadriceps activity is not required for knee stability.30,31

During GAITRite testing with a group of 161 persons with stroke, Patterson and colleagues29 found an average comfortable velocity of 0.72 ± 0.33 m/s, whereas SSWV in the current report ranged from 0.84 to 1.19 m/s. Gait velocity (6MWT) for persons at least 6 months after stroke has been documented to be between 0.73 ± 0.36 m/s35 and 0.76 ± 0.29 m/s,36 whereas speeds of 1.83 ± 0.19 m/s35 have been reported for healthy persons walking the same distance. The velocity of the persons in this report on the 6MWT at final LAB testing was between 1.0 and 1.48 m/s.

Esquenazi et al.11 evaluated the effects of AFOs on gait symmetry in persons who could walk without their AFOs and found improved symmetry with the devices. Hesse et al.16 found no improvements in gait symmetry using a Valens caliper in a group of persons 1.5 to 16 months after stroke, whereas Pohl and Mehrholz21 found modest improvements in symmetry with the use of a custom-designed AFO. At IP discharge, these three subjects demonstrated significant gait asymmetry. Using the same symmetry formula, Patterson et al.29 evaluated symmetry in 161 community ambulators after stroke and compared the findings with those of 81 healthy persons. Persons with stroke were able to walk without an assistive device, and 12 used AFOs. In the nonstroke group, mean step length ratio was 1.03 (SD, 0.02), swing time ratio was 1.02 (SD, 0.02), and stance time ratio was 1.02 (SD, 0.02). In contrast, the persons with stroke had a mean step length ratio of 1.13 (SD, 0.20), swing time ratio of 1.24 (SD, 0.34), and stance time ratio of 1.09 (SD, 0.10). Lewek and Randall13 also evaluated symmetry in a group of 26 individuals in the chronic phase of recovery and found greater asymmetry than Patterson et al.29 did. In the current report, step length symmetry ratios obtained at the time of final testing ranged from 1.0 to 1.08 across conditions. Swing symmetry ranged from 1.0 to 1.05, and stance symmetry, from 1.0 to 1.03. These values are surprisingly close to the values of the healthy participants reported by Patterson et al.29 This may represent the effects of continued practice of optimum gait kinematics afforded by extended use of the DAJ AFO.

The favorable gait outcomes for the three subjects may reflect a symbiotic relationship between the early standardized intervention (ESTT) and the DAJ AFO. There is growing evidence in the literature that early, task-specific training is critical for neuroplastic changes.22,24 Equally compelling is the evidence suggesting that changing gait after it is well established in persons after stroke is significantly less effective.37–39 Kautz and colleagues40 suggested that it is critical when designing therapeutic interventions to be able to determine if a measured functional effect is the result of a compensatory strategy or of true restitution of preinjury motor function. This combination of ESTT and DAJ AFO was designed to promote preinjury motor function. In these three individuals, EMG findings, coupled with gait velocity and symmetry data, suggest restitution of function rather than compensation.


Application of AFOs to address gait dysfunction after stroke is common. However, we would assert that a comprehensive approach to gait training necessitates awareness of the impact of AFO design on motor recovery. Given the fact that most persons regain the ability to walk within the first 5 weeks after stroke,41 early training efforts become critical with regard to long-term outcome. The persons in this case report were all trained exclusively with a DAJ AFO that was designed to retrain typical gait, including typical muscle firing patterns. Each demonstrated a very fast, symmetrical, and typical motor pattern (EMG) at the time of final testing, and each was able to do so without an assistive device or AFO. Additional studies are needed to evaluate the contribution of AFOs on restitution of lower limb motor function after stroke.


This is a case series with a small number of subjects, all with right-sided lesions, which does not allow for the findings to be generalized to all persons with stroke. Interestingly, the sensory deficits associated with right-sided lesions may have predicted worse outcomes than those experienced by these persons.


1. Friedman PJ. Gait recovery after hemiplegic stroke. Int Disabil Stud 1990; 12: 119–122.
2. Abe H, Michimata A, Sugawara K, et al.. Improving gait stability in stroke hemiplegic patients with a plastic ankle-foot orthosis. Tohoku J Exp Med 2009; 218: 193–199.
3. Leung J, Moseley A. Impact of ankle-foot orthoses on gait and leg muscle activity in adults with hemiplegia: systematic literature review. Physiotherapy 2002; 89: 39–55.
4. Woolley SM. Characteristics of gait in hemiplegia. Top Stroke Rehabil 2001; 7: 1–18.
5. Chen G, Patten C, Kothari DH, Zajac FE. Gait differences between individuals with post-stroke hemiparesis and non-disabled controls at matched speeds. Gait Posture 2005; 22: 51–56.
6. De Quervain IA, Simon SR, Leurgans S, et al.. Gait pattern in the early recovery period after stroke. J Bone Joint Surg Am 1996; 78: 1506–1514.
7. Goldie PA, Matyas TA, Evans OM. Gait after stroke: initial deficit and changes in temporal patterns for each gait phase. Arch Phys Med Rehabil 2001; 82: 1057–1065.
8. Knutsson E, Richards C. Different types of disturbed motor control in gait of hemiparetic patients. Brain 1979; 102: 405–430.
9. Roth EJ, Merbitz C, Mroczek K, et al.. Hemiplegic gait. Relationships between walking speed and other temporal parameters. Am J Phys Med Rehabil 1997; 76: 128–133.
10. Wall JC, Turnbull GI. Gait asymmetries in residual hemiplegia. Arch Phys Med Rehabil 1986; 67: 550–553.
11. Esquenazi A, Ofluoglu D, Hirai B, Kim S. The effect of an ankle-foot orthosis on temporal spatial parameters and asymmetry of gait in hemiparetic patients. PM R 2009; 1: 1014–1018.
12. Ohata K, Yasui T, Tsuboyama T, Ichihashi N. Effects of an ankle-foot orthosis with oil damper on muscle activity in adults after stroke. Gait Posture 2011; 33: 102–107.
13. Lewek MD, Randall EP. Reliability of spatiotemporal asymmetry during overground walking for individuals following chronic stroke. J Neurol Phys Ther 2011; 35: 116–121.
14. Den Otter AR, Geurts AC, Mulder T, Duysens J. Abnormalities in the temporal patterning of lower extremity muscle activity in hemiparetic gait. Gait Posture 2007; 25: 342–352.
15. Buurke JH, Nene AV, Kwakkel G, et al.. Recovery of gait after stroke: what changes? Neurorehabil Neural Repair 2008; 22: 676–683.
16. Hesse S, Luecke D, Jahnke MT, Mauritz KH. Gait function in spastic hemiparetic patients walking barefoot, with firm shoes, and with ankle-foot orthosis. Int J Rehabil Res 1996; 19: 133–141.
17. Mulroy SJ, Eberly VJ, Gronely JK, et al.. Effect of AFO design on walking after stroke: impact of ankle plantar flexion contracture. Prosthet Orthot Int 2010; 34: 277–292.
18. Hesse S, Werner C, Matthias K, et al.. Non–velocity-related effects of a rigid double-stopped ankle-foot orthosis on gait and lower limb muscle activity of hemiparetic subjects with an equinovarus deformity. Stroke 1999; 30: 1855–1861.
19. Perry J. Basic Functions Gait Analysis: Normal and Pathological Function. 2nd Ed. Thorofare, NJ: SLACK, Inc; 2010: 19–47.
20. Lairamore C, Garrison MK, Bandy W, Zabel R. Comparison of tibialis anterior muscle electromyography, ankle angle, and velocity when individuals post stroke walk with different orthoses. Prosthet Orthot Int 2011; 35: 402–410.
21. Pohl M, Mehrholz J. Immediate effects of an individually designed functional ankle-foot orthosis on stance and gait in hemiparetic patients. Clin Rehabil 2006; 20: 324–330.
22. Kleim JA, Jones TA. Principles of experience-dependent neural plasticity: implications for rehabilitation after brain damage. J Speech Lang Hear Res 2008; 51: S225–S239.
23. Hornby TG, Straube DS, Kinnaird CR, et al.. Importance of specificity, amount, and intensity of locomotor training to improve ambulatory function in patients poststroke. Top Stroke Rehabil 2011; 18: 293–307.
24. McCain K, Smith P, Pollo F, et al.. Excellent outcomes for adults who experienced early standardized treadmill training during acute phase of recovery from stroke: a case series. Top Stroke Rehabil 2011; 18: 428–436.
25. Duncan P, Richards L, Wallace D, et al.. A randomized, controlled pilot study of a home-based exercise program for individuals with mild and moderate stroke. Stroke 1998; 29: 2055–2060.
26. Daley K, Mayo N, Wood-Dauphinee S. Reliability of scores on the Stroke Rehabilitation Assessment of Movement (STREAM) measure. Phys Ther 1999; 79: 8–19; quiz 20-13.
27. Bilney B, Morris M, Webster K. Concurrent related validity of the GAITRite walkway system for quantification of the spatial and temporal parameters of gait. Gait Posture 2003; 17: 68–74.
28. Rao N, Chaudhuri G, Hasso D, et al.. Gait assessment during the initial fitting of an ankle foot orthosis in individuals with stroke. Disabil Rehabil Assist Technol 2008; 3: 201–207.
29. Patterson KK, Gage WH, Brooks D, et al.. Evaluation of gait symmetry after stroke: a comparison of current methods and recommendations for standardization. Gait Posture 2010; 31: 241–246.
30. Perry J, Burnfield JM. Knee Gait Deviations Gait Analysis: Normal and Pathological Function. Second ed. Pomona, California: Slack Incorporated; 2010: 221–222.
31. Perry J, Burnfield JM. Ankle-Foot Complex Gait Analysis: Normal and Pathological Function. Sonoma, CA: Slack, Incorporated; 2010: 51–82.
32. Miyazaki S, Yamamoto S, Kubota T. Effect of ankle-foot orthosis on active ankle moment in patients with hemiparesis. Med Biol Eng Comput 1997; 35: 381–385.
33. Yamamoto S, Ebina M, Kubo S, et al.. Quantification of the effect of dorsi-/plantar flexibility of ankle-foot orthoses on hemiplegic gait: a preliminary report. J Prosthet Orthot 1993; 3: 88–94.
34. Patten C, Lexell J, Brown HE. Weakness and strength training in persons with poststroke hemiplegia: rationale, method, and efficacy. J Rehabil Res Dev 2004; 41: 293–312.
35. Dean CM, Richards CL, Malouin F. Walking speed over 10 metres overestimates locomotor capacity after stroke. Clin Rehabil 2001; 15: 415–421.
36. Sibley KM, Tang A, Patterson KK, et al.. Changes in spatiotemporal gait variables over time during a test of functional capacity after stroke. J Neuroeng Rehabil 2009; 6: 2n7.
37. Griffin M, Olney S, McBride I. Role of symmetry in gait performance of stroke subjects with hemiplegia. Gait Posture 1995; 3: 132–142.
38. Patterson SL, Rodgers MM, Macko RF, Forrester LW. Effect of treadmill exercise training on spatial and temporal gait parameters in subjects with chronic stroke: a preliminary report. J Rehabil Res Dev 2008; 45: 221–228.
39. Buzzelli S, Di Francesco L, Villani S, Giaquinto S. Gait in stabilized hemiplegic outpatients in rehabilitation. Disabil Rehabil 2003; 25: 964–967.
40. Kautz SA, Duncan PW, Perera S, et al.. Coordination of hemiparetic locomotion after stroke rehabilitation. Neurorehabil Neural Repair 2005; 19: 250–258.
41. Baer G, Smith M. The recovery of walking ability and subclassification of stroke. Physiother Res Int 2001; 6: 135–144.

gait; stroke; ankle-foot orthosis (AFO); EMG

© 2012 by the American Academy of Orthotists and Prosthetists.