Effect of Botulinum Toxin Injection in the Rectus Femoris on Stiff-Knee Gait in People With Stroke: A Prospective Observational Study

Article Outline

• Abstract
• Methods
  • Study Population
  • BTX-A Treatment
  • Clinical Examination
  • Instrumented Gait Analysis
  • Protocol
  • Statistical Analysis
• Results
• Discussion
• Conclusions
• References
• Copyright

Abstract 

Stoquart GG, Detrembleur C, Palumbo S, Deltombe T, Lejeune TM. Effect of botulinum toxin injection in the rectus femoris on stiff-knee gait in people with stroke: a prospective observational study.

Objective

To study the effect of botulinum toxin type A (BTX-A) injection in the rectus femoris on the decreased knee flexion during the swing phase of gait (stiff-knee gait) in people with stroke.

Design

Intervention study (before-after trial) with an observational design.

Setting

Outpatient rehabilitation clinic and gait laboratory.

Participants

Nineteen chronic hemiparetic adults presenting with stiff-knee gait.

Intervention

Injection of 200U of BTX-A (Botox) into the rectus femoris.

Main Outcome Measures

Before and 2 months after BTX-A rectus femoris injection: Stroke Impairment Assessment Set (SIAS), Duncan-Ely test, and an instrumented gait analysis.

Results

Median SIAS score improved from 53 (range, 36−65) to 57 (range, 42−70) (signed-rank test, P=.005) and the Duncan-Ely score from 3 (range, 1−3) to 1 (range, 0−3) (P<.001). In gait analysis, mean (± standard deviation) maximum knee flexion improved from 26°±13° to 31°±14° during the swing phase (paired t test, P<.001), knee flexion speed at toe-off improved from 82°±63° to 112°±75°/s (P=.009), and knee negative joint power (eccentric muscular contraction) improved from −.27±.23 to −.37±.26W/kg (P<.001). The 4 patients who almost did not flex the knee (<10°) before the BTX-A rectus femoris injection did not improve after the injection. The other 14 patients who flexed the knee more than 10° before the BTX-A rectus femoris injection decreased the walking energy cost from 5.4±1.6 to 4.6±1.3J·kg⁻¹·m⁻¹ (P=.006).

Conclusions

BTX-A rectus femoris injection may be beneficial in patients with a stiff-knee gait after stroke, particularly in patients with some knee flexion (>10°).

Key Words: Botulinum toxin type A, Gait, Muscle spasticity, Rectus femoris, Rehabilitation, Stroke

STIFF-KNEE GAIT IS characterized by a lack of knee flexion during the swing phase of the gait cycle. This gait frequently affects hemiparetic and spastic people with stroke, compromising foot clearance. Lack of knee flexion leads to toe dragging, which interferes with gait stability and increases the risk of falls. The compensatory movements performed to clear the foot during the swing phase, such as ipsilateral hip circumduction or contralateral vaulting, can result in increased vertical displacement of the center of body mass and greater energy expenditure.

The physiopathology of this stiff-knee gait is only partly understood, and several hypotheses are mentioned in the literature. The role of overactivity of the rectus femoris is often cited, but other possible mechanisms include hip flexor weakness and overactivity of the ankle plantarflexors at terminal stance.¹ Using dynamic simulation, Riley and Kerrigan² showed that knee motion in the swing phase was more sensitive to knee moment than to hip moment, suggesting that rectus femoris overactivity should decrease knee flexion. Sung and Bang³ confirmed this implication by measuring an increase in knee flexion after rectus femoris motor branch block (MBB) with phenol. Goldberg et al⁴ showed that the iliopsoas and gastrocnemii had the potential to increase the speed of knee flexion at the end of the stance phase, which was considered as a determinant of stiff-knee gait.⁵ The vasti, rectus femoris, and soleus were likely to decrease this speed, and the rectus femoris had the largest potential to decrease knee flexion. Albert et al⁶ showed that combined MBB of both vastus intermedius and vastus lateralis nerves could decrease quadriceps femoris spasticity and increase knee flexion slightly during walking.

Recognition of the involvement of the rectus femoris in stiff-knee gait in children with cerebral palsy has led to effective treatments targeted at this muscle: surgical transfer, release, or botulinum toxin injection.⁷ In contrast, in adults with stroke, treatments remain limited and poorly investigated.

The present study was designed to study the effect of botulinum toxin type A (BTX-A) injection in the rectus femoris on the stiff-knee gait of adults with stroke. It aimed to determine clinical relevance of BTX-A treatment on stiff-knee gait and to provide more information on the physiopathology of this condition.

Back to Article Outline

Methods 

Study Population 

Nineteen people with stroke with chronic hemiparesis and stiff-knee gait were enrolled in the present study between November 2003 and December 2005 (age, 53±15y; time since stroke, 52±56mo; height, 170±11cm; weight, 76±18kg). Anthropometric and historical data are presented in table 1. Inclusion was based on clinical examination and observation of gait: lack of knee flexion during swing and the ability to walk independently, without any assistive device, on a treadmill for sufficient time to complete a metabolic analysis. Patients were included at least 6 months after stroke, when the spontaneous recovery had ended and when patients were considered chronic. The exclusion criteria were the presence of significant impairments in other joints that restrained walking, such as an equinovarus foot, and the presence of other significant neurologic conditions or knee contracture. In the acute phase, all patients had been treated in a stroke unit and, thereafter, followed an intensive rehabilitation program. Patients were still following regular physical therapy (30min per session, 3−5 times/wk) aimed at maintaining functional status. This treatment was provided in the community and was maintained unchanged throughout the study. Patients’ drug therapies remained unchanged throughout the study.

Table 1. Anthropometric Data

Abbreviation: SD, standard deviation.

After giving their informed consent, all patients participated freely in the study, which was approved by the ethics board of our medical school.

BTX-A Treatment 

BTX-A (Botox, 200U in 2mL physiologic solution) was injected through 3 punctures in the rectus femoris (at the junction between the proximal and the middle third, at the midpoint, and at the junction between the middle and the distal third). One proximal and 1 distal injection were performed at each site (6 injections of 33U).

Clinical Examination 

Neurologic impairments (upper- and lower-limb motor function, muscle tone, sensitivity, range of motion, deep tendon reflexes) were assessed using the Stroke Impairment Assessment Set (SIAS),⁸ validated by Rasch analysis.⁹ The Duncan-Ely test¹⁰ was also performed to specifically evaluate rectus femoris spasticity. In 1 patient (patient 16), SIAS data were missing from the clinical examination performed before the treatment. They were the only missing data for this patient, and all other variables were acquired and analyzed.

Instrumented Gait Analysis 

Gait was assessed by 3-dimensional analysis, including synchronous kinematic, kinetic, electromyographic, and energetic measurements. All data were acquired simultaneously on a force-measuring treadmill.^{11, a} For each patient, 10 consecutive walking cycles were recorded to analyze each variable.

Segmental kinematics was measured with the Elite system.^b Six infrared cameras measured, at 100Hz, the coordinates in the 3 spatial planes of 20 reflective markers positioned on specific anatomic landmarks.¹² These measurements allowed computation of the angular displacement and angular speed of the knee in the sagittal plane during the walking cycle using Euler angles.¹²

Ground reaction force was recorded by 4 strain gauges located under each corner of the treadmill.¹¹ Positions of the center of pressure and ground reaction force under each foot were determined by following the algorithm described by Davis et al¹³ and Raison et al.¹⁴ By using an inverse dynamics approach, ground reaction force, kinematic, and anthropometric data enabled computation of the net joint moment of the knee in the sagittal plane. The joint power of the knee was calculated as the product of the angular speed and the net joint moment.

The electric (electromyographic) activity of the vastus lateralis, vastus medialis, and biceps femoris was recorded by a telemetry electromyography system (Telemg)^b with surface electrodes.^c The electric activity of the rectus femoris was recorded with fine-wire electrodes. The signal was digitized at 1000Hz, full-wave rectified, and filtered (bandwidth, 25−300Hz). The onset and cessation of muscle activity were determined as described by Van Boxtel et al.¹⁵

Kinematic, kinetic, and electromyographic data were normalized to 100% of the time for the full walking cycle, with 0% corresponding to the initial contact of the foot of the affected side. Foot contact periods were determined from the forward direction velocity curve of both fifth metatarsal markers, computed from kinematic data.¹⁶ As shown in figure 1, the gait cycle starts with the stance phase of the paretic lower limb. Step frequency was derived from these foot contact periods.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 1. 

  The typical trace of 1 patient walking at .55m/s. (A) The knee angular displacement (d), (B) knee angular speed (s), (C) net knee joint moment (m), (D) knee joint power (p), and (E) vastus medialis (VM), vastus lateralis (VL), biceps femoris (BF), and rectus femoris (RF) activation time are presented as a function of gait cycle time (in percent). Grey areas represent values (mean ± SD) of healthy subjects walking at .55m/s. Dashed grey lines and solid black lines represent, respectively, pre– and post–BTX-A data of the patient. The vertical dashed line symbolizes the transition between the stance and swing phases. Arrows show key data selected for each variable in all patients on every assessment. Abbreviation: SL, slope of the linear regression fitted to the knee angular speed between s1 and s3.

The metabolic cost of walking was determined by each patient’s oxygen consumption (V̇o₂) and carbon dioxide production (V̇co₂) measured throughout the treadmill test. Values were automatically converted by the software^d to standard temperature, pressure, and dry V̇o₂. The respiratory exchange ratio (RER), computed as the ratio between V̇co₂ and V̇o₂, always remained less than 1. Each energy measurement started with a rest period while the patient was standing on the treadmill. Thereafter, patients walked until a steady state was reached and maintained for at least 2 minutes. The joules of energy expended per liter of oxygen consumed were computed depending on the RER, according to the Lusk equation.¹⁷ The energy expended above the resting value (standing subtracted from walking consumption) was divided by the walking speed to obtain the net energy cost of walking (in J·kg⁻¹·m⁻¹).

Protocol 

The protocol started with an initial clinical examination and gait analysis (pre–BTX-A). BTX-A was then injected, and second clinical and gait evaluations were performed 2 months later to assess the effects of the BTX-A injection (post–BTX-A). The walking speed on the treadmill was determined before pre–BTX-A as the most comfortable speed for each patient; both assessments were then performed at the same walking speed. All other conditions and medical care remained identical during the study for each patient. Time since stroke was greater than 6 months, and the spontaneous recovery had ended.

Normative values were obtained with the same protocol in 12 subjects without stroke (4 men, 8 women; age, 23±2y; weight, 58±10kg; height, 167±7cm) walking slowly (.55m/s).

Statistical Analysis 

A paired t test was computed to compare pre– and post–BTX-A data (mean ± standard deviation [SD]) and identify the effect of BTX-A injection. Signed-rank tests were used to compare nonparametric data (median [range]). The 95% confidence intervals (CIs) for difference of means are presented for each parametric statistic.

Back to Article Outline

Results 

The procedure was well tolerated by all patients. Eighteen of the 19 patients included completed the study. One patient (patient 6) dropped out because of a pelvic fracture secondary to a fall not related to a quadriceps femoris paresis. BTX-A injection into the rectus femoris was associated with beneficial effects on the SIAS, with scores increasing from 53 (pre–BTX-A range, 36–65) to 57 (post–BTX-A range, 42–70) (P=.005), and on rectus femoris spasticity, with a decrease in the Duncan-Ely test from 3 (range, 1–3) to 1 (range, 0–3) (P<.001) (table 2).

Table 2. Results Obtained for All Patients

NOTE. Values are median (range) or mean ± SD.

Abbreviations: d, knee angular displacement; m, knee joint moment; p, knee joint power; s, knee angular speed; TAT, total activation time.

The mean speed adopted by patients was only .54±.14m/s. Pre–BTX-A step frequency increased slightly (85±14 steps/min) compared with normative values (75±7 steps/min) (see table 2). Post–BTX-A results were similar (82±16 steps/min; 95% CI, −0.635 to 6.083; P=.105).

In figure 1, a typical trace of mean knee displacement, angular speed, net joint moment, joint power, and electromyographic timing are presented as functions of stride duration (in percent) in 1 patient and in the 12 subjects without stroke at the same speed (.55m/s). The grey area represents normative values (mean ±1 SD). The dashed dark grey line and the solid black line represent the pre– and post–BTX-A data of the patient, respectively. Arrows show key data selected for each variable in all patients on every assessment. These data were submitted to the statistical analysis, and mean values for all subjects are presented in table 2.

In subjects without stroke walking at .55m/s, the knee was extended during the stance phase and a large peak of flexion occurred during the swing phase to allow foot clearance (see fig 1A). Pre–BTX-A knee flexion (d2) (26°±13°) was half that of subjects without stroke (50°±12°). The shape of the representative curve was also abnormal. Knee flexion stopped at the beginning of the swing phase, and the curve presented a characteristic double bump shape. After the BTX-A injection knee flexion increased, and this double bump disappeared. The maximum knee flexion during the swing (d2) increased from 26°±13° (pre–BTX-A) to 31°±14° (post–BTX-A) (95% CI, −8 to −3; P<.001).

Knee angular speed is presented in figure 1B. In subjects without stroke, knee flexion speed reached a maximum at the end of the stance phase and then decreased steadily up to 90% of the gait cycle. Pre–BTX-A knee flexion speed was highly abnormal. The maximum speed reached at the end of the stance phase (s2) (82°±63°/s) corresponded to one third of normative values (275°±127°/s). The shape of the representative curve was also abnormal and characteristic of stiff-knee gait. The speed of knee flexion decreased at the beginning of the swing phase and reached negative values (s3) corresponding to an extension movement. Knee flexion then reaccelerated so that the knee angular speed also showed a double bump pattern, related to that observed on the kinematic curve (see fig 1A). After the BTX-A injection, knee flexion speed improved and this double bump disappeared. The knee flexion speed at the end of the stance phase (s2) increased from 82°±63° (pre–BTX-A) to 112°±75°/s (post–BTX-A) (95% CI, −52 to −9; P=.009). To quantify the effect of BTX-A on the double bump shape, a linear regression was fitted to the knee speed between s1 and s3. The r² of this regression increased from .47±.32 to .59±.33 (95% CI, −.23 to −.01; P=.031), indicating a more regular and normal decrease in the knee speed during the swing phase.

The net joint moment of the knee (fig 1C) was similar between subjects with and without stroke during the second part of the gait cycle. The joint power of the knee is presented in figure 1D. In subjects without stroke, the knee power showed a large peak of negative power related to the eccentric contraction of the quadriceps femoris to decelerate knee flexion at the end of the stance phase. Pre–BTX-A knee joint power was highly abnormal in patients: the maximum negative power (p1) reached at the end of the stance phase (−.27±.23W/kg) was only one third of normative values (−.88±.23W/kg). Post–BTX-A peak increased significantly (−.37±.26W/kg; 95% CI, .06 to −.14; P<.001).

The timing of the vastus medialis, vastus lateralis, biceps femoris, and rectus femoris activity is presented in figure 1E. The typical trace shows that the pre–BTX-A total activation time (the sum of both periods of activation) was increased and even continuous for the vastus medialis, vastus lateralis, and rectus femoris and was almost continuous for the biceps femoris. Post–BTX-A vastus medialis activation time was slightly decreased, whereas those of the vastus lateralis and biceps femoris were more markedly reduced. Taking the mean results of all patients, the total activation times of the vastus lateralis and biceps femoris decreased significantly from 69%±17% (pre–BTX-A) to 60%±14% (post–BTX-A) (95% CI, 2−17; P=.021) and from 80%±16% to 67%±18% (95% CI, 1−25; P=.04), respectively. The activation time of the vastus medialis did not change significantly, from 69%±21% to 65%±16% (95% CI, −9 to 17; P=.503). The pre–BTX-A rectus femoris activation time was 76%±26%. This muscle was not analyzed post–BTX-A to save patients pain and discomfort.

The pre–BTX-A energy cost (5.4±1.5J·kg⁻¹·m⁻¹) was 2.3 times greater than in data for subjects without stroke (2.3±0.8J·kg⁻¹·m⁻¹). The post–BTX-A net energy cost of walking decreased slightly but not significantly (4.9±1.5J·kg⁻¹·m⁻¹; 95% CI, −0.2 to 1.2; P=.12). However, the 18 patients can be divided into 2 groups. On 1 hand, 4 patients did not show any pre–BTX-A knee flexion (5°±1°). Their post–BTX-A knee flexion (95% CI, −6 to 2; P=.229) and net energy cost of walking (95% CI, −4.2 to 3.6; P=0.8) remained unchanged (nonresponders). On the other hand, 14 patients presented a small pre–BTX-A flexion (25°±13°) and responded to the treatment (responders, 33°±16°; 95% CI, −12 to −3; P=.005). Their net energy cost of walking decreased significantly after the BTX-A, from 5.4±1.6 to 4.6±1.3J·kg⁻¹·m⁻¹ (95% CI, 0.3−1.3, P=.006).

Back to Article Outline

Discussion 

The present study shows the beneficial effects of BTX-A injection into the rectus femoris on stiff-knee gait in people with stroke. The treatment improved knee joint displacement, angular speed, and power. These improvements, and the decrease in the rectus femoris tone, decreased the vastus lateralis and biceps femoris activation time.

Although it might have been preferable to perform a double-blind randomized controlled trial (RCT) to assess the effects of BTX-A, this would have been prohibitively expensive and long given the number of patients required and the cost of the motion analysis measurements. Because of the restrictive inclusion and exclusion criteria, only 19 patients were included in our study; it would have been difficult to split our patients into 2 similar groups (BTX-A rectus femoris injection, control) and perform an RCT.

To our knowledge, our study is the first to study the efficacy of BTX-A injection in the rectus femoris on stiff-knee gait among people with stroke. The capacity of BTX-A to decrease muscle tone in spastic patients has been proven by several RCTs,¹⁸ but the effect of the decreased rectus femoris tone after BTX-A rectus femoris injection on stiff-knee gait was unknown and never studied. Several researchers wrote that observational studies are useful as preliminary studies. If this preliminary study shows a positive effect of this new therapy, this effectiveness should then be confirmed by an RCT.¹⁹, ²⁰ Moreover, it is not obvious that results from observational studies differ from those of RCTs.²¹, ²²

Before the BTX-A injection, we performed MBB with anesthetic according to the method described by Sung et al.²³ This method was used to avoid falls due to paresis after the BTX-A treatment and to predict BTX-A results. No patient had paresis during MBB, and all were treated. Gait analysis was performed during selective MBB (2% lidocaine [Xylocaine], 1.5mL) in 12 of our 18 patients (post-MBB) to study the ability of MBB to predict post–BTX-A results. It was considered predictive when 1-way repeated-measures analysis of variance showed that post-MBB and post–BTX-A data differed from pre–BTX-A data and there were no differences in post–BTX-A and post-MBB results. Sung²³ showed that MBB was efficient at improving maximum knee flexion and speed of knee flexion at the end of the stance phase. These results were partly confirmed in the present study. Indeed, only post-MBB speed range (s4, see fig1B)—computed as the difference between maximal knee flexion speed (s1, see fig1B) and minimal knee flexion speed (s3, see fig1B) (P<.001)—and negative power (p1, see fig1B) (P=.002) showed the required predictive characteristics (see table 3). Mean values of speed range (s4) were 179°±98°/s (pre–BTX-A), 238°±127°/s (post-MBB), and 263°±124°/s (post–BTX-A). Mean values of negative power were −.33±.26W/kg (pre–BTX-A), −.42±.26W/kg (post-MBB), and −.46±.26W/kg (post–BTX-A). The other variables did not satisfy the predictive requirements; however, these variables had post-MBB values located between pre–BTX-A and post–BTX-A results. These results show a nonsignificant tendency for MBB to predict post–BTX-A results for these variables. The results were probably not significant, because patients did not have enough time to get accustomed to the decreased muscle tone. The effect of MBB is rarely evaluated by gait analysis²⁴ but more often by clinical assessment.²⁵ Use of MBB remains interesting: its efficiency can be assessed by the Duncan-Ely test, and it allows the clinical assessment of rectus femoris paresis, which is difficult if the rectus femoris is spastic. Other muscles potentially responsible for stiff-knee—for example, triceps surae (soleus)—should also be evaluated. This extensive evaluation would help to determine precisely which muscles have to be treated.

Table 3. Results Obtained for Patients Who Underwent a Rectus Femoris MBB

NOTE. Values are mean ± SD.

The 4 nonresponders represented only a small proportion of the patients studied, but their results are interesting. Neurologic impairments were similar in both groups. The absence of pre–BTX-A knee flexion in these patients could be due to the spasticity or paresis of other muscles involved in flexing or extending the knee or to diffuse muscle spasticity around the knee (and eventually the whole lower limb). BTX-A injection into the rectus femoris would then not be sufficient to improve knee flexion, and BTX-A should perhaps be injected into more than 1 muscle. The small number of these patients is not surprising, because isolated lack of knee flexion—which is probably not common in these patients, frequently suffering from diffuse spasticity—was an inclusion criterion. In patients who did have some knee flexion, BTX-A was effective.

Back to Article Outline

Conclusions 

For the first time, the beneficial effect of BTX-A rectus femoris injection on stiff-knee gait was shown in most adults with stroke. This study shows the importance of the rectus femoris in the physiopathology of stiff-knee gait. The result of this preliminary observational study should be confirmed by an RCT. Future studies should also focus on the long-term benefits of BTX-A.

Suppliers

Back to Article Outline

References 

1. Perry J. Gait analysis: normal and pathological function. Thorofare: Slack; 1992;
   • View In Article
2. Riley PO, Kerrigan DC. Torque action of two-joint muscles in the swing period of stiff-legged gait: a forward dynamic model analysis. J Biomech. 1998;31:835–840
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (254 KB)
   • CrossRef
3. Sung DH, Bang HJ. Motor branch block of the rectus femoris: its effectiveness in stiff-legged gait in spastic paresis. Arch Phys Med Rehabil. 2000;81:910–915
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (58 KB)
   • CrossRef
4. Goldberg SR, Anderson FC, Pandy MG, Delp SL. Muscles that influence knee flexion velocity in double support: implications for stiff-knee gait. J Biomech. 2004;37:1189–1196
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (351 KB)
   • CrossRef
5. Goldberg SR, Ounpuu S, Arnold AS, Gage JR, Delp SL. Kinematic and kinetic factors that correlate with improved knee flexion following treatment for stiff-knee gait. J Biomech. 2006;39:689–698
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (503 KB)
   • CrossRef
6. Albert TA, Yelnik A, Bonan I, Lebreton F, Bussel B. Effectiveness of femoral nerve selective block in patients with spasticity: preliminary results. Arch Phys Med Rehabil. 2002;83:692–696
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (55 KB)
   • CrossRef
7. Waters RL, Garland DE, Perry J, Habig T, Slabaugh P. Stiff-legged gait in hemiplegia: surgical correction. J Bone Joint Surg Am. 1979;61:927–933
   • View In Article
   • MEDLINE
8. Chino N, Sonoda S, Domen K, Saitoh E, Kimura A. Stroke impairment assessment set. In:  Chino N,  Melvin JL editor. Functional evaluation of stroke patients. Tokyo: Springer; 1996;p. 19–31
   • View In Article
9. Tsuji T, Liu M, Sonoda S, Domen K, Tsujuchi K, Chino N. The stroke impairment assessment set: its internal consistency and predictive validity. Arch Phys Med Rehabil. 2000;81:863–868
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (133 KB)
   • CrossRef
10. Marks MC, Alexander J, Sutherland DH, Chambers HG. Clinical utility of the Duncan-Ely test for rectus femoris dysfunction during the swing phase of gait. Dev Med Child Neurol. 2003;45:763–768
    • View In Article
    • MEDLINE
    • CrossRef
11. Dierick F, Penta M, Renaut D, Detrembleur C. A force measuring treadmill in clinical gait analysis. Gait Posture. 2004;20:299–303
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (155 KB)
    • CrossRef
12. Davis RB, Ounpuu S, Tyburski D, Gage JR. A gait analysis data collection and reduction technique. Hum Mov Sci. 1991;10:575–587
    • View In Article
    • CrossRef
13. Davis BL, Cavanagh PR. Decomposition of superimposed ground reaction forces into left and right force profiles. J Biomech. 1993;26:593–597
    • View In Article
    • MEDLINE
    • CrossRef
14. Raison M, Detrembleur C, Fisette P, Samin JC, Willems PY. Determination of three-dimensional joint efforts of a subject walking on a treadmill. Comput Methods Biomech Biomed Engin. 2005;(Supp 1):229–230
    • View In Article
15. Van Boxtel GJ, Geraats LH, Van den Berg Lenssen MM, Brunia CH. Detection of EMG onset in ERP research. Psychophysiology. 1993;30:405–412
    • View In Article
    • MEDLINE
    • CrossRef
16. Mickelborough J, van der Linden ML, Richards J, Ennos AR. Validity and reliability of a kinematic protocol for determining foot contact events. Gait Posture. 2000;11:32–37
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (231 KB)
    • CrossRef
17. McArdle WD, Katch FI, Katch VL. Exercise physiology: energy, nutrition, and human performance. 4th ed.. Philadelphia: Williams & Wilkins; 1996;
    • View In Article
18. Ward AB, Molenaers G, Colosimo C, Berardelli A. Clinical value of botulinum toxin in neurological indications. Eur J Neurol. 2006;13(Suppl 4):20–26
    • View In Article
    • CrossRef
19. Tate DG, Findley T, Dijkers M, Nobunaga AI, Karunas RB. Randomized clinical trials in medical rehabilitation research. Am J Phys Med Rehabil. 1999;78:486–499
    • View In Article
    • MEDLINE
    • CrossRef
20. Horn SD, DeJong G, Ryser DK, Veazie PJ, Teraoka J. Another look at observational studies in rehabilitation research: going beyond the holy grail of the randomized controlled trial. Arch Phys Med Rehabil. 2005;86(12 Suppl 2):S8–S15
    • View In Article
    • MEDLINE
21. Benson K, Hartz AJ. A comparison of observational studies and randomized, controlled trials. N Engl J Med. 2000;342:1878–1886
    • View In Article
    • MEDLINE
    • CrossRef
22. Concato J. Observational versus experimental studies: what’s the evidence for a hierarchy?. NeuroRx. 2004;1:341–347
    • View In Article
    • MEDLINE
    • CrossRef
23. Sung DH, Jung JY, Kim HD, Ha BJ, Ko YJ. Motor branch of the rectus femoris: anatomic location for selective motor branch block in stiff-legged gait. Arch Phys Med Rehabil. 2003;84:1028–1031
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (163 KB)
    • CrossRef
24. Lejeune TM, Detrembleur C, De Nayer JA, Plaghki LH. Gait analysis and motor point block in dynamic varus of the rear foot in a head injured adult. Arch Phys Med Rehabil. 1996;77:827–829
    • View In Article
    • Abstract
    • Full-Text PDF (318 KB)
    • CrossRef
25. Buffenoir K, Decq P, Lefaucheur JP. Interest of peripheral anesthetic blocks as a diagnosis and prognosis tool in patients with spastic equinus foot: a clinical and electrophysiological study of the effects of block of nerve branches to the triceps surae muscle. Clin Neurophysiol. 2005;116:1596–1600
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (187 KB)
    • CrossRef

• a Mercury LTmed; H/P/Cosmos Sports & Medical Gmbh, Am Sportplatz 8, 83365 Nussdorf-Traunstein, Germany.
• b BTS SpA, Viale Forlanini 40, Garbagnate Milanese, 20024, Italy.
• c Medi-Trace; Graphic Controls Corp, 400 Exchange St, Buffalo, NY 14204-2064.
• d Quark b² win, version 5.1a; Cosmed Via dei Piani di Mt Savello 37, Pavona di Albano – Rome, I-00040, Italy.