Docstoc

Title Evaluation of conventional and dynamic ankle foot orthosis

Document Sample
Title Evaluation of conventional and dynamic ankle foot orthosis Powered By Docstoc
					                                  CHAPTER 1


                     GENERAL INTRODUCTION



Dynamic equinus is one of the commonest deformities in children with diplegic

cerebral palsy (CP) and it is due to the spasticity of calf muscles. Ankle foot orthoses

are externally applied devices which are used to modify structural and functional

characteristics of neuromuscular and skeletal systems by the external forces to

control the equinus deformity. Conventional ankle foot orthosis (AFO) is commonly

prescribed for children with dynamic equinus and it creates a three-point pressure

system to control the deformity. However, in the past 10 years, dynamic ankle foot

orthosis (DAFO) has been increasingly used as an alternative to the conventional

ankle foot orthosis (AFO) in the neuromuscular disorders; the DAFO redistributes

the pressure under the foot and alter the sensory input of the foot reflexes which

control the equinus deformity.



Both the AFO and DAFO have different mechanisms to control the equinus, but the

differences between these two orthoses are still unclear. Hence, in order to compare

the effect of DAFO and AFO on the patient, gait analysis is needed to find out the

orthotic effect. Gait analysis is the systematic measurement, description, and

assessment of quantities that characterize human locomotion including kinematics,

kinetics and electromyography (EMG) information. With the gait analysis

information gained, physician can integrate clinical assessment with gait analysis,

and thus the most suitable treatment could be prescribed.



                                          -1-
The root mean square value (RMS) and median frequency (MF) are the most

frequently used parameters in functional assessment of skeletal muscles (spinal

muscles, upper limb muscles), because of their good objectivity and reproducibility.

In this study, the RMS and MF values were adopted to study the EMG findings: Root

mean square represents the square root of the average power of the EMG signal for a

given time while median frequency reflects the number of motion unit recruitment.



Gait analysis has proven valuable in evaluating the effect of orthosis on CP.

However, not many studies were done investigating the various aspects of gait

including kinetics and electromyography (EMG) interpretation. In particular, only a

few studies comparing the differences between the AFO and DAFO were published.

It was believed that the different treatment mechanisms by these orthoses should be

evaluated to help clinician to know the mechanisms.



The main aim of this study was to evaluate the orthotic effects of DAFO and AFO

with walking kinetics, kinematics and EMG information. The alterations of diplegic

spastic gait would also be mentioned, as well as try using the RMS and MF value to

interpret the walking EMG.




                                        -2-
                                CHAPTER 2

                       LITERATURE REVIEW



The following topics were discussed in this chapter. Firstly, background information

of cerebral palsy was introduced. Secondly, the theories of the orthotic treatments

(ankle foot orthosis and dynamic ankle foot orthosis) were highlighted. Thirdly, the

equipments, parameters and clinical uses of gait analysis were discussed. Finally,

researches on AFO and DAFO were reviewed.




                                        -3-
2.1 Background on Cerebral Palsy


Cerebral palsy is a non-progressive brain injury before, during or shortly after birth.

This interferes with the sensory, motor as well as intellectual development depending

on the severity of the lesion in brain. The etiologic insult can occur in the prenatal,

perinatal and postnatal period. An insult happening in prenatal period includes causes

such as maternal infection, maternal drug or alcohol; an insult occurring in perinatal

period can be due to trauma, placental complications, hypoxia, anoxia or low

birth-weight; postnatally, the insult can resulted from head trauma and vascular

accident in brain (Renshaw TS, 1996). The main causes for cerebral palsy were low

birth-weight and prematurely, particularly in the spastic diplegic type (Nelson KB,

2001; Renshaw TS et.al., 1995).



The overall prevalence of cerebral palsy was 1.4 to 2.3 per thousand in the US

population (Nelson KB, 2001); Worldwide, Renshaw TS (1996) summarized that

cerebral palsy was found in from 1 to 7 per 1000 children. Cerebral palsy can be

categorized into its clinical subgroups: hemiplegia, diplegia and quadriplegia in

spastic, dyskinetic, ataxic and mixed forms.



Gage JR (1991) mentioned the percentage of the different anatomic subgroups, and

the most common subgroup was diplegia (32%); the percentage in other subtypes like

hemiplegia, quadriplegia, and all other types were 29%, 24% and 14% at age seven

respectively. Quadriplegia generally has more global brain damage than diplegia or

hemiplegia, and they are usually wheelchair bounded and unable to ambulate

independently.

                                         -4-
The hemiplegia is diagnosed as one side of the body being involved, and the upper

limb being more affected than the lower. On the other hand, the diplegia is defined as

the involvement primarily in both lower extremities with a relatively normal function

of upper extremities. The function of their lower limbs are involved similarly,

therefore, the gait data of the left and right sides were combined to one mean in some

gait studies with diplegia (Carlson et al, 1997; Crenshaw S et al, 2000)




                                         -5-
2.2 Common Deformities of Cerebral Palsy


The deformities depend on severity and location of brain injury, these deformities

may be found at foot, knee, hip, upper extremities, or a combination of the above.

The deformities arise mainly due to muscle imbalance (spasticity) between affected

and antagonist muscles. Lima D (1990) summarized the Bishop`s theory that the

spasticity occurred because the inhibition normally provided by the suppresser areas

of the brain was affected, and the major consequence of this disruption results in

hypersensitive phasic stretch reflexes, hyperactive tonic reflexes and clonus. Such

hypersensitivity is triggered during walking, and results in dynamic equinus

(Lohman M et al, 1993).



Rab GT (1992) pointed out that spasticity is a common presentation in diplegic gait.

Dynamic equinus resulting from spasticity of calf muscle was one of the most

common deformities (Makley JT et al, 1983; Goldstein M et al, 2001), this is due to

involuntary muscle imbalance between ankle plantarflexors and dorsiflexsors, in

which the gastrocnemius and soleus muscles tend to be much stronger than the

dorsiflexsors. If the dynamic equinus is not properly treated, the muscle length will

become short and eventually contracture will develop. Makley JT et al (1983) further

illustrated with Heuter-Volkmann`s principle: In normal individuals, muscles growth

keep up with skeletal growth by serial addition of sarcomere units, which permits

optimal length-tension relationships between the muscle motor and bone lever-arm.

In cerebral palsy patients, this mechanism does not appear work normally. This

results in progressive muscle shortening in relation to bone and equinus deformity.

To prevent contracture, earlier treatment in the dynamic phase is recommended.

                                        -6-
2.3 Measuring Spasticity


The spasticity resulting from various disease processes like cerebral palsy,

cerebrovascular accidents, spinal cord injury, head trauma and multiple sclerosis, are

all quite different. Smyth MD et al (2000) reviewed that the spasticity can be defined

as a condition in which there is a velocity-dependent increase in resistance of the

muscle group to passive stretch with a “clasp-knife” type component associated with

the hyperactive tendon reflexes.



In clinical setting, the most commonly used measurement is the modified Ashworth

scale. This scale had been proposed by Bohannon RW et al (1987), which modified

from the Ashworth scale. Bohannon RW et al (1987) suggested that adding an

additional level of measurement (1+) will increase the sensitivity of this method, the

difference between the two scales are shown in Table 2.1. It is based on previous

studies that modified Ashworth scale could be regarded as an ordinal and nominal

level measure of resistance to passive movement (Johnson GR, 2001).



Other measurements of spasticity, such as measurements of resistance of motion with

dynamometer (Engsberg JR et al, 2000), electromyographic technique (Crenna P,

1998; Pirpiris M 2001), measurement of deep tendon reflexes with combination of

angular accelerometer and force transducer (Skinner SR, 1992), are impractical to be

applied in clinical setting but they are still valuable in scientific researches. In this

clinical study, the spasticity was measured with the modified Ashworth Scale.




                                          -7-
Score   Ashworth Scale (Ashworth, Modified Ashworth scale
        1964)                     (Bohannon RW et al, 1987)
0       No increase in tone.             No increase in muscle tone.
1       Slight increase in tone giving a Slight increase in muscle tone,
        catch when the limb was manifested by a catch and release or by
        moved in flexion or extension. minimal resistance at the end of the
                                         range of motion.
1+                                       Slight increase in muscle tone,
                                         manifested by a catch, followed by
                                         minimal resistance throughout the
                                         remainder (less than half) of the ROM
2       More marked increase in tone More marked increase in muscle tone
        but limb easily flexed.      through most of the ROM, but affected
                                     part(s) easily moved.
3       Considerable increase in tone – Considerable increase in muscle tone,
        passive movement difficult.     passive movement difficult.
4       Limb rigid     in   flexion   or Affected part (s) rigid in flexion or
        extension.                       extension.
Table 2.1: Definitions of the Ashworth and modified Ashworth scales. The score 1+

shown the main difference between two scales (Adopted from Johnson GR, 2001).




                                        -8-
2.4 Clinical Problems of Dynamic Equinus


The dynamic equinus (spastic equinus) is mainly due to the spasticity of the

plantarflexors. The patient with dynamic equinus will only present with excessive

plantarflexion during walking, but not in static situation. For these patients with

dynamic equinus, the passive range of ankle motion (ROM) of these patients was

full.



During walking, the excessive ankle plantarflexion interferes with foot clearance in

swing phase and foot positioning at initial contact. Moreover, those patients who

present the excessive ankle plantarflexion had a shorter stride length and smaller

total range of ankle motion, comparing with the normal. These patients walk with a

toe-toe gait or a toe-heel gait which drag and fall on the floor easily (Gage JR, 1991;

Perry J, 1992).



In order to balance, secondary (coping) gait deformities may develop. For example,

to overcome an excessive plantarflexion during swing phase, an excessive hip

flexion, excessive knee flexion, hip circumduction or lateral trunk bending (David JR

et al, 1996) may occur. If the dynamic equinus is not treated properly, plantarflexion

contracture may develop.




                                         -9-
2.5 Treatment of the Equinus


The treatment is individualized including surgery, orthosis, casting, physical and

occupational therapy depending upon the nature and severity of the problem

(Goldstein M et al, 2001). In general, patients with dynamic equinus will be treated

conservatively, but surgery may be required for patients who developed an equinus

contracture.




2.5.1 Conservative treatment


2.5.1.1 Physical and Occupational Treatment


Rehabilitation plays an essential role in the treatment of cerebral palsy patients,

regardless of whether surgical or orthotic treatments have been done. Physical and

occupational therapy is needed to help the patient to regain their muscle strength and

achieve an independent daily living as possible. In treatment of dynamic equinus, the

dorsiflexors must be strengthened to prevent the occurrence of plantarflexion

contracture. To reduce hospitalization, the patient should undergo a home exercise

programme with regular follow up. Physical and occupational therapy have been

offered to the patients in most of the treatment protocols (Renshaw TS et al, 1995;

Smyth MD et al, 2000).




                                        - 10 -
2.5.1.2 Casting


The inhibitive cast method has been used over decades for children with spasticity.

This aims to keep the ankle in the neutral anatomical position and reduce the

abnormal foot reflex, so that the development of fixed plantarflexion deformity can

be prevented. Harris SR et al (1986) reviewed that the inhibitive cast had been

recommended for developmental treatment approach, with reducing the tone of the

plantarflexors and facilitating walking. She also stated that such cast appeared to

reduce reflex-induced foot deformities and hypertonus in children with cerebral palsy.

Moreover, it was concluded that the child had a better performance with the

tone-reducing casts than that with the standard casts (which same as tone-reducing

cast but no tone-reducing modification) (Hinderer KA et al, 1988). However, when

casting was discontinued, the dynamic equinus returned (Zachazewski JE et al, 1982).

The inhibitive cast has been gradually replaced by the dynamic ankle foot orthoses,

which is lighter and can be fitted inside shoes (Diamond MF et al, 1990).




2.5.1.3 Drug Injection


Injection of Botox (Botulinum A toxin) has been widely applied to different muscle

groups (from tiny facial muscles to the big hamstrings muscles) to reduce muscle

tone. This technique is easy to apply and avoids an operation, and has been

recommended for the management of spasticity by many authors (Graham HK et al,

2000; Metaxiotis D et al, 2002; Sutherland DH et al, 1999). However, the Botox

effect will diminish in about 5 months.



                                          - 11 -
2.5.1.4 Orthotic treatment


Orthosis is used to inhibit an abnormal pattern by restricting abnormal movement,

promoting normal alignment and minimizing the restriction of movement, as well as

preventing contracture. Further details will be discussed in session 2.5.3.




2.5.2 Surgical Treatment

Selective dorsal rhizotomy and muscle-tendon lengthening are commonly done for

the dynamic equinus. Selective dorsal rhizotomy (Smyth MD et al, 2000) interrupted

the stretch reflex of the afferent limb at the posterior spinal nerve root or rootlet level,

and thus reduced the spasticity. Muscle tendon lengthening involved a step-cut or

sliding of the Archilles tendon (Renshaw TS et al, 1995). Previous studies showed

the equinus was improved after the muscle-tendon lengthening surgery (Damron TA

et al, 1994; Granata KP et al, 2000; Rose SA et al, 1993; Saraph V et al, 2000).




                                           - 12 -
2.5.3 Orthotic Treatment

Orthoses are externally applied devices which are used to modify structural and

functional characteristics of neuromuscular and skeletal systems by exerting forces

on the body (Bowker et al, 1993; International Organization for Standardization,

1989). The orthoses are usually custom-fabricated from thermoplastics that are

moulded over the rectified plaster models of affected part of the body (Morris C et al,

2002). The main functions of the orthoses are to inhibit an abnormal pattern by

restriction on abnormal movement, promoting normal alignment and movement, as

well as prevent contracture, so as to improve the patient’s quality of life.



Ankle foot orthosis (AFO) is the commonest used orthosis for patient with

neurological deficits such as cerebral palsy, spinal injury, stroke and head trauma.

Approximately 53,000 ankle foot orthoses (AFOs) per year were prescribed in the

United States for children with cerebral palsy to provide an external support to prevent

foot drop during walking (Parker K et al. 1994). In 1996, number of AFO prescribed

has been increased to about 100,000 per year in the United States (Parker K et al

1996).




2.5.3.1 Mechanism of Conventional Ankle Foot Orthosis


Conventional ankle foot orthosis (AFO) has been used since 1960s. This was

tailored-made by thermoplastic and kept the ankle at neutral position. The AFO may

be prescribed not only to compensate for problems encountered during the swing

phase, but also to influence the characteristics of the stance phase. Condie DN et al

                                          - 13 -
(1993) demonstrated the development of the three-point pressure system required to

control equinus during swing phase by simple biomechanics (Fig 2.1) in the

following paragraph:



Considering first, the normal leg during midswing, the principal external force

present is from the weight of the lower leg and foot (W) (Fig 2.1a). If the foot is now

separated free body diagrams and deduced into the internal forces and moments at

the ankle joint (Fig 2.1b). The patient with spasticity at calf muscle, it will a have

larger plantarflexion moment (Fig 2.1c). Therefore, it is possible to propose

additional external forces (by the AFO) to counter balance the excessive

plantarflexion due to spasticity of calf muscle R1 (Fig 2.1d). Finally, R1, R3 and R5

form a three-point pressure system to control drop foot during swing.



To achieve the above purpose, the rigidity of the AFO provides the passive pressure

on the bottom of the forefoot and pressure on the calf muscle posteriorly, in addition

to the ankle strap which provides a 45 degree anterior-superior force pressure on the

ankle. These pressures formed a three-point pressure system (Fig 2.2: the arrows) to

prevent drop foot in the swing phase (Charlton PT et al, 2001; Condie DN et al,

1993).




                                         - 14 -
        (a)




Fig 2.1: (a)-(e) Development of the three-force system required to control drop
foot during midswing. R1, R3 and R5 formed the three-force system to correct
equinus. (Adopted from Condie DN et al, 1993)




                                    - 15 -
                                 45 degrees      Fig 2.2: Polypropylene ankle
                                                 foot orthosis with full piece
                                                 and 45 degree ankle strap.




The orthosis has been repeatedly evaluated since 1960. Meadows (1984) provided a

very good description of AFO that the straps of the orthoses encouraged better

control and anchor the hindfoot into the orthosis, and the additional sole wedge or

rocker made more fluent mobility from initial contact to toe off. The author further

concluded that the necessity for surgery was reduced; from the fundamentals of the

AFO design, some other types of orthoses have been developed, like hinged AFO,

posterior leaf AFO, carbon-fiber type AFO, tone-inhibiting (dynamic) AFO and

spring AFO. Many researches on functional differences between different types of

the AFOs have been investigated.




2.5.3.2 Mechanism of Dynamic Ankle Foot Orthosis


It was summarized from Duncan WR (1960) that four tonic movements of foot could

be elicited in cerebral palsy children including (1) toe grasp reflex; (2) inversion

reflex; (3) eversion reflex and (4) dorsiflexion reflex. It was shown that stimulation

of particular reflexogenous areas of the foot (Fig 2.3) had an effect on more proximal

                                        - 16 -
musculature. Through a change of stimulus under the foot, the proximal muscle

could be activated or inhibited, and therefore the walking pattern altered.




                                   Fig 2.3: Reflexogenous areas on the
                                   plantar surface of the foot: 1) toe
                                   grasp; 2) inversion; 3) eversion; and 4)
                                   dorsiflexion. (Adopted from Lima D,
                                   1990)




Base on the findings from Duncan WR (1960), Hylton NM (1990) designed a special

dynamic ankle foot orthosis (DAFO or tone-inhibitive orthosis) which implemented

the inhibitive cast technique onto the new orthosis. She advocated the use of a

footplate that was manufactured accurately to reinforce all of the dynamic arches of

the foot, including the transverse arch, medial and lateral longitudinal arches; and to

relieve pressure under the metatarsal fat pad and calcaneal fat pad (Hylton NM,

1990). This modification aimed to inhibit the plantarflexors reflex and trigger of

dorsiflexors action. As a result, the dynamic equinus could be controlled. Lohman M

et al (1993) and Pratt DJ (2000) further emphasized on the design benefits of the

tone-reducing orthosis; since then DAFO has been increasingly prescribed for

patients with dynamic equinus.

                                         - 17 -
The design and construction of AFOs has undergone a significant change in recent

years due to the development of material science and technology. From a review on

efficacy of lower-limb orthoses for cerebral palsy, use of DAFOs was claimed to

improve postural control and balance because of the highly contoured footplate that

was proposed to influence muscle tone (Morris C, 2002).



The ‘dynamic’ in DAFO does not mean that the brace is mobile, but it only means

that the orthosis has less restriction on ankle motion. Although the evidence

supporting the physiological mechanism by reduction in tone was not too much,

these techniques have been widely adopted (Morris C, 2002). It is obviously

necessary to provide more evidence on the physiological benefits of DAFO by

kinetics, kinematics or electromyography parameters.




                                       - 18 -
2.6 Gait Analysis


Gait analysis is the systematic measurement, description and assessment of quantities

that characterize human locomotion (Perry J, 1992). The human gait has been well

studied in the past two decades (Patla AE, 1991; Braune W et al, 1987). Gait analysis

can be done by observation and instrumentation. Observational gait analysis begins

with a gross analysis of the child’s gait, focusing on velocity, cadence, step length

and stability. This is followed by a serial horizontal analysis from the ankle joint

upwards to the hip. Sagittal plane analysis is observed from the right and then the left

side. Coronal plane analysis is performed with the child walking toward and away

from the observer (David JR, 1996). Combining this observation with the clinical

history of the patient and physical findings, the functional deficits with respect to the

sub-phases of the gait cycle can be identified.



Instrumentation (Computerized) gait analysis can provide an objective and

comprehensive documentation with kinematics, kinetics and electromyography

parameters in different joints and planes. It is commonly used as a tool to study the

effect of treatments on patients (Gage JR, 1991; Gage JR et al, 1995; Whittle MW,

1996a; Whittle MW, 1996b; Winter DA, 1991) and as a diagnostic tool like

radiotherapy in the clinical setting (Adam JM et al, 1994; Gage JR, 1994; Kay RM et

al, 2000). The gait analysis consists of three measurement systems: (1) Motion

analysis system; (2) Force platform and (3) Electromyography (EMG). Each system

provides information on different aspect of the parameters. To gain a complete

understanding of gait, all systems should be synchronized and analyzed.



                                          - 19 -
2.6.1 Motion Analysis System

The motion analysis system measures the kinematic parameters of the movement.

According to Winter DA (1990), the kinematic variables are involved in the

description of the movement but independent of forces that cause the movement. The

output parameters are in the form of temporal (timing), linear (distance) and angular

measurements that describe the movement of the body segments and joint angles.

Measurement techniques could be classified as direct measurement or imaging

measurement technique.



Goniometer and accelerometer are the devices for direct measurement. The

goniometer is a special name given to the electrical potentiometer that can be

attached on the body segments to measure joint angles (Winter DA, 1990). As such,

one bar is strapped to the proximal segment and the second to the distal segment, so

the relative angular data can be recorded. This revolution has been from mechanical

goniometer    to   electrical   goniometer,   and   it   further   extended   into   3D

electrogoniometer (Bontrager EL, 1998). This is generally inexpensive and output

signal is available immediately for conversion into computer, but limiting factor is

that they do not give absolute angles (Olsson EC, 1990). On the other hand, Winter

DA (1990) reviewed that the accelerometer is a device that measures acceleration of

the limb segment. Output signal is also available immediately, but great artifact will

be made from the rapid movement and it easily breaks. A major technical difficulty

for both goniometer and accelerometer is found when complex motion studied more

than one joint.



                                         - 20 -
To overcome the difficulties of measuring multi-joints at the same time, imaging

measurement has been used. Among the imaging measurement systems available,

cinematography, optoelectronics, videography and automated motion analysis

systems are commonly employed. There is no limitation in the number of markers

used so that more complicated motion can be studied simultaneously. The imaging

measurement systems can provide a record for comparison between pre and post

treatment, as well as between different subjects groups.



With the advancement of the technology, the automated motion analysis systems

have been developed into the stage that imaging data can be directly digitized and

fed into the computer. It permits a simultaneous measurement of sagittal, coronal and

transverse motion of several joints. In addition, the raw data can be automatically

calculated. Within the automated motion analysis systems, markers should be placed

on the subject before capturing.



According to Furnee H (1998), two categories of marker system are available: active

marker system and passive marker system. Active (wired) markers are uniaxial

arrays and magnetic and electroacoustic sensing system. It may be Light Emitting

Diode (LEDs) or infrared emitting diode. Passive (unwired) markers are

hemisphere-shaped or sphere-shaped, and they are covered by retroflective adhesive

sheet. Both systems require the placement of surface skin markers in a location

which represent positions of the underlying joints and segments. Finally, comparing

two marker systems, the passive marker system has the advantage of using

lightweight reflective markers without the need for electrical cables or batteries on

the users (Bontrager EL, 1998). At the present, the passive markers are available in

                                         - 21 -
our gait laboratory, and hence the passive markers are applied in this study.



For motion analysis, the marker location is automatically identified by the system

through determination of the center of the ‘bright area’ recorded for each marker. For

this achievement, at least two cameras are needed to identify the 3-Dimensional (3D)

coordinate of the marker’s image (Perry J, 1992). It should be said that the more

cameras the better, six cameras are enough in most situation (Whittle MW, 1996b).

Capturing frequency of the cameras depends on how fast the motion is tested. In

general, 60 Hz should be normally adequate for gait analysis.



With the software provided, the coordinates of the markers system can be reduced

into some meaningful parameters: tempo-distance parameters (stride length, walking

speed, stance-swing ratio and cadence) and range of motion parameters at ankle joint,

knee joint and hip joint.




2.6.2 Force Platform

Force platform captures the kinetic data (external forces and moments) which are

acting on the subject. When the kinetic data are combined with kinematic data in an

appropriate model, the net joint forces and moments of movement can then be

determined. The force platform is a rectangular steel plate embedded at the floor to

avoid vibrations during loading (Aissaoui R et al, 1998). Three force transducers are,

perpendicular to each other, symmetrically located at the corners of the force

platform. These transducers measures the loading force which are equal in the

intensity but opposite in the direction of the loading limb; The purpose of such

                                         - 22 -
perpendicular arrangement is to determine the vertical and horizontal forces of

loading limb by the computer programme automatically (Winter DA, 1990).



Different types of transducer are commercially available in the market including

strain gauge, piezoelectric, piezoresistive, capacitive and others (Winter DA, 1990).

The use of the transducer depends on which manufacturer used, and the main

manufacturers are summarized by Bontrager EL (1998): AMTI, Bertec and Kistler.

The Kistler force platform, using piezoelectric transducers (made of Quartz) is

installed in our laboratory. This deforms its crystalline structure and changes its

electrical characteristics when an external force is applied; such changes are directly

proportional to the external applied force. Consequently, the ground reaction forces

in vertical, shear horizontal in fore-aft and mediolateral can be calculated and

recorded in the computer (Olsson EC, 1990; Perry J, 1992; Winter DA, 1990).



The raw data can be reduced into the vertical and horizontal forces, and the position

of center of pressure can be determined by calculating the coordinates of the force

vectors. Furthermore, by combining these data with kinematics as well as

anthropometric measurements, joint forces and moments can be calculated by an

inverse dynamic calculation. Work, energy and power during movements can also be

found (Winter DA, 1990).




                                         - 23 -
2.6.3 Electromyography

Previous studies stated that the electromyography plays an important role in gait

analysis (Gage JR et al, 2001; Kay RM et al, 2000; Chambers HG et al, 1997). It is

because electromyography can provide the clinician an objective and quantitative

assessment of which muscles are contributing to the gait deviations, and it can

combine with the information from kinetics and kinematics to give a more

comprehensive evaluation. The origin of EMG, electrode types and signal

transmission are discussed in the following.




2.6.3.1 The Origin of Electromyography


Muscles produce active movement by conversion of metabolic energy into muscle

fiber contraction. The processes of the depolarization and repolarization of muscle

fibers produce a lot of electrical activities, which call muscle action potential (MAP).

Motor unit devotes the most fundamental grouping in muscles, and its contraction is

initiated by an action potential. When a muscle performs an active task, thousands of

motor units are recruited and contracted in a period. The electromyography (EMG) is

actually a record of spatial-temporal algebraic sum of all muscle action potential

along the muscle fibers at that period (Perry J, 1992; Rab GT, 1994).




2.6.3.2 Choice of Electrode


There are three different types of electrodes for EMG recording: needle, wire and

surface electrodes. Different types of electrodes have their own advantages and

                                         - 24 -
disadvantages. The needle electrodes are too insecure and uncomfortable in gait

analysis (Perry J, 1992; Pullman SL et al, 2000). Hence, the choice of electrode

would lie between wire and surface electrode. The wire electrodes are a pair of

50-micron, nylon insulated nickel-chromium alloy wire with the distal 2-mm bare

tips, placed in a 3.81 cm 25- or 30-gauge needle for intramuscular insertion (Perry J,

1992). The wire electrode has two main advantages: The relatively small pickup area

enables the electrode to detect an individual EMG signal from small muscles. In

addition, they can be used effectively when the activity of deep muscles is to be

examined. However, only the clinician was allowed to apply this electrode (Perry J,

1992; Whittle MW, 1996a). Most importantly, this wire electrode may be quite

uncomfortable or even be painful, which may cause further alteration in gait pattern

(Young CC et al 1989).



Although the main limitation of surface electrode is cross talk, the surface electrode

is preferable in our study because of its non-invasive and easily adopted technique.

Furthermore, no pain or discomfort from the subject is expected with this technique

(Cram JR et al, 1998; Pullman SL, 2000; Tosh RW, 1991). The EMG signal obtained

from surface electrode also provided a good reliability on EMG recording

(Kollmitzer J et al, 1999).




2.6.3.3 Surface Electrode


Silver / Silver Chloride (Ag/AgCl) electrode is the most common type of surface

electrode used in the gait analysis and give a quite consistency EMG signal (Kleissen

RFM et al, 1997; Kollmitzer J et al, 1999). The electrodes are typically round-shaped

                                        - 25 -
and attached with doubled adhesive tape to the patient. The electrode is available in

various sizes ranging from about 7 mm to 20 mm in diameter. The choice of the

diameters depends on which muscle groups are going to be studied, and about 7 to 20

mm diameters should be appropriate in most situations (Kollmitzer J et al, 1999;

Dupont L et al, 2000). Conductive electrode gel is required to increase conductivity

of the EMG signal (Bontrager EL, 1998).




2.6.3.4 Preparation of Skin


Because of the thickness of callous skin, the amount of the dead tissues, oiliness and

moisture in the skin surface which all increase the skin resistance. Winter DA (1992)

and Cram JR et al (1998) advised that to lower the skin resistance (the impedance

should be controlled at less than 10 kΩ), the skin preparation including washing,

shaving, wiping and abrading is needed to have better EMG recordings. After the

preparation of skin, surface electrode should then be placed in the mid-point of the

muscle belly and parallel to the muscle fiber, so that the maximum signal can be

collected (Cram JR, 1998; Tosh RW, 1991).




                                        - 26 -
2.7 Gait Parameters


Gait cycle is defined as the time interval between two successive occurrences of one

of the repetitive events of walking (Whittle MW, 1996a). In gait analysis, temporal

distance parameters, range of motion parameters, kinetic parameters and

electromyography parameters are investigated. It is better to know the basic

definition of gait cycle because some of the parameters are normalized with gait

cycle (Fig 2.4).




       Fig. 2.4: Time dimensions of gait cycle. (adopted from Malanga G, 1998)


The gait cycle is logically divided into two phases, stance and swing, respectively

defined by the presence or absence of floor contact for the limb being considered.

Stance phase constitutes the first 60% of the gait cycle and contains two periods of

“double support”; swing phase takes the remaining 40% of the gait cycle and begins

                                       - 27 -
at the point where the limb is advanced from behind of the body to front of the body

(Fig 2.4). Gait cycle can be further divided into eight sub-phases (Perry J, 1992): The

stance phase is sub-divided into five sub-phases, each of which has a functional

objective: initial contact, loading response, mid-stance, terminal stance and

pre-swing; The swing phase is sub-classified as initial swing, mid swing and terminal

swing.




                                         - 28 -
2.7.1 Temporal Distance Parameters

The most commonly measured variables are temporal and length measurements. The

reliability of temporal distance parameters in gait studies was judged as high

(Steinwender G et al, 2000; Stolze H et al, 1998). The definition of the commonest

temporal distance parameters are summarized as follows (Winter DA, 1991).



Cadence is defined as the number of steps per minute (steps/min). Natural or free

cadence for 20 children 6 to 13 was about 116 steps/min (Winter DA, 1991).

Sutherland et al (1988) concluded that cadence in the 1-year-old subjects was

approximately 22.5 % higher than in the 7-year-olds, and the 7-year-olds cadence

was still 26% higher than normal adult mean.



Stride length is the horizontal distance travel along the plane of progression during

one gait cycle. That is measured as the distance travel from initial contact to another

initial contact of the same limb (meter). Stride length is very much a function of the

subject’s height and possibly weight, age and sex. Sutherland et al (1988) revealed

that the stride length increased with age.



Stride time is the period of time for one gait cycle, and it is also called as cycle time

(second). It is often used to normalize the gait parameters in term of 0 to 100% for

interpretation of range of motion, kinetics and electromyography parameters. The

stride time is gradually increasing from children to adult (Sutherland et al, 1988).



Walking speed is the horizontal speed of body traveling along the plane of

                                             - 29 -
progression measured for one gait cycle (m/s). The velocity is increasing by age from

children to adult (Sutherland et al, 1988). It is dependent of stride length and cadence,

as demonstrated in the formula below:


                                    stride.length × cadence
                       velocity =                           (m / s )
                                              120



Stance time is the period of time when the foot is in contact with ground. It can be

expressed as second or the percentage of gait cycle. All researchers support that the

stance time starts from 0% and ends at about 60% of the gait cycle (Winter DA,

1991).



Swing time is the period of time when the foot is off the ground. It can be expressed

as second or the percentage of gait cycle. The swing time starts from about 60% and

ends at 100% of the gait cycle.



Stance/swing ratio is the ratio of stance time to swing time.




                                           - 30 -
2.7.2 Range of Motion Parameters

The definition and calculation of range of motion (ROM) parameters of different

joint levels are summarized by Perry J (1992), Sutherland DH et al (1998) and

Winter DA (1991). The parameters are normalized with the gait cycle, in order to

describe the motion event throughout the gait cycle and facilitate comparison

between subjects. The repeatability of ROM parameters was studied in normal and

spastic children, the repeatability of ROM parameters were acceptable (Steinwender

G et al, 2000).



Joint angles can be calculated in different planes: sagittal plane, coronal plane and

transverse plane. Most of the gait events and gait disorders can be observed in the

sagittal plane. Consequently, many studies have been focused on ROM parameters in

sagittal plane. All segment angles can be calculated using conventions by Winter DA

(1991). From figure 2.5, all segments are measured in a counter clockwise direction

from the horizontal. It is designated positive meaning of flexion (dorsiflexion) and

negative meaning of extension (plantarflexion). The hip angle is the sagittal joint

angle between trunk and thigh; the knee angle is the sagittal joint angle between

thigh and leg segments; and the ankle angle is the sagittal joint angle between leg and

foot. The motion events of hip, knee and ankle joints were normalized with gait cycle

and comprehensively described (Perry J, 1992) and those figures (Winter DA, 1991)

as follows:



Hip angle: Hip moves through two arcs motion during a normal stride (gait cycle):

extension during stance and flexion in the swing phase. The thigh is flexed about 20˚

                                         - 31 -
at initial contact, and then it reaches a neutral position at 38 % and a peak extension

(20˚) at 50 % of gait cycle. During pre-swing, the hip move toward neural position at

60% of gait cycle again, and finally 25˚ flexion of the thigh is reached and

maintained within five degrees through the terminal swing (Fig 2.6).



Knee angle: Knee passes through four arcs of motion, with flexion and extension

occurring in an alternating fashion, two flexion peaks and one trough can be

observed in figure 2.6. An initial contact, the knee is flexed about 5˚. Following the

onset of stance, the knee rapidly flexes throughout the loading response and reaches

the first peak (18˚) at 15% of gait cycle. During the rest of mid stance phase, the knee

gradually extends and reaches minimum stance phase flexion (3˚, 40% of GC). After

that, the second peak begins from the end of terminal stance. The final position of

60˚ is the maximum knee angle occurring during the swing phase, and it then extends

back to 5˚ flexion at the terminal swing.



Ankle joint: The arcs of ankle motion are not large, but they are critical for

progression and shock absorption during stance. In stance, ankle motion contributes

to limb advancement. The entire range of motion averages 30˚ (20˚ to 40˚). Ankle is

at neutral position at initial contact, and this is followed by the first plantarflexion arc

during loading response (0-20˚ of GC). Dorsiflexion continues through the mid

stance and the first half of terminal stance, reaching the maximum 10˚ angle by 48%

of GC. There is a rapid ankle plantarflexion, reaching a maximum 30˚ at the terminal

stance. Toe-off initiates the final dorsiflexion action, and finally a neutral position is

maintained during the rest of the mid swing (Fig 2.6).



                                            - 32 -
                                               where:

                                               θh = hip angle

                                               θth = thigh angle

                                               θtr = trunk angle

                                               θk = knee angle

                                               θlg = leg angle




Fig 2.5: Joints angle definition (adopted from Winter DA, 1991)




Joint angle
(degree)



                                                   % Gait Cycle




 Fig 2.6: Typical hip, knee and ankle joint movements through the gait cycle
 with 19 normal adult subjects (adopted from Winter DA, 1991)


                                      - 33 -
2.7.3 Kinetic Parameters

Ground reaction forces are basically obtained from force platform. Combining this

information with kinematics and segmental properties, the joint moment and power

can be calculated by an inverse dynamic method (Siegler S et al, 1998). Hence, the

kinetic parameters can be further divided into three different subgroups: Ground

reaction force (GRF), joint moment and power. These definition and calculation are

suggested by Perry J (1992) and Winter DA (1991).




2.7.3.1 Ground Reaction Force


Ground reaction forces (GRF) are measurements of the net vertical and shear forces

acting on the force platform, they are an algebraic summation of the

mass-acceleration products of all body segments while the foot was in contact with

the platform. This is divided into three perpendicular components including vertical,

anterior-posterior (AP) and medial-lateral (ML) of GRF. White R et al (1999)

concluded that the most repeatable component was the vertical component of GRF in

intra-day and intra-subject testing, and the least was medial-lateral component.

Therefore, the vertical GRF was investigated in this study.



The normal stance phase pattern of vertical forces generated by the customary

walking speed of 82 m/min has two peaks separated by a valley (Winter DA, 1991).

The first peak (F1) occurs at the onset during loading response. At this time the

body’s center of gravity is rapidly dropping; an action that adds the effect of

acceleration to the body weight. In late mid stance, the valley (F2) is created by rise

                                         - 34 -
of the center of gravity as the body rolls forward over the stationary foot. The second

peak (F3), occurring in late terminal stance, indicates downward acceleration and the

lowering of the center of gravity as body weight falls forward over the forefoot

rocker in terminal stance (Perry J, 1992).



These actions can be explained mathematically by following equation:

F = M (g + a)

where, F is the vertical GRF

       M is the mass

       g is the gravitational constant

       a is the vertical acceleration

The peaks and valley of the pattern of vertical GRF varies with acceleration




                       F1                    F3
 Force
 (N/Kg)




                                F2




 Fig 2.7: The vertical component of the GRF with 19 normal adult subjects
 (adopted from Winter DA, 1991)
                                         - 35 -
2.7.3.2 Joint Moment


Joint moments are the net result of a muscular, ligament and friction forces acting to

alter the angular rotation of a joint. The moments can be calculated by an inverse

dynamic method (Siegler S et al, 1998). As the joint angles do not reach their

extreme limits and the friction forces are minimal, it could be assumed that the net

moment could be due to the muscle forces only. External moment represents the

resultant external forces producing a joint rotation, whereas the internal moment

means the response of muscles to such external moment. The internal moment was

used throughout this study.



Hip and knee moments were quite varied (Winter DA, 1991), whereas Steinwender

G et al (2000) examined the repeatability of gait data that kinetic parameter was also

repeatable in the study. Therefore, the typical diagrams of hip and knee moments are

shown in figure 2.9.



Ankle moment was consistent: a small dorsiflexors moment at initial contact

followed by a plantarflexor moment increasing from the foot flat to the peak during

push-off (fig. 2.9).




                                        - 36 -
Internal

moments
                                                     % Gait cycle




Fig 2.9: The typical internal moment from 19 normal adults (Adopted
from Winter DA, 1991).




   Total

   Power
                                                     Gait cycle




   Fig 2.10: The typical total power of different joints from 19 normal
   adult subjects. (Adopted from Winter DA, 1991)
                                    - 37 -
2.7.3.3 Joint Power


Power is the rate of work done generated by the muscles. However, because of rapid

time-course changes, it has been necessary to calculate muscle power as a function of

time: Muscle power is defined as the product of net muscle moment and angular

velocity about joint (Winter DA, 1990, 1991).



Power = Moment x Angular Velocity (watts)



Power in our study was normalized with their body weight. Power can be either a

positive or negative. Positive power means the work generated from the muscle

during a concentric contraction and indicates the muscle moment acts in the same

direction as the angular velocity of the joint. Negative power means the work

absorbed during an eccentric contraction and indicates that the muscle moment and

angular velocity are in an opposing direction (Winter DA, 1990).



Hip power was quite varied across the subjects and the typical diagram is shown

(Winter DA, 1991) in figure 2.10.



Knee power: From initial contact to 15% of stride (gait cycle), the knee flexes under

control of knee extensors and results in K1, which is the first major absorption phase.

From 15% to 40% stride, the knee extends partially under control of quadriceps. This

concentric contraction results in K2, which is the only positive burst of power from

the knee extensors and represents only about 10%-15% of the energy generation in a

level walking. At 40% of stride, the knee starts to flex and continues into an early

                                         - 38 -
swing and produces a small K3 absorption burst. Finally, in the latter half of swing,

the knee flexors (hamstrings) turn, and they absorb most of the energy from the

swinging leg and foot during K4 burst (Fig 2.10).



Ankle power has two main bursts in the gait cycle: Absorption burst in stance and

generation burst in swing phase. From initial contact to 5% stride, the foot

plantarflexes under control of a small dorsiflexion moment. This results in a relative

small power absorption which is not labelled. From 5% to 40% of stride, the leg

moves over the foot under the control of an increasing plantarflexor moment and

produces energy absorption (A1). At 35% of stride, the rapid and active plantarflexor

moment has increased to cause heel off, so that a high A2 power burst is produced.

At about 60%, toe off occurs and the strong plantarflexor moments ends. The power

generation in association with an extreme small movement is maintained about zero

during the swing phase (Fig 2.10).




                                        - 39 -
2.7.4 Electromyography Parameters

Electromyography (EMG) records the myoelectrical signal in a period of time. Perry

J (1992) and Scribner G et al. (1992) suggested that electromyography data were

essential to show how the effect of time and status of the muscle firing during the gait

analysis. Appropriate interpretation of the timing (phasic) and intensity of effort can

identify the functional effectiveness of muscular activities. The period of muscle

activities in gait may be defined by three different reference scales: Gait cycle

interval, stance and swing periods, and the functional phases (Perry J, 1992).



Expressing EMG activities with respect to the gait cycle interval is the simplest

determination and is frequently used in the gait studies (Crenna P, 1997; Gage JR,

1991; Ounpuu S, 1995; Rab GT, 1994; Sutherland DH, 2001). The timing of four

major muscle groups are commonly conducted with surface electrodes in gait

analysis: Quadriceps, hamstrings, tibialis anterior and calf muscles (fig 2.11).

Furthermore, needle or wire electrodes are applied to interpret the phasic of EMG

signal from tiny muscles (Ounpuu S, 1995; Rab GT, 1994; Sutherland DH, 2001). An

example of EMG timing is shown in figure 2.12.



The amplitude of electromyography is also worthwhile to analyze, which represents

varying level of effort. As more muscle strength is required, the additional motor

units are recruited. Finsterer J (2001) reviewed that power spectrum analysis of EMG

including root mean square (RMS) and median frequency (MF) is one objective

method: The RMS represents the square root of the average power of the EMG signal

for a given period of time, and it found the strong correlations with force in the

                                         - 40 -
quadriceps (Karlsson S et al, 2001). It is one of the most widely used variables, along

with the mean absolute value, for measuring the amplitude of the EMG signal. The

MF is defined as that frequency that divides the power density spectrum into two

regions having the same amount of power; it may reflect the number of motion unit

recruitment and interpret the fatigue characteristics (Finsterer J, 2001; Karlsson S et

al, 2001; Ng JKF et al, 1996).



The power spectral analysis of the electrical signals produced by a contracting

muscle is emerging as a useful clinical tool. According to Kollmitzer J et al (1999),

both RMS and MF of rectus femoris muscle showed excellent repeatability within a

day, particularly correlation of MF between different trials were excellent (0.98). The

RMS and MF have been proven to have a good reliability and reproducibility in

many studies using dynamometer (Gerdle B et al, 2000; Goodwin PC et al, 1999;

Karlsson et al, 2001; Larsson B et al, 1999,2003).



Both RMS and MF has been used in studying flexion and extension of lower limbs

with dynamometer, which controlled the angular velocity of the quadriceps

(Ebenbichler G et al, 1998; Gerdle B et al, 2000; Karlsson S et al, 2001; Larsson B et

al, 1999,2003; Pincivero DM et al, 2000). Both RMS and MF value were also used

for spinal muscle (Lu WW et al, 2002), jumping motion (Goodwin PC et al, 1999),

and even for upper limbs motion (Dupont L et al, 2000; Feng CJ et al, 1998;

Serratrice G et al, 1995; Spaulding SJ et al, 1990).



For gait study involved EMG, only phasic of EMG was investigated (Rose J et al,

1999; Policy JF et al, 2001). The RMS and MF of the EMG have seldom been

                                         - 41 -
studied with gait analysis. It would be helpful for the clinician to have a

comprehensive understanding of EMG characteristic when the muscle firing RMS,

MF and duration were investigated.




                                     - 42 -
                                                  Gait Cycle


Fig 2.11: This showed the muscles firing (L) in the gait cycle. L1 and L2
indicate of first and second firing period. (adopted from Crenna P, 1997)



                        Initial Contact                                     Toe-off




 Fig 2.12: The EMG timing for muscles of normal children (adopted from
 Ounpuu S, 1995)


                                    - 43 -
2.8 Gait Studies with Orthoses


With the advancement of technology in the past 15 years, many authors have carried

out the gait analysis to study the functional performance of patients with different

orthoses. The reviews of gait studies on AFO and on DAFO are summarized below.




2.8.1 Gait Studies with AFO


2.8.1.1 Studies on AFO only


In 1996, Ribaudo T et al assessed the solid ankle foot orthosis (AFO) with kinetic

and kinematic data on 8 hemiplegic subjects (mean age = 28.5). They studied on the

differences between the braced (ipsilateral) and unbraced (contralateral) limb to

examine the orthotic effect on the contralateral limb, and with barefoot as a baseline.

The results showed that there were similar values for the ROM and kinetics

parameters of contralateral limb both with and without bracing of the ipsilateral limb

(P >0.05). A smaller contralateral knee flexion angle with ipsilateral bracing was

noted, but results were not significant (P = 0.07). The subjects did, however, exhibit

statistical differences in their stance and swing time duration on the non-braced side

comparing to braced side (P<0.01). Stance time and swing time of the non-braced

limb was increased by 15.9% and 9.1% respectively when using an ipsilateral AFO;

this implied that there was an increase in the total gait duration for the non-braced

side. The findings suggested that the use of ipsilateral AFO for neurologic or

musculoskeletal disorders posed no specific risks for the unbraced limb in terms of

kinetics and kinematics parameters.

                                         - 44 -
Thirty five diplegic CP with mean age of 8.7 were studied (Abel MF et al, 1998).

Eighteen CP wore braces to control equinus and 17 CP to control pes planovalgus

and crouch. Only the kinematics and kinetics were compared with the barefoot and

AFO. It was summarized that the AFO increased the velocity, stride length and ankle

moment in stance phase, but reduced in the abnormal power burst in early stance

phase (P<0.05). The benefits of AFO were the elimination of premature plantar

flexion and improved the progression of foot contact. The authors also stated that the

effects on proximal joint kinetics and kinematics were not significant. They

concluded that the main difference will be found at ankle joint.



Eighteen children (mean age was 8 years and 5 months) with different types of

hemiplegia wearing ankle foot orthosis was studied (Thompson NS et al, 2002).

They used the kinematic data to compare barefoot and AFO with different types of

hemiplegia. The AFO was found to improve the cadence, step length and walking

velocity. The author concluded that the AFO provided a better foot positioning at

initial contact for the patients but restrict their ankle motion.




2.8.1.2 Studies Compared AFO with Other Orthosis


Fourteen children with CP diplegia (mean age = 10.7, range: 6.9-16) were tested with

solid AFO, hinged AFO and posterior leaf-spring AFO (Smiley SJ et al, 2002). With

the shoe as a baseline, the kinetic and kinematic parameters were evaluated. It was

found that significant differences at ankle joint occurred at initial contact, maximum

dorsiflexion in stance, midswing, maximum dorsiflexion in swing, and toe-off. All

braces were found to have a significant difference at maximum dorsiflexion in stance.

                                           - 45 -
At maximum dorsiflexion in swing, there was a significant difference between the

hinged AFO than baseline. The authors concluded that each brace differed

significantly from baseline, which also contrast to normal, in other words: the braces

held the foot in abnormal dorsiflexion at initial contact, in the stance and swing phase.

None of the orthoses were significantly different from each other or from the shoe in

terms of velocity, cadence, and stride length. The similar findings in temporal

distance parameters comparing between orthoses from some authors (Radtka SA et al,

1997) who support the use of ankle-foot orthoses for various populations.



Middleton EA et al (1988) conducted a single-CP-subject study of age 4.5. They

employed a video system and Kistler force platform to compare the kinematic and

kinetic data between solid ankle foot orthosis (AFO) and hinged ankle foot orthosis

(HAFO). It was concluded that the HAFO was more effective than the AFO: The

subject exhibited a more natural ankle motion during the stance phase of gait, greater

symmetry of segmental lower extremity motion and decrease knee moments during

stance while wearing the HAFO. It seems that the orthosis with more range of

motion would have better gait.



In 1996, Lunsford T et al compared the AFO with 3 other hinged AFOs (three

subjects age from 5-12). Motion analysis revealed that the AFO was more capable

than all of the HAFO at restraining excessive terminal stance dorsiflexion. Mean

dorsiflexion at terminal stance for the rigid AFO was 13˚ versus three hinged AFOs:

17˚, 15˚ and 15.5˚ for patients respectively. Mean stride length provided by four

brace conditions was also very similar. Ground reaction force showed less knee

flexion moment during the loading response for the AFO compared to the HAFO. It

                                         - 46 -
was thought that a hinged AFO would permit a more normal transition from swing to

stance by avoiding the excessive knee flexion thrust caused by the AFO. EMG data

showed that quadriceps activity was increased during the terminal swing and loading

with a rigid AFO. It was suggested that the orthosis with less restriction on ankle

range allowed natural walking from swing phase to stance phase without an

excessive knee flexion thrust.



Patients (mean age = 11.42 and ranged from 6.46-20.08) walking on barefoot, with

Spring AFO and with AFO were studied by Brunner R et al (1998). The kinetics and

kinematics, but no EMG data were evaluated. Toe-heel-toe gait was significantly

improved the heel toe gait pattern by using both types of ankle foot orthoses. Spring

type AFO was significantly superior to the AFO; this could be expressed in the

cadence, velocity and step length. Kinetic data gave a significant improvement with

any type of AFO. Although little difference between orthoses was found, the author

observed almost normal rocking of foot and suggested that a spring type of orthosis

rendered the gait more dynamic and best corrects the pathologic gait. It may further

support the hypothesis that the more dynamic, the better in gait improvement.



In 2001, Buckon CE et al compared hinged AFO (HAFO), posterior leaf spring AFO

(PLS) and solid ankle foot orthosis (AFO). Thirty patients (mean age 9 years 4

months) with hemiplegia were participated in gait analysis. The HAFO and PLS

increased the passive ankle dorsiflexion and better ankle rocker pattern during gait,

but AFO did not. The authors recommended that the HAFO was the most effective in

controlling the knee stability. As HAFO and PLS allow more ankle motion than the

AFO, they provide a better ankle rocker function.

                                        - 47 -
A retrospective review of 115 patients with CP (diplegia n = 97; hemiplegia n = 18)

on temporal distance parameters was conducted by White H et al (2002). They

compared the parameters for the following conditions: barefoot, with AFO, with

Hinged AFO. The results showed that parameters of velocity, stride length, step

length and single limb stance were significantly increased (P<0.01) with the use of

AFOs versus the barefoot walking. Cadence was the only parameter found not to be

statistically different. They also theorized that AFO provided increase stability to the

ankle that allowed for longer step and stride length. Therefore, some authors

(Diamond MF et al, 1990; Dieli J et al, 1997; Radtka SA et al, 1997) also revised that

the longer step or stride length was the orthotic benefit.



From the gait studies, the AFO was recommended to be useful because this provided

the stability on ankle joint and controlled equinus. More importantly, AFO has main

influence on ankle joint. To summarize the studies among the ankle foot orthosis

(AFO), hinged ankle foot orthosis (HAFO) and posterior leaf-spring orthosis (PLS):

Orthoses allowing more ankle ROM than AFO, like HAFO and PLS, could provide a

more natural gait and better ankle rocker function.




                                          - 48 -
2.8.2 Gait Studies on DAFO



The dynamic ankle foot orthosis (DAFO) is particularly prescribed for the patients

with dynamic equinus deformity. To achieve this, the DAFO aims to alter the sensory

input on the foot reflex, so that the equinus can be controlled. In 1990, Diamond MF

et al conducted a case study that compared walking with barefoot, ankle foot orthosis

(AFO) and dynamic ankle foot orthosis with plantarflexion stop (P-DAFO). The

subject was 32-year-old who had suffered from stroke, and temporal distance

parameters were evaluated. There was a significant improvement in walking velocity,

step length, and stance time and a significant decrease in cadence when either AFO

or P-DAFO condition was compared with barefoot. The subject also reported that

P-DAFO was more comfortable and less restrictive than AFO. It was suggested that

P-DAFO improved the recruitment and sequencing of muscle activity, providing an

external stability with correct biomechanical alignment so that the patient no longer

need to use an abnormal muscle activity to compensate for lack of stability. In

addition to the patient’s view, although there is no functional difference between

AFO and P-DAFO, the P-DAFO may be preferred. Evaluation with a larger sample

size is needed.



The effect of DAFO on the foot-loading pattern of a stroke patient was studied

(Mueller K et al, 1992). They investigated total foot force, total foot area and total

foot contact time with A-B-B-A single subject design. The testing conditions were:

shoe-1 (baseline-1), DAFO-1, shoe-2 (baseline-2) and DAFO-2. The subject was

instructed to wear the DAFO during all walking hours for next 14 days, shoe-2 and

                                        - 49 -
DAFO-2 was tested after 14 days. The total stance duration was significantly

decreased and total foot area was increased as a result of wearing the DAFO,

suggesting prevention of spasticity-induced foot postures. The author concluded that

DAFO resulted in increasing improvements over time and suggested that kinematics

and EMG with larger samples can also be investigated.



Radtka et al. (1997) compared the solid AFO with the plantarflexion stop type of

DAFO (P-DAFO) in cerebral palsy children (4 hemiplegia and 6 diplegia, mean age =

6.5, range from 3.5-8.5) and provided a good foundation for the comparison of

P-DAFO (which have the same proximal trimline as AFO) with rigid ankle foot

orthosis (AFO) and with baseline (barefoot). They showed the effect on the

temporal-distance, range of motion and electromyography parameters with AFO and

P-DAFO with plantarflexion stop. Both orthoses significantly increased stride length,

reduced cadence and excessive ankle plantarflexion, and there was a significant

premature muscle firing in calf muscle. They concluded that although no differences

between the two orthoses were found, patients and their relatives also preferred to use

P-DAFO with plantarflexion stop due to cosmetic effect and light in weight. This

conclusion is quite similar with the previous studies with Diamond MF et al (1990).

Furthermore, kinetics and comprehensive EMG parameters can be conducted in the

future study.



In the studies done by Carlson WE et al (1997), they measured the kinetics, kinematics

and time-distance parameters in 11 CP patients with diplegia and compared the

difference among the patients with shoe only, AFO and DAFO. They found both

orthoses significantly reduced an ankle excursion and improved plantarflexion angle.

                                         - 50 -
However, it was found that the supramalleolar type of dynamic ankle foot orthosis

(DAFO) did not provide an effective control of dynamic equinus in children with

cerebral palsy in considering the maximum power at push-off phase as AFO did. The

authors reviewed their small sample size, the ANOVA with repeated measurement

design was sufficient to provide the necessary statistical power to draw valid

conclusion. They found no orthotic effect in the temporal distance parameters, which

was contradictory to previous studies (Diamond MF et al, 1990; Mueller K et al,

1992; Radka et al, 1997). Therefore, more evidence should be provided to determine

whether temporal distance parameters were altered by wearing orthosis.



Dieli J et al (1997) compared barefoot, AFO and DAFO with plantarflexion stop on

three hemiplegic patients with stroke, using footswitch stride analysis. The results

displayed a significant increased walking velocity and longer stride length, but

reduced cadence and delayed time of single-limb support on the affected side when

using both orthoses. The DAFO group showed the highest velocity. They

recommended that DAFO to be an alternative treatment to conventional

thermoplastic orthosis (AFO) on patients with stroke.



In 2000, Crenshaw S et al compared the effects of hinged AFO (HAFO), DAFO with

plantarflexion stop (P-DAFO which has the same proximal trimline as AFO),

supramalleolar DAFO (DAFO) on patients with diplegic CP (n=8, CP diplegia). In

the temporal-distance parameters, only the stride length between the shoes-only

condition and P-DAFO had a significant difference. The orthoses improved the

positioning at initial contact compared to baseline; whereas the DAFO showed a

significant more plantarflexion at toe-off and maximum plantarflexion at swing

                                        - 51 -
phase than the other orthoses. For the kinetics parameters, the majority of significant

changes occurred at the ankle, the DAFO produced lower maximum moment than

the HAFO and P-DAFO conditions; The HAFO condition also produced a significant

higher value of peak power absorption than the baseline and DAFO. Such difference

may be due to the HAFO and P-DAFO having a longer lever arm and providing a

better positioning to start third rocker motion of metatarsophalangeal joint than the

DAFO.



The supramalleolar DAFO and hinged AFO were compared (Romkes J et al, 2002)

in twelve hemiplegic CP patients (mean age was 11.9 ± 4.9 years). Stride length and

step length increased significantly when compared with barefoot walking when using

the DAFO and HAFO, but there was no significant difference between the two

orthoses. Both orthoses significantly improved the excessive plantarflexion at swing

phase and plantarflexion moment of ankle than the barefoot. Furthermore, two

orthoses produced a positive change towards normal at ankle power absorption. The

difference between DAFO and HAFO was not significant, but more normal value

was produced by HAFO. Therefore, authors suggested that DAFO did not improve

gait function as effective as HAFO. This could be explained that the reduced

lever-arm of the DAFO could not be fully compensated by the tone-reducing

foot-plate.



The DAFO has been compared with AFO, P-DAFO and HAFO. Some studies

showed that the DAFO seems to be less effective than other AFO, P-DAFO or

HAFO comparing different type of parameters, because of shorter posterior trimline

than AFO, P-DAFO or HAFO. However, this shorter trimline would actually allow

                                         - 52 -
more ankle motion, and hence more natural gait would be produced. Furthermore,

from the clinical experience, such short trimline of DAFO is undoubtedly increasing

the patient compliance on the orthotic treatment, especially in the adolescent age.



EMG data have been rarely interpreted in gait studies of orthosis; the root mean

square value (RMS) and median frequency (MF) could be applied to evaluate the gait

EMG because of common use in EMG study and good reliability (Gage JR et al,

2001; Kay RM et al, 2000; Chambers HG et al, 1997). By combining EMG,

kinematic and kinetic interpretation is to determine different orthotic treatments.




                                         - 53 -
                                 CHAPTER 3


                   MATERIALS AND METHODS


3.1 Subjects



Two groups of subjects were recruited for the study. There were 16 patients with

cerebral palsy (CP) and 18 healthy normal control (normal) subjects. The study was

approved by the appropriate ethical committee of The University of Hong Kong, and

all subjects gave informed consent before participation (Appendix 3.1 – 3.2). For the

healthy participants, their barefoot walking was performed. For the CP patients, three

sessions of data collection in random order were done on the same day: barefoot

(group 1), AFO (group 2) and DAFO (group 3). A minimum of three individual gait

cycles were averaged for each condition; if more cycles were available, they were

included in the averages. Thus, for each condition studied, at least three successful

walking trials (including only one foot completely on a given force platform) for

each of the left and right sides were calculated and then combined to a single mean. 1)

An independent t-test was used to compare CP patients’ barefoot and normal’s

barefoot data. 2) To compare the treatment effect, ANOVA with repeated

measurement, which is a powerful statistical test to minimize the intra-subject

variability (Steinwender G et al, 2000) among the measurements were used.




                                        - 54 -
3.1.1 Cerebral Palsy Group

Seven girls and nine boys (mean age = 6.83 years, ranged from 3.3 to 13.7 years)

were recruited from the Duchess of Kent Children's Hospital (DKCH) and the

department of Prosthetics and Orthotics in Maclehose Medical Rehabilitation Center

(MMRC), and all of them followed the same treatment protocol (App 3.3).

Participants with dynamic equinus deformity of moderate level at calf muscles

(modified Ashworth scale 1-3, Table 2.1) were recruited and the most severe group

(modified Ashworth scale 4) is considered to have contracture already. Patients with

significant observed coronal or rotational deformities were excluded because these

deformities may alter the motion in sagittal plane. The subjects had good vision and

were able to comprehend instruction and walk independently. All of them had no foot

and ankle surgery done. If the subjects had previous Botox injection, they were

recruited at least 5 months after the injection so that the drug effects would not affect

our investigation. General information of the subjects is summarized in table 3.1.




3.1.2 Normal Group

Subjects were recruited from the St. Paul Primary School, and they did not have any

symptomatic problem on walking. One female and seventeen male participated to

form a healthy control group and the mean age was 7.17 years (range form 6 to 9.7

years). General information of the subjects is shown in table 3.2.




                                          - 55 -
                                        muscle tone
                                        hamstring Calf
No. Sex Age        weight(kg) height(cm) L       R    L   R
1      F   5.3     16.1      107        2        2    1+ 1+
2      M   5.6     15.3      106        1        1    3   3
3      M   6.5     15.3      106        1+       1+   1+ 1+
4      F   6.2     24.2      115.2      1        1    1+ 1+
5      M   4       13        94.5       1        1    3   3
6      F   8.1     27.5      133.6      0        0    1   1
7      M   3.4     13        94.5       1+       1+   3   3
8      F   3.3     14.9      95         0        0    1   2
9      M   7.3     18.7      113.4      0        0    1   1
10     M   6.6     26.4      115.1      2        2    1+ 1+
11     F   6       16.4      106        1        1+   2   2+
12     F   5.5     19        109        0        0    1   1
13     M   9.7     28.5      124.8      0        0    1+ 1+
14     M   8.1     20.4      112.6      0        0    1+ 2
15     M   13.70 45.5        154.2      0        0    1+ 2
16     F   10      28.9      125.8      0        0    1   1
Mean       6.83    21.44     113.29
SD         2.68    8.49      15.58

Table 3.1: General information of cerebral palsy group




                                        - 56 -
Nos.        Sex          Age       Height (cm) Weight (kg)
1           M            6.9       120           245
2           M            7.3       123.5         197
3           M            7         129           215
4           M            7.5       133.2         314
5           F            9.1       155.1         415
6           M            6.8       130           276
7           M            6.7       115.9         192
8           M            6.8       123.2         245
9           M            6.8       131.3         271
10          M            7.1       123.8         195
11          M            6.7       125.2         268
12          M            6         114.7         266
13          M            6.3       121           230
14          M            6.1       115.1         206
15          M            7.3       112           206
16          M            9.7       129.8         335
17          M            7.8       123.5         265
18          M            7.2       115           205
Mean                     7.17      124.52        252.56
SD                       0.94      9.94          57.97

Table 3.2: General information of normal group




                                     - 57 -
3.2 Orthoses



All CP participants followed the study protocol (App 3.3). Two pairs of orthoses,

were custom-made by the orthotists, followed the fitting guideline of the orthoses (no

pressure point and immediate complaint after wearing 10-15 minutes) at the

Maclehose Medical Rehabilitation Center. The details of the orthoses are as follows:




3.2.1 AFO

The AFO with neutral position (foot-shank ankle was 90 degree) was fabricated from

4.8mm thick polypropylene extending distally under the toes and on the mediolateral

border of the foot, its proximal trimline of the posterior part of the leg extended up to

about 2.5cm below the fibula head. The trimline was just over the tip of malleolli

(Fig 3.1).




3.2.2 DAFO

The DAFO (originated from Company of Cascade Prosthetics and Orthotics, 1998),

with a contoured footplate, was made from 2.4 mm thick polypropylene enclosing

the dorsum of the forefoot and ankle; and covered the lateral part of the leg to about

5 to 7.5 cm above the malleolli (Fig 3.2). The modification of the footplate was

demonstrated in the figure 3.3. Such modification tried to alter the sensory input of

the foot reflexes in order to control the dynamic equinus deformity.




                                          - 58 -
   Anterior view of AFO                      Lateral view of AFO
Fig 3.1: A picture of AFO: Anterior and lateral views.




   Anterior View of DAFO                 Lateral View of DAFO

Fig 3.2: A picture of DAFO: Anterior and lateral views.




                                        - 59 -
                                                      6                  5    1


                  Dotted line



                                         3
     6
                                                      4
                                                               1




              2


                                5



Fig. 3.3: Top left: The footplate for dynamic ankle foot orthosis (DAFO), Top right:

The cross section view from the dotted line. The bottom: the top-side view of the

foot plate.

Build-up area under the toes (1), lateral arch (2), medial arch (3) and transverse

metatarsal arch (4). Recessed areas under the metatarsal fat pad (5) and calcaneal fat

pad (6). (Adopted from Radtka et al (1997))


                                        - 60 -
3.3 Apparatus



The apparatus used in this study consisted of a motion analysis system, a force

platform and a surface electromyographic system. The data from all systems were

synchronized and collected in the computer. The details of each system are described

below.




3.3.1 Motion Analysis System

The motion analysis system consisted of 6-camera with infra-red light strobe.

(Oxford Metrics Ltd, Oxford, UK), sampling at 60Hz (Fig 3.4), was applied to

acquire the kinematic information. The cameras were aligned in a circular fashion

and the angle between any two cameras’ optical axes was at least 45 degrees. The

layout of the environment is shown in figure 3.5. A total of 15 passive spherical

retroreflective markers (diameter 25 mm, fig 3.6) were attached according to the

original marker set from Helen Hayes (Fig 3.7, Davis RB et al, 1991). When the

subject walked along the walkway, the infra-red signal strike on the markers were

reflected and captured by the cameras. The captured images of each marker were

recorded in the computer in the form of three dimensional coordinates (Fig 3.8). Both

static and dynamic calibrations were performed before any capturing to determine

the position and orientation, and to minimize the lens distortion of each camera.




                                         - 61 -
3.3.2 Force Platform

A 40cm x 60cm piezoelectric force platform (Kistler Instrument AG, Winterthur,

Switzerland), sampling at 60Hz, was embedded inside the floor to prevent noise. The

platform which embedded under ground and covered with large carpet is to avoid

reflecting infra-red light from the floor which may affect marker recognition of

motion analysis system, and prevent the notice of the location of force platform

which may lead to an unnatural gait. Instantaneous ground reaction forces acting on

subject’s foot were recorded during capturing (Fig 3.5).




3.3.3 Electromyographic System

An electromyographic system (Telemg, BTS Inc, Milan, Italy) was used for surface

EMG measurement in this study. This consists of 8 pairs of electrodes, 8 channels

pre-amplifiers, patient unit, an optical fiber cable and a control unit. Each pair of

surface EMG electrode (Ag-AgCL Disc electrodes, φ=2cm) was applied to skin

cleaned with an alcohol swab, and the impedance was always controlled at less than

10 kΩ. The inter-electrode distance was 3cm. The electromyographic signal was

detected by the electrode and was differentially amplified by the pre-amplifier (Input

impedance > 1 MΩ and common-mode rejection ratio >100dB). The pre-amplifier

(Fig 3.10) was filtered with cut off frequency of 240 Hz so as to reduce the

interference from electrical appliances.



The signal from pre-amplifier then transmitted into the 8-channel EMG patient unit,

the EMG signal was then converted from analog to digital data inside the patient unit


                                           - 62 -
(Fig 3.9) before transmitted to the condition unit through optical fiber cable. It should

be noted that two optical fibers were enclosed in the optical fiber cable (Fig 3.11):

Infra-red radiation at 820nm generated from the condition unit was transmitted to the

patient unit through one of the fiber, and the EMG signal collected by the control unit

via another fiber. Such optical transmission of EMG signal allowed a complete

galvanic isolation of the subject from the system and avoided electric currents

generation when the metallic type of electric cable was applied.



Within the condition unit, the EMG signal was digital-to-analog converted and

transmitted to the computer. Eventually, in the computer, the EMG signal was

analog-to-digital converted again for storage. During acquisition, the on and off time

of capturing EMG data were synchronized with that of the Vicon System. Eight pairs

of surface electrodes were placed on lower limb muscles -- over the left and right

side of quadriceps, hamstrings, tibialis anterior and calf muscles. The surface

electrodes were placed in the mid-point of the muscle belly and parallel to the muscle

fiber, so that the maximum signal can be collected (Cram JR, 1998; Tosh RW,

1991).The real time monitoring of the EMG signal to detect any detached electrodes

or artifact during every walking trials. The overall transmission of the EMG signal

was illustrated in the figure 3.13.




                                          - 63 -
Fig. 3.4: Camera with infra-red light strobe.
             C
                 C




                                                                                            C
                                                                                                C




             C                                                                      C
                 C                                                                      C




                                                 F o r c e p la t f o r m




                                                    w a lk w a y




     C
         C



                                                                            C
                                                                                C




                                   EMG
                     V ic o n   a m p lif ie r




Fig 3.5: The layout of the Gait Laboratory in DKCH: Six cameras were mounted at

wall and force platform was embedded in the middle of the walkway.


                                                 - 64 -
  Fig 3.6: A total 15 retroreflective markers.




                                                     PSIS
     R. ASIS                      L. ASIS




R. M. Thigh                         L. M. Thigh

 R. Knee
                                L. Knee

R. M. Shank                         L. M. Shank



     R. Ankle                    L. Ankle

      R. Toe                    L. Toe



  Fig 3.7: The marker set from Helen Hayes.




                                            - 65 -
Fig 3.8: Computer’s monitor displaying video image of the subject.




Fig 3.9: The electrodes and patient unit




                                           - 66 -
Fig 3.10: The electrode and cable




                                        Fig 3.11: The Optical fibers of the EMG.




Fig 3.12: The condition unit (TelEMG ) system for EMG data collection.



                                      - 67 -
      EMG signal from muscle




          EMG channels 1 to 8                   EMG electrodes




   Low passed filtered and amplified             Pre-amplifier




          A/D convention                        EMG Patient Unit



                     Optical fiber cable


          D/A convention                        EMG condition unit




    A/D convention for storage                  Computer



Fig 3.13: The summary of EMG circuit




                                       - 68 -
3.4 Experimental Procedures



Prior to the gait analysis, static and dynamic calibrations of the motion analysis

system were performed to ensure minimal missing data. For CP subject, the

spasticity of the muscles (hamstrings and calf muscles) was measured by the

physiotherapists at DKCH. The gait analysis was then performed and divided into

two phases: Preparatory and data acquisition phases. Six categories of gait

parameters were obtained after data reduction.




3.4.1 Calibration of the System

Both static and dynamic calibration of cameras aimed to define an absolute spatial

coordinate system in the laboratory.



The static calibration was performed using an L-frame with four reflective markers.

The L-frame should be properly placed over the corner of the force plate which

defined as the origin of force plate system (Fig 3.14). The distance of the reflective

markers (diameter = 2.5cm) from the reference point was shown in the figure 3.15.

The image from the markers of the L-frame was captured by the cameras so that

coordinates and relative distance could be calibrated.



The dynamic calibration was performed by the calibration rod, and the distance

between the two reflective markers (Diameter = 5 cm) was 50 cm (Fig 3.16). Moving

the calibration rods and walking around the force platform create moving images for


                                         - 69 -
the cameras (Fig 3.17). To improve calibration, increase the walking speed and

distance from the force platform to simulate different orientations and speeds. The

cameras captured the moving images and calculated the overall residual

(automatically calculated by the Vicon System) of the image which detected the

acceptance level. If the residual of any camera was greater than two which was

recommended by the manufacturer, another set of calibration should be performed.




                                       - 70 -
Fig 3.14: L-frame for static calibration placing on the corner of force platform.




      d




             a        b                                                 c




Fig 3.15: The diagram of L-frame. The distance of reflective markers from reference

points a = 9.5 cm, b = 88 cm, c = 439 cm and d = 300cm.




                                         - 71 -
                                                  Markers distance


Fig 3.16: The calibration rod for dynamic calibration. Diameter of markers is 5 cm

and distance between markers is 50 cm.




Fig 3.17: Perform the dynamic calibration with calibration rod.




                                         - 72 -
3.4.2 Gait Analysis

Gait Analysis was divided into two different phases: Preparatory phase and data

acquisition phase.




3.4.2.1 Preparatory Phase


Positioning of reflective markers, anthropometric measurements and EMG electrodes

placement was accurately prepared. The accuracy and validity of every trial depends

upon accurate and appropriate placement of markers and surface electrodes.



1) Marker placement: The subject should wear a short swimming pant to attach 15

reflective markers on their lower limbs with double adhesive tape (Fig 3.18). The

landmarks developed by Helen Hayes (Fig 3.7) were mid-point of posterior superior

iliac spine (PSIS), anterior superior iliac spine (ASIS), mid-thigh, knee, mid-shank,

lateral malleolli of ankle joint, the heel and the head of second metatarsal. These

specific landmarks were used to locate the joint centers of the hip, knee and ankle.

The markers should be placed on the corresponding position when wearing orthosis

(Fig 3.19) (Buckon CE et al, 2001; Crenshaw S, 2000; Radtka SA et al, 1997;

Romkes J et al, 2002). The anthropometric measurements such as body weight,

height, leg length, width of ASIS, knee and ankle width were obtained for calculation

in the computer (Fig 3.20, Fig 3.21).



2) Electrode placement: Before attaching the EMG electrodes, the skin preparation

including washing, shaving, wiping and abrading is needed to lower the skin


                                        - 73 -
resistance and have better EMG recordings. The rectus femoris (one of the

quadriceps), hamstrings, tibialis anterior and calf muscle were investigated in this

study (Fig 3.22). The pair of electrodes was placed over the middle of the muscle

along the direction of the muscle fiber and using double adhesive tape with an

inter-electrode spacing of 3 cm. Conductive gel was used to increase conductivity of

EMG. The muscle signal collected from electrode was connected to the patient unit

by cables which also incorporate the pre-amplifier within (Fig 3.23). The patient unit

was secured over the subject’s back and the optical fiber cable connected the patient

unit with the condition unit. Subject preparation was showed in the figure 3.24 and

fig 3.25.




   Fig 3.18: Attach reflective                    Fig 3.19: Attach reflective
   markers on subject with                        markers on subject with AFO.
   barefoot.




                                        - 74 -
Fig 3.20: Measuring the ankle width




 Fig 3.21: Measuring the distance between ASIS




                                      - 75 -
Fig 3.22: The electrode placement on the muscles: Rectus femoris (Top left),

hamstring (Top right), tibialis anterior (Bottom left) and calf muscles (Bottom right).

The dark spot devotes surface electrode (Cram JR et al, 1998)




                                         - 76 -
Fig 3.23: The procedure of electrode
placement and cable connection.




Fig 3.24: Anterior view of the            Fig 3.25: lateral view of the
subject after preparation of              subject after preparation of
markers and electrodes                    markers and electrodes



                                 - 77 -
3.4.2.2 Data Acquisition Phase

Data acquisition consisted of static standing trial and walking trial. Prior to walking

trial, the static trial was required for the calculation of the offset values on the angle

parameters. This static trial was used to calibrate certain internal axes of the limb

segments by placing the extra markers on the subject, in addition to those target

markers worn during a walking trial.



1) Static (standing) trial: Stand the subject in the center of the calibration volume

and checked each of camera views to assure every marker could being seen and that

a strobe was not obliterating markers. While the subject was standing quietly in the

center of the calibration volume (in position of force platform), three seconds of

motion data were collected. Reconstructed and viewed these data immediately to be

positive which was at least three frames of data where every marker was visible. If

the standing data were severely fragmented, another static trial should be performed.



2) Walking Trial: For each session, the subject was asked to walk along the

walkway with their natural walking speed without a previous notice of the force

platform. The investigator observed and optimized the best starting position to ensure

the foot of subject strike on the force platform in each walking trial. With

approximately two minute break provided between 2 successive trials, at least three

successful walking trials for each limb were recorded for averaging. Eventually, data

acquisition of camera, force platform and EMG was synchronized within the

computer. For CP subject, they received three sessions of walking trial with the

random sequences: Barefoot, with AFO and with DAFO. For normal subject, only a

session of barefoot was performed.


                                          - 78 -
3.4.3 Data Reduction

The raw data from the cameras and force platform was reduced with the Vicon

Clinical Manager (VCM) software, which has been based on the biomechanical

model proposed by Kadaba MP et al (1990) and Davis RB et al (1991), for definition

of the gait cycle (Fig 3.26) and gave kinematics and kinetics data. The raw data from

the surface EMG was analyzed by MatLab based program (Math Works Inc., Natrick,

MA). The procedure of data reduction was in the followings.




3.4.3.1 Procedure of Kinematics and Kinetics Analysis


In the first stage, the 3D data file was checked for the presence of the marker

trajectories. Gaps in trajectories were permitted, but if any trajectory was very

broken up, a warning message was displayed.



In the second stage, the spatial position and orientation of each segment was

computed from the trajectory marker positions, at 2% steps through the defined gait

cycles. This process of normalization involved interpolation and smoothing of the 3D

trajectory marker.



In the third stage, the joint angles were calculated from the absolute and the relative

orientations of the segments axes (Davis R et al, 1991). Joint moment and power

were calculated from the segments masses, forces and orientation (Winter DA, 1990).

In which, the values of joint moment and power were normalized with bodyweight.




                                         - 79 -
Fig 3.26: Defined gait cycle with VCM program




Fig 3.27: Abstract firing segment with red line.




                                         - 80 -
3.4.3.2 Procedure of EMG Analysis


It is well established that the amplitude of the EMG signal is a stochastic (random)

with zero mean in nature and can be reasonably represented by a Gausian distribution

function, with a program of MatLab (Fig 3.27).



In this program, the muscle firing segments were manually determined in figure 3.27.

The segment of muscle firing was calculated by determining the starts and the end

muscle activity. These segments were also used to calculate muscle contracting root

mean square (RMS) value and median frequency (MF).



After the raw EMG signals were adjusted to zero mean in order to remove the offset

from the amplifier, the root mean square value (RMS) of each interval was then

calculated by the formula:
                1
    RMS =
                n
                        ∑x          2
                                        (i )           (1)

where n is the total number of samples within the window considered for processing

and x is the amplitude of the raw EMG. The RMS value during walking was

normalized with their RMS value during standing.



After spectral analysis, the median frequency (MF) was calculated with the formula:
                               fc
     f median                       2

      ∑ P( f    k   )          ∑ P( f
                            k = f median
                                               k   )
      k =0              =                               (2)

where P(f) is the power density spectrum, fK the frequency and fc the sampling

frequency while fmedian the median frequency. The median frequency is defined as

that frequency that divides the power density spectrum in two regions having the

                                                              - 81 -
same amount of power.



To sum up, the muscle firing duration was measured manually, and both root mean

square (RMS) and median frequency (MF) of useful segments were calculated, and

eventually saved in Microsoft Excel file. It was illustrated in figure 3.38 that the start

and end of the muscle firing duration, it showed a single continue muscle firing

period which overlapped two individual gait cycles. We assumed that the first and

second period of muscle firing segment of this gait cycle should be the same as the

next gait cycle. It is therefore, the RMS and MF in the same gait cycle were averaged,

and the total contraction duration was summated to a single value throughout this

study.




                                          - 82 -
3.5 Measured Parameters



By combining data with the camera, force platform and EMG systems. Six categories

of parameters are commonly used for clinical gait evaluation, and they are consisting

of 1) temporal distance parameters, 2) joint angle, 3) ground reaction force, 4) joint

moment, 5) joint power and 6) electromyography. These variables were discussed in

the literature review (session 2.7) and shown in the follows.



1) Temporal distance parameters

Temporal distance parameters included stride length, stride time, walking speed,

stance time, swing time, stance/swing ratio and cadence.



2) Joint angle

Angle patterns from hip, knee and ankle joints along the gait cycle were calculated

by the Vicon Clinical Manager (VCM) system, with convention suggested by Winter

DA (1991). For all joint angles, the maxima, minima and their corresponding times

of the occurrence in a gait cycle (Fig 3.28-3.30) were measured. In particular, for hip

and ankle angles, the difference between the maxima and minima was also measured.



3) Ground reaction force

The vertical component of the ground reaction force was investigated. The

parameters included first peak, valley, second peak and their corresponding times of

the occurrence within a gait cycle (Fig 3.31)




                                         - 83 -
4) Joint moment

The moments of the hip, knee and ankle joints were evaluated by solving the

equations of motion for the six segments of the lower limbs (Kadaba MP et al, 1987)

The details were reviewed in the chapter 2.7.3.2. This medical convention, also

called internal joint moments, was generated by active muscles. For each joints, the

maxima, minima and corresponding time of the occurrence within a gait cycle was

reduced as the parameters (Fig 3.32- 3.34).



5) Joint Power

The total powers of the hip, knee and ankle joints were made by solving the

equations for the six lower limbs segments (Kadaba MP et al, 1987). Power was

calculated as the scalar product of moment and angular velocity and the details were

in chapter 2.7.3.3. Positive power meant the muscle was undergoing the concentric

contraction, the rate of work done generated by the muscle, and vice versa. For each

joint, the maxima, minima and the corresponding times of the occurrence within a

gait cycle were determined (Fig 3.35- 3.37).



6) Electromyography

The starting point and end point of muscle firing phase of each muscle channel were

obtained in the figure 3.38. For each of the muscle firing duration, the length of

firing duration, firing root mean square (RMS) and firing median (MF) was

calculated with the MatLab programme. Firing duration was defined as end of

muscle firing minus start of muscle firing; The RMS represented the averaged power

generated by the muscle; The MF reflected the pattern of the motor unit recruitment.


                                        - 84 -
If a muscle group had two contracting durations within gait cycle, the RMS and MF

were averaged, and total contraction duration was summated to a single value.




                                       - 85 -
                                                                                    HA3
                             HA1
                   Flexion
          (Degree)




                                                                  HAT2
  Extension




                                                                                     HAT3

                                                                  HA2     Toe-Off



  Fig 3.28: Definition of hip angle parameters.




Legends:


HA1 (degree)                       Hip angle at initial contact
HA2 (degree)                       Maximum hip extension angle at stance phase
HA3 (degree)                       Maximum hip flexion angle at swing phase
HAT2 (%GC)                         Time of occurrence at HA2
HAT3 (%GC)                         Time of occurrence at HA3
HA3-HA2 (degree) Total range of hip motion



Note: GC = Gait Cycle



Data source: Build-in normal curve from Vicon

Definition method was modified by Benedetti MG et al (1998)



                                                       - 86 -
                Flextion
         (Degree)
  Extension




                                        KAT3

                                                  Toe-Off


  Fig 3.29: Definition of knee angle parameters




Legends:


KA1 (degree) Knee angle at initial contact
KA2 (degree) Maximum knee flexion angle in the stance phase
KA3 (degree) Minimum knee angle at stance phase
KA4 (degree) Maximum knee flexion angle in the swing phase
KAT2 (%GC) Time of occurrence at KA2
KAT3 (%GC) Time of occurrence at KA3
KAT4 (%GC) Time of occurrence at KA4



Note: GC = Gait Cycle




                                       - 87 -
                    dorsiflexion



                                                                     Toe-Off
             (Degree)




                                                                               AA4
  Plantarflexion




                                                                               AAT4




Fig 3.30: Definition of ankle angle parameters




Legends:


AA1 (degree)                       Ankle angle at initial contact
AA2 (degree)                       Maximum dorsiflexion in stance phase
AA3 (degree)                       Maximum plantarflexion in terminal phase
AA4 (degree)                       Maximum dorsiflexion during swing phase
AAT2 (%GC)                         Time of occurrence at AA2
AAT3 (%GC)                         Time of occurrence at AA3
AAT4 (%GC)                         Time of occurrence at AA4
AA2-AA3 (degree) Total range of ankle motion



Note: GC = Gait Cycle




                                                       - 88 -
   Vertical component




                                                         Toe-Off


  Fig 3.31: Definition of ground reaction force parameters.




Legends:


GR1 (%BW) First peak during loading response
GR2 (%BW) Valley at the stance phase
GR3 (%BW) Second peak at the terminal stance
GRT1 (%GC) Time of occurrence at GR1
GRT2 (%GC) Time of occurrence at GR2
GRT3 (%GC) Time of occurrence at GR3



Note: BW = Body Weight; GC = Gait Cycle




                                       - 89 -
      Extensor
  (Nm/Kg)
  Flexor




                                                              Toe-Off


     Fig 3.32: Definition of internal hip moment parameters




Legends:


HM1 (Nm/BW)        Maximum hip extensor moment at stance phase
HM2 (Nm/BW)        Maximum hip flexor moment at stance phase
HMT1 (%GC)         Time of occurrence at HM1
HMT2 (%GC)         Time of occurrence at HM2



Note: BW = Body Weight; GC = Gait Cycle




                                        - 90 -
                                   Knee

 Extensor




                                                       Toe-Off

  Fig 3.33. Definition of internal knee moment parameters.




Legends:


KM1 (Nm/ BW) Maximum knee extensor moment at stance phase
KM2 (Nm/ BW) Maximum knee flexor moment at stance phase
KMT1 (%GC)        Time of occurrence at KM1
KMT2 (%GC)        Time of occurrence at KM2



Note: BW = Body Weight; GC = Gait Cycle




                                       - 91 -
                                      Ankle

    Dorsiflexor




                                                               Toe-Off


   Fig 3.34: Definition of internal ankle moment parameters.




Legends:


AM1 (Nm/BW)       Maximum dorsiflexor moment in stance phase
AM2 (Nm/BW)       Maximum plantarflexor moment in stance phase
AMT1 (%GC)        Time of occurrence at AM1
AMT2 (%GC)        Time of occurrence at AM2



Note: BW = Body Weight; GC = Gait Cycle




                                       - 92 -
 Generatio




                                                          Toe-Off




    Fig 3.35: Definition of total hip power parameters.




Legends:


HP1 (Watt/BW) Maximum hip power generation in stance phase
HP2 (Watt/BW) Maximum hip power absorption in stance phase
HPT1 (%GC)        Time of occurrence at HP1
HPT2 (%GC)        Time of occurrence at HP2



Note: BW = Body Weight; GC = Gait Cycle




                                          - 93 -
                                   Total




                                                          Toe-Off


   Fig 3.36: Definition of total knee power parameters.




Legends:


KP1 (Watt/BW)     Maximum knee power generation at stance phase
KP2 (Watt/BW)     Maximum knee power absorption at stance phase
KP3 (Watt/BW)     Maximum knee power absorption at swing phase
KPT1 (%GC)        Time of occurrence at KP1
KPT2 (%GC)        Time of occurrence at KP2
KPT3 (%GC)        Time of occurrence at KP3



Note: BW = Body Weight; GC = Gait Cycle




                                       - 94 -
   Absorption Generation
          (Watt/Kg)




                                                                Toe-Off


  Fig 3.37: Definition of total ankle power parameters.




Legends:


AP1 (Watt/BW)              Maximum ankle power absorption at stance phase
AP2 (Watt/BW)              Maximum ankle power generation at stance phase
APT1 (%GC)                 Time of occurrence at AP1
APT2 (%GC)                 Time of occurrence at AP2



Note: BW = Body Weight; GC = Gait Cycle




                                               - 95 -
Fig 3.38: The start point and end point of muscle firing duration in the gait cycle

from our data. The quadriceps, hamstrings, tibialis anterior and calf muscle were

studied, with different type of conditions 1) barefoot, 2) AFO 3) DAFO and 4)

normal.




                                       - 96 -
3.6 Data Analysis



For all gait parameters collected, the left and the right sides in all groups were tested

with paired t-test (App. 3.4.1-3.4.12) and then averaged into a single data. Two-tailed

independent t-test was performed to compare the difference between the barefoot

from CP and from normal group to document the natural difference of CP. The

ANOVA with repeated measurement (with Bonferroni`s confidence interval

adjustment) was conducted to compare the differences among the barefoot, AFO and

DAFO to evaluate the orthotic effect. All the results were testing with 95% of

confident interval. The flow of statistical testing was followed.




                          CP (n=16)                                 Normal (n=18)




   Left vs Right        Left vs Right              Left vs Right         Left vs Right




     Averaged              Averaged                 Averaged            Averaged
       AFO                  DAFO                   CP Barefoot       Normal Barefoot




      ANOVA with repeated measurement                         Independent t-test




                                          - 97 -
                                 CHAPTER 4


                                   RESULTS



The gait analysis of 16 patients with cerebral palsy (CP) with three conditions

(Barefoot, with AFO and with DAFO) and 18 healthy normal controls were analyzed.

Details of the whole experimental procedures were described in the chapter 3. Six

categories of the gait parameters were measured in the chapter 3 and reported in this

chapter.



In each category of gait parameters, the results were reported as two levels. At first,

the barefoot from CP and that from normal control were compared with independent

samples t-test to document the alteration of CP. Then, the CP subjects with barefoot,

AFO and DAFO were compared with ANOVA with repeated measurements to

observe any treatment effect. All the results were presented in a table format. In

particular, some of the important findings were shown graphically for better

observations.



The results of different categories: Temporal distance, range of motion, ground

reaction force, joint moment, joint power and electromyography parameters were

presented in the following paragraphs.




                                         - 98 -
4.1 Temporal Distance Parameters


i) For comparison of CP and normal (Table 4.1), CP had a significant shorter stride

length and slower walking speed than the normal control subject, whereas there was

no significant difference found in other parameters between the two groups.



ii) To evaluate the orthotic effect on the CP subjects, AVOVA with repeated

measurement was performed (Table 4.2). The CP patients had a significant increase

in stride length when wearing DAFO, such increase being toward the normal value

which indicated improvement. There was no other significant difference found in

other temporal distance parameters.




                                        - 99 -
Table 4.1: Comparing the temporal distance parameters between barefoot of CP

and normal using the independence t-test.



                                 CP                    Normal
                                                                       t-value   p-value
                         Mean          SD           Mean        SD
  Stride length (m)      0.75         0.193          1.05    0.139     -5.388    0.000
    Stride time (s)      1.02         0.274          0.95    0.084     0.936     0.362
 Waking Speed (m/s)      0.78         0.255          1.14    0.142     -4.970    0.000
   Stance time (s)       0.63         0.033          0.61    0.015     1.606     0.124
    Swing time (s)       0.40         0.246          0.35    0.088     0.849     0.402
 Stance/Swing ratio      2.00         0.893          1.91    0.547     0.358     0.723
 Cadence (steps/min)    103.58        21.001        105.79   9.456     -0.388    0.702




Table 4.2: Comparing the temporal distance parameters among bare, AFO and
DAFO using the ANOVA with repeated measures. (Raw SPSS output in the
appendix 4.1)


                              BARE                         AFO             DAFO
                        Mean       SD                Mean       SD     Mean     SD
 Stride length (m)       0.75*    0.193               0.81     0.196   0.86*   0.174
  Stride time (s)        1.02     0.274              1.07      0.293    1.04   0.262
Walking speed (m/s)      0.78     0.255              0.81      0.262    0.87   0.258
  Stance time (s)        0.63     0.033               0.64     0.044    0.63   0.029
  Swing time (s)         0.40     0.246               0.44     0.261    0.41   0.237
Stance/Swing ratio       2.00     0.893              1.78      0.803    2.02   1.064
Cadence (steps/min)     103.58   21.001              98.49    17.870   101.39 21.226
* mean significant difference between barefoot and DAFO




                                          - 100 -
4.2 Range of Motion Parameters


The range of motion (ROM) parameters at the hip, knee and ankle joints were

respectively evaluated and reported as follows:



4.2.1 Hip Motion


i) Comparing barefoot of the hip walking motion from CP and from normal control,

the maximum hip extension angle at stance phase (HA2) from CP was significantly

less than the normal. On the other hand, a significantly greater hip flexion angle at

swing phase (HA3) than the normal was found in the table 4.3.



ii) For the orthotic effect, the hip angle at the initial contact (HA1), maximum hip

flexion angle in swing phase (HA3) and total range of hip motion were found to have

a significant difference (Table 4.4). Both orthoses significantly increased in HA1 and

total range of motion of hip (range) from the baseline (barefoot). Furthermore, in the

DAFO group, there was significantly greater maximum hip angle in swing phase

(HA3) against the baseline. There was no significant difference between the two

orthoses.




                                        - 101 -
4.2.2 Knee Motion

i) Comparing CP with normal control, CP had significantly greater knee flexion

angle at initial contact (KA1). The CP had a significantly earlier onset in the time of

occurrence of maximum knee flexion angle in stance (KAT2) and that of knee angle

at terminal stance phase (KAT3). On the contrary, CP subjects significantly delayed

the time of occurrence of maximum knee flexion angle in the swing phase (KAT4)

against the normal control (Table 4.5).



ii) For comparison within CP groups, DAFO had a significant higher knee angle at

the terminal stance phase (KA4) than the barefoot and AFO, which greater than the

normal control. The DAFO significantly increased the knee flexion angle at initial

contact (KA1) and the maximum knee flexion angle in the stance phase (KA2) than

the barefoot. In addition, the AFO found a significant delay in the time of knee angle

at terminal stance (KAT3) than DAFO (Table 4.6).




                                          - 102 -
4.2.3 Ankle Motion


i) Comparing the barefoot of CP with the normal control, the CP had significantly

excessive plantarflexed position at initial contact (AA1), in stance phase (AA2) and

during the swing phase (AA4) than the normal control. In addition, the CP

significantly preserved a smaller total range of ankle joint (Table 4.7).



ii) For comparison among the barefoot, AFO and DAFO, both orthoses significantly

reduced the excessive plantarflexed position at theses positions (AA1, AA2 and

AA4).The push-off in the pre-swing phase were limited by the orthoses resulting in

reduced plantarflexion at AA3. AFO had a significant limitation of total range of

motion (ROM) compared to the barefoot and the DAFO. For the difference between

orthoses, the AFO significantly limited the phase of AA3 than the DAFO. Moreover,

the DAFO had significant less restriction on ROM than AFO (Table 4.8 and Fig 4.1).




                                         - 103 -
Table 4.3: Comparing the ROM of hip parameters between barefoot of CP and
normal using the independence t-test.


                            CP                       Normal
                                                                   t-value   p-value
                   Mean           SD          Mean            SD
 HA1 (degree)      38.32         8.040        33.60        5.729   1.989     0.055
 HA2 (degree)       -6.13        8.380       -14.63        5.662   3.500     0.001
 HA3 (degree)      43.20         9.309        36.67        6.478   2.396     0.023
 HAT2 (% GC)       27.69         2.394        28.03        0.813   -0.542    0.595
 HAT3 (%GC)        47.03         1.072        46.39        1.378   1.503     0.143
RANGE(degree)      49.33         7.772        51.30        6.895   -0.784    0.439

Note: GC = Gait Cycle



Table 4.4: Comparing the ROM of hip parameters among bare, AFO and DAFO
using the ANOVA with repeated measures (Raw SPSS output in the appendix 4.2).


                        BARE                         AFO                DAFO
                  Mean           SD         Mean              SD   Mean        SD
 HA1 (degree)    38.32*†     8.040          42.79†         9.577   43.07*    9.307
 HA2 (degree)      -6.13     8.380           -6.69         8.587   -6.65     8.480
 HA3 (degree)     43.20*     9.309          46.86          9.861   47.83*    10.665
 HAT2 (% GC)       27.69     2.394          28.03          2.566   27.47     2.579
 HAT3 (%GC)        47.03     1.072          47.34          1.106   46.81     1.196
RANGE(degree) 49.33*†        7.772          53.55†         6.632   54.48*    9.244
Note: GC = Gait Cycle


* mean significant difference between barefoot and DAFO.
† mean significant difference between barefoot and AFO.




                                         - 104 -
Table 4.5: Comparing the ROM of knee parameters between barefoot of CP and
normal using the independence t-test.

                        CP                 Normal
                                                              t-value   p-value
               Mean           SD     Mean         SD
KA1 (degree)   22.99         7.865    4.63      5.765         7.826      0.000
KA2 (degree)   21.48         9.530   18.07      6.480         1.232      0.227
KA3 (degree)    3.69         9.211    3.17      5.330         0.206      0.838
KA4 (degree)   57.58         8.020   59.77      6.539         -0.877     0.387
KAT2 (%GC)      5.22         1.251    7.83      0.875         -7.127     0.000
KAT3 (%GC)     19.38         4.085   21.81      1.457         -2.256     0.036
KAT4 (%GC)     41.59         2.215   37.97      0.675         6.285      0.000
Note: GC = Gait Cycle



Table 4.6: Comparing the ROM of knee parameters among bare, AFO and DAFO
using the ANOVA with repeated measures. (Raw SPSS output in the appendix 4.3)

                      BARE                     AFO                  DAFO
                Mean       SD        Mean             SD      Mean       SD
KA1 (degree) 22.99*       7.865       25.71          6.704    28.04*    8.701
KA2 (degree) 21.48*       9.530       25.69          7.829    27.43*    9.032
KA3 (degree)     3.69     9.211       2.21           7.545     5.57     8.433
KA4 (degree) 57.58*       8.020      60.37‡          6.290   65.91*‡    8.526
KAT2 (%GC)       5.22     1.251       5.22           1.378     5.41     1.452
KAT3 (%GC) 19.38          4.085      20.47‡          3.324    18.38‡    3.956
KAT4 (%GC) 41.59          2.215      41.22           1.505    41.00     1.623
Note: GC = Gait Cycle


* mean significant difference between barefoot and DAFO
† mean significant difference between barefoot and AFO
‡ mean significant difference between AFO and DAFO




                                     - 105 -
Table 4.7: Comparing the ROM of ankle parameters between barefoot of CP and
normal using the independence t-test.




Note: GC = Gait Cycle



Table 4.8: Comparing the ROM of ankle parameters among bare, AFO and DAFO
using the ANOVA with repeated measures (Raw SPSS output in the appendix 4.4).

                              BARE                   AFO             DAFO
                         Mean      SD           Mean     SD      Mean     SD
                        -6.87*†  6.946          3.28†   4.922    3.94*  4.693
                        2.04*†    10.122        11.66†   5.844   14.10*    6.003
                     -20.24*†     11.997        0.09†‡   4.873   -5.17*‡   6.605
                        -4.57*†   8.348         3.29†    4.533   5.62*     5.566
                         21.25    7.983         25.13    2.772   21.72     7.872
                        33.56     5.856         36.50    2.569   35.84     1.710
                         45.25    1.197      45.53       2.225    45.59    1.635
                        22.29†    7.372     11.56†‡      5.558   19.26‡    4.505
Note: GC = Gait Cycle


* mean significant difference between barefoot and DAFO
† mean significant difference between barefoot and AFO
‡ mean significant difference between AFO and DAFO




                                      - 106 -
                  dorsiflexion


                                                                            Toe-Off                 Range


                                                                                            AA4
         (Degree)
Plantarflexion




                                                                                         AAT4




                                 1   2   3   1    2   3      1      2   3   1    2   3          1   2    2


                                 *+              *+          *+‡                *+                  +‡




             Fig 4.1: The significance value of AA1, AA2, AA3 and range from ANOVA
             with repeated measurement shown at the bottom, with 1 (barefoot), 2 (AFO)
             and 3 (DAFO). The dotted line indicated as the mean value of our normal
             control.
             * Sig diff between bare - DAFO
             + Sig diff between bare - AFO
             ‡Sig diff between AFO - DAFO




                                                          - 107 -
4.3 Ground Reaction Force Parameters

i) To compare CP with normal control, CP significantly produced smaller second

peak at terminal stance (GR3) than the normal (table 4.9). No significant difference

was found in other GRF parameters.



ii) When evaluating the orthotic effect, only the AFO significantly increased the

second peak at the terminal stance (GR3) against their barefoot (table 4.10), such

increase was closer toward our normative value; while the DAFO did not. No

significant difference between two orthoses was found in other ground reaction force

parameters.




                                       - 108 -
Table 4.9: Comparing the ground reaction force parameters between barefoot of
CP and normal using the independence t-test.


                         CP               Normal
                                                           t-value        p-value
                 Mean          SD     Mean         SD
 GR1 (%BW)        1.14        0.259   1.13        0.100        0.067       0.948
 GR2 (%BW)        0.71        0.118   0.94        0.929        -0.992      0.328
 GR3 (%BW)        0.90        0.107   0.99        0.077        -2.913      0.006
GRT1 (%GC)        7.69        3.750   7.56        0.889        0.137       0.892
GRT2 (%GC)       14.72        4.722   15.83       1.350        -0.912      0.375
GRT3 (%GC)       22.50        3.642   24.28       0.600        -1.929      0.072
Note: BW = Body Weight;       GC = Gait Cycle



Table 4.10: Comparing the ground reaction force parameters among bare, AFO and
DAFO using the ANOVA with repeated measures (Raw SPSS output in the appendix
4.5)

                      BARE                       AFO                         DAFO
                Mean       SD          Mean             SD              Mean      SD
GR1 (%BW)        1.14    0.259         1.26            0.339            1.29     0.317
GR2 (%BW)        0.71    0.118         0.72            0.191            0.68     0.161
GR3 (%BW)       0.90†    0.107         0.98†           0.058            0.92     0.140
GRT1 (%GC)      7.69     3.750         7.63            3.207            7.31     2.575
GRT2 (%GC)      14.72    4.722         13.75           4.251            14.47    3.713
GRT3 (%GC)      22.50    3.642         21.31           3.816            22.00    3.291
Note: BW = Body Weight;       GC = Gait Cycle


† mean significant difference between barefoot and AFO




                                       - 109 -
4.4 Joint Moment Parameters


The joint moment parameters at the hip, knee and ankle joints were evaluated and

reported as follows:



4.4.1 Hip Moment


i) Comparing CP with the normal control, the CP significantly reduced the maximum

hip flexor moment at stance phase (HM2), in which the less flexor moment was

generated in CP. No significance difference was detected in other hip moment

parameters (Table 4.11).



ii) For comparison among barefoot, AFO and DAFO, only the maximum hip

extensor moment at stance phase (HM1) was significant higher in DAFO than the

barefoot (Table 4.12), which approached normative value and indicative of

improvement. No significant difference between two orthoses was found.




                                      - 110 -
4.4.2 Knee Moment


i) Comparing CP with the normal control, the CP significantly produced less

maximum knee flexor moment at stance phase (KM2) and postponed the time of

maximum knee extensor moment at stance phase (KMT1) than the normal control

group (Table 4.13).



ii) There was no significant difference among all CP conditions (Table 4.14).




                                       - 111 -
4.4.3 Ankle Moment


i) For comparison between CP and normal control, the CP had a significant early

onset in the time of the maximum dorsiflexor moment in stance phase (AMT1) and

the maximum plantarflexor moment in stance phase (AMT2) (Table 4.15).



ii) For comparison among all CP conditions, both orthoses significantly increased the

maximum plantarflexor moment in stance phase (AM2) against the barefoot, which

was a positive effect. For the time of maximum dorsiflexor moment in stance phase

(AMT1), the AFO significantly delayed the onset time than the barefoot, which

tended to our normative value. The AFO also had significantly earlier onset the time

than DAFO (Table 4.16 and Fig 4.2).




                                       - 112 -
Table 4.11: Comparing the hip moment parameters between barefoot of CP and
normal using the independence t-test.


                           CP                      Normal
                                                                   t-value   p-value
                  Mean            SD        Mean            SD
HM1 (Nm/BW)        1.15          0.624       1.45       0.385      -1.746    0.090
HM2 (Nm/BW)        -0.90         0.400      -1.48       0.277      4.963     0.000
 HMT1 (%GC)        3.97          1.746       4.64       1.109      -1.352    0.186
 HMT2 (%GC)        26.72         1.472      26.69       0.645      0.064     0.950
Note: BW = Body Weight;      GC = Gait Cycle



Table 4.12: Comparing the hip moment parameters among bare, AFO and DAFO
using the ANOVA with repeated measures (Raw SPSS output in the appendix 4.6)

                          Bare                AFO                     DAFO
                   Mean        SD       Mean       SD            Mean      SD
HM1 (Nm/BW) 1.15*             0.624      1.33     0.628          1.52*    0.749
HM2 (Nm/BW) -0.90             0.400     -0.83     0.308          -0.93    0.402
 HMT1 (%GC)         3.97      1.746      5.00     1.033           4.22    1.673
 HMT2 (%GC)        26.72      1.472     26.56     1.721          26.28    1.538
Note: BW = Body Weight; GC = Gait Cycle
* mean significant difference between barefoot and DAFO




Table 4.13: Comparing the knee moment parameters between barefoot of CP and
normal using the independence t-test.

                           CP                    Normal
                                                                   t-value   p-value
                   Mean           SD        Mean        SD
KM1 (Nm/BW)        -0.66         0.393      -0.74     0.214         0.748     0.46
KM2 (Nm/BW)         0.34         0.276       0.62     0.190        -3.435    0.002
KMT1 (%GC)          5.44         4.370       2.92     1.240         2.229    0.039
KMT2 (%GC)         27.56         1.537      27.58     0.772        -0.049    0.961
Note: BW = Body Weight;      GC = Gait Cycle




                                         - 113 -
Table 4.14: Comparing the knee moment parameters among bare, AFO and DAFO
using the ANOVA with repeated measures (Raw SPSS output in the appendix 4.7).

                       BARE               AFO                           DAFO
                 Mean       SD      Mean       SD                 Mean       SD
 KM1 (Nm/BW)     -0.66     0.393    -0.56     0.318               -0.68     0.399
 KM2 (Nm/BW)      0.34     0.276     0.31     0.122                0.39     0.205
 KMT1 (%GC)       5.44     4.370     7.56     5.180                4.19     2.926
 KMT2 (%GC)      27.56     1.537    28.34     2.079               27.72     1.472
Note: BW = Body Weight; GC = Gait Cycle



Table 4.15: Comparing the ankle moment parameters between barefoot of CP and
normal using the independence t-test.


                          CP                       Normal
                                                                  t-value   p-value
                   Mean         SD         Mean             SD
AM1 (Nm/BW)        0.04        0.060       -0.02        0.140     1.493     0.145
AM2 (Nm/BW)        0.76        0.300             0.9    0.111     -1.941    0.061
 AMT1 (%GC)        1.06        0.170        1.36        0.540     -2.235    0.037
 AMT2 (%GC)        15.5        7.050       22.22        4.800     -3.208    0.004
Note: BW = Body Weight;    GC = Gait Cycle



Table 4.16: Comparing the ankle moment parameters among bare, AFO and DAFO
using the ANOVA with repeated measures (Raw SPSS output in the appendix 4.8)

                     BARE                 AFO                         DAFO
                Mean      SD       Mean        SD                Mean      SD
AM1 (Nm/BW)     0.04     0.060      0.03      0.216              0.11     0.258
AM2 (Nm/BW) 0.76*†       0.301      1.05†     0.330              1.18*    0.412
AMT1 (%GC)      1.06†    0.171     1.47†‡     0.531              1.13‡    0.289
AMT2 (%GC)      15.50    7.055     17.00      6.733              14.72    7.486
Note: BW = Body Weight; GC = Gait Cycle


* mean significant difference between barefoot and DAFO
† mean significant difference between barefoot and AFO
‡ mean significant difference between AFO and DAFO


                                       - 114 -
                                     Ankle
     Dorsiflexor




                                                        Toe-Off




Fig 4.2: The significance value of AMT1 and AM2 from ANOVA with repeated
measurement shown at the bottom, with 1 (barefoot), 2 (AFO) and 3 (DAFO).
The dotted line indicated the mean value of our normal control.
The bottom left showed the significant difference from the AFO (AMT1)
The bottom right showed the significant difference from the barefoot (AM2).




                                     - 115 -
4.5 Joint Power Parameters


The joint power parameters at the hip, knee and ankle joints were evaluated and

reported as follows:




4.5.1 Hip Power


i) For comparison between CP and the normal control, the CP had significantly less

hip power absorption in stance phase (HP2) than normal; The CP was had a

significantly earlier onset at the hip power generation in the stance phase (HPT1)

than the normal (Table 4.17).



ii) No significant difference was found among barefoot, AFO and DAFO in all hip

power parameters (Table 4.18).




                                      - 116 -
4.5.2 Knee Power


i) For comparing CP with normal, the CP showed significantly less maximum knee

power generation at stance phase (KP1), the maximum knee absorption at stance

phase (KP2) and the maximum knee power absorption at swing phase (KP3) than the

normal. Furthermore, CP indicated a significant delay at the time of occurrence of

maximum knee power absorption in swing phase (KPT3) against normal (Table

4.19).



ii) There was no significant difference found among barefoot, AFO and DAFO in all

knee power parameters (Table 4.20).




                                      - 117 -
4.5.3 Ankle Power


i) For CP and normal, a significantly less maximum ankle power generation at stance

phase (AP2) was found on the barefoot of CP than that of the normal (Table 4.21).



ii) For comparison among barefoot, AFO and DAFO, the AFO decreased

significantly in the maximum ankle power generated in stance phase (AP2) than their

barefoot and DAFO; such change was deviated away from the normal control and

indicated that was not a functional benefit of AFO. The AFO had a significantly less

ankle power than the DAFO (Table 4.22 and Fig 4.3).




                                       - 118 -
Table 4.17: Comparing the hip power parameters between barefoot of CP and
normal using the independence t-test.


                         CP                       Normal
                                                                 t-value    p-value
                 Mean            SD       Mean             SD
HP1 (Watt/BW)     2.33          1.509      2.04          0.594   0.731      0.473
HP2 (Watt/BW)    -1.35          0.928      -2.06         0.571   2.741       0.01
 HPT1 (%GC)       4.91          2.806      7.17          1.485   -2.883     0.009
 HPT2 (%GC)      24.28          2.352     24.94          0.969   -1.098      0.28
Note: BW = Body Weight;       GC = Gait Cycle



Table 4.18: Comparing the hip power parameters among bare, AFO and DAFO
using the ANOVA with repeated measures (Raw SPSS output in the appendix 4.9).


                         Bare                      AFO                    DAFO
                 Mean            SD       Mean             SD     Mean           SD
HP1 (Watt/BW)    2.33           1.509      2.81          1.423     3.22       2.196
HP2 (Watt/BW)    -1.35          0.928     -1.27          0.770    -1.34       0.797
 HPT1 (%GC)      4.91           2.806      5.88          2.284     5.44       2.190
 HPT2 (%GC)      24.28          2.352     24.06          2.845    24.94       1.515
Note: BW = Body Weight;       GC = Gait Cycle



Table 4.19: Comparing the knee power parameters between barefoot of CP and
normal using the independence t-test.

                         CP                    Normal
                                                                 t-value     p-valve
                 Mean            SD       Mean        SD
KP1 (Watt/BW)    0.35           0.512     2.06      0.965         -6.574      0.000
KP2 (Watt/BW)    -1.13          0.589     -1.99     0.637         4.075       0.000
KP3 (Watt/BW)    -1.54          0.698     -2.75     0.794         4.704       0.000
 KPT1 (%GC)       4.59          2.728      3.67     0.786         1.312       0.207
 KPT2 (%GC)      28.84          3.239     28.75     0.712         0.113       0.911
 KPT3 (%GC)      49.00          0.931     47.42     0.772         5.421       0.000
Note: BW = Body Weight;       GC = Gait Cycle




                                        - 119 -
Table 4.20: Comparing the knee power parameters among bare, AFO and DAFO
using the ANOVA with repeated measures (Raw SPSS output in the appendix 4.10)

                          Bare                     AFO                 DAFO
                  Mean            SD       Mean           SD     Mean       SD
KP1 (Watt/BW)     0.35           0.512     0.51          0.571    0.90     1.028
KP2 (Watt/BW)     -1.13          0.589     -1.61         1.127   -1.37     0.755
KP3 (Watt/BW)     -1.54          0.698     -1.49         0.846   -1.67     0.992
 KPT1 (%GC)        4.59          2.728      4.47         1.866    5.28     2.646
 KPT2 (%GC)       28.84          3.239     29.38         1.688   29.22     1.712
 KPT3 (%GC)       49.00          0.931     48.97         0.921   48.97     0.846
Note: BW = Body Weight;        GC = Gait Cycle



Table 4.21: Comparing the ankle power parameters between barefoot of CP and
normal using the independence t-test.


                          CP                       Normal
                                                                 t-value   p-value
                 Mean             SD       Mean             SD
AP1 (Watt/BW)     -1.26          0.889     -0.90         0.403   -1.502    0.148
AP2 (Watt/BW)     0.83           0.364      1.30         0.615   -2.686    0.011
 APT1 (%GC)       9.28           6.703     12.47         7.107   -1.342    0.189
 APT2 (%GC)       26.97          4.752     28.08         1.252   -0.911    0.375
Note: BW = Body Weight;        GC = Gait Cycle



Table 4.22: Comparing the ankle power parameters among bare, AFO and DAFO
using the ANOVA with repeated measures (Raw SPSS output in the appendix 4.11)

                          Bare                     AFO                 DAFO
              Mean              SD        Mean            SD     Mean       SD
AP1 (Watt/BW) -1.26            0.889      -1.08          0.797   -1.65     1.267
AP2 (Watt/BW) 0.83†            0.364      0.54†          0.219    0.89     0.544
 APT1 (%GC)    9.28            6.703      11.38          7.835    8.63     6.467
 APT2 (%GC) 26.97              4.752      28.06          4.449   25.30     8.260
Note: BW = Body Weight;        GC = Gait Cycle


† mean significant difference between barefoot and AFO


                                         - 120 -
    Absorption Generation
           (Watt/Kg)




                                                        Toe-Off




                                                    1       2     3




Fig 4.3: The significance value of AMT1 and AM2 from ANOVA with repeated
measurement shown at the bottom, with 1 (barefoot), 2 (AFO) and 3 (DAFO).
The dotted line indicated the mean value from our normal control.
The line (bottom) showed the significant difference from the AFO (AMT1)




                                    - 121 -
4.6 Electromyography Parameters


The electromyography signals (EMG) from quadriceps, hamstrings, tibialis anterior

(TA) and calf muscles were evaluated with three parameters: total contracting

duration (duration), contracting root mean square value (RMS) and contracting

median frequency (MF).



i) For comparison of CP and normal: for contracting duration, it was found that the

CP had significantly longer firing duration within a gait cycle for all the muscle

groups. For median frequency, all the muscle groups in the CP had significantly

higher median frequency than the normal control. For root mean square value, the CP

produced significantly smaller amplitude in tibialis anterior muscle than the normal

(Table 4.23 and Fig 4.4).



ii) For comparison among the CP groups: For the contracting duration, both orthoses

had a significant reduction in the calf muscle’s contracting duration against barefoot,

and these improved the excessive muscle firing for walking (Fig 4.5). There was no

difference in the duration between two orthoses. Concerning the parameter of MF,

the DAFO showed a significant decrease in the hamstrings’ MF against their

barefoot. For calf’s MF, the AFO was significant lower than both barefoot and

DAFO, which indicated as an improvement. For parameter of RMS, the AFO

significantly increased the RMS of the hamstrings and calf muscles against the

barefoot, while the DAFO did not. Furthermore, a significant difference between two

orthoses was found in the RMS of calf muscle, with DAFO higher than the AFO,

which approached toward the normative value (Table 4.24 and Fig 4.6).



                                        - 122 -
Table 4.23: Comparing the EMG parameters between barefoot of CP and normal

using the independence t-test.


                                   CP                   Normal
  Channel                                                             t-value   p-value
                          Mean           SD          Mean    SD
            Duration
 quadriceps               65.80         14.175       47.52   12.289   4.027       0
            (%GC)
             RMS           4.04          2.208        3.94    1.789   0.147     0.884
            MF (Hz)       96.96         22.346       75.92   16.341   3.101     0.004
            Duration
 hamstrings               67.51         14.699       56.55   13.431   2.272      0.03
            (%GC)
             RMS           2.97          1.023        4.87    6.014   -1.242    0.223
            MF (Hz)       100.50        18.966       76.20   16.015    4.05       0
            Duration
    TA                    76.93         7.429        66.65   9.765    3.418     0.002
            (%GC)
             RMS           3.37          1.481        5.57   1.281    -4.642      0
            MF (Hz)       111.26        24.530       80.46   9.381     4.725      0
            Duration
    calf                  74.30         8.914        55.79   9.147    5.958       0
            (%GC)
             RMS           3.55          1.308        4.26   1.372    -1.535    0.135
            MF (Hz)       113.65        16.520       81.40   9.360     6.886      0
Note: GC = Gait Cycle




                                           - 123 -
% GC


                *          *                *         *




            *               *               *         *




                                            *




Fig 4.4: Firing duration (Top), Median frequency (Middle) and Root Mean
Square (Bottom) between CP barefoot and normal barefoot at 1)
Quadriceps, 2) Hamstrings, 3) Tibialis Anterior and 4) Calf Muscle
* meant significant difference between Bare and Normal (P<0.05)

                                  - 124 -
Table 4.24: Comparing the EMG parameters among bare, AFO and DAFO using the
ANOVA with repeated measures (Raw SPSS output in the appendix 4.12)


                                Bare                       AFO              DAFO
 Channel
                        Mean            SD             Mean    SD        Mean    SD
             Duration
quadriceps              65.80      14.175              62.94   15.859    61.44    11.883
             (%GC)
              RMS        4.04       2.208               4.84    1.508    4.33     1.996
             MF (Hz)    96.96      22.346              89.42   22.453    88.32    24.914

             Duration
hamstrings            67.51        14.699              63.25   14.918    61.21    9.967
             (%GC)
              RMS     2.97†         1.023              3.82†   1.876     3.48     1.624
             MF (Hz) 100.50*       18.966              93.78   19.020   89.93*    19.657

             Duration
    TA                76.93            7.429           73.14   11.008    70.19    9.965
             (%GC)
              RMS      3.37         1.481           4.00        1.482    3.71     1.698
             MF (Hz) 111.26        24.530          104.96      27.586   103.70    24.564

             Duration
   calf               74.30*†          8.914       67.20†      10.561   68.14*    6.152
             (%GC)
              RMS      3.55†        1.308          5.13†‡       1.667    4.12‡     1.358
             MF (Hz) 113.65†       16.520         88.03†‡      29.761   113.15‡   14.481
Note: GC = Gait Cycle

* mean significant difference between barefoot and DAFO
† mean significant difference between barefoot and AFO
‡ mean significant difference between AFO and DAFO




                                             - 125 -
   % GC


                                                    +   *




Fig 4.5: Contracting duration among barefoot, AFO and DAFO at 1)
Quadriceps, 2) Hamstrings, 3) Tibialis Anterior and 4) Calf Muscle. The
dotted line represented the normative value.
* mean significant difference between Bare and AFO (P<0.05)
+ mean significant difference between Bare and DAFO (P<0.05)




                                    - 126 -
                                                           *       ‡
                                  +




                                                          *    †
                           *




Fig 4.6: Median frequency (Top) and Root Mean Square (Bottom) among
barefoot, AFO and DAFO at 1) Quadriceps, 2) Hamstrings, 3) Tibialis Anterior
and 4) Calf Muscle. The dotted line represented the normative value.
* mean significant difference between Bare and AFO (P<0.05)
+ mean significant difference between Bare and DAFO (P<0.05)
‡ mean significant difference between AFO and DAFO (P<0.05)




                                       - 127 -
                                 CHAPTER 5

                                DISCUSSION



The following aspects will be discussed in this chapter. Firstly, the methodology will

be including subject inclusion criteria, choice of the baseline and the parameters.

Secondly, our findings (temporary distance, range of motion, ground reaction force,

moment, power and electromyography) will be interpreted and reviewed with other

publications. Thirdly, the limitations and the clinical significance of these findings

will be stated.




                                        - 128 -
5.1 Methodology

5.1.1 Inclusion Criteria of Subject

The diplegic CP participants with dynamic equinus deformity of moderate level

(modified Ashworth scale 1-3) were recruited. The most severe group (modified

Ashworth scale 4) is considered to be having early contracture already; while the

mildest group (modified Ashworth scale 0) is considered to have no spasticity and

these two groups were excluded from the study.



Patients with significant (observable) coronal or rotational deformities at the hip,

knee and ankles were excluded to avoid any compensatory effects toward the sagittal

plane alignment (David JR, 1996). However, this exclusion might undoubtedly limit

the results to be generalized to CP subjects with multiple planes deformities.



It is necessary to understand the natural history of mature walking before defining

and interpreting pathological gait in children. It is well accepted that a mature gait is

present in normal children by the age of 5. The study of Keen M (1993) further

suggested that a 3-year-old child could achieve an adult pattern of base of support

and joint angles throughout the gait cycle. This has been further supported by

Johnson DC et al (1997). Sutherland DH in 1994 and 1997 confirmed that the

parameters of step length, cadence and walking velocity showed evidence of both

central nervous system maturation and growth at approximately 4 years of age.



Nevertheless, gait patterns from adults and children had been reported to be different

(James B et al, 1989). In assessing the gait pattern in CP children, it is better to

recruit normal children of similar age for valid comparison. Consequently, normal
                                         - 129 -
children from a local primary school were invited to be our normal control. The age

range of normal subjects (6 - 9.7 years) was however slightly different from that of

CP subjects (3.3 – 13.7 years);




5.1.2 Choice of Baseline

Either barefoot or shoe alone can serve as the baseline in a gait study. The shoe alone

condition has been used as the baseline in some studies, but shoes which fit the

orthosis are too big for the subject’s feet and such loose shoes may cause unnatural

gait. Barefoot was therefore chosen as our baseline.



Many authors agreed that barefoot can be used as the baseline to determine the

orthotic effect (Abel MF et al, 1998; Brunner R et al, 1998; Dieli J et al, 1997;

Radtka et al, 1997; Romkes J et al, 2002; Thompson NS et al, 2002; White H et al,

2002) because the original walking pattern could be obtained. This helped to

determine the actual alteration of gait. Moreover, the comparison of gait with and

without shoes in children was investigated by Oeffinger D et al (1999), who

confirmed that barefoot was sufficient to be a baseline for most clinical studies and

shoe wearing was not necessary.




                                        - 130 -
5.1.3 Design of Study
The major source of error could be errors from inconsistencies in the placement of

the reflective markers on the subjects. A further possible source of error may arise

from the attachment of markers on the orthoses. To improve the accuracy of the

placements of reflective markers and electrodes, the placements had been repeatedly

practiced under supervision by a senior physiotherapist before starting the study.

Movement of electrodes was minimized by good attachment of electrodes and

monitoring the raw EMG signal during walking trial.



For each CP patient, the three gait sessions of data collection were done in random

order on the same day: barefoot, AFO and DAFO. This eliminated marker placement

variations from doing the tests on different days. The ANOVA with repeated

measurement was used because of its sensitivity to detecting differences in the

dependent variables (Barefoot, AFO and DAFO for same subject) and eliminates the

bias to participants in one group being different from the participants in other groups.




                                         - 131 -
5.1.4 Investigated Parameters

Six categories of parameters were investigated including temporal distance, joint

range of motion (ROM), ground reaction force, joint moment, joint power and

electromyography (EMG) parameters. Each of the parameters was averaged with

three recordings rather than five recordings because no significant difference was

determined between the three and five recordings (Kirkpatrick M et al, 1994), and

three recordings shortened the period of acquisition to prevent fatigue, especially for

the CP group because they needed to receive three conditions (barefoot, AFO and

DAFO) on the same day. Steinwender G et al (2000) studied the intra-subject

repeatability of gait analysis data in normal and spastic children, on the same day.

The intra-subject variability of the temporal distance, kinematic and kinetic

parameters was acceptable.




5.1.4.1 Component of Ground Reaction Force


The ground reaction force can be resolved into vertical force, medial-lateral shear

and anterior-posterior shear force. According the findings of Kirkpatrick M et al

(1994) and White R et al (1999), they found that the vertical component of force was

the most reproducible parameters, and the medial-lateral shear force was the least

reproducible when done between days or between subjects. It is because the

magnitude of the shear force is so much smaller than the vertical force component.

Therefore, only the vertical force vector component was investigated in this study.




                                        - 132 -
5.1.4.2 Choice of Sagittal Parameters


During walking, movement occurs in the sagittal, coronal and traverse plane. The

kinematics and kinetics of the hip, knee and ankle joint in sagittal plane, coronal

plane as well as transverse plane can be generated by the computer. However, the

sagittal joint ROM, and joint moment were evaluated in the present study. It is

because the data from coronal or transverse plane was less accurate than those from

sagittal plane (Kadaba MP et al, 1989; Kirkpatrick M et al, 1994; Ounpuu S, 1995;

Whittle MW, 1996). This could be explained that the coronal or transverse motion is

actually very small, resulting in high signal-to-noise ratio (Benedetti MG et al, 1998;

Meglan D et al, 1994). On the other hand, the sagittal plane measurements are the

best understood and the most accurately reproducible (Sutherland DH et al, 1994).

Therefore, most gait studies on ankle foot orthoses only interpreted the joint range of

motion and the moment in the sagittal plane (Carlson WE et al, 1997; Crenshaw S et

al, 2000; Radtka SA et al, 1997; Romkes J et al, 2002). More importantly, this study

focuses on the orthotic effect toward equinus which occurs in the sagittal plane

(Gage JR, 1991).




5.1.4.3 Parameters for Electromyography


The measurement of walking EMG is now a routine element in clinical gait analysis

(Perry J, 1992). Most commonly, only the duration of muscle firing is interpreted in

gait studies. The methods of interpreting EMG amplitude is still controversies: Perry

J et al (1998) introduced the use of maximum voluntary contraction (MVC) as the

basic to normalize the EMG signals. However, MVC has not been standardized to

interpret the amplitude of EMG activity during walking and MVC is so difficult to
                                        - 133 -
perform in CP patients. That is why only a few reports have been published on the

interpretation of EMG amplitude during walking. The root mean square (RMS) is

believed to be an objective method to determine the EMG amplitude (Finsterer J,

2001) and is thought to be related to force generation level under certain constrain. In

our study, walking root mean square (RMS) was normalized with standing RMS to

determine the amplitude of walking EMG.



In addition, the power spectrum analysis of EMG signals (such as median frequency)

has been recognized as a useful tool for the assessment of muscular function (Josef F,

2001; Lu WW, 2002). The root mean square (RMS) and the median frequency (MF)

are the most frequently used parameters in the functional interpretation of skeletal

muscles because of their good objectivity and reproducibility (Ahern DK et al, 1986;

Donker SF, 2002; Gerold E et al, 1998; Ludovic D et al, 2000; Lu WW et al, 2002).

In published literature, both RMS and MF have not been applied to interpret the

walking EMG. We therefore attempted to analyze and evaluate the EMG data in

terms of the firing duration, the RMS and the MF to give a clear EMG understanding

on walking.




                                         - 134 -
5.2 Results


The aim of this study was to document alterations of gait in spastic CP patients with

healthy normal children and to evaluate the biomechanical and electromyographic

effects of different orthotic treatments on walking. The findings were discussed

according to their categories.




5.2.1 Temporal Distance Parameters

The CP’s barefoot had significant shorter stride length and slower walking speed than

the normal’s barefoot (Table 4.1). Such differences were explained by Bache CE et al

(2003) that stride length was shorter in the CP group because the second and third

ankle rockers were curtailed in the stance limb. Walking speed is the product of step

length multiplied by cadence (Maezawa Y et al, 2001; Winter DA, 1990, 1991). The

cadence measured in our patients and in the healthy control subjects showed no

significant difference. Consequently, the slower walking speed depends on the

shortening of the step length.



When we evaluated the orthotic effect on CP subjects, the DAFO significantly

increased stride length than the baseline, while the AFO did not have a significant

difference against baseline. This finding was consistently shown in many studies

(Diamond MF et al, 1990; Dieli J, et al, 1997; Radtka SA et al, 1997; Romkes J et al,

2002). Such changes approached the normative value which indicated improvement

caused by wearing the DAFO. This may be due to the better ankle positioning at the

initial contact and the greater stability provided by the DAFO (White H et al, 2002).

The same conclusion on stride length, made in other treatment like Botox injection in
                                       - 135 -
calf muscle, was found that increase in stride length would enhance the patient’s

quality of life (Metaxiotis D et al, 2002). This further supported that the increase in

stride length is one of the benefits obtained on wearing the orthosis.




5.2.2 Range of Motion Parameters


5.2.2.1 Ankle Joint


The CP significantly presented an excessive plantarflexed position at the initial

contact (AA1), in the stance phase (AA2) and during the swing phase (AA4) than the

normal. The excessive plantarflexed position causes a floor clearance problem (Gage

JR, 1991, Perry J, 1992) and the abnormal ankle positioning was due to the spasticity

of calf muscles and it leads to a shortened stride length. In addition, the CP

significantly preserved a smaller total range of the ankle joint (Table 4.7), which was

consistency with the findings by Bell K et al (2002).



Both orthoses significantly improved the excessive plantarflexed position at the initial

contact (AA1), the maximum dorsiflexion in stance phase (AA2) and the maximum

dorsiflexion in the swing phase (AA4) than their barefoot (Table 4.8 and Fig 4.1). The

same conclusions were reached by the gait studies (Buckon CE et al, 2001; Carlson

WE et al, 1997; Radtka SA et al, 1997; Romkes J et al, 2002) that orthoses could

provide better ankle positioning in the stance phase and control the excessive

plantarflexion in the swing phase.



AFO had a significant limitation of total range of ankle motion (ROM) compared to

the barefoot and the DAFO. The DAFO found significantly less restriction on ROM
                                         - 136 -
than AFO (Table 4.8 and Fig 4.1). This restricted ankle motion is due to the

mechanism of the AFO which has been reviewed in the section 2.5.3.1. From clinical

experience, such limitation of ankle movement produced an unnatural walking

pattern (Brunner R et al, 1998; Buckon CE et al, 2001) and would lead to muscle

weakness (Thomson JD et al, 1999), and muscle atrophy due to disuse. On the

contrary, more mobility allowed better ankle and knee movement and less limitation

on their daily activities especially ability to walk up or down stairs (Thomas SS et al,

2002) and the movement of sit to stand (Holly W et al, 1997).



From the ankle kinematics point of view, the DAFO would be beneficial to this type

of the patient because it is equally effective as the AFO but with a significant less

restriction on the ankle ROM.




5.2.2.2 Knee and Hip Joint


The maximum hip extension angle at stance phase (HA2) from CP was significant less

than the normal, but a significant more hip flexion angle at swing phase (HA3) than

the normal was found in the table 4.3. For the knee joint, the CP had a significant

greater knee flexion angle at the initial contact (KA1), (Table 4.5). Such differences

between the CP and normal control resulted from the secondary or coping response

to the dynamic equinus deformity (David JR, 1996; Gage JR, 1991; Gage JR et al,

1996). For patients with equinus deformity, the hip or knee had to be increasingly

flexed in order to accomplish the foot clearance during the swing phase (Goldstein M

et al, 2001).




                                         - 137 -
Concerning the orthotic effect, the hip angle at the initial contact (HA1), the maximum

hip flexion angle in swing phase (HA3) and the total range of hip motion (range) were

found to have a significant difference (Table 4.4). For Knee kinematics, both orthoses

caused a significant difference in the knee angle at terminal stance (KA4), the knee

flexion angle at initial contact (KA1), the maximum knee flexion angle in the stance

phase (KA2) and the time of knee angle at terminal stance (KAT3) in the table 4.6.

The kinematic changes at the hip and the knee joint were secondarily influenced by

the orthoses (Romkes J et al, 2002; Thomson JD et al, 1999). Therefore, the orthotic

effects on the knee and the hip will not be discussed in details.




5.2.3 Ground Reaction Force Parameters

Ground reaction force is a response to the muscular actions and the body weight

transmitted through the foot. This varies with walking speed and reflects their ability

for the loading response and push off (Meglan D et al, 1994; Perry J, 1992; White H

et al, 2002). This can also combine with kinematic data to generate the joint moment

and the joint power.



Abnormal pre-positioning of the foot at the initial contact (Table 4.7) will result in

restriction of the first rocker, the second rocker or the third rocker. The abnormal

pre-positioning altered the second peak of ground reaction force (Gage JR, 1991).

Moreover, increase in velocity also suggested higher second peak of the ground

reaction force (Perry J, 1992). Therefore, the CP subjects significantly produced a

smaller second peak at the terminal stance (GR3) than the healthy control (table 4.9)

because of limitation of the third rocker and lower speed (table 4.1).


                                         - 138 -
When evaluating the orthotic effect, only the AFO significantly increased the second

peak at the terminal stance (GR3) against their barefoot (table 4.10); while the DAFO

did not. The AFO produced the higher second peak value than the barefoot. This may

not be due to increase of the walking speed, because no significant difference on

speed was found between the AFO and their barefoot (Table 4.2). As a result, the

AFO improvement of the second peak would be due to altering the push off pattern

at the terminal stance.




5.2.4 Moment and Power Parameters

In general, specific joint kinetic patterns are associated with specific abnormalities as

defined by the kinematic patterns. The study of joint kinetic patterns and the

associated kinematic, EMG patterns as well as clinical information will help clinician

to understand the mechanisms of pathological gait. Ounpuu S et al (1996) illustrated

how to use kinetic information to evaluate the orthotic effect and provide excellent

information about the function of the ankle foot orthosis. For kinetic studies on ankle

foot orthosis, they always focused on the ankle joint (Crenshaw S et al, 2000;

Romkes J et al, 2002). Furthermore, because of the primary influence on the ankle

joint (Metaxiotis D et al, 2002; Thomson JD et al, 1999; Ounpuu S et al, 1996), the

ankle moment and the ankle power will be focused.



For the ankle joint, our CP subjects showed an abnormal pattern (time of occurrence

parameters) rather than the amplitude compared to the normal: We found a

significant early onset in the time of the maximum dorsiflexor moment in stance phase

(AMT1) and the maximum plantarflexor moment in stance phase (AMT2) (Table


                                         - 139 -
4.15). For power, the CP also showed a significant reduction of maximum ankle

power generation at stance phase (AP2) (Table 4.21). The third foot rocker of CP has

been disturbed by their equinus deformity, so CP demonstrated a significant

reduction in the ankle power (Gage JR, 1991; Romkes J et al, 2002).



Both AFO and DAFO significantly increased the maximum plantarflexor moment in

stance phase (AM2) toward the normative value. Abel MF et al (1998) and Carlson

WE et al (1997) stated that the increased plantarflexor moment in the stance phase

was the biomechanical benefit of both orthoses in the group of spastic diplegia. The

AFO significantly delayed the time of maximum dorsiflexor moment in stance phase

(AMT1) than the barefoot, which tended to our normative value (Table 4.16 and Fig

4.2).



Concerning ankle power, the AFO decreased significantly the maximum ankle power

generation in stance phase (AP2) than their barefoot and DAFO (Table 4.22 and Fig

4.3). It was suggested that increase in ankle power generation indicated an orthotic

improvement (Buckon CE et al, 2001; Carlson WE et al, 1997; Romkes J et al, 2002).

Hence, reduction in ankle power generation indicates a limitation of the AFO. This

change could be due to the plantarflexion limitation at the terminal stance by the

AFO (Crenshaw S et al, 2000).




                                       - 140 -
5.2.5 Electromyography Parameters

Spasticity can be defined as a condition in which there is a velocity-dependent

increase in resistance of a muscle group to passive stretch with a “clasp-knife” type

component associated with hyperactive tendon reflexes (Smyth MD et al, 2000). This

always alters the electromyography pattern, including phasic and amplitude.




5.2.5.1 EMG Characteristics of Cerebral Palsy


For all muscles, CP showed a significant higher median frequency (MF) and longer

total contracting duration when compared to the normal (Table 4.23 and Fig 4.4). The

higher muscle median frequency reflects more motor units were recruited for a certain

action or a certain period of time. The duration of muscle contraction reflects the

period of muscle firing in each gait cycle: the start and the end point of muscle firing

duration for all muscles group have been shown in the figure 3.38. Long contracting

duration and high MF were a characteristic of muscular spasticity in CP, as indicated

by the results of previous studies (Feng CJ et al, 1998; Gamet E et al, 1998; Policy JF

et al, 2001; Rose J et al, 1999). This may imply the muscular function was not

effective, it takes much more non-functional muscle firing for walking than normal,

and such a high MF could further indicate that the CP’s muscles will fatigue easily.



On the contrary, the RMS in CP was significantly smaller than the normal in tibialis

anterior (TA) muscle (Table 4.23 and Fig 4.4), which is consistent with the previous

findings (Sutherland DH et al, 1996).The RMS value represents the square root of the

average variance or power of the myoelectric signal (Kleine BU et al, 2000), or

simply represents the averaged EMG amplitude. Karlsson S et al (2001) found a
                                         - 141 -
strong correlation between RMS and force in the quadriceps under isokinetic

contraction, and showed the higher RMS, the more effort done by the muscles. On

the other hand, decreased RMS was indicative of muscle atrophy (Riley DA et al,

1990; Lu WW et al, 2002). Because of the muscle imbalance (spasticity) between the

calf muscle and their antagonist muscles, with calf muscles stronger than tibialis

anterior (TA), this induces the muscle wasting of the TA muscle and development of

muscle atrophy (Sutherland DA et al, 1996), and a drop in RMS value in the TA

muscle.




5.2.5.2 Orthotic Effect on EMG


In comparison among the CP groups: for the contracting duration, both orthoses

significantly reduced the excessive duration of calf muscles within a gait cycle (Fig

4.5). There was no significant difference in the duration of muscle firing between the

two orthoses.



Concerning the parameter of MF, the DAFO showed a significant improvement in the

hamstrings’ MF against the barefoot; this may be secondarily influenced by the

DAFO. In calf’s MF, the AFO was significantly lower than both barefoot and DAFO

(Fig 4.5). Such dropped MF may be due to muscle fatigue when the subjects had to

walk for several times for the gait analysis (Ebenbichler G et al, 1998; Gerdle B et al,

2000). However, we randomized the sequences (barefoot, with AFO and with DAFO)

provided two-minute break between 2 successive trials to avoid ordering and fatigue

effect, we could also make sure no detachment of electrode throughout the testing

period to prevent a poor data collection.


                                            - 142 -
Consequently, we assumed that median frequency (MF) of muscles decreased in

compared to barefoot condition should not due to the muscle fatigue but proposed

that is a benefit over the DAFO; this probably prolong the fatigue time and increases

the walking endurance for the CP patients, which was supported by the findings of

Moglia A et al (1991). The reduction of excessive MF could result from the

plantarflexion stop in the AFO.



For the parameter of RMS, no orthotic effect was found in the tibialis anterior muscle.

Interestingly, the AFO significantly increased the RMS of the hamstrings and calf

muscles against the barefoot, while the DAFO did not. A significant difference

between two orthoses was found in the RMS of calf muscle, with AFO higher than the

DAFO, the AFO improving RMS toward the normative value (Table 4.24 and Fig 4.6).

The AFO may alter the amplitude of the muscle activity, this may due to the changes

of the walking pattern.



In summary, long firing duration and high median frequency in the muscle was the

characteristic of the spasticity (Feng CJ et al, 1998; Gamet E et al, 1998; Policy JF et

al, 2001; Rose J et al, 1999). Therefore, reduced muscle firing duration or lowered

median frequency could be an indicator to inhibit the spasticity. Both the AFO and

DAFO had an advantage to reduce the excessive muscle firing duration in calf

muscles. The AFO had a benefit over the DAFO in terms of the calf muscle’s MF

and RMS, the patient wearing AFO would reduced the calf muscle’s MF and

increased calf muscle’s RMS, this imply that the calf muscle become more effective

in muscle power generation (less motor unit recruit for larger force level generation).




                                         - 143 -
5.3 Limitations of Study


More CP and normal subjects are needed to meet the need of statistical requirement

A control group of wider age range is desirable. With the provision of one force

platform, the ground reaction force data for the left and right limbs were recorded

separately. Hence, in this study, double support phase which reflected the stability of

walking could not be studied.



Some parameters were found significantly different between the left and right limbs

(Appendix 3.4.1 – 3.4.12) for both CP and normal subjects. Therefore, the left and

right limbs data were averaged to a single mean in our study; although we admit that

averaged data could affect the results. However, the benefit of averaging the left and

right limb in our study was to prevent cumbersome data calculation when too many

groups of data were compared (Crenshaws S et al, 2000; Romkes J et al, 2002).




                                        - 144 -
5.4 Clinical Significance


Ankle foot orthosis (AFO) and dynamic ankle foot orthosis (DAFO) are commonly

prescribed for patients with dynamic equinus deformity, and aimed to improve

pathological gait. The normal walking pattern has five major attributes which are

frequently lost in pathological gait (Gage JR, 1991): 1) stability in stance, 2)

sufficient foot clearance during swing, 3) appropriate swing phase pre-positioning of

the foot, 4) an adequate step length and 5) energy conservation. All of these attributes

should be considered to decide which orthosis is better.



In this study, both AFO and DAFO achieved the primary function (control equinus):

they could sufficiently control foot positioning in the stance phase and foot clearance

in the swing phase. They also reduced the excessive muscle firing duration of the calf

muscles within a gait cycle, which may reflect a conservation of the muscle activity.

The advantages of the AFO was an increase in the second peak of ground reaction

force and a reduction in the median frequency of the calf muscle, whereas the DAFO

had less restriction on ankle movement which facilitates daily activities and

minimize muscle atrophy.



The DAFO may be preferred to AFO because it is equally effective in controlling the

equinus, has a less restriction on ankle motion and is lighter and less bulky.



Barnett SL et al (1993) demonstrated that oxygen consumption per unit distance and

oxygen consumption rate had significant positive linear correlations with added

weight (P = 0.001, P = 0.007, respectively). They concluded that the advantage of a

less bulky and lighter orthosis is that it can minimize oxygen consumption (Barnett
                                         - 145 -
SL et al, 1993), which means conservation of energy consumption during walking.



All subjects agreed that wearing DAFO was more comfortable than the AFO. As for

appearance, most of them (7 girls and 4 boys) preferred the DAFO due to the better

look; while the rest (5 boys) had no preference..



Furthermore, both median frequency (MF) and root mean square (RMS) value are

objective methods to interpret the characteristics of EMG and are recognized as a

useful tool for the assessment of muscular function (Josef F, 2001; Lu WW, 2002),

because of their good objectivity and reproducibility. Employing RMS and the MF as

parameters could serve as a basis for further gait studies.




                                         - 146 -
                                 CHAPTER 6


                                CONCLUSIONS



In Hong Kong, both conventional ankle foot orthosis (AFO) and supramalleolar type

of dynamic ankle foot orthosis (DAFO) are prescribed for the CP patient with

dynamic equinus. They control the equinus deformity with two very different

mechanisms. Therefore, it is necessary to evaluate the orthotic effect on the CP

patient using gait analysis.



Our CP subjects presented a significant shorter stride length, slower walking speed

and abnormal ankle positioning than our normal control. Due to the abnormal ankle

position at terminal stance, it led to lower the second peak of ground reaction force

and generate less power at the terminal stance than our healthy control. More

interestingly, it was found that the muscle firing duration and muscle median

frequency (MF) for all muscle groups were much higher than the normal, which

could be the characteristics of spasticity. These alterations provide better

understanding on the characteristics of diplegic CP patients and thus better treatment

prescription.



This study showed that both AFO and DAFO can achieve the primary function:

improving the ankle positioning at initial contact and during swing phase. They also

improved the plantarflexor moment in stance and reduced the excessive muscle firing

duration in the calf muscles.



                                        - 147 -
The advantage of the AFO was an increase in the second peak of ground reaction

force and a reduction in the median frequency of the calf muscle, but the limitation

of this orthosis was the significantly reduced ankle motion which may induce muscle

weakness. The advantages of the DAFO were an increase in stride length which

reflected better stability, and less restriction on ankle movement which facilitate

daily activities (up to stand motion, up and down the stair) and minimize muscle

weakness. The DAFO is preferred because it is equally effective in controlling the

equinus, with less restriction on ankle motion as well as being lighter and less bulky.



This study comprehensively evaluated the advantages and disadvantages of the AFO

and DAFO, and provided an orthotic suggestion for patients with dynamic equinus

deformity. Furthermore, this study tried to adopt the use of root mean square (RMS)

and median frequency (MF) to gait study because of their objectivity and

repeatability. It was admitted that the clinical interpretation of RMS and MF in this

study is a preliminary.




                                         - 148 -
                                APPENDICES

3.1a Consent form for patient with cerebral palsy (English version)

3.1b Consent form for patient with cerebral palsy (Chinese version)

3.2a Consent form for normal (English version)

3.2b Consent form for normal (Chinese version)

3.3 Flow of patient with cerebral palsy

3.4.1 Paired t-test for the left and right sides in the temporal distance parameters

3.4.2 Paired t-test for the left and right sides in the ROM of hip parameters

3.4.3 Paired t-test for the left and right sides in the ROM of knee parameters

3.4.4 Paired t-test for the left and right sides in the ROM of ankle parameters

3.4.5 Paired t-test for the left and right sides in the ground reaction force parameters

3.4.6 Paired t-test for the left and right sides in the moment of hip parameters

3.4.7 Paired t-test for the left and right sides in the moment of knee parameters

3.4.8 Paired t-test for the left and right sides in the moment of ankle parameters

3.4.9 Paired t-test for the left and right sides in the power of hip parameters

3.4.10 Paired t-test for the left and right sides in the power of knee parameters

3.4.11 Paired t-test for the left and right sides in the power of ankle parameters

3.4.12 Paired t-test for the left and right sides in the EMG parameters

4.1 SPSS results of the temporal distance parameters using ANOVA with repeated

   measure for CP conditions

4.2 SPSS results of the ROM of hip parameters using ANOVA with repeated measure

   for CP conditions
                                          - 149 -
4.3 SPSS results of the ROM of knee parameters using ANOVA with repeated

   measure for CP conditions

4.4 SPSS results of the ROM of ankle parameters using ANOVA with repeated

   measure for CP conditions

4.5 SPSS results of the ground reaction force parameters using ANOVA with

   repeated measure for CP conditions

4.6 SPSS results of the moment of hip parameters using ANOVA with repeated

   measure for CP conditions

4.7 SPSS results of the moment of knee parameters using ANOVA with repeated

   measure for CP conditions

4.8 SPSS results of the moment of ankle parameters using ANOVA with repeated

   measure for CP conditions

4.9 SPSS results of the power of hip parameters using ANOVA with repeated

   measure for CP conditions

4.10 SPSS results of the power of knee parameters using ANOVA with repeated

    measure for CP conditions

4.11 SPSS results of the power of ankle parameters using ANOVA with repeated

    measure for CP conditions

4.12 SPSS results of the EMG parameters (for all channels) using ANOVA with

    repeated measure for CP conditions

4.13 Curve of ankle motion among barefoot, AFO, DAFO and normal groups.

4.14 Curve of ankle moment among barefoot, AFO, DAFO and normal groups.

4.15 Curve of ankle power for among barefoot, AFO, DAFO and normal groups.


                                        - 150 -
Appendix 3.1a Consent form for patient with cerebral palsy (English version)



                                      Consent Form


                                The Duchess of Kent Children Hospital




Research title: Gait analysis approach: to compare the immediate effect of rigid
ankle foot orthosis (AFO) and supramalleolar type dynamic ankle foot orthosis
(DAFO) in diplegic cerebral palsy children.


I, __________________ (name), hereby consent to participate in as a subject for the
research project entitled “Gait analysis approach: to compare the immediate effect of
rigid ankle foot orthosis (AFO) and supramalleolar type dynamic ankle foot orthosis
(DAFO) in diplegic cerebral palsy children”. The participant will wear two types of
orthosis for comparing. I also understand the effect and details of the experimental
procedures that have been explained to me.


I understand that I have the right to discontinue, with no reasons given, my
participation any time, even during the experiment. I realized that any findings of the
study will be only used by The Department of Orthopaedic Surgery of The
University of Hong Kong for research purpose.



Signature     ______________
Date         ______________
Researcher    ______________
Witness      ________________




                                        - 151 -
Appendix 3.1b Consent form for patient with cerebral palsy (Chinese version)




                                         參




    :               兩                              兩                力

    ______________ ( 參           /            )
______________ (     參               )

            參           兩   不                      行
              了


                    行            不                        療
    料




參               ______________
                ______________
                ______________
見               ______________




                                         - 152 -
Appendix 3.2a Consent form for normal (English version)



                                       Consent Form


                                The Duchess of Kent Children Hospital



Research title:   Gait analysis approach: to compare the immediate effect of rigid
ankle foot orthosis (AFO) and supramalleolar type dynamic ankle foot orthosis
(DAFO) in diplegic cerebral palsy children.


I, __________________ (name), hereby consent to participate in as a subject for the
research project entitled “Gait analysis approach: to compare the immediate effect of
rigid ankle foot orthosis (AFO) and supramalleolar type dynamic ankle foot orthosis
(DAFO) in diplegic cerebral palsy children”. The participant will have a session of
gait analysis with barefoot in order to create a normal control (aged from 6 to 13) for
the cerebral palsy patient. I also understand the effect and details of the experimental
procedures that have been explained to me.


I understand that I have the right to discontinue, with no reasons given, my
participation any time, even during the experiment. I realized that any findings of the
study will be only used by The Department of Orthopaedic Surgery of The
University of Hong Kong for research purpose.



Signature         ________________
Date             ________________
Researcher        ________________
Witness          ________________




                                         - 153 -
Appendix 3.2b Consent form for normal (Chinese version)




      :            兩                           兩          力



    ______________ ( 參          /         )
  了       立六                                                  參
                   行       行
               了


                   行                           料




參              ______________
               ______________
               ______________
               ______________
見              ______________




                                     - 154 -
Appendix 3.3 Flow of patient with cerebral palsy


  Flow of Patient with Cerebral Palsy


                                 Aged 4-13
 Subject criteria                Spasticity at calf muscles
                                 No significant coronal or rotational
                                 deformity
                                 Independent walking without aids
                                 No ankle surgery done
                                 Pre-botox or post-botox 5 months




                              If patient agree to be a subject




     Go to P&O dept at MMRC
     and provide AFO and DAFO (2weeks)




     Go back to Gait Laboratory at DKCH for gait analysis (1 week)



   Overall time required for the study will be 2-3 weeks




                                      - 155 -
Appendix 3.4.1 Paired t-test for the left and right sides in the temporal distance

parameters



                                 Left                  Right
     Bare (n=16)                                                    t-value p-value
                         Mean           SD       Mean          SD
  Stride length (m)       0.74      0.192        0.74      0.195    1.065    0.304
    Stride time (s)       1.02      0.272        1.02      0.278    0.463    0.650
 Walking speed (m/s)      0.78      0.251        0.78      0.258    -0.415   0.684
    Stance time(s)        0.63      0.038        0.62      0.032    1.243    0.233
    Swing time (s)        0.39      0.242        0.40      0.252    -0.312   0.760
  Stance/swing ratio      1.97      0.830        2.03      0.966    -0.948   0.358
 Cadence (steps/min) 103.08 20.612 104.08 21.447                    -1.706   0.109


                                 Left                  Right
     AFO (n=16)                                                     t-value p-value
                         Mean           SD       Mean          SD
  Stride length (m)       0.81      0.194        0.81      0.198    0.963    0.351
    Stride time (s)       1.07      0.291        1.06      0.296    0.726    0.479
 Walking speed (m/s)      0.81      0.263        0.80      0.259    0.149    0.884
    Stance time(s)        0.63      0.044        0.63      0.047    -0.093   0.927
    Swing time (s)        0.44      0.258        0.43      0.265    0.764    0.456
  Stance/swing ratio      1.71      0.626        1.85      1.029    -0.921   0.372
 Cadence (steps/min)     98.08      17.877       98.91    18.045    -0.909   0.378


                                 Left                  Right
    DAFO (n=16)                                                     t-value p-value
                         Mean           SD       Mean          SD
  Stride length (m)       0.85      0.170        0.85      0.178    0.587    0.566
    Stride time (s)       1.04      0.264        1.04      0.260    0.364    0.721
 Walking speed (m/s)      0.87      0.255        0.86      0.263    0.921    0.371
    Stance time(s)        0.63      0.033        0.63      0.030    0.490    0.631
    Swing time (s)        0.40      0.238        0.41      0.239    -0.065   0.949
  Stance/swing ratio      2.09      1.264        1.95      0.895    1.139    0.273
 Cadence (steps/min) 101.54 21.764 101.23 20.764                    0.453    0.657


                                             - 156 -
                             Left                  Right
  Normal (n=18)                                                 t-value p-value
                      Mean          SD       Mean          SD
 Stride length (m)    1.06      0.140        1.05      0.140    2.233    0.039
  Stride time (s)     0.95      0.086        0.95      0.085    0.789    0.441
Walking speed (m/s)   1.14      0.140        1.13      0.143    2.970    0.009
  Stance time(s)      0.61      0.016        0.61      0.014    0.000    1.000
  Swing time (s)      0.34      0.088        0.34      0.088    0.642    0.529
Stance/swing ratio    1.90      0.557        1.92      0.550    -0.579   0.570
Cadence (steps/min) 105.58      9.481       106.00     9.533    -0.896   0.383




                                         - 157 -
Appendix 3.4.2 Paired t-test for the left and right sides in the ROM of hip parameters


                           Left            Right
  Bare (n=16)                                               t-value p-value
                   Mean           SD   Mean         SD
 HA1 (degree)      38.08      8.950    38.57       8.906    -0.252   0.804
 HA2 (degree)      -7.06      9.750    -5.20       8.227    -1.108   0.285
 HA3 (degree)      42.17     10.985    44.22       9.220    -1.017   0.325
 HAT2 (%GC)        27.69      2.089    27.69       2.892    0.000    1.000
 HAT3 (%GC)        47.25      1.238    46.81       1.047    2.150    0.048
 Range (degree)    49.23      9.818    49.42       7.440    -0.100   0.922


                           Left            Right
  AFO (n=16)                                                t-value p-value
                   Mean           SD   Mean         SD
 HA1 (degree)      43.06      9.370    42.52       10.807   0.337    0.741
 HA2 (degree)      -7.21      8.246    -6.17       10.620   -0.508   0.619
 HA3 (degree)      47.25     10.939    46.46       10.031   0.441    0.666
 HAT2 (%GC)        28.13      2.605    27.94       2.816    0.426    0.676
 HAT3 (%GC)        47.44      1.209    47.25       1.183    0.824    0.423
 Range (degree)    54.46     10.482    52.63       6.379    0.653    0.523


                           Left            Right
 DAFO (n=16)                                                t-value p-value
                   Mean           SD   Mean         SD
 HA1 (degree)      43.00     11.308    43.14       8.835    -0.071   0.945
 HA2 (degree)      -6.96      9.368    -6.34       9.093    -0.343   0.737
 HA3 (degree)      47.17     11.519    48.48       10.522   -0.927   0.368
 HAT2 (%GC)        27.75      2.113    27.19       3.250    1.209    0.245
 HAT3 (%GC)        46.88      1.455    46.75       1.291    0.368    0.718
 Range (degree)    54.13     10.581    54.82       9.277    -0.372   0.715


                           Left            Right
 Normal (n=18)                                              t-value p-value
                   Mean           SD   Mean         SD
 HA1 (degree)      34.65      5.732    32.56       6.883    1.640    0.119
 HA2 (degree)     -14.30      6.673    -14.96      5.275    0.693    0.498
 HA3 (degree)      37.77      6.452    35.57       7.227    2.090    0.052

                                         - 158 -
HAT2 (%GC)       28.06   0.725   28.00       1.188   0.212    0.834
HAT3 (%GC)       46.33   1.455   46.44       1.504   -0.437   0.668
Range (degree)   52.06   8.321   50.53       6.316   1.228    0.236




                                   - 159 -
Appendix 3.4.3 Paired t-test for the left and right sides in the ROM of knee

parameters


                          Left                    Right
  Bare (n=16)                                                  t-value   p-value
                  Mean           SD     Mean              SD
 KA1 (degree)     22.75      8.852      23.23         9.298    -0.209    0.837
 KA2 (degree)     21.12      10.327     21.83        10.670    -0.323    0.751
 KA3 (degree)      2.63      9.615      4.75         10.494    -1.045    0.312
 KA4 (degree)     56.75      10.542     58.40         8.592    -0.624    0.542
 KAT2 (%GC)        5.19      1.328      5.25          1.238    -0.436    0.669
 KAT3 (%GC)       19.50      3.830      19.25         4.568    0.480     0.638
 KAT4 (%GC)       41.44      2.632      41.75         2.082    -0.735    0.474


                          Left                    Right
  AFO (n=16)                                                   t-value   p-value
                  Mean           SD     Mean              SD
 KA1 (degree)     25.60      7.302      25.82         7.463    -0.139    0.891
 KA2 (degree)     25.70      7.214      25.68         9.804    0.011     0.991
 KA3 (degree)      0.97      8.550      3.46          8.383    -1.298    0.214
 KA4 (degree)     60.54      8.805      60.20         6.811    0.145     0.887
 KAT2 (%GC)        5.25      1.238      5.19          1.721    0.212     0.835
 KAT3 (%GC)       20.56      3.366      20.38         3.631    0.341     0.738
 KAT4 (%GC)       41.69      1.622      40.75         2.049    1.749     0.101


                          Left                    Right
 DAFO (n=16)                                                   t-value   p-value
                  Mean           SD     Mean              SD
 KA1 (degree)     27.79      10.743     28.28         8.726    -0.219    0.829
 KA2 (degree)     28.00      10.322     26.86        11.895    0.351     0.731
 KA3 (degree)      4.58      7.595      6.56         11.312    -0.849    0.409
 KA4 (degree)     65.06      9.121      66.76        10.142    -0.754    0.463
 KAT2 (%GC)        5.38      1.500      5.44          1.504    -0.324    0.751
 KAT3 (%GC)       19.19      4.020      17.56         5.253    1.302     0.212
 KAT4 (%GC)       40.88      1.928      41.13         1.784    -0.553    0.588


                                        - 160 -
                        Left                  Right
Normal (n=18)                                              t-value   p-value
                Mean           SD   Mean              SD
KA1 (degree)    6.38       4.000    2.87          8.552    2.205     0.041
KA2 (degree)    18.56      5.644    17.58         8.344    0.700     0.493
KA3 (degree)    3.63       4.957    2.70          7.243    0.617     0.546
KA4 (degree)    61.63      5.847    57.90         8.890    2.128     0.048
KAT2 (%GC)      7.83       0.924    7.83          0.924    0.000     1.000
KAT3 (%GC)      22.00      1.495    21.61         1.852    0.979     0.341
KAT4 (%GC)      38.06      0.873    37.89         0.832    0.678     0.507




                                    - 161 -
Appendix 3.4.4 Paired t-test for the left and right sides in the ROM of ankle

parameters


                             Left                     Right
   Bare (n=16)                                                     t-value   p-value
                     Mean            SD       Mean            SD
  AA1 (degree)       -5.81          6.977     -7.94       8.252    1.336     0.202
  AA2 (degree)       2.32       10.839         1.77       10.577   0.312     0.759
  AA3 (degree)      -20.15      13.952       -20.33       11.343   0.088     0.931
  AA4 (degree)       -4.01          8.743     -5.13       8.728    0.871     0.398
  AAT2 (%GC)         21.38          8.609     21.13       7.544    0.374     0.713
  AAT3 (%GC)         33.38          6.185     33.75       5.663    -0.808    0.432
  AAT4 (%GC)         45.63          1.455     45.38       1.455    0.889     0.388
 Range (degree)      22.46          9.245     22.10       6.714    0.218     0.830


                             Left                     Right
   AFO (n=16)                                                      t-value   p-value
                     Mean            SD       Mean            SD
  AA1 (degree)       3.42           5.682      3.13       5.645    0.211     0.836
  AA2 (degree)       11.27          6.310     12.03       6.518    -0.575    0.574
  AA3 (degree)       -0.69          6.186      0.87       5.562    -0.945    0.360
  AA4 (degree)       3.34           5.193      3.24       5.138    0.082     0.936
  AAT2 (%GC)         24.75          3.715     25.50       2.608    -0.927    0.368
  AAT3 (%GC)         36.19          2.562     36.81       2.738    -1.908    0.076
  AAT4 (%GC)         45.81          2.971     45.44       2.756      1       0.333
 Range (degree)      11.96          7.686     11.16       4.586    0.528     0.605


                             Left                     Right
  DAFO (n=16)                                                      t-value   p-value
                     Mean            SD       Mean            SD
  AA1 (degree)       3.42           6.966      4.47       5.035    -0.543    0.595
  AA2 (degree)       13.02          7.397     15.17       5.826    -1.494    0.156
  AA3 (degree)       -6.48          9.408     -3.86       5.922    -1.231    0.237
  AA4 (degree)       4.72           7.873      6.53       4.841    -1.053    0.309
  AAT2 (%GC)         21.75          8.037     21.69       7.948    0.090     0.929
  AAT3 (%GC)         35.81          1.834     35.88       1.784    -0.212    0.835

                                            - 162 -
AAT4 (%GC)       46.19          1.760     45.63       1.928    1.379     0.188
Range (degree)   19.50          6.610     19.03       3.823    0.317     0.756


                         Left                     Right
Normal (n=16)                                                  t-value   p-value
                 Mean            SD       Mean            SD
AA1 (degree)     1.11           5.345      0.31       4.049    0.881     0.391
AA2 (degree)     13.04          5.139     11.23       4.950    2.832     0.011
AA3 (degree)     -17.36         7.661    -17.74       7.478    0.359     0.724
AA4 (degree)     2.46           4.978      1.82       4.350    0.745     0.467
AAT2 (%GC)       20.50          5.874     20.89       5.476    -0.562    0.581
AAT3 (%GC)       33.56          0.856     33.78       0.943    -1.166    0.260
AAT4 (%GC)       46.06          2.182     46.06       2.128    0.000     1.000
Range (degree)   30.40          5.454     28.97       6.218    1.520     0.147




                                        - 163 -
Appendix 3.4.5 Paired t-test for the left and right sides in the ground reaction force
parameters
                             Left                     Right
   Bare (n=16)                                                     t-value   p-value
                    Mean             SD     Mean              SD
  GR1 (%BW)          1.15           0.209    1.12         0.336    0.473     0.643
  GR2 (%BW)          0.72           0.121    0.69         0.168    0.722     0.481
  GR3 (%BW)          0.89           0.133    0.91         0.093    -1.231    0.237
  GRT1 (%GC)         7.69           3.420    7.69         4.223    0.000     1.000
  GRT2 (%GC)         14.81          4.446   14.63         5.175    0.380     0.709
  GRT3 (%GC)         22.44          4.163   22.56         3.444    -0.217    0.831


                             Left                     Right
  AFO (n=16)                                                       t-value   p-value
                    Mean             SD     Mean              SD
  GR1 (%BW)          1.23           0.329    1.28         0.382    -0.887    0.389
  GR2 (%BW)          0.72           0.186    0.71         0.208    0.548     0.592
  GR3 (%BW)          1.00           0.054    0.96         0.078    2.261     0.039
  GRT1 (%GC)         7.75           2.769    7.50         3.830    0.532     0.603
  GRT2 (%GC)         14.00          3.652   13.50         5.073    0.826     0.422
  GRT3 (%GC)         21.00          3.464   21.63         4.646    -0.837    0.416


                             Left                     Right
 DAFO (n=16)                                                       t-value   p-value
                    Mean             SD     Mean              SD
  GR1 (%BW)          1.30           0.345    1.28         0.326    0.307     0.763
  GR2 (%BW)          0.68           0.186    0.68         0.202    -0.023    0.982
  GR3 (%BW)          0.92           0.157    0.90         0.172    0.472     0.644
  GRT1 (%GC)         7.19           2.613    7.44         3.326    -0.329    0.747
  GRT2 (%GC)         14.69          3.807   14.25         4.810    0.390     0.702
  GRT3 (%GC)         22.00          3.225   22.00         3.652    0.000     1.000


                             Left                     Right
 Normal (n=16)                                                     t-value   p-value
                    Mean             SD     Mean              SD
  GR1 (%BW)          1.15           0.099    1.11         0.118    1.783     0.092
  GR2 (%BW)          1.13           1.815    0.74         0.088    0.941     0.360
  GR3 (%BW)          0.98           0.096    1.00         0.061    -1.388    0.183

                                            - 164 -
GRT1 (%GC)   7.44    1.042    7.67     0.908   -1.166   0.260
GRT2 (%GC)   16.06   1.830   15.61     1.145   1.325    0.203
GRT3 (%GC)   24.17   0.786   24.39     0.608   -1.288   0.215




                             - 165 -
Appendix 3.4.6 Paired t-test for the left and right sides in the moment of hip
parameters


                             Left                     Right
   Bare (n=16)                                                     t-value   p-value
                     Mean            SD      Mean             SD
 HM1 (Nm/BW)         1.17           0.405     1.12        0.874    0.360     0.724
 HM2 (Nm/BW)         -0.90          0.401     -0.89       0.467    -0.096    0.925
 HMT1 (%GC)          4.00           1.826     3.94        1.982    0.164     0.872
 HMT2 (%GC)          26.88          1.500    26.56        2.032    0.618     0.546


                             Left                     Right
  AFO (n=16)                                                       t-value   p-value
                     Mean            SD      Mean             SD
 HM1 (Nm/BW)         1.23           0.577     1.42        0.749    -1.600    0.130
 HM2 (Nm/BW)         -0.83          0.335     -0.82       0.334    -0.155    0.879
 HMT1 (%GC)          5.38           1.500     4.63        1.455    1.419     0.176
 HMT2 (%GC)          26.31          2.496    26.81        1.642    -0.816    0.427


                             Left                     Right
 DAFO (n=16)                                                       t-value   p-value
                     Mean            SD      Mean             SD
 HM1 (Nm/BW)         1.50           0.678     1.53        0.916    -0.168    0.869
 HM2 (Nm/BW)         -0.94          0.391     -0.92       0.452    -0.404    0.692
 HMT1 (%GC)          4.38           1.928     4.06        1.569    1.159     0.264
 HMT2 (%GC)          26.69          1.580    25.88        2.062    1.619     0.126


                             Left                     Right
 Normal (n=18)                                                     t-value   p-value
                     Mean            SD      Mean             SD
 HM1 (Nm/BW)         1.42           0.431     1.48        0.393    -0.920    0.370
 HM2 (Nm/BW)         -1.47          0.299     -1.49       0.295    0.435     0.669
 HMT1 (%GC)          4.61           1.145     4.67        1.237    -0.270    0.790
 HMT2 (%GC)          26.61          0.698    26.78        0.732    -1.144    0.269




                                            - 166 -
Appendix 3.4.7 Paired t-test for the left and right sides in the moment of knee

parameters


                           Left                     Right
  Bare (n=16)                                                    t-value   p-value
                   Mean            SD     Mean              SD
KM1 (Nm/BW)        -0.75          0.539   -0.56         0.481    -1.141    0.272
KM2 (Nm/BW)         0.31          0.300    0.38         0.296    -1.172    0.260
KMT1 (%GC)          5.19          4.564    5.69         4.715    -0.641    0.531
KMT2 (%GC)         27.94          1.652   27.19         1.870    1.732     0.104


                           Left                     Right
  AFO (n=16)                                                     t-value   p-value
                   Mean            SD     Mean              SD
KM1 (Nm/BW)        -0.47          0.343   -0.63         0.369    1.977     0.067
KM2 (Nm/BW)         0.31          0.128    0.31         0.148    0.227     0.824
KMT1 (%GC)          7.75          5.310    7.38         5.201    0.841     0.414
KMT2 (%GC)         28.50          2.450   28.19         2.228    0.581     0.570


                           Left                     Right
 DAFO (n=16)                                                     t-value   p-value
                   Mean            SD     Mean              SD
KM1 (Nm/BW)        -0.64          0.337   -0.72         0.529    0.903     0.381
KM2 (Nm/BW)         0.41          0.237    0.37         0.215    0.700     0.495
KMT1 (%GC)          4.38          3.384    4.00         2.556    1.145     0.270
KMT2 (%GC)         28.13          1.500   27.31         2.182    1.403     0.181


                           Left                     Right
Normal (n=18)                                                    t-value   p-value
                   Mean            SD     Mean              SD
KM1 (Nm/BW)        -0.70          0.248   -0.77         0.242    1.217     0.240
KM2 (Nm/BW)         0.64          0.175    0.60         0.242    0.933     0.364
KMT1 (%GC)          3.00          1.372    2.83         1.383    0.589     0.564
KMT2 (%GC)         27.44          0.705   27.72         1.074    -1.230    0.236




                                          - 167 -
Appendix 3.4.8 Paired t-test for the left and right sides in the moment of ankle
parameters


                            Left                      Right
  Bare (n=16)                                                      t-value   p-value
                   Mean             SD        Mean            SD
AM1 (Nm/BW)         0.03           0.085       0.04       0.054    -0.466    0.648
AM2 (Nm/BW)         0.80           0.321       0.71       0.316    1.815     0.089
AMT1 (%GC)          1.13           0.342       1.00       0.000    1.464     0.164
AMT2 (%GC)         14.56           7.183      16.44       7.703    -1.571    0.137


                            Left                      Right
  AFO (n=16)                                                       t-value   p-value
                   Mean             SD        Mean            SD
AM1 (Nm/BW)         -0.03          0.263       0.08       0.250    -1.599    0.131
AM2 (Nm/BW)         0.96           0.339       1.14       0.396    -2.140    0.049
AMT1 (%GC)          1.56           0.727       1.38       0.500    1.145     0.270
AMT2 (%GC)         17.44           6.572      16.56       7.339    0.979     0.343


                            Left                      Right
 DAFO (n=16)                                                       t-value   p-value
                   Mean             SD        Mean            SD
AM1 (Nm/BW)         0.09           0.234       0.13       0.292    -1.168    0.261
AM2 (Nm/BW)         1.19           0.495       1.16       0.378    0.315     0.757
AMT1 (%GC)          1.13           0.342       1.13       0.342    0.000     1.000
AMT2 (%GC)         14.31           7.709      15.13       7.813    -0.792    0.441


                            Left                      Right
 Normal(n=18)                                                      t-value   p-value
                   Mean             SD        Mean            SD
AM1 (Nm/BW)         -0.07          0.163       0.04       0.232    -1.610    0.126
AM2 (Nm/BW)         0.89           0.140       0.92       0.113    -0.913    0.374
AMT1 (%GC)          1.39           0.698       1.33       0.594    0.325     0.749
AMT2 (%GC)         22.22           4.797      22.22       4.845    0.000     1.000




                                           - 168 -
Appendix 3.4.9 Paired t-test for the left and right sides in the power of hip parameters


                            Left                  Right
  Bare (n=16)                                                  t-value   p-value
                    Mean            SD     Mean           SD
HP1 (Watt/BW)       2.33           1.498   2.33       1.847    -0.013    0.989
HP2 (Watt/BW)       -1.29          0.823   -1.41      1.196    0.545     0.594
 HPT1 (%GC)         5.06           2.568   4.75       3.587    0.459     0.653
 HPT2 (%GC)         24.50          2.280   24.06      2.840    0.835     0.417


                            Left                  Right
  AFO (n=16)                                                   t-value   p-value
                    Mean            SD     Mean           SD
HP1 (Watt/BW)       2.77           1.584   2.85       1.626    -0.199    0.845
HP2 (Watt/BW)       -1.18          0.867   -1.35      0.867    0.839     0.414
 HPT1 (%GC)         6.00           2.366   5.75       2.408    0.719     0.483
 HPT2 (%GC)         23.63          3.030   24.50      3.098    -1.537    0.145


                            Left                  Right
 DAFO (n=16)                                                   t-value   p-value
                    Mean            SD     Mean           SD
HP1 (Watt/BW)       3.01           1.718   3.42       2.970    -0.789    0.442
HP2 (Watt/BW)       -1.23          0.761   -1.44      0.908    1.686     0.113
 HPT1 (%GC)         5.81           2.228   5.06       2.489    1.695     0.111
 HPT2 (%GC)         23.75          2.769   26.13      3.667    -1.652    0.119


                            Left                  Right
 Normal (n=18)                                                 t-value   p-value
                    Mean            SD     Mean           SD
HP1 (Watt/BW)       2.07           0.635   2.00       0.701    0.545     0.593
HP2 (Watt/BW)       -2.09          0.694   -2.03      0.630    -0.38     0.708
 HPT1 (%GC)         7.00           2.000   7.33       1.879    -0.566    0.579
 HPT2 (%GC)         24.89          1.023   25.00      1.237    -0.399    0.695




                                           - 169 -
Appendix 3.4.10 Paired t-test for the left and right sides in the power of knee
parameters


                             Left                       Right
   Bare (n=16)                                                       t-value   p-value
                     Mean            SD        Mean             SD
 KP1 (Watt/BW)        0.29          0.588       0.40        0.563    -0.887    0.389
 KP2 (Watt/BW)       -1.02          0.780       -1.24       0.853     0.77     0.453
 KP3 (Watt/BW)       -1.60          0.852       -1.47       0.784    -0.64     0.532
  KPT1 (%GC)          4.50          2.633       4.69        2.892    -0.824    0.423
  KPT2 (%GC)         28.56          5.597      29.13        2.363    -0.399    0.696
  KPT3 (%GC)         48.63          1.821      49.38        0.885    -1.379    0.188


                             Left                       Right
  AFO (n=16)                                                         t-value   p-value
                     Mean            SD        Mean             SD
 KP1 (Watt/BW)        0.26          0.467       0.75        0.881    -2.37     0.032
 KP2 (Watt/BW)       -1.71          1.893       -1.50       1.276    -0.357    0.726
 KP3 (Watt/BW)       -1.49          1.063       -1.49       0.909    -0.003    0.997
  KPT1 (%GC)          4.13          1.148       4.81        2.926    -1.139    0.273
  KPT2 (%GC)         29.56          1.825      29.19        2.105    0.739     0.471
  KPT3 (%GC)         48.69          1.078      49.25        1.065    -2.058    0.057


                             Left                       Right
  DAFO (n=16)                                                        t-value   p-value
                     Mean            SD        Mean             SD
 KP1 (Watt/BW)        0.76          0.809       1.04        1.341    -1.369    0.191
 KP2 (Watt/BW)       -1.51          0.899       -1.22       0.791     -1.5     0.154
 KP3 (Watt/BW)       -1.67          1.198       -1.67       1.100    -0.021    0.984
  KPT1 (%GC)          5.31          3.092       5.25        2.955    0.085     0.933
  KPT2 (%GC)         29.63          1.962      28.81        2.344    1.232     0.237
  KPT3 (%GC)         48.94          1.124      49.00        0.966    -0.202    0.843


                             Left                       Right
 Normal (n=18)                                                       t-value   p-value
                     Mean            SD        Mean             SD
 KP1 (Watt/BW)        1.90          1.079       2.22        1.089    -1.374    0.187
 KP2 (Watt/BW)       -2.10          0.699       -1.88       0.704    -1.585    0.131

                                            - 170 -
KP3 (Watt/BW)   -3.00   0.818       -2.49   0.883   -3.477   0.003
KPT1 (%GC)      3.78    1.215       3.56    0.705   0.776    0.449
KPT2 (%GC)      28.67   0.840      28.83    0.707   -1.144   0.269
KPT3 (%GC)      47.39   0.979      47.44    0.922   -0.212   0.834




                                - 171 -
Appendix 3.4.11 Paired t-test for the left and right sides in the power of ankle
parameters


                             Left                       Right
  Bare (n=16)                                                           t-value   p-value
                     Mean            SD         Mean             SD
AP1 (Watt/BW)        -1.47          1.384       -1.05           0.799    -1.2     0.249
AP2 (Watt/BW)         0.90          0.487        0.74           0.406   1.222      0.24
  APT1 (%GC)          9.63          7.228        8.94           6.846   0.639     0.532
  APT2 (%GC)         25.19          8.408       28.75           4.203   -1.533    0.146


                             Left                       Right
  AFO (n=16)                                                            t-value   p-value
                     Mean            SD         Mean             SD
AP1 (Watt/BW)        -0.77          0.483       -1.38           1.425   1.729     0.104
AP2 (Watt/BW)         0.48          0.201        0.58           0.276   -2.026    0.061
  APT1 (%GC)         11.44          8.016       11.31           7.872    0.19     0.852
  APT2 (%GC)         28.06          3.768       28.06           5.335     0         1


                             Left                       Right
 DAFO (n=16)                                                            t-value   p-value
                     Mean            SD         Mean             SD
AP1 (Watt/BW)        -1.75          1.565       -1.55           1.102   -0.842    0.413
AP2 (Watt/BW)         0.89          0.642        0.89           0.593   0.004     0.997
  APT1 (%GC)          8.75          6.728        8.50           6.460   0.385     0.705
  APT2 (%GC)         25.25          8.737       24.75           8.918   0.913     0.376


                             Left                       Right
 Normal (n=18)                                                          t-value   p-value
                     Mean            SD         Mean             SD
AP1 (Watt/BW)        -0.93          0.428       -0.87           0.468   -0.628    0.539
AP2 (Watt/BW)         1.16          0.593        1.43           0.756   -1.918    0.072
  APT1 (%GC)         12.61          7.382       12.33           6.920   0.719     0.482
  APT2 (%GC)         27.61          2.304       28.56           0.984   -1.598    0.129




                                            - 172 -
Appendix 3.4.12 Paired t-test for the left and right sides in the EMG parameters


                                      Left             Right
CHANNEL       Bare (n=16)                                            t-value p-value
                              Mean           SD    Mean        SD
quadriceps      Duration      64.40      14.306    67.19    17.128   -0.806   0.433
                  RMS          4.33      2.475      3.75    2.103    1.827    0.088
                  MF           96.50     22.685    97.42    23.696   -0.296   0.771
 hamstring      Duration      65.60      16.119    69.42    14.988   -1.494   0.156
                  RMS          3.01      0.960      2.93    1.202    0.435    0.669
                  MF          100.92     20.238    100.07   21.304   0.201    0.844
    TA          Duration      77.12      11.314    76.73    7.715    0.124    0.903
                  RMS          3.34      1.754      3.40    1.499    -0.195   0.848
                  MF          110.51     22.028    112.01   27.822   -0.565   0.581
    calf        Duration      73.45      7.978     75.14    11.606   -0.76    0.459
                  RMS          3.53      1.450      3.57    1.238    -0.226   0.824
                  MF          116.54     19.646    110.75   18.805   1.177    0.258


                                      Left             Right
CHANNEL AFO (n=16)                                                   t-value p-value
                              Mean           SD    Mean        SD
quadriceps      Duration      63.15      17.058    62.72    16.931   0.142    0.889
                  RMS          4.61      1.647      5.07    1.818    -1.069   0.302
                  MF           92.37     23.280    86.47    25.365   1.254    0.229
Hamstring       Duration      64.76      16.523    61.73    16.243   0.894    0.385
                  RMS          3.78      1.642      3.86    2.247    -0.297   0.77
                  MF           95.63     18.609    91.92    21.922   1.032    0.318
    TA          Duration      75.00      13.428    71.29    12.909   1.026    0.321
                  RMS          3.77      1.696      4.23    1.950    -0.862   0.403
                  MF          105.40     30.547    104.51   29.379   0.152    0.881
   Calf         Duration       65.13     13.395    69.28    10.405   -1.461   0.165
                  RMS          4.78      1.754      5.48    2.022    -1.558   0.14
                  MF           86.10     34.093    89.95    32.785   -0.504   0.621


                                      Left             Right
CHANNEL DAFO (n=16)                                                  t-value p-value
                              Mean           SD    Mean        SD

                                         - 173 -
quadriceps   Duration   62.69      13.779    60.20    14.019   0.689    0.501
              RMS        4.45      2.247      4.21    1.990    0.687    0.503
               MF       85.27      26.826    91.37    27.677   -1.103   0.288
Hamstring    Duration   64.31      15.129    58.11    10.598   1.469    0.163
              RMS        3.42      1.415      3.54    2.061    -0.372   0.715
               MF       89.71      27.667    90.14    22.400   -0.055   0.957
   TA        Duration   71.00      13.836    69.38    10.316   0.462    0.651
              RMS        3.65      2.130      3.76    1.614    -0.249   0.807
               MF       108.03     28.259    99.38    33.543   0.913    0.375
   Calf      Duration   67.95      8.433     68.33    8.239    -0.137   0.893
              RMS        3.91      1.609      4.33    1.509    -1.094   0.291
               MF       113.68     18.284    112.62   18.697   0.184    0.856


                                Left             Right
CHANNEL Normal (n=18)                                          t-value p-value
                        Mean           SD    Mean        SD
quadriceps   Duration   49.53      15.499    45.51    10.598   1.697    0.108
              RMS        4.15      1.895      3.73    1.800    1.931    0.07
               MF       71.33      15.798    80.50    17.705   -5.114     0
hamstrings   Duration   54.29      16.686    58.80    17.329   -0.916   0.372
              RMS        4.72      5.180      5.02    6.912    -0.596   0.559
               MF       78.34      18.332    74.06    17.543   1.121    0.278
   TA        Duration   69.46      9.118     63.84    17.102   1.241    0.231
              RMS        5.02      1.550      6.12    1.360    -3.326   0.004
               MF       79.14      10.831    81.78    11.449   -0.934   0.363
   Calf      Duration   58.97      11.982    52.62    9.091    2.484    0.024
              RMS        3.63      0.775      4.88    2.277    -2.648   0.017
               MF       82.85      14.992    79.96    14.119   0.549    0.59




                                   - 174 -
Appendix 4.1 SPSS results of the temporal distance parameters using ANOVA with

repeated measure for CP conditions


                                             Mean               Std. Error   Sig.
                                             Difference (I-J)


Parameters          (I)        (J) FACTOR1
                    FACTOR1
Stride length (m)   barefoot   AFO           -6.500E-02         .030         .146
                               DAFO          -.110              .026         .002
                    AFO        barefoot      6.500E-02          .030         .146
                               DAFO          -4.500E-02         .031         .502
                    DAFO       barefoot      .110               .026         .002
                               AFO           4.500E-02          .031         .502
Stride time (s)     barefoot   AFO           -4.500E-02         .041         .857
                               DAFO          -1.687E-02         .035         1.000
                    AFO        barefoot      4.500E-02          .041         .857
                               DAFO          2.813E-02          .031         1.000
                    DAFO       barefoot      1.687E-02          .035         1.000
                               AFO           -2.813E-02         .031         1.000
Walking speed (m/s) barefoot   AFO           -2.625E-02         .052         1.000
                               DAFO          -8.875E-02         .052         .326
                    AFO        barefoot      2.625E-02          .052         1.000
                               DAFO          -6.250E-02         .051         .714
                    DAFO       barefoot      8.875E-02          .052         .326
                               AFO           6.250E-02          .051         .714
Stance time (s)     barefoot   AFO           -9.375E-03         .011         1.000
                               DAFO          -8.125E-03         .006         .616
                    AFO        barefoot      9.375E-03          .011         1.000
                               DAFO          1.250E-03          .008         1.000
                    DAFO       barefoot      8.125E-03          .006         .616
                               AFO           -1.250E-03         .008         1.000
Swing time (s)      barefoot   AFO           -3.750E-02         .033         .832


                                          - 175 -
                                    DAFO         -8.750E-03   .031    1.000
                      AFO           barefoot     3.750E-02    .033    .832
                                    DAFO         2.875E-02    .024    .749
                      DAFO          barefoot     8.750E-03    .031    1.000
                                    AFO          -2.875E-02   .024    .749
Stance/Swing ratio    barefoot      AFO          .218         .169    .649
                                    DAFO         -2.125E-02   .241    1.000
                      AFO           barefoot     -.218        .169    .649
                                    DAFO         -.239        .193    .699
                      DAFO          barefoot     2.125E-02    .241    1.000
                                    AFO          .239         .193    .699
Cadence (Steps/min)   barefoot      AFO          5.088        3.979   .662
                                    DAFO         2.194        3.827   1.000
                      AFO           barefoot     -5.088       3.979   .662
                                    DAFO         -2.894       3.137   1.000
                      DAFO          barefoot     -2.194       3.827   1.000
                                    AFO          2.894        3.137   1.000
Based on estimated marginal means
a Adjustment for multiple comparisons: Bonferroni.




                                               - 176 -
Appendix 4.2 SPSS results of the ROM of hip parameters using ANOVA with
repeated measure for CP conditions


                                       Mean               Std. Error   Sig.
                                       Difference (I-J)
Parameters   (I) FACTOR1 (J) FACTOR1
HA1 (degree) Barefoot    AFO           -4.468             1.597        .040
                         DAFO          -4.747             1.250        .005
             AFO         barefoot      4.468              1.597        .040
                         DAFO          -.279              .964         1.000
             DAFO        barefoot      4.747              1.250        .005
                         AFO           .279               .964         1.000
HA2 (degree) Barefoot    AFO           .560               1.494        1.000
                         DAFO          .519               1.240        1.000
             AFO         barefoot      -.560              1.494        1.000
                         DAFO          -4.125E-02         .974         1.000
             DAFO        barefoot      -.519              1.240        1.000
                         AFO           4.125E-02          .974         1.000
HA3 (degree) Barefoot    AFO           -3.659             1.780        .173
                         DAFO          -4.629             1.525        .025
             AFO         barefoot      3.659              1.780        .173
                         DAFO          -.970              1.372        1.000
             DAFO        barefoot      4.629              1.525        .025
                         AFO           .970               1.372        1.000
HAT2 (%GC) Barefoot      AFO           -.344              .508         1.000
                         DAFO          .219               .335         1.000
             AFO         barefoot      .344               .508         1.000
                         DAFO          .563               .376         .466
             DAFO        barefoot      -.219              .335         1.000
                         AFO           -.563              .376         .466
HAT3 (%GC) Barefoot      AFO           -.313              .350         1.000
                         DAFO          .219               .278         1.000
             AFO         barefoot      .313               .350         1.000
                         DAFO          .531               .279         .230
                                       - 177 -
              DAFO           barefoot      -.219      .278    1.000
                             AFO           -.531      .279    .230
Range (degree) Barefoot      AFO           -4.219     1.350   .021
                             DAFO          -5.150     1.540   .013
              AFO            barefoot      4.219      1.350   .021
                             DAFO          -.931      1.520   1.000
              DAFO           barefoot      5.150      1.540   .013
                             AFO           .931       1.520   1.000
Based on estimated marginal means
a Adjustment for multiple comparisons: Bonferroni.




                                            - 178 -
Appendix 4.3 SPSS results of the ROM of knee parameters using ANOVA with
repeated measure for CP conditions


                                       Mean          Std. Error   Sig.
                                       Difference
                                       (I-J)
parameter    (I) FACTOR1 (J) FACTOR1
KA1 (degree) barefoot   AFO            -2.721        1.582        .318
                        DAFO           -5.048        1.291        .004
             AFO        barefoot       2.721         1.582        .318
                        DAFO           -2.327        1.572        .478
             DAFO       barefoot       5.048         1.291        .004
                        AFO            2.327         1.572        .478
KA2 (degree) barefoot   AFO            -4.213        1.926        .135
                        DAFO           -5.951        .983         .000
             AFO        barefoot       4.213         1.926        .135
                        DAFO           -1.738        1.950        1.000
             DAFO       barefoot       5.951         .983         .000
                        AFO            1.738         1.950        1.000
KA3 (degree) barefoot   AFO            1.477         1.868        1.000
                        DAFO           -1.883        1.128        .348
             AFO        barefoot       -1.477        1.868        1.000
                        DAFO           -3.359        1.551        .140
             DAFO       barefoot       1.883         1.128        .348
                        AFO            3.359         1.551        .140
KA4 (degree) barefoot   AFO            -2.796        2.269        .710
                        DAFO           -8.335        1.994        .002
             AFO        barefoot       2.796         2.269        .710
                        DAFO           -5.539        1.538        .008
             DAFO       barefoot       8.335         1.994        .002
                        AFO            5.539         1.538        .008
KAT2 (%GC) barefoot     AFO            .000          .246         1.000
                        DAFO           -.188         .232         1.000
             AFO        barefoot       .000          .246         1.000

                                           - 179 -
                           DAFO         -.188         .176   .911
             DAFO          barefoot     .188          .232   1.000
                           AFO          .188          .176   .911
KAT3 (%GC) barefoot        AFO          -1.094        .576   .231
                           DAFO         1.000         .668   .465
             AFO           barefoot     1.094         .576   .231
                           DAFO         2.094         .552   .005
             DAFO          barefoot     -1.000        .668   .465
                           AFO          -2.094        .552   .005
KAT4 (%GC) barefoot        AFO          .375          .462   1.000
                           DAFO         .594          .464   .660
             AFO           barefoot     -.375         .462   1.000
                           DAFO         .219          .428   1.000
             DAFO          barefoot     -.594         .464   .660
                           AFO          -.219         .428   1.000
Based on estimated marginal means
a Adjustment for multiple comparisons: Bonferroni.




                                            - 180 -
Appendix 4.4 SPSS results of the ROM of ankle parameters using ANOVA with

repeated measure for CP conditions



                                         Mean               Std. Error   Sig.
                                         Difference (I-J)
Parameters     (I) FACTOR1 (J) FACTOR1
AA1 (degree)   barefoot   AFO            -10.148            2.414        .002
                          DAFO           -10.817            1.498        .000
               AFO        barefoot       10.148             2.414        .002
                          DAFO           -.669              1.739        1.000
               DAFO       barefoot       10.817             1.498        .000
                          AFO            .669               1.739        1.000
AA2 (degree)   barefoot   AFO            -9.611             2.539        .005
                          DAFO           -12.051            1.853        .000
               AFO        barefoot       9.611              2.539        .005
                          DAFO           -2.440             1.611        .452
               DAFO       barefoot       12.051             1.853        .000
                          AFO            2.440              1.611        .452
AA3 (degree)   barefoot   AFO            -20.336            2.752        .000
                          DAFO           -15.072            1.955        .000
               AFO        barefoot       20.336             2.752        .000
                          DAFO           5.264              1.720        .024
               DAFO       barefoot       15.072             1.955        .000
                          AFO            -5.264             1.720        .024
AA4 (degree)   barefoot   AFO            -5.535             2.093        .055
                          DAFO           -7.984             1.545        .000
               AFO        barefoot       5.535              2.093        .055
                          DAFO           -2.449             1.475        .353
               DAFO       barefoot       7.984              1.545        .000
                          AFO            2.449              1.475        .353
AAT2 (%GC)     barefoot   AFO            -3.875             2.184        .289
                          DAFO           -.469              1.447        1.000
               AFO        barefoot       3.875              2.184        .289

                                     - 181 -
                            DAFO          3.406        1.924   .291
               DAFO         barefoot      .469         1.447   1.000
                            AFO           -3.406       1.924   .291
AAT3 (%GC)     barefoot     AFO           -2.938       1.730   .331
                            DAFO          -2.281       1.471   .426
               AFO          barefoot      2.938        1.730   .331
                            DAFO          .656         .701    1.000
               DAFO         barefoot      2.281        1.471   .426
                            AFO           -.656        .701    1.000
AAT4 (%GC)     barefoot     AFO           -.281        .518    1.000
                            DAFO          -.344        .378    1.000
               AFO          barefoot      .281         .518    1.000
                            DAFO          -6.250E-02   .518    1.000
               DAFO         barefoot      .344         .378    1.000
                            AFO           6.250E-02    .518    1.000
Range (degree) barefoot     AFO           10.723       2.390   .001
                            DAFO          3.022        1.756   .317
               AFO          barefoot      -10.723      2.390   .001
                            DAFO          -7.701       1.117   .000
               DAFO         barefoot      -3.022       1.756   .317
                            AFO           7.701        1.117   .000
Based on estimated marginal means
a Adjustment for multiple comparisons: Bonferroni.




                                         - 182 -
Appendix 4.5 SPSS results of the ground reaction force parameters using ANOVA

with repeated measure for CP conditions



                                          Mean               Std. Error   Sig.
                                          Difference (I-J)
Parameter    (I) FACTOR1 (J) FACTOR1
Gr1 (%BW)    barefoot      AFO            -.119              .066         .269
                           DAFO           -.157              .064         .079
             AFO           barefoot       .119               .066         .269
                           DAFO           -3.750E-02         .048         1.000
             DAFO          barefoot       .157               .064         .079
                           AFO            3.750E-02          .048         1.000
Gr2 (%BW)    barefoot      AFO            -1.000E-02         .044         1.000
                           DAFO           2.625E-02          .033         1.000
             AFO           barefoot       1.000E-02          .044         1.000
                           DAFO           3.625E-02          .033         .859
             DAFO          barefoot       -2.625E-02         .033         1.000
                           AFO            -3.625E-02         .033         .859
Gr3 (%BW)    barefoot      AFO            -8.250E-02         .027         .025
                           DAFO           -1.437E-02         .026         1.000
             AFO           barefoot       8.250E-02          .027         .025
                           DAFO           6.812E-02          .032         .149
             DAFO          barefoot       1.437E-02          .026         1.000
                           AFO            -6.812E-02         .032         .149
Grt1 (%GC)   barefoot      AFO            6.250E-02          .888         1.000
                           DAFO           .375               .659         1.000
             AFO           barefoot       -6.250E-02         .888         1.000
                           DAFO           .313               .825         1.000
             DAFO          barefoot       -.375              .659         1.000
                           AFO            -.313              .825         1.000
Grt2 (%GC)   barefoot      AFO            .969               .753         .654
                           DAFO           .250               .740         1.000
             AFO           barefoot       -.969              .753         .654

                                      - 183 -
                             DAFO          -.719     .810   1.000
              DAFO           barefoot      -.250     .740   1.000
                             AFO           .719      .810   1.000
Grt3 (%GC)    barefoot       AFO           1.188     .774   .438
                             DAFO          .500      .579   1.000
              AFO            barefoot      -1.188    .774   .438
                             DAFO          -.688     .781   1.000
              DAFO           barefoot      -.500     .579   1.000
                             AFO           .688      .781   1.000
Based on estimated marginal means
a Adjustment for multiple comparisons: Bonferroni.




                                         - 184 -
Appendix 4.6 SPSS results of the moment of hip parameters using ANOVA with

repeated measure for CP conditions



                                           Mean               Std. Error   Sig.
                                           Difference (I-J)
Parameters    (I) FACTOR1 (J) FACTOR1
HM1 (Nm/BW) barefoot         AFO           -.182              .130         .544
                             DAFO          -.373              .123         .025
              AFO            barefoot      .182               .130         .544
                             DAFO          -.191              .126         .448
              DAFO           barefoot      .373               .123         .025
                             AFO           .191               .126         .448
HM2 (Nm/BW) barefoot         AFO           -6.875E-02         .071         1.000
                             DAFO          3.313E-02          .049         1.000
              AFO            barefoot      6.875E-02          .071         1.000
                             DAFO          .102               .060         .334
              DAFO           barefoot      -3.313E-02         .049         1.000
                             AFO           -.102              .060         .334
HMT1 (%GC) barefoot          AFO           -1.031             .402         .064
                             DAFO          -.250              .246         .976
              AFO            barefoot      1.031              .402         .064
                             DAFO          .781               .329         .094
              DAFO           barefoot      .250               .246         .976
                             AFO           -.781              .329         .094
HMT2 (%GC) barefoot          AFO           .156               .397         1.000
                             DAFO          .438               .341         .658
              AFO            barefoot      -.156              .397         1.000
                             DAFO          .281               .306         1.000
              DAFO           barefoot      -.438              .341         .658
                             AFO           -.281              .306         1.000
Based on estimated marginal means
a Adjustment for multiple comparisons: Bonferroni.


                                         - 185 -
Appendix 4.7 SPSS results of the moment of knee parameters using ANOVA with
repeated measure for CP conditions


                                           Mean               Std. Error   Sig.
                                           Difference (I-J)
Parameters    (I) FACTOR1 (J) FACTOR1
KM1 (Nm/BW) barefoot         AFO           -.102              .098         .943
                             DAFO          2.500E-02          .098         1.000
              AFO            barefoot      .102               .098         .943
                             DAFO          .127               .077         .359
              DAFO           barefoot      -2.500E-02         .098         1.000
                             AFO           -.127              .077         .359
KM2 (Nm/BW) barefoot         AFO           3.125E-02          .074         1.000
                             DAFO          -5.000E-02         .052         1.000
              AFO            barefoot      -3.125E-02         .074         1.000
                             DAFO          -8.125E-02         .044         .251
              DAFO           barefoot      5.000E-02          .052         1.000
                             AFO           8.125E-02          .044         .251
KMT1 (%GC) barefoot          AFO           -2.125             1.133        .241
                             DAFO          1.250              .833         .463
              AFO            barefoot      2.125              1.133        .241
                             DAFO          3.375              1.456        .105
              DAFO           barefoot      -1.250             .833         .463
                             AFO           -3.375             1.456        .105
KMT2 (%GC) barefoot          AFO           -.781              .616         .671
                             DAFO          -.156              .315         1.000
              AFO            barefoot      .781               .616         .671
                             DAFO          .625               .473         .619
              DAFO           barefoot      .156               .315         1.000
                             AFO           -.625              .473         .619
Based on estimated marginal means
a Adjustment for multiple comparisons: Bonferroni.




                                         - 186 -
Appendix 4.8 SPSS results of the moment of ankle parameters using ANOVA with

repeated measure for CP conditions



                                          Mean               Std. Error   Sig.
                                          Difference (I-J)
Parameters    (I) FACTOR1 (J) FACTOR1
AM1 (Nm/BW)barefoot         AFO           6.875E-03          .047         1.000
                            DAFO          -7.375E-02         .060         .710
              AFO           barefoot      -6.875E-03         .047         1.000
                            DAFO          -8.063E-02         .043         .246
              DAFO          barefoot      7.375E-02          .060         .710
                            AFO           8.063E-02          .043         .246
AM2 (Nm/BW)barefoot         AFO           -.297              .075         .004
                            DAFO          -.420              .106         .004
              AFO           barefoot      .297               .075         .004
                            DAFO          -.123              .053         .104
              DAFO          barefoot      .420               .106         .004
                            AFO           .123               .053         .104
AMT1 (%GC) barefoot         AFO           -.406              .123         .014
                            DAFO          -6.250E-02         .063         1.000
              AFO           barefoot      .406               .123         .014
                            DAFO          .344               .118         .033
              DAFO          barefoot      6.250E-02          .063         1.000
                            AFO           -.344              .118         .033
AMT2 (%GC) barefoot         AFO           -1.500             1.112        .593
                            DAFO          .781               .542         .510
              AFO           barefoot      1.500              1.112        .593
                            DAFO          2.281              1.369        .349
              DAFO          barefoot      -.781              .542         .510
                            AFO           -2.281             1.369        .349
Based on estimated marginal means
a Adjustment for multiple comparisons: Bonferroni.


                                         - 187 -
Appendix 4.9 SPSS results of the power of hip parameters using ANOVA with

repeated measure for CP conditions



                                            Mean               Std. Error   Sig.
                                            Difference (I-J)
Parameters     (I) FACTOR1 (J) FACTOR1
HP1 (Watt/BW) barefoot       AFO            -.479              .346         .560
                             DAFO           -.885              .489         .271
               AFO           barefoot       .479               .346         .560
                             DAFO           -.406              .449         1.000
               DAFO          barefoot       .885               .489         .271
                             AFO            .406               .449         1.000
HP2 (Watt/BW) barefoot       AFO            -8.250E-02         .177         1.000
                             DAFO           -1.375E-02         .144         1.000
               AFO           barefoot       8.250E-02          .177         1.000
                             DAFO           6.875E-02          .111         1.000
               DAFO          barefoot       1.375E-02          .144         1.000
                             AFO            -6.875E-02         .111         1.000
HPT1 (%GC)     barefoot      AFO            -.969              .603         .387
                             DAFO           -.531              .560         1.000
               AFO           barefoot       .969               .603         .387
                             DAFO           .438               .502         1.000
               DAFO          barefoot       .531               .560         1.000
                             AFO            -.438              .502         1.000
HPT2 (%GC)     barefoot      AFO            .219               .720         1.000
                             DAFO           -.656              .547         .747
               AFO           barefoot       -.219              .720         1.000
                             DAFO           -.875              .605         .505
               DAFO          barefoot       .656               .547         .747
                             AFO            .875               .605         .505
Based on estimated marginal means
a Adjustment for multiple comparisons: Bonferroni.


                                         - 188 -
Appendix 4.10 SPSS results of the power of knee parameters using ANOVA with

repeated measure for CP conditions



                                             Mean Difference Std. Error   Sig.
                                             (I-J)
Parameters      (I) FACTOR1 (J) FACTOR1
KP1 (Watt/BW)   barefoot    AFO              -.163            .138        .767
                            DAFO             -.551            .214        .064
                AFO         barefoot         .163             .138        .767
                            DAFO             -.388            .205        .234
                DAFO        barefoot         .551             .214        .064
                            AFO              .388             .205        .234
KP2 (Watt/BW)   barefoot    AFO              .473             .279        .333
                            DAFO             .235             .148        .402
                AFO         barefoot         -.473            .279        .333
                            DAFO             -.238            .311        1.000
                DAFO        barefoot         -.235            .148        .402
                            AFO              .238             .311        1.000
KP3 (Watt/BW)   barefoot    AFO              -5.000E-02       .087        1.000
                            DAFO             .136             .115        .766
                AFO         barefoot         5.000E-02        .087        1.000
                            DAFO             .186             .159        .788
                DAFO        barefoot         -.136            .115        .766
                            AFO              -.186            .159        .788
KPT1 (%GC)      barefoot    AFO              .125             .734        1.000
                            DAFO             -.688            .877        1.000
                AFO         barefoot         -.125            .734        1.000
                            DAFO             -.813            .454        .281
                DAFO        barefoot         .688             .877        1.000
                            AFO              .813             .454        .281
KPT2 (%GC)      barefoot    AFO              -.531            .827        1.000
                            DAFO             -.375            .606        1.000
                AFO         barefoot         .531             .827        1.000

                                       - 189 -
                               DAFO             .156         .460   1.000
                 DAFO          barefoot         .375         .606   1.000
                               AFO              -.156        .460   1.000
KPT3 (%GC)       barefoot      AFO              3.125E-02    .244   1.000
                               DAFO             3.125E-02    .294   1.000
                 AFO           barefoot         -3.125E-02   .244   1.000
                               DAFO             .000         .306   1.000
                 DAFO          barefoot         -3.125E-02   .294   1.000
                               AFO              .000         .306   1.000
Based on estimated marginal means
a Adjustment for multiple comparisons: Bonferroni.




                                          - 190 -
Appendix 4.11 SPSS results of the power of ankle parameters using ANOVA with

repeated measure for CP conditions



                                                Mean               Std. Error   Sig.
                                                Difference (I-J)
Parameters       (I) FACTOR1 (J) FACTOR1
AP1 (Watt/BW)    barefoot      AFO              -.183              .264         1.000
                               DAFO             .388               .299         .643
                 AFO           barefoot         .183               .264         1.000
                               DAFO             .571               .312         .263
                 DAFO          barefoot         -.388              .299         .643
                               AFO              -.571              .312         .263
AP2 (Watt/BW)    barefoot      AFO              .289               .093         .021
                               DAFO             -6.500E-02         .119         1.000
                 AFO           barefoot         -.289              .093         .021
                               DAFO             -.354              .137         .061
                 DAFO          barefoot         6.500E-02          .119         1.000
                               AFO              .354               .137         .061
APT1 (%GC)       barefoot      AFO              -2.094             1.642        .665
                               DAFO             .656               1.358        1.000
                 AFO           barefoot         2.094              1.642        .665
                               DAFO             2.750              1.452        .233
                 DAFO          barefoot         -.656              1.358        1.000
                               AFO              -2.750             1.452        .233
APT2 (%GC)       barefoot      AFO              -1.094             1.460        1.000
                               DAFO             1.656              1.140        .501
                 AFO           barefoot         1.094              1.460        1.000
                               DAFO             2.750              1.995        .565
                 DAFO          barefoot         -1.656             1.140        .501
                               AFO              -2.750             1.995        .565
Based on estimated marginal means
a Adjustment for multiple comparisons: Bonferroni.


                                          - 191 -
Appendix 4.12 SPSS results of the EMG parameters (for all channels) using

ANOVA with repeated measure for CP conditions



A) Quadriceps
                                             Mean Difference Std. Error   Sig.
                                             (I-J)
Parameters       (I) FACTOR1 (J) FACTOR1
Duration (%GC) barefoot        AFO           2.860            4.304       1.000
                               DAFO          4.350            3.815       .816
                 AFO           barefoot      -2.860           4.304       1.000
                               DAFO          1.491            2.401       1.000
                 DAFO          barefoot      -4.350           3.815       .816
                               AFO           -1.491           2.401       1.000
RMS              barefoot      AFO           -.797            .406        .206
                               DAFO          -.290            .215        .597
                 AFO           barefoot      .797             .406        .206
                               DAFO          .507             .429        .766
                 DAFO          barefoot      .290             .215        .597
                               AFO           -.507            .429        .766
MF (Hz)          barefoot      AFO           7.531            3.629       .167
                               DAFO          8.634            4.811       .279
                 AFO           barefoot      -7.531           3.629       .167
                               DAFO          1.103            3.215       1.000
                 DAFO          barefoot      -8.634           4.811       .279
                               AFO           -1.103           3.215       1.000
Based on estimated marginal means
a Adjustment for multiple comparisons: Bonferroni.




                                          - 192 -
B) Hamstrings
                                                  Mean Difference Std. Error   Sig.
                                                  (I-J)
Parameters         (I) FACTOR1 (J) FACTOR1
Duration (%GC)     barefoot      AFO              4.261            3.913       .880
                                 DAFO             6.300            3.521       .281
                   AFO           barefoot         -4.261           3.913       .880
                                 DAFO             2.039            4.253       1.000
                   DAFO          barefoot         -6.300           3.521       .281
                                 AFO              -2.039           4.253       1.000
RMS                barefoot      AFO              -.847            .314        .049
                                 DAFO             -.506            .214        .095
                   AFO           barefoot         .847             .314        .049
                                 DAFO             .341             .296        .805
                   DAFO          barefoot         .506             .214        .095
                                 AFO              -.341            .296        .805
MF (Hz)            barefoot      AFO              6.719            3.914       .320
                                 DAFO             10.572           3.884       .047
                   AFO           barefoot         -6.719           3.914       .320
                                 DAFO             3.853            2.544       .452
                   DAFO          barefoot         -10.572          3.884       .047
                                 AFO              -3.853           2.544       .452
Based on estimated marginal means
a Adjustment for multiple comparisons: Bonferroni.




                                            - 193 -
C) TA muscles
                                                      Mean               Std. Error   Sig.
                                                      Difference (I-J)
PARAMET          (I) FACTOR1     (J) FACTOR1
duration         barefoot        AFO                  3.781              3.340        .826
                                 DAFO                 6.734              2.555        .056
                 AFO             barefoot             -3.781             3.340        .826
                                 DAFO                 2.953              3.606        1.000
                 DAFO            barefoot             -6.734             2.555        .056
                                 AFO                  -2.953             3.606        1.000
RMS              barefoot        AFO                  -.625              .275         .115
                                 DAFO                 -.334              .369         1.000
                 AFO             barefoot             .625               .275         .115
                                 DAFO                 .291               .476         1.000
                 DAFO            barefoot             .334               .369         1.000
                                 AFO                  -.291              .476         1.000
MF               barefoot        AFO                  6.303              5.029        .688
                                 DAFO                 7.556              4.119        .260
                 AFO             barefoot             -6.303             5.029        .688
                                 DAFO                 1.252              5.510        1.000
                 DAFO            barefoot             -7.556             4.119        .260
                                 AFO                  -1.252             5.510        1.000
Based on estimated marginal means
a Adjustment for multiple comparisons: Bonferroni.




                                            - 194 -
D) Calf mucles
                                             Mean               Std. Error   Sig.
                                             Difference (I-J)
Parameters       (I) FACTOR1 (J) FACTOR1
Duration (%GC) barefoot        AFO           7.094              2.399        .029
                               DAFO          6.154              2.172        .038
                 AFO           barefoot      -7.094             2.399        .029
                               DAFO          -.941              2.240        1.000
                 DAFO          barefoot      -6.154             2.172        .038
                               AFO           .941               2.240        1.000
RMS              barefoot      AFO           -1.584             .455         .010
                               DAFO          -.571              .275         .167
                 AFO           barefoot      1.584              .455         .010
                               DAFO          1.014              .365         .042
                 DAFO          barefoot      .571               .275         .167
                               AFO           -1.014             .365         .042
MF (Hz)          barefoot      AFO           25.618             6.811        .006
                               DAFO          .494               1.484        1.000
                 AFO           barefoot      -25.618            6.811        .006
                               DAFO          -25.125            7.136        .009
                 DAFO          barefoot      -.494              1.484        1.000
                               AFO           25.125             7.136        .009
Based on estimated marginal means
a Adjustment for multiple comparisons: Bonferroni.




                                          - 195 -
Appendix 4.13 Curve of ankle motion among barefoot, AFO, DAFO and normal

groups.

    Dorsiflexion
    Extension




                                   - 196 -
Appendix 4.14 Curve of ankle moment among barefoot, AFO, DAFO and normal

groups.

    Extensor (Nm/Kg)
    Dorsiflexor




                                   - 197 -
Appendix 4.15 Curve of ankle power among barefoot, AFO, DAFO and normal

groups.

    Generation (Watt/Kg)
    Absorption




                                   - 198 -

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:32
posted:3/31/2011
language:English
pages:199