Home About Journal AHEAD OF PRINT Current Issue Back Issues Instructions Submission Search Subscribe Blog    

Users Online: 1688 
Print this page  Email this page Small font sizeDefault font sizeIncrease font size 

MISCELLANEOUS Table of Contents   
Year : 2005  |  Volume : 39  |  Issue : 3  |  Page : 182-186
Instrumented gait analysis for planning and assessment of treatment in cerebral palsy

Department of Physical Medicine and Rehabilitation, Christian Medical College, Vellore, India

Click here for correspondence address and email


Background: To improve the walking efficiency of children with cerebral palsy, gait must be documented accurately so that the abnormalities can be assessed and the best treatment option can be selected.
Methods: Gait was analysed using Selspot kinematic system with a Kistler force plate and the motion analysis ambulatory EMG system connected to a personal computer.
Results: Walking speed and stride length had improved in children.
Conclusion: Gait analysis can help to accomplish improvement in walking efficiency of children with cerebral palsy.

Keywords: Instrumented gait analysis; Cerebral palsy; Video-analysis; Kinematics; Kinetics; Dynamic EMG.

How to cite this article:
Thomas R, Ganesh T, George J, Venugopal K, Macaden A, Poonnoose P, Tharion G, Devasahayam S R, Bhattacharji S. Instrumented gait analysis for planning and assessment of treatment in cerebral palsy. Indian J Orthop 2005;39:182-6

How to cite this URL:
Thomas R, Ganesh T, George J, Venugopal K, Macaden A, Poonnoose P, Tharion G, Devasahayam S R, Bhattacharji S. Instrumented gait analysis for planning and assessment of treatment in cerebral palsy. Indian J Orthop [serial online] 2005 [cited 2020 Mar 29];39:182-6. Available from:

   Introduction Top

Gait is a dynamic process constantly adapting to the demands placed on it and involves the ability to coordinate the various joints in a correct sequence. Gait may be impaired due to anatomical or physiological causes. Gait analysis is an important tool to evaluate walking and to monitor the progress of treatment.

Clinical and video gait observation helps to evaluate several parameters of gait like stance phase stability, swing limb clearance, pre-positioning of foot in terminal swing and adequacy of stride length. However, instrumented gait analysis is essential to measure joint moments, power, energy efficiency as well as dynamic EMG. One of the important applications of gait analysis is the evaluation of pathological gait of children with cerebral palsy [1] .

In this article, we record our experience of how gait analysis has been useful in the treatment of two children with cerebral palsy, one with spastic hemiplegia and one with spastic diplegia.

   Subjects and methods Top

The our Gait Analysis Lab consists of a Selspot kinematic system with a Kistler force plate and the motion analysis ambulatory EMG system connected to a personal computer. The computer tracks the position of infrared light markers placed on anatomical land marks on the patient's legs using 3 special cameras and simultaneously collects data from 8 surface EMG preamplifiers from the muscles. Physiological cost index is calculated from the walking speed and heart rate which is acquired using one of the EMG channels to record the ECG from the chest.

Gait analysis was done on two children with cerebral palsy. Infra red light emitting markers were placed at key anatomical positions on the lower limbs at the ankle, knee and hip. EMG recordings were obtained from 8 muscles in each leg (gluteus maximus, rectus femoris, tensor fasciae latae, adductor longus, vastus lateralis, medial hamstrings, gastrocnemius, tibialis anterior). A force plate embedded in the ground recorded ground reaction forces when the patient stepped on the plate. The magnitude of this force as well as the direction in all three dimensions was recorded. All the data were recorded at 100 samples per second.

The joint positions were obtained in 3 dimensions from the camera recordings. Using the joint positions, the angles, velocities and acceleration of the limb segments were calculated. Using these and estimated segment masses, the moments as well as the net power at each joint were calculated. Since the EMG from the muscles was recorded synchronously, the contribution of each muscle to the power could be inferred.

When summarizing the data, stride length data was normalized with the patient's height and forces were normalized with patient's weight. Three methods of presenting the data have been used. The first is tabulation of stride length, speed, stance swing ratio and single limb support as shown in Tables I and II. The second method is superimposed stick figures over a few strides as shown in [Figure - 1] and [Figure - 3]. The third method is graphs of the joint angles, moments and power as shown in [Figure - 2] and [Figure - 4].

A fourth important method of data review not shown in this paper is video recording which is usually played back at slow speed and individual frames are examined carefully.

   Results Top

Case 1 (Child with spastic hemiplegia)

A six year old boy born 6 weeks pre-term by Caesarian section presented with delayed motor developmental sequences and spastic right hemiplegia with equinus contracture of 20 degrees at the ankle. Video analysis showed the ankle to be stiff, in equinus and in varus. The knee was not extending completely at initial contact [Figre 1]a.

Kinematics (Joint angle graphs in [Figure - 2]) showed that the ankle was plantar flexed throughout the gait cycle and the knee was flexed at initial contact with limited knee flexion in the swing phase of gait.

Kinetics (Moment and Power graphs in [Figure - 2]) showed a burst of plantar flexion moment at the ankle occurring much earlier in stance due to power absorption by the gastrocnemius followed by an early power generation. There were two bursts of power generation: first in mid-stance where one would expect power absorption and second, in the terminal stance phase. At the knee, the extensor moment was less than expected with an increased power absorption by the hamstring muscles in midstance. Hip moments and power were nearly normal.

Dynamic EMG showed early and prolonged activity of the gastrocnemius muscle and prolonged activity of the hamstrings with co-contraction of the vastus lateralis in stance. The rectus femoris muscle showed prolonged activity in swing limiting the knee flexion. Physiological cost index was 0.8. (Normal range: 0.12 to 0.45).

Based on the gait analysis, treatment was planned for restoration of appropriate ankle dorsiflexion and knee extension in stance and knee flexion in swing. A Hoke's procedure was done at the right ankle to lengthen the tendoachilles and distal fractional lengthening of the semimembranosus and gracilis muscles and a semitendinosus tenotomy was done at the knee. Post operatively, above knee casts were applied for three weeks and the gluteus maximus, gluteus medius, knee extensors and the anterior tibial muscles were strengthened. He was sent home with an ankle foot orthosis. Clinically the gait was considerably improved following surgery [Figure - 1]b.

Gait analysis was repeated one year after surgery and the postoperative outcome was compared with the preoperative analysis. Walking speed and stride length had improved [Table I]. The knee was more extended at initial contact and flexion was better in swing. Ankle dorsiflexion was restored. The abnormal power absorptions in initial stance at the ankle and knee were reduced to half their preoperative level [Figure - 2]. Physiological cost index improved to 0.53 indicating the improved energy efficiency of the child's gait.

Postoperatively, as the child's gait was apparently normal, the need for continuation of appliances and exercises were corroborated by the residual abnormalities which were evident only in the kinetic studies of the ankle and knee.

Case 2 (Child with spastic diplegia)

A 14 year old child presented with a history of birth asphyxia and neonatal sepsis with delayed motor developmental sequences.

He had spastic diplegia and he was ambulant with a crouched gait pattern with hips and knees flexed and calcaneo-valgus feet [Figure - 3]a. On examination, the lower limbs were spastic with poor voluntary control of the hip and knee extensors. There was no hip flexion contracture. The popliteal angle at the knees was 80 degrees on each side. Gait analysis was done and the various temporal parameters are given in [Table - 2].

Sagittal plane kinematics showed the ankle to be in 20 degrees of dorsiflexion and the knees to be in 80 degrees of flexion throughout the gait cycle. The hip angle was normal [Figure - 4].

The kinetics at the ankle showed that the moments were reduced with power absorption throughout the stance phase by eccentric gastrocnemius contraction and the normal power generation in the ankle at terminal stance was absent. Continuous extensor moment was present in the stance phase at the knee and the power graphs showed continuous absorption in stance due to the eccentric contraction of the quadriceps. Evaluation of dynamic EMG showed prolonged stance phase activity of the quadriceps, hamstrings and gastrocnemius muscles confirming that they were generating the abnormal moments. This abnormal activity required extra energy and the physiological cost index was calculated to be 1.69 (Normal range: 0.12 to 0.45).

Power generation at the hip was more than normal and this is common in cerebral palsy because cortical control of proximal muscles is better and they use the proximal muscles to compensate for the poorly controlled distal muscles[2] . The usual way that these children generate power at the hip when the hip extensors are poor is to use their hamstrings as hip extensors[3] during the first 50-60 percent of stance. Since the hamstrings span both the hip and the knee, overactive hamstrings lead to knee flexion. Normally during stance, the ground reaction force provides stability from midstance to toe off. Since both knees are flexed throughout stance, eccentric quadriceps action is required to prevent collapse. The extensor moment at the knee indicates that the quadriceps must be active as the external flexor moment produced by the ground reaction force needs to be balanced by muscle activity. Though the firing time of the vasti is increased, these muscles are under normal voluntary control as the muscle activity by dynamic EMG is synchronous with the calculated extensor moment. Eccentric contraction of the gastrocnemius was required at the ankle instead of the normal concentric contraction at terminal stance to possibly slow down the forward fall of the tibia. The knee was flexed in stance because the triceps surae could not generate an adequate plantar flexion-knee extension couple. Loss of knee flexion in swing was due to the co-spasticity of rectus femoris and hamstrings throughout the swing phase. He walked with excessive dorsiflexion at the ankles to maintain a foot flat gait in the presence of bilateral knee flexion in stance.

The patient's hamstring shortening was treated with stretching and casting; the weak hip extensors were strengthened and AFOs were prescribed to prevent the forward fall of the tibia. He returned for review every year and there was a gradual improvement in the crouch gait, which was present prior to treatment. A repeat gait analysis was done three years after the first [Figure - 3]b. This showed that though the gait was slightly slower, the other stride parameters were almost the same as before. The ankle dorsiflexion and the knee flexion were found to be reduced. Following treatment the ankle power absorption was absent, but the normal power generation at terminal stance was absent as well. At the knee, with the improved extension, the extensor moment and power absorption by the quadriceps in stance had significantly reduced. Knee motion in swing also improved resulting in improved foot clearance. The dynamic EMGs were also seen to be more normal [Figure - 4] The PCI was calculated to be 0.88 and was 0.66 with AFOs.

Among children with diplegic and quadriplegic cerebral palsy, crouch gait is one of the common abnormalities, the primary cause of which is pathological Hamstrings [4] . Even though this child walked with a slower speed following treatment, he had a more energy efficient gait. However, as the kinetics and kinematics are still abnormal, he needs continued strengthening exercises and exercises to maintain and improve range of motion especially as he enters his adolescent growth spurt. Gait analysis in this child has thus provided an objective assessment of the treatment outcome and has helped to monitor ongoing treatment.

   Discussion Top

Instrumented Gait Analysis includes video analysis, kinematic studies, kinetics, dynamic EMG measurements and estimation of energy consumption [5] . Video analysis is an objective method of recording a patient's gait, which is difficult to record on paper. Walking can be studied at length without making the patient walk for long periods because the video can be played back in slow motion many times [6] .

Kinematics describe the spatial movements of limb segments using reflective markers, and measures stride length, step length, single limb support, stance-swing ratio and joint angles. Kinetics describes forces during gait and is reported as ground reaction forces, joint moments and power.

The data is collected as the patient steps on the force plate. Dynamic EMG of selected muscles is collected using surface or needle electrodes and is used to study whether the particular muscle is contracting appropriately or pathologically [6] . If EMG and kinetic moment data are used in conjunction with each other, specific information about the patient's gait can be obtained. The fifth component of gait analysis is the measurement of Energy consumption and physiological cost index (PCI) is used as an index of the energy efficiency of gait. Since normal gait is energy efficient, any deviation from the normal, whether primary or secondary, results in excessive energy consumption.

Each of the above components of Instrumented Gait analysis provides a critical part of the total picture, but alone they provide only limited application. A comprehensive gait analysis system must consist of a combination of the various methods to measure all the necessary parameters of gait required for complete analysis.

Despite the many causes of cerebral palsy, these patients have been found to exhibit several common gait abnormalities which are due to muscle spasticity and joint contractures [1] . The spastic muscles may inhibit movement by firing prematurely, inappropriately, or for a prolonged period. The associated abnormalities may compromise either stance phase stability or limb propagation in swing phase or both. The energy requirement of ambulation in cerebral palsy may increase many fold due to the poor control of body movements and inefficient transfer of kinetic and potential energies. Gait analysis in cerebral palsy has two major functions.

  1. Precise assessment of gait pathophysiology for treatment planning.
  2. Accurate outcome assessment and planning of further intervention.

Once gait analysis is done, a functional baseline is established and an individualised treatment plan can be made. The difference between primary gait deviations and adaptive coping mechanisms can be distinguished. To optimise gait efficiency, one needs to correct the former;the latter will disappear spontaneously when no longer required. Continual follow up of patients with cerebral palsy is crucial[7] because growth produces changes in gait, which need to be monitored. In the case of the two children described above, gait analysis has helped to plan treatment, objectively assess the outcome and to monitor ongoing treatment.

The management of cerebral palsy has radically changed with gait analysis. A careful pre treatment analysis can result in better decision making for long term post intervention results. Human gait is complex and quantitative gait analysis offers a clinical tool to better understand these complexities and thus prescribe optimal rehabilitation intervention. The technology required for acquiring and analysing the data is continually improving and low cost gait analysis systems are also being made available. It is anticipated that instrumental gait analysis will soon become a routine extension of the clinical examination and management of children with cerebral palsy will be routinely based on objective evidence of the precise dysfunction.

Acknowledgement: This work was made possible through a research grant from the Department of Science and Technology of the Government of India.

   References Top

1.Sutherland DH, Davids JR. Common gait abnormalities of the knee in cerebral palsy. Clin Orthop. 1993; 288:139-147.  Back to cited text no. 1    
2.Gage JR. Gait Analysis in Cerebral Palsy. Clinics in Developmental Medicine.No 121. London: MacKeith Press. 1991; 20.  Back to cited text no. 2    
3.Hoffinger SA, Rab GT. Hamstrings in cerebral palsy crouch gait. J Paediatr Orthop. 1993;13: 722-726.  Back to cited text no. 3    
4.Thompson NS, Baker RJ, Cosgrove AP. Musculoskeletal modelling in determining the effect of botulinum toxin on the hamstrings of patients with crouch gait. Developmental Med Child Neurol. 1998;40: 622-625.  Back to cited text no. 4    
5.Perry, J. Gait analysis systems . In: Gait Analysis. Normal and patho­logical function. NJ: Slack Inc.1992; 351-354.  Back to cited text no. 5    
6.Kerrigan CD, Schaufele M, Wen MN. Gait analysis. In: Rehabilitation Medicine .Principles & Practice.3 rd ed. Philadelphia: Lippincott-Raven Publishers. 1998; 167-187.  Back to cited text no. 6    
7.Sutherland DH, Santi M, Abel MF. Treatment of stiff knee gait in cerebral palsy: a comparison by gait analysis of distal rectus femoris transfer versus proximal rectus release. J Paediatr Orthop. 1990;10: 433-441.  Back to cited text no. 7    

Correspondence Address:
Raji Thomas
Department of Physical Medicine and Rehabilitation, Christian Medical College,Vellore- 632004
Login to access the Email id

Source of Support: None, Conflict of Interest: None

Rights and PermissionsRights and Permissions


  [Figure - 1], [Figure - 2], [Figure - 3], [Figure - 4]

  [Table - 1], [Table - 2]


    Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
    Email Alert *
    Add to My List *
* Registration required (free)  

    Subjects and methods
    Article Figures
    Article Tables

 Article Access Statistics
    PDF Downloaded354    
    Comments [Add]    

Recommend this journal