Gait Analysis and Parkinson's Disease

Subscribe | Advertise | About Us | Contact Us | Home

August/September 2002

By Marsha E. Melnick, PT, PhD; Sandra Radtka, PT, PhD; and Melinda Piper, MS, PT

Marsha E. Melnick, PT, PhD, and Melinda Piper, MS, PT, prepare pa1tient for 3-D computerized gait analysis. Piper places reflective markers on the patient prior to the analysis as Melnick reviews the pathway recorded from an overhead view.

As therapists increase their emphasis on evidence-based practice, they must increase the use of objective, quantitative methods to demonstrate efficacy. Gait analysis is an excellent method for demonstrating change from treatment or from disease progression. It can be performed with modest equipment, such as a stopwatch and a measured walkway, or a sophisticated, computerized three-dimensional analysis system. Gait analysis for the person with Parkinson’s disease (PD) is an excellent tool for determining the treatment efficacy of pharmacologic, surgical, or physical therapeutic interventions.

The hallmark symptoms of PD—bradykinesia, rigidity including truncal rigidity, and postural instability—are clearly evident in the gait disorders presented in PD patients. The typical gait of the person with PD is characterized by decreased velocity and stride length, and a normal stride width. The person with PD also demonstrates decreased foot clearance, which becomes more obvious as stride length and gait velocity decrease. Frequently, increased cadence is also reported and as the cadence increases and the steps shorten with progression of walking, the person may demonstrate festination or anteropulsion. The patient may also experience freezing episodes, which increase with frequency as the disease progresses, as does the incidence of falls. Absent arm swing and decreased trunk rotation with gait and turning are also the result of bradykinesia and rigidity.¹

WHAT ASPECTS CAN BE MEASURED?

Gait analysis includes the temporal, kinetic, and kinematic aspects of gait. The temporal characteristics of gait, eg, stride length, width, cadence, and velocity, are easy to measure and should be measured routinely. Ideally, these would be measured at the PD patient’s self-selected speed as well as at a faster and slower speed. It is very difficult for people with PD to change speed or stride length, ie, to vary their motor program once it is engaged. Consequently, preferred, fast, and slow speeds are all the same with the same stride length and cadence.^1,2 Effective treatment would generate an increase in walking speed and stride length and better foot clearance. Effective treatment might also increase the ability of the patient with PD to modify velocity and stride length on demand without freezing. The temporal characteristics of gait are sensitive measures of change. We demonstrated an increase in speed and stride length after 3 and 6 months of once-a-week group exercise for our patients.³ Others have demonstrated improved gait characteristics following levodopa administration (improvements in stride length and velocity but not cadence or arm swing) or surgical interventions.^1,4 Other temporal characteristics of gait include time of double limb support and the ratio of swing and stance time. As postural stability decreases, there is an increase in double limb support but there is no report of a change in the ratio of swing time and stance time. All symptoms of gait dysfunction are worsened when the patient is asked to perform a complex task requiring attention during walking. This has been most vividly demonstrated in a study by Bond and Morris,⁵ which showed greatly decreased ability to walk when the patient was asked to carry a tray with drinking glasses. Interestingly, there was no decline in walking when the patient with PD carried the tray alone.

The kinetic characteristics of gait include external forces, either ground or external loads, acting on the joints or body segments causing motion during gait. Kinetic characteristics are measured with one or two force plates. These are sometimes difficult to measure in the patient with PD because the shortened stride length is below the length of the force plate. Consequently, two feet may be on the force plate simultaneously, which makes calculations of force moments less accurate than when there is a “clean” step. If the patient shuffles during gait, this further decreases the accuracy of kinetic data.

When reading the literature, it is, therefore, important to pay attention to the methodology used. As expected, there are decreased ground reaction forces. Other changes in the three-dimensional vector profile of the patient with PD have been reported. The person without impairments has a typical two-peak (or butterfly) pattern for ground reaction forces; the person with PD has decreased forward movement, especially at the initial and final stages of limb support. There is a corresponding decrease in the center of pressure progression even without a frozen gait.⁶ Morris et al^7,8 found that there is also a change in the hip extensor moment, especially in the early part of stance. They, likewise, found decreased ankle power generated at push-off. These changes indicate a difficulty with propulsive forces and may provide clues to appropriate rehabilitation strategies, such as increasing activity in the plantar flexors and pleiometric activities.

Kinematic analysis enables the therapist to measure the movement of the joint or body segment without reference to the cause of motion. Kinematic factors include linear and angular displacements, velocities, and acceleration. Kinematic data are dependent on gait velocity. As velocity decreases, the joint excursions also decrease.

Again, pay careful attention to the methodology and whether velocity was controlled. Ideally, age- and gender-matched control subjects should be asked to walk at their self-selected speed and then a speed matching that of the patient. In several studies, including work in our laboratory,^1,4,6 patients with PD had a decreased angular excursion at the hip, knee, and ankle and this correlated with disease progression. This is usually manifested as decreased hip extension, knee extension, and decreased ankle plantar flexion. Decreased joint excursion would be expected to coincide with decreased stride length and velocity. In fact, Siegel and Metman⁴ reported that an increase in knee and hip excursion along with an increase in peak heel rise explained 96% of the improvement in stride length seen following bilateral pallidotomy.

Muscle activity can be measured with electromyography using either surface (for the larger muscles) or fine-wire electrodes. The information on muscle activity enables the therapist to determine muscle sequencing and interactions as well as the timing of activation; information that could not be gathered from visual inspection and qualitative analysis. The study by Mitoma et al⁶ showed that there was a decrease in activation of the distal musculature, ie, the tibialis and gastroc-soleus group,⁵ as would be expected based on the kinematic studies mentioned above.

GAIT ANALYSIS METHODOLOGY

Therapists are quite familiar with measuring gait simply and inexpensively. The temporal characteristics of gait can be measured with a stopwatch and a measured pathway. It is best if the acceleration phase and deceleration phase are measured separately. For example, if you want to measure a 25-meter walk, the patient walks for 3-5 meters before timing, and is instructed to walk beyond the measured distance. The therapist can count the number of steps and, thereby, get an estimate of average stride length (number of strides/distance as well as the cadence). In one study where we measured the acceleration time as well as the steady state time, we found both measures were sensitive to change following a one time per week group exercise program.³ We also found that taking measurements from a videotaped walk was as accurate as taking the measurements with the person present.

Temporal characteristics of gait can also be measured using the timed up and go (TUG) test. If you are interested only in gait, the time for standing, walking, turning, etc, needs to be measured separately. The TUG is a more complex task than simple over-the-ground walking and so may actually be a more sensitive measure of functional walking in everyday life.

Gait patterns have also been studied from analysis of footprints. Paint or ink can be applied to a person’s foot or sole of the shoe and then analyzed from these footprints. Whereas this method is useful for persons with a stroke or brain injury or children with spina bifida, it is not as useful for persons with PD. This technique may provide some indication of gait pattern and toe clearance; however, it pales in comparison with 2-D or 3-D motion analysis. Measuring the kinematic and kinetic aspects of gait requires more sophisticated equipment. Two- and three-dimensional gait analysis uses markers on the body, a computer interface, and, at least for kinetic analysis, one or two force plates. In utilizing these sophisticated systems, the force plate should blend in with the walkway and the person should be walking at a stable velocity during data collection. Usually, several steps are recorded and an average of three or more walks is computed. Whereas gait can be measured even if the person requires stand-by or contact-guard assist with the simpler methods, the computerized method does not permit such assistance as the therapist’s or caretaker’s presence will block the markers seen by the computer. Kinetic measurements are often difficult to collect in the person with advanced disease even if the person is still able to walk unassisted. This is because kinetic measurements require a force plate and a clean single foot on the plate. A person with short, shuffling footsteps will have difficulty achieving clean contact with the force plates.

In all studies of gait analysis, it is important to remember that those with PD have improved ability when auditory or visual cues are given. Putting a line or dowel on the floor for a patient who is frozen enables that patient to walk. I had one patient who actually threw pennies in front of him so he could walk over the pennies. Even telling the person to take large steps is helpful and the effects of this instruction may last 2 hours.⁸ When analyzing the literature, the instructions given to the patient need to be noted for accurate assessment of results.

It is important to realize that computerized gait analysis is done in a laboratory setting and, although the results are important and can lead to testable hypotheses regarding treatment of the gait disorder, the patient also needs to be observed at home and in the community for a complete picture of function. What is seen in gait analysis is the effects of the disease process. They are the summed results of the bradykinesia, rigidity, and, when applicable, tremor. We can hypothesize about the cause of the disorder, but gait analysis is a sensitive and comprehensive quantitative measurement tool that should be used more frequently as an outcome measure for PD.

References

Melnick ME. Basal ganglia disorders: metabolic hereditary and genetic disorders in adults. In: Umphred D, ed. Neurological Rehabilitation. 3rd ed. St Louis: CV Mosby Co; 2001:661-695.
Morris ME, Iansek I, Matyas TA, Summers JJ. Stride length regulation in Parkinson’s disease: normalization strategies and underlying mechanisms. Brain. 1996;119:551-568.
Melnick ME, Dowling GA, Baum WC, Piper MS, Rust LL. Effects of rhythmic exercise on balance, gait, and depression in patients with Parkinson’s disease. Gerontologist. 1999;39:293.
Siegel KL, Metman LV. Effects of bilateral posteroventral pallidotomy on gait in subjects with Parkinson’s disease. Arch Neurol. 2000;57:198-204.
Bond JM, Morris ME. Goal-directed secondary motor tasks: their effects on gait in subjects with Parkinson’s disease. Arch Phys Med Rehabil. 2000;81:110-116.
Mitoma H, Hayashi R, Yanagisawa N, Tsukagoshi H. Characteristics of parkinsonian and ataxic gaits: a study using surface electromyograms, angular displacements and floor reaction forces. J Neurol Sci. 2000;174:22-39.
Morris ME, Iansek I, Matyas TA, Summers JJ. The pathogenesis of gait hypokinesia in Parkinson’s disease. Brain. 1994;117:1169-1181.
Morris ME, McGinley J, Huxhan F, Collier J, Iansek I. Constraints on the kinetic, kinematic and spatiotemporal parameters of gait in Parkinson’s disease. Hum Mov Sci. 1999;18:461-468.

Marsha E. Melnick, PT, PhD, is a professor, and Sandra Radtka, PT, PhD, is associate professor in the graduate program in physical therapy at the University of California, San Francisco/San Francisco State University. Melinda Piper, MS, PT, is a research associate at the San Francisco Veterans Administration Medical Center.

LOOKING FOR EXPERT ADVICE?

Experts here are available to answer all your questions!

Please contact us for more information about this feature, or to become an expert.

MEDIA CENTER

Interactive Media

Resources

	Calendar
	Consumer Resources
	Media Kit
	Advertiser Index EAB Reprints Submit an Article

ADDITIONAL ONLINE RESOURCES

Allied Media
24X7mag Clinical Lab Products (CLP) Orthodontic Products The Hearing Review Hearing Review Products Rehab Management		Physical Therapy Products Plastic Surgery Practice Imaging Economics RT Magazine Sleep Review

Subscribe | Advertise | About Us | Contact Us | Home

August/September 2002