5 Applications of gait analysis
The aim of this chapter is to provide a broad overview of some of the ways in which gait analysis is used, particularly in a clinical setting. It is not intended to transform the reader into an expert on the subject. Anyone planning to use gait analysis in clinical decision making should read all the texts listed at the end of this chapter, attend courses and training on the interpretation of gait analysis data, and if possible spend some time studying or working in a clinical gait laboratory. There are several national and international societies that specialise in such courses and training, such as the European Society of Movement Analysis for Adults and Children (ESMAC), the Gait and Clinical Movement Analysis Society (GCMAS) and the Clinical Movement Analysis Society (CMAS).
The applications of gait analysis are conveniently divided into two main categories: clinical gait assessment and gait research. Clinical gait assessment has the aim of helping individual patients directly, whereas gait research aims to improve our understanding of gait, either as an end in itself or in order to improve medical diagnosis or treatment in the future. There is obviously some overlap, in that many people performing clinical gait assessment use it as the basis for research studies. Indeed, this is the way that most progress in the use of clinical gait assessment is made.
Davis (1988) pointed out that there are considerable differences between the technical requirements for clinical gait assessment and those for gait research. For example, an intrusive measurement system and a cluttered laboratory environment might not worry a fit adult, who is acting as an experimental subject, but could cause significant changes in the gait of a child with cerebral palsy. In gait research, it might be acceptable to spend a whole day preparing the subject, making the measurements and processing the data, whereas in the clinical setting patients often tire easily and the results are usually needed as quickly as possible. The requirements for accuracy are generally not as great in the clinical setting as they are in the research laboratory, so long as the measurement errors are not large enough to cause a misinterpretation of the clinical condition. However, it is essential that those interpreting the data appreciate the possible magnitude of any such errors. Finally, the system must be able to cope with a wide variety of pathological gaits. It is much easier to make measurements on normal subjects than on those whose gait is very abnormal, which may explain why the literature of the subject is dominated by studies of normal individuals! A final and important point is that there is no value in using a complicated and expensive measurement system, unless it provides information which is useful and which cannot be obtained in an easier way.
Gait research may be divided into clinical and fundamental research, the former concentrating on disease processes and methods of treatment, the latter on methods of measurement and the advancement of knowledge in biomechanics, human performance and physiology.
Clinical gait assessment seeks to describe, on a particular occasion, the way in which a person walks. This may be all that is required, if the aim is simply to document their current status. Alternatively, it may be just one step in a continuing process, such as the planning of treatment or the monitoring of progress over a period of time.
Rose (1983) made a distinction between gait analysis and gait assessment. He regarded gait analysis as ‘data gathering’ and gait assessment as ‘the integration of this information with that from other sources for the purposes of clinical decision making’. This usage of the term ‘analysis’ differs from that in more technical fields, in which it means ‘the processing of data to derive new information’. However, Rose’s use of the term is helpful, because it points out that gait assessment is simply one form of clinical assessment. Medical students are taught that clinical assessment is based on three components: history, physical examination and special investigations. In this context, gait analysis is simply a special investigation, the results of which will augment other investigations, such as X-ray reports and blood biochemistry, to provide a full clinical picture. The term ‘gait evaluation’ is sometimes used instead of gait analysis.
The simplest form of gait assessment is practised every day in physician offices, physiotherapy and rehabilitation clinics, orthotic and prosthetic clinics, sports centres, and many other settings throughout the world. Every time a clinician watches a client or patient walk up and down a room, they are performing an assessment of the patient’s gait. However, such assessment is often unsystematic and the most that can be hoped for is to obtain a general impression of how well the patient walks, and perhaps some idea of one or two of the main problems. This could be termed an ‘informal’ gait assessment. To perform a ‘formal’ gait assessment requires a careful examination of the gait, using a systematic approach, if possible, augmented by objective measurements. Such a gait assessment will usually produce a written report and the discipline involved in preparing such a report is likely to result in a much more carefully conducted assessment.
The gait analysis techniques which are used in clinical gait assessment vary enormously, with the nature of the clinical condition, the skills and facilities available in the individual clinic or laboratory and the purpose for which the assessment is being conducted. In general, however, clinical gait assessment is performed for one of three possible reasons: it may form the basis of clinical decision making, it may help with the diagnosis of an abnormal gait, or it may be used to document a patient’s condition. These will be considered in turn.
this starts with a full clinical history, both from the patient and from any others involved, such as doctors, therapists or family members. Where a patient has previously had surgery, details of this should be obtained, if possible from the operative notes. History taking is followed by a physical examination, with particular emphasis on the neuromusculoskeletal system. In many laboratories, physical examinations are performed by both a doctor and a physiotherapist. Finally, a formal gait analysis is carried out.
the next stage is the development of a hypothesis regarding the cause or causes of the observed abnormalities. This hypothesis is often informed by the specific questions raised by the referring doctor. Time needs to be set aside to review the data, and consultation between colleagues, particularly those from different disciplines, is extremely valuable. Indeed, almost all of those using gait assessment as a clinical decision-making tool stress the value of this ‘team approach’. In forming a hypothesis as to the fundamental problem in a patient with a gait disorder, Rose (1983) emphasised that the patient’s gait pattern is not entirely the direct result of the pathology, but is the net result of the original problem and the patient’s ability to compensate for it. He observed that the worse the underlying problem, the easier it is to form a hypothesis, since the patient is less able to compensate.
this stage is sometimes omitted, when there is little doubt as to the cause of the abnormalities observed. However, where some doubt does exist, the hypothesis can be tested in two different ways – either by using a different method of measurement or by attempting in some way to modify the gait. Some laboratories routinely use a fairly complete ‘standard protocol’, including video recording, kinematic measurement, force platform measurements and surface electromyography (EMG). They will then add other measurements, such as fine wire EMG, where this is necessary to test a hypothesis. Other clinicians start the gait analysis using a simple method, such as video recording, and only add other techniques, such as EMG or the use of a force platform, where they would clearly be helpful. Rose (1983) opposed the use of a standard protocol for all patients, since some of the procedures turn out to be unnecessary and there is a risk of ending up with ‘an exhausted subject in pain’. The other method of testing a hypothesis is to re-examine the gait after attempting some form of modification, typically by the application of an orthosis to limit joint motion, a medication such as botulinum toxin to decrease spasticity, or by anaesthetising a muscle. The ultimate form of gait modification is the surgical operation, with retesting following recovery. However, this is a rather drastic form of ‘hypothesis testing’, which can be used only where there is a good reason to suppose that the operation will lead to a definite improvement.
Different types of gait analysis data may be useful for different aspects of the gait assessment. Information on foot timing may be useful to identify asymmetries and may indicate problems with balance, stability, pain etc. The general gait parameters give a guide to the degree of disability and may be used to monitor progress or deterioration with the passage of time. The kinematics of limb motion describe abnormal movements, but do not identify the ‘guilty’ muscles.
The most useful measures are probably the joint moments and joint powers, particularly if this information is supplemented by EMG data. Hemiparetic patients may show greater differences between the two sides in muscle power output than in any of the other measurable parameters, including EMG. Winter (1985) stressed the need to work backwards from the observed gait abnormalities to the underlying causes in terms of the ‘guilty’ motor patterns, using both the EMG and the moments about the hip, knee and ankle joints. He offered a method of charting gait abnormalities and a table giving the common gait disorders, their possible causes and the type of evidence which would confirm or refute them (Table 5.1). Although the next step, that of treatment, was not considered in detail, he suggested that once an accurate ‘diagnosis’ had been made, the therapist would be challenged to ‘alter or optimise the abnormal motor patterns’ which requires the understanding of the biomechanical cause–effect relationships necessary to improve gait.
|Foot slap at heel contact||Below normal dorsiflexor activity at heel contact||Below normal tibialis anterior EMG or dorsiflexor moment at heel contact|
|Forefoot or flatfoot initial contact||(a) Hyperactive plantarflexor activity in late swing||(a) Above normal plantarflexor EMG in late swing|
|(b) Structural limitation in ankle range||(b) Decreased dorsiflexor range of motion|
|(c) Short step length||(c) See (a–d) immediately below|
|Short step||(a) Weak push off prior to swing||(a) Below normal plantarflexor moment or power generation or EMG during push off|
|(b) Weak hip flexors at toe off and early swing||(b) Below normal hip flexor moment or power or EMG during late push off and early swing|
|(c) Excessive deceleration of leg in late swing||(c) Above normal hamstring EMG or knee flexor moment or power absorption late in swing|
|(d) Above normal contralateral hip extensor activity during contralateral stance||(d) Hyperactivity in EMG of contralateral hip extensors|
|Stiff-legged weightbearing||Above normal extensor activity at the ankle, knee or hip early in stance*||Above normal EMG activity or moments in hip extensors, knee extensors or plantarflexors early in stance|
|Stance phase with flexed but rigid knee||Above normal extensor activity in early and mid-stance at the ankle and hip, but with reduced knee extensor activity||Above normal EMG activity or moments in hip extensors and plantarflexors in early and mid-stance|
|Weak push off accompanied by observable pull off||Weak plantarflexor activity at push off. Normal, or above normal, hip flexor activity during late push off and early swing||Below normal plantarflexor EMG, moment or power during push off. Normal or above normal hip flexor EMG or moment or power during late push off and early swing|
|Hip hiking in swing (with or without circumduction of lower limb)||(a) Weak hip, knee or ankle dorsiflexor activity during swing||(a) Below normal tibialis anterior EMG or hip or knee flexors during swing|
|(b) Overactive extensor synergy during swing||(b) Above normal hip or knee extensor EMG or moment during swing|
|Trendelenburg gait||(a) Weak hip abductors||(a) Below normal EMG in hip abductors: gluteus medius and minimus, tensor fascia lata|
|(b) Overactive hip adductors||(b) Above normal EMG in hip adductors, adductor longus, magnus and brevis, and gracilis|
Reproduced with permission from Winter (1985).
Many others working in the field of clinical gait assessment have noted the difficulty of deducing the underlying cause from the observed gait abnormalities, because of the compensations which take place. A number of attempts have been made to simplify this process, by using a systematic approach. Computer-based expert systems are very suitable for this type of application and a number of such systems have been developed for clinical gait assessment. Since gait patterns are seldom clear-cut, such expert systems cannot generally use a fixed set of rules but rather need to learn to recognise patterns within complex sets of data. Techniques such as neural networks and fuzzy logic have been explored for this purpose (Chau, 2001). No doubt the number and quality of such systems will increase in the future. The following paragraphs describe how gait assessment is used for clinical decision making in a ‘typical’ laboratory. The details will, of course, differ from one laboratory to another, based on the skills and interests of the laboratory personnel, the facilities and equipment available, and the types of patient seen.
When the patient arrives at the facility informed consent is taken, a thorough history is then obtained and a physical examination is performed, by both a doctor, a physiotherapist, and sometimes others such as a prosthetist. Height, weight and a number of other measurements are made. The patient’s gait is video-recorded, viewing the patient from both sides and from the front and back. The ‘technological’ element of the gait analysis is performed, using a television/computer kinematic system and one or more force platforms. The number of cameras used is dictated largely by economics. Ideally, at least six cameras should be used but three can give acceptable data, particularly if measurements are made from only one side of the body at a time. Most laboratories record surface EMG, either on muscles which are selected on a case-by-case basis or on a standard set, such as gluteus maximus, quadriceps (in particular rectus femoris), medial and lateral hamstrings, triceps surae, tibialis anterior and the hip adductors. Depending on the clinical condition, fine wire EMG of selected muscles may be recorded, either at the same time or later. For example, Gage et al. (1984) reported that where a hip flexion contracture is present, their laboratory routinely records fine wire EMG from the iliopsoas, however this is not common clinical practice. There is variation in the protocols used, with some facilities recording kinetic, kinematic and EMG data at the same time, whereas others find it more convenient to record EMG separately.
Since great variability may exist between one walk and another, any single walk may not be representative, particularly if the patient hesitates or momentarily loses their balance. For this reason, a number of walks are recorded and the results examined for consistency. Depending on the clinical status of the patient, walks may be made under a number of different conditions, for example with and without shoes, with or without an orthosis, or using different walking aids. Activities such as rising from a chair or walking up and down steps can also be studied, however, it may not be possible to obtain force plate data.
Once preliminary reports on the history and physical examination have been prepared, the data processed and charts printed, the ‘team’ meets to discuss the case. The composition of the team varies considerably from one facility to another, but it would commonly consist of a doctor, a physiotherapist and a kinaesiologist or bio-engineer, with the optional addition of other doctors, physiotherapists, prosthetists, orthotists and podiatrists. Indeed, anyone with an interest in providing the best possible care for the patient may be invited to join the team to discuss a particular patient. In many facilities, the team includes the orthopaedic surgeon who will ultimately perform any necessary surgery.
The assessment begins with a careful review of the patient’s history and the physical examinations by the doctor and physiotherapist. The gait video recording is watched frame by frame and discussed and notes are made for inclusion in a final report. The general gait parameters are noted to determine the degree of disability and the effects of any changes in conditions, such as orthoses or walking aids. Different gait analysis systems provide different amounts of technical data on the patient’s gait and examination of the ‘charts’ can be a long and painstaking process. One of the most important parts of the assessment is to identify deviations from normal in the joint angles and to determine the cause of such deviations. This process is easier if joint moments and powers are available. Joint moments indicate in general terms which structures in the region of a joint are coming under tension and the degree of tension in them. Joint powers may help to discriminate between concentric muscle contraction, eccentric muscle contraction and passive tension in soft tissues. They can also distinguish between powerful and weak muscle contractions. The EMG also contributes to this process by identifying which muscles are electrically active and are thus candidates for developing tension at different times during the gait cycle.