3 Pathological and other abnormal gaits
Although some variability is present in normal gait, particularly in the use of the muscles, there is an identifiable ‘normal pattern’ of walking and a ‘normal range’ can be defined for all of the variables which can be measured. Pathology of the locomotor system frequently produces gait patterns which are clearly ‘abnormal’. Some of these abnormalities can be identified by eye, but others can only be identified by the use of appropriate measurement systems.
In normal walking, all of these are achieved without any apparent difficulty and with a modest energy consumption. However, in many forms of pathological gait they can be accomplished only by means of abnormal movements, which usually increase the energy consumption, or by the use of walking aids such as canes, crutches or orthoses (calipers and braces). If even one of these four requirements cannot be met, the subject is unable to walk.
The pattern of gait is the outcome of a complex interaction between the many neuromuscular and structural elements of the locomotor system. Abnormal gait may result from a disorder in any part of this system, including the brain, spinal cord, nerves, muscles, joints and skeleton. Abnormal gait may also result from the presence of pain, so that although a person is physically capable of walking normally, they find it more comfortable to walk in some other way.
The term limp is commonly used to describe a wide variety of abnormal gait patterns. However, dictionary definitions are unhelpful, a typical one being ‘to walk lamely’. Since the word has no clearly defined scientific meaning, it should only be used with caution in the context of gait analysis. The most appropriate use of the word is probably for a gait abnormality involving some degree of asymmetry, which is readily apparent to an untrained observer.
Since gait is the end result of a complicated process, a number of different original problems may manifest themselves in the same abnormality of gait. For this reason, the abnormal gait patterns will be described separately from the pathological conditions which cause them. This chapter describes, in some detail, the most common abnormal gait patterns. This is followed by a description of the use of walking aids such as canes and walkers, and treadmill gait.
The following sections are based on a manual of lecture notes for student orthotists published by New York University (1986). Despite being over 25 years old, the manual includes a very useful list of common gait abnormalities, all of which can be identified by eye, which still hold today. The manual criticises the common practice of identifying gait abnormalities by their pathological cause, for example ‘hemiplegic gait’, which immediately suggests that all hemiplegics walk in the same way, which is far from true, and also neglects the changes in gait which may occur with the passage of time, or result from treatment. The manual suggests that it is preferable to use purely descriptive terms, such as ‘excessive medial foot contact’. This practice will be adopted in the following sections. Some of the gait abnormalities described in the New York University publication apply only to the gait of subjects wearing orthoses; these descriptions have been omitted from the present text.
The pathological gait patterns to be described may occur either alone or in combination. If in combination, they may interact, so that the individual gait modifications do not exactly fit the description. The list that follows is not exhaustive; a subject may use a variation of one of the general patterns or may use another gait pattern which is not listed here.
When studying a pathological gait, particularly one which does not appear to fit into one of the standard patterns, it is helpful to remember that an abnormal movement may be performed for one of two reasons:
Bending the trunk towards the side of the supporting limb during the stance phase is known as lateral trunk bending, ipsilateral lean or, more commonly, a Trendelenburg gait. The purpose of the manoeuvre is generally to reduce the forces in the abductor muscles and hip joint during single leg stance.
Lateral trunk bending is best observed from the front or the back. During the double support phase, the trunk is generally upright but as soon as the swing leg leaves the ground, the trunk leans over towards the side of the stance phase leg, returning to the upright attitude again at the beginning of the next double support phase. The trunk bending may be unilateral, being restricted to the stance phase of one leg, or it may be bilateral, the trunk swaying from one side to the other, to produce a gait pattern known as waddling.
In the examples which follow, the weight of the trunk is 452 N, corresponding to a mass of 46 kg, and the weight of the right leg is 147 N, corresponding to a mass of 15 kg. As explained in Chapter 1 (p. 000), weight is a force, calculated by multiplying an object’s mass by the acceleration due to gravity, 9.81 m/s2.
Figure 3.1 shows a schematic of the trunk, pelvis and hip joints when standing on both legs. The abductor muscles are inactive and the weight of the trunk is divided equally between the two hip joints. Figure 3.2 shows what happens in a normal individual when the right foot is lifted off the ground: the force through the left hip joint increases by a factor of six, from 226 N (23 kgf or 51 lbf) to 1510 N (154 kgf or 339 lbf). This increase in force is made up of three components:
3. The left hip abductors (primarily gluteus medius) contract, produces a clockwise moment to keep the pelvis from dropping on the unsupported side. The reaction force to this contraction passes through the left hip.
Fig. 3.2 • Schematic of single legged stance on the left (the right leg is held up). The force in the left hip (1510 N) is the sum of: (i) weight of trunk (452 N), (ii) weight of right leg (147 N) and (iii) contraction force of abductor muscles (911 N).
The effect of lateral trunk bending on the joint force is shown in Figure 3.3. There is no effect on components (1) and (2) of the increased force, but if the centre of gravity of the trunk is moved directly above the left hip, this eliminates the anti-clockwise moment produced by the mass of the trunk. The abductors are now only required to contract with a force of 363 N (37 kgf or 81 lbf) to balance the anti-clockwise moment provided by the weight of the right leg. There is thus a reduction of 548 N (56 kgf or 123 lbf) in the abductor contraction force and a corresponding reduction in the total joint force, from 1501 N down to 962 N. The numbers in the illustrations refer to standing; during the stance phase of walking, higher forces are to be expected due to the vertical accelerations of the centre of gravity, which cause the force transmitted through the leg to fluctuate above and below body weight (see Fig. 2.19). However, these fluctuations tend to be less in pathological than in normal gait, since the vertical accelerations are less in someone walking with a shorter stride length. The numbers also suppose that the bending of the trunk brings its centre of gravity exactly above the hip joint. This is unlikely to happen in practice, of course, but the principles remain the same, whether the centre of gravity is not deviated as far as the hip joint or even if it passes lateral to it.
Fig. 3.3 • Lateral trunk bending: bringing the trunk across the supporting hip reduces the anti-clockwise moments about the left hip from 62 Nm to 25 Nm, permitting the pelvis to be stabilised by a smaller abductor force. The force in the left hip (962 N) is the sum of: (i) weight of trunk (452 N), (ii) weight of right leg (147 N), and (iii) contraction force of abductor muscles (363 N).
If the hip joint is painful, as in osteoarthritis and rheumatoid arthritis, the amount of pain experienced usually depends to a very large extent on the force being transmitted through the joint. Since lateral trunk bending reduces the total joint force, ‘Trendelenburg gait’ is extremely common in people with arthritis of the hip. Although it produces a useful reduction in force and hence in pain, the forces still remain substantial (962 N in Fig. 3.3) and some form of definitive treatment is usually required.
If the hip abductors are weak, they may be unable to contract with sufficient force to stabilise the pelvis during single leg stance. In this case, the pelvis will dip on the side of the foot which is off the ground (Trendelenburg’s sign, as opposed to ‘Trendelenburg gait’). In order to reduce the demands on the weakened muscles, the subject will usually employ lateral trunk bending, in both standing and walking, to reduce joint moment as far as possible (Fig. 3.4). Hip abductor weakness may be caused by disease or injury affecting either the muscles themselves or the nervous system which controls them.
Three conditions around the hip joint will lead to difficulties in stabilising the pelvis using the abductors: congenital dislocation of the hip (CDH, also known as developmental dysplasia of the hip), coxa vara and slipped femoral epiphysis. In all three, the effective length of the gluteus medius is reduced because the greater trochanter of the femur moves proximally, towards the pelvic brim. Since the muscle is shortened, it is unable to function efficiently and thus contracts with a reduced tension. In CDH and severe cases of slipped femoral epiphysis, a further problem exists in that the normal hip joint is effectively lost, to be replaced by a false hip joint, or pseudarthrosis. This abnormal joint is more laterally placed, giving a reduced lever arm for the abductor muscles, and it may fail to provide the ‘solid and stable fulcrum’ required. The combination of reduced lever arm and reduced muscle force gives these subjects a powerful incentive to walk with lateral trunk bending (Fig. 3.5). In many cases, particularly in older people with CDH, the false hip joint becomes arthritic and they add a painful hip to their other problems. Pain is frequently also a factor in slipped femoral epiphysis.
If the walking base is abnormally wide, there is a problem with balance during single leg stance. Rather than tip the whole body to maintain balance, as in Figure 2.28A, lateral bending of the trunk may be used to keep the centre of gravity of the body roughly over the supporting leg. In most cases, this will need to be done during the stance phase on both sides, leading to bilateral trunk bending and a waddling gait. A number of conditions, which will be described later, may cause a wide walking base.
When walking with an unequal leg length, the pelvis tips downwards on the side of the shortened limb, as the body weight is transferred to it. This is sometimes described as ‘stepping into a hole’. The pelvic tilt is accompanied by a compensatory lateral bend of the trunk.
In anterior trunk bending, the subject flexes his or her trunk forwards early in the stance phase. If only one leg is affected, the trunk is straightened again around the time of opposite initial contact, but if both sides are affected, the trunk may be kept flexed throughout the gait cycle. This gait abnormality is best seen from the side.
One important purpose of this gait pattern is to compensate for an inadequacy of the knee extensors. The left panel of Figure 3.6 shows that early in the stance phase, the line of action of the ground reaction force vector normally passes behind the axis of the knee joint and generates an external moment which attempts to flex it. This is opposed by contraction of the quadriceps, to generate an internal extension moment. If the quadriceps are weak or paralysed, they cannot generate this internal moment and the knee will tend to collapse. As shown in the right panel of Figure 3.6, anterior trunk bending is used to move the centre of gravity of the body forwards, which results in the line of force passing in front of the axis of the knee, producing an external extension (or hyperextension) moment. In addition to anterior trunk bending, subjects will sometimes keep one hand on the affected thigh while walking, to provide further stabilisation for the knee.
Fig. 3.6 • Anterior trunk bending: in normal walking, the line of force early in the stance phase passes behind the knee; anterior trunk bending brings the line of force in front of the knee, to compensate for weak knee extensors.
One form of posterior trunk bending is essentially a reversed version of anterior trunk bending, in that early in the stance phase, the whole trunk moves in the sagittal plane, but this time backwards instead of forwards. Again, it is most easily observed from the side. The purpose of this is to compensate for ineffective (weak) hip extensors. The line of the ground reaction force early in the stance phase normally passes in front of the hip joint (Fig. 3.7, left panel). This produces an external moment which attempts to flex the trunk forward on the thigh and is opposed by contraction of the hip extensors, particularly the gluteus maximus. Should these muscles be weak or paralysed, the subject may compensate by moving the trunk backwards at this time, bringing the line of action of the external force behind the axis of the hip joint, as shown in the right panel diagram of Figure 3.7.
Fig. 3.7 • Posterior trunk bending: in normal walking, the line of force early in the stance phase passes in front of the hip; posterior trunk bending brings the line of force behind the hip, to compensate for weak hip extensors.
A different type of posterior trunk bending may occur early in the swing phase, where the subject may throw the trunk backwards in order to propel the swinging leg forwards. This is most often used to compensate for weakness of the hip flexors or spasticity of the hip extensors, either of which makes it difficult to accelerate the femur forwards at the beginning of swing. This manoeuvre may also be used if the knee is unable to flex, since the whole leg must be accelerated forwards as one unit, which greatly increases the demands on the hip flexors. Posterior trunk bending may also occur when the hip is ankylosed (fused), the trunk moving backwards as the thigh moves forwards.
Many people have an exaggerated lumbar lordosis, but it is only regarded as a gait abnormality if the lordosis is used to aid walking in some way, which generally means that the degree of lordosis varies during the course of the gait cycle. Increased lumbar lordosis is observed from the side of the subject and generally reaches a peak at the end of the stance phase on the affected side.
The most common cause of increased lumbar lordosis is a flexion contracture of the hip. It is also seen if the hip joint is immobile due to ankylosis. Both of these deformities cause the stride length to be very short, by preventing the femur from moving backwards from its flexed position. This difficulty can be overcome if the femur can be brought into the vertical (or even extended) position, not through movement at the hip joint but by extension of the lumbar spine, with a consequent increase in the lumbar lordosis (Fig. 3.8).
Fig. 3.8 • Increased lumbar lordosis: when there is a fixed flexion deformity of the hip (left panel), the whole pelvis must rotate forwards for the femur to move into a vertical position (right panel), with a resulting increase in lumbar lordosis.
The orientation of the pelvis in the sagittal plane is maintained by the opposing pulls of the trunk muscles above and the limb muscles below. If there is muscle imbalance, for example a weakness of the muscles of the anterior abdominal wall, weakness of the hip extensors or spasticity of the hip flexors, the subject may develop an excessive anterior pelvic tilt, again with an increase in the lumbar lordosis.
Four gait abnormalities (circumduction, hip hiking, steppage and vaulting) are closely related, in that they are designed to overcome the same problem – a functional discrepancy in leg length. A review on the topic of leg length discrepancy was published by Gurney (2002).
An ‘anatomical’ leg length discrepancy occurs when the legs are actually different lengths, as measured with a tape measure or, more accurately, by long-leg X-rays. A ‘functional’ leg length discrepancy means that the legs are not necessarily different lengths (although they may be) but that one or both are unable to adjust to the appropriate length for a particular phase of the gait cycle. In order for natural walking to occur, the stance phase leg needs to be longer than the swing phase leg. If it is not, the swinging leg collides with the ground and is unable to pass the stance leg. The way that a leg is functionally lengthened (for the stance phase) is to extend at the hip and knee and to plantarflex at the ankle. Conversely, the way in which a leg is functionally shortened (for the swing phase) is to flex at the hip and knee and to dorsiflex at the ankle. Failure to achieve all the necessary flexions and extensions is likely to lead to a functional leg discrepancy and hence to one of these gait abnormalities. This usually occurs as the result of a neurological problem. Spasticity of any of the extensors or weakness of any of the flexors tends to make a leg too long in the swing phase, as does the mechanical locking of a joint in extension. Conversely, spasticity of the flexors, weakness of the extensors or a flexion contracture in a joint makes the limb too short for the stance phase. Other causes of functional leg length discrepancy include musculoskeletal problems such as sacroiliac joint dysfunction.
An increase in functional leg length is particularly common following a ‘stroke’, where a foot drop (due to anterior tibial weakness or paralysis) may be accompanied by an increase in tone in the hip and knee extensor muscles.
The gait modifications designed to overcome the problem may either lengthen the stance phase leg or shorten the swing phase leg, thus allowing a normal swing to occur. They are not mutually exclusive and a subject may use them in combination. The gait modification employed by a particular person may have been forced on them by the underlying pathology or it may have been a matter of chance. Two people with apparently identical clinical conditions may have found different solutions to the problem.
Ground contact by the swinging leg can be avoided if it is swung outward, in a movement known as circumduction (Fig. 3.9). The swing phase of the other leg will usually be normal. The movement of circumduction is best seen from in front or behind. Circumduction may also be used to advance the swinging leg in the presence of weak hip flexors, by improving the ability of the adductor muscles to act as hip flexors while the hip joint is extended.