➢ Spasticity is a term commonly used to describe a collection of muscle overactivity patterns associated with the upper motor neuron syndrome including actual spasticity, clonus, dystonia, co-contraction, associated reactions, and flexor or extensor spasms.
➢ Gait dysfunction in the upper motor neuron syndrome can be due to a combination of paresis, impaired coordination and balance, and muscle overactivity and contracture.
➢ Treatment options include physiotherapy, assistive devices, orthotic devices, oral and intrathecal medications, intramuscular chemodenervation, neurolysis, and/or neuro-orthopaedic surgical procedures.
➢ The objective evaluation of walking and its underlying muscle activation patterns can be performed by the simultaneous collection of joint kinematics, kinetics, and dynamic electromyography (EMG) data that serve as an extension of the physical examination to better discern primary gait deviations from compensatory gait deviations as well as underlying muscle overactivity from contracture.
➢ Despite the science behind instrumented three-dimensional gait analysis, its specific contribution to clinical and surgical decision-making is not well utilized because of its associated cost, the incorrect view by some insurance companies of gait analysis being a research technique, the dearth of qualified clinical gait and motion analysis laboratories accessible to adult patients, and limited access to orthopaedic surgeons who have experience in the treatment of neurological disorders for this patient population.
Abnormal limb postures and gait dysfunction are common consequences of the upper motor neuron syndrome due to cerebrovascular accident, spinal cord injury, cerebral palsy, and traumatic brain injuries. The objective of this article is to review the upper motor neuron syndrome, the abnormal gait, and the application of instrumented gait analysis in the management of adults with spastic gait dysfunction.
The upper motor neuron syndrome is characterized by damage to the corticospinal system resulting in a collection of negative and positive signs1. Negative signs refer to the phenomena of absences such as paresis and loss of coordination. Positive signs refer to the phenomena of presences such as hyperreflexia and spasticity. From a clinical point of view, functional problems caused by the upper motor neuron syndrome have more to do with negative signs. However, it is much easier to reduce antagonistic muscle overactivity or contracture that restricts motion than it is to strengthen paretic muscles and to retrain muscle control.
Abnormal muscle activity includes spasticity, clonus, dystonia, co-contraction, associated reactions, and flexor and extensor spasms (Table I)2. For simplification, these abnormal muscle activation patterns are collectively referred to as “spasticity.”3 However, the classical definition of spasticity, as characterized by Lance4, is “an increase in velocity dependent tonic stretch reflexes with exaggerated tendon jerks.” The issue of terminology is more than semantic, as spasticity may be amenable to surgical lengthening, whereas dystonia and co-contractions are less likely to improve with a surgical procedure5. We have elected to use the term “muscle overactivity” as it is a more encompassing term6-8. Changes may also occur in the structural and mechanical properties of the muscles, tendons, and other connective tissues about a joint resulting in increased stiffness and contracture that may be confused for muscle overactivity7-10. Thus, the resistance perceived by an examiner stretching a muscle may be increased because of muscle overactivity or rheological changes in muscles and other soft tissues. Differentiating between abnormal muscle activity and soft-tissue contracture can help to direct treatment with regard to physiotherapy, oral or intrathecal medications, chemodenervation, orthotics, and/or surgical interventions2,6,7,10.
There are four treatment considerations that are useful for guiding management of upper motor neuron syndrome-related problems: (1) the treatment approach will vary as a function of the stage of recovery, (2) the treatment approach will differ for a focal problem compared with a diffuse problem, (3) many pharmacological agents that are useful in treating sequelae of upper motor neuron syndrome have adverse effects that may worsen function, and (4) the treatment should address passive, active, or functional objectives10.
Neurological recovery can be arbitrarily divided into an early period during which motor recovery may be expected and a late period during which, for all practical purposes, motor recovery has ended. During the period of motor recovery, temporizing interventions such as targeted chemodenervation with botulinum toxin A or neurolysis with phenol or ethanol can be used to address muscle overactivity. When these agents wear off in approximately three to six months, reevaluation is performed to determine whether additional motor recovery has taken place and whether there is further indication for treatment. Functional recovery is different from neurological recovery and can occur many years after the initial injury. For instance, a patient who is unable to walk because of an equinovarus deformity may regain the ability to walk through orthopaedic interventions to correct the problem.
The treatment of focal problems compared with diffuse problems will also differ in approach. When restricted motion can be attributed to a small number of muscles, chemodenervation or surgical lengthening or releases of these muscles is feasible. If the problem is diffuse or generalized, oral or intrathecal antispasticity agents may be considered. However, oral agents have adverse effects, such as sedation, weakness, and impaired cognition, and possibly interfere with neurological recovery11.
Abnormalities, such as traumatic brain injuries and cerebrovascular accident, produce highly variable presentations of coexisting residual voluntary function and muscle overactivity and must be carefully assessed using an objective and functional approach for each patient prior to planning surgical interventions. Observational gait analysis is commonly used but has been identified as an inadequate diagnostic method in the evaluation of gait abnormalities12-14. Therefore, these distinctions should not be based on physical examination and observation alone, but the clinician should consider including instrumented gait analysis to provide a more sound basis for implementing conservative and surgical interventions15-19. The clinical examination and instrumented gait analysis should attempt to answer the following questions. Which muscles can the patient voluntarily activate and to what degree? Is the muscle spastic, that is, activated in response to a stretch? Is the muscle co-contracting as an antagonist during active movement? Does the joint have motion limitations due to contracture20?
Normal human locomotion consists of complex interaction of the head, trunk, arms, and lower limbs with the fundamental goal of moving from one point to another in a safe and efficient manner. These interactions are cyclic and can be characterized by the timing of foot contact with the ground. An entire sequence of functions by one limb is identified as a gait cycle (Fig. 1).
Instrumented gait analysis is a useful clinical tool in the assessment and treatment of gait dysfunction and is not an experimental methodology. Advances in technology have allowed for large amounts of data to be obtained simultaneously from a variety of sources and to be processed rapidly for clinical interpretation. Despite these advances, the evaluation of a single patient may take approximately one hour because of thorough clinical examination, application of instrumentation, and multiple trials of walking. Multiple trials are recommended to account for increased variability in instrumented gait analysis data in patients with abnormalities compared with normal subjects21. Instrumented gait analysis is not indicated in every patient with gait dysfunction secondary to upper motor neuron syndrome. For instrumented gait analysis to be useful in clinical evaluation, the measured parameters should meet the following five criteria: (1) gait analysis should supply additional and more pertinent information than that of the clinical examination, (2) gait analysis should correlate with the functional capacity of the patient, (3) gait analysis should be accurate and repeatable, (4) gait analysis should result from a test that minimally alters the natural performance of the patient, and (5) gait analysis should be interpreted by experienced clinicians familiar with the scope of the test protocol, instrumentation, and limitations of the equipment.
Caution should also be exercised with regard to patients with severe cognitive deficits as the unfamiliar environment and instrumentation may alter the patient’s performance. In addition, patients with severe balance dysfunction may not be able to undergo testing. The clinician should also have a clear understanding of instrumented gait analysis data and should be able to relate these features to the abnormal motion that is observed during walking to effectively diagnose and to address the problems of abnormal gait22.
The analysis of motion, regardless of what external or internal forces are producing it, is referred to as kinematics. Temporal spatial parameters are simple to obtain, are useful for evaluating gait, and can be recorded with instrumented walkways (Fig. 2) and wearable inertial sensors23,24. A step period is defined as from an event in one foot until the subsequent occurrence in the other foot (e.g., right initial contact to left initial contact). Step length is the distance covered from one foot to the other. The stride period is defined as from an event in one foot until the recurrence of the same event in the same foot. Cadence refers to the number of steps in a period of time (steps per minute).
Three-dimensional motion analysis refers to a quantitative description of the movement of body segments. Most modern gait laboratories use specialized optoelectronic tracking systems to determine anatomic marker positions in space and time (Fig. 3). Joint angles (Fig. 4-A), linear and angular velocities, and accelerations are some of the calculated measures. However, caution must be exercised during instrumentation, as variability in marker placement can be a cause of decreased repeatability and reliability of motion analysis data21,25.
Joint kinetics are calculated when kinematic data are combined with ground reaction forces. The ground reaction force is a reflection of the body’s mass and acceleration as it contacts the ground. Whereas gravity is mostly responsible for the generation of normal forces, friction is responsible for the generation of shear forces. Moment is the tendency of a force to rotate a joint. It is the product of force and the distance from the center of rotation. Internal forces, generated primarily by muscles, ligaments, joint capsule, and the geometry of the articulating joint surfaces, counteract external rotational forces (Fig. 4-B). For example, the ground reaction force, when positioned posterior to the ankle during initial contact, produces a plantar flexor moment. To counter this moment, the tibialis anterior will produce a dorsiflexor moment22,26. Work is defined as the product of moment and angular displacement. Work is positive when the trajectory of the limb corresponds to the direction of the moment22. Power is defined as the product of joint moment and angular velocity and reflects the rate of work22. Concentric contractions correlate with power generated for propulsion, whereas eccentric contractions correlate with power absorbed for deceleration. For example, the ankle plantar flexors demonstrate power absorption during the loading response and power generation during push-off (Fig. 4-C).
The EMG signal is a recording of motor unit action potentials that reflects skeletal muscle activation. These electrical signals can be recorded and can be analyzed to determine the timing and relative intensity of the muscular activity. Although the patterns of muscle activity during walking have been well documented (Fig. 5), understanding the functional importance of these activation patterns during gait is important and can be challenging. This is most likely because clinicians have a tendency to think of muscle activity in terms of concentric muscle contraction in the open kinetic chain. This approach can be applied to the swing phase of the gait cycle where hip flexion observed throughout the swing phase is generated mainly by active concentric muscle activation of the hip flexors. However, the same does not hold true for swing-phase knee flexion, which is largely an inertial byproduct of hip flexion and late-stance-phase ankle plantar flexion22,27,28. In addition, when the foot is in contact with the ground, a closed kinetic chain is formed that allows muscles to influence joints that they do not cross, thus increasing the complexity of the relationship between muscle activity and joint movement. For example, the ankle plantar flexors generate push-off in the late stance phase via concentric contraction but also contract isometrically and eccentrically earlier in the stance phase to control forward progression of the tibia, contributing to stance-phase knee stability22,27.
It is important to emphasize that EMG provides only information about the timing of muscle activation. Abnormalities in activation patterns include absence of activity, abnormal onset of activation (early or delayed), abnormal duration of activation (abbreviated or prolonged), and presence of abnormal activation patterns (out-of-phase activity)29,30. EMG can also assist in identifying spasticity as evidenced by inappropriate activity associated with motion or stretch. Clonus is demonstrated electromyographically as short-duration bursts of repetitive electrical activity that occur in response to stretch2. Dynamic polyelectromyographic studies should consider including all muscles capable of producing the target movement, not just the muscles that cross the affected joint, to reveal problems with timing and coordination that may be responsible for a gait dysfunction8.
Studies of patients with hemiparesis due to cerebrovascular accident or traumatic brain injury have shown that weakness of the hip flexor, knee extensor, and ankle plantar flexor muscles is the primary factor contributing to impaired walking31-33. Initial contact at the forefoot and decreased ankle dorsiflexion are typically observed. The plantar flexed attitude of the affected ankle can be caused by muscle overactivity and/or contracture. Excessive knee extension or knee flexion may also occur. These deviations result in impaired limb stability that correlates with findings of decreased stance time on the affected limb, increased double support duration, and shorter step length for the unaffected limb33,34. The initiation of the swing phase is delayed and more effortful, consistent with the findings of increased swing time on the affected side31-35. From a biomechanical standpoint, this is due to an abnormal force transfer from the hindfoot to the forefoot and reduced or absent push-off in the terminal stance33. During the swing phase, decreased hip and knee flexion and ankle equinovarus can lead to impaired clearance on the affected side with compensatory hip hiking or circumduction (Fig. 6).
When there is bilateral involvement (e.g., spinal cord injury or traumatic brain injury), the resultant lower-limb involvement may or may not be symmetric. Individuals with spinal cord injury who are able to walk usually have a lower-level or incomplete injury. Traumatic brain injury is more likely to have a more variable presentation because of the diffuse nature of the injury with the possibility of involvement in other structures such as the basal ganglia and cerebellum. A scissoring gait pattern can be observed in the swing phase, leading to a reduced base of support and stability. However, using the hip adductors may also be a compensatory strategy to assist with hip flexor weakness. This differentiation becomes particularly important when considering treatment, as elimination of hip adduction may render a paraparetic patient unable to walk36,37. The knees can be flexed or can be hyperextended with the ankles in equinus, further impairing clearance and stability.
The goal of management and treatment of gait dysfunction following a central nervous system injury should focus on improving lower-limb stability, clearance, and advancement. Instrumented gait analysis can help to delineate the role of weakness, muscle overactivity, and contracture in gait dysfunction. Clinical examination and instrumented gait analysis can be repeated after a temporary diagnostic nerve block, taking care not to disturb the instrumentation. This approach can assist with differentiating between the contributions of contracture and muscle overactivity to a deformity and gait dysfunction and can guide further treatment. The rest of this section will review common joint deformities that contribute to gait dysfunction in the upper motor neuron syndrome and their underlying biomechanics, muscle overactivity patterns, and treatment.
Equinovarus foot deformity is commonly observed with upper motor neuron syndrome but can also be the result of ankle immobilization, fractures, and surgical procedures. The foot and ankle assume a toe-down (plantar flexed) posture and are also frequently inverted (varus). Gait observation reveals initial contact with the forefoot, with weight borne primarily on the anterior and lateral border of the foot with or without coexisting toe flexion. Restricted ankle dorsiflexion during midstance prevents forward progression of the tibia over the stationary foot, producing knee hyperextension and increasing metatarsal pressure. During the swing phase, plantar flexion of the foot may result in limb clearance problems.
Overactivation of ankle plantar flexors during the swing and/or stance phase, or underactivation of the ankle dorsiflexors during the swing phase, can lead to this deformity (Fig. 7). Muscles to consider studying include the gastrocnemius, soleus, tibialis anterior, tibialis posterior, extensor hallucis longus, flexor digitorum longus, or flexor hallucis longus38,39. When it is difficult to differentiate between the contribution of the tibialis anterior and tibialis posterior to a varus deformity, a diagnostic tibial nerve block with lidocaine can be performed. Surgical intervention in the form of Achilles tendon lengthening, split tibialis anterior tendon transfer, and myotendinous lengthening of the extensor hallucis longus may be considered in these cases40-44. To supplement the weak ankle plantar flexors and avoid toe curling, a release and transfer of the long toe flexors to the calcaneus should be considered45.
Hitchhiker’s (Hyperextended) Great Toe
Observation of barefoot walking reveals the great toe held in extension during the stance and swing phases. Ankle equinovarus posture may accompany this deformity. When wearing shoes, the patient may report pain at the first metatarsal head and the tip of the great toe. Dynamic polyelectromyography will demonstrate overactivity of the extensor hallucis longus. However, a number of other muscles may also contribute, including weakness of the flexor hallucis longus. When equinovarus is present, analysis of the tibialis anterior, tibialis posterior, gastrocnemius, soleus, and long toe flexors should be included7. A trial of an intramuscular injection of botulinum toxin A to the extensor hallucis longus can be effective in reducing this deformity and relieving the patient’s symptoms; however, treatment of the equinus deformity should also be considered to prevent worsening plantar flexion deformity due to further dorsiflexor weakness. Surgical lengthening of the extensor hallucis longus tendon may be beneficial, and, in selected cases, transfer to the midfoot will supplement dorsiflexion weakness. When the flexor hallucis longus is also overactive, surgical lengthening to both the extensor hallucis longus and the flexor hallucis longus may be required to prevent development of an opposite deformity10.
Stiff Knee Gait Pattern
Stiff knee gait results from a dynamic deformity created by muscle contraction and external moments rather than a structural deformity of the knee joint. In this gait deviation, the knee is maintained in extension throughout the swing phase and the moment of inertia of the lower limb is increased, further impairing hip flexion as well. The lack of adequate limb clearance due to reduced hip and knee flexion can result in a foot drag, even if the ankle has adequate dorsiflexion. Compensatory mechanisms in the trunk, ipsilateral hip (e.g., circumduction), and contralateral limb (e.g., vaulting or early heel rise) may also be present. Kinematic studies will typically demonstrate diminished and/or delayed peak swing-phase knee flexion.
Abnormalities in both the stance and swing phases at multiple joints can contribute to this dynamic deformity. Out-of-phase activation of the rectus femoris in the swing phase can be a major contributor to this pattern as it crosses both the hip and knee joint and can restrict knee flexion. Out-of-phase activation of the vastae muscles in the swing phase can also contribute to the deviation. At the hip, overactivity of the gluteus maximus and hamstrings during the swing phase may act to restrain hip flexion, resulting in a stiff knee gait. A weak iliopsoas can also result in reduced hip flexion in the swing phase, which can further impair knee flexion. Ankle equinus can contribute to knee hyperextension in the stance phase by preventing forward progression of the tibia and delaying knee flexion. In general, when a concomitant ankle deformity is observed, the ankle should be addressed first, as this may result in subsequent improvement in the stiff knee gait pattern.
If a stiff knee is present in absence of ankle deformity or persists despite treatment of the ankle, a diagnostic lidocaine block of the femoral nerve motor branch to the rectus femoris can be performed to help differentiate the force contribution of the knee muscles compared with the hip extensor muscles. This can be followed by treatment with phenol neurolysis or intramuscular botulinum toxin A injections of the offending muscles and aggressive stretching46,47. Selective surgical release of the rectus femoris or of the rectus femoris and vastus intermedius can be considered48. Transfer of the rectus femoris to the semitendinosus, to the gracilis, medial to the sartorius, or lateral to the iliotibial band may be considered to promote knee flexion49,50. Surgical release of all four heads of the quadriceps could result in postoperative knee instability and is not recommended.
Flexed Knee Deformity
The flexed knee deformity refers to the flexed posture of the knee in the stance and swing phases. Not only does this deformity impair limb stability and contralateral limb clearance, but the lack of knee extension in terminal swing also limits limb advancement. There may also be a compensatory increase in contralateral hip and knee flexion to assist with clearance. This gait pattern is often associated with medial and lateral hamstring muscle overactivity or contracture. Other factors that can contribute to this dysfunction include overactivity of the gastrocnemius in the stance phase and of the iliopsoas in the swing phase. Weakness of knee extensors or plantar flexors may also lead to knee flexion in the stance phase.
If the hamstring muscles are the primary contributors, a temporary sciatic nerve block can differentiate between muscle overactivity and contracture. Phenol injections to the motor points of the hamstrings (direct injection of the sciatic nerve is not recommended) or intramuscular injection of botulinum toxin A may be performed to reduce knee flexor overactivity. If knee flexion deformity is persistent or is due to contracture, the biceps femoris, gracilis, and semimembranosus can be fractionally lengthened and the semitendinosus can be divided or undergo Z-lengthening51.
Adducted Thigh Deformity
The adducted thigh deformity can also be referred to as a scissoring gait pattern. Severe hip adduction during the swing phase can interfere with limb advancement and result in a narrow base of support during the stance phase, increasing the risk for falls. Furthermore, severe adductor overactivity promotes hip subluxation or dislocation. In addition, this deformity may interfere with hygiene, dressing, sexual intimacy, sitting, standing, and transfers. Urinary tract infections, heterotopic ossification, hip subluxation, undiagnosed pelvic or long bone fractures, and other painful stimuli should be ruled out, as they can further exacerbate overactivity of the hip adductors. Kinematic studies will demonstrate diminished hip abduction in the swing phase. Overactive adductor longus, adductor brevis, adductor magnus, gracilis, and/or pectineus muscles or weak iliopsoas, sartorius, and/or gluteus muscles in the swing phase can contribute to this deformity. A diagnostic lidocaine block of the obturator nerve can help differentiate hip adductor muscle overactivity from contracture. Phenol neurolysis of the obturator nerve or botulinum toxin A injections can reduce dynamic hip adduction deformity. Surgical release of the offending adductor muscles, usually by proximal myotomy, can be performed52. Obturator neurectomy can also be performed if the hip range of motion is adequate but increased tone is present53,54.
Flexed Hip Deformity
Hip extension impairments are a common physical examination finding on assessment and may be due to contracture from prolonged sitting or overactivity of the hip flexor muscles (e.g., iliacus, psoas, pectineus, and rectus femoris). In addition, heterotopic ossification, hip dislocation, and undiagnosed pelvic or long bone fractures or other painful stimuli should be ruled out, as pain can increase overactivity of the hip flexors and hip flexor posturing.
Hip flexion deformities may also contribute to knee flexion posturing and can restrain hip extension during the late stance phase. The rectus femoris can also restrict hip flexion during the early swing phase. Kinematic studies may reveal excessive anterior pelvic tilt and reduced hip extension in middle and terminal stances, and kinetic studies may reveal diminished hip flexor power generation. Other muscles that may contribute to this deformity are the adductor longus, adductor brevis, sartorius, tensor fasciae latae, and anterior portion of the gluteus medius10. Clinical examination, in combination with dynamic EMG recordings, can help to determine if the deformity is obligatory or compensatory in nature. Intramuscular botulinum toxin A injections or phenol motor point injections to the offending muscles may be helpful. Surgical release of persistent overactive or contractured flexor muscles should be considered and should be accompanied by postoperative rehabilitation to promote correction of the residual deformity. If knee flexion deformity is present, it should be addressed at the same time to reduce the risk of recurrence.
Gait dysfunction resulting from muscle overactivity, contracture, impaired motor control, and balance after upper motor neuron injury is complex because of its multimuscle involvement and the difficulty discerning primary from compensatory gait deviations. Treatment options include physiotherapy, assistive devices, orthotics, oral and intrathecal medications, intramuscular chemodenervation, neurolysis, and/or neuro-orthopaedic surgical procedures.
Despite the logic behind instrumented gait analysis, it has had limited acceptance by some because of its associated cost42,43,55,56. Obtaining financial coverage for instrumented gait analysis from insurance companies can be challenging, as its specific contribution to clinical and surgical decision-making in adult patients with spastic gait dysfunction is viewed as controversial and as an unproven diagnostic study by some. This claim is supported by a study that found only moderate to substantial agreement among physicians at different institutions with regard to identification of soft-tissue problems and treatment recommendations in a pediatric population57. Subjectivity in the interpretation of objective data and variability in raw data collection among institutions were suspected to be responsible for these differences25,57. However, other studies have found instrumented gait analysis to provide valuable diagnostic information, to alter treatment considerations, to confirm surgical planning, and even to improve outcomes15,18,19,44,57. Another hurdle to the incorporation of instrumented gait analysis is the dearth of qualified clinical gait and motion analysis laboratories accessible to patients as well as access to orthopaedic surgeons with experience in the treatment of neurological disorders. Thus, future efforts should be directed at the following objectives: continuing to develop and to refine instrumented gait analysis to improve repeatability and reliability in data collection and interpretation; establishing the benefits of incorporating instrumented gait analysis into the treatment of spastic gait dysfunction, especially with regard to outcomes and cost; and increasing accessibility of patients to these laboratories and the training of orthopaedic surgeons in neuro-orthopaedic management.
These efforts should help to achieve better functional outcomes and to improve the quality of life and independence for adults with disability due to upper motor neuron syndrome.
Investigation performed at the MossRehab Sheerr Gait and Motion Analysis Laboratory, Elkins Park, Pennsylvania
Disclosure: There was no source of external funding for this study. The Disclosure of Potential Conflicts of Interest forms are provided with the online version of the article.
- Copyright © 2016 by The Journal of Bone and Joint Surgery, Incorporated