Principles and Practices of Neurological Rehabilitation

Published on 12/04/2015 by admin

Filed under Neurology

Last modified 22/04/2025

Print this page

rate 1 star rate 2 star rate 3 star rate 4 star rate 5 star
Your rating: none, Average: 1 (1 votes)

This article have been viewed 3965 times

Chapter 48 Principles and Practices of Neurological Rehabilitation

Neurological rehabilitation fosters assessments and practices that extend into every aspect of the care of patients with acute and chronic neurological disabilities. Managing the rehabilitation needs of patients should be a priority for the clinical neurologist who seeks opportunities to lessen impairments and disabilities and to meet the requests of patients to increase their ability to participate in daily home and community activities (Dimyan et al., 2008). To best assist patients, the clinician must determine how a patient’s physical and cognitive deficits cause disabilities; consider what tasks patients can and cannot perform independent of assistance and at the speed and accuracy necessary for daily activities; design practice and training paradigms with health-related professionals to lessen disabilities that are important to patients and their caregivers; consider interventions that manipulate the fundamental mechanisms that induce neural adaptations for cerebral reorganization and learning; and anticipate and manage the neuromedical and psychosocial complications of immobility, loss of motor control, cognitive impairment, and functional dependence.

Goals and Structure of Rehabilitation

Rehabilitation training reduces physical and cognitive impairments and their related disabilities in an effort to improve functional independence and health-related quality of life. Training involves an active learning process that requires motivation, guidance, goal setting, progressive practice, and social support.

Aims

Neurological rehabilitation employs multidisciplinary services to improve functional and cognitive skills such as walking and language, reduce disability in personal care and other daily activities, lessen the burden of care provided by family and society for disabled persons, and prevent and manage complications such as dysphagia, contractures, pressure sores, and depression. Although the links between disease pathology, physical and cognitive impairments, disabilities, participation, and handicap are not always clear, physicians and therapists ultimately target the health-related quality of life of patients by maximizing functional independence for home and community pursuits. Patients, caregivers, and families must be fully involved in the rehabilitation process if they are to successfully facilitate self-management, compensatory adaptations, and more independent skills.

The quality of a rehabilitation care plan is only as good as the assessment it is based on. Accurate assessment means getting to the bottom of the mechanisms behind problems that adversely affect patient functioning. Neurorehabilitation assessment includes identifying the most productive focus for interventions and the most appropriate setting in which better outcomes can be achieved within expected time frames. Clinical evaluation initiates a treatment program that is continually revised in light of successive assessments. Over the long run, the clinician monitors for complications and functional changes that present new opportunities to return patients to a higher level of function and participation.

An expert assessment acquires sufficient information to allow a reasonably accurate initial prediction of the potential outcome for the patient after a course of rehabilitation. Both short- and long-term goals take into account the amount of likely neurological recovery and the amount of residual disability. The long-term goal often can be broken down into component steps that move steadily toward the final outcome. Progressive goal setting is a technique to encourage the patient as each short-term objective is achieved, as well as serving to monitor efficacy and identify emerging confounders of gains. Short-term goals must be relevant, motivating, explicit, attainable, measurable, and agreeable to the patient.

To achieve these aims, the rehabilitation process differs from the usual medical model of care by including personnel from multiple disciplines, problem-solving strategies that include methods to engage mechanisms of neuroplasticity, standardized outcome measures, and the organization of home and community services to meet the patient’s needs.

Personnel and Strategies

A team approach to inpatient and outpatient care best manages the diverse problems faced by disabled patients and their families (Fig. 48.1). In a multidisciplinary model, each member with specialty training treats particular disabilities. In an interdisciplinary model, roles blend. An interdisciplinary approach is oriented toward problem-solving to improve functional outcomes, rather than being bound by individual disciplines. For example, training procedures for motor and cognitive learning or behavioral modification are reinforced by all members of an interdisciplinary group, using agreed-on strategies. These interaction styles are not mutually exclusive. Most teams move between the two models when they formally meet to discuss the patient’s progress and to adjust goals and treatments.

The care milieu created by the team of therapists, nurses, social workers, neuropsychologists, and physicians, with its emphasis on lessening disability, is one of rehabilitation’s most powerful tools. Studies of inpatient stroke rehabilitation, for example, support the team approach as an efficient way to organize services for patients with functional disabilities. With traumatic brain injury (TBI) or spinal cord injury (SCI), the special needs of affected patients suggest that interdisciplinary inpatient and outpatient care will lead to fewer medical and psychosocial complications.

Physicians

An understanding of the underlying disorder, including the mechanisms of disability, potential outcomes, and natural history of the disease being managed, is critical in planning a rehabilitation program for a patient. This expertise may be provided by the growing number of neurologists with expertise in neurorehabilitation who can bring principles from their increasing storehouse of knowledge to bear on recovery, and by rehabilitation physicians or physiatrists, who also have broad experience in musculoskeletal, orthopedic, and cardiopulmonary rehabilitation issues. Orthopedists, urologists, psychiatrists, plastic surgeons, neurosurgeons, and podiatrists often are consulted during rehabilitation and for long-term management of disabled patients.

The clinician superimposes the contributions of neurological, musculoskeletal, cardiopulmonary, and other impairments on a map of the patient’s functional abilities and disabilities. For example, does tender musculoligamentous tissue cause pain or limit movement, or does spasticity lead to loss of motor control? Does a medication or metabolic abnormality lessen concentration, the ability to learn, or endurance for exercise? Physicians tend to be the facilitators of the multidisciplinary team, especially during inpatient care. Here, the physician may conduct a weekly team conference that reviews the patient’s progress in reaching the functional goals that will permit a discharge to the home. To do this well, the physician must help build the team’s infrastructure and understand the practices of its disciplines. Rehabilitation physicians should serve as clinician-scientists as well. The physician can encourage therapists to weigh, formulate, and test strategies. Drawing on current literature and collaborating with basic and clinical researchers, the neurological rehabilitation specialist can optimally assess and develop interventions and creative solutions (Dobkin, 2003).

Physicians should explain to both patient and primary care doctor the indications for medications, measures for secondary prevention of complications, management of risk factors for recurrence or exacerbation of the disease, and the type and duration of rehabilitative interventions. Increasingly, doctors must address the risks and possible benefits of not only medications and usual rehabilitative approaches to care but also potential research interventions such as cellular transplantation (Dobkin et al., 2006b). During outpatient care, physicians must provide informed counseling about exercise and home practice for motor and cognitive retraining. The clinician reviews the details of what the patient is practicing to improve walking, the functional use of an affected upper extremity, language and memory skills, and socialization. Education should be offered about how task-specific practice may alter the brain’s representations of these activities and improve the patient’s abilities, even years after the initial neurological illness. For patients with chronic diseases that progress, such as multiple sclerosis (MS), practice is perhaps even more important because it may spur gradual neural reorganization to maintain function. Clinicians can encourage patients to increase their strength, speed, and precision of multijoint movements and to build cardiovascular fitness. The physician also should monitor outcomes with serial tests of the activities that are targeted to best determine the optimal dose of a treatment. For example, if the gait pattern is suboptimal, the clinician can test walking speed for 50 feet or the distance walked in 2 minutes to reassess progress in mobility at each visit. By documenting the effects of treatment, the physician can best advocate the continuing goals of rehabilitation to patients and insurers.

The Internet has many sites from which to develop educational materials that physicians and the therapy team can offer patients (e.g., http://www.uclahealth.org/body.cfm?id=1174), as well as lectures and articles about recovery and about experimental interventions.

Rehabilitation Nursing

Traditionally, the nursing role has been one of providing care and support during a phase of illness and doing for others the things they would normally do for themselves. Nurses have particular expertise in bowel and bladder management and have developed the post of continence advisor, particularly for teaching chronic intermittent self-catheterization (CISC) and scheduled voiding programs. They teach skin care and pressure sore management. In an inpatient unit, nurses are in constant contact with the patient undergoing rehabilitation. This extended contact with patients allows nurses to address the issue of carryover of skills from physical and occupational therapy sessions to other areas fundamental for function in the community. Each activity is integrated with others; for instance, continence management through CISC may require improvements in upper limb coordination, trunk control, mobility, lower limb tone, medications to optimize bladder and sphincter control, and strategies to facilitate problem solving. Nurses in community programs also have become involved in the management of individual chronic diseases including MS, amyotrophic lateral sclerosis (ALS), and Parkinson disease (PD), especially attending to gaps in services needed by patients. A nurse practitioner can be a valuable asset to the physician and team on a busy inpatient service, especially in a university hospital, where patients tend to have complex medical illnesses and needs. The Association of Rehabilitation Nurses has excellent resources for continuing education (www.rehabnurse.org).

Physical Therapists

Physical therapists (PTs), or physiotherapists relate voluntary motor control and patterns of multijoint movements, sensory appreciation, range of motion of joints, strength, balance, and endurance to the training needs for bed and wheelchair mobility, standing up, walking, and functional mobility during activities. PTs bring expertise to the team in wheelchair design, assistive devices, and orthoses. They manage compensatory strategies for carrying out activities of daily living (ADLs) such as the use of a walker and offer interventions to lessen specific impairments. PTs play a primary role in managing musculoskeletal and radicular pain, contractures, spasticity, and deconditioning.

Two broad categories of physical therapy, therapeutic exercise and the so-called neurophysiological and neurodevelopmental techniques, were the bulwark of the approaches used by therapists in the past (Box 48.1). Newer concepts related to practice-induced neuroplasticity, motor control, and skills learning have taken greater hold beginning in the past decade.

Conditioning and Strengthening

Light resistance exercises for any UMN or lower motor neuron (LMN) disease, from stroke and SCI to ALS, the postpolio syndrome, and the muscular dystrophies, generally are safe and effective in improving strength and sometimes function. Strength can be increased without inducing spasticity in patients with UMN diseases and without muscle tissue injury in those with neuromuscular diseases. Concern about falls, disability, and muscle atrophy in older adults has led to many studies that show that a strengthening program can benefit any sedentary person. Resistance training can lead to an increase in strength without any improvement in muscle bulk, probably by augmenting the amount of supraspinal input that is recruited to the task. Thus, strengthening can be considered a form of motor learning. Isometric resistance exercises probably are the safest approach for weak patients and can be performed without equipment. For example, flexing the elbow of one arm about 60 degrees and pressing down on that forearm with the palm of the other arm to reach an equilibrium of tension in each arm will enhance strength in the shoulder girdle, elbow flexors and extensors, and forearm groups. To build muscle mass requires the subject to perform 1 or 2 sets of 8 to 15 repetitions 3 times a week against 50% or more of the maximum resistance manageable in a single lift. Pool therapy can be used to augment fitness and strengthening exercises. Patients may work against the resistance of the water, for example, by repeatedly flexing, extending, abducting and adducting each leg at the hip while standing on the other or by using swim strokes. Practice walking in a pool allows the patient to make use of the water’s buoyancy, but light refraction within the water may make visual cues for foot placement less reliable.

Small trials of medications that act on the neuromuscular junction (e.g., pyridostigmine), those that work on muscle metabolism (e.g., β2-adrenergic agonists, creatine), potassium channel blockers that act on demyelinated axons (e.g., aminopyridines), and hormones such as androgens and human growth factor, which may lessen disuse atrophy, have revealed modest efficacy when combined with resistance exercise.

Fitness training is valuable in UMN, extrapyramidal, and LMN diseases. Repetitive exercise, at least in animal models of stroke and SCI, also has induced potential reparative biological effects such as neurogenesis and increased expression of neurotrophic factors. This finding may be used to motivate patients. Treadmill walking can be used as an aerobic workout for the older adult with hemiparetic stroke. After a complete SCI, cardiovascular conditioning exercise is limited by the upper body’s small muscle mass, by pooling of blood in the leg muscles, which reduces cardiac preloading, and by impaired cardiovascular reflexes. Functional electrical stimulation (FES) of the lower extremities during cycle ergometry can improve both peripheral muscular and central cardiovascular fitness. Clinical trials suggest that FES exercise in sets of 10 to 15 repetitions against an increasing load resistance of 1 to 15 kg over 12 weeks will increase muscle bulk, improve strength, and reduce fatigability for the FES activity. Although psychological and other physiological benefits have been attributed to FES in paraplegic subjects, long-term home programs require much motivation.

Neurophysiological Schools

Many schools of physical therapy have developed approaches that focus on enhancing the movement of paretic limbs affected by UMN lesions. These approaches may be especially valuable when initiating motor therapies in patients with profound weakness. Techniques include using sensory stimuli and reflexes to facilitate or inhibit muscle tone and single- and whole-limb muscle movements in and out of mass actions called synergies. Most approaches try to sequence therapy in a progression reminiscent of the neurodevelopmental evolution in infants from reflexive to more complex movements. The emphasis is on normal postural alignment before any movement. Some techniques permit mass movement patterns early in treatment; others inhibit spastic overflow synergistic movements. For example, mobility activities may proceed in a developmental pattern from rolling onto the side with arm and leg flexion on the same side, followed by extension of the neck and legs while prone, then lying prone while supported by the elbows, and then doing static and weight-shifting movements while crawling on all four extremities. These mat activities are followed by efforts for sitting, standing, and finally walking. This progression is used most often in children with cerebral palsy, but some therapists also apply it to stroke and TBI rehabilitation.

Different schools vary in their attempts to activate or minimize reflexive movements and how they train the functional movements needed for ordinary activities. These schools have not emphasized strengthening exercises. Their use of reflexive movements, vibration to stimulate a muscle contraction, cutaneous stimulation to facilitate a voluntary contraction, and loading a joint to increase extension is reasonable from a physiological point of view, but any carryover of responses into functional or volitional movement is uncertain.

Neurodevelopmental practitioners believe that the optimal facilitation of movement requires normal postural responses, that abnormal motor behaviors are compensatory, and that the quality of motor experiences and the integration of whole-body somatosensory information will train patients for more normalized movements. The Bobath hands-on approach is a particularly popular neurodevelopmental technique. It aims to facilitate normal movement and desired automatic reactions and to restore postural control while inhibiting abnormal tone and reflex activity using specific motor patterns. Bobath therapists avoid provoking mass flexor synergies from the shoulder, elbow, and wrist or extensor synergies at the knee and ankle. The coordination of patterns of muscle group activity is viewed as more important than the actions of individual muscles. Most practitioners take a problem-solving and task-oriented approach with patients. A typical exercise routine may work on stance and trunk control with a large ball and careful hand placement by the therapist to evoke movements out of synergy (Fig. 48.2). These methods originally were developed for children with cerebral palsy but have been adapted for stroke and other neurological disorders. In comparison with other neurophysiological approaches and task-oriented motor learning for gait, upper extremity function, and overall level of functional independence, use of Bobath techniques has led to equivalent outcomes or, in several small trials, modestly inferior outcomes (van Vliet et al., 2005). Given the underlying theory of Bobath, however, outcome measures ought to examine the quality of task-related movements as much as functional gains.

Task-Oriented Practice

Motor learning emphasizes visual, verbal, and other sensory feedback to achieve task-specific movements, in contrast with neurophysiological techniques, which rely on cutaneous, proprioceptive, and other sensory stimuli to elicit facilitation and inhibition of movement patterns. A key aim of the PT is to put the patient in the best position to be able to practice progressively. Constraint-induced therapy (CIT) for the upper extremity, body weight–supported treadmill training (BWSTT) for walking, mental practice, electromyography (EMG) and other forms of biofeedback (BFB) during functional movements, and training in a virtual reality (VR) environment are among many task-oriented approaches to improve motor control for particular tasks.

Studies of the efficacy of particular schools of therapy have not revealed differences between approaches. These studies, primarily in patients with stroke, used outcome measures that emphasize independence in ADLs and not an outcome directly related to the primary focus of the specific technique of physiotherapy, which is motor performance and patterns of movement. Studies of efficacy should concentrate on the best well-defined practice for an important goal such as reaching, grasping items, and walking that may in theory be modulated by the intervention. Another emphasis must be on the optimal intensity and duration of sessions of training. Most therapists take an eclectic experimentalist’s approach, not unlike what physicians do in daily practice, but this may not be conducive to an optimal outcome for the patient. Indeed, the best evidence to date is that a mixed approach leads to more functional independence than a placebo or no active therapy, but no one approach to a particular set of upper extremity or mobility functions has been shown to be superior to another (Pollack et al., 2008).

Adaptive Equipment

Canes and walkers improve stability through a lever arm that can share the body’s weight between the leg and device, keep the pelvis level during stance on the weak leg, and generate a joint moment to assist the hip abductors and reduce loading on the knee. Devices must be fitted properly. For example, handgrips should be at a height that allows approximately 20 to 30 degrees of elbow flexion. The cane should swing forward with the involved limb and bear the most weight during stance on that leg.

Wheelchairs are of two main types, companion-operated and patient-operated. The latter can be manual or power-assisted. Lightweight and very lightweight patient- or companion-operated wheelchairs must be fitted with at least a dozen characteristics in mind (Box 48.2). Severe spasticity, poor head or trunk control, the amount of upper extremity function, and the type of work and sports engaged in may necessitate additional modifications. Very lightweight wheelchairs tend to be most manageable and durable for the patient with paraplegia. Models with power-assisted wheels have become more affordable and are practical for use by patients with weakness or pain in the upper extremities. Motorized wheelchairs can be maneuvered by joystick switches and chin or sip-and-puff mouth controls. The high cost of custom-designed wheelchairs means that therapists, vendors, patients, and families need to work together to obtain what is most appropriate and cost-effective. Wheelchairs also require maintenance. A wobbly front wheel or poorly aligned main rear wheel adds to the energy cost of mobility and may cause shoulder and wrist injuries. Most rehabilitation centers that manage patients with myelopathies offer a wheelchair clinic and have close links to durable medical equipment suppliers.

Occupational Therapists

Occupational therapists (OTs) facilitate the practical management of disability. The philosophical foundation of OT is that purposeful activity helps prevent and remediate dysfunction and elicits maximum adaptation. Goal-oriented tasks are meant to be culturally meaningful and important to the needs of clients and their families. Activities include daily life and work skills, exercise, recreation, and crafts. Occupational therapy also is concerned with improving the patient’s interaction with the environment and maximizing the patient’s role in society in terms of relationships, occupation, and personal standing. The OT implements a program to enable patients to learn or relearn specific activities, develop new or compensatory skills, adapt their behavior to what is feasible, make adjustments to increase the accessibility of their environment, and perform leisure activities.

A program may include the use of appliances to improve independence, ranging from simple devices (e.g., a thickened grip to better grasp cutlery or a pen) to complex ones (e.g., use of an environmental control unit). Such adaptive aids for daily living are listed in (Box 48.3). Hemicuff and Bobath slings are used to reduce shoulder subluxation and prevent pain in patients with upper limb paralytic disorders (Fig. 48.3). A balanced forearm orthosis can support the upper arm and allow a modest biceps and triceps contraction to swing the arm over a table, which may be especially effective for the patient with a level C5 SCI. For patients with stroke and brain injury, OTs work closely with the neuropsychologist to address visuospatial inattention, memory loss, apraxia, difficulties in problem solving, and the skills needed for return to school or employment. Some OTs manage dysphagia and interpret modified barium swallow (MBS) studies.

Task-oriented and motor learning strategies have gained attention in formal occupational therapy research. Using this approach, the OT presents activities in a way that elicits the retention and transfer of particular skills for use in a functional setting. For example, in one study, limb kinematics improved in normal and hemiparetic subjects when training included purposeful goals with relevant items used daily. Thus, practice in object-related tasks, rather than simple repetition of reaching and grasping of items that have no significance for the client, may provide more concrete sensory information and offer rewards that motivate performance. In many instances, OT strategies evolve from problems that arise in daily living that require a solution. For example, adaptive equipment and an OT educational intervention in stroke patients to remediate the lack of confidence and increase the amount of information patients had available to them reduced the barriers to outdoor mobility and participation in the community (Logan et al., 2004).

Speech and Cognitive Therapists

Speech and language therapists are trained in many aspects of communication and cognition, including phonetics and linguistics, attention and memory, audiology, and developmental psychology, and provide expertise in the investigation and management of dysphagia. These therapists treat primarily patients with dysarthria, aphasia, and cognitive dysfunction that interferes with daily activities.

Interventions to improve the patient’s speech intelligibility, volume, and fluency include exercises of affected oromotor structures. For example, patients may be trained to slow their articulation, use shorter sentences, maximize breath support, extend jaw motion, adjust placement of the tongue, and exaggerate articulatory movements. Communication aids include voice amplification and computer assistive and voice recognition devices. These therapists also provide guidance for persons with swallowing difficulties; assessment may include the MBS study during videofluoroscopy.

Treatment for aphasia generally is based on clinical evaluation of the patient’s cognitive and linguistic assets and deficits. The therapy plan is fine-tuned according to standardized language and neuropsychological test results, knowledge of the cortical and subcortical structures damaged, and the ongoing response to specific therapies. Speech therapists attempt to circumvent, deblock, or help the patient compensate for defective language behaviors. Stimulation-facilitation approaches, listed in (Box 48.4), are commonly employed. Views on the value of speech therapy for aphasia vary. Most randomized controlled trials demonstrate a significant benefit for aspects of expression and comprehension in moderately impaired subjects. The amount of practice is a key variable for enhancing outcomes for any particular approach.

Orthotists and Bracing

Expertise in the manufacture, selection, and application of orthotic devices is another key component of a rehabilitation service. The PT or the OT works with an orthotist to select external devices that modulate directional forces from the body and joints in a controlled manner. Although many orthotic devices are mass-produced, the expertise of a trained orthotist is invaluable in choosing and constructing orthoses and supervising their fitting and adjustment. Orthoses include ankle and ankle-knee braces, finger-wrist and shoulder splints, spinal braces, collars, and corsets. The material most often used in manufacture of these devices is a malleable type of plastic, but light metals may be used when large biomechanical forces have to be managed. The effects of pressure, shear forces, and heat retention with sweating also must be considered during fitting to protect the skin.

With shortened inpatient rehabilitation stays, especially after stroke, ankle-foot orthoses (AFOs) that fit inside a shoe tend to be used early to more quickly assist foot clearance during ambulation in patients with a central or peripheral lesion. Observation of gait usually is enough to determine the need for a trial with an AFO in a patient with hemiparesis or footdrop. Indications include inadequate dorsiflexion for initial heel contact or for toe clearance during early and mid-swing, excessive hip-hiking during swing, medial-lateral subtalar instability during stance, tibial instability during stance, and uncontrolled foot placement caused by sensory loss. An orthosis also may be needed after operative heel cord lengthening. If the knee of the hemiplegic buckles during stance, angling the AFO in slight plantar flexion will extend the knee earlier. Dorsiflexing the AFO by 5 degrees can decrease knee hyperextension and help prevent the snapping back that causes instability and pain in midstance. Ankle inversion may necessitate greater rigidity and longer anterior foot trim. The AFO worn in a shoe ought to improve weight-bearing on the affected leg, increase single-limb stance time, and perhaps lessen postural sway. This may improve safety, especially on uneven surfaces, and walking velocity. Fig. 48.4 shows a thermoplastic AFO that fits in a shoe to limit plantar flexion and rotation and help control the knee. Orthoses may be static, such as a rest splint worn at night, or dynamic, with joints that may be lockable or free moving. Fig. 48.5 shows a thermoplastic AFO with a hinged joint that allows flexibility on rising to stand and at heel strike to start the stance phase of the gait cycle. Toe clawing can be managed with a metatarsal pad that spreads the toes. Some patients who have footdrop from a neuropathy such as Charcot-Marie-Tooth disease or diabetes may find that a fashionable boot with a flat heel that fits snugly above the ankle can improve gait by lessening its steppage quality, yet allow toe clearance. An orthotist can assess the potential for shoe modifications and inserts to improve balance and pressure points. Multiple small trials in patients with hemiplegic stroke show that walking activity increases and impairments in walking and balance decrease with the use of an AFO (Tyson and Kent, 2009). Functional electrical stimulation of the peroneal nerve timed to the gait cycle has equivalent or somewhat better effects on walking speed.

A metal double-upright brace offers greater rigidity for mediolateral foot instability and allows more versatility in adjustments for the amount of plantar and dorsiflexion but can be expensive, heavy, and cosmetically unappealing to the hemiplegic patient. Metal bracing systems are used more often by selected subjects with paraplegia from spinal injuries and in those with polio. Lightweight plastic knee-ankle-foot orthoses (KAFOs) with locking metal knee joints also can assist patients with severe polyneuropathy, muscular dystrophy, meningomyelocele, or SCI. Reciprocal gait orthoses (RGOs) with wire cables that link flexion of one hip to extension of the opposite hip are available for paraplegics. Short-distance ambulation for exercise also can be aided in the patient with SCI by variations on a KAFO or other devices. Bracing in persons with SCI has been combined with functional electrical stimulation to improve stepping and decrease its energy cost.

The thoracolumbar support shown in Fig. 48.6 was molded to the patient to lessen neck motion after a high thoracic vertebral injury and surgical stabilization. Static orthoses allow no motion of the primary joint. Solid wrist-hand orthoses usually are set between neutral and 30 degrees of extension. Upper limb orthoses, however, have not been shown to improve arm or hand function, increase range of motion or lessen the incidence of pain, based on small trials after stroke. Dynamic orthoses use elastic, wire, or powered levers that compensate for weakness or an imbalance in strength and allow some controlled movement. Fig. 48.7 shows the paretic left hand of a patient with a level C6 SCI holding a playing card with the aid of a thumb opposition splint. The weaker right hand needs wrist extension to mechanically oppose the thumb to the second and third digits. Such custom-made devices can be produced with lightweight metals and plastics. Functional neuromuscular electrical stimulation electrodes can be embedded in a wrist splint to enable grasp and release or over the peroneal nerve to trigger ankle dorsiflexion at appropriate times during the stepping cycle.

Measurement Tools

Outcome measurement is essential to demonstrate the effectiveness of an intervention for an individual and for clinical trials that aim to develop better evidence-based practices. Many of these measurement tools can also be employed within usual clinical care to monitor for declines. The complete discussion of this topic is available in the online version of this book at www.expertconsult.com.

Measures

Neurorehabilitation employs measurement tools to characterize the impairments, disabilities, activities, and psychosocial needs of patients, as well as to monitor interventions, assess outcomes, and document services. To provide a comprehensive assessment of the impact of disease, outcomes may be considered at four levels: pathophysiology and related impairments, disability and level of functional activity, handicap and ability to participate, and health-related quality of life. The design and applications of these scales across diseases are described in Table 48.1.

What to measure depends on the clinical setting and must be based on the anticipated effect of the intervention. Quantitative clinical end points such as death and presence or absence of paralysis are important in evaluating early neuroprotective interventions in acute SCI, TBI, and stroke but tell little about disability and quality of life in survivors.

Each outcome level addresses distinct aspects of the disease process. The relationships among them are complex. The first two levels entail physician-oriented outcomes, whereas those at the third and fourth levels are more accurately called patient-based outcomes, although only the fourth incorporates the patient’s perspective. Rehabilitation is especially interested in measures of the impact of disease, which is contained within the World Health Organization (WHO) International Classification of Impairments, Disabilities, and Handicaps. As shown in Table 48.1, the focus is on the consequences of the disease or health condition rather than the disease alone (i.e., a classification of disablements and functioning). A classification available on the WHO website (www.who.int/icidh/) emphasizes whether people actually perform the tasks, especially those that require a lot of time and energy. The dimensions of the WHO classification also include personal and environmental factors that affect functioning. The emphasis on activities, participation, and contextual factors interacting with impairments and disabilities may influence the design of new measurement tools.

Two types of measures exist for each of these domains: measures that are specific to a given disease and generic ones that may, with caution, be applied across a range of diseases (see Table 48.1). The former may be more sensitive in detecting changes particular to the effects of an individual disease, whereas the latter may allow comparisons across different diseases and serve as an off-the-shelf tool. Two generic measures of disability are the 10-item Barthel Index (BI), which can be totaled to a maximum score of 20 or 100 (Table 48.2), and the 18-item Functional Independence Measure (FIM) (Box 48.5), which includes a modestly responsive cognitive component for a total of 126 points (Box 48.6). The FIM is used more often in the United States, especially for studies with large numbers of subjects, than in Europe or Asia, where its applicability for clinical trials has been questioned. The FIM, the BI, and two other scales—the Glasgow Outcome Scale for Recovery of Consciousness and Function and the Rankin Disability Scale (Tables 48.3 and 48.4)—commonly are used in clinical trials. The latter correlate closely with one other, in part because their definitions at the levels of 3 and 4 not well defined in terms of residual impairments and specific disabilities. In many respects, the BI and FIM reflect the level of care needed by patients. These and most scales have floor or ceiling effects and variable sensitivity to change, especially if used during outpatient care.

Table 48.2 Barthel Index

  Score
Activity Help Independent
1.Feeding (if food has to be cut up = help) 5 10
2.Moving from wheelchair to bed and return (includes sitting up in bed) 5-10 15
3.Personal toilet (wash face, comb hair, shave, clean teeth) 0 5
4.Getting on and off toilet (handling clothes, wipe, flush) 5 10
5.Bathing self 0 5
6.Walking on level surface (or if unable to walk, propel wheelchair*) 10 (0*) 15 (5*)
7.Ascend and descend stairs 5 10
8.Dressing (include tying shoes, fastening fasteners) 5 10
9.Controlling bowels 5 10
10.Controlling bladder 5 10

* Score only if patient is unable to walk.

Box 48.6 Description of Functional Independence Measure Levels of Function

Table 48.3 Glasgow Outcome Scale for Recovery of Consciousness and Function

Score Outcome
1 Death
2 Vegetative: unresponsive and speechless
3 Severe disability: dependent on others for all or part of care or supervision because of mental or physical disability
4 Moderate disability: disabled but independent in activities of daily living and community
5 Good recovery: resumption of normal life; may have minor neurological or psychological deficits

Table 48.4 Rankin Disability Scale

Score Disability
1 No disability
2 Slight disability: unable to carry out some previous activities but looks after own affairs without assistance
3 Moderate disability: needs some help but walks without assistance
4 Moderately severe disability: unable to walk and do bodily care without help
5 Severe disability: bedridden, incontinent; constant nursing care needed

How to Measure

Irrespective of what is to be measured, two essential issues must be addressed if the data are to be clinically meaningful. The outcome measure must be (1) clinically useful and (2) scientifically sound.

Clinical Usefulness

Factors that determine whether an instrument can be successfully incorporated into clinical practice or, for a clinical trial, is relevant for the study sample include the method of administration, appropriateness, user acceptability, and score distributions. The most widely used methods of administration are patient completion of a paper-and-pencil tool (as with the Medical Outcome Study 36 Item Short Form [SF-36] Survey), interview rating (as with standard neuropsychological testing), behavioral observation (as with the BI), examination rating (as with the Ashworth Scale [described later on]), and rating by team consensus by trained observers (as with the FIM). The method of administration often has important implications for the design of clinical research studies and their cost.

Appropriateness of a measurement tool takes into account the anticipated effect of the intervention and the particular patient population, including factors such as disease type, duration, and severity. During neurorehabilitation training, an intervention may result in a change in disability and health-related quality of life and also may have an impact on mood, emotional well-being, coping skills, and self-efficacy. Each of these outcomes may have to be measured separately, because they may change independently of each other; that is, a rehabilitation intervention may have a different effect on disability than on general well-being, or may affect mobility but not disabilities related to functional use of the upper extremities.

User acceptability is equally important. If an instrument is to be clinically useful and acceptable so it can be incorporated into daily practices, it must be brief, user-friendly, practical to administer, and cost-effective. Unwieldy, time-consuming, or resource-heavy instruments have limited utility, especially for clinical trials.

Finally, if an instrument is to be a useful measure of an attribute such as disability, it must adequately represent the true distribution of that attribute in the study, must discriminate between individuals in the study sample, and must detect change in the attribute being measured. These issues can be addressed by examining the distribution of scores in the sample of interest, including the range, mean score, standard deviation and standard error, and floor and ceiling effects. Many tools have been evaluated for their possible meaningfulness by calculating the minimal detectable change (MDC) or the minimal clinically important difference (MCID). The MDC is the smallest amount of change from baseline that likely reflects a reliable change after an intervention, rather than an error of measurement inherent in the score. Thus, it is the safest threshold for showing statistically detectable individual changes. It is derived from the z score, standard error of measurement (SEM), and standard deviation. Some controversy accompanies how best to calculate and interpret the MCID. For a within-person change, the MCID value should be greater than the MDC to indicate that the measure is precise enough to indicate a meaningful clinical change. Some authors have used coefficients of the SEM, ranging from 1 to 2.77. Even if the magnitude of change on a scale were a clinically meaningful difference, however, the scale may still not inform the investigator about what the gain means to patients.

Scientific Soundness

Clinical usefulness does not guarantee scientific soundness in terms of rigorous measurement. The outcome measure must contain three essential and intimately related scientific properties: reliability, validity, and responsiveness. A reliable measure produces results that are accurate, consistent, stable over time, and reproducible. The purpose of reliability testing is to determine the extent to which random measurement error is present. Random error includes all chance factors of produced variations on repeated measurement. The potential sources of random error include the design of the measure itself, the person doing the measuring, and the person being measured. Measures have four types of reliability: internal consistency, test-retest interrater and intrarater reliability, and parallel test forms. Reliability also is necessary for an instrument to be valid.

Validity concerns the relationship between the concept being measured and the instrument being used to assess that concept. It is broadly defined as the extent to which the instrument measures the concept it purports or is intended to measure. Three types of evidence support validity: content-related, criterion-related, and construct-related.

Responsiveness is the ability of an instrument to measure clinically important change and change over time. It is essential in evaluating the relative benefits of differing interventions. Sensitivity to change is particularly important when treatments are associated with small but significant differences that may be undetected by insensitive measures. Responsiveness often is reported in the form of an effect size, which most commonly is defined as the mean change score divided by the standard deviation of the baseline score.

Measures for Clinical Trials

Rehabilitation interventions have progressed considerably in conforming to scientifically designed clinical trials. Although the efficacy of general rehabilitation has been the subject of trials in patients with stroke, MS, and other diseases, trials are especially needed to assess a particular intervention for subjects with a particular set of disabilities that are the target of the well-defined intervention. Physical interventions are far more difficult to provide in a reproducible fashion than drug therapies. Outcome measures also are less easy to develop and apply when the primary outcome is levels of disability or severity of a specific disability, rather than outcomes such as mortality, morbidity, survival time, or subjects who survive without disability, which are typical measures of drug and surgical studies. Recent multicenter randomized controlled trials (MRCTs) of locomotor training after stroke (Duncan et al., 2007; Hidler et al., 2009) and SCI (Dobkin et al., 2006a), of constraint-induced movement therapy (Wolf et al., 2006), and of cortical electrical stimulation (Harvey et al., 2009) have demonstrated the feasibility of developing complex treatment strategies provided by many therapists at many cooperating sites. Ordinal scales like the BI dominate the outcome measures for most trials, but continuous or ratio measurement scales such as walking speed or time to complete tasks were the primary outcomes for these MRCTs. The relationship between walking faster over ground after an intervention and how well you can participate in home and community as a result of a mobility intervention is moot. Dichotomous outcome measures (Duncan et al., 2007) that divide subjects into those who achieve walking speeds compatible above or below a level that correlates with levels of home and community ability to ambulate is closer to an ideal measure of participation. Future trials related to mobility and arm use, however, may draw upon tiny triaxial accelerometers, gyroscopes, and global positioning satellite or cell phone devices. Using mathematical algorithms, these and other inexpensive wireless sensors may allow clinicians and trialists to directly monitor the quality and amount of activity and participation.

Organization of Services

Service Provision

The way in which services are organized depends to some extent on the disease being managed. The most fundamental differentiation is between acute-onset diseases with different pathophysiological characteristics such as brain and spinal cord trauma and stroke; chronic conditions such as cerebral palsy and polio, in which disability may increase with aging and overuse of muscles and joints; and progressive disorders such as MS, muscular dystrophies, ALS, and PD. Service provision also should be driven by the philosophy underlying rehabilitation: to return the patient home and optimize home and community activities as soon as possible. The speed with which this is done depends as much on the services available in the community and the capacity of the family and caregivers to look after the patient as it does on the severity of the disability. Some tension arises between inpatient services in which all of the necessary ingredients are gathered under one roof, which appeals to patients and families, and community services, which are less centralized but support patients in their own environment. Most important is the smooth interface between these two settings when patients are discharged from an inpatient service.

Inpatient Rehabilitation Unit

The most efficient rehabilitation setting is an inpatient unit designed and staffed specifically for this purpose. In the United States, approximately 1200 inpatient sites are covered under Medicare. The benefits of dedicated units for stroke management have been convincingly demonstrated. Patients managed in stroke units are significantly less likely to die than those cared for on ordinary wards. Death, institutionalization, and dependency all are significantly less common in stroke unit patients, in part because of a reduction in secondary complications of stroke and a milieu dedicated to managing disability. Early functionally oriented therapies by an organized team tend to get patients independent enough to return home sooner than sporadic therapy that does not articulate specific attainable goals for mobility, self-care, and family training. These benefits persist for up to 5 years after discharge and improve quality of life. With inpatient stroke rehabilitation in the United States usually starting within 10 days of onset, and with the average length of stay less than 24 days, planning is imperative.

Several studies of TBI suggest the benefit of coordinated care starting during the inpatient stay. Investigators randomly assigned 120 active-duty military personnel who suffered moderate TBI to a standardized inpatient milieu–based cognitive rehabilitation program or to a home program with weekly phone support from a nurse and mental and physical exercises carried out on their own for about 30 minutes a day (Salazar et al., 2000). The inpatient program used both didactic cognitive and functional experiential approaches. Treatment lasted 8 weeks. Outcomes did not differ 1 year later in standard cognitive tests, social adaptation, mood, behavior, or fitness for military duty. More than 90% returned to work. Aggression increased in both groups, suggesting the need for ongoing support.

The role of inpatient rehabilitation units in managing progressive conditions such as MS and PD is much less well defined, although some evidence suggests a benefit in progressive MS. Inpatient care also may return patients home at a higher level of functioning after implantation of a deep brain stimulator for PD or after an exacerbation of walking disability caused by a hip fracture in a patient with a chronic hemiparesis.

Community-Based Services

Although community-based rehabilitation appears to have a number of advantages from the perspective of the disabled patient, few studies have addressed its efficacy (Legg, 2004). This lack may relate at least in part to the methodological difficulties in defining the training team’s level of expertise, the patient’s level of disability, the amount of caregiver support, and the frequency, duration, and type of therapy. One large randomized stroke trial compared rehabilitation at home after an average 12-day inpatient stay with another week of inpatient care followed by hospital-based outpatient treatment. Patients who lived alone either were independent in transfers when they left the hospital or needed to be assisted by a caregiver. Similar outcomes 12 months after a stroke were achieved at lower cost because of less use of hospital beds by the early discharge group. An intention-to-treat randomized trial with 250 subjects showed that rehabilitation in an inpatient unit after a brief stay in an acute stroke unit or general medical ward produced better outcomes in moderate to severely disabled patients (BI score less than 50) compared with rehabilitation treatment in the community. No differences in quality of life were found, and levels of activity outside the home were not measured. Smaller trials confirm similar positive outcomes at 3 to 6 months for home to those for various forms of outpatient care, with the home group having fewer in-hospital days and greater gains in instrumental ADLs.

A day treatment program for approximately 100 patients who were unable to work for 1 to 2 years after TBI provided a range of interventions during group therapy for a mean of 190 days. The investigators found a significant reduction in physical disability, increased self-awareness and emotional self-regulation, and more effective participation in interpersonal activities. At 1 year after completion, 72% lived independently and 57% were employed.

A brief stint of outpatient therapy for a well-defined goal, such as to improve transfers in a patient who declines from MS or to make walking safe again after an illness that causes deconditioning in a patient with chronic hemiparesis, is an invaluable means for patients to maintain their highest level of independence and avoid placement in a nursing home. A few therapy sessions a week for a few weeks plus a home program may accomplish this when provided for a task-specific goal that relieves caregivers of new burdens. For example, a bout of therapy for aerobic conditioning and muscle strengthening lessened disability in patients with the chronic effects of a prior stroke. A randomized trial assigned 110 patients with mostly chronic TBI to an outreach program in the community or to provision of written material about resources for patients and families (Powell et al., 2002). The outreach group met about twice a week for a mean of 27 weeks. The outreach group made modest but significant gains in scores on the BI and a brain injury outcome measure 2 years after the start of this late intervention.

The most efficient structure for rehabilitation services in many countries includes a two-tiered system of service delivery. Regional specialists can manage complex and profound disabilities, and local community services provide neurological disability teams from local hospitals or the community. Regional neurological rehabilitation centers provide experts familiar with the management of complex and severe disabilities; serve as a focus for education, training, and research; and should be linked to university teaching centers. Such centers can offer care for patients with high-level SCIs and severe TBI, manage specialized orthopedic and plastic surgery procedures, and provide wheelchair and special seating needs, custom orthotics and prosthetics, rehabilitation engineering, functional electrical stimulation devices for walking and upper extremity movement, neuroprostheses, communication aids and environmental controllers for quadriparetic patients, and driving assessments.

Most disability services can be provided locally, which is likely to be both cost-efficient and cost-effective. Services provided by a multidisciplinary neurological disability team based in a local hospital can develop community outreach into clinical centers and homes. Resources often are not adequate for this approach, however. A community-based rehabilitation model developed by the WHO has suggested how communities can develop their own support mechanisms for disabled people, often with locally trained workers supervised by qualified staff. Telerehabilitation services may prove especially useful in supporting patients who live far from available services or are too disabled to leave the home.

Biological Bases for Rehabilitative Interventions

The potential to enhance neurological recovery by manipulating the biological adaptability of the brain, spinal cord, and peripheral nerves has become remarkably relevant to clinical practice (Dobkin, 2003). Basic neuroscience studies suggest that physical and cognitive training, extrinsic stimulation, and pharmacological interventions, along with natural biological reactions to injury, could inhibit or enhance the restoration of motor and cognitive functions. Box 48.7 lists some of the potential mechanisms that contribute to changes in impairments and disabilities. These are not discrete mechanisms. They overlap, and many depend on each other over periods of variable duration. In addition, biological therapeutic approaches such as use of tissue implants may enhance these mechanisms when applied to human studies of recovery in the near future. These potential mechanisms, drawn mostly from in vitro and in vivo experiments with invertebrates and vertebrates, are the subject of much ongoing investigation (Carmichael, 2008; Blesch and Tuszynski, 2008). Although care must be taken in extrapolating from animal studies of recovery to their implications for human interventions, at least a few of these potential mechanisms suggest strategies that the rehabilitation team can use to improve outcomes.

Box 48.7

Mechanisms That May Support Recovery of Function

Adapted with permission from Dobkin, B., 2003. The Clinical Science of Neurologic Rehabilitation. Oxford University Press, New York.

Recovery of Neuronal Excitability

Reversal of toxic-metabolic insults to neurons, axons, and glia plays an early role in recovery of function. Recovery of impairments may be delayed until intracellular and extracellular edema, acidosis, and ion fluxes resolve and until protein synthesis restarts, the mass and toxic effects of blood from a hemorrhage lessen, and other intracellular and membrane functions return to normal. Some of the gains in the first days to few weeks after TBI, SCI, and stroke seem related to these mechanisms. Acute medical interventions to spare the penumbra of hypometabolic tissue on the edge of an infarct have been a mainstay of approaches to neuroprotection in stroke. The extent of the recovery of cortical penumbral and peri-infarct tissue has been modestly correlated with recovery of function in patients after stroke. The surviving penumbra may be a locus for long-term potentiation (LTP), representational plasticity and synaptogenesis, and activity-driven structural reorganization, as well as for the promotion of axonal regeneration and the attraction of new neural cells from the subventricular zone (Cramer, 2008). Infarcts, diffuse axonal injury after TBI, and much of the damage after SCI occur within white matter tracts, so concepts of penumbral sparing may not apply.

Remote effects of an injury, caused by a lack of transsynaptic activity along a neural pathway after one of its links has been damaged or due to the loss of modulation by noradrenergic, serotonergic, dopaminergic, or cholinergic neurotransmission, also could transiently limit recovery. This transsynaptic functional deactivation, called diaschisis, has been demonstrated in autoradiography and positron emission tomography (PET) studies as neurovascular uncoupling and aberrant neurotransmission. After a thalamic infarction, for example, hypometabolism of frontoparietal cortex that had received input from thalamic projections often is found. Hypometabolism of the dorsolateral frontal cortex after a stroke in the caudate nucleus and anterior limb of the internal capsule is another example of diminished activity in spared tissue. Although resolution of remote functional deactivation has been suggested as a mechanism of motor recovery in animal studies, human studies with functional imaging have not clearly revealed a relationship between the degree of impairment of recovery and regions of anatomically remote hypometabolism.

Activity in Partially Spared Pathways

Partial sparing of neuronal clusters, as in periinfarct tissue or in tracts such as the posterior limb of the internal capsule, provides a substrate to lessen impairment and disability over days to months after injury. Experimental studies reveal that as little as 20% sparing of one corticospinal tract can provide the minimal residual structure necessary for fair upper or lower extremity motor control. For example, symptoms of PD evolve after approximately 75% of dopamine-containing neurons have died, and locomotor function decompensates when 95% are lost. In a classic human study, Bucy relieved hemiballismus by incising a portion of the cerebral peduncle. Within 24 hours, the patient’s flaccid hemiplegia began to improve. By 7 months, the patient plateaued with a very mild hemiparesis, independent gait, and the ability to hop on the contralateral leg. At autopsy, only 17% of the axons in the medullary pyramid persisted. An estimated 90% of precentral giant cells of Betz suffered retrograde degeneration.

Partial wallerian degeneration of the corticospinal tract sometimes can be seen on MRI scans after stroke (Fig. 48.8). On computed tomography (CT) images, sparing of more than 60% of the cerebral peduncle, including the medial portion, predicts the recovery of a precision grip and, to a lesser degree, the force of the grip. Greater sparing of the primary motor cortex (M1) and less extensive wallerian degeneration are associated with better hand function in studies that involved functional neuroimaging. The typical hemiplegic posture of elbow, wrist, and finger flexion followed 60% shrinkage, which corresponds to a loss of roughly 88% of the descending fibers. Better motor and functional outcomes also have been associated with sparing of metabolic activity in the ipsilesional primary motor cortex, the thalamic-and-basal ganglia–frontal network, and with sparing of the basal ganglia ipsilateral to the stroke. Approximately 20% sparing of the ventral or ventrolateral funiculi after SCI permits fair motor function, and 35% sparing of the dorsal funiculus allows appreciation of dorsal column sensory inputs.

The recovery of vision after an occipital stroke may depend on spared parts of the vision-associated areas of the cortex. Patients who recover from a hemianopsia tend to activate bilateral extrastriatal cortex during hemifield stimulation to the affected occipital lobe. Involvement of primary and extrastriatal cortex leaves a persistent field loss.

Residual neurons and axons of an injured tract may need rehabilitation to help drive training-induced, activity-dependent plasticity. When the number of corticospinal fibers that synapse with spinal motor pools is too small to generate adequate excitatory postsynaptic potentials, a descending volley will not excite the spinal neurons. Recovery of an adequate excitatory postsynaptic potential, however, may follow practice that strengthens synaptic efficacy. Other contributors to spared pathway functioning include changes in ion channels along partially demyelinated axons, increased strength of output from undamaged cortical and brainstem neurons that also descend onto spinal neurons such as propriospinal pathways, dendritic sprouting, and up-regulation of the number of motor neuron and interneuron excitatory receptors. In addition, the ipsilateral motor cortex, via the uncrossed ventral corticospinal tract, often has been invoked as a pathway that may provide some of this input, thereby compensating for a contralateral cerebral injury. The fibers of the ventral tract synapse especially with motor neurons for axial and limb girdle muscles. Some projections of the lateral corticospinal tract cross and then recross under the central canal of the cord. Thus, deafferentation of one side of the ventral horn from loss of descending fibers may induce further sprouting of recrossing fibers to enhance motor control of the affected limb (Rosenzweig et al., 2009). Functional neuroimaging studies of the upper extremity, leg, and language suggest that better gains with training are associated with greater cortical activation near the primary representation than when homologous regional activity of the intact hemisphere is most apparent (Baron et al., 2004).

Alternative Behavioral Strategies

Many functional activities can be accomplished by compensatory behavioral adaptations that allow greater independence in ADLs despite persistent impairments. Motor functions can appear to have fully recovered when in fact residual neural activity is supporting behavioral plasticity (Levin et al., 2009). For example, after a unilateral pyramidal lesion in the rodent or monkey, reaching for a pellet of food gradually improves and at first glance may appear to have recovered fully. A closer analysis of the movement reveals better control of the proximal than of the distal portion of the affected limb. The animal reaches with a grasp, brings the pellet to its mouth without the normal supination of the hand and forearm, turns its head to chase after the food, and cannot easily release its grip. The hand-to-mouth movement pattern of the hemiparetic patient is similar to the lesioned animal’s strategy. Functional ambulation often improves in part by biomechanical adaptations that are revealed by kinematic deviations in the gait pattern and by use of braces and assistive devices. Behavioral adaptations, of course, also have a neural substrate that includes training-induced cortical representational reorganization and increased efficiency in the incorporation of residual pathways. Cognitive impairments that affect learning and problem solving, however, may limit the use of compensatory strategies. Rehabilitation efforts often train patients to use compensatory behavioral strategies to accomplish a task. Rather than performing tasks using the original neural network, alternative strategies may be emphasized. Substitution by a different neural network, however, may interfere with the activity-dependent reorganization that is needed to achieve further recovery from a specific impairment or disability.

Distributed Networks of Neuronal Circuitry

A distributed system is a collection of processing units (i.e., dynamic neuronal assemblies with similar functional properties and anatomical connections) that are spatially separate and exchange messages. Hierarchical, parallel, and quasi-serial linking operations are made with the afferent and efferent systems of the brain (Box 48.8). The nodes in such systems cooperate to manage the diverse information necessary for the rapid and highly flexible control of, for example, cognition and multijoint movements. These circuits lend themselves to an understanding of the neuroplasticity that may contribute to spontaneous and training-induced recovery of function.

The executive motor system has been the subject of recovery-related studies. The primate primary motor cortex (M1) has separate clusters of output neurons that can facilitate the activity of a single spinal motoneuron. Also, a single cortical motor neuron can project to the spinal motoneurons for several muscles, even those that may act across a joint. In addition, neurons in overlapping M1 territories are activated to produce movements of different body parts. This overlapping organization contributes to the control of the complex muscle synergies for voluntary movement of the arm and leg within the ordinary workspace of the body (Graziano et al., 2002). Thus, complex movement representations within cortical and subcortical maps may be reexpressed after injury as usual patterns of movements. It seems likely, then, that retraining ought to engage patients in the practice of complex movements rather than single joint actions or exercise of single muscles.

Multiple representations also have been demonstrated in nonprimary motor regions. They are found in the premotor cortex of Brodmann area (BA) 6, the supplementary motor area, and the region immediately rostral to it, and in BA 23 and BA 24 by the cingulate sulcus. These motor regions have interrelated and some overlapping functions, with direct and indirect anatomical connections. Each area has an independent set of inputs from adjacent and remote regions, and most have parallel but separate outputs to the brainstem and spinal cord. The effective unit of operation in such a distributed system is not a particular neuron and its axon but a group of cells with similar functional properties and anatomical connections.

The capacity of motor-related cortices to redistribute their function is apparent from PET and functional magnetic resonance imaging (fMRI) studies that have looked at subjects with recovered hand function after a stroke. For example, after a striatocapsular infarction in patients who recovered finger tapping and hand squeezing, bilateral rather than just contralateral activation of cortical motor neurons was found, along with greater involvement of motor areas that ordinarily would not be visibly activated for that task. Other regions related to selective attention and intention also show increases in blood flow and metabolism, which suggests that they must play a larger role when the substrates for a movement reorganize. Regions that are activated in normal subjects when greater force is exerted, such as insular, premotor, and anterior opercular cortex, may also be activated during movement of a hemiparetic limb (Baron et al., 2004), in parallel with the increased effort needed to move that some patients describe.

Parallel segregated circuits process different variables for movement throughout their integrated pathways in the striatum, thalamus, cerebellum, brainstem, and spinal cord. The cerebellum receives and modulates locomotor cycle–related signals. The neocerebellum monitors the outcome of every movement and optimizes movements using proprioceptive feedback. In view of the great computational interest the cerebellum has in the details of afferent information from joints and muscles, rehabilitation therapies for walking and upper extremity actions should aim to provide this key motor center with the sensory feedback that the spinal cord and cerebellum recognize as typical of normal walking and reaching-related inputs. The motor functions that the cerebellar inputs and outputs attend to, such as timing and error correction for accuracy as the hand approaches an object, are especially important for patients to practice when a lesion undermines motor control. Thus, individual channels can control separate functional units of motor cortex, and in turn each is independent in its control of subcortical motor nuclei. After an injury, the balance of activities of these networks is reset. Although these systems are not likely to be highly redundant, they may provide a partially reiterative capacity for some sparing or compensation after a sensorimotor network injury. The intact parallel systems from cortical and subcortical areas may partially compensate or substitute for nonfunctional ones when their activity is enhanced by specific cognitive and motor rehabilitative retraining. For example, if the premotor cortex is damaged, compromising visually cued movements, the patient may be taught to use an internally cued strategy mediated especially by the spared supplementary motor area. Patients with a paretic hand, then, can be encouraged to pre-shape the hand before they reach for an object, to lessen the need for automatic visuomotor input from the premotor hand region.

The brainstem and spinal cord also include important subcomponents of distributed motor functions. For example, they contain their own intrinsic networks for aspects of locomotion. Although cortical and peripheral sensory input is essential for normal locomotion under disparate environmental conditions, the timing of synergist and antagonist muscle activity for stepping appears to be primarily the task of a self-oscillating lumbar interneuronal network. Even an isolated section of lumbar spinal cord in mammals can produce cyclical outputs that rhythmically flex and extend a joint. This locomotor circuit is called a central pattern generator and especially depends on glutamate and glycine for alternating excitation and inhibition.

Both distributed and hierarchical organization and plasticity are just as evident for higher cortical functions. Cognitive domains appear to be mapped at multiple sites that are highly connected with feedback connections. PET and fMRI studies have begun to reveal changes in the organization of these interactions for particular tasks after a cerebral injury. Cognitive rehabilitation strategies have been developed based on the notion that recovery may be mediated by tapping into the localizable and distributed grids of connectivity that are intact.

Cortical and Subcortical Representational Adaptations

Many lines of animal research suggest that partial restoration after a central or peripheral injury can result from a functional shift to neighboring neurons. In the adult as well as the developing animal, and in humans, the sensory and motor representations of the brain are distributed, dynamic, and capable of much physiological and structural reorganization. As noted earlier, the neurons of the ischemic penumbra of an acute cerebral infarction may play an important role in functional gains. Very early overuse of these neurons was shown to cause greater injury in rat models of stroke and trauma. Such profound overuse shortly after an injury is unlikely in human subjects, however. Indeed, early practice appears important. A primate model of a less than 1-mm injury to the hand area of M1 found that perilesional representations for the digits decreased when the monkey failed to practice using the hand to scoop pellets out of a narrow well. Neighboring representations for the digits, wrist, and forearm increased with practice. Thus, cortical representational changes are especially likely to arise during training paradigms that involve learning and the acquisition of specific skills, although the relationship between the regional cortical change and the behavioral gain may not coincide, because the focal cerebral adaptation may be one of many. This plasticity probably arises from the unmasking of previously silent synapses and increases in synaptic efficacy in thalamocortical and intracortical circuits. Over weeks and months, it might arise from sprouting of dendrites over short distances. In animal studies of recovery of forelimb function, only 12 hours of a specific therapy for the affected limb leads to behavioral gains and cortical representational changes. In human subjects after stroke, representational plasticity for the ankle movers after locomotor training or the wrist and fingers after upper extremity task-related training can be shown after as little as 12 to 18 hours of practice.

The combination of mutable neuronal assemblies that represent movements and sensation and multiple representational maps in a parallel distributed system offers a sound basis for developing rehabilitation interventions. Behaviorally relevant tasks that are shaped by an optimal schedule of practice and feedback to neural networks may potentially increase functional gains. The optimal duration and intensity of training are uncertain for human rehabilitation strategies, but greater intensity of practice seems to enhance subsequent performance (Kwakkel et al., 2004). Unfortunately, most patients get only a few months of formal inpatient and outpatient retraining of modest intensity that is spread across many tasks.

Biological Interventions

The molecular processes that are induced by injury and by activity in neurons, axons, and dendrites are under intense investigation. Morphological changes such as axonal regeneration over short distances, dendritic arborization, and synaptogenesis have been observed after a brain and cord injuries. When inputs from one pathway to the dendritic tree of a neuron are lost, intact axons can sprout and form synapses on denervated receptors. The net effect of this change in the weight of inputs could have a positive or a detrimental effect on neural function. Can such changes be manipulated? Basic neuroscience studies of biological approaches offer some exciting approaches to complement neurorehabilitation strategies.

Sprouting and regeneration of injured axons in the central nervous system has been demonstrated over very short distances, perhaps no more than 1 cm in rodent models. Growth depends in part on production of regeneration-associated genes, receptors on the surface of cells and the axon growth cone, and pro- and antigrowth signaling molecules in the extracellular matrix. Tissue culture and vertebrate models have shown that substances in the extracellular matrix and on the surface of oligodendrocytes inhibit elongation of the axonal growth cone. Antibodies and receptor blockers have been made to these substances, including the inhibitor Nogo-A, the receptor Nogo-66, Rho, and other membrane and intracellular steps that affect cytoskeletal proteins. Blockers lead to some growth of axons in the rat and mouse after a partial spinal or a brain injury, but no one blocker is enough to drive robust regeneration in these complex signaling systems. When they are combined with a growth factor such as neurotrophin-3 or brain-derived neurotrophic factor, axonal growth may increase further, although this is more evident in serotonergic and adrenergic axons than in corticospinal fibers. In nonhuman primates after stroke, focal TBI, and SCI, inosine has led to greater sprouting in the cord and brainstem by driving growth-promoting gene activity, and chondroitinase injected locally has reduced milieu inhibitors to enable sprouting and regeneration within the injection site. Clinical trials have been proposed for these and similar techniques. Neurotrophic factors have been used to protect neurons from apoptosis, as well to signal neuronal machinery to make the intracellular scaffolds necessary for neurite outgrowth.

Cultured Schwann cells, transplanted olfactory ensheathing cells, neural progenitors, embryonic tissue implants, neurotrophic factors delivered by engineered fibroblasts, and peripheral nerve bridge implants, with a combination of other biological manipulations, also have met with some success in permitting several populations of axons to regenerate across or around a rodent SCI. Bone marrow stromal cells have been injected into the blood after experimental stroke and seem to produce a trophic factor or other support for enhancing plasticity. Cortical and subcortical transplants of stem cells, engineered cells, human embryonal cells that take on the characteristics of local neurons, and progenitor cells for neurons and glia also have been introduced into the brain to promote recovery in animal models. Physical activity alone can regulate the expression of neurotrophic factors in the cortex and spinal cord and induce endogenous neurogenesis, especially in the hippocampus and spinal cord.

Published cell transplant experiments in humans include numerous interventions for PD; most have been unsuccessful overall, but recent trials are more promising and aim to eliminate induced movement disorders. Studies in animal models already have led to several safety studies of human interventions for stroke. These include intravenous injection of autologous mesenchymal stem cells about 40 days after a hemispheric stroke and implantation of human neuronal cells (LBS neurons) into the edge of deep infarcts near the basal ganglia. Further studies of bone marrow–derived stem cells are anticipated. Safety studies in SCI have proceeded with injection of human fetal spinal cord tissue into a syrinx, autologous activated macrophage injections into the acute injury, and injections of porcine oligodendrocyte progenitors into chronic injuries, as well as recent dural or intrathecal infusions of Rho and Nogo inhibitors. No serious complications have been described to date, but reports are often vague. Recent uncontrolled but prospective safety trials from Australia and Portugal point to the safety of autologous olfactory ensheathing glia implanted in the spinal cord. At conferences and on websites, but not in peer-reviewed published reports, clinicians at hospitals in China (www.stemcellschina.com) and at least 100 sites around the world have built stem-cell spas to treat a broad range of neurological and other diseases. They claim to use fetal, olfactory ensheathing glia, bone marrow stromal, and other cell types for injection or implantation in uncontrolled and poorly documented experimental therapies in patients (Dobkin et al., 2006b). The sites offer hope at $20,000 to $50,000 for their services. These unpublished interventions tend to be based on a misinterpretation of preclinical experiments in rodents. Of interest, the patients who go for these interventions and then commit to a lengthy postoperative course of rehabilitation are usually the ones who report very modest changes.

Clinical trials in neural repair will be complex. No single intervention is likely to succeed, a similar situation to trials of neuroprotective agents. Too many biological pathways must be manipulated together and in series over time to get robust regrowth and effective connectivity or to incorporate new cells. To optimize the applicability of animal models to human translational research, clinicians will need to put the results of models into perspective and participate with basic neuroscientists in combining training and transplant technologies. In addition, strict adherence to best scientific practices for the conduct of randomized clinical trials will be needed, as suggested by guidelines from the International Campaign for Cures of Spinal Cord Injury Paralysis. Clinicians should encourage patients to read the educational materials about participating in neural repair experiments, such as the guidelines available at www.icord.org.

Pharmacological Interventions

Drugs that affect neurotransmission, intracellular second messenger signaling, excitation and inhibition, and the cascades associated with long-term potentiation and long-term depression are candidates to enhance learning and the reacquisition of skills after a brain injury or an SCI. At least five neurotransmitter projections have modulating effects on wide regions of the cortex and spinal cord. A lesion may transsynaptically interrupt this neurotransmitter output to cortex and cord, limiting the drive to uninjured regions and producing behavioral deficits. For example, dextroamphetamine augments specific cognitive processes by increasing the signal-to-noise ratio; cholinergic projections serve as a gate for behaviorally relevant sensory information. Cholinergic modulation is essential for motor cortex learning (Conner et al., 2005). Animal studies have provided preliminary evidence that a variety of pharmacological agents may facilitate or inhibit the rate or degree of recovery of sensorimotor function and walking after a cerebral injury. After being given dextroamphetamine, both rats and cats that underwent a unilateral or bilateral ablation of the sensorimotor or frontal cortex exhibited an accelerated rate of recovery, although not necessarily greater gains than in the control subjects, in the ability to walk across a beam. This effect endured well past the single or intermittent dosing schedule of the drug. Amphetamine and other drugs to be discussed have shown promise in small human trials. Thus, specific training paradigms combined with agents may enhance gains. Functional imaging studies with transcranial magnetic stimulation and fMRI may aid in the selection of drugs to combine with rehabilitation (Butefisch et al., 2002).

Drugs that may affect neural repair have also been tried, such as fibroblast growth factor, but side effects have exceeded expectations. Erythropoietin, granulocyte stimulating growth factor, an anti-Nogo antibody, and other factors are currently in clinical trials for patients with stroke and spinal cord injury. The designs of these trials suggest uncertainty about the mechanism of action, neuroprotection versus plasticity, in that they are initiated within 1 to 5 days of onset. Some drugs have more clear-cut mechanisms of action. For example, the conduction of action potentials along demyelinated axons may be increased by pharmacological agents such as 4-aminopyridine, which blocks potassium channels. This approach has met with some success in the responders with MS, who improved their ability to walk (Goodman et al., 2009). In some patients with motor neuron disease and Guillain-Barré syndrome, peripheral anticholinesterase-inhibiting medications, acting on neuromuscular junctions with altered structure and function, have reduced fatigability and improved strength in a modest way.

Genetic studies may become of value in identifying patients who have polymorphisms in a single nucleotide that affects memory, learning, cortical morphology, and other critical functions. If patients are predisposed to lower levels of brain catecholamines as a result of enzyme activity (catechol-O-methyltransferase Val versus Met polymorphism) or have higher or lower neurotrophin activity (brain-derived neurotrophic factor [BDNF] Val versus Met polymorphism), these differences may be amplified after brain injury, potentially affecting outcomes. Medications could be used in a more focused fashion with genetic screening to identify persons most likely to benefit.

Drug studies in humans pose confounding problems. The type, location, extent, and age of the lesion and the specific drug, its dosage, time of initiation, duration of use, and adverse effects and the accompanying physical or cognitive therapy that may add to a drug’s effect must be determined. What is clear is that clinicians should select medications with special care in the months after a cerebral injury and must monitor for effects that may impede recovery.

Neuromedical Problems During Rehabilitation

Neurological and systemic complications often interfere with progress during inpatient and outpatient rehabilitation. With shorter acute hospital and inpatient rehabilitation stays, physicians, nurses, and therapists must anticipate, recognize, and treat medical conditions that may impede progress in the rehabilitation process. Some of these problems will arise from new medications, neurological impairments, immobility, transient infections and metabolic abnormalities, and underlying systemic illness. In addition, patients and caregivers must be trained to prevent errors of omission and commission that arise in using medication, performing daily care, and managing risk factors such as hypertension that lead to late morbidity.

As noted, some but not all studies suggest that specialized stroke, SCI, and TBI hospital programs appear to lead to better early outcomes than with treatment on general medical units. Differences in morbidity, mortality, and length of hospital stay have been associated with more organized services. For example, protocols that use prophylactic subcutaneous heparin, hold oral intake until completion of a screening test for safety of swallowing, avoid indwelling bladder catheters, and assess postvoid residuals by ultrasound examination to avoid unnecessary intermittent catheterizations or overfilling, can reduce medical complications.

Frequency of Complications

Complications in Patients with Stroke

Medical complications often interfere with a patient’s ability to participate in therapy (Table 48.5). Medical and neurological complications occur at rates of approximately 4 and 0.6 per patient, respectively, during an average course of inpatient rehabilitation. A urinary tract infection, urinary retention, musculoskeletal pain, or depression will develop in approximately one-third of patients. Up to 20% will experience a fall, exhibit a rash, or need continuous management of blood pressure, hydration, nutrition, and glucose levels. In approximately 10%, a transient toxic-metabolic encephalopathy, pneumonia, cardiac arrhythmias, pressure sores, or thrombophlebitis will develop. Up to 5% will have a pulmonary embolus, seizures, gastrointestinal bleeding, heart failure, or other systemic complications. When feasible, prophylactic measures for these potential problems are essential. Because many patients in the United States are transferred from the acute hospital to a rehabilitation unit less than 7 days after a stroke, the rate of neuromedical complications may be higher at some centers. Side effects during the adjustment of new medications are especially prevalent, including orthostatic hypotension from antihypertensives or dialysis, drowsiness from anticonvulsants and analgesics, and a statin-induced myopathy with normal serum creatine kinase. Across rehabilitation centers, 5% to 15% of patients must be transferred back to an acute hospital setting.

Table 48.5 General Frequency of Inpatient Stroke Rehabilitation Neuromedical Complications

Complication % of Patients
Urinary tract infection 40
Musculoskeletal pain 30
Depression 30
Urine retention 25
Falls 25
Fungal rash 20
Hypertension 20
Hypotension 15
Incipient pressure sores 15
Hypoglycemia or hyperglycemia 15
Azotemia 15
Toxic-metabolic encephalopathy 10
Pneumonia 10
Arrhythmia 10
Congestive heart failure 5
Angina 5
Thrombophlebitis 5
Allergic reaction 5
Gastrointestinal bleeding 5
Pulmonary embolus <5
Myocardial infarction <5
Decubitus ulcer <5
Recurrent stroke <5
Seizure <5

Adapted with permission from Dobkin, B., 2003. The Clinical Science of Neurologic Rehabilitation. Oxford University Press, New York.

Complications in Patients with Spinal Cord Injury

Medical complications of a somewhat different nature are common in the acute and chronic phases after SCI. In this younger group of patients, chronic comorbid systemic medical problems are less common than in patients with stroke. Prior substance abuse and emotional and behavioral disorders, however, are much more likely to complicate therapy. In the first 6 weeks after SCI, reparative operative procedures affect what can be done in rehabilitation. Approximately half of patients undergo spinal fusion and internal fixation. Acute spinal care also entails the use of external stabilization techniques such as a halo vest and Philadelphia collar for cervical injuries and a thermoplastic, custom-molded, thoracolumbar fixation shell (see Fig. 48.6). These devices limit head and trunk mobility, which makes self-care tasks that involve balance, management of the lower extremities, and CISC more difficult. Lower extremity fractures, especially of the femur, occur in about 5% of patients with acute SCI, which can further limit mobility and increase the risk of deep vein thrombosis and skin breakdown. Table 48.6 lists the complications found during a prospective clinical trial of methylprednisolone for acute SCI in 487 patients.

Table 48.6 Medical Complications Within 6 Weeks of Acute Spinal Cord Injury

Complication % of Patients
Urinary tract infection 46
Pneumonia 28
Decubitus ulcer 18
Paralytic ileus 9
Arrhythmia 6
Sepsis 6
Thrombophlebitis 5
Wound infection 4
Gastrointestinal hemorrhage 3
Pulmonary embolus 3
Congestive failure 1

Early morbidity is greater with cervical and upper thoracic injuries and with complete lesions than with lower level or incomplete lesions. Ventilatory dysfunction, aspiration, dysautonomia with upright hypotension or paroxysmal hypertension, a neurogenic bowel with impactions, a neurogenic bladder with retention and infections, a catabolic state, and gastric atony are especially likely to complicate early inpatient rehabilitation. Hypercalciuria or hypercalcemia related to immobilization may also necessitate therapy. Central and musculoskeletal pain and grief reactions warrant immediate attention.

Complications in Patients with Traumatic Brain Injury

Systemic and neurological complications are common after serious TBI. The older patient carries more systemic comorbidity, whereas the younger patient may have alcohol or drug abuse as a comorbid condition. Unrecognized fractures and heterotopic ossification as a cause of pain, in addition to other bodily injuries may complicate rehabilitation along with any of the complications that may accompany stroke and SCI. Physicians must also monitor for pituitary-hypothalamic dysfunction with endocrinopathies and disorders of homeostasis including cerebral salt wasting and inappropriate secretion of antidiuretic hormone, as well as cerebral hygromas and obstructive hydrocephalus. Ventricular enlargement develops in 30% to 70% of patients with severe TBI. Most have hydrocephalus ex vacuo, which is passive enlargement of the ventricles from the loss of gray and white matter. In the patient with enlarging ventricles who reaches an early plateau or declines in mobility and cognition, a diagnosis of symptomatic normal- or high-pressure hydrocephalus must be considered.

Patients who are in a persistent vegetative state often are evaluated by the rehabilitation team for prognostication and for a trial of stimulation. After TBI, the prognosis is not quite so grim as after hypoxic-ischemic coma unless hypotension accompanied the TBI. Table 48.7 shows the outcomes during the first year after brain injury determined for a 1994 study by the Multi-Society Task Force on Persistent Vegetative State.

Management of Neuromedical Problems

Dysphagia

Neurogenic dysphagia is the potential cause of a pulmonary infection in any patient with stroke, TBI, motor neuron disease, MS, advanced PD, cervicomedullary disorders such as a syrinx, Guillain-Barré syndrome, myasthenia gravis, and most neuromuscular diseases. Indeed, swallowing disorders affect 10% of acutely hospitalized older adults and 30% of nursing home residents. Even transient dysphagia can lead to malnutrition, dehydration, aspiration pneumonia, and airway obstruction with asphyxiation. It increases the patient’s risk of death and institutionalization.

The natural history of recovery from dysphagia after stroke and TBI is good. For example, a British study diagnosed dysphagia in 30% of 357 conscious patients within 48 hours of a unilateral hemispheric stroke; the patients were rated as having impaired deglutition if they exhibited delayed and prolonged swallowing or if they coughed on 10 mL of water. Lethargy, gaze paresis, and sensory inattention were present more often than in those who swallowed normally. By 1 week, 16% had dysphagia. At 1 month only 2% and at 6 months only 0.4% of survivors were still impaired. Symptoms and signs that suggest a risk of aspiration include lethargy, coughing or a hoarse and gurgly voice after feeding, slow eating or drinking (<10 mL of water per second drunk from a cup), dysphonia, and poor oropharyngeal movement. The limitations of bedside indicators have led to the use of the MBS with videofluoroscopy as the method of choice to rule out silent aspiration. The relationship between small-volume aspiration as seen on MBS and clinical complications is uncertain, however, and requires clinical judgment. The MBS also visualizes the effects of dietary texture and compensatory techniques. In attentive stroke rehabilitation inpatients, coughing or a wet-hoarse quality of the voice noted within 1 minute of continuously swallowing 90 mL of water from a cup had a sensitivity of 80% and specificity of 54% for aspiration detected by an MBS study. The bedside test had a sensitivity of 88% and specificity of 44% for large-volume aspiration, which may be clinically more significant. Fiberoptic endoscopy also can identify mechanisms of dysphagia and can be performed at the bedside.

Dysphagia management must be addressed through all phases of recovery from stroke and TBI and during the progression of disorders such as PD, ALS, and myasthenia gravis (see Chapter 12B). Lesions in the pathways for swallowing interfere with the oral, oral preparatory, and reflex or pharyngeal phases of deglutition. Common deviations include loss of the ability to form a bolus and inadvertent trapping of liquids in the vallecula, with subsequent trickling over the vocal cords. The initial emphasis for care is placed on protecting the airway and maintaining adequate alimentation and hydration, with use of a nasopharyngeal feeding tube if necessary. A large single-site randomized trial found that a standardized daily therapy for swallowing and dietary modification for up to 1 month after stroke, compared with usual care or with standard care provided only 3 days a week, reduced dysphagia-related medical complications such as pulmonary infection and significantly increased the number of patients who regained swallowing function. As in many other rehabilitation trials, it may be that the intensity of a treatment is as important as the strategy to achieve greatest efficacy.

Swallowing may improve by downsizing a tracheostomy tube or eliminating it as soon as possible. Oral pooling of secretions and drooling, which predispose the patient to aspiration, can be managed with suctioning and sometimes with use of a low titrated dose of an anticholinergic drug such as glycopyrrolate. Good oral hygiene and prevention of tooth caries by oral rinsing with chlorhexidine gluconate and stannous fluoride can lessen the likelihood of carriage of bacterial infection into the lungs by oral secretions. Therapeutic efforts focus on the use of stimulation and compensatory approaches designed to reduce the swallowing impairment or to minimize the functional disability resulting from that impairment (Table 48.8). For example, postural adjustments may be made on the basis of the MBS findings. A chin tuck narrows the airway opening, tilting the head to the stronger side of the pharynx directs a bolus away from the weak side, and head rotation toward the weak pharyngeal muscle channels a bolus toward the stronger side. Modification of diet texture includes choosing thickened or gelled liquids, which are less likely to be aspirated than thin liquids. Purees and formable solids such as applesauce or mashed potatoes usually are safer than foods that require chewing and oral manipulation. Pharyngeal stimulation by the therapist may help drive representational plasticity for these muscles on the unaffected side of the brain after stroke to improve swallowing.

Table 48.8 General Therapies for Dysphagia

Compensation Sensorimotor Exercises Direct Interventions
Postural adjustments: Oral sensory stimulation (thermal, vibration) Palatal prosthesis
Head positioning Resistance and placement exercises of tongue and jaw Surgery
Chin tuck Chewing, oral manipulation of bolus Cricopharyngeal myotomy
Head rotation to weak pharyngeal side Laryngeal adduction Epiglottopexy
Dietary modifications: Biofeedback  
Softer food    
Thicker liquids    
Lower intake volume    
Slower intake pace    
Compensatory maneuvers:    
Double swallow    
Supraglottic swallow    
Laryngeal elevation    

The FOOD Trial Collaboration reported that early placement of a nasogastric tube to feed aphagic patients after stroke improved nutrition and reduced deaths and poor outcomes compared with no enteral feeding, but early use of percutaneous gastrostomy caused a 1% absolute increase in complications. Gastrostomy may not lessen the risk of reflux aspiration any more than a nasopharyngeal feeding tube for patients with persisting aphagia. Percutaneous endoscopic gastrostomy can be complicated by gastric perforation, peritonitis, hematoma, fistula formation involving the lung, stomal infection, cellulitis, and bleeding at the insertion site. Jejunostomy may lessen the risk of reflux but increases the risk of the dumping syndrome. Rarely, esophagostomy or pharyngostomy may be better suited for the patient with neurological dysfunction and prior gastrointestinal disease or surgery. Clinical trials related to the efficacy of types of tube feedings and other interventions for dysphagia are monitored by the Cochrane Review. Dysphagia therapies provided by the family under the guidance of a therapist may be as efficacious as hands-on therapy by a professional. With outpatient care, patients and family members assume greater responsibility for integrating and adapting recommended procedures to suit their individual needs and priorities.

Skin Ulcers

Education in skin management during rehabilitation provides an important opportunity for preventing later morbidity and mortality. Ischemia of the skin and underlying tissues occurs particularly in weight-bearing areas adjacent to bony prominences. The American Model Systems data for patients hospitalized within 24 hours of traumatic SCI showed that pressure sores subsequently developed in 4%, and 13% of these were graded as severe. The lesions occurred over the sacrum, heel, scapula, foot, and greater trochanter of the hip. Lower-grade skin lesions developed over the genitals. Sores related to sitting most often are located over the ischial tuberosities, where tissue pressure can exceed 300 mm Hg on an unpadded seat. A 2-inch-thick foam pad decreases the local pressure to 150 mm Hg. Even with the use of cushions designed to distribute pressure evenly over weight-bearing skin surfaces, pressures in the sitting position are far above the pressures of 11 to 33 mm Hg in the capillaries and venules. Raising the head of the bed by only a few inches especially increases shearing forces over the sacrum.

A standard classification for degrees of integument breakdown, prophylactic measures, and wound care is available from the Agency for Health Care Policy and Research in Washington, DC. Rubor, induration, and blistering are signs that precede a break in the skin. Pressure relief by turning and repositioning is the best approach, performed every half hour after a complete SCI and every 2 to 3 hours in patients with intact sensation after stroke or other disabling injuries. Patients must develop a skin care program based on their general health, nutrition, continence, toughness of their skin, most commonly used positions, type of wheelchair seat, presence of old skin scars, and other factors.

Deep Vein Thrombosis and Pulmonary Embolus

In controlled trials of prophylaxis, DVT has been found in 20% to 75% of untreated patients within 2 weeks of stroke; 5% to 20% suffered a pulmonary embolus (PE), which was fatal in approximately 10%. Intermittent calf compression for the paretic leg, intermittent low-dose heparin given as 5000 units every 8 or 12 hours, and low-molecular-weight heparinoid are far more effective than no intervention or the use of antiembolism hose alone. Across several studies, anticoagulants have reduced the incidence of DVT by a factor of two- to sevenfold, and of PE by about two- to fourfold. If a DVT is diagnosed, patients usually can restart activities out of bed after 2 days of intravenous heparin.

For those with SCI or head injury complicated by bone fractures, the incidence of DVT and PE is particularly high. After an SCI without a fracture, the risk of DVT, detected by a radiolabeled fibrinogen scan, impedance plethysmography, or Doppler blood flow study, appears greatest in the first 12 weeks, especially in combination with paraplegia and flaccidity. Symptomatic thrombophlebitis and PE are less common. In one study, thromboembolism was detected in 31% of plegic patients who were randomized to receive 5000 units of subcutaneous heparin twice a day within 72 hours of injury. It was detected in only 7% of patients whose activated partial thromboplastin time was prolonged to 1.5 times control values by dosage adjustment every 12 hours. Over 7 weeks of anticoagulation, the incidence of bleeding complications was greater in the adjusted-dosage group, especially at trauma sites. Most patients with uncomplicated SCI continue anticoagulation until discharge from inpatient rehabilitation.

Contractures

Across studies, approximately 15% of patients with SCI admitted for rehabilitation and 80% admitted after moderate to severe TBI lost more than 15% of the normal range of motion of at least one joint. Hemiparetic patients with stroke fall between these extremes. Contractures are found especially in the lower extremities in neuromuscular diseases, affecting at least 70% of outpatients with Duchenne muscular dystrophy. Contractures limit functional use of a limb and impair hygiene, mobility, and self-care. Serious contractures can cause pressure sores, pain, and, especially in youngsters, emotional distress when odd postures distort the body. After an acute UMN lesion, proper positioning of the arm in abduction and hand in extension and of the leg in hip abduction with knee flexed and ankle in neutral position can protect the affected extremities. Any source of pain will increase tone and predispose to contractures. Serial casting and surgeries occasionally are necessary for contractures that interfere with skin care or functional use of a limb.

Ectopic bone formation—heterotopic ossification (HO)—below the neurological level of injury may cause functional impairment in up to 20% of patients with SCI, usually in the first 4 months after injury. Patients with a complete lesion, pressure sores, spasticity, and age older than 30 years may be at greatest risk. After TBI, HO especially tends to affect the proximal joints of comatose patients. During rehabilitation of less responsive and cognitively impaired or aphasic patients, pain caused by undetected musculoskeletal injury and HO can add greatly to agitation and limit participation. HO develops when multipotential connective tissue cells transform into chondroblasts and osteoblasts, presumably under the influence of locally induced growth factors. The hips, knees, and shoulders are affected most often. Swelling, erythema, and decreasing range of motion are among the first clinical signs. A three-phase technetium 99m–labeled methylene-diphosphonate bone scan reveals focal uptake before radiographic visualization of bone. Early treatment with disodium etidronate suppresses mineralization of the osteoid. Range-of-motion exercises, aspirin, nonsteroidal antiinflammatory drugs, and a wedge resection of mature heterotopic bone can decrease pain and immobility.

Dysautonomia

Bedrest, dehydration, cardiac and antihypertensive medications, antidepressants, and autonomic reflex dysfunction from diabetes mellitus contribute to postural hypotension in the patient with stroke and TBI. Supine and standing blood pressures should be checked as mobilization proceeds during rehabilitation. Autonomic reflexes may fail in patients with SCI levels above T6. Symptomatic postural hypotension is common in the first weeks after injury and may persist in quadriplegic patients.

Initial therapies for OH include gradual reconditioning of postural reflexes on a tilt table, sleeping in a reverse Trendelenburg position to prevent overnight diuresis, wearing full compression leg hose, and application of an abdominal binder. In the inpatient setting, fluid loading with saline or albumin may aid the effort to compensate for venous pooling, decreased cardiac output, and impaired vasoconstriction and venoconstriction from interruption of sympathetic outflow. Fludrocortisone increases the intravascular volume and peripheral vascular resistance. The dosage can be pushed gradually to 0.5 mg per day. Salt tablets should be added to the diet. Hypokalemia and edema with pressure sores can complicate use of mineralocorticoids. While the patient is upright, ephedrine, 25 to 100 mg, up to every 3 hours; ergotamine, 2 mg, up to several times daily; or midodrine, up to 10 mg 3 times a day can be tried. Episodic autonomic hyperreflexia related to uninhibited sympathetic outflow may affect 50% to 90% of SCI patients with high-level SCI, usually beginning several months after injury. It is caused by visceral and joint pain, HO, pressure sores, bowel and bladder distention, fecal impaction, urinary infection and cystitis, ingrown toenails, pregnancy and labor, venous thrombosis, and late development of a syrinx. Wearing tight clothing, a particular supine position, or oropharyngeal suctioning also can cause bradyarrhythmias. Often it has no evident precipitating cause. Hypertension, headache, diaphoresis, anxiety, reflexive bradycardia, nasal congestion, flushing above and pallor below the SCI level, extensor spasms, and piloerection can result. The instigating agent must be removed. Acute therapies include upright positioning, search for an unemptied bladder or rectum using lidocaine on a catheter or finger when probing, and treatment of blood pressure that exceeds 180/100 mm Hg. A beta-blocker such as labetalol, a short-acting calcium channel blocker, vasodilators such as hydralazine, and occasionally phenoxybenzamine, prazosin, clonidine, or nitroglycerin usually lower the pressure safely. Some quadriplegic patients have very labile responses to antihypertensive drugs and suddenly become hypotensive with these treatments. For frequent bouts of hypertension, maintenance therapy includes low dosages of any of these oral agents but with dosage adjustments based on the finding of supine hypertension coupled with sitting hypotension. For paroxysmal bradycardia, propantheline or a pacemaker may be needed. Scopolamine and propantheline can prevent bouts of sweating.

Bowel and Bladder Dysfunction

Urinary incontinence occurs in up to 60% of patients in the first week after a stroke, but function tends to improve without a specific medical treatment. This likelihood must be considered in the context of an incidence of urinary dribbling and involuntary emptying of approximately 30% in the population of healthy, noninstitutionalized adults older than 65 years of age. Across studies, about 18% of those who were incontinent at 6 weeks after a stroke are still so at 1 year. By the end of inpatient rehabilitation, the incidence is about 10% in patients with a motor-only stroke and about 30% with large hemispheric strokes. A Cochrane Database systematic review of the prevention and treatment of urinary incontinence after stroke found insufficient evidence for most physical, behavioral, and pharmacological interventions, but specialist management tended to reduce the number of urinary tract infections.

After SCI, the bladder detrusor reflex may not return for 6 weeks to 12 months. In the absence of spontaneous bladder emptying, intermittent catheterization is done on a schedule that prevents the accumulation of more than 400 mL. Patients should measure their output from time to time and develop a voiding schedule that takes into account variations in fluid intake and the use of alcohol and caffeine. Catheters can be washed, stored in a plastic bag, and reused. If sensorimotor impairments persist, nearly all patients with an SCI lesion that spares the S2 to S4 micturition center will develop dyssynergia between the detrusor and the external sphincter. These uncoordinated contractions lead to incontinence and intermittent outlet obstruction. Urodynamic studies and an intravenous pyelogram are indicated as a baseline and to assist in therapy, especially in patients with SCI (see Chapter 50C). Obstruction can cause recurrent infections, urosepsis, vesicourethral reflux, urolithiasis, and hydronephrosis.

Patients should learn the signs of bladder fullness, such as sweating, changes in temperature, increased spasticity, and an increase in heart rate. Some palpate the area of the bladder to determine fullness. Once awareness of fullness develops, the person can aid in initiating or completing micturition by tapping over the bladder, rubbing the skin over the pubis or on the inner thighs, pressing on the abdominal wall (Credé maneuver), and bearing down or coughing. These maneuvers are particularly helpful in those with an LMN bladder with an open sphincter. Most patients with SCI also learn the signs of an early bladder infection, such as a change in clarity or odor, as well as more spasms, and work out a system with their physician to obtain a culture and antibiotics immediately.

Pharmacological treatment can be understood in relation to problems in bladder filling, storage, and bladder emptying with any neurological disease (Table 48.9). Urodynamic studies often aid in the choice of drug trial. The goals are continence and regular emptying that is achieved without high intravesicular pressure and with less than 100 mL of residual volume after voiding. Many males after stroke who have an enlarged prostate benefit from an alpha-blocker. If the goal for micturition is not achieved, several alternatives to the use of an indwelling catheter on a long-term basis are available. To prevent inadequate emptying at low pressure at the price of external sphincterotomy and resultant incontinence, long-term CISC can be used in combination with anticholinergic medications or bladder infusion of botulinum toxin to partially paralyze detrusor function. This procedure has become increasingly accepted as an effective alternative in the management of low intravesicular pressure. An external collecting device also can be used to ensure dryness. The patient must be trained to adjust fluid intake and the timing of catheterization to meet the flexible needs of community living. The absence of an effective external collecting device for female patients makes it necessary to continue long-term intermittent catheterization in women with paralysis of detrusor function. A waterproof undergarment may be worn between catheterizations to avoid embarrassment. The difficulties of this regimen cause many women to choose constant indwelling catheter drainage despite its drawbacks. Augmentation enterocystoplasty, reservoirs, conduits, and electrical stimulation techniques are less often needed. The VOCARE Bladder System for patients with UMN SCI allows patients to stimulate sacral nerves that have been implanted with electrodes to empty the bladder and bowel.

Table 48.9 Pharmacological Manipulation of Bladder Dysfunction

Medication Indication(s) Mechanism of Action
Bethanechol, 25 mg bid-50 mg qid Facilitate emptying Increase detrusor contraction
Prazosin, 1 mg bid-2 mg tid; or tamsulosin, 0.4 mg daily Decrease outlet obstruction Alpha blockade of external sphincter to decrease tone
Prostatic hypertrophy
Hyoscyamine, 0.125 mg hs-0.25 mg tid; oxybutynin 2.5 mg hs-5 mg qid Urge incontinence Relax detrusor, increase internal sphincter tone, decrease detrusor contractions
Tolterodine, 2 mg daily; or imipramine, 25-100 mg hs Frequency  
Enuresis  

bid, Twice daily; hs, at bedtime; qid, four times daily; tid, three times daily.

The optimal management of spina bifida in children includes self-catheterization. Through cartoons and drawings, children can be taught how to prevent germs from growing in the bladder. Intermittent self-catheterization of the bladder usually can begin by age 5 years or by the time the child starts school.

Bowel evacuation can be brought under control in a majority of cases of SCI or other causes of myelopathy or cauda equina injury. The goals include continence, the prevention of impaction and discomfort caused by inadequate elimination, prevention of rectal bleeding, and a reasonable amount of time for bowel care. Some people with SCI can identify a signal of rectal distention, such as sweating or an increase in spasticity. It is particularly useful to establish a fixed time, usually after a meal, for evacuation. Once a pattern has been established with stimulatory suppositories, patients often get by with dilatation of the anal sphincter, either digitally or by a glycerin suppository. Those who cannot develop enough intraabdominal pressure to defecate may need manual evacuation several times a week.

Fatigue

Although most commonly associated with MS, various forms of fatigue occur in all patients with central and peripheral neurologic diseases. Fatigue may encompass malaise, lack of energy, and depression. Underlying mechanisms may include disease-related autonomic, endocrine, and inflammatory dysregulation, as may occur in MS. Cognitive processing effort that is increased by the presence of cerebral lesions may induce a form of fatigue in carrying out mental and physical activities. One of the least recognized forms of fatigue that will interfere with ADLs occurs with repetitive use of muscles. This fatigability especially affects overused girdle muscles during walking but can be so insidious in the course of a motor activity that patients do not realize why they employ a slow gait with short stride. Clinically, this fatigability can be detected by testing strength immediately before and after 10 leg lifts by a supine or prone subject or after a 30-meter walk. It arises from problems as diverse as the more familiar fatigability found in myasthenia gravis and in lumbar stenosis with intermittent compression of the cauda equina. Impaired drive of motor units from the motor cortex due to less cortical excitability or greater inhibition, loss of corticospinal projections, conduction blocks along the descending projections, spinal mechanisms of reciprocal inhibition, disruption of firing rates of fast and slow motor units, and muscle atrophy can contribute. Medications such as the statins may cause a myopathy that induces weakness and greater fatigability that takes 6 to 12 weeks to resolve. Strengthening and conditioning exercises, as well as use of techniques for energy conservation, may lessen motor fatigability. Medications such as antidepressants, modafinil, and methylphenidate or noradrenergic modulation may reduce symptomatic fatigue.

Spasticity and Upper Motor Neuron Syndrome

Spasticity is found in less than 20% of hemiplegic patients after stroke. Patients with bilateral weakness from TBI, especially if hypoxic injury occurred, are likely to have signs of spasticity as well. After SCI, spasticity is more prominent with incomplete than with complete motor and sensory impairments, especially with cervical and upper thoracic lesions.

The most important UMN problems that cause disability are the decrement in motor control associated with dyssynergic patterns of muscle activation and the coactivation of agonist and antagonist groups during movements, which are associated with paresis, slow movements, loss of dexterity, and fatigability with repetitive contractions. Spasticity per se does not produce weakness and these other aspects of impaired motor control. Exaggerated cutaneous and autonomic reflexes and involuntary flexor and extensor spasms are the most disabling episodic problems associated with spasticity for patients. During rehabilitation, the most visually striking signs associated with the UMN syndrome tend to be treated, and sometimes overtreated. These include hyperreflexia; flexor synergy of the elbow, wrist, and fingers; plantar flexion of the ankle; and the dystonic postures, rigidity, and contractures that may accompany weakness.

Assessments

For routine assessments and for clinical trials of antispasticity interventions, a number of measures of hypertonicity have been used. The Ashworth Scale (Table 48.10) has had good interrater reliability in studies of stroke, SCI, and MS. The relationship between this score and disability is not so evident, however. Hypertonicity, clonus, and spasms vary in relation to positioning of the limbs, posture, and activities. A clenched, plegic hand may lead to disability if pain or maceration of the palm develops, but a treatment that lowers the Ashworth score is unlikely to improve motor control. Other measurement techniques require instrumentation.

Table 48.10 Clinical Measures of Spasticity

Score Measure
ASHWORTH SCALE
1 No increase in tone
2 Slight increase, producing a catch when joint is moved in flexion or extension
3 More marked increase in tone but easily flexed
4 Considerable increase, passive movement difficult
5 Affected part rigid in flexion or extension
SPASM SCORING SCALE
0 No spasms
1 Mild spasms induced by stimulation
2 Spasms less than 1/h
3 Spasms more than 1/h
4 Spasms more than 10/h

Treatment

Therapists usually can manage pathologically increased tone, spasms, and poor range of motion in patients with hemiplegia by aiming to maintain normal length of the muscle and soft tissue across a joint, eliminating shoulder and other sources of pain, and helping patients to avoid abnormal flexor and extensor patterns at rest and during movement. Splinting to extend the wrist and the long finger flexor muscles is a common practice in the first month after stroke, but in randomized trials, this intervention has not clearly reduced the likelihood of a wrist contracture without employing other modalities. Spasms and dystonic postures should be treated more aggressively when they interfere with nursing care and perineal hygiene or evolve into contractures and pressure sores. Treatment often is needed for patients with myelopathies who endure painful spasms or involuntary flexor or extensor trunk and leg movements during transfers and after minor cutaneous stimulation. Measures to lessen hypertonicity can be useful when hypertonicity appears to restrain voluntary upper or lower limb movements through co-contraction of agonist and antagonist groups. In most cases, however, upper extremity flexor postures during voluntary movement may be best explained by abnormal muscle contraction evoked by action. For example, the typical synergistic response to attempted shoulder abduction causes flexion torques at the elbow, probably because the usual coactivation that stiffens the elbow to stabilize it fails as a result of an imbalance in strength of the opposite extensor action.

Hypertonicity and spasms have potential value. For example, spasms can decrease muscle atrophy and bone demineralization and increase venous return. An extensor thrust can provide the rigidity needed for weight-bearing stance. Learning to induce an extensor spasm can assist bed transfers in patients with a myelopathy. Determining how and when hypertonicity interferes with a patient’s activities is the most useful way to determine whether an intervention is needed. Bouts of clonus and flexor and extensor spasms during ambulation, driving, wheelchair push-up pressure releases, transfers, self-care activities, bed mobility, sleep, and sexual activities can be counted over the course of a day or week. Any intervention should aim to greatly lessen recurrences.

Nociception can exacerbate hypertonicity and trigger flexor and extensor spasms and dystonic postures. A painful shoulder can cause the hemiplegic arm to flex at the elbow and wrist. Even an ordinarily innocuous stimulus like tight clothing or sunburn can abruptly increase tone, much as it can cause autonomic dysreflexia in the patient with a cervicothoracic SCI. Treatable pain stimuli include bowel and bladder distention, urinary tract infection, epididymitis, joint pain especially on range of motion, unrecognized fractures, pressure sores, ingrown toenails, and DVT. Resistance exercises, although generally useful during rehabilitation, can increase flexor or extensor tone, especially if the exercise brings out associated movements. Selective serotonin receptor inhibitors used for managing anxiety and depression can exacerbate spasms. The only way to make this diagnosis is to taper off the possibly offending drug and see if improvement follows by 2 weeks after cessation.

An overall approach to the management of pathological hypertonicity and spasms includes reversing any noxious stimulus, using physical interventions before adding drug trials, and reserving more invasive techniques such as nerve blocks and orthopedic or neurosurgical procedures for a few recalcitrant situations.

Pharmacotherapy

Controlled trials of antispasticity agents have varied widely in the target symptoms managed and the outcome assessments used. Functional gains related to locomotion and voluntary use of the upper extremity usually are marginal in any UMN disease. A medication that prevents disabling spasms, however, may improve quality of life. Table 48.11 lists useful first- and second-line drugs.

Table 48.11 Dosages of Medications for Symptomatic Spasticity

Drug Dosage
FIRST-LINE AGENTS
Diazepam 2 mg bid-15 mg qid
Dantrolene 25 mg bid-100 mg qid
Baclofen 5 mg bid-40 mg qid
Clonidine 0.05 mg daily-0.2 mg tid
Tizanidine 2 mg bid-8 mg qid
Gabapentin 400 mg tid-900 mg qid
OCCASIONALLY USEFUL ADDITIONS
Intrathecal baclofen 50-150 µg trial dosages; intrathecal infusion with pump
l-Dopa or carbidopa 25 or 100 mg bid, respectively
Phenytoin Serum concentration 10-20 µg/dL
Phenobarbital Serum concentration 10-30 µg/dL
Cyproheptadine 4 mg bid-8 mg qid
Chlorpromazine 10 mg daily-50 mg tid
Dronabinol 2.5 mg daily–tid

bid, Twice daily; qid, four times daily; tid, three times daily.

After an SCI, about 25% of patients are discharged with an antispasticity agent, and half are still using medication by 1 year. Patients with American Spinal Injury Society (ASIA) grades A and D (see Chapter 50C) are less likely to have been treated than those with grades B and C. Baclofen, tizanidine, and clonidine are especially useful in reducing clonus and extensor spasms caused by a myelopathy. The latter two drugs are α2-agonists that inhibit the excitatory influences of peripheral sensory inputs on motoneurons. Medications with short-lasting effects, such as tizanidine, may be especially useful in limiting spasms during sleep or brief activities like transferring from wheelchair to bed. Baclofen, dantrolene, and the benzodiazepines can cause muscular weakness and difficulty with weight bearing. Children with cerebral palsy and patients with hemiplegic stroke often need their extensor tone to ambulate on a paretic leg. Dantrolene tends to be most useful in managing hypertonicity of the upper extremity after stroke and TBI. Less than 0.5% of users develop hepatotoxicity after several months of intake. l-Dopa may be of additive value to lessen spasms in selected adults after stroke or SCI. All these drugs can also cause sedation, confusion, or hypotension; may add to bowel and bladder dysfunction; and can produce other central and systemic side effects. Great care must be taken in using them in the patient who has neurogenic dysphagia or a pseudobulbar palsy and is at risk for aspiration. Whenever a drug appears to be useful, it is worth tapering the dosage down from time to time so that the patient can help reassess continued benefits.

For refractory spasms and pain, intrathecal baclofen or clonidine given by an implanted programmable pump infusion generally has replaced a surgical myelotomy, intrathecal morphine, and electrical spinal stimulation. It also has replaced selective dorsal rhizotomy, except in some children with spastic diplegia from cerebral palsy. The functional effects on mobility and self-care are more difficult to discern, but some patients achieve modestly better walking speeds with intrathecal agents.

Chemical Blocks

Chemical agents such as phenol have been injected into the lumbar theca, nerve, motor point, or muscle to lessen inappropriate muscle co-contraction, spasms, and dystonic postures. Because motor point blocks can partially spare voluntary movement and could reduce reciprocal inhibition when given to an antagonist muscle, they could improve some aspects of motor control. Intramuscular infiltrative injections of 50% ethanol or botulinum toxin reduce focal resistance to passive movement for approximately 3 months.

Botulinum injections have seemed most efficacious for the wrist and finger flexors and plantar flexors of the ankle. Interpretation of the results of clinical trials using botulinum toxin requires attention to how well the outcome measure reflects clinical effectiveness for an important problem. Is a change in ease of passive range of motion, as in the Ashworth Scale, as meaningful as an increase in functional use of the limb? A few trials in children with cerebral palsy and spastic diplegia reveal modest 10% increases in walking speed after injections. The great majority of studies report a 1- or 2-point decrease in the Ashworth Scale score and support this finding by offering a global physician score that in reality probably reflects a perception of a decrease in tone around one or more joints. A randomized trial compared injection of type A toxin into forearm muscles with vehicle injection in patients after stroke who scored 3 or more on the Ashworth Scale for the wrist and 2 or more for the fingers (Brashear et al., 2002). A statistically significant decrease in the Ashworth score occurred at 6 and 12 weeks (e.g., a change from 0.1 in controls to 0.8 in the finger flexors of the treated group). Significant changes also were found in the Disability Assessment Scale, which was said to reflect functional disability. This disability is only a subjective measure of change in hand hygiene, pain, positioning, and dressing that does not require use of the hand. It would be easy to misinterpret the data as revealing better functional use of the hand after botulinum toxin injection, but the real meaning is that with the wrist and finger flexors loosened, the hand was easier to keep open passively, and the arm may then be easier to manage when, for example, the patient tries to put it through a shirt sleeve. Trials of botulinum toxin that are sponsored by the pharmaceutical industry have not included a control intervention that uses rehabilitative passive or active range of motion or treatments for pain that may decrease tone. Also, these studies have not tried to maintain the effect of greater passive ranging by adding physical therapy after an injection to try to prolong any benefit.