Neuromuscular disorders are defined by impaired proprioception, or perception of muscle length changes. This results when sensory information from muscle length and tendon force receptors is disrupted. Biofeedback has been most effectively used in stroke, incomplete spinal cord lesion, cerebral palsy, lower motor neuron lesion, and peripheral nerve problems.

Stroke is a representative neuromuscular application. A typical stroke patient shows hyperactive agonist and weak antagonist muscles. The clinician attempts to down train spastic agonist muscles, up train weak antagonist muscles, and restore functional movement. Taub's Constraint Induced Movement Therapy is a revolutionary approach to treating stroke and other disorders, which combines, physical therapy, behavioral principles, and biofeedback to overcome learned non-use of an affected limb.



This unit covers the Central nervous system (IV-B), General treatment considerations (IV-D), and Target muscles, typical electrode placements, and SEMG treatment protocols for specific neuromuscular conditions (IV-E).
 
Students completing this unit will be able to discuss:

  1. Central nervous system
    A. Neuroanatomy, neurophysiology, and pathology
    B. The types of central nervous system injuries that produce spasticity, rigidity, and flaccidity
  2. General treatment considerations
    A. Standard physical rehabilitation techniques and procedures
    B. SEMG biofeedback techniques used to improve motor control
    C. Use of SEMG biofeedback to promote relaxation in clients with spasticity and rigidity
  3. Target muscles, typical electrode placements, and SEMG treatment protocols for specific
    neuromuscular conditions
    A. Immobilization/weakness of limb segments







The prefrontal cortex controls attention and planning. Most complex behaviors are planned in the prefrontal cortex. The frontal lobes develop strategies for diverse contingencies, monitor events, predict outcomes, and switch to new strategies as needed.

The supplementary motor area programs movements into sequences during learning. The premotor cortex adjusts existing motor programs under the guidance of external sensory information.

The primary motor cortex executes motor programs through two descending tracts: the lateral group (corticospinal tract, corticobulbar tract, and rubrospinal tract) and the ventromedial group (vestibulospinal tract, tectospinal tract, reticulospinal tract, and ventral corticospinal tract).

The corticospinal tract provides discrete control of fingers, hands, arms, trunk, and upper legs. The axons of upper motor neurons descend from the primary motor cortex to the medulla where most fibers cross to the opposite side (lateral corticospinal tract) and a minority descend on the same side (ventral corticospinal tract).

These neurons synapse, directly or via interneurons, on lower motor neurons in the ventral gray matter of the spinal cord. The connections are one-sided (lateral tract) or two-sided (ventral tract). The two-sided connection provides the basis for rehabilitation using Wolf's motor copy procedure which will be described in the next section.

The lateral and ventral tracts activate central pattern generators in the spinal cord that direct the recruitment of motor units to perform repetitive movements.



Cerebrovascular accident (CVA) or stroke involves destruction of brain tissue (upper motor neurons) due to disorders of blood vessels that supply the brain. CVAs show abrupt onset and involve temporary or permanent neurological symptoms like aphasia, paralysis, or loss of sensation.





CVAs are produced by cerebral hemorrhage and cerebral ischemia. Cerebral hemorrhage involves the rupture of a cerebral blood vessel, often due to bursting aneurisms (ballooning of blood vessel walls), that release blood that damages adjacent neural tissue. A subarachnoid hemorrhage releases blood into the subarachnoid space between the middle meninx (membrane) that protects the brain and the brain. An intracerebral hemorrhage releases blood into the brain. Cerebral hemorrhage accounts for 15% of strokes and is fatal in up to 50% of patients.

Cerebral ischemia disrupts blood circulation to a brain region and accounts for 85% of strokes. The three main causes of cerebral ischemia are arteriosclerosis, thrombosis, and embolism.

Arteriosclerosis is a narrowing of cerebral blood vessels, often due to lipid buildup, that can result in total blockage. Thrombosis involves the formation of a plug called a thrombus, which can be comprised of air bubbles, blood clots, fat, oil, or tumor cells (or their combination), that blocks circulation where it forms. An embolism is a thrombus that moves downstream (from a larger to a smaller blood vessel) where it becomes stuck and causes almost three of every four ischemic strokes.



                                        
 

Most of the damage inflicted by cerebral ischemia does not occur immediately, but rather develops over 1-2 days. While the mechanisms of ischemia-induced damage may vary across brain structures and regions like the hippocampus are more vulnerable than others, most neuron loss may be due to excessive release of excitatory amino acids, especially glutamate. Glutamate release may produce lethal swelling in neurons due the entry of ions and water. Glutamate also increases calcium entry into neurons and may raise the levels of enzymes that produce apoptosis or cell death (Freberg, 2006; Pinel, 2006). Glutamate may also be responsible for ischemic damage to oligodendrocytes (CNS glial cells that myelinate axons) via overactivation of NMDA receptors (Salter & Fern, 2005).



CVA is the most common brain disorder, affecting 700,000 Americans each year, two-thirds of these 60 or older. CVA is currently the third leading cause of death, following heart attacks and cancer (Newsweek, March 8, 2004).



The pyramidal motor system starts in the primary motor cortex and projects to spinal cord motor neurons. Fibers from the primary motor cortex cross over to the opposite side in structures on the ventral medulla that are shaped like pyramids. Damage to any part of the pyramidal system will affect movement, especially fine motor control.

Damage to any part of the corticospinal tract results in transient flaccid paralysis, which is complete loss of muscle tone, with paralysis, seen immediately after damage to the pyramidal motor system. Transient flaccid paralysis usually lasts for only a few days or weeks and is gradually replaced by a more permanent state of hyperreflexia, in which reflexes are extremely reactive and exaggerated. This can produce spastic paralysis.

Damage to areas of the parietal and prefrontal cortex in the left hemisphere that relay information to the primary motor cortex about the sequence of movements to be performed can produce apraxia. In apraxia, a person cannot organize movements into a productive sequence and can no longer perform previously familiar hand movements (Wilson, 2003).



The extrapyramidal motor system arises from neurons in the cerebral cortex, basal ganglia, cerebellum, and reticular formation. The extrapyramidal motor system controls bilateral, gross movements (lifting an object) and postural changes.

Damage to the extrapyramidal system causes hyperreflexia, which leads to spasticity. Spasticity interferes with normal smooth movement of the limbs. As the injured person attempts to move a limb, stretch reflexes in antagonist muscles are called into action, and the limb moves in a jerky fashion (clonus) until rigidity sets in, halting movement altogether (Wilson, 2003).



Research by Wolf has shown that impaired proprioception (perception of body position) may be central to loss of motor control following a CVA. If a CVA impairs a patient’s ability to understand, follow, and remember instructions, this would prevent successful biofeedback training. For this reason, neurologists must evaluate a patient’s cognitive performance before starting biofeedback-assisted physical therapy.




           


The main requirements for biofeedback-assisted rehabilitation are intact motor units that can be recruited and patient ability to understand and remember instructions, localize a limb or joint in space, and interact with the therapist or instrument.



The availability of intact motor units can be confirmed by inserting percutaneous electrodes in a muscle and instructing the patient to perform voluntary movements.

These electrodes may be required because muscle output is weak, the muscle is lies deep beneath the skin, adjacent muscles contaminate the EMG signal, or the EMG signal is absorbed by subcutaneous fat.



From 22% to 25% stroke patients die within a year of their first stroke. About 90% of these survivors will suffer mild to catastrophic long-term disabilities in their movement, sensation, memory, or cognition (Newsweek, March 8, 2004).



A study of patients 65 years and older six months following their strokes (Newsweek, March 8, 2004) concluded that:
    50% were paralyzed on one side
    35% were depressed
    30% could not walk without assistance
    26% required assistance with daily activities
    26% lived in nursing homes
    19% suffered aphasia



Physical therapy should be started as soon as the patient is stable, as early as 2 days poststroke. While some patients will experience the fastest recovery in the first few days, many will improve for six months or longer. Physical rehabilitation procedures include diathermy, heat packs, or hot tubs to reduce spasticity; muscle relaxants like Flexoril to reduce spasticity; evaluation of range of motion and exercises to increase range of motion; immobilizing the unaffected limbs; and strength training.



Constraint-Induced (CI) therapy represents a revolutionary paradigm shift that produces large gains in limb use following a cerebrovascular accident. The rationale for CI therapy is that a patient following a CVA or injury experiences "learned non-use," which contributes to loss of limb function. When a patient's efforts to use the affected limb result in failure, this reinforces its non-use and reliance on the healthy limb. These behavioral changes result in cortical reorganization that ensures the continued non-use of the affected limb.

CI therapy constrains movement of the unaffected limb by placing it in a sling or mitt for 90% of waking hours for 2-3 weeks. Taub's team constrains the healthy limb to force the patient to rely on the affected one and uses operant procedures to restore functional use of the affected limb. The most important operant principle used in CI therapy is shaping, where therapists reinforce successive approximations of the target behavior. They intensively train patients to use the more-affected limb many hours daily (massed practice) using biofeedback as a adjunctive procedure.

Therapists provide patients with continuous auditory and visual feedback, including verbal reports and "shaping data forms" which graphically summarize their progress. In CI therapy for lower limbs, limb load monitors and goniometers can provide immediate and continuous feedback about gait (Shaffer & Moss, in press).

Taub (2005) described the use of biofeedback in CI therapy: "The limb load monitor provides continuous feedback of the force and timing of each foot fall. We use it to correct stance time on the more affected leg, increase weight supported by it and increase the cadence of gait. The electric goniometer gives either continuous or error feedback of knee joint angle."





           


Neuroimaging and transcranial magnetic stimulation studies show that CI therapy increases the cortical area controlling the more affected limb through use-dependent cortical reorganization.

CI therapy has been effectively used to treat the upper limb of CVA and chronic traumatic brain injury patients and the lower limb of CVA, incomplete spinal cord injury, and hip fracture patients. CI therapy has also been applied to treat musician focal hand dystonia and phantom limb pain.



Doses of drugs, like Flexoril, that reduce spasm can produce fatigue, sleep, lethargy, and hypotension. Therapists should ask patients to hold finger up in the air to prevent falling asleep. EMG training is still possible even though these drugs will lower EMG.

EMG biofeedback is incorporated in physical therapy exercises to reduce spasticity, strengthen paretic muscles, and restore functional movement.



Neuromuscular disorders should be treated within the context of physical therapy. A rehabilitative medicine specialist should evaluate the patient and design the treatment plan. Biofeedback should be integrated in assessment and rehabilitative exercises. Biofeedback can provide real-time information for the clinician and patient.

Normal movement is prevented by hyperactive agonist and weak antagonist muscles. Stroke produces hemiplegia, the paralysis of upper extremity, trunk, and lower extremity muscles on one side. Muscle testing reveals agonist spasticity (excessive muscle tone and tendon reflexes) and antagonist paresis (muscle weakness). EMG evaluation using percutaneous electrodes can establish whether antagonist muscles are innervated and can benefit from therapy. Treatment strategy is to reduce spasticity, increase recruitment in paretic (weak) muscles, and restore functional movement.



In Wolf's (1985) Concurrent Assessment of Muscle Activity (CAMA), the clinician uses biofeedback information to suggest changes in ongoing patient posture, position, or movement. As the patient attempts the recommended change, the clinician can evaluate its effectiveness and provide new instructions.

Alternatively, the clinician can directly display information to the patient to modify performance. A feedback display provides more specific and immediate information to the patient than a therapist's instructions to "relax" or "try harder," which may follow observation and palpation (examination by touch) by seconds.



Narrow spacing (1 cm apart) is advised for weak antagonists (tibialis anterior) if spastic muscle (gastrocnemius) interference is a problem. If spastic muscles do not contaminate EMG readings, clinicians start with wider spacing (2-3 cm) over weak antagonists and progress from wide to narrow as the patient recovers. Wolf (1985) recommends use of two EMG channels to simultaneously monitor agonist and antagonist muscle groups

The gastrocnemius muscle is shown below at the back of the calf.



     
                                           
               




Clinicians should use a wide bandpass whenever possible, small sensors (12-15 mm), and placement of active sensors parallel to muscle striations.



Rehabilitation proceeds from proximal (trunk) to distal (extremities). Training may start with the anterior deltoid and then move to the middle deltoid, wrist extensors, fingers, and thumb. This sequence provides a stable platform required for effective use of extremities like the fingers.

Training starts in a supine (on the back) position and then moves to sitting, standing, and walking. The therapist follows the proximal to distal sequence in each position. The therapist trains the patient to reduce spastic muscle activity at rest and then to maintain low EMG levels when the muscles are passively stretched. When spastic agonists are relaxed in each position, the same proximal to distal sequence is repeated to strengthen weak antagonists. The final stage of training shapes purposeful movements like walking or grasping a cup.



Hemiplegia involves paralysis of only half of the body. Wolf (1985) uses a patient's bilateral control (rubrospinal and ventromedial tracts) over skeletal muscles in the motor copy technique.

The clinician asks the patient to contract an extensor on the unaffected side and then displays an EMG tracing of this effort. Next, the clinician asks the patient to match this tracing during contraction of a mirror image extensor on the affected side. When patients successfully copy the healthy movement, they may have substituted bilateral control for contralateral (lateral corticospinal tract) control.



The main requirements for biofeedback-assisted rehabilitation are intact motor units that can be recruited and patient ability to understand and remember instructions, localize a limb or joint in space, and interact with the therapist or instrument. The availability of intact motor units can be confirmed by inserting percutaneous electrodes in a muscle and instructing the patient to perform voluntary movements. These electrodes may be required because muscle output is weak, the muscle lies deep beneath the skin, adjacent muscles contaminate the EMG signal, or the EMG signal is absorbed by subcutaneous fat.



Loss of proprioception, receptive aphasia, and active shoulder range of motion may limit treatment outcome in upper extremity hemiplegic patients. Treatment outcome seems unrelated to age, sex, time since stroke, length of previous treatment, degree of expressive aphasia, or lesion site. For hemiplegics (paralysis of one side of the body), reduced muscle hyperactivity during passive stretching and ability to isolate wrist and finger movements may best predict improvement.



Clinicians routinely monitor EMG activity, force, joint angle, and velocity when treating neuromuscular disorders.

EMG biofeedback measures muscle action potentials which precede the mechanical contraction of a muscle. Force feedback indexes vertical force conducted through the body, hand, lower limb, or assistive device.

Three approaches include limb load monitors (LLM), feedback canes, and force transducers.

Joint angle is observed when we are concerned with degree of knee extension or wrist pronation. This parameter is measured by goniometers which reflect change in joint angle by changing resistance to current. Finally, velocity feedback monitors variation in the movement of a body part. Symptoms, like hand tremor, can be measured by an accelerometer.

EMG biofeedback is not always the modality of choice. Muscles that control patient performance cannot always be monitored by surface or percutaneous electrodes. Further, the most direct index of performance may be force, joint angle, or velocity instead of EMG level. Clinicians should use the biofeedback modality that most directly and immediately measures the treated symptom.

The best biofeedback modality can amplify the effectiveness of physical therapy by providing specific information in real time. Biofeedback measurements can assist the clinician during diagnosis, physical therapy exercises, and evaluation of patient improvement, supplementing less precise and immediate data from observation and palpation.



Force feedback is feedback of force conducted through the body or hand during work or exercise, or through a lower limb (force plate) or assistive device (feedback cane) to train patients to stand or walk. Three forms of force feedback may supplement EMG biofeedback: a feedback cane, limb load monitor (LLM), and force transducer.

A feedback cane uses a strain gauge to measure pressure applied to the cane when it is used to assist walking. When the patient uses the cane inappropriately (force exceeds a preset threshold), a warning tone reminds the patient to bear more weight. A therapist can gradually adjust a threshold downward to gradually shape independent ambulation (walking). The feedback cane also may be used after hip replacement to initially increase reliance on the cane to prevent injury.

A limb load monitor (LLM) is a force feedback device used to train hemiplegic patients to balance their weight distribution (rising from a chair, sitting, and standing) and to shift weight during the “stance phase” of walking. A force transducer can be inserted in a shoe or placed inside a platform (shown below).




Wolf (1985) proposed that clinicians use this device diagnostically to record force output to evaluate gait parameters like cadence and velocity. The LLM can be used in physical therapy to provide auditory and visual feedback when the patient meets the therapist's goals for amount of force and weight distribution during standing and walking. The therapist can adjust the LLM parameters to shape ambulation as with the feedback cane.

A force transducer (shown below) detects the force conducted through the body or hand during work or exercise.





This sensor is invaluable in ergonomic assessment and in training patients to reduce the risk of work-related injuries.





An electrogoniometer is a potentiometer that translates change in joint angle into a change in electrical resistance. A threshold can be selected to provide joint position feedback.

For example, when training elbow extension, a high-pitched tone can signal unwanted flexion, while silence means correct extension.



 



An accelerometer measures the the linear motion of a body part, like the wrist. A gyroscope (shown below) measures rotational motion and is not influenced by gravity. These instruments can measure the variation in wrist movement (lap to table surface) and hand tremor. A patient watches the feedback display, instead of the involved hand, to suppress tremors. Velocity feedback is more direct than EMG biofeedback since it monitors the motor behavior we want to modify (tremor).

 


 



Treatment outcomes are better for the lower than upper extremities. Wolf (1982) reported 20% of stroke patients regained independent use of the upper extremity, 30-40% improved, and 30-40% did not improve. In contrast, 60% of Wolf’s sample could walk independently, 30% required fewer assistive devices (walkers), and 10% did not improve.

Why the dramatic difference in outcome? Movement of upper extremities requires thumb movement. Unfortunately, the cortical area that controls the thumb is a large target that a stroke is likely to damage, surface EMG recording is prevented by the thenar eminence's (thumb) small surface area, and there are no good substitution systems.



Reduced spasticity and relationship with the patient may be important variables in patient recovery. Wolf (1982) believes that reduced agonist spasticity accounts for most improvement in neuromuscular rehabilitation. Normal movements like extension (by the triceps) are prevented if the agonist (biceps) is spastic. Wolf (1985) suggests that the therapist's relationship with the patient may also influence recovery.

Reduced spasticity and relationship with the patient may be important variables in patient recovery. Wolf believes that reduced agonist spasticity accounts for most improvement in neuromuscular rehabilitation. This makes sense since the agonist is opposed by the antagonist at a joint. Normal movements like extension (by the triceps) are prevented if the agonist (biceps) is spastic.

Wolf (1985) suggests that the therapist's relationship with the patient may also influence recovery. Patients may be more strongly motivated to participate in rehabilitation exercises when they perceive that the therapist is competent and cares about them.



Foot drop is common in hemiplegia due to stroke. This disorder involves failure to dorsiflex and evert the foot. The cause is weakness or paralysis of dorsiflexor (tibialis anterior) and evertor (extensor digitorum longus) muscles of the foot and toes combined with spasticity in the plantar flexors (gastrocnemius). EMG-assisted rehabilitation attempts to strengthen dorsiflexors and evertors while reducing spasticity in opposing plantar flexors.

The tibialis anterior muscle is shown below.



     
                                           
               





Evidence-Based Practice in Biofeedback and Neurofeedback (2004) rates EMG biofeedback for stroke at level 2 efficacy, possibly efficacious. The criteria for level 2 efficacy include "At least one study of sufficient statistical power with well identified outcome measures, but lacking randomized assignment to a control condition internal to the study" (pp. 31-33).

Meta-analytical studies have found conflicting results for EMG biofeedback in the treatment of stroke patients. EMG biofeedback has been shown to improve lower extremity function.

Schleenbaker and Mainous (1993) analyzed 8 studies that involved 192 patients and found that EMG biofeedback was useful in treating hemiplegic stroke patients. The effect size was an impressive 0.81.

Moreland, Thomson, and Fuoco (1998) concluded that EMG biofeedback produced greater ankle dorsiflexor strength than physical therapy, but failed to improve ankle range of motion or angle during gait, gait quality or speed, or stride length.

EMG biofeedback has not been effective for improving upper extremity function or range of motion.

Moreland and Thomson (1994) found that EMG biofeedback and physical therapy produced equivalent changes in upper extremity function.

Glanz et al. (1995) analyzed 8 studies and found no evidence that EMG biofeedback could restore the range of motion of hemiparetic upper or lower extremity joints.



Musculoskeletal weakness is a common symptom and can be produced by diverse causes including:

In true muscle weakness, called paresis, maximum effort fails to generate normal strength. This can be diagnosed by asking a patient to squeeze your hand as hard as possible. If the patient suffers from paralysis, there will be no muscle activity. Paresis or paralysis are frequently associated with spasticity, excessive muscle activity, in agonist muscles. This is illustrated when a patient involuntarily contracts the triceps when attempting to contract the triceps.

The availability of intact motor units can be confirmed by inserting percutaneous electrodes in a muscle and instructing the patient to perform voluntary movements. These electrodes may be required because muscle output is weak, the muscle is lies deep beneath the skin, adjacent muscles contaminate the EMG signal, or the EMG signal is absorbed by subcutaneous fat. 

Physical therapy should be started as soon as the patient is stable, as early as 2 days poststroke. While some patients will experience the fastest recovery in the first few days, many will improve for six months or longer. EMG biofeedback is incorporated in physical therapy exercises to reduce spasticity, strengthen paretic muscles, and restore functional movement.

Neuromuscular disorders should be treated within the context of physical therapy. A rehabilitative medicine specialist should evaluate the patient and design the treatment plan. Biofeedback should be integrated in assessment and rehabilitative exercises because it can provide real-time information for the clinician and patient. 



When trauma severs the spinal cord, muscle groups below the site of injury may be disconnected. No motor commands can reach them, and they cannot send proprioceptive information back to the brain. Rehabilitation can be attempted only when spinal cord lesion is incomplete and there are intact pools of motor units that can be recruited through exercise.

Fortunately, almost all spinal cord injuries are incomplete, and the remaining motor neuron cell bodies can be trained. Marked flexor spasticity is seen 3-6 months after spinal cord injury. Without treatment, this can result in muscle shortening (contracture). Extensor spasticity starts about 6 months following spinal cord injury. A clinician can use percutaneous EMG evaluation in acute (recent) cases to identify muscle groups that are still connected and can benefit from isometric strengthening.

EMG biofeedback-assisted rehabilitation is promising for both quadriparetic patients (weakness in four limbs) and paraparetic patients (weakness in lower limbs). The hemiplegia treatment protocol described earlier is also used in incomplete spinal cord lesions. The main clinical objective is to reduce spasticity of agonists in response to passive stretching (stretching produced by the therapist). As with hemiplegics, the clinician moves from proximal to distal in supine, sitting, and standing positions, and then during functional movement. Force feedback using the biofeedback cane and limb load monitor may supplement EMG biofeedback.
 


Evidence-Based Practice in Biofeedback and Neurofeedback (2004) rates EMG biofeedback for incomplete spinal cord lesion at level 1 efficacy, not empirically supported. The criteria for level 1 efficacy include "Supported only by anecdotal reports and/or case studies in non-peer reviewed venues. Not empirically supported" (p. 31). This rating was awarded due to insufficient investigation.

Brucker and Bulaeva (1996) reported increasing triceps EMG activity following only one EMG biofeedback training session, with further gains over subsequent sessions.

Petrofsky (2001) reported on the success of a 2-month program involving daily  training to increase muscle strength and improve gait. Five patients received 30 minutes of EMG biofeedback a day, while the remaining 5 patients received continuous EMG biofeedback from a portable electromyograph whenever they walked. At the end of 2 months, patients receiving only 30 minutes of biofeedback reduced hip drop by 50% and those who received continuous feedback achieved near-normal gait.



Cerebral palsy (CP) is a family of motor disorders that involve irreversible motor disability due to damage before or soon after birth.

These patients show flaccid paralysis (loss of muscle tone, loss or reduction of tendon reflexes, atrophy, and muscle degeneration), spastic paralysis (increased muscle tone, increased tendon reflexes, and pathological reflexes), and athetoid movements (slow, continuous twisting movements). The neurological deficits resemble stroke. Unlike stroke, lower limb spasticity may be greater and the antagonist may not be weak, but are overpowered by a hypertrophied (enlarged) agonist.



The prevalence of CP in the United States is 1.5-2 cases in 1000 live births, which is about 10,000 cases per year. While the initial injury occurs between birth and age 2, diagnosis is usually made after age 1 following failure to meet developmental milestones (Thorogood & Alexander, 2005).



More than half of the cases are due to prenatal insults that include infection from the mother to the fetus, maternal stroke, environmental toxins, and problems in brain development. The remaining cases are due to adverse events, including traumatic birth delivery, complications of premature birth, meningitis, and head injury during child abuse.



The three main types of cerebral palsy are spastic, dyskinetic, and ataxic. The mixed type combines these symptoms.



About 70-80% of CP patients experience spasticity or limited movement due permanently contracted muscles. Spasticity is produced by damage to voluntary motor control pathways.

About 10-15% of CP patients present with dyskinetic symptoms. These include uncontrolled athetoid movements (slow, continuous twisting movements). Athetoid movements are produced by damage to neurons that inhibit muscle contraction.

Less than 5% of CP patients present with the ataxic type of cerebral palsy. Their coordination is impaired while walking and moving their arms. Ataxia is produced by injury to the cerebellum which maintains balance and smoothes motor performance. Children have decreased muscle tone in hypotonic cerebral palsy. This results in muscle weakness and joint instability.

The spastic type of CP is subdivided into hemiplegia (20-30%), which involves the arm and leg on one side, diplegia (30-40%), which involves both lower extremities more often than the upper extremities, and quadriplegia (10-15%), which involves all four limbs and the trunk. Monoplegia, which involves an arm or leg is rare (Thorogood & Alexander, 2005).

Mental retardation is common, but is not present in all cases. Patients with cerebral palsy may also experience epilepsy, visual disturbance, hearing impairment, language difficulty, and slow growth.



Physical therapy is used to improve strength, range of motion, and joint mobility. Braces are often used to keep joints in appropriate positions. Medications like Botox are used to temporarily treat symptoms like muscle spasticity.







Cerebral palsy patients often lack normal afferent (incoming sensory) information about limb position and movement. Biofeedback can complement physical therapy for spasticity, athetoid movements, and ataxia by replacing incomplete information from afferent nerves with auditory and visual displays.



Treatment may be designed to correct posture and gait, and prevent athetoid movements. SEMG biofeedback may be integrated with rehabilitation exercises to address these problems.

Positional feedback may be invaluable in correcting abnormal head position. A feedback helmet with position sensors may sound a warning tone when a patient's head deviates from normal posture. A feedback helmet provides more direct and effective feedback than SEMG sensors placed over the sternocleidomastoid or semispinalis capitis muscles that move the head.

Colbourne, Wright, and Naumann (1994) reported that SEMG biofeedback for the triceps surae (leg) increased gait symmetry more than physical therapy.

Toner, Cook, and Elder (1998) reported that SEMG biofeedback to increase recruitment of ankle dorsiflexors improved tapping, which was used to assess ankle performance.

Bolek, Moeller-Mansour, and Sabet (2001) described a "minimax" procedure that utilizes SEMG biofeedback to teach disabled children how to correctly recruit and relax the gluteus medius and gluteus maximus muscles during sitting. This protocol immediately reinforces correct posture to teach children awareness of good sitting form.

Bolek (2003) reported successful real-time SEMG biofeedback training to recruit and relax the anterior tibialis at the correct points of the gait cycle. He trained two children diagnosed with CP to walk with increased toe clearance during the swing phase of gait.




                  




Evidence-Based Practice in Biofeedback and Neurofeedback (2004) rates SEMG biofeedback for cerebral palsy at level 2 efficacy, possibly efficacious. The criteria for level 2 efficacy include "At least one study of sufficient statistical power with well identified outcome measures, but lacking randomized assignment to a control condition internal to the study" (p. 14). This rating was assigned due to the limited number of studies of SEMG biofeedback efficacy.




Multiple sclerosis (MS) involves progressive destruction of the myelin sheaths that insulate axons, short-circuiting conduction.



There are approximately 350,000 cases of MS in the United States, with 10,000 new cases diagnosed each year. MS is more frequently seen in females than males (1.6 -2 to1). While females experience more relapses, males more often experience the primary progressive form. MS mainly affects individuals from 18-50 years (Dangond, 2005).



Both SEMG-assisted rehabilitation and velocity feedback may be used to correct intention tremor (quivering that appears or worsens when a patient attempts coordinated movement). The SEMG treatment protocol teaches patients to down train involved muscles and moves from proximal to distal. This approach is time consuming and suffers from the lack of standardized muscle relaxation criteria.



Peripheral nerves include the lower motor neurons that innervate skeletal muscles. These may be damaged by trauma, edema (swelling), and infection. Clinical signs of peripheral nerve injury include flaccid paralysis and muscle atrophy (wasting). Examples include Bell palsy (also called Bell's palsy), brachial plexus injuries, and poliomyelitis.



Bell palsy is responsible for about 60-75% of cases of acute unilateral facial paralysis. The annual incidence of Bell Palsy in the United States is 23 per 100,000 cases. This disorder affects the right side of the face in 63% of patients. While Bell palsy generally affects both sexes equally, females 10-19 are at greater risk than their male counterparts. Pregnant women have a 3.3 times greater risk than nonpregnant women. The lowest incidence of Bell palsy is under 10 and the highest incidence is 60 and older.

Prognosis for Bell palsy patients is usually good. About 80-90% of patients recover within 6 weeks to 3 months without perceptible disfigurement. About 40% of patients 60 years or older completely recover and have a greater risk of complications like synkinesis (involuntary contraction when another muscle is contracted) and crocodile tears syndrome (profuse tear production when tasting strongly-flavored food), and rare hemifacial spasm (Monnell, Zachariah, & Khoromi, 2005).



SEMG-assisted rehabilitation has been widely used in Bell palsy. These patients experience paralysis on one side of the face. The affected half of the face is weak, control over lacrimination (crying) and salivation is lost, and facial expression is abnormal. Possible causes include inflammation, ischemia, and nerve compression of the facial (seventh cranial) nerve.

This partial paralysis often remits in several months. The treatment goal is to restore muscle strength and facial symmetry. This is only possible if the paralysis remits.

Bell palsy seems ideally suited for Wolf's (1985) motor copy procedure. The clinician can show the patient an SEMG tracing of a muscle group (orbicularis oculus) on the healthy side of the face and then ask the patient to duplicate it using the corresponding muscle group on the affected side. The healthy muscle group provides a template to correct facial muscle tone.

The orbicularis oculi muscles, which surround the eyes, are shown below.


                                                   

               





Torticollis is a form of cervical dystonia characterized by a twisted neck and involuntary neck muscle contraction that results in abnormal head movements and postures.

Torticollis symptoms include head turning (80%), neck pain (50%), head shaking (50%), and abnormal posture (25%). Torticollis has multiple causes and is divided into two major categories: idiopathic torticollis (80-90%) and acquired torticollis (10-20%).

Idiopathic spasmodic torticollis (IST) is a neurodegenerative disease that destroys dopaminergic neurons in basal ganglia circuits, which disinhibits projections from the thalamus to the cortex and produces postural dystonia.



The incidence of IST in the United States is estimated at 3 per 10,000 cases. This disorder affects women more frequently than men (4.5 to 1). While IST is diagnosed in both children and adults, symptom onset for 90% of cases is between 31 and 60 years (Ross & Dufel, 2005).



Torticollis patients typically deviate the head to one side with slight neck flexion. Severe muscle contraction and spasm may be observed in the following muscles: sternocleidomastoid, levator scapulae, splenius capitis, and trapezius.

The sternocleidomastoid muscles have their origin in the sternum and clavicle, and insertion in the mastoid process of the temporal bone. Acting together, they flex the cervical vertebral column and extend the head. Acting individually, they laterally flex and rotate the head to the opposite side.

The levator scapulae muscle is located in the posterior neck and lies below the sternocleidomastoid and trapezius muscles. Its origin lies in the cervical vertebrae and insertion in the superior border of the scapula. The levator scapulae elevates the scapula and produces downward rotation.

The splenius capitis muscles (also called cervical muscles) originate in cervical and thoracic vertebrae and insert into the occipital bone and mastoid process of the temporal bone. Acting together, they extend the head. Acting individually, they laterally flex and rotate the head to the same side as the contracting muscle

The trapezius, the most superficial back muscle, is a triangular muscle sheet that covers the posterior neck and superior trunk. The two trapezius muscles form a trapezoid. The upper trapezius elevates the scapula and helps extend the head (Tortora & Derrickson, 2006).



Following medical evaluation, bilateral SEMG biofeedback can be provided within the context of physical therapy. The therapist should first determine which muscle groups are involved using bilateral SEMG assessment and then train the affected muscle groups.

For example, in spasmodic torticollis where a client's head is rotated to the left, a clinician may down train both the right sternocleidomastoid and left cervical (splenius capitis) muscles. Recall that sternocleidomastoid contraction rotates the head to the opposite side and cervical contraction rotates the head to the same side as the contracting muscle.

Each muscle group should be trained bilaterally, one at a time. Where there is significant asymmetry, the therapist should gradually down train the overactive muscle and up train its weak counterpart. Where spasticity is present, training should gradually reduce SEMG levels and then teach the patient to inhibit spasm in during passive stretching and functional activity.

Cleeland’s (1989) protocol uses four channels of SEMG to monitor the left and right sternocleidomastoid and trapezius groups. Cleeland reported a treatment protocol where patients receive two daily 45-minutes (Monday-Friday) for 2-3 weeks. After SEMG training allows a patient to return the head to midline, spasm is seen. A 3 to 5 mA shock (unpleasant, but not painful) is delivered to the first two fingers of either hand when spasm is detected. This often reduces the spasm.

The therapist emphasizes the patient’s active contribution to spasm reduction and motivates the patient to continue to work learning spasm control. After a patient shows spasm reduction (usually after 4-6 sessions), the patient is instructed to perform spasm reduction exercises in front of a mirror, twice a day, for 15 minutes each session. To increase control of spasm, once a patient has gained significant head control, he or she will be monitored with a portable SEMG device while walking and assigned hand-eye coordination tasks like reading or sewing.

After training, patients are treated on an outpatient basis on the following schedule: 2 sessions per day/2 days a week for the first month and 2 sessions per day/1 day a week for the second and third months.

Results for 52 patients who received combined SEMG biofeedback and contingent cutaneous shock showed that at 30.5 month follow-up:

Patients who had torticollis for less than 12 months showed the greatest gains: 70% (10) showed moderate to marked improvement.

Jahanshahi, Sartory, and Marsden (1991) reported a controlled outcome study in which 12 torticollis patients were randomly assigned to SEMG biofeedback for both sternocleidomastoids or relaxation training and graded neck exercises (RGP). The treatment conditions included 15 sessions of SEMG or RGP, 6 sessions of SEMG or RGP plus home-management, or home-management only. Follow-up was 3 months after the end of treatment.

The authors reported comparable SEMG reductions in both groups from pre- to posttreatment, feedback-specific neck muscle reduction in the SEMG biofeedback group, comparable improvements in head deviation and head range of motion, and that no patient was asymptomatic.






Now that you have completed this module, locate the muscles discussed above in a muscle atlas.




Adler, J. (March 8, 2004). The war on strokes. Newsweek, 42-48.

Basmajian, J. V. (Ed.). (1989). Biofeedback: Principles and practice for clinicians. New York: Williams & Wilkins.

Bolek, J. E. (2003). A preliminary study of modification of gait in real-time using surface electromyography. Applied Psychophysiology and Biofeedback, 28(2), 129-138.

Bolek, J. E. (2001), Moeller-Mansour, L., & Sabet, A. (2001). Enhanced proper sitting position using a new sEMG protocol, the "minimax" procedure, with Boolean logic. Applied Psychophysiology and Biofeedback, 26(1), 9-16.

Brucker, B. S., & Bulaeva, N. V. (1996). Biofeedback effect on electromyograph responses in patients with spinal cord injury. Archives of Physical Medical Rehabilitation, 77(2), 133-137.

Colbourne, G. R., Wright, F. V., & Naumann, S. (1994). Feedback of triceps surae EMG in gait of children with cerebral palsy: A controlled study. Archives of Physical Medicine Rehabilitation, 75(1), 40-45.

Dangond, F. (2005). Multiple sclerosis. eMedicine.

Freberg, L. A. (2006). Discovering biological psychology. Boston: Houghton Mifflin Company.

Glanz, M., Klawansky, S., Stason, W., Berkey, C., Shah, N., Phan, H., et al. (1995). Biofeedback therapy in posttroke rehabilitation: A meta-analysis of the randomized controlled trials. Archives of Physical Medical Rehabilitation, 76(6), 508-515.

Hartje, J. (1997) Biofeedback: Specific neuromuscular intervention (Module 4). Wheat Ridge: AAPB.

Jannett, T. C., Taub, E., & Groom, E. (2002). Microcomputer-based devices for feedback control of constraint-induced movement therapy in lower extremity stroke rehabilitation. Presentation to the Southeastern Symposium on Systems Theory, Nashville.

Knapp, H. D., Taub, E., & Berman, A. J. (1963). Movement in monkeys with deafferented forelimbs. Experimental Neurology, 7, 307-315.

Monnell, K., Zachariah, S. B., & Khoromi, S. (2005). Bell palsy. eMedicine.

Moreland, J. D., & Thomson, M. A. (1994). Efficacy of electromyographic biofeedback compared with conventional physical therapy for upper-extremity function in patients following stroke: A research overview and meta-analysis. Physical therapy, 74(6): 534-547.

Moreland, J. D., Thomson, M. A., & Fuoco, A. R. (1998). Electromyographic biofeedback to improve lower extremity function after stroke: A meta-analysis. Archives of Physical Medicine Rehabilitation, 79(2), 134-140.

Petrofsky, J. S. (2001). The use of electromyogram biofeedback to reduce trendelenburg gait. European Journal of Applied Physiology, 85(5), 491-495.

Pinel, P. J. (2006). Biopsychology (6th ed.). Boston: Allyn and Bacon.

Ross, M. J., & Dufel, S. (2005). Torticollis. eMedicine.

Salter, M. G., & Fern, S. (2005). NMDA receptors are expressed in developing oligodendrocyte processes and mediate injury. Nature, 438(22), 1167-1171.

M. S. Schwartz, & F. Andrasik (Eds.). (2003). Biofeedback: A practitioner's guide (3rd ed.). New York: The Guilford Press.

Schleenbaker, R. E., & Mainous, A. G. (1993). Electromyographic biofeedback for neuromuscular reeducation in the hemiplegic stroke patient: A meta-analysis. Archives of Physical Medicine Rehabilitation, 74(12), 1301-1304.

Shaffer, F., & Moss, D. (in press). Biofeedback.

Taub, E. (1994). Overcoming learned non-use: A new approach to treatment in physical medicine. In J. G. Carlson, A. R. Seifert, & N. Birbaumer (Eds.), Clinical applied psychophysiology (pp. 185-220). New York: Plenum.

Taub, E. (2005). Personal communication to Don Moss.

Taub, E., Uswatte, G., Elbert, T. (2002). An impending paradigm shift: The melding of basic research in neuroscience and behavioral science to produce new treatments in neurorehabilitation. Nature Reviews Neuroscience, 3, 226-236.

Taub, E., Uswatte, G., & Pidikiti, R. (1999). Constraint induced movement therapy: A new family of techniques with broad application to physical rehabilitation—A clinical review. Journal of Rehabilitation Research and Development, 36 (3), 237-251.

Thorogood, C., & Alexander, M. A. (2005). Cerebral palsy. eMedicine.

Toner, L. V., Cook, K., & Elder, G. C. (1998). Improved ankle function in children with cerebral palsy after computer-assisted motor learning. Developmental Medicine and Child Neurology, 40(12), 829-835.

Tortora, G. J., & Derrickson, B. H. (2006). Principles of anatomy and physiology (11th ed.). New York: John Wiley & Sons.

Wilson, J. (2003). Biological foundations of human behavior. Belmont, CA: Wadsworth/Thompson Learning.

Yucha, C. B., & Gilbert, C. D. (2004). Evidence-based practice in biofeedback and neurofeedback. Wheat Ridge: AAPB.