Impaired postural control

Improving proximal stability to enhance distal mobility has long been a tenet of occupational therapy interventions. Postural adjustments stabilize supporting body parts while other parts (e.g., the upper extremities) are being moved.⁷³ The following studies describe the effect of postural adjustments on arm function.

In the classic study by Belenkii, Gurfinkle, and Paltsev,¹⁶ unimpaired subjects who were evaluated while standing were asked to raise their arm to a horizontal position after they heard an external signal. Various EMG studies were performed to trace the pattern of muscle activation. The results demonstrated that the postural muscle synergies of the trunk and lower extremities were activated before (by 90 milliseconds) the anterior deltoid, the primary muscle used to perform this motion. The subjects then were evaluated in the supine position while performing the same task. No lower extremity activation was detected in this position (i.e., a different pattern of postural adjustments). The following conclusions can be inferred from this study:

Bouisset and Zattara²⁹ replicated the previous study and demonstrated that an upward and forward trunk movement resulting from spine and/or lower limb extension precedes upper limb movement. This movement pattern is familiar to therapists who cue their patients to focus on spinal extension and the associated anterior pelvic tilt while treating arm function.

Horak and colleagues⁸⁸ compared postural adjustments of subjects with and without hemiplegia during a variety of tasks with different parameters. The hemiplegic subjects demonstrated the same sequence of muscle activation as the subjects without hemiplegia, although activity on the hemiparetic side was delayed. In addition, the hemiparetic individuals were not capable of making rapid movements with the unimpaired arm. This was hypothesized to result from a delay in the anticipatory activity of the contralateral hemiplegic muscles. This study dispels the myth of “good” and “bad” sides after a stroke, especially when postural control is compromised.

In their study of postural adjustments during arm movements, Cordo and Nashner⁵² were able to demonstrate that when subjects’ postural stability was increased (e.g., by outside shoulder support or placing a finger lightly on a support rail), postural activity was reduced, and voluntary movement enhanced. This concept is crucial to understand when treating upper extremity dysfunction. As support is increased, the postural demands of the task are decreased and vice versa. The therapist can control the patient’s level of postural stability by manipulating the following treatment environment factors: positioning—supine to sitting to standing; type of support surface—stationary or unstable surfaces; positioning of objects used in activities—near or far, base of support, and amount of external stability.

Cordo and Nashner⁵² also made a critical distinction between associated postural adjustments that precede voluntary movements (e.g., reaching) and automatic postural adjustments that follow external perturbations (e.g., standing on a bus that stops at a light or being moved by the therapist). Training in one type of adjustment cannot be assumed to carry over into other types of adjustments.

Woollacott, Bonnet, and Yabe¹⁹⁴ demonstrated that their subjects’ postural activity varied depending on the task being performed (pushing, pulling) and whether they received information in advance regarding the goal of the task.

Massion¹¹⁹ points out that voluntary movements are “accompanied by postural adjustments which show three main characteristics: (1) they are ‘anticipatory’ with respect to movement and minimize the perturbations of posture and equilibrium due to the movement, (2) they are adaptable to the conditions in which the movement is executed, and (3) they are influenced by the instructions given to the subject concerning the task to be performed.”

Postural control disorders in stroke patients have been well-documented. Lee¹⁰⁹ emphasized the detrimental impact that postural dysfunction has on free arm movements and therefore ADL. Although a variety of muscles can serve as postural stabilizers, postural control of the trunk is critical for upper extremity function.¹⁹ See Chapter 7.

Occupational therapists must use their activity analysis skills to help patients develop the missing trunk control components. (See Table 7-9 for examples of the effects of object positioning on trunk control and weight shifting during reaching activities.) Functional mobility patterns requiring increased trunk control (e.g., scooting) should be incorporated into treatment plans for upper extremity function. See Chapter 14.

Postural control evaluations should be performed within the context of upper extremity tasks such as reaching or performing ADL and IADL. Evaluating postural control separately does not provide the therapist with sufficient information for intervention. (See Chapters 7 and 8 for more information related to postural control.)

Weakness

Until relatively recently, the impact of weakness (a negative symptom) on stroke patients’ functional status has long been ignored. The motor control deficits in patients previously were attributed exclusively to spasticity, which resulted in treatment focused on inhibiting the spasticity. Many therapists considered upper extremity muscle tests for strength difficult to interpret because of common “synergy patterns.” Bourbonnais and colleagues³¹ demonstrated that the patterns of activity in the elbow flexor muscles were not consistent with established synergistic patterns. Weakness of the upper extremity musculature plays a major role in upper extremity dysfunction, most likely more than the positive symptoms after stroke. Muscle weakness is reflected by the inability of patients to generate normal levels of muscle force.³⁰ Stroke survivors who have written about their experiences focus on the difficulty in force production. Brodal³⁵ reflected on his own stroke: “It was a striking and repeatedly made observation that the force needed to make a severely paretic muscle contract is considerable. . . . Subjectively this is experienced as a kind of mental force, a power of will. In the case of a muscle just capable of being actively moved the mental effort needed was very great.”

Bourbonnais and Vanden Noven³⁰ reviewed the physiological changes in the nervous system that contribute to muscle weakness in patients with hemiparesis. They summarized specific changes at the motor neuron and muscle levels that decrease a patient’s ability to produce force. Box 10-7 summarizes these changes.

Bohannon and colleagues²¹ found that static strength deficits of the shoulder medial rotator and elbow flexor muscles did not correlate with antagonist muscle spasticity. They concluded that therapists might determine the capacity for force production for an agonist muscle based on its own tone rather than that of its antagonist.

Gowland and colleagues⁷⁹ studied agonist and antagonist activity during upper limb movements in stroke patients and concluded that treatment should be aimed at improving motor neuron recruitment rather than reducing antagonist activity. In their study, patients who could not perform select upper extremity tasks had EMG values significantly and consistently lower than those of patients who were successful at the task.

Indeed recent empirical evidence highlights the relationship between weakness and loss of function. Findings include:

From a treatment planning perspective, integrating strengthening interventions is imperative in efforts to regain limb function. Bohannon and Smith²³ analyzed strength deficits in stroke patients and verified that muscle strength improves in stroke patients with hemiplegia who are undergoing rehabilitation. Empirical evidence supports the use of strengthening interventions in this population without deleterious effects:

The debate about which type of muscle contraction (eccentric, concentric, or isometric) is the most effective in strengthening patients has been long-standing. Muscle groups need to contract in a variety of ways to complete functional tasks successfully. For example, when a person reaches for a can of soup on a high shelf, the shoulder musculature must contract (concentrically) to bring the hand to the level of the shelf, maintain the contraction (isometrically) to locate the correct item, and control the weight of the arm and item in gravity (eccentrically) as the can is placed with control on the countertop.

In a study of dynamical muscle strength training in stroke patients, Engardt and colleagues⁶⁵ found that eccentric contractions were more effective than concentric contractions. Twenty patients with hemiparesis resulting from strokes participated in activities that elicited concentric or eccentric contractions. After the treatment, significant improvements resulted in the relative strength of paretic muscles during eccentric and concentric actions in the group that was trained solely with eccentric contractions (i.e., eccentric training increased the strength of both types of contractions); this was not true for the group that only received concentric contraction training. Therefore, the authors determined eccentric contraction training to be more advantageous and efficient (Box 10-8). See Table 10-1 for a review of evidence-based interventions related to strengthening.

Spasticity

Spasticity, which is a positive symptom according to Jackson classification system, has been a subject of debate by various authors. Although an abundance of research has been done on spasticity, disagreements still exist about its definition, physiological basis, treatment, and evaluation. Glenn and Whyte⁷⁵ define spasticity as “a motor disorder with persistent increase in the involuntary reflex activity of a muscle in response to stretch. Four specific phenomena may be variably observed in the constellation of spasticity: hypertonia (frequently velocity dependent and demonstrating the clasp-knife phenomenon), hyperactive (phasic) deep tendon reflexes, clonus, and spread of reflex responses beyond the muscle stimulated.” In addition, Babinski sign is characteristic, and hyperactive tonic neck or vestibular reflexes may be present.¹¹⁶

Several different phenomena commonly observed in stroke rehabilitation including hyperactive stretch reflexes, increased resistance to passive movement, posturing of the extremities, excessive cocontraction, and stereotypical movement synergies are clumped together in the category of spasticity. Spasticity has become a catchall term for a variety of problems. Rather than being a specific symptom, spasticity is related to a variety of neural and nonneural factors. Therefore, spasticity cannot be treated uniformly by surgical, physical, or pharmacological procedures. Spastic paresis is a commonly used term that implies a cause-and-effect relationship (i.e., a cause-and-effect relationship between positive and negative symptoms). This belief has been challenged recently.

Preston and Hecht¹⁴⁷ provide further information regarding the clinical presentation of spasticity to include the following:

Bobath¹⁸ stated that there is “An intimate relationship between spasticity and movement . . . spasticity must be held responsible for much of the patient’s motor deficit.” Treatment techniques were based on “helping the patient gain control over the released patterns of spasticity by their inhibition.” Patients were treated under the assumption that “Weakness of muscles may not be real, but relative to the opposition by spastic antagonists.” A variety of studies have been published that refute these assumptions. See Chapter 6.

Sahrmann and Norton¹⁵⁹ studied normal subjects and subjects with upper motor neuron symptoms. The movement pattern studied was alternating flexion and extension of the elbow. The analysis of their EMG findings showed that the primary cause of impaired movement was not antagonist stretch reflexes but was limited and prolonged agonist contraction recruitment and delayed cessation of agonist contractions after movement had stopped. Rather than focusing treatment on inhibiting spasticity, therapists should train patients to perform alternating movement patterns (e.g., hand-to-mouth patterns) efficiently.

Fellows, Kaus, and Thilmann⁶⁷ studied the importance of hyperreflexia and paresis on voluntary arm movements in normal subjects and subjects with spasticity resulting from a unilateral ischemic cerebral lesion. The subjects with spasticity showed a lower maximum movement velocity; the more marked the paresis, the greater the reduction in maximum velocity. No relationship was found between the degree of voluntary movement impairment and level of passive muscle hypertonia in the antagonist. The conclusion was that agonist muscle paresis, rather than antagonist muscle hypertonia, had the most significant effect on impaired voluntary movement.

In their study on overcoming limited elbow movement in the presence of antagonist hyperactivity, Wolf and colleagues¹⁹⁰ concluded that functional elbow improvements could be made without first training the patient specifically to inhibit hyperactivity.

Landau¹⁰⁸ performed pharmacological interventions that effectively abolished the hyperactive stretch reflexes in his patients. This intervention did not result in a corresponding improvement in motor behavior.

Ada, O’Dwyer, and O’Neill⁶ examined the relationship between the motor impairments (spasticity and weakness) and their impact on physical activity. They specifically aimed to study the contribution of weakness and spasticity to contracture, and the contribution of all three impairments to limitations in physical activity during the first 12 months after stroke. The authors followed 27 stroke survivors for one year. They found that “the major independent contributors to contracture were spasticity for the first four months after stroke (p = 0.0001-0.10) and weakness thereafter (p = 0.01-0.05). However, the major and only independent contributor to limitations in physical activity throughout the year was weakness (p = 0.0001-0.05).” “For the first time, from a longitudinal study, the findings show that spasticity can cause contracture after stroke, consistent with the prevailing clinical view. However, weakness is the main contributor to activity limitations.”

In the traditional evaluation of spasticity, the therapist moves the patient’s limb quickly in a direction opposite to the pull of the muscle group being tested, and the examiner feels for a resistance to the movement. The gold standard for rating resistance is the Ashworth Scale⁹ or the Modified Ashworth Scale²⁴ (Box 10-9).

The response of a spastic muscle to stretch has been argued not to be the same during passive and active movement. In addition, spasticity is a multidimensional problem that incorporates neural and nonneural components (e.g., altered soft-tissue compliance). Therefore, some authors have questioned the usefulness of test measures such as the Ashworth Scale and are investigating a more comprehensive evaluation of spasticity.

Although the research on spasticity does not support focusing treatment on suppressing stretch reflexes, it does support treatment focusing on preventing secondary structural muscle changes in patients with spasticity.

Hufschmidt and Mauritz’s study⁹⁰ suggested that spastic contracture is the result of degenerative changes (e.g., atrophy and fibrosis) and changes of the passive and contractile muscle properties.

In their study on spastic and rigid muscles, Dietz, Quintern, and Berger⁶¹ concluded that the actual muscle fibers undergo changes, which explains the increased muscle tone in spastic patients.

For treating patients with spasticity, Perry¹⁴² emphasized early mobilization and assistance with developing evolving motor control into effective function. These two interventions result in minimal contractures and prevent improper use of patients’ available control mechanisms. Hummelsheim and colleagues⁹¹ studied the results of sustained stretch in spastic patients. They found that sustained muscle stretch of approximately 10 minutes led to significant reduction in the spastic hypertonus in the elbow, hand, and finger flexors. They hypothesized that this benefit is due to stretch receptor fatigue or adaptation to the new extended position.

Little and Massagli¹¹⁶ also emphasized using a stretching program incorporating pain prevention and patient education focusing on the adverse effects of spasticity (contracture), use of slow movements, and importance of daily stretching.

In addition to the mentioned techniques, specific modalities and their physiological bases have been described in the literature and include local cooling, vibration therapy, and electrical stimulation.

Perry¹⁴² summarized the effective rehabilitation of a patient with spasticity by using five categories: contracture minimization, realistic planning, muscle strength preservation and restoration, enhancement of returning control, and substitution for permanent functional loss.

Carr, Shepherd, and Ada⁴³ summarized their treatment approach based on the assumption that clinical spasticity is a manifestation of length-associated muscle changes and disordered motor control: “The development of spasticity will be less severe if soft-tissue length can be maintained and if motor training emphasizes elimination of unnecessary muscle force and training muscle synergies as part of specific actions.”

Again, the point must be emphasized that many of the observed phenomena that occur during treatment should not be attributed automatically to spasticity and require more in-depth evaluations and treatment plans (Box 10-10; Table 10-6).

Table 10-6 Suggested Interventions for Problems Commonly Thought to Be Caused by Spasticity*

* This table represents a variety of functional limitations and problems traditionally considered to be the direct result of spasticity. Although sometimes interconnected, these problems stem from different sources and must be treated accordingly.

Preston and Hecht¹⁴⁷ have comprehensively reviewed the literature related to spasticity management, including topics such as oral and intrathecal medications, nerve blocks, orthopedic surgery, and neurosurgical interventions.

Nerve blocks are being used increasingly as an adjunct therapy in the rehabilitation process. Preston and Hecht¹⁴⁷ differentiated between short-term blocks such as procaine and bupivacaine used to diagnose and assist in the evaluation process and long-term blocks such as phenol and botulinum toxin type A (Botox).

Rousseaux, Kozlowski, and Froger¹⁵⁴ assessed the efficacy of Botox treatment on disability, especially in manual activities, and attempted to identify predictive factors of improvement in 20 patients with stroke. They concluded that botulinum toxin A is efficient in improving hand use in patients with relatively preserved distal motricity and in increasing comfort in patients with severe global disorders. Similarly, Bakheit and colleagues¹¹ completed a randomized controlled trial to assess the efficacy of Botox in decreasing spasticity in stroke survivors. They concluded that treatment with Botox reduced muscle tone in patients with poststroke upper limb spasticity.

While the positive effect at the impairment level (i.e., reduction of spasticity) has been well-documented, the effect on functional limitations is not as clear.^164,183 As spasticity increases, the risk for soft-tissue shortening is heightened, which in fact may lead to a vicious circle of problems such as spasticity, soft-tissue shortening, overrecruitment of shortened muscles, and increased stretch reflexes. Secondary problems that may occur if the spasticity is not managed in a therapy program include the following:

In summary, although reducing spasticity does not appear to result in automatic improvements related to function, therapists must manage spasticity to prevent soft-tissue contracture, prevent deformity, and maintain a flexible and mobile arm.

Loss of soft-tissue elasticity (contractures and deformities)

Contracture in stroke patients results from immobilization and may be attributed to spasticity (particularly during the first four months after stroke) and weakness thereafter,⁶ improper positioning, postural malalignment, a lack of variation in limb postures (e.g., prolonged sling use), or a combination of various factors. The formation of contractures indicates a poor prognosis for limb function. Perry¹⁴² discussed the vicious circle of contracture and spasticity: “contractures stiffen tissues, immobility creates contractures. Spasticity preserves the contracture by excluding the intramuscular fibrous tissues from the stretching force.”

Botte, Nickel, and Akeson²⁸ have reviewed the literature correlating spasticity and contracture. As the stroke patient progresses to a state of spasticity, the increased activity of the spastic muscles may result in characteristic posturing of the limb, resulting in increased stiffness of the soft-tissue surrounding the joint and the eventual formation of fixed contracture. The authors further pointed out that contracture is associated with loss of elasticity and fixed shortening of involved tissues. Contracture may occur in a variety of soft-tissues including the following: skin, subcutaneous tissue, muscle, tendon, ligament, joint capsule, vessels, and nerves.

Halar and Bell⁸² categorized contracture as arthrogenic (resulting from cartilage damage, joint incongruency, or capsular fibrosis), soft-tissue related (skin, tendons, ligaments, subcutaneous tissue), and myogenic (shortening of the muscle by intrinsic or extrinsic factors). The therapist must consider the difference between myogenic and joint contracture, especially if the muscle spans two or more joints (e.g., the wrist and hand). The therapist can differentiate contractures by flexing the proximal joint and noting the resulting position of the distal joints. Joint contracture is not affected by changes in proximal joint position. See Chapter 13.

Booth²⁷ reviewed the physiological and biochemical effects of immobilization on muscle. His findings indicate that muscle strength rapidly declines during limb immobilization because of a decrease in muscle size; muscle fatigability increases rapidly after immobilization. His observations also indicate that muscle atrophy in immobilized limbs begins rapidly, and a decrease in muscle size is greatest in the early phases of immobilization.

Passive range of motion.

Soft-tissue and joint mobilization are the treatments of choice for preventing contracture. The benefits of mobilization include maintenance of joint lubrication,²⁸ prevention of secondary orthopedic problems (impingements), maintenance of soft-tissue length, and possible reduction of spasticity by acting on the nonneural components of spasticity.

Contracture is prevented by deliberate and frequent limb movement, with active movement being preferred over passive when possible. Perry¹⁴² pointed out that it is essential to move the patient through complete ROM and not just the middle ranges. Therapists must determine what a full ROM is for each patient and must consider age-related factors. Determining the full ROM on the less affected side may be helpful. A joint that moves or is moved through its full ROM several times daily develops almost no deformities. Although the therapist should maintain the patient’s ability to participate in all ranges of trunk and upper extremity activities, the therapist should pay particular attention to the following ranges:

The mobility of the scapula on the thoracic wall with emphasis on protraction and upward rotation should be maintained because this range is critical in the prevention of soft-tissue impingement in the subacromial space during overhead movements of the arm and in preparation for forward reach patterns. Overhead ranges should not be attempted unless the scapula is freely gliding in upward rotation.

Maintaining external (lateral) rotation of the glenohumeral joint allows abduction of the arm as the humerus rotates laterally to permit the greater tuberosity of the humerus to clear the acromial process. Bohannon and colleagues,²² Ikai and colleagues,⁹³ and Zorowitz and colleagues²⁰³ concluded that loss of external rotation ROM was the factor most significantly correlated to shoulder pain.

Elbow extension is important because the majority of stroke patients favor elbow flexion as a rest posture.

The therapists also should maintain wrist extension with concurrent radial deviation. During wrist ROM exercises, therapists must realize that the range of wrist deviation is at a maximum when the wrist is slightly flexed and a minimum when the wrist is fully flexed. Wrist extension is at a maximum during neutral deviation and a minimum during ulnar deviation.¹⁰⁰

Composite flexion of the digits leads to collateral ligament elongation. Therapists must maintain this length to prevent deformity and prepare the hand for return of motor function.

Composite extension of the wrist and digits results in long flexor elongation.

Digits ranged in intrinsic plus (metacarpophalangeal flexion and interphalangeal extension) and intrinsic minus (metacarpophalangeal extension and interphalangeal flexion).

Halar and Bell⁸² recommended active ROM and passive ROM combined with a terminal stretch at least twice per day if contracture is beginning to develop. Therapists must use low-load prolonged stretch if a contracture has developed (see Chapter 13). During the terminal stretch, the therapist should stabilize the proximal body part well. The therapist may distract the joint slightly during the stretch to prevent soft-tissue impingement. The therapist must monitor scapula position during passive ROM activities. If necessary, the therapist should support the scapula in a position of protraction and upward rotation. In addition, the therapist must support the humerus in an external rotation position. The elbow crease should be facing up (not medially toward the trunk) to ensure proper alignment (Fig. 10-12).

Figure 10-12 Passive range of motion activities with strict attention to the biomechanical alignment of the scapulothoracic and glenohumeral joints. Therapist’s right hand assists with mobilization (upward rotation) of the scapula, while left arm keeps humerus externally rotated.

Positioning.

Positioning is another effective means of maintaining soft-tissue length and can be used to promote low-load, prolonged stretch. Therapists must address positioning needs of patients while they are in bed or wheelchairs/armchairs (see Chapter 26) and anytime they are in a recumbent position. Effective positioning encourages proper joint alignment, variations in joint position, comfort, and the maintenance of stretch in areas at risk for contracture. Common areas of concern during patient positioning include head and neck alignment, trunk alignment, glenohumeral joint alignment, scapula alignment, maintenance of abduction, external rotation, elbow extension, and maintenance of long flexor length.

A thorough literature review comparing authors’ strategies on bed positioning has been published.⁴¹ This review found no consensus on some issues and multiple discrepancies on strategies. Many of the positioning protocols are based on the principle of inhibiting primitive reflexes, a topic of considerable debate.

Patients are engaged in therapy only a portion of the day. Studies have shown that patients in rehabilitation units spend almost half of their days engaged in passive pursuits including sitting unoccupied and lying in bed.¹⁵ Therefore, patients at risk for developing contracture because of limb immobilization are good candidates for participation in a positioning program in addition to therapy.

The positioning suggestions in Box 10-11 are based on Carr and Kenney’s review⁴¹ of the positioning literature and highlights the consensus of reviewed authors.

Box 10-11 Suggested Bed Positioning

POSITIONING OF PATIENT
On unaffected side	Head/neck: neutral and symmetrical Affected upper limb: protracted and forward on pillow-wrist neutral, fingers extended, and thumb abducted Trunk: aligned Affected lower limb: hip forward, flexed, and supported; knee forward, flexed, and supported
On affected side	Head/neck: neutral and symmetrical Affected upper limb: protracted forward with elbow extended, hand supinated, wrist neutral, fingers extended, and thumb abducted Trunk: straight and aligned Affected lower limb: knee flexed Unaffected lower limb: knee flexed and supported by pillows
In supine	Head/neck: slight flexion Affected upper limb: protracted and slightly abducted with external rotation with wrist neutral and fingers extended Trunk: straight and aligned
	Affected lower limb: hip forward on pillow; nothing against soles of the feet

Although the positioning suggestions in Box 10-11 represent the consensus of many authors, major areas of intervention are missing, which result in the controversies surrounding this area of intervention. For example, glenohumeral joint support remains controversial. Although most authors agree that the scapula should be protracted with a pillow, no consensus exists about support of the humerus. If only the scapula is protracted with a pillow, the humerus takes on a position of relative extension. Therefore, only support of both the scapula and humerus achieves the original goal of proper joint alignment (Fig. 10-13).

Figure 10-13 A, Bed positioning with only the scapula supported. The humerus takes on a position of relative extension, with the head of the humerus migrating anteriorly. B, Proper support of scapula and humerus ensures proper biomechanical alignment of shoulder joint.

At this point, no definitive studies support one type of positioning more than another with few exceptions. Ada and colleagues⁵ determined that positioning patients in supine with the affected shoulder abducted to 45-degrees and the elbow flexed to 90-degrees and placed in maximum comfortable external rotation, with towels or pillows to support the forearm 30 minutes per day prevented contracture of the internal rotators. Occupational therapists must decide what their intervention goals are and critically analyze their effectiveness. Therapists should not use general, generic strategies for bed positioning; instead, they should evaluate each patient’s positioning needs individually.

Patient management of the extremity.

Strategies to teach patients safe ROM activities they can perform themselves need to be initiated as soon as patients are medically stable. Although the clasped-hand position followed by overhead movements of both extremities has been advocated by some authors, this position may not be the most effective, especially for trauma prevention. This movement pattern does not account for factors such as scapula-humeral rhythm (especially if weakness, malalignment, or tightness around the scapula exists), overzealous patients who do not or cannot respect their pain, or critical shoulder biomechanics. Many patients observed performing this type of ROM activity have their trunk hyperextended, scapula retracted, and humerus internally rotated. This type of alignment does not correspond with an ROM pattern that emphasizes forward flexion of the humerus; it promotes proximal patterns (e.g., retraction) that should be discouraged (Fig. 10-14). Recommended techniques for patients performing ROM activities by themselves safely include the following:

1. “Towel on table”: The patient is seated at a table with both arms on top of a towel. The less affected arm guides the towel around the table, with the majority of movement occurring in the trunk and from hip flexion. The patient’s goal is to “polish the table” while holding positions at the end of desired ROM. The farther the patient’s chair is positioned from the table, the greater the ROM. This technique not only enhances the range of the glenohumeral and elbow joint but also encourages scapula protraction and weight-shifting. Excessive effort is minimized because the towel assists the movement (Fig. 10-15).

2. “Rock the baby”: The patient’s less affected arm cradles the more affected arm, lifts it to 90 degrees, and places it into positions of horizontal abduction and adduction. Increased horizontal adduction on the more affected side encourages scapula protraction. This technique also encourages trunk rotation (Fig. 10-16).

3. While seated or standing, the patient reaches down to the floor and allows both arms to dangle. This position encourages extension of the elbow, wrist, and digits and forward flexion of the humerus with scapula protraction. The activity is an especially useful technique for patients after they have performed an excessively difficult activity (e.g., gait, transfer, or dressing) that results in stereotypical arm posturing (Fig. 10-17).

4. While seated or standing, the patient places the more affected extremity onto a table or counter so that the forearm is bearing the weight. With the extremity in this position, the patient turns the trunk away from the supported extremity. As the trunk turns farther away and is enhanced by the posterior reach of the less affected arm, the external rotation of the more affected shoulder increases (Fig. 10-18).

5. Davis⁵⁷ has advocated rolling over the protracted scapula (from supine to side lying) several times to mobilize the scapula.

6. If the scapula of a patient is mobile and stays mobile, the range of abduction and external rotation may be increased by having the patient lie supine, placing the hands behind the head, and allowing the elbows to fall toward the bed (Fig. 10-19). This is a common resting position for an individual who has unimpaired upper extremity function. This technique should be used judiciously and only for patients who move slowly, respect pain, and have a mobile scapula. Therapists may use the five techniques outlined previously for almost all patients because they inherently follow biomechanical principles.

7. Avoid the use of overhead pulleys.¹⁰⁴

Figure 10-14 Because of multiple biomechanical concerns (e.g., impingement), self-overhead range of motion is discouraged.

Figure 10-15 “Towel on table.” Therapist is training patient to perform safe self range-of-motion activity. As the patient pushes the towel toward bottle, range of motion is gained in humeral flexion, scapular protraction, and elbow extension (which are ranges required for functional reach). Much of the range is gained by hip and trunk flexion.

Figure 10-16 “Rock the baby.” The patient lifts upper extremity to chest level (A) and abducts (B) and adducts (C) horizontally, allowing trunk rotation.

Figure 10-17 Patient performs self range-of-motion activity by reaching to floor. This pattern is especially effective after a difficult task that results in stereotypical posturing.

Figure 10-18 External rotation of the left glenohumeral joint is achieved by reaching to side and behind with opposite arm.

Figure 10-19 Internal rotators stretch to be used judiciously for patients who respect their own pain. This rest posture is effective at maintaining external rotation and abduction of the glenohumeral joint. If range is lacking, the humerus can be supported with a towel until patient gains increased external rotation and horizontal abduction.

The ultimate strategy used to decrease contracture and maintain ROM is encouraging functional use of the trunk and upper extremity. A person who has never had a stroke maintains ROM of an extremity by incorporating it into ADL. Activities that eliminate maladaptive positions during activities, improve balanced muscle activity on both sides of the joints, and focus on activities that encourage ROMs that are commonly decreased in stroke patients (e.g., external rotation, forward flexion, abduction, and protraction) should be incorporated into a comprehensive upper extremity program. (See Chapter 13 for other adjunct treatments to prevent or correct soft-tissue shortening.)

Superimposed orthopedic injuries

Orthopedic problems associated with stroke have been well-documented. These complications have a negative impact on functional outcomes, prolong rehabilitation, and are one of the main causes of upper extremity pain syndromes after stroke. Indeed a recent magnetic resonance imaging (MRI) study of 89 chronic stroke survivors documented:¹⁶²

Rotator cuff and biceps tendon lesions.

The rotator cuff guides and leads the movements of the shoulder joint. The cuff supplies the strength needed to complete the ROM in the shoulder joint and seats the head of the humerus into the glenoid fossa.

Najenson, Yacubovic, and Pikielni¹³¹ studied 32 hemiplegic patients with severe upper limb paralysis; 18 patients served as controls by having their less affected side evaluated. Forty percent of the patients had a rotator cuff tear on the affected side. None of the patients had complaints about the affected shoulder before the stroke. Only 16% of the patients in the control group had ruptured rotator cuffs on the less involved side; all three seemed to be long-standing tears.

Najenson, Yacubovic, and Pikielni¹³¹ also discussed the pathophysiology of a rotator cuff tear in hemiplegic patients. Many older patients are predisposed to rotator cuff ruptures because of degenerative changes associated with aging. Cuff tears commonly result from impingement of the cuff between the greater tuberosity and acromial arch (Fig. 10-20), which occurs when the humerus is forced into abduction without external rotation (e.g., during inappropriate passive ROM activities or activities that are not sensitive to shoulder biomechanics [reciprocal pulleys]). Therapists who have a thorough understanding of joint alignment can prevent impingement during treatment.

Figure 10-20 Impingement of soft tissues located in subacromial space. Impingement occurs between the head of humerus and acromion/coracoid. Impingement occurs during forced humeral flexion/abduction without concurrent upward rotation of scapula and/or external rotation of humerus.

Nepomuceno and Miller¹³⁴ found seven rotator cuff tears and one a transverse bicipital tendon tear in 24 subjects with painful hemiplegic shoulders. None of the patients had premorbid pathological conditions of the shoulder. With one exception, all patients with soft-tissue lesions had left-sided hemiplegia. (This study did not evaluate the presence of visual field loss or neglect.)

Therapists should note that a relationship between rotator cuff age and wear has been documented. After age 50-years-old, the percentage of lesions significantly increases.

Pain syndromes

Although pain syndromes have been discussed previously in the context of orthopedic injuries and SHS, their impact on functional recovery is significant, so this section specifically reviews the literature on hemiplegic shoulder pain.

The incidence of shoulder pain in hemiplegic patients has been reported to be as high as 72%.^22,155,184 Roy, Sands, and Hill¹⁵⁵ identified strong associations between hemiplegic shoulder pain and prolonged hospital stays, arm weakness, poor recovery of arm function, ADL, and lower rates of discharge to the home. Those responsible for stroke patients have the onus to be aware of hemiplegic shoulder pain and to diagnose, relieve, and prevent this syndrome. Although shoulder pain is obviously not the only variable leading to prolonged hospital stays, it is a potentially preventable variable over which occupational therapists have much control.

Pain can limit patient’s activities, such as rolling in bed, transferring, putting on a shirt or blouse, and bending to reach the feet to put on shoes and socks. The occurrence of shoulder pain also has been linked to depression.¹⁶⁰

The literature concerning hemiplegic shoulder pain is confusing at times and often contradictory. The following review was obtained from a selection of articles from a variety of disciplines. The focus of the review is clinical correlations associated with hemiplegic shoulder pain.

In their study of 55 patients, Roy, Sands, and Hill¹⁵⁵ found positive correlations between hemiplegic shoulder pain and “glenohumeral malalignment without descent of the humeral head” and between hemiplegic shoulder pain and reflex sympathetic dystrophy (SHS). The study did not confirm a strong association between spasticity (measured by the Ashworth Scale) and hemiplegic shoulder pain.

Joynt⁹⁹ found significant correlations between loss of motion and shoulder pain and questioned the relationship between neglect/perceptual dysfunction and pain. His left-sided hemiplegic subjects had a higher incidence of shoulder pain, which led him to question whether the incidence of trauma was increased. He found no correlation between shoulder pain and subluxation, spasticity, strength, or sensation.

Joynt⁹⁹ identified the subacromial area as a pain-producing location in a significant number of cases. Of 28 patients who received a subacromial injection of 1% lidocaine, more than half obtained moderate or significant pain relief and improved ROM. The author suggested that physical agent modalities, steroid injections, and careful ROM activities focusing on impingement prevention were significant in reducing pain.

The subacromial area is prone to trauma during therapy and patient handling. The subacromial space includes the supraspinatus tendon, long head of the biceps, and subacromial bursa⁴⁰ (Fig. 10-21). All of these structures are prone to impingement and inflammation. Structure impingement can develop easily in hemiplegic patients during ROM activities because the normal scapulohumeral rhythm becomes impaired. If the scapula is not rotated upward (by therapist’s manipulation or active control), the humerus becomes blocked by the acromion and causes impingement, inflammation, and pain (see Fig. 10-20). Combined motions of scapula retraction with forward flexion should be avoided to prevent impingement. Instead, the scapula should glide freely and be protracted and upwardly rotated during upper extremity activities. Objects for reaching activities should be placed in front or below waist level of the patient to encourage humeral forward flexion with scapular protraction. Indeed, Dromerick and colleagues⁶⁴ designed a study to clarify the pathophysiology of hemiplegic shoulder pain by determining the frequency of abnormal shoulder physical diagnosis signs and the accuracy of self-report. They found:

Figure 10-21 Subacromial space.

Some patients may develop inflammation around the biceps tendon and supraspinatus insertion because of impingements. Palpation skills are important for determining which structures are involved (Fig. 10-22). To palpate the biceps tendon, the therapist palpates the acromion and drops one finger to the anterior shoulder; the biceps tendon lies in the groove between the greater and lesser tuberosities of the humerus. If the patient feels pain on application of pressure, the biceps tendon probably has been affected. (Passively rotating the humerus while palpating assists the therapist with locating the tuberosities.)

Figure 10-22 Palpation point. The x on left anterior) is palpation point for long head of biceps. The x on right (more lateral) is palpation point for supraspinatus tendon.

To palpate the supraspinatus tendon, the therapist palpates the acromion, but this time drops one finger to the lateral shoulder right below the center of the acromion. If pressure or slight friction elicits pain, the supraspinatus most likely has been affected.

Bohannon and colleagues²² studied the relationship of five variables (age, time since onset of hemiplegia, range of external rotation of the hemiplegic shoulder, spasticity, and weakness) to shoulder pain. In their study of 50 patients, 36 had shoulder pain. Range of shoulder external rotation was considered the factor related most significantly to shoulder pain. They hypothesized that hemiplegic shoulder pain was in part a manifestation of adhesive capsulitis. In this study, only patients with full external rotation were free of pain. The suggested treatment was elimination of inflammation and maintenance of ROM.

Hecht⁸⁶ treated 13 patients with limited ROM and shoulder pain with percutaneous phenol blocks to the nerves of the subscapularis (a major shoulder internal rotator). Immediate and significant improvements were observed in the flexion, abduction, and external rotation ROMs; pain relief also was noted. This study indicates that the subscapularis is a key muscle and should be addressed during treatment focusing on maintaining soft-tissue length. The subscapularis muscle may tighten in patients with the previously mentioned pain syndrome. If the humerus resists external rotation with the arm at the side during evaluation, the therapist can presume the subscapularis to be a factor contributing to the deformity. Similarly, subscapularis injection of botulinum toxin A appears to be of value in the management of shoulder pain in spastic hemiplegic patients.²⁰⁰ These studies adds more support to the concept of focusing on maintaining the range of humeral external rotation to prevent resulting complications.

Bohannon and Andrews²⁰ studied 24 patients in an effort to establish a relationship between subluxation and pain. Despite the emphasis placed on reduction of subluxation, the relationship between shoulder pain and subluxation has not been established. Their study did not find an association between shoulder pain and subluxation (which was defined in this study as the separation between the acromion and the humeral head). A study by Arsenault and colleagues⁸ also found no significant relationship between subluxation and shoulder pain.

A more recent study by Zorowitz and colleagues²⁰³ also focused on the correlation between subluxation and pain. Results showed that shoulder pain did not correlate with age, vertical or horizontal subluxation, shoulder flexion, abduction, or Fugl-Meyer Assessment scores, but it did correlate with the degree of shoulder external rotation. Wanklyn, Forster, and Young¹⁸⁶ also found an association between reduced external rotation and hemiplegic shoulder pain, with an incidence as high as 66%. This association was believed to be due to abnormal muscle tone or structural changes, namely adhesions. Similarly, Ikai and others⁹³ evaluated 75 subjects and found no correlation between subluxation and pain.

Kumar and colleagues¹⁰⁴ demonstrated a positive correlation between shoulder pain and therapy programs that did not consider biomechanical shoulder alignment during treatment. Patients were assigned to one of three exercise groups: ROM initiated by the therapist, skateboard treatment, and overhead pulley treatment. Of the patients who developed pain during the treatment programs, 8% were in the ROM group, 12% in the skateboard group, and 62% in the overhead pulley group. The probable cause of this discrepancy was soft-tissue damage resulting from forced abduction without external rotation. This study showed that poorly prescribed activities by the therapist could be the cause of pain syndromes. This study found no significant relationship between subluxation and pain.

In a three-year study of 219 hemiplegic patients, Van Ouwenaller, Laplace, and Chantraine¹⁸⁴ found that 85% of the patients who developed pain had spasticity (an increased myotactic reflex) compared with 18% of flaccid patients. They also found that 50% of the patients who developed pain had anteroinferior subluxations (which were not defined). The authors advocated use of muscle relaxation techniques for the shoulder girdle.

Jensen⁹⁸ attributed shoulder pain to traumatic tendinitis resulting from unskilled and strenuous joint treatment during ADL (e.g., bathing, dressing, and bed mobility) and bilateral ROM activities of more than 90 degrees resulting in “jamming [of] soft-tissue against the acromion resulting in lesions” (see Fig. 10-20). Jensen suggested the following precautions: educating all staff members, placing signs over patients’ beds to warn staff of the shoulder instability, supporting the arm during the acute stage, avoiding treatment that may cause soft-tissue impingement, having a thorough understanding of shoulder anatomy, and dissuading use of pulley exercises and self-ROM activities.

Lastly, Wanklyn, Forster, and Young¹⁸⁶ found a 27% increased incidence of shoulder pain in dependent patients after discharge, which may reflect improper handling at home by caregivers. They suggested a greater emphasis on patient and caregiver education regarding proper transfer techniques and correct handling of the hemiplegic arm (Box 10-14).

Loss of biomechanical alignment

Immediately after a stroke, patients lose their ability to maintain upright control and become malaligned because of the effects of gravity, weakness, and muscle imbalance.

Occupational therapists must be able to identify malalignments to treat upper extremity dysfunction effectively. The following section discusses common trunk and upper extremity alignment problems and reviews activities to counteract the adverse effects of malalignment.

Loss of pelvic/trunk alignment.

After a stroke, patients commonly lose their ability to perform postural adjustments and maintain postural alignment because of weakness, a loss of equilibrium, and righting reactions; the trunk assumes an asymmetrical posture.^18,19,55

The first area to observe is the patient’s pelvis and its effect on spinal alignment. Patients typically bear weight asymmetrically through their pelvis (by one ischial tuberosity accepting more weight than the other), which results in lateral spine flexion. This lateral flexion causes the trunk musculature to become shortened on the nonweight-bearing side and lengthened on the weight-bearing side⁵⁵ (Fig. 10-23). At the same time, patients tend to assume a posterior pelvic tilt, which results in spinal flexion. Again the result is a muscle imbalance, with the anterior musculature (abdominals) becoming shortened and the posterior muscles (extensors) becoming elongated. Davies⁵⁵ hypothesized that patients sit with posterior pelvic tilt to compensate for weak abdominals. Patients assume this “safe” posture to prevent themselves from falling backward. The spinal flexion that results from the posterior tilt leads to loss of natural lumbar spine lordosis and accentuated thoracic spine kyphosis.

Figure 10-23 Asymmetrical trunk posture in patient with left hemiplegia. Note the left trunk shortening, right trunk elongation/overstretching, rib cage shift, loss of scapula stability on rib cage, relative downward rotation of scapula, increased weight-bearing on right ischial tuberosity, and shoulder asymmetry (left hemiplegia).

Abdominal weakness (especially the obliques) results in a destabilization of the rib cage. A lack of balance between the obliques results in trunk and rib cage rotation.¹⁰⁰ See Chapter 7.

Loss of scapula alignment.

Upper extremity malalignment commonly results from pelvic and trunk malalignments. When in a resting position, the scapula is flush on the rib cage (the scapulothoracic joint) and upwardly rotated. When one palpates the scapula, the distance between the inferior angle and the vertebral column should be greater than the distance between the medial border of the scapular spine and the vertebral column¹⁰⁰ (Fig. 10-24). In the resting position the glenoid fossa of the scapula faces upward, forward, and outward. Therefore, the trunk and rib cage must be stable to support the scapula properly. In hemiplegic patients, the scapula loses its orientation on the thoracic wall and assumes a position of relative downward rotation.⁴⁰

Figure 10-24 Normal resting posture of the scapula in upward rotation. A is the distance (in finger breadths or centimeters) from the medial border of the spine of the scapula to the vertebral column. B is the distance from the inferior angle of the scapula to the vertebral column. Distance B should be greater than distance A if the scapula is aligned appropriately. If A equals B or A is greater than B, then the scapula has assumed a position of relative downward rotation.

Cailliet⁴⁰ described several events that result in a downwardly rotated scapula (Fig. 10-25), such as lateral flexion toward the hemiparetic side. The lateral flexion may be due to trunk weakness, perceptual dysfunction that results in an inability to perceive midline, or excess activity in unilateral trunk flexors (i.e., latissimus dorsi). Downward rotation also can be caused by unopposed muscle activity that depresses and downwardly rotates the scapula (i.e., rhomboids, levator scapulae, and latissimus dorsi) or by generalized weakness in the muscles that orient the scapula in a position of upward rotation (i.e., serratus anterior, upper and lower trapezius).

Figure 10-25 A, Scapular alignment with a straight spine (xy glenoid angle). B, Paresis with downward rotation of scapula (AB glenoid angle). C, Relative downward rotation of scapula with functional scoliosis (CD glenoid angle).

(From Cailliet R: The shoulder in hemiplegia, Philadelphia, 1980, FA Davis.)

Loss of glenohumeral joint alignment.

Thus far the loss of pelvic/trunk, rib cage, and scapula control have been reviewed. All of the aforementioned alignment changes have an effect on the stability and alignment of the glenohumeral joint. The mechanisms of glenohumeral joint subluxation remains controversial. As reviewed by Cailliet⁴⁰ and Basmajian,¹⁴ the following factors assist in maintaining glenohumeral joint stability: the angle of the glenoid fossa when facing forward, upward, and outward; the support of the scapula on the rib cage; the seating of the humeral head in the fossa by the supraspinatus; possible support from the superior capsule; and contraction of the deltoid and cuff muscles when passive support is eliminated by slight abduction of the humerus.⁴⁰ Cailliet stated that any change in these factors may play a role in causing subluxation (Fig. 10-26).

Figure 10-26 Possible biomechanics of subluxation from malalignment. Line AB indicates an aligned spine (the goal of treatment). Instead the spine assumes a position of lateral flexion (curve CB). The scapula downwardly rotates (GH), resulting in a downward angulation of the glenoid fossa (XY). Because of the scapula position, the supraspinatus (S) loses its mechanical line of pull, making it ineffective and prone to overstretching. The result is a subluxation of the glenohumeral joint.

(Modified from Cailliet R: Shoulder pain, Philadelphia, 1990, FA Davis.)

Basmajian’s EMG studies¹⁴ confirmed that the supraspinatus prevents downward migration of the humeral head when a downward load is applied to the upper extremity (e.g., when a person holds a briefcase). Authors previously believed that the deltoid performed this function, but the deltoid actually shows no activity during this function. The author pointed out that the supraspinatus is a horizontally positioned muscle that runs through the supraspinous fossa and can be effective only if the scapula is oriented correctly on the thorax.

The upward orientation of the glenoid fossa creates a “cradle” for the humeral head. As the humerus is pulled downward, it is forced to move laterally by the slope of the fossa.¹⁴ The supraspinatus (and superior portion of the capsule) prevents this lateral movement and therefore downward migration. Basmajian¹⁴ also pointed out that this mechanism is not effective if the humerus is abducted. This position predisposes patients to subluxation by eliminating the described mechanism. Many patients are positioned so that their humerus is abducted slightly because of the lateral trunk flexion toward the more affected side or due to passive positioning.

The relationship between scapula rotation and inferior subluxation has been challenged. Prevost and colleagues¹⁴⁸ evaluated both shoulders of 50 stroke survivors with inferior subluxations using tridimensional radiograph. Results included the following:

The affected and nonaffected shoulders were different in terms of the vertical position of the humerus vis-a-vis the scapula.

The orientation of the glenoid cavities was also different; the subluxed one faced less downward.

The angle of abduction of the arm of the affected side was significantly greater than on the nonaffected side, but the relative abduction of the arm was on the same order of magnitude for both sides.

No significant relationship existed between the orientation of the scapula and the severity of the subluxation.

The abduction of the humerus was weakly (r = 0.24) related to the subluxation, which partly explained the weak association found between the relative abduction of the arm and the subluxation.

Overall, the authors concluded that the position of the scapula and the relative abduction of the arm cannot be considered important factors in the occurrence of inferior subluxation in hemiplegia.

Similarly, Culham, Noce, and Bagg⁵⁴ examined 17 subjects with high tone and 17 subjects with low tone based on the Ashworth Scale. Linear and angular measures of scapular and humeral orientation were calculated from tridimensional coordinates of bony landmarks collected using an electromagnetic device with subjects in a seated position with arms relaxed by their sides. Glenohumeral subluxation was measured from radiographs. They found the following:

The scapula was farther from the midline and lower on the thorax on the affected side in the low-tone group.

Glenohumeral subluxation was greater in the low-tone group.

The scapular abduction angle was significantly greater on the nonaffected side in the low-tone group compared with the affected side in this group and with the nonaffected side in the high-tone group.

In the high-tone group, no differences were found between the affected and nonaffected sides in the angular or linear measures.

No significant correlation was found between scapular or humeral orientation and glenohumeral subluxation in either group.

Chaco and Wolf⁴⁵ confirmed that the supraspinatus did not respond to loading in the hemiplegic patients they studied. Although not immediate, subluxation developed later in the study in the patients who remained flaccid. They inferred that the joint capsule holds the head of the humerus in relation to the glenoid fossa, but unless the supraspinatus starts responding, it cannot prevent subluxation indefinitely. Therefore, subluxation appears to be caused by the weight of the arm and mechanical stretch to the joint capsule and traction to unresponsive shoulder musculature.

Ryerson and Levit^156,157 described three patterns of subluxation in the glenohumeral joint. They emphasized that the therapist must assess trunk posture, determine the position of the scapula on the trunk, evaluate scapular mobility and rhythm, and examine the alignment and mobility of the glenohumeral joint before setting treatment goals for the shoulder. Table 10-7 reviews Ryerson and Levit’s subluxation classifications, including inferior, anterior, and superior subluxations.

Table 10-7 Subluxation/Malalignment Patterns in the Upper Extremity after Stroke

Hall, Dudgeon, and Guthrie⁸³ assessed the validity of three clinical measures (palpation, arm length discrepancy, and thermoplastic jig measurement) for evaluating shoulder subluxation in adults with hemiplegia resulting from a stroke. These measures were combined with anterior/posterior radiographic examinations of the hemiplegic shoulder; results indicated that palpation had the highest correlation with successful subluxation evaluation. In their technique for palpating subluxation, the patient is seated with the upper extremity unsupported at the side in neutral rotation; trunk stability was maintained during the evaluation. During palpation, the therapist measured subluxation by palpating the subacromial space (the distance between the acromion and the superior aspect of the humeral head) with the index and middle fingers. The authors concluded that their findings provided cautious optimism in terms of measuring and identifying subluxation. Prevost and colleagues¹⁴⁹ also validated that palpation is a reliable measurement tool in the evaluation of subluxation. One should note that the evaluator should palpate both shoulders for comparison.

Hall, Dudgeon, and Guthrie⁸³ used a 0 (no subluxation) to 5 (2½ finger widths of subluxation) scale during their study. Bohannon and Andrews²⁰ used a 3-point scale to demonstrate interrater reliability for measuring subluxation: none, 0; minimal, 1; and substantial, 2.

Interdependence of trunk and limb alignment.

Anatomically, therapists must remember that only one bony attachment connects the entire limb to the axial skeleton, the sternoclavicular joint. (The scapulothoracic joint is not a true joint; the scapula rides on the thoracic cage and is maintained by muscular attachments only.) Therefore, the clavicle serves as an anatomical link between the shoulder complex and trunk. This point should solidify the interdependence between the trunk and upper extremity. Any malalignments in the proximal segments have deleterious effects on the upper extremity (Fig. 10-27).

Figure 10-27 Shoulder anatomy. Seven joints make up the shoulder complex. The sternoclavicular joint is the only bony attachment of the shoulder to the trunk, with the clavicle serving as a bridge between the trunk and shoulder. Skeletal alignment of the shoulder joint depends on trunk alignment and stability. For example, if the pelvis becomes malaligned (pelvic obliquity), the vertebral column, the rib cage, and other components lose their alignment (see Fig. 10-23).

The musculature acting on the shoulder has proximal points of attachment. A group of upper extremity muscles (the trapezius, rhomboids, serratus anterior, and levator scapulae) runs between the trunk and scapula, and another (the pectoralis and latissimus dorsi) runs between the trunk and humerus. Another group of muscles (the deltoid, rotator cuff, and coracobrachialis) attaches from the humerus to the scapula. These attachments emphasize the interdependence of trunk alignment and extremity control.

Mohr¹²⁷ pointed out that biomechanical malalignment produces a pattern of movement that looks like stereotypical patterns used by patients with spasticity. For example, patients who gain early control of scapula elevation and humeral abduction continue to use this pattern and also flex the trunk, resulting in more elevation and abduction. As the scapula tips forward, it predisposes the humerus to internal rotation and extension because of its position in the fossa. The distal arm follows into elbow flexion, pronation, and wrist and digit flexion. The author stated that if a normal individual only activates the scapula elevators with humeral abductors, the resulting pattern looks similar to the patterns used by stroke survivors.

These alignment problems need to be addressed before and throughout the treatment session. Therapists should correct them by mobilization techniques, positioning, and appropriate activity choices. The therapist needs to ensure alignment during ROM activities and maintain appropriate alignment during functional activities. For example, the alignment of the trunk and pelvis of patients who are trying to feed themselves has a direct effect on the quality of the extremity movement pattern. Even in persons without a known neuropathological condition, the quality of the eating activity clearly is compromised if they assume a forward flexed and laterally flexed static posture rather than an aligned and active trunk posture.

Ryerson and Levit¹⁵⁶ suggest patients perform activities that maintain enhanced trunk alignment and simultaneously coordinate movements of the scapula, trunk, and humerus.

Shoulder supports

Shoulder supports include any devices used to align, protect, or support an affected proximal limb. Shoulder supports include bed-positioning devices, adaptations to seating systems, and slings. The use of shoulder supports, especially slings, has been debated in the literature for at least 30 years. A recent review concluded that “There is insufficient evidence to conclude whether slings and wheelchair attachments prevent subluxation, decrease pain, increase function or adversely increase contracture in the shoulder after stroke.”⁴

Much of the debate is fueled by the variety of available slings, the controversy regarding their effectiveness, when and how they should be used, and whether they add to the already numerous complications resulting from an extremity affected by stroke.

Boyd and Gaylard³² published the results of their survey of Canadian occupational therapists who prescribe slings. The respondents most frequently indicated that the goals of using a sling were to decrease and prevent subluxation and pain. The respondents frequently measured the effectiveness of their interventions by the level of resulting pain relief, subluxation assessments, and the amount of hand swelling. Less frequent measures of effectiveness included ROM, spasticity, and body awareness.

In light of the previously proposed cause of subluxation (see Loss of Biomechanical Alignment), Cailliet⁴⁰ suggested that if the goal of treatment is to provide glenohumeral joint stability, then the device must support the scapula on the rib cage with the glenoid fossa facing upward, forward, and outward and must compensate for a lack of support by the rotator cuff and possibly the superior capsule. At this point, no slings are available on the market that assist in realigning the scapula on the rib cage. Therefore, slings cannot be prescribed to “reduce a subluxation.” They may lift the head of the humerus to the level of the glenoid fossa, but the scapular and trunk alignments (the key to correcting shoulder malalignment) remain impaired. This reduction may be seen as treating a symptom of a larger problem. The therapist must realize that they may find cases in which treating this symptom is appropriate. Analysis is critical for determining which goals certain interventions are achieving. Palpating the subluxation before and after the sling is donned is not sufficient. The therapist must evaluate the effect (if any) of the sling on the more proximal segments.

In their review of the literature, Smith and Okamoto¹⁶⁵ identified desirable and undesirable features of slings. Proper positioning of the humeral head in addition to humeral abduction, external rotation, and elbow extension are cited as desirable positions as opposed to humeral adduction, internal rotation, and elbow flexion. The latter positions typically cause problems in the maintenance of tissue length in the stroke population. The sling also should permit the impaired extremity to provide postural support when the patient is seated and should allow self-ROM. In terms of positioning, the sling should provide neutral wrist support, unobstructed hand function, finger abduction, and scapula protraction and elevation.

Smith and Okamoto¹⁶⁵ emphasized that if a therapist expects compliance with sling use, comfort, cosmetic appeal, and easy donning and doffing are crucial. The authors published a checklist to assist therapists in analyzing the slings they provide. The percentage of therapists using slings has been reported to be as high as 94%,³² despite the fact no definitive studies support or reject the use of slings. Several studies have compared and contrasted the effectiveness of various supports. Zorowitz and colleagues²⁰⁴ compared the following four supports:

In this study, 20 patients were evaluated in the listed supports with anteroposterior shoulder radiography. The authors evaluated the vertical, horizontal, and total asymmetries of glenohumeral joint subluxation compared with the opposite shoulder. In terms of vertical asymmetry, the single-strap hemisling corrected the vertical displacement, the Cavalier support did not alter vertical displacement, and the remaining supports significantly reduced but did not correct vertical displacement.

Although as a group, the subjects had no significant horizontal asymmetry when no supports were used, the Bobath roll and the Cavalier support produced a significant lateral displacement of the humeral head of the more affected shoulder. This fact is of interest because one proposed goal of a sling is to decrease or prevent subluxation; this study demonstrated that equipment not well-researched actually may cause shoulder asymmetry in patients who previously had none.

In terms of total asymmetry, the Rolyan humeral cuff sling was the only support that significantly decreased (although it did not eliminate) total subluxation asymmetry.

Moodie, Brisbin, and Morgan¹²⁸ evaluated the effectiveness of five shoulder supports: the Bobath roll, an acrylic plastic lap tray on a wheelchair, a wheelchair-mounted arm trough, a conventional triangular sling (which is much like an arm cast support), and the Hook Hemi Harness (which has two adjustable shoulder cuffs with a suspension strap that are tightened while the affected arm is lifted, resulting in shoulders of equal height). Anteroposterior radiographs of 10 subjects demonstrated that the conventional sling, lap tray, and arm trough were effective in decreasing the width of the glenohumeral space to normal. The Bobath roll and the Hook Hemi Harness were not effective in reducing the subluxation. The authors pointed out that although the conventional sling decreased the subluxation, it reinforced the flexor pattern found in the upper extremity.

Brook and colleagues³⁶ compared the effects of three supports: the Bobath sling, an arm trough/lap board, and the Harris hemisling (which has two straps and cuffs that cradle the elbow and wrist, holding the arm in a position of adduction, internal rotation, and elbow flexion). The Harris hemisling resulted in good vertical correction; in comparison the Bobath sling did not correct the subluxation as well, the arm trough/lap board was less effective and tended to overcorrect, and the Bobath sling tended to distract the joint horizontally.

An important note is that none of the mentioned studies discussed scapular or trunk alignment; they only addressed the glenohumeral joint.

Hurd, Farrell, and Waylonis⁹² alternately placed 14 patients into a control group (which used no sling) or treatment group (which used a sling). These patients were treated identically in all other respects. The patients were evaluated initially and again two to three weeks later and three to seven months later. No appreciable difference in shoulder ROM, shoulder pain, or subluxation was found between the treated or control groups. No evidence of increased incidence of peripheral nerve or plexus injury was noted in the control group. The authors concluded that the hemisling does not need to be used uniformly by all patients with a flaccid limb after a stroke. They suggested that a sling might be useful when used with discrimination but did not elaborate on this point.

Some authors have suggested that slings be prescribed to prevent overstretching of soft tissue. Chaco and Wolf⁴⁵ proposed that permanent subluxation of the glenohumeral joint could be prevented by avoiding loading on the joint when the limb is flaccid. They concluded that the joint capsule holds the head of the humerus in relation to the fossa when the supraspinatus is not responding but cannot prevent subluxation for an unlimited time unless the cuff responds.

If the joint capsule is prevented from stretching during the stage in which the limb is flaccid, patients may have a better opportunity to develop adequate muscle function to maintain joint alignment. Kaplan and colleagues¹⁰¹ advised using a sling during the flaccid stage to prevent distraction of the joint resulting in a possible brachial plexus injury.

Some therapists have suggested that sling use may increase body neglect and interfere with body image, although this hypothesis has not been researched. Although they have not been specifically related to sling use, the learned nonuse studies of Taub, Uswatte, and Pidikiti¹⁷³ may influence therapists’ decisions about whether to prescribe a sling, especially for a patient in the acute phase.

Zorowitz²⁰⁴ stated that “although supports are commonly used during the rehabilitation of stroke survivors, there is no absolute evidence that supports prevent or reduce long-term shoulder subluxation when spontaneous recovery of motor function occurs, or that a support will prevent supposed complications of shoulder subluxation. Without proper training in the use of a support, stroke survivors may face potential complications such as pain and contracture.” Although the literature does not give definitive answers about when or whether to use slings, one can infer the following guidelines:

Figure 10-28 A, Pouch sling. Sling is only to be used for short periods with patients in upright postures and frees therapist’s hands to control trunk and lower extremities. This sling may be appropriate for initial phases of walking, transfer, and upright function training. B, Shoulder saddle sling. Sling supports distal weight of extremity and can be worn under clothing. This style of sling can be worn all day because it does not block distal function or hold extremity in a flexor pattern. C, The GivMohr Sling (www.givmohrsling.com, 505-292-1144). (A and B courtesy of Sammons Preston Rolyan, Inc, Bolingbrook, Ill.)

Figure 10-29 Taping/strapping is being used more commonly to treat shoulder instability. Further research is required to determine its effectiveness.

One may infer from the literature that the most effective way to reduce the level of subluxation is to provide the patient with activities that enhance trunk and scapula alignment, activate the rotator cuff, and enhance functional use of the extremity during weight-bearing and reach patterns.