Skeletal system

This section reviews the bony components of trunk anatomy, including articulations and range of motion (ROM).

Vertebral column.

The vertebral column is made up of 26 vertebrae, which are classified as follows:

Cervical: 7

Thoracic: 12

Lumbar: 5

Sacral: 5 (fused into one bone, the sacrum)

Coccygeal: 4 (fused into one or two bones, the coccyx)

As a whole, the vertebral column from sacrum to skull is equivalent to a joint with three degrees of freedom³³ in the directions of flexion and extension, right and left lateral flexion, and axial rotation. Kapandji³³ has documented the ROM throughout the vertebral column (Table 7-1).

Table 7-1 Range of Motion of the Vertebral Column

MOVEMENT	RANGE OF MOTION
Flexion	Cervical: 40 degrees Thoracolumbar: 105 degrees Total: 145 degrees
Extension	Cervical: 75 degrees Thoracolumbar: 60 degrees Total: 135 degrees
Lateral flexion	Cervical: 35 to 45 degrees Thoracic: 20 degrees Lumbar: 20 degrees Total: 75 to 85 degrees
Rotation	Cervical: 45 to 50 degrees Thoracic: 35 degrees Lumbar: 5 degrees Total: 85 to 90 degrees

An understanding of spinal alignment is necessary for effective evaluation and treatment planning. Normal alignment of the vertebral column implies that the appropriate spinal curvatures are present. In the sagittal plane, the vertebral column shows four curvatures³³ (Table 7-2 and Fig. 7-1).

Table 7-2 Spinal Curvatures

	POSTERIOR MOVEMENT
CURVATURE	CONVEX	CONCAVE
Sacral (fixed)	X
Lumbar		X
Thoracic	X
Cervical		X

Figure 7-1 Lateral view of spine with spinal curvatures.

Muscular system

Muscles of the abdominal wall

The general functions of the abdominal muscles are as follows:

Abdominal viscera support

Respiration assistance

Trunk control in the directions of flexion, lateral flexion, and rotation

Although these muscles are situated primarily on the anterior aspect of the trunk, they also are situated laterally and slightly posteriorly, forming a girdle around the abdomen. The abdominal muscles consist of three groups: the rectus abdominis, the obliques (internal and external), and the transversus abdominis (Fig. 7-2).

Figure 7-2 Anterior anatomy of trunk with resected layers.

Rectus abdominis.

The rectus abdominis consists of right and left sides that are separated by a fibrous band called the linea alba, which runs from the xiphoid process to the pubis.

The proximal attachment is the xiphoid process of the sternum and adjacent costal cartilage, whereas the distal attachments are the pubic bones near the pubic symphysis.⁴⁷

The muscle is palpated easily in the following two cases:

1. When the subject is supine and is asked to lift the head and shoulders off the support surface in a straight plane (sit-up)

2. During backward sway in sitting or standing position

When this muscle is activated and not opposed by the extensors, the pelvis and sternum are approximated, the pelvis is pulled into a posterior tilt, and the lumbar curves flatten. Because of its multisegmental arrangement, the rectus abdominis can contract in part or as a whole, making a variety of postures possible. De Troyer’s work¹⁹ demonstrated that “abdominal muscle recruitment which naturally occurs in response to posture in most individuals does not uniformly involve the whole of the muscles.”

The rectus abdominis (and the other muscles of the trunk) require a stable origin to function efficiently.¹⁷ This stable origin can be the pelvis or thorax, depending on the posture and which part of the trunk is moving. Davies¹⁷ further explains, “The pelvis is stabilized in lying, sitting, and standing by the activity of the muscles around the hips, and in sitting and lying the stabilization is helped by the weight of the legs themselves. Stabilization of the thoracic origin for activities in which the abdominals contract to move or prevent movements of the pelvis requires selective extension of the thoracic spine.” Davies further points out that the abdominal muscles cannot function effectively when their origin and insertion are approximated (e.g., in patients with an exaggerated thoracic kyphosis). Winzeler-Mercay and Mudie⁵⁷ noted weakness based on electromyographic recordings in the rectus abdominis after stroke during dynamic trunk movements such as donning shoes. The weakness was noted particularly on the involved side. Similarly, Tanaka, Hachisuka, and Ogata⁴⁹ found that peak torque of the flexors was significantly less than in healthy controls.

The rectus abdominis can be self-palpated by assuming a recumbent posture in a chair (slumping in the chair) and then pulling up and forward to an aligned position. One should notice that the burst of activity diminishes when leaning forward (shoulders move in front of hips).

Obliques.

The obliques consist of three interwoven muscles: internal obliques, external obliques, and transversus abdominis.

External obliques.

The external oblique forms the superficial layer of the abdominal wall. Its fibers run an oblique course superoinferiorly and lateromedially.³³ The muscle is lateral to the rectus abdominis and covers the anterior and lateral regions of the abdomen. The attachments are as follows:⁴⁷

Proximal attachment: Anterolateral portions of ribs where the muscle interdigitates with serratus anterior and slips from latissimus dorsi

Distal attachment: Upper fibers run down and forward and attach to an aponeurosis that connects them to the linea alba; lower fibers attach to the crest of the ilium

If the external oblique contracts unilaterally, the trunk rotates to the opposite side. Therefore, if one rotates to the left, the right external oblique is active and vice versa. Bilateral contraction assists in trunk flexion and a resultant posterior pelvic tilt. This muscle is also active during straining and coughing.⁴⁷ The muscle is palpated easily while rotating the trunk to the opposite side.

Internal obliques.

The internal obliques also are located laterally and are covered by the external obliques. In essence, the internal obliques constitute the second layer of muscles on the abdominal wall. This muscle covers the same area as the external oblique, but its fibers cross those of the external oblique. Attachments are as follows:⁴⁷

Proximal attachment: Inguinal ligament, crest of ilium, and thoracolumbar fascia

Distal attachments: Pubic bone, an aponeurosis connecting to linea alba, and last three or four ribs

This muscle groups is activated during trunk rotation, but contraction occurs toward the same side (i.e., rotation to the left occurs following contraction of the left internal oblique). Clearly the external and internal obliques are synergists in the action of trunk rotation. The right external and left internal oblique work together to rotate the trunk to the left and vice versa. “The efficient action of the muscles of one side of the abdominal wall is therefore very much dependent upon the fixation or anchorage provided by the activity of the muscles on the other side, particularly for activities involving rotation of the trunk.”¹⁶

Tanaka, Hachisuka, and Ogata⁴⁹ examined trunk rotation performance in poststroke hemiplegic subjects and found significantly lower muscle performance in the subjects compared with the health controls. No differences were found when comparing right and left rotation in terms of angular velocities, the side of hemiplegia, or gender, but muscle performance in both directions was decreased compared with controls.

The internal obliques are difficult to palpate. However, the therapist may feel tension under the fingertips when palpating the lateral abdominal wall on the side toward which the trunk is rotating. This tension is due in part to activation of the internal obliques.

Transversus abdominis.

The transversus abdominis is the deepest layer of the abdominal wall. Its fibers run transversely, and the muscle has been called the corset muscle because it encloses the abdominal cavity like a corset. Attachments are as follows:⁴⁷

Proximal attachments: Lower ribs, thoracolumbar fascia, crest of the ilium, and inguinal ligament

Distal attachments: Via an aponeurosis fuses with other abdominal muscles into linea alba

The main action of the transversus abdominis is forced compression; the muscle acts like a girdle to flatten the abdominal wall and compress the abdominal viscera. Weakness of this muscle permits bulging of the anterior abdominal wall, thereby indirectly leading to an increase in lordosis.³⁸ The therapist may palpate this muscle between the lower ribs and the crest of the ilium during forced expiration.

Posterior trunk muscles

The posterior trunk muscles include the quadratus lumborum, the erector spinae group, and latissimus dorsi (Fig. 7-3). The actions of this group of muscles include trunk extension, lateral flexion, rotation of the trunk, and assistance with balancing the vertebral column.

Figure 7-3 Posterior anatomy of trunk.

Quadratus lumborum.

The quadratus lumborum is lateral and posterior (i.e., on the posterior abdominal wall); it lies between the psoas major and the erector spinae group. The attachments are as follows:⁴⁷

Proximal attachment: Crest of ilium

Distal attachments: Twelfth rib and transverse processes of first to third lumbar vertebrae

The main action of this muscle is to assist in “hip hiking.” Therefore, the muscle is active during lateral trunk flexion. The easiest way to palpate the quadratus lumborum is to have the subject prone, to palpate superior and lateral to the iliac crest, and to ask the subject to hike the hip.

Erector spinae group.

The erector spinae group of muscles is a large mass that fills the spaces between the transverse and spinous processes of the vertebrae and extends laterally covering a large portion of the posterior thorax. Multiple muscles make up this group, and they are named according to attachments, shape, and action. Muscles such as the transversospinales, the interspinales, the longissimus, and the iliocostalis are included in this group.

Collectively, these muscles connect the back of the skull to the posterior iliac crest and sacrum. Unopposed contraction of the back extensors approximates the head and the sacrum. The pelvis is pulled into an anterior tilt (accentuating the lumbar curve), and the ribs are forced to flare. These muscles also contract during lateral flexion (to balance the abdominals), and they may assist in trunk rotation during unilateral contraction (e.g., assist the trunk with rotating to the ipsilateral side). The therapist easily can palpate this muscle group with patients in the prone position if the head and shoulders are lifted from the support surface;⁴⁷ palpation also is possible during forward sway in sitting or standing. The therapist easily can palpate the lower back extensors during low back extension (accentuating the lumbar curve) while the patient is sitting.

Winzeler-Mercay and Mudie⁵⁷ found increased activity in the erector spinae on both sides during work activities (such as reaching and donning shoes) and at rest. This increased activity was particularly evident on the involved side. They hypothesized that this abnormal response might reflect a disruption of cortical influences on motor unit activity. Similarly, Tanaka, Hachisuka, and Ogata⁴⁸ found that peak torque of the extensors was significantly less than in healthy controls.

Latissimus dorsi.

The latissimus dorsi is superficial and covers the posterior/lateral trunk. Its attachments include the following:⁴⁷

Proximal attachments: Spinous processes of T6 down, dorsolumbar fascia, posterior crest of ilium, lower ribs, interdigitations with external oblique; fibers converge toward axilla, passing over the inferior angle of the scapula.

Distal attachments: Tendon attaches to crest of lesser tubercle of humerus, proximal to the teres major.

Acting unilaterally, the latissimus dorsi adducts, extends, and internally rotates the humerus and laterally flexes the trunk (approximates the shoulder and the pelvis). Bilateral contraction helps hyperextend the spine and anteriorly tilt the pelvis.

Trunk muscle contractions

To achieve full trunk control and to use this control during functional tasks, patients must regain the ability to contract their trunk muscles under three different circumstances outlined by Davies.¹⁷ The task of lower extremity bathing from a seated position illustrates these points:

The previous examples show that functional independence requires control of all three trunk muscle contractions and combinations. Successful treatment plans must include activities that elicit a variety of trunk muscle contractions. Self-care training inherently challenges a variety of trunk postures and muscle contractions.

Musculoskeletal components

Control of the trunk depends on several musculoskeletal variables including ROM, biomechanical alignment, strength, and muscle length. These variables are interdependent and can create a vicious circle in stroke patients.

Postural malalignment

Stroke patients commonly assume postural malalignments that first must be identified via observations and palpations. After identification, the causative factors must be determined before determining the most appropriate intervention (Table 7-3).

Table 7-3 Common Postural Alignments and Potential Causes

Prolonged postural malalignment results in muscle shortening on one side of the trunk and muscle overstretching on the opposite side. For example, a posterior pelvic tilt with lumbar flexion results in shortening of the anterior musculature and elongation (overstretching) of the posterior muscles. Lateral flexion on the right side results in muscle shortening on the right side and muscle elongation on the left side of the trunk.

Postural malalignment may occur because of unilateral weakness (specifically around the pelvis), unbalanced skeletal muscle activity, perceptual dysfunction and an inability to perceive midline, and soft-tissue shortening.

Prolonged postural malalignment can result in soft-tissue shortening, loss of ROM, and an inability to generate enough force to contract the muscle group in question. The total force of muscle (active tension) is high at the rest length of the muscle (i.e., when the trunk is aligned properly) and less when the muscle is tested at shorter lengths. Therefore, the force-generating mechanism within the muscle works optimally at the rest length of the muscle³⁸ (i.e., a symmetrical and aligned trunk).

Managing stiffness and the degrees of freedom problem

Mohr⁴⁰ states, “Normal control in any body part demands the ability to dissociate (separate) different parts of the body.” She gives the examples of dissociating the head from the body, one side of the body from the other, and the upper trunk from the lower trunk. Instead, patients often appear stiff, have nonfluid movements, and move their body segments as a unit

Examples of dissociation during functional tasks include upper trunk rotation with lower trunk stability while reaching for toilet paper, counterrotation of the trunk during ambulatory activities, and upper trunk rotation with concurrent lower trunk lateral flexion to increase the range of reach beyond the arm span when reaching for a phone positioned on the left side of a desk with the right hand.

Difficulty with dissociation/postural stiffness may result from soft-tissue tightness, bony contracture, or efforts by the patient to decrease the degrees of freedom in the trunk⁴⁵ during functional activities. It is critical to determine why the person is not able to dissociate. A typical clinical problem is determining if trunk stiffness and lack of dissociation is due to soft-tissue tightness (which may require soft tissue stretching and mobilization) or if the person is freezing the degrees of freedom in an effort to maintain stability (which requires core stabilization activities). One method to differentiate the cause is to provide various levels of postural support, for example, sitting in a high back chair versus sitting unsupported on a therapy table, or side lying versus sitting unsupported. If the underlying cause of the stiffness is related to freezing the degrees of freedom, substantial differences will be noted for both passive and active movements under the various conditions of postural support. In the situations in which the patient has the most support (side lying and supported seating), he or she will be able to “free” the degrees of freedom and move with increased ease and fluidity, and will be able to separate body parts. If the same person is placed in a condition of decreased postural support, the system will respond by “freezing” the degrees of freedom, and stiffness will emerge.

Motor adaptation

Concerning motor adaptation, Smith, Weiss, and Lehmkuhl⁴⁷ state, “Normal postural control requires the ability to adapt responses to changing tasks and environmental demands. This flexibility requires the availability of multiple movement strategies and the ability to select the appropriate strategy for the task and environment. The inability to adapt movements to changing task demands is a characteristic of many patients with neurological disorders. Patients become fixed in stereotypical patterns of movement, showing a loss of movement flexibility and adaptability.”

Motor adaptation can occur in response to an external perturbation or in anticipation of potentially destabilizing forces. Unexpected external perturbations include bumping into someone in a crowded lobby, being in a vehicle that unexpectedly turns or decelerates, and being on a moving platform, such as an escalator, that stops unexpectedly.

Activities that lead to trunk movements in anticipation of destabilizing forces (i.e., internal perturbations) include reaching for a heavy book on a shelf, reaching beyond the arm span, and preparing to push or pull a chair into place. Shumway-Cook and Woollacott⁴⁶ point out that anticipatory postural control depends heavily on previous experience and learning. Research focusing on anticipatory postural responses during reach activities is presented in Chapter 10.

Subjective interview

Therapists should question patients about their perceived stability limits. Stability limits have been defined as the “boundaries of an area of space in which the body can maintain its position without changing the base of support.”⁴⁷ Patients’ perceived stability limits may or may not be consistent with their actual limits. If patients’ perceived limits of stability are greater than their actual limits, they are at risk for falls. If their perceived limits of stability are less than their actual limits, they may be reluctant to attempt tasks with progressively greater demands on their postural system (e.g., lower extremity dressing without assistive devices and picking up objects from the floor without a reacher).

Perceived stability limits may have a direct correlation with observed neurobehavioral deficits. Body scheme disorders commonly occur in the stroke population. These deficits include body neglect, somatoagnosia, and impaired right/left discrimination.² Ayres³ has defined body scheme as a postural model on which movements are based. Knowledge of body parts and their relationships are necessary for deciding what and where to move and in what way to perform.² Spatial relation deficits including spatial neglect, depth perception, and spatial relation disorders also may have an effect on patients’ perceived stability limits (gaining and regaining midline orientation and position in space) (see Chapter 18).

Other components of the subjective interview include determining patients’ insights into their trunk malalignments and their ability to perceive and assume midline positions.⁴³ The therapist’s goal in this interview is to gain insight into the patients’ ability to make accurate observations about their postural dysfunction. This is difficult for many patients because trunk control does not occur at a conscious level in the majority of daily tasks.

Standardized assessments

The use of valid and reliable tools is always recommended. The following section reviews available measurement instruments related to trunk control. The first three instruments specifically evaluate trunk control after stroke and are therefore highly recommend for this area of practice/research, while the others are comprehensive measures that include items related to the trunk. See Table 7-4 for a review of the psychometric properties of these three measures.

Table 7-4 Psychometric Properties of the Trunk Control Test and Two Trunk Impairment Scales

Observations of trunk alignment/malalignment

For the purposes of this chapter, observations concern the seated posture. The patient’s trunk should be exposed as much as possible, and the patient should be asked to “sit up nice and straight and gently rest your hands in your lap.” (Table 7-8 outlines the ideal alignment of the trunk and extremities and common asymmetries observed after stroke during static sitting.)

Table 7-8 Typical Alignment and Common Malalignments after Stroke

Following the evaluation of postural malalignments during static sitting, the therapist should begin to hypothesize the cause of these malalignments. Causes may include increased skeletal muscle activity on one side of the trunk, inability to recruit muscle activity or weakness, soft-tissue shortening, fixed deformity, body scheme disorder, and inability to perceive midline.

The therapist must remember that observed postures may be caused by more than one impairment (see Table 7-3). For example, stroke patients tend to sit in a posterior pelvic tilt position with resultant hip extension and thoracic spine flexion. This posture may result from one of or a combination of the following:

Another example of a commonly observed malalignment is trunk shortening on the side affected by the stroke. The patient may assume this posture for several reasons:

Following the observation of the patient in a static posture, the occupational therapist must observe trunk responses during functional activities. The two most effective methods of making these observations are observing patients during self-care and mobility in a variety of positions and controlled reach pattern activities (Table 7-9). During functional reach patterns, trunk responses are required to provide proximal stability for distal function, enhance the ability to interact with the environment by increasing reaching distance (i.e., extend the arm span with an appropriate trunk response), and prevent falls.

Table 7-9 Effects of Object Positioning on Trunk Movements and Weight Shifts during Reaching Activities*

POSITION OF OBJECT	TRUNK RESPONSE/WEIGHT SHIFT
Straight ahead at forehead level, past arm’s length	Trunk extension, anterior pelvic tilt Anterior weight shift
On floor, between feet	Trunk flexion Anterior weight shift
To side at shoulder level, past arm’s length	Left trunk shortening, right trunk elongation, left hip hiking Weight shift to right
On floor, below right hip	Right trunk shortening, left trunk elongation Weight shift to right
Behind right shoulder, at arm’s length	Trunk extension and rotation (right side posteriorly) Weight shift to right
At shoulder level, to left of left shoulder	Trunk extension and rotation (left side posteriorly) Weight shift to left
On floor, to left of left foot	Trunk flexion and rotation (left side posteriorly) Weight shift to left
Above head, directly behind	Trunk extension, shoulders move behind hips Posterior weight shift

* These examples are for a patient with left hemiplegia. The left-hand column indicates where to position objects during a reaching task (using the right upper extremity). The right-hand column indicates the resultant trunk position and weight shift.

An individual’s reaching ability is limited to within the arm span by static trunk postures. When an object is placed beyond arm’s length (e.g., on a floor, across a dining table, or under a sink), a trunk response is required to pick up the object successfully.

In general, picking up an object from the floor or from in front of an individual requires an anterior trunk shift. Picking up objects placed beyond the arm span to the right or left of the individual requires a lateral weight shift from the trunk primarily onto one of the ischial tuberosities. Retrieving objects placed behind the trunk requires a posterior weight shift. Rotational trunk responses result from reaching across the midline or for objects posterior to the shoulders or hips.

The therapist’s goals while observing the patient perform functional reach patterns are the following:

Evaluation of specific trunk movement patterns

In addition to performing each movement pattern, the reader should refer to the appropriate figures while reading this section. The following evaluation procedures are based on the work of Mohr,⁴⁰ Boehme,⁸ Davies,¹⁶ and Basmajian and DeLuca.⁵

Trunk flexor control

The trunk flexors are evaluated by the five different methods that follow:

1. Patients assume a seated, upright position. The therapist asks them to move their shoulders behind their hips slowly and with control (Fig. 7-4, A); this movement pattern occurs in the sagittal plane, is initiated from the upper trunk,⁴⁰ and elicits an eccentric contraction of the trunk flexors.^25,50 Holding the end range of this posture results in an isometric contraction of the trunk flexors. Observations should include resistance to movement, fall potential, and symmetry of the posterior weight shift. Unilateral weakness causes the weak side to become posterior to the stronger side (i.e., it results in rotation of the trunk).

2. From the end position of the first movement pattern, the therapists asks patients to move their shoulders forward so that they are sitting in proper alignment within the sagittal plane (see Fig. 7-4, B); this movement pattern is achieved by a concentric contraction of the trunk flexors.²⁵ The therapist should note symmetry during the movement pattern. Unilateral weakness causes the stronger side to lead the pattern.

3. In an aligned, seated position, patients assume a controlled lumbar flexion posture (posterior tilt with flattening of the lumbar curve and spinal flexion) (see Fig. 7-4, C). Mohr⁴⁰ states that this movement pattern is initiated by the lower trunk and pelvis. If this pattern is performed actively, the final posture is assumed by concentric flexor contraction. Patients also may achieve this posture by a relaxation response of the low back extensors, so the therapist should palpate the flexors to ensure the pattern is due to active movement. At the end range of this pattern—posterior tilt and spinal flexion (a recumbent posture)—little to no muscle activity exists, and patients maintain this posture by support of their vertebral ligaments.⁵

4. The therapist also should evaluate control of the trunk flexors when the patient is supine (during rolling and bed mobility activities). While the patient is in a supine position, the therapist asks the patient to sit up in a straight plane. This movement pattern, which is controlled primarily by the rectus abdominis, allows the therapist to evaluate antigravity control of the trunk flexors. The therapist also can ask the patient to roll by lifting one shoulder up and across the trunk in a position of trunk flexion and rotation. This movement pattern also gives the therapist insight into the antigravity control of the flexors (primarily the obliques).³⁸

5. Although the first four movement patterns to test flexor control were initiated by the patient, testing the response of the flexors to being moved by the therapist also is useful. The therapist lifts the lower legs of the patient into a position of increased hip flexion. For the patient to refrain from falling backward, the trunk flexors must be activated isometrically (see Fig. 7-4, D).

Figure 7-4 Trunk flexor control. Dotted lines indicate trunk starting position, solid lines indicate trunk final position, arrows indicate movement direction, and plus signs indicate muscle groups primarily responsible for control of pattern.

(Skeletal muscle activity occurs on both sides of the trunk; that is, reciprocal innervation.)

As a rule of thumb the trunk flexors are activated in the seated position when the shoulders move posterior to the hips (backward sway), when the trunk is moving away from the support surface (supine starting point), and during rotational activities.

Trunk extensor control

The following four movement patterns are used to evaluate trunk extensor control during seated activities and bridging.

1. To start this movement pattern, the patient assumes a flexed spine posture with a posterior pelvis tilt (the resting posture for many stroke survivors). The patient initiates the movement with the lower trunk and pelvis⁴⁰ and assumes an extended spine posture with a neutral to slight anterior tilt, which accentuates the lumbar curve (Fig. 7-5, A). The patient completes the movement pattern by a concentric contraction of the trunk extensors, which is the trunk pattern required for forward reach.

2. Patients assume an aligned, seated starting position and are asked to keep their spine straight as they lean forward, keeping the shoulders in front of the hips in the sagittal plane (see Fig. 7-5, B). They assume this posture by an eccentric contraction of the trunk extensors,^5,25,50 and if they hold the posture between the middle to end range, the back extensors isometrically contract. The patient has unilateral weakness if the trunk moves forward asymmetrically. Unilateral weakness causes the weaker side to lead the movement pattern (e.g., to fall into gravity). If the movement continues in a forward direction (e.g., patient reaches down to the floor), the back extensors become inactive at the end range, and the tension of the vertebral ligaments maintains the position.⁵

3. While patients are in the end posture of the second movement pattern, the therapists asks them to move their shoulders back to assume a seated, aligned position (see Fig. 7-5, C). To assume this posture, the trunk extensors contract concentrically, although the hip extensors initiate the movement;^5,50 this movement occurs in the sagittal plane.

4. The therapist also should test the back extensors by observing the patient in a bridge posture. While the patient is in a supine position with the hips and knees flexed, the therapist asks the patient to assume a bridge position, which is accomplished by a concentric contraction of the back and hip extensors¹⁶ and is maintained by an isometric contraction of the same muscles. The release of the posture is controlled by eccentric contraction of the back and hip extensors.

Figure 7-5 Trunk extensor control. Dotted lines indicate trunk starting position, solid lines indicate trunk final position, arrows indicate movement direction, and plus signs indicate muscle groups primarily responsible for control of pattern.

(Skeletal muscle activity occurs on both sides of the trunk; that is, reciprocal innervation.)

As a rule of thumb, in the seated posture the back extensors are active during anterior weight shifts (in which the shoulders move in front of the hips), during correction of posture to a position of alignment from an anterior weight shift, and during bridging activities.

Control of the lateral flexors

Lateral flexion occurs in the coronal plane; therefore, a balance of control between the flexors and extensors is required to maintain movement. Electromyographic studies have demonstrated that dorsal and ventral muscles coactivate during lateral flexion.⁵⁰ Electromyographic activity of the right and left erector spinae has been documented during lateral trunk flexion.^5,25

Mohr⁴⁰ states, “Two different movement strategies occur when you reach down to the side: (1) the initiation may occur in the upper trunk and the ipsilateral spine shortens, or (2) the movement can be initiated with your lower trunk and pelvis, resulting in ipsilateral elongation.” Three movement patterns are used to evaluate control of lateral trunk flexion:

1. The first movement pattern is initiated from an aligned, seated position. The pelvis remains stable, and the upper trunk initiates lateral flexion toward the floor with the shoulder approximating the hip (Fig. 7-6A). The end posture (one of ipsilateral trunk shortening) occurs by an eccentric contraction of the side of the elongating trunk.^40,50 In Fig. 7-6, A, the right side of the trunk is shortening, but the predominant control is on the left side, which is elongating eccentrically. Holding this posture between the middle and end ranges allows evaluation of isometric lateral flexion control. Therapists should evaluate both sides of the trunk using this movement pattern.

2. While patients are in the end position of the first movement pattern, the therapist asks them to realign themselves by sitting up straight (see Fig. 7-6, B). The trunk is realigned by a concentric contraction of the lateral flexors³⁸ (the left lateral flexors in Fig. 7-6, B).

3. The last movement pattern evaluates lateral flexion, which initiates the movement from the lower trunk and pelvis.⁴⁰ This movement pattern allows reach beyond the arm span in the frontal plane. During this movement, the majority of weight is shifted to one ischial tuberosity; the shoulder and hip approximate in this pattern. In the resulting posture the trunk is elongated on the weight-bearing side, and trunk shortening occurs on the nonweight-bearing side (see Fig. 7-6, C). The predominant control comes from concentric contraction of the lateral flexors on the shortening side. Fig. 7-6, C, illustrates the contraction on the right side of the trunk. The therapist must evaluate both sides of the trunk.

Figure 7-6 Lateral flexor control. Dotted lines indicate trunk starting position, solid lines indicate trunk final position, arrows indicate movement direction, and plus symbols indicate muscle groups primarily responsible for control of pattern.

(Skeletal muscle activity occurs on both sides of the trunk; that is, reciprocal innervation.)

Rotation control

Concerning rotation control, Kapandji³³ states, “Rotation of the vertebral column is achieved by the paravertebral muscles and the lateral muscles of the abdomen. Unilateral contraction of the paravertebral muscles causes only weak rotation... During rotation of the trunk, the main muscles involved are the oblique muscles. Their mechanical efficiency is enhanced by their spiral course around the waist and by their attachments to the thoracic cage away from the vertebral column, so that both the lumbar and lower thoracic vertebral columns are mobilised.” During rotation of the trunk to the left, the right external and left internal obliques are activated (Fig. 7-7). The fibers of both of these muscles run in the same direction and are synergistic. Basmajian’s⁴ review of the literature on electromyography demonstrates that bilateral activity in the extensors at the thoracic level is evident during rotation.

Figure 7-7 Rotation control. IO, Internal oblique; EO, external oblique.

(From Kapandji IA: The physiology of the joints, vol 3, The trunk and vertebral column, New York, 1974, Churchill Livingstone.)

Mohr⁴⁰ states, “Stroke patients will very rarely rotate because normal rotation requires extensors and flexors to be active simultaneously on opposite sides of the trunk.” Rotational trunk control depends on muscle fixation on one side of the trunk, resulting in efficient muscle action on the opposite side.

Trunk rotation can occur in two positions: flexion with rotation and extension with rotation.⁷ Mohr⁴⁰ points out that rotation can be initiated by the upper trunk or the lower trunk/pelvis. Rotation control is evaluated by five movement patterns:^8,40

1. In the first movement pattern, the patient sits upright, and the pelvis remains stable on the support surface. The patient reaches across midline so that the shoulder moves toward the opposite hip (e.g., reaching with the right arm across the body toward the floor). The result is a position of flexion and rotation. The primary control is by concentric contraction of the obliques and contraction of the back extensors (especially at the thoracic level). The therapist must evaluate both sides of the trunk.

2. In the second movement pattern, the upper trunk remains stable, and the lower trunk and pelvis initiate a forward movement on one side (e.g., scooting forward). The result is a position of extension with rotation.

3. In the third movement pattern the patient reaches behind at the shoulder level (upper trunk initiation), and the resulting posture is rotation and extension.

4. The fourth movement pattern involves initiating a backward shift with the lower trunk and pelvis (scooting backward) while shifting to one side and rotating the opposite side posteriorly; this posture is flexion with rotation.

5. The final movement pattern is similar to a pattern reviewed in the section on trunk flexion control. The patient is supine and initiates a segmental roll by lifting the shoulders up from the support surface and toward the opposite side of the body. This pattern is controlled by a concentric contraction of the abdominal muscles (the obliques).

Trunk control during activities of daily living

There is a clear relationship between the loss of trunk control and the loss of functional independence. Conclusions from empirical research include:

The previous section focused on select movement patterns of the trunk. Evaluating the trunk in this manner is useful for identifying specific problem areas and focusing treatment plans. However, the impact that impaired trunk control has on functional tasks is more relevant to all rehabilitation professionals. Most, if not all, of the reviewed movement patterns (and combinations of them) are used during ADL performance. Therefore, the evaluation of trunk control can take place during skilled observations of ADL.

For clarification, an infinite number of variations are observed in movement patterns during task performance. Therefore, the focus of evaluation and treatment should be on observing, evaluating, and treating the patient in a variety of different environments and with tasks that include multiple variables. The situational context and task demands determine which components of trunk control are necessary for successful task performance. Box 7-2 has an example of task variables that affect trunk control patterns.

The list of trunk control variations during ADL performance in the following section are not considered exhaustive, but are guidelines for observing trunk patterns and inherent variations during various tasks. The reader should mimic performing each task to ensure understanding of the posture descriptions (Table 7-10).

Table 7-10 Trunk Control to Support Participation

Upper extremity dressing

Pullover shirt.

Putting on a pullover shirt requires the following movements:

Trunk flexion: Required for the patient to manipulate the shirt in the lap and reach down toward the lap to insert an arm into the sleeve

Trunk extension: Observed as the patient realigns the trunk, continues to pull up the sleeve, and inserts the head into the shirt

Trunk rotation with extension: May be necessary for reaching posteriorly and adjusting the orientation of the shirt and/or tucking the shirt into the pants

Button-down shirt.

Putting on a button-down shirt requires the following movements:

Trunk flexion: Used to orient the shirt correctly on the lap for preparation of donning and to guide the arm into the sleeve when the trunk is inclined forward

Trunk extension: Required to realign the trunk from the previous position

Trunk rotation with extension: Used to reach with the more functional arm behind the head and to the opposite shoulder to grasp the collar of the shirt and pull it to the opposite side (Fig. 7-8); also used to move the second arm through the sleeve and tuck the shirt into the pants

Trunk flexion: Used as the patient attempts to button the shirt; more often used as relaxation position (a slumped posture) rather than an active flexion pattern

Figure 7-8 Trunk control during upper extremity dressing.

Lower extremity dressing (seated).

Putting on pants, underwear, shoes, and socks requires the following movements:

Trunk flexion: Required to reach down toward the feet (Fig. 7-9)

Trunk rotation with flexion: Required to reach the more functional arm toward the opposite foot

Trunk extension: Required to realign the trunk from the previous positions

Lateral flexion: Required when using a crossed-leg method to don/doff pants, underwear, or footwear (the crossed-leg position shifts the patients’ center of gravity posteriorly, placing increased demand on the abdominal muscles [i.e., controlling the trunk in flexion while preventing a posterior fall]) (Fig. 7-10); also required to pull pants and underwear up or down over the buttocks and hips successfully

Figure 7-9 Trunk control during lower extremity dressing.

Figure 7-10 Trunk adjustments during lower extremity dressing.

Eating

Eating requires the following movements:

Trunk flexion and extension: Used in varying degrees with a hand-to-mouth pattern in which an anterior weight shift of the trunk toward the table occurs (Fig. 7-11) to position the mouth over the plate as food enters. (The degree to which this weight shift occurs depends on the type of food being eaten. Food that is hot or liquid requires increased flexion toward the plate or bowl. The increased flexion reduces the distance the food must be transported, thereby reducing spillage opportunities.)

Trunk rotation: May be used in flexion and extension to reach for condiments that are across the midline of the trunk

Lateral flexion: Lower trunk initiation may be used in reaching for condiments that are positioned to the side of the place setting and beyond arm’s length and also may be used with trunk rotation postures (Fig. 7-12); upper trunk initiation may be used when reaching for an object that drops on the floor to the side of the patient.

Figure 7-11 Trunk control while eating.

Figure 7-12 Trunk adjustments while reaching for utensils or condiments.

Assuming an appropriate starting posture

Before initiation of tasks and retraining of trunk control, the trunk must be in a proper biomechanical alignment. Therapists should observe patients anteriorly, posteriorly, and laterally to detect deviations from normal alignment (see Table 7-8).

Physically or verbally cueing patients to assume an appropriate starting posture should place them in a position of readiness for function-optimal symmetry in the trunk that is usually in midline, depending on the task (Box 7-3).

This neutral starting posture is similar to the position the trunk and lower extremities assume when a person begins a typing task.

An aligned and upright trunk posture has been shown to recruit muscle activity in the trunk. Floyd and Silver’s electromyographic studies²⁵ demonstrated that a “slumped” position while sitting (simultaneous trunk flexion and extension of the hip joint) resulted in trunk extensor relaxation. In contrast, sitting upright in a chair without a backrest resulted in increased activity of the erector spinae muscle group. This muscle activity persisted as long as the trunk remained in extension, despite adjustments of the head and shoulders. A “slumped” posture, which consists of trunk flexion, posterior pelvic tilt, and resulting hip extension, is observed commonly during evaluation of posture in the stroke population; the position requires minimal skeletal muscle activity.

Andersson and Ortengren’s review of the literature¹ demonstrated that the position of the feet had an effect on the myoelectric activity of the trunk extensors. Knee flexion (causing the feet to come toward the chair) increased muscle activity in the trunk, whereas knee extension resulted in a decrease in muscle activity. Stroke patients commonly assume a seated posture in which their feet (especially the more affected lower extremity) are positioned on the floor in front of their knees (e.g., in knee extension). The position of the feet tends to have an effect on pelvic tilt and resultant trunk postures. When the feet are positioned under the knees and toward the chair, an anterior pelvic tilt and trunk extension are enhanced. The opposite is also true: when the feet are positioned in front of the knees and the knees are extended, a posterior pelvic tilt and resulting trunk flexion are enhanced.

More recently, there is further empirical evidence that a neutral spine/starring position should be encouraged from both a neuromuscular and functional perspective. Cholewicki and colleagues¹³ demonstrated that antagonistic trunk flexor-extensor muscle coactivation was present around the neutral spine posture in healthy individuals, and this coactivation increased with added mass to the torso. Gillen and colleagues²⁹ examined the effects of various seated trunk postures on upper extremity function. Fifty-nine adults were tested using the Jebsen Taylor Hand Function Test while in three different trunk postures. Significant mean differences between the neutral versus the flexed and laterally flexed trunk postures were noted during selected tasks. Specifically, dominant hand performance during the tasks of feeding and lifting heavy cans was significantly slower while the trunk was flexed and laterally flexed than when performed in the neutral trunk position. Performance of the nondominant hand during the tasks of picking up small objects, page turning, and the total score was slower while the trunk was flexed compared with performance in the neutral trunk position. These findings support the assumption that neutral trunk posture improves upper extremity performance during daily activities, although the effect is not consistent across tasks.

Patients should be encouraged to feel the difference between an aligned and a malaligned posture. The patient should be able to assume an appropriate posture automatically. Demonstration of the effect that a slumped posture has on reaching activities performed with the side less affected by the stroke may be helpful. Patients may realize that the distance and quality of their reach is enhanced when they are sitting in a proper alignment.

Although the use of mirrors for visual feedback may be appropriate for some patients, mirrors should be used with caution for patients with neurobehavioral deficits. Another technique for assisting patients with gaining symmetry is to have the therapist positioned in front of the patient and to assume the patient’s postures to provide feedback for the patient. Therapists should slowly correct their posture, instructing the patient to mimic the movement. The therapist may state, “Keep your shoulders in line with mine” or “Keep your forehead at the same level as mine.”

Mohr⁴⁰ emphasizes use of activities that encourage rotation and lateral flexion to gain midline control: “the active movements of the trunk into rotation and lateral flexion are caused by the same muscles that flex and extend the trunk. The different movements occur as a result of different interactions of these muscles with each other . . . In order for patients to achieve midline postural control, the therapist must work with the patient in the higher levels of lateral and rotational planes of movement.”

Maintaining or increasing trunk range of motion through mobilization and movement

Concerning ROM in the trunk, Mohr⁴⁰ states, “If there is not full range in all trunk movements (flexion extension, lateral flexion, and rotation), it will be more difficult to gain full control of the trunk. Any lack of ROM in the trunk will lead to decreased function.”

Although limited ROM in the extremities is commonly evaluated and treated, the ROM in the spine often is overlooked. After acute strokes, patients lose the ability to shift their weight and make postural adjustments. Evaluating patients who have trunks influenced completely by gravity and who demonstrate only static trunk postures is common. In these cases, prolonged immobilization of the trunk because of loss of control can result in loss of softtissue elasticity, joint play, and ultimately function. These problems, compounded by inappropriate trunk positioning and support in upright postures, lead to a cycle of immobility, soft-tissue changes, loss of range, and impaired functional abilities.

Specific trunk mobilization techniques are beyond the scope of this chapter but are discussed in the literature.^8,16,22,40 Just as therapists train patients to perform self-ROM activities for their extremities, therapists must promote patient awareness of trunk mobility and educate them about specific movement patterns that maintain and/or increase their trunk ROM. Following are examples of movement patterns that patients can perform to meet this goal:

Using various postures

Therapists may use various postures as an adjunct treatment during patients’ performance of functional tasks. Therapists should select postures based on specific patient needs. The chosen posture should accentuate and challenge the movement and control patterns interfering with independent performance of life activities. If the patient is not engaging in a specific activity (self-care tasks, games, and adapted sports), the use of these postures in isolation is not encouraged. Examples of varying postures include the following:

Conversely, positioning the patient with the knees above the hips (increasing the amount of hip flexion) results in a position of trunk flexion and a posterior weight shift, which places greater demands on the trunk flexors.

Treating the “pusher syndrome” or contraversive pushing

The “pusher syndrome” is a phrase coined by Davies,¹⁹ who derived the name from what she felt was the most striking aspect of this syndrome: the patient pushes heavily toward the hemiplegic side in all positions and resists any attempt at passive correction (i.e., a correction that would bring the weight toward or over the midline of the body to the unaffected side). This phenomenon has been documented in patients with both right and left hemispheric lesions. Further analysis has revealed that the brain structure typically damaged in patients with pusher syndrome is the left or right posterolateral thalamus.³⁶

Danells and colleagues¹⁵ defined pushing as “resistant to accepting weight on and actively ‘push’ away from the nonparetic side.” The authors identified pushers (n=65) from stroke patients with moderate to severe hemiparesis and examined longitudinal changes in symptoms, level of impairment, and functional independence. Assessments were performed within 10 days postonset, at six weeks, and at three months. The authors found:

Perennou and colleagues⁴² investigated whether the pusher syndrome affects only the trunk for which gravitational feedback is given by somesthetic information, or the head as well (gravitational information given by the vestibular system). The results of their pilot study indicated that that the pusher syndrome does not result from disrupted processing of vestibular information but from a higher-order disruption in the processing of somesthetic information originating in the left hemibody, which could be an extinction phenomenon. The authors felt that this disruption leads pushers actively to adjust their body posture to a subjective vertical bias to the side opposite the lesion.

Pedersen and colleagues⁴¹ also examined the pusher syndrome. The study examined the incidence of the syndrome, the relation of this syndrome to neurobehavioral deficits, and the effect of the syndrome on the rehabilitation process in 327 patients. The study revealed a 10% incidence and found no significant differences in hemineglect or anosognosia in patients with and without ipsilateral pushing. The study discovered that patients who demonstrated ipsilateral pushing required 3.6 weeks longer to reach the same outcome as patients who did not demonstrate ipsilateral pushing. Of note in this study are the Barthel index scores at admission and discharge. On admission, patients who demonstrated ipsilateral pushing scored an average of 13.7 on the Barthel index compared with 46.8 for patients without evidence of pushing. On discharge, the average score for pushers was 43.9 compared with 66.8 for patients without pushing. The discharge scores (in terms of ADL function) of the patients who were pushers were still below the admission scores for patients who did not push.

Davies¹⁷ summarizes the typical signs of the pusher syndrome as the following:

Pushing can be quantified via the Scale for Contraversive Pushing (SCP).³⁵ It is scored 0 to 6 with higher the scores indicating a greater severity of pushing. There are three domains (posture, extension, and resistance) that are assessed for both sitting and standing positions (i.e., six scored items) (Box 7-4). Using a cut score of greater than 0 in each section appears to increase the agreement of clinical and SCP observations and has been suggested.⁴

Unfortunately at this point, specific interventions are based on anecdotal evidence only, but they may still be helpful to clinicians. Davies¹⁷ recommends the following specific treatments for the pusher syndrome:

Karnath and Broetz³⁷ recommend the following intervention sequence:

A hands-on approach does not seem to be effective with patients who have the pusher syndrome; therapists’ attempts to assist patients with gaining midline by handling is met by further patient resistance. Manipulating the environment and providing external cues (verbal) seem to be more effective. Examples include the following:

Further empirical intervention studies are needed in this area of practice.

Engaging in reaching tasks

Therapists can use placement of objects in reaching activities as a way to place a variety of demands on the trunk. The key to eliciting a trunk response is to place the object slightly beyond the arm’s reach. Fisher²⁴ observes that when subjects without brain injuries reach, they anteriorly tilt the pelvis, slightly extend the upper back, and move the trunk in the direction of the arm. Patients with brain injuries do not incorporate trunk movements into arm movements and reach only to arm’s length as they maintain a slumped posture.

Therapists have the ability to control the desired response by the way they set up the activity. Setting up the activity includes placing items required during ADL in specific places, choosing the appropriate environment (e.g., kitchen with upper and lower shelves, bookcase, desk space, meal table), deciding how far beyond the arm span the activity should be placed, and deciding on the characteristics of the objects the patient is reaching for (number of objects, weight of the objects, and whether objects require one or two hands). Table 7-11 includes examples of activity placement and resulting trunk response.

Table 7-11 Examples of Grading Activities during Reaching Tasks

Dean and Shepard¹⁸ investigated, via a randomized placebo-controlled trial of task-related training after stroke, the effect of a training program designed to improve the ability to balance in sitting after stroke. The training program was designed to improve sitting balance and involved emphasis on appropriate loading of the affected leg while at the same time practicing reaching tasks using the unaffected hand to grasp objects located beyond arm’s length. The reaching tasks were performed under varied conditions. Changing the location of the object that the subject was reaching for varied the distance and direction of reach. Seat height, movement speed, object weight, and extent of thigh support were also varied. Increasing the number of repetitions and complexity of the tasks advanced the training. The authors found that after training, subjects were able to reach faster and farther, increase load through the affected foot, and increase activation of affected leg muscles compared with the control group (highlighting the critical contribution of the lower extremities in promoting sitting balance). The experimental group also improved in sit to stand. The control group did not improve in reaching or sit to stand, and finally, neither group improved in walking. See Tables 7-9 and 7-11 for suggestions to grade reaching tasks.

Handling

Handling is a common technique used in the clinic. This intervention commonly is associated with neurodevelopmental therapy⁷ and may allow the patient to feel the desired movement pattern, gain range, assist weak movement patterns, and provide external support to prevent falls. To be effective, the patient must be aware of the goal associated with handling, and the therapist should use handling within the context of a functional task. In addition, the handling from the therapist should be graded to allow the patient to perform as much of the movement pattern as possible. A variety of texts that review specific techniques of handling are available.^{7,8,16,17,22,32} Although handling is commonly used, research does not support use of this technique (see Chapter 6).

Using activities of daily living and mobility tasks

Clearly the most effective tools that occupational therapists can use to help patients regain trunk control are self-care, instrumental ADL, and mobility tasks.

The therapist first must perform a thorough evaluation as described in previous sections. Following the evaluation, therapists and patients should identify the most problematic movement patterns that occur during the patients’ daily activities. At this point, therapists use their activity analysis skills to choose appropriate tasks that incorporate the desired patterns and postures as described previously. For example, if the identified problematic patterns are lateral flexion and lateral weight shifts, therapists may choose the following activities for the patient to practice:

The majority of ADL and mobility tasks encompass a variety of postures and movements. For the mentioned strategies to be effective, therapists should initially focus patients’ attention on the desired components of trunk movements. As the patient progresses, the obvious goal is for the trunk responses to be relearned and become automatic.

Therapeutic exercise

A recent pilot randomized controlled trial⁵⁵ (N = 33) documented that conventional occupational and physical therapy combined with additional trunk exercises aimed at improving sitting balance and selective trunk movements has a positive effective on the selective performance of lateral flexion after stroke. Specifically, the experimental group received 10 hours of additional trunk exercises that included:

Adapting the environment

Some patients may have little improvement in trunk control. Environmental adaptations are necessary for these patients to enhance independent performance. Examples include

Use of outside supports can help maintain trunk stability while the extremities are engaged in functional tasks. Supports such as lateral supports, anterior chest straps, arm chairs, pillows and cushions for propping, and lap trays are examples of equipment used to compensate for compromised trunk control (see Chapter 26).

Rearrangement of the environment can decrease demands on the trunk. Placing required equipment within the patient’s reach (arm reach) not only increases independence but also may prevent falls. Storing dishes on the counter instead of in a cabinet, placing utensils in front of the patient, and keeping grooming items on top of the sink instead of in a medicine cabinet are examples of this strategy.

Provision of adaptive equipment is a common strategy to increase independent performance and minimize safety risks. ADL equipment issued to compensate for poor trunk control may include the following: long-handled shoe horns, elastic laces, adapted bath brushes, soap on a rope, reachers, tub seats, and commodes. See Chapter 28 for more information on adaptive devices.

Home modifications such as grab bars and bed rails also may be indicated. See Chapter 27 for a full review of home adaptations and equipment recommendations.

CASE STUDY Regaining Trunk Control after a Stroke

S.G. is a 64-year-old female who came to the rehabilitation unit after a right middle cerebral artery cerebrovascular accident. The following data were collected from the initial evaluation (specific to the trunk):

Sensation was intact.

Static postural malalignments included posterior pelvic tilt, retracted left rib cage, and increased weight-bearing on left ischial tuberosity.

Dynamical posture was difficult to assess because the patient was afraid to move. The patient could not reach beyond her arm span when reaching with her right arm. She had a tendency to fall backward and to the left during lower extremity dressing and when reaching for objects behind her.

S.G.’s personal occupational therapy goals were to be able independently to don her shoes, be able to reach for objects she had dropped without falling, and decrease the amount of spillage that occurred during meals.

The initial treatment plan included adapting her wheelchair with lateral supports (which were removed when she was supervised by friends, family, and staff) and a lumbar roll to maintain optimal alignment while functioning in the wheelchair, trunk mobilizations (specifically in the directions of extension and lateral flexion to both sides), activities to recruit abdominal activity (rolling, games that encouraged trunk rotation, reaching for objects positioned overhead and behind her), and activities that presented unexpected challenges to her trunk control (e.g., balloon volleyball and catch). During the initial stages of treatment, the therapist sat behind S.G. in a straddle position, which increased feelings of security and allowed the therapist to provide outside assistance with difficult patterns and to prevent falls. S.G. was observed while eating, and the therapist noted that she did not shift her weight anteriorly when bringing food to her mouth. Instead she kept her trunk supported against the back of the chair. Because of the increased distance the food had to travel to reach her mouth, spillage was considerable, especially of liquids on a spoon (e.g., soup and cereal milk). S.G. was trained to shift her weight forward as she brought food to her mouth. Although spillage still occurred, it happened less frequently, with the food falling to the plate or table and not in her lap.

As S.G. progressed, the therapist was able to sit next to or in front of her during activities. She engaged in graded reaching activities in which the distance S.G. was required to reach was increased progressively. Activities included reaching to lower shelves in the refrigerator and reaching for objects positioned at specific levels (e.g., knee level, midshin level, and floor level).

On discharge from inpatient rehabilitation, S.G. was able to perform all basic ADL with distant supervision and without assistive devices, reach to the floor when she propped her upper trunk with her more affected forearm against her knees, and eat independently by using a rocker knife to cut her food, with spillage occurring only 10% of the time.