Background and Purpose. This case report documents a rare opportunity to observe the motor function of an individual for nearly 6 months following a primary pontine hemorrhage in the medial pontine tegmentum of the brain stem. The purpose of this report is to illustrate how knowledge of the location of the hemorrhage, in conjunction with knowledge of brain-stem structure-function relationships, informs physical therapist examination and intervention. Case Description. RM, a right-handed 81-year-old man with hypertension, had a hemorrhagic brain-stem stroke that severely compromised control of posture and whole-limb movements. Some residual ability to use the right hand and fingers remained, provided the trunk and right upper arm were stabilized. RM had undiminished intellectual abilities and unaltered memory because of sparing of cerebral cortices. RM's cognitive abilities, however, were obscured by severe impairments in interpersonal communication because of extensive damage to cranial nerve structures. Computed tomographic scans verified that the hematoma crossed the midline and was confined to the medial pontine tegmentum. Discussion. We interpret motor deficits resulting from stoke in the medial pontine tegmentum in terms of damage to brain-stem descending motor systems and ascending somatosensory systems. Recognition of cognitive and residual motor abilities following brain-stem stroke can aid in the development of rehabilitation strategies.

Recent autobiographies of people who have survived a brain-stem stroke illustrate the unique challenges they face.13 Often, people with brain-stem stroke are left with their cognitive abilities fully intact, but with an inability to control movements of their body. In contrast, strokes that affect cerebral cortices (or their output fibers in the internal capsule) often result in both cognitive and motor impairments. This case report documents motor impairments of an individual following a primary pontine hemorrhage bilaterally in the medial pontine tegmentum of the brain stem. This case is unique because most individuals with lesions in this location do not survive, or they remain unconscious. A review of neuroanatomic, neurophysiologic, and pathophysiological studies of descending motor pathways and clinical manifestations resulting from brain-stem versus cerebral cortical stroke provides the background and support for the motor deficits described in the case report. The case report emphasizes the importance of taking into account the location of the lesion and known structure-function relationships in the clinical management of people with brain-stem stroke.

Functional Organization of Descending Motor Systems

The now classic studies of the Dutch neuroscientist Kuypers and colleagues47 have provided a conceptual framework for functional organization of descending motor pathways. The studies demonstrated that cerebral cortical motor areas and brain-stem nuclei in nonhuman primates give rise to descending motor pathways that are functionally distinct. Kuypers classified brain-stem motor pathways into medial and lateral systems. The medial system includes the reticulospinal, vestibulospinal, interstitiospinal, and tectospinal tracts, which originate from the medial reticular formation, the vestibular nuclei, the interstitial nucleus of Cajal, and the superior colliculus, respectively. Medial-system pathways project bilaterally, via ventral columns, to ventromedial regions of the ventral horn where motoneurons for axial and proximal muscles are located. The lateral system includes the rubrospinal tract, which originates from the red nucleus. The rubrospinal tract projects contralaterally, via lateral columns, to dorsolateral regions of the ventral horn where motoneurons for distal muscles are located. Cerebral cortical motor areas influence spinal circuitry directly via corticospinal tracts and indirectly via medial and lateral brain-stem systems. The ventral corticospinal tract projects, via ventral columns, to ipsilateral spinal cord segments and then terminates bilaterally in ventromedial regions of the ventral horn. The lateral corticospinal tract projects contralaterally, via lateral columns, to dorsolateral regions of the ventral horn, where its terminations largely overlap those of the lateral brain-stem system. Based on the combined evidence of neuroanatomic and behavioral observations, Kuypers concluded that the medial brain-stem system is the basic motor system upon which controls exerted by cerebral cortical motor areas and the lateral brain-stem system are superimposed.7

Studies conducted over the past 3 decades have provided ample evidence in support of Kuypers' view of descending motor pathway organization. All major components of the medial brain-stem system, for example, contribute importantly to control of posture and whole-body movements. Vestibulospinal fibers in the medial longitudinal fasciculi (MLF) project to extensor and lateral neck motoneurons and, together with interstitiospinal projections, function to stabilize the head in space.8,9 Vestibulospinal fibers from the lateral vestibular nuclei excite ipsilateral extensor motoneurons of the limbs and trunk,10 and they are important for maintaining upright posture and for extending the limbs when falling.8 Reticulospinal neurons receive input from various sources, including peripheral afferents and the superior colliculus, vestibular and deep cerebellar nuclei, and cerebral cortical motor areas.11 The medial reticulospinal system contributes to the control of behaviors that involve synergistic activation of broad groups of muscles, such as neck and vestibular reflexes, orienting responses, and locomotion.1113 Tectospinal fibers originate from large cells in deep layers of the superior colliculus,14 which are important for orienting the eyes, head, and trunk to visual, auditory, and somatosensory stimuli.15,16 Thus, the medial system provides the basic control of posture and whole-body movements upon which cerebral cortical motor areas organize more highly differentiated movement (eg, for integrating voluntary limb movements with posture in cats12 and for control of relatively independent finger movements in primates6,1719). The lateral system preferentially influences distal muscles2023 and provides control of voluntary movements of the arm, hand, and fingers for reaching and manipulating objects in primates5,22,2426 and for contact placing27,28 and for limb trajectory and foot placement during locomotion in cats.13,2932

In summary, a large body of work has firmly established the basic functional organization of descending motor pathways. In addition, recent findings have provided significant new insights in the contribution and functional overlap of different descending pathways important for the control of multijoint, coordinated movements such as locomotion and reaching to grasp and have provided new perspectives on Kuypers' classic view of descending motor pathway organization.

Brain-Stem Versus Cortical Stroke

People with brain-stem stroke are a minority of a larger group of people who have sustained strokes. Stroke is one of the leading causes of adult disability and is our nation's third leading cause of death behind heart disease and cancer.33 Each year, more than 750,000 Americans have a first or recurrent stroke,34 resulting in an age-adjusted mortality rate of 25.1 deaths per 100,000 people in the population in 1998.33 In a recent study,35 83% of strokes were ischemic, 10% were intracerebral hemorrhages, and 7% were subarachnoid hemorrhages. Many fewer people have brain-stem stroke as compared with cortical stroke. Primary pontine hemorrhage, for example, accounts for less than 8% of incidences of intracerebral hemorrhage (7.5%, 50/667 cases over a period from 1935 to 196436; 7.9%, 61/771 cases over a period from 1985 to 199037). Moreover, brain-stem stroke is associated with a much higher mortality rate as compared with cortical stroke because ascending and descending projections from the reticular formation and vital cardiovascular and respiratory centers are located in the brain stem. Primary pontine hemorrhage, for example, is highly fatal, with overall case mortality rates as high as 61%37 to 75%.36 The prognosis of primary pontine hemorrhage, however, depends on the size, location, and extent of the hematoma. Bilateral lesions involving the medial pontine tegmentum were a minority of cases (14%, 7/50 cases36; 11.5%, 7/61 cases37) and had the lowest case survival rates (0/7 cases, survival 2–10 days36; 14.3%, 1/7 cases37). Furthermore, brain-stem strokes are frequently abrupt in onset and produce coma, which precludes study of associated motor deficits. Coma is especially common following bilateral lesions that involve the medial pontine tegmentum.38 Because of low survival rate and poor prognosis, few people with brain-stem stroke enter rehabilitation, and most accounts of motor deficits and rehabilitation following stroke concern people with cortical stroke rather than brain-stem stroke.

Clinical manifestations following cortical stroke and brain-stem stroke differ. Following stroke in cerebral cortices (or their output fibers), people often exhibit hemiplegia or hemiparesis, upper motor neuron facial weakness, hemi-somatosensory loss, and loss of vision in one hemifield.39,40 Weakness and somatosensory loss are most prominent on the side of the body contralateral to the damaged cortex, and typically distal muscles are more strongly affected than proximal muscles.41 Higher-level perception and cognition also may be deficient following a cortical stroke, depending on which regions of the cerebral cortex are damaged. Stroke of the dominant hemisphere, for example, frequently results in aphasia,42 whereas stroke of the nondominant hemisphere frequently results in anosognosia and contralateral hemineglect.43 Following brain stem stroke, coma or death are common. If a patient does survive, a seemingly small lesion often has devastating consequences because many nuclei and neural pathways, including cranial nerve nuclei, descending motor pathways, ascending somatosensory pathways, and widespread ascending and descending projections from the reticular formation, are densely packed in the brain stem. Motor deficits and signs following brain-stem stroke may include contralateral hemiparesis and ipsilateral lower motor neuron facial weakness or sensory loss.44,45 Motor deficits following brain-stem stroke, however, also may be bilateral and, depending on the extent and location of damage, may include one or several of the following signs: quadriplegia, pupillary changes, diplopia, gaze palsies, internuclear ophthalmoplegia, dysphagia, dysarthria, vertigo, and ataxia.44,4648 Following cortical stroke, axial and proximal limb muscles often remain relatively more functional than distal muscles because the corticospinal system primarily innervates distal musculature. In contrast, a brain-stem stroke may result in the reverse situation. That is, following a brain-stem stroke, control of axial and proximal limb musculature may be severely affected, whereas control of distal musculature may be relatively more spared.

The neural bases underlying therapy-induced improvements in motor function following stroke are, at present, incompletely understood. A large body of evidence from neurophysiologic, neuroanatomic, and neuroimaging studies in animals and humans supports the view that cerebral cortical circuitry is highly plastic.4951 The demonstrated plasticity in neuronal circuitry may provide a scientific basis for commonly used therapeutic approaches following cortical stroke. For example, techniques such as constraint-induced movement therapy5254 and repetitive rehabilitative training of the impaired limb55,56 are thought to promote functional reorganization following neuronal damage. In addition, robot-aided neurorehabilitation reduces impairment and has a positive effect on reorganization of the adult brain by concomitantly controlling the amount of therapy delivered to a patient and measuring the patient's performance.57,58 Furthermore, transcranial cortical stimulation,59 cognitive rehabilitation,60,61 neuromuscular stimulation,62,63 biofeedback therapy,64,65 and motor imagery66 are all based on the idea that sensorimotor stimulation will enhance cortical reorganization following injury, thereby improving motor function. Although much less is known about plasticity of brain-stem circuitry and use-dependent reorganization of neural circuitry following brain-stem stroke, current therapeutic approaches following brain-stem stroke are similar to those following cortical stroke (eg, Hummelsheim and Eickhof67). Clearly, a more refined understanding of the neural bases underlying cortical and brain-stem strokes is needed to optimize physical therapist examination and intervention.

Our case is an excellent one to study the unique manifestations of a brain-stem stroke and the intervention considerations required. The case exemplifies a cognitively, socially, and emotionally intact individual with extremely limited motor abilities who needed special consideration.

Case Description

Patient History

“RM” was a right-handed 81-year-old man with hypertension and poorly controlled atrial fibrillation that was managed with anticoagulant medications on a long-term basis. He was doing quite well until the morning of the day he was admitted for emergency care with headache, diaphoresis, dizziness, diplopia, sudden onset of right arm tingling, numbness, and weakness, followed by progressive slurred speech. Computed tomographic (CT) scans of the head showed progressive hemorrhagic stroke intrinsic to the pontine tegmentum of the brain stem, with rupture into the fourth ventricle (Fig. 1A). The observed signs of damage to cranial nerve structures and ascending somatosensory pathways are summarized in the Table. Pupils were equal in size and reactive to light. Horizontal eye movements and conjugated gaze were severely restricted as a result of bilateral abducent nerve paralysis. Vertical eye movements were normal. The jaw was deviated to the right. He showed bilateral facial weakness (right greater than left) and difficulties wrinkling the forehead and closing the eyelids. He had severe dysphagia. His oral pharynx was dry. His tongue and palate moved normally. His gag reflex was good. Respirations were of the Cheyne-Stokes type. Tactile discrimination and sensation of position of limbs were impaired on the left and intact on the right. A distorted and heightened reaction to noxious stimuli was noted on the left. Bilateral segmental static reflexes (eg, flexor withdrawal and crossed extension) were present. Deep tendon reflexes were decreased. Extensor plantar reflexes (Babinski sign) were negative. Automatic labyrinthine and neck reflexes were greatly impaired or absent. Although RM was oriented, attentive, and cooperative, his level of arousal fluctuated. At times, he was somnolent and difficult to arouse. Verbal expression was difficult due to dysarthria. Audition and oral comprehension were normal, and no aphasia or signs of cognitive impairment were evident.

Figure 1.

Location and extent of RM's hemorrhagic stroke. (A) Head computed tomography (CT) scan showing the extent of hemorrhage in the medial pontine tegmentum of the brain stem (arrow). Abbreviations: IV=fourth ventricle, CB=cerebellum. (B) Placement of a section through the sella turcica and the caudal third of the fourth ventricle that corresponds to the plane of the CT scan in panel A and the histological section in panel C. The plane of section (thick line) is at a 25-degree angle with respect to the horizontal stereotaxic plane (thin line). (C) Histological section in the plane of the CT scan in panel A. The superimposed white lines indicate the extent of hemorrhage reconstructed from 2 successive CT scans. The broken white line corresponds to the CT scan in panel A; the solid white line corresponds to a CT scan at a level 6 mm superior to that of the CT scan in panel A. Abbreviations: AbdNu=abducent nucleus, AbdNr=abducent nerve roots, CTT=central tegmental tract, CST=corticospinal and corticobulbar tracts, FacNr=facial nerve, FacNu=facial nucleus, ForVen=fourth ventricle, LL=lateral lemniscus, LVN=lateral vestibular nucleus, MCP=middle cerebellar peduncle, ML=medial lemniscus, MLF=medial longitudinal fasciculus, PonNu=pontine nuclei, PonRet=pontine reticular formation (inferior part), RaNu=raphé nuclei, RST=rubrospinal tract, SL=spinal lemniscus (spinoreticular, spinotectal, and spinothalamic tracts), SONu=superior olivary nucleus, SpTTr=trigeminal nerve spinal tract, SYNu=superior vestibular nucleus, TecSp=tectospinal tract, TrapB=trapezoid body, TriMoNu=trigeminal motor nucleus, VesCochNr=vestibulocochlear nerve roots. The diagram in panel B is reproduced with permission from Hanaway J, Scott WR, Strother CM. Atlas of the Human Brain and the Orbit for Computed Tomography. St Louis, Mo: Warren H Green Inc; 1977. The histological section in panel C is provided courtesy of the Digital Anatomist Program of the Department of Structural Anatomy of the University of Washington.


Localization of the Hemorrhage in the Medial Pontine Tegmentuma

RM's movement dysfunction, described in detail below, became progressively more evident over the first few days following the stroke. Upon stabilization of his condition, he was unable to walk, stand, or sit upright. Although some ataxia of the left upper extremity was noted, he did not develop dysmetria, dysdiadochokinesia, intention tremor, or asynergia. He showed good judgment concerning his functional limitations. Post-stroke depression was treated with antidepressant medication (Zoloft,* 100 mg/d). Over the next few months, RM developed a gross, low-frequency, rest and intention tremor of the head, neck, trunk, and left extremities, termed “Claude syndrome” or “rubral tremor.”68,69 Approximately 6 months following the initial stroke, RM had another stroke and died of pneumonia.


Clinical findings were obtained from emergency department, hospital, and nursing home records; from neurological examinations; from physical therapy, occupational therapy, and speech-language pathology reports; and from testing described below. RM received physical therapy, occupational therapy, and speech therapy from licensed therapists for a total of 2 to 3 hours of therapy a day, 5 days a week, from 3 days to approximately 4½ months post-stroke. Therapy sessions were spaced throughout the day to minimize fatigue. When improvement appeared to reach a plateau, he was discharged from therapy but continued to receive restorative nursing intervention, including training in activities of daily living (ADL) and mobility, range of motion (ROM), and splint management. Generally, RM was cooperative and motivated during examination and therapy sessions. He responded to questions and commands correctly and without delay, and he called for assistance when necessary. His emotional and social responses were appropriate. He became understandably frustrated with his inability to successfully perform some tasks but exhibited good abstract thinking and problem-solving abilities in devising alternative strategies. For example, he was resourceful in using environmental support to stabilize his body or objects he manipulated, and when unable to accomplish a task one-handed, he used both hands. Physical therapist examination detailed below was performed 3 months after his stroke, at Meriter Health Center, Madison, Wis.

Posture and Whole-Body Movements

Neuronal damage associated with the hemorrhage in the medial pontine tegmentum resulted in severe deficits of control of posture, balance, and locomotion. Rolling and transitions from sitting to supine postures were poorly coordinated, and they were not accompanied by sequential eye, head, or trunk movements and protective assistance of the upper extremities. RM was unable to sit unsupported, and attempts to use his arms or legs for postural support were ineffective. He sat with a posterior pelvic tilt, increased thoracic kyphosis, increased upper cervical extension, and with his head forward. During transitions from sitting to standing, his hips and knees were poorly coordinated as evidenced by high variability in the temporal sequence of hip and knee extension. Unless he was assisted to keep his hips and knees somewhat flexed, his center of mass tended to shift posteriorly and to the left. Videotape review indicated that self-induced sway as well as postural adjustments in response to externally imposed perturbations were absent. Postural alignment and the ability to remain upright while sitting or standing deteriorated further with the eyes closed. When supported upright by a platform walker that was propelled forward for him, RM was able to take steps. His base of support was wide. He advanced the left leg slowly and stiffly, with lateral (external) rotation and adduction of the hip. He advanced the right leg using hip and knee flexion. Step length and height were small and irregular. Although he accepted weight on the legs with the knees flexed, the knee and hip frequently buckled further into flexion, especially on the right. During stepping, the legs advanced faster than the trunk, thereby further shifting his center of mass posteriorly and to the left.

Independent Limb Movements

Despite severe postural deficits of head, trunk, and limb girdle, RM made relatively good use of the right hand and fingers in isolation, provided the trunk and right upper arm were appropriately stabilized. For example, with environmental support, he could manipulate small objects such as dominoes or simple tools such as a toothbrush. When an object was placed near his right hand in an accessible orientation, he used thumb and forefinger apposition to retrieve and hold on to it. In contrast, he was unable to grasp an object in his left hand, and when placed there it easily slipped from his grip. He did attempt to use his left arm and hand in conjunction with his right arm and hand for bimanual activities, such as retrieving or repositioning an object in the hand, although he used his left limb primarily to assist activities of the right limb. Thus, some functional use of the right hand and fingers remained, yet skillful manipulation of objects with either hand was poor. RM's voluntary limb movements were poorly integrated with posture. Videotape review indicated that elevation of either arm was not preceded by anticipatory postural adjustments of the trunk or shoulder girdle and that pointing or reaching toward visual targets was not accompanied by sequential eye, head, and arm movements. RM pointed without moving his eyes, head, or trunk. Aiming with both right and left limbs was accurate.

Reaching to Grasp

In light of the functional specialization of medial versus lateral descending motor systems reviewed in the preceding text, we considered it important to formally test RM's ability to perform reach-to-grasp tasks. RM and his wife gave permission to videotape the testing. A belt around the waist provided support to maintain an upright sitting posture (Figs. 2A and 2D). Prior to reaching, the arm was held loosely at the side, with the palm down and the forearm pronated. On verbal cue, RM reached to grasp a cylinder (height=15 mm, diameter=25 mm) positioned at arm's length and at shoulder height, using 1 of 2 types of grasp. One grasp, the whole-hand grasp, required an overhand scooping motion of all 4 fingers to retrieve the cylinder from a clear container (height=50 mm, diameter=100 mm). The other grasp, the precision grasp, required thumb and forefinger apposition to retrieve the cylinder from a horizontally oriented slot (width=80 mm, height=30 mm). RM's reach-to-grasp movements were slow and labored with either arm. Figures 2A through 2F show videotaped images and stick-figure reconstructions of the right arm (Figs. 2A2C) and left arm (Figs. 2D2F) during individual trials of performance of the whole-hand task (Figs. 2B and 2E) and precision task (Figs. 2C and 2F). Records of angles of the metacarpophalangeal, wrist, elbow, and shoulder joints are plotted versus time in Figures 2G through 2J respectively, for the trials illustrated in Figures 2B, 2C, 2E, and 2F.

Figure 2.

Kinematics of the right and left upper extremities during performance of the reach-to-grasp task. (A and D) videotaped images showing the right and left limbs during performance of the whole-hand task. (B and E, C and F) stick figures of the upper extremity during task performance. Stick figures of the moving limb were reconstructed and joint angles were calculated from the x-y coordinates of the following landmarks: the head of the humerus, the rotation point of the elbow, the proximal end of the carpals, the proximal phalanges, and the proximal interphalanges. The x-y coordinates for these points were identified on successive video frames of the moving limb (resolution: 33.3 milliseconds). Each panel illustrates an individual trial of performance of the whole-hand task (panels B and E) and the precision task (panels C and F). Individual lines connect the shoulder, elbow, wrist, metacarpophalangeal, and proximal interphalangeal joints and the tip of the index finger. (G–J) Individual trial records of approximate angles of metacarpophalangeal (MCP), wrist, elbow, and shoulder joints are plotted versus time for the trials illustrated in panels B and E and panels C and F. Time 0 corresponds to reach onset. The time at which RM contacted the container (whole-hand task) or slot (precision task) corresponds to reach offset.

During reaching to grasp with the right arm, RM did not extend the wrist to preshape the hand in preparation for grasp (Fig. 2H, solid lines), and he did not supinate the right forearm, regardless of task (whole-hand or precision). Therefore, the right hand was not oriented appropriately relative to the orientation of the target. Despite the impaired preshaping of the right hand, he was able to complete a grasp. In contrast, during reaching to grasp with the left arm, the wrist did extend (Fig. 2H, broken lines) and the forearm did supinate to preshape and orient the hand to the target, regardless of task. Typically, the left hand opened wide but failed to close around the object grasped.

Localization of the Lesion

As neither structural magnetic resonance imaging nor postmortem histology was available, our estimate of the location and extent of the neurological damage that resulted from the hemorrhagic stroke is based on the known structure-function relationships of brain-stem structures in combination with correlation of CT scans of the head with matched histological sections. The CT scan shown in Figure 1A was made in the evening of the day of admission when the extent of hemorrhage was maximal. An outline of the hyperdense area of the CT scan in Figure 1A is overlaid (broken white line) on a matched histological section in Figure 1C. The plane of section (Fig. 1B) is through the sella turcica and the caudal third of the fourth ventricle, at a 25-degree angle with respect to the horizontal stereotaxic plane. The area of hemorrhage was most extensive medially and caudally in the pons, including both the right and left pontine reticular formation. The hemorrhage extended farther into the right half than the left half of the pontine tegmentum, with little or no involvement of the basilar pons.

The extent of the area of hemorrhage agreed well with the observed clinical signs of damage to cranial nerve structures and motor and somatosensory pathways described in the preceding text and summarized in the Table. The oculomotor and trochlear nuclei in the midbrain were not included in the area of hemorrhage, which is consistent with retention of vertical eye movements. The abducent nuclei, the medial longitudinal fasciculi, and the abducent nerve roots, were included in the area of hemorrhage bilaterally, which is consistent with the observed abnormalities in horizontal eye movements and conjugated gaze. The genu of the facial nerve were included in the area of hemorrhage bilaterally, which accounts for the observed facial weakness and difficulties wrinkling the forehead and closing the eyelids. Components of the medial system of brain-stem descending motor pathways passed through or were included in the area of hemorrhage bilaterally, which is consistent with the observed deficits in posture and whole-limb movements. The entire medial lemniscus and parts of the spinal lemniscus were included in the area of hemorrhage on the right, whereas only a dorsomedial portion of the medial lemniscus was included on the left, which is consistent with the observed deficits in somatosensory responsiveness. The raphé nuclei were included in the area of hemorrhage bilaterally, which is consistent with the observed fluctuations in level of arousal. The right central tegmental tract was included in the area of hemorrhage, which is consistent with the observed rubral tremor. The central tegmental tract connects the red nucleus with the ipsilateral inferior olivary nucleus which, in turn, projects to the contralateral cerebellum.70,71

Tests and Measures

Examination results (detailed below) were based on observations of the first author. Although validity and reliability of measurements were not established formally for the purpose of this case report, independent testing and retesting by individual members of RM's rehabilitation team over the 6-month period after the stroke yielded consistent results.


RM's mobility was quantified by the Minimum Data Set (MDS) and the Functional Independence Measure (FIM), administered 2 weeks post-stroke and again 3 months post-stroke. The MDS is an instrument that was developed to rate severity of patient disability and outcomes of medical rehabilitation of individual nursing home residents. Evidence for validity and reliability of measurements obtained with the MDS has been reported.7274 In brief, MDS scores have been validated by correlation with various independently obtained measurements of basic behavioral and mental health functions (eg, Mini-Mental State Examination, Spearman correlation coefficient [r]=.45; Alzheimer's Disease Patient Registry, r=.50; Dementia Rating Scale for ADL, r=.59).72 The MDS scores met a standard for excellent reliability in key areas of functional status, such as cognition, ADL, continence, and diagnoses (eg, Spearman-Brown intraclass correlation coefficients were .4 or higher for 89% of items and .6 or higher for 63% of items).75 RM assisted in but contributed less than 25% of effort required to complete mobility tasks tested, which classified him as requiring maximal assistance as defined by Keith et al.76 The MDS scores for mobility (eg, the ability to move in bed, perform toilet transfers and hygiene while toileting) improved over the 3-month period from requiring maximal assistance of 2 people (3/8) to requiring maximal assistance of 1 person (2/8).

The FIM is another widely used scale that yields valid and reliable measurements of mobility, locomotion, self-care, sphincter management, communication, and social cognition.7780 Its 7-level scale ranges from “total dependence” (0) to “complete independence” (7). RM's bed mobility improved from requiring maximal assistance (2/7) to requiring moderate to maximal assistance (2.5/7). Transfers improved from requiring maximal assistance (2/7) to requiring moderate assistance (3/7). Walking improved from complete dependence (1/7) to requiring moderate to maximal assistance (2.5/7). Grooming, bathing, toileting, and upper- and lower-body dressing improved from requiring maximal assistance (1.7/7) to requiring moderate to maximal assistance (2.5/7). Although the increase in scores over the 3-month interval between tests appears modest, we consider the corresponding improvements in mobility important because they facilitated interactions of RM with his caregivers, thereby reducing the level of frustration RM experienced as a result of his disabilities.

Psychosocial well-being.

Both the MDS and FIM incorporate measures of psychosocial well-being. Scores of dichotomous items such as restlessness, variability in mental status, insomnia, and depression improved from being present to being absent. In addition, the ability to make himself understood and interact with others; participation in planned and structured, and self-initiated, activities; and involvement in the social life of the facility all improved from being absent to being present. RM's improvements in dysarthria were large. Three days post-stroke, he was 20% intelligible in conversation; following 3 months of speech therapy, he was 60% to 70% intelligible, and he used compensatory strategies such as slow, exaggerated articulation and spelling. Speech therapy also addressed his dysphasia. Despite extensive therapy, however, he continued to experience difficulties swallowing while feeding on pureed solids and thin liquids. Therefore, he received his nutrition with enteral feedings through a gastric tube and only under special circumstances (“recreational feeding”), and with careful monitoring was he allowed to consume thin liquids by mouth.


Psychiatric and neurological consultations done approximately 1 year prior to RM's stroke indicated that he had mild cognitive impairment due to early-onset dementia related to age, alcohol abuse, or ischemic accident. Until the morning of his stroke, however, he continued to function well, at home as well as at his work as a parking lot attendant. Three days after the stroke, RM's comprehension of written and spoken words was good, he was able to answer complex yes/no questions with an 85% success rate, and he followed 2-and 3-step commands. Cognitive performance was formally tested using the Ross Information Processing Assessment (RIPA),81 an instrument used to identify, describe, and quantify cognitive-linguistic deficits in geriatric populations. Procedures for establishing norms and evidence for reliability and validity of RIPA scores have been reported.81 Content validity has been established through professional review.81 Analysis of internal consistency reliability yielded alpha coefficients of .88 to .97.81 Concurrent validity and interrater reliability of RIPA scores have not been evaluated. RIPA scores indicated mild impairment in recent memory (20/24), spatial orientation (14/18), and problem solving/abstract reasoning (22/27) and no impairment in temporal orientation (30/30). A diagnosis of mild cognitive impairment was confirmed 6 weeks after the stroke by the same physician consulted 1 year earlier. In conclusion, no evidence indicated that RM's brain-stem stroke had altered his prestroke cognitive abilities.

Somatosensory responsiveness.

Examination of somatosensory responsiveness included testing discriminative touch, pain and temperature sensation, and kinesthesia.82(pp143–148) Discriminative touch and pain and temperature sensation were intact in the right extremities and were decreased in the left extremities. Kinesthesia was intact in the right extremities and absent in the left extremities.

Muscle function.

Muscle tone (ie, resistance to passive stretch) was examined by observing trunk and extremity posture, by palpation, and by monitoring resistance to imposed movements of extremity joints.82(pp183–184) Tone appeared to be decreased in muscles of the trunk and extremities. Manual testing of muscle force, performed according to the procedures described by Kendall et al,83 showed that RM was able to hold his hip joint in a flexed position against moderate pressure (4/5) bilaterally. He was able to hold his knee joint in a flexed position and in an extended position against maximal pressure (5/5) and his ankle joint in a dorsiflexed position against minimal pressure (3+ to 4-/5) bilaterally. On the right, RM was able to hold his shoulder joint in a flexed position and in an abducted position against moderate pressure (4/5). He was able to hold his elbow joint against maximal pressure (5/5) and his wrist and finger joints against minimal pressure (3+ to 4-/5). On the left, he was able to hold his shoulder, wrist, and finger joints against minimal pressure (3+ to 4-/5) and his elbow joint against moderate pressure (4/5).

Range of motion of extremity joints.

Goniometric measurements84 indicated that the range of active supination/pronation and flexion/extension of the elbow, wrist, and finger joints of both upper extremities was normal, except for wrist extension on the right, which was limited to neutral. With trunk and scapular support, isolated active and passive shoulder flexion was limited to 155 degrees on the right and 145 degrees on the left. Without trunk and scapular support, isolated active shoulder flexion was 100 degrees on the right and 80 degrees on the left. Passive ROM of the joints of all 4 extremities was normal, except for right wrist dorsiflexion, which was limited to 20 degrees, and the limitations in ROM of shoulder flexion noted above.


Evaluation of the examination results was in keeping with the task-oriented approach of Shumway-Cook and Woollacott.85 Based on the examination results, we hypothesized that both motor and somatosensory impairments contributed to the severe functional limitations that RM experienced as a result of hemorrhagic stroke in the medial pontine tegmentum. Postural control was largely ineffective because of the absence of both anticipatory and compensatory postural adjustments. Major components of the medial system of brain-stem descending motor pathways (eg, the MLF, which carry vestibulospinal fibers and descending fibers of the interstitial nucleus of Cajal; the predorsal bundles, which carry tectospinal fibers; the medial pontine reticular formation, which gives rise to reticulospinal fibers) as well as ascending somatosensory pathways (eg, medial lemniscus and anterolateral tracts) passed through or were included in the area of hemorrhage. Therefore, RM's abnormalities in control of posture and whole-body movements most likely reflected combined damage to medial brain-stem and ascending somatosensory systems. Damage to the MLF helps explain the observed horizontal gaze paralysis (vertical eye movements were retained), because the MLF contains fibers important for coordinating horizontal but not vertical eye movements. Some residual control of the right hand and fingers and, to a lesser extent, the left hand and fingers remained, which probably reflected sparing of corticospinal tracts. RM was able to grasp and hold on to objects with the right hand but not with the left hand, provided the trunk and proximal arm were supported. However, both the right and left hands were largely ineffective when reaching to grasp. The asymmetry in residual use of the right hand versus the left hand may be accounted for by more extensive damage to medial lemniscus or rubrospinal fibers on the right than on the left.

Intervention and Outcomes

Intervention during the period from 3 days to approximately 4½ months after the stroke was based on the approach advocated by Shumway-Cook and Woollacott.85 They suggested that 4 key elements contribute to comprehensive clinical practice: (1) The problems and needs of the patient drive the gathering of information and the development of the plan of care; (2) recognition that disablement is affected by disease at several levels, such as impairment, motor strategy, and functional limitation, allows the therapist to develop a list of deficits at each of the levels, toward which intervention can be directed; (3) the nature and cause of deficient motor control are systematically tested (hypothesis-oriented clinical practice); and (4) assumptions related to the nature and causes of deficient motor control have their foundation in a scientifically based theory of motor control.

Because RM's comprehension, memory, judgment, and problem-solving abilities were retained after the stroke, the care plan emphasized that he be allowed, and encouraged, to direct his activities and to make as many decisions for himself as possible. He chose to spend less time and energy on self-care activities, such as dressing and personal hygiene, which were energy consuming and not rewarding to him. Instead, he spent more time and energy on the things he really enjoyed, such as one-on-one social activities with family members. Visiting his family required training family members in how to assist him with car and chair transfers. Sliding board and squat pivot transfers were safer and more efficient than standing pivot transfers because, when assisted to remain in a flexed position, RM did not lose his balance. RM's wife learned how to help him get in and out of his chair and in and out of the car, so they were able to temporarily leave the nursing facility. RM's wife was emotionally and physically supportive of her husband. She visited him daily, attended therapy sessions, and assisted RM with mobility and daily cares. She appeared and reported being physically and mentally healthy, and she appeared to handle this difficult situation well. Use of an electric wheelchair would have provided RM with increased mobility and a sense of independence and control, but was not implemented because of a prognosis for short life expectancy and financial considerations.

Adaptive equipment and specific compensatory movement strategies were implemented in an attempt to assist RM to regain control of some movements, largely because of his undiminished cognitive abilities after the stroke. He and his caregivers were trained to support his trunk and proximal limb joints to allow use of the right hand in playing games, such as dominoes, or to assist in self-care activities that he chose to do. During the day, RM wore a lightweight cock-up hand splint that positioned his wrist in 20 degrees of dorsiflexion. The splint was designed to improve wrist alignment and stability to promote use of the hand and fingers. (Grasping and manipulating objects with the wrist flexed is difficult because of passive insufficiency of finger flexor muscles.) We taught RM simple strategies, such as resting his arm on a support surface and placing his work close to his body, to further improve hand use. Stabilization of joints proximal to the hand is important to manual dexterity in general86; however, our observations suggest that providing stabilization is essential following stroke in the medial pontine tegmentum of the brain stem.

The specific problems RM experienced were not limited to motor deficits but extended to difficulties in interpersonal communication and a sense of diminished control over his life. RM's appearance (resulting from damage to cranial nerves), low level of arousal, difficulty speaking, and extreme dependence in mobility (detailed in the preceding text) led people to believe—erroneously—that he had aphasia and cognitive impairment. It seemed important for his quality of life that those working with him realized that he comprehended what had happened to him and that he was devastated by the inability to control movements of his body. Motor rehabilitation, to some extent, also served to fulfill RM's social and emotional needs and his desire for purposeful activity in his life. For example, when RM was assisted in forward propulsion by a platform walker, he was able to bear weight on his legs and activate large muscle groups. Thus, gross motor physical activity (“walking for exercise”) was incorporated into the care plan both for general exercise and to attempt to promote positive feelings of well being, accomplishment, and self-esteem. Such “quality-of-life experiences” may influence positively the mental and physical health of people who are struggling with loss of functioning and diminished control over their lives.87,88

In summary, during the 4½-month period after the stroke, RM regained his ability to effectively communicate his needs and wishes, thereby restoring control over many aspects of his daily life. Mobility and ability to perform self-care activities improved to the point that he required assistance of 1 person rather than 2 people. With environmental support and the use of a wrist splint, he regained some hand use. Up to the time of his death, he continued to participate in a restorative program of exercise and ambulation designed to promote physical and mental well-being.


RM's deficits in control of posture and whole-body movements are consistent with functional contributions of medial brain-stem descending motor pathways as revealed by studies of postural control in humans and by animal experiments. Studies of automatic postural responses to mechanical perturbations in control subjects and in people with postural deficits have provided insights into the neural control of postural stability,8992 and they have implications for physical therapy practice.93 Current understanding of the neural substrates that underlie control and integration of posture and voluntary movements, however, remains fragmentary and depends largely on results of animal experiments.

RM's severe deficits in control of posture and whole-body movements and less severe deficits in control of hand and finger movements were similar to but more severe than those observed in monkeys with lesions of medial-system pathways.5 Thus, our observations indicate that the contributions of the medial brain-stem and corticospinal systems to motor control are functionally distinct in humans, as they are in monkeys. In monkeys, the corticospinal system does compensate to some extent for loss of function of the medial brain-stem system because motor deficits following lesions of the medial brain-stem system were less severe in animals that had not undergone a bilateral pyramidotomy before the lesion as compared with animals that had fully recovered from prior bilateral pyramidotomy.5 The severity of the observed deficits in RM's posture and whole-body movements and their persistence for the entire 6-month survival period, in combination with the observation that the corticospinal system was largely spared, suggest that the corticospinal system provides little or no compensation for loss of function following damage to the medial brain-stem system in humans.

One caveat is that RM's fluctuating level of alertness may have slowed or impeded developing compensatory movement strategies. Invoking corticospinal pathways may require adequate motivation and attention, which RM may only have had periodically due to damage to the raphé nuclei and the locus coeruleus, which give rise to extremely widespread ascending and descending projections that terminate in structures throughout the central nervous system. These projections mediate diffuse influences on a number of behavioral, physiological, and neuroendocrine functions that are important for regulating arousal, sensory awareness, motor responsiveness, and the level of consciousness (for reviews, see Brodal94 and Saper95). Pharmacological intervention, using drugs that modulate the level of specific neurotransmitters, such as noradrenaline (neurotransmitter in the locus coeruleus) and serotonin (neurotransmitter in the raphé nuclei), combined with physical therapy has been reported to enhance motor performance following stroke.96102 Use of drugs such as amphetamine assumes that sufficient numbers of noradrenergic projections are retained to facilitate neurotransmitter release. Pharmacological intervention might have improved RM's attention and ability to utilize sensory information but was not implemented.

The asymmetry in RM's residual control of the right hand versus the left hand may reflect different functional contributions of rubrospinal versus corticospinal tracts to the control of hand use similar to those observed in monkeys.5 Alternatively, or in addition, RM's nonfunctional grasp of the left hand may have resulted from more extensive damage to the right than left medial lemniscus at the level of the pons, as use of hand and fingers is severely compromised by loss of proprioception.103 RM's preserved ability to produce a functional grasp with the right hand in isolation may have reflected sparing of corticospinal tracts. RM's deficit in extending the right wrist and fingers to preshape the hand during reaching to grasp may have resulted largely from damage to the right rubrospinal tract. The view that rubrospinal neurons are important for controlling hand preshaping during reaching to grasp is consistent with recent results of single-unit recording studies in nonhuman primates.25

Although the rubrospinal system is important for hand use in nonhuman primates,25,26 little is known about the contribution of rubrospinal fibers to control of movements in humans. Rubrospinal contributions are commonly considered less important in humans than in monkeys because the number of large red nucleus cells is much smaller in humans104 and, correspondingly, fewer large-diameter rubrospinal fibers have been observed in humans than in monkeys.71 In addition, human rubrospinal fibers could not always be traced into the spinal cord, and in cases in which a spinal projection was recognized, few rubrospinal fibers were observed below C3.71 If one assumes that rubrospinal fibers originate primarily from large red nucleus neurons, the above histological studies support the view that the rubrospinal tract may be rudimentary in humans.71 However, the size of the rubrospinal tract in humans may be underestimated if a substantial number of rubrospinal neurons are small.105 There is some support for this view as there is evidence that rubrospinal fibers originate from small cells in humans,70,106 monkeys,107110 and cats.111 Furthermore, observations of few rubrospinal fibers below C371 are consistent with the view that rubrospinal neurons in humans may exert their influence on limb movements indirectly, via projections to propriospinal neurons in upper cervical segments rostral to C3. This view derives support from histological observations in humans that propriospinal neurons in upper cervical segments project extensively into the cervical enlargement.112 Thus, the conclusion that the human rubrospinal tract is rudimentary may be premature—damage to rubrospinal pathways may have contributed to the observed deficits in RM's hand use.


Therapeutic intervention aimed at optimizing residual control of the distal extremities appears beneficial for rehabilitation of people with either cortical or brain-stem stroke. This case suggests, however, that people with a stroke affecting the medial brain-stem system need to be provided with support of the trunk and proximal limb during activities. Positive social interactions and a sense of well-being, accomplishment, and self-esteem appear to be equally important. This case suggests that the latter are especially important following brain-stem stroke because the patients' appearance and fluctuating level of arousal may obscure their undiminished cognitive and emotional functions. We believe that physical therapist examination and recognizing intact motor and cognitive functions following brain-stem stroke can aid in developing intervention strategies that maximize rehabilitation and quality of life.


  • Ms Ruhland and Dr van Kan provided concept/project design, writing, data analysis, project management, and facilities/equipment. Ms Ruhland provided data collection.

  • * Pfizer Inc, 235 E 42nd St, New York, NY 10017.

  • Received August 15, 2002.
  • Accepted January 11, 2003.


View Abstract