I The earliest reference to spinal cord injury (SCI) is found in
the Edwin Smith Surgical Papyrus, written between 2500 and 3000
S.c., where it is described as "an ailment not to be treated" (1).
Much has changed in spinal cord care over the centuries,
particularly in the last 50 years as it relates to increasing
survival, life expectancy, community reintegration, and quality of
life. Major advances include the specialized spinal cord centers of
care, model SCI centers funded by the National Institute on
Disability and Rehabilitation Research (NIDRR) in the United States
Department of Education, establishment and growth of organizations
and journals dedicated to SCI care, and the development of the
subspecialty of SCI medicine in 1998. This subspedalty addresses
the prevention, diagnosis, treatment, and management of traumatic
and nontraumatic etiologies of spinal cord dysfunction (2). The
advances of the last 20 years alone have been dramatic in terms of
the understanding of the pathology of the initial and secondary
aspects of the injury, and the barriers that must be overcome to
enhance recovery. Newer techniques to improve function and
intervene at the cellular level for possible cure are being
developed. These will further allow individuals I who sustain an
SCI to be more independent in the future. I t EPIDEMIOLOGY OF
TRAUMATIC SPINAL CORD INJURY I Incidence and Prevalence The
National Spinal Cord Injury Statistical Center (NSCISC) 1. database
consists of data contributed by model.SCI systems (3). This
database captures approximately 15% of all new traumatic 11 SOs
that occur in the United States each year, and has been i used to
develop an epidemiological profile (4-6). When comI pr1red to
population-based studies, persons in this database are I
representative of all SCIs except that more severe injuries, pert
Sons of color, and injuries due to acts of violence are slightly !
overrepresented (4). ; The overall incidence of traumatic SCI in
the U.5. has re-I mained relatively constant, at approximately 40
new cases per r million population, or just over 11,000 cases per
year. The incij (l ..nce of SCI in the rest of the world is
consistently lower than I in the U.S. (7). The prevalence of SCI is
estimated to be approx 250,00 persons by 20D.!, with the growth
resulting from .mproved life expL'c.[,mcy rather than any increase
in incidence. CHAPTER 79 Rehabilitation of Spinal Cord Injury
Steven Kirshblum Age, Gender, Race, Marital, and Occupational
Status The mean age at injury is 32.1 years, with the most common
age at injury being 19 years (Table 79-1). Approximately 60% of all
persons enrolled in the NSCISC database are 30 y.ears of age or
younger at the time of injury. The percentage of new persons
injured who are older than 60 has increased tq over 10% in the last
decade. Men suffer traumatic SCI much more commonly than women, at
a 4:1 ratio. State registries and NSCISC data reveal higher
incidence rates of SCI for African Americans than whites, which is
predominately due to injuries that result from acts of violence.
Approximately 50% of persons enrolled in the NSCISC database have
never been married at the time of their injury. ApprOXimately 60%
of persons enrolled between the ages of 16 and 59 are employed at
the time of injury. Etiology and Time of Injury Motor vehicle
crashes (MVCs) rank first (accounting for 38.5% of cases since
1990), followed by acts of violence (primarily gunshot wounds, or
GSWs), falls, and recreational sporting activities. MVCs cause a
lower percentage of cases among men than women while men have a
higher percentage ofSCI due to GSW, diving mishaps, and motorcycle
crashes. MVCs have decreased, while SCI caused by GSWs almost
doubled (from 12.7% to 24.1%) among men from the mid 1970s to the
early 1990s, before declining slightly in subsequent years. Diving
mishaps account for the majority of SCls due to recreational
sports, followed by snow skiing, surfing, wrestling, and football.
MVC is the leading cause of SCI until age 45; however, beginning
with the 46- to 60-year-old age group, falls represent the leading
cause of SCI. Recreational sports and acts of violence decrease
with advancing age as a cause of injury. Traumatic SCI occurs with
greater frequency on weekends, with the greatest incidence on
Saturday. Seasonal variation exists, with peak incidence occurring
in July followed closely by August and June. The seasonal pattern
in incidence is more pronounced in the northern part of the U.5.
where seasonal variation in climate is greatest. 1716 IV: SPECIFIC
CONDITIONS TABLE 791. Epidemiology ofTraumatic Spinal Cord Injury
(Since 1990) Incidence: Approximately 11 ,OOO/y Prevalence:
200,000-250,000 Average age: 32.1 Gender: 80.5% male .....
Etiology: Motor vehicle accidents (38.5%); followed by violence
(primarily GSW)i falls; and sports Current Percentage
Classification, 1. Incomplete tetraplegia (29.6%) 2. Complete
paraplegia (27.3%) 3. Incomplete paraplegia (20.6%) 4. Complete
tetraplegia (18.6%) ASIA, American Spinal Injury Association; GsW,
gun shot wounds. . From Facts and figures: National Spinal Cord
Injury Statistical Center. J Spinal Cord Med 2002;25:139-140.
Associated Injuries SCls are often accompanied by other significant
injuries. The most common include broken bones (29.3%), loss of
consciousness (28.2%), traumatic pneumothorax (17.S%), and head
injury sufficient to. affect cognitive or emotional functioning
(11.5%) (4). The nature and frequency of these injuries is
significantly associated with the etiology of the SCL For example,
pneumothorax occurs more frequently with GSWs as compared with
other causes of SCl. Neurological Level and Extent of Lesion
Traumatic SCI most commonly causes cervical lesions (approximately
50%) followed by thoracic and then lumbosacral lesions. The C5
segment is the most common lesion level, followed by C4, C6, n2,
C7, and L1 at the time of discharge from inpatient rehabilitation
programs (7). At time of discharge from inpatient rehabilitation,
4S.6% of persons have neurologically complete injuries [American
Spinal Injury Association (ASIA) A classification], followed by
incomplete injuries of ASIA D, C, B, and E classifications (7). The
etiology of injury is strongly associated with level and severity
of the injury. Most recreational sports-related injuries, falls,
and approximately 50% of MVCs result in tetraplegia, whereas acts
of violence usually result in paraplegia. Neurologically complete
injuries are more likely to occur as a result of acts of violence
and among younger age grouPs. Thoracic injuries are the most likely
to be neurologically complete while most lower level lesions are
incomplete injuries. Cervical injuries are most commonly classified
as either ASIA A or ASIA D. There has been a trend toward increased
likelihood of cervical injury since 1994 (5). Marital and
Occupational Status after Spinal Cord Injury The annual marriage
rate after SCI is below average for the general population and the
annual divorce rate is higher (S,9). ApprOXimately 25/., of persons
with SCI are employed, but the percentage varies substantially by
neurological level and extent of injury (10). The higher the level
of and the more severe the injury, the less chance one has of
returning to gainful ployment. Most employed individuals with SCI
have ..un . .....rather part-time jobs. gredictors .of to work'
elude bemg of younger age, male, white, Wlth a greater education
and being employed at the time of injury- havI...l greater
motivation to return to work; having a -:.. jury etiology; being
able to drive; and greater time postin (11-13). Persons who return
to work within the first year Ju;1 jury usually return to the same
job with the same employ while those who return to work after more
than 1 year .: elapsed usually acquire a different job with a
different ell\ployer, often after retraining.
Professional/technical and clerical/sales jobs are the most common.
! Discharge Placement !Model SCI systems have a higher percentage
of persons discharged to the community relative to nonmodel system
centers (6-14). Approximately 90% of -pe.rsons discharged from "
model system are discharged to a private residence within the
Icommunity; 5% being discharged to a nursing home, and 3% tdying
during hospitalization. Predictors of nursing home placement
include ventilator dependence, older age, cervical level of injury
with non-useful motor recovery, being unmarried and unemployed, and
having either Medicaid or health maintenance organization (HMO)
insurance (15). There has been a significant trend toward an
increasing percentage of persons being discharged to nursing homes
since 1995 with the advent of shorter inpatient rehabilitation
lengths of stay (15,16). Life Expectancy Life expectancy of persons
with SCI has improved significantly over the past few decades but
remains below normal. Predictors of mortality after injury include
male gender, advanced age, ventilator dependence, injury sustained
by an act of violence, high injury level (particularly C4 or
above), neurologi .. cally complete injury, poor self-rated
adjustment to disability, and having either Medicare or Medicaid
third-party sponsorship of care (17,lS). Life expectancy estimates
are typically based on neurological level of injury, degree of
injury completeness, age at injury, and ventilator dependence.
Estimates from the NSCISC database appear in Table 79-2. Causes of
Death Diseases of the respiratory system are the leading cause of
death following SCI, with pneumonia being the most common. Heart
disease ranks second, followed by septicemia associated with
pressure ulcers, urinary tract, or respiratory Illfections) and
cancer. The most common location of cancer is the lung, followed by
bladder, prostate, and colon/rectum. Pneumonia is by far the
leading cause of death for . with tetraplegia while heart disease,
septicemia, and sUlClde are more common among persons with
paraplegia (19). Among persons with incomplete motor-function (ASIA
D) at any neurological level, heart disease again ranks as the
cause of death (24.1%), followed by pneumonia (11.0%). Whi e '
genitourinary disease (Le., renal failure) was the leading of death
30 years ago, this has declined dramatically, m05 likely due to
improved care. 79: REHABILITATION OF SPINAL CORD INJURY 1717 TABLE
79-2. Life Expectancy (years) for Postinjury by Severity of Injury
and Age at Injury (for Persons Surviving at Least 1-Year
Postinjury) Motor Ventilator-Age at Injury NoSCI Functional at Any
Level Para Low Tetra (CSCS) High Tetra (C1C4) Dependent at Any
level 20 57.2 52.5 46.2 41.2 37.1 26.8 40 38.4 34.3 28.7 24.5 21.2
13.7 60 21.2 18.1 13.7 10.6 8.4 4.0 SCI, spinal cord injury. From
Facts and figures: National Spinal Cord Inju ry Statistical Center.
JSpinal Cord Med 2002;25:1 39-140. Lifetime Costs Data are
available from the model system that include only the direct costs
of the SCI, with the indirect costs (Le., lost wages, fringe
benefits) not included in these estimates (6). Costs vary by year
postinjury (first year versus subsequent years), by the level of
injury, and severity of injury. Estimates of the total annual costs
of SCI are $9.73 billion (12). This includes first-year direct
costs estimated at $2.58 billion (based on 10,600 new cases),
recurring direct costs of $4.55 billion, and $2.59 billion in lost
productivity. ACUTE MEDICAL AND SURGICAL MANAGEMENT OF SPINAL CORD
INJURY Aft.:r injury prompt resuscitation, stabilization of the
injury, and avoidance of additional neurological injury and medical
complications are of greatest importance. During the first seconds
after SCI there is release of catecholamines with an initial
hypertensive phase. This is rapidly followed by a state of spinal
shock, defined as flaccid paralysis and extinction of muscle
stretch reflexes below the injury level (20), although this may not
occur in all patients. The cardiovascular manifestation of spinal
shock is neurogenic shock, consisting of hypotension, bradycardia,
and hypothermia. Hypotension is especially common in persons with a
neurological level of injury (NU) at or above T6. The para
sympathetic influences on the heart rate are unopposed in persons
with a level of injury above Tl; thus heart rates are typically
less than 60 per minute. Treatment of hypotension involves fluid
resuscitation to produce adequate urine output of more than 30
cc/hour, however, in neurogenic shock further fluid administration
must proceed cautiously, as the patient is at risk for neurogenic
pulmonary eckma. The use of vasopressors is preferred in order to
maintain a mean arterial blood pressure above 85 mm Hg, as this has
been associated with enhanced neurological outcome (21).
Bradycardia is common in the acute period in cervical spinal injury
and may be treated, if symptomatic, with intravenous atropine, or
prevented with atropine given prior to any maneuver that may cause
further vagal stimulation (such as nasotracheal suctioning). In
cases of severe bradycardia that may Jead to cardiac arrest;
temporary cardiac pacing, as well as permanent pacemakers, may be
required (22,23). While most ~ pronounced in the first few weeks
after injury, by 6 weeks the . : rhythm usually returns to normal.
Respiratory assessment is critical for acute SCI patients and
should include arterial blood gases and measurement of forced vital
capacity (VC) as an assessment of respiratory muscle strength. A VC
of less than 1 L indicates ventilatory compromise, and the patient
usually: ...requires assisted ventilation. Close serial assessments
should be obtained for those with borderline values. A nasogastric
tube should be inserted during the initial assessment period to
prevent emesis and potential aspiration. Aspiration occurs in
apprOXimately 5% of all SCI individuals, with a higher risk in
patients over the age of 60, those who have undergone anterior
approach spine surgery, or in the presence of a tracheostomy
(24,25). A Foley catheter should be inserted for urinary drainage
and will allow for an accurate assessment of output. . . Once the
initial resuscitative measures have been taken, attention is turned
to spinal realignment. A full initial neurological assessment
should be documented, and provides a baseline on which changes in
neurological status can be gauged. A standard trauma series
includes cross-table laterals and anteroposterior views of the
cervical and thoracolumbar spine. There is a 12% incidence of
noncontiguous fractures, therefore once one fracture is identified
careful inspection of the rest of the spine is imperative (26).
Forty-seven percent of patients with spine trauma and 64% of
patients with SCI have concomitant injuries, including head, chest,
and long bone fractures (27). Computerized tomography (CT scanning)
can help evaluate for the presence of cervical fractures most often
at the Cl or C7 levels, provide information for surgical
stabilization or decompression, and facilitate selection of
appropriately sized hardware for operative stabilization. Magnetic
resonance imaging (MRI) is helpful in cases in which a ruptured
disc or epidural hematoma is suspected. In addition, an MRI is
usually recommended prior to attempting closed reduction after
cervical spine injury to identify disc herniation and its potential
for neurological worsening with manipulation. Closed cervical
reduction is accomplished with traction. The patient is placed in
either a Stryker frame (Stryker Corp., Kalamazoo, MI) or a RotoRest
bed (Kinetic Concepts Inc., San Antonio, TX). Gardner-Wells tongs
or a Halo ring are applied. Once reduction is confirmed, 10 to 15
pounds of weight is used to maintain reduction. The patient can
later be brought to the operating room for open stabilization. The
indications for and types of spinal orthotic devices are discussed
in Chapter 62. Stab wounds and GSWs generally do not produce spinal
instability and therefore may not require surgical stabilization or
orthotic immobilization. Plain films and CT are used not only to
assess the extent of bony injury, but to provide information
regarding the location and path of the bullet and bone fragments,
and the extent of soft-tissue/vascular damage, in which case
angiography would also be required. Objects that are em1 1 1718 IV:
SPECIFIC CONDIT10NS bedded around the spinal canal (i.e., knife)
should be left in place with removal performed in the operating
room under direct visualization of the spinal canal. Bullets that
pass through the abdominal viscera are treated with broad-spectrum
antibiotics and tetanus prophylaxis (28,29). Bullets do not have to
be removed, however, they can be removed if accessible while
performing another surgical procedure. In most trauma centers IF
III 3'r ' ba (lid; is given after an acute SCI. Mechanisms of
action for MP include improving blood flow to the spinal "Cord,
preventing lipid peroxidation, being a free radical scavenger, and
having antiinflamInatory function. The National Acute Spinal Cord
Injury Study (NASCIS) 2 demonstrated that MP given within 8 hours
of injury (30 mg/kg bolus and 5.4 mg/kg/hour for 23 hours)
marginally improves neurological recovery at 6 weeks, 6 months, and
1 year, although functional recovery was not clearly studied (30).
The NASCIS 3 reported that if MP is initiated within 3 hours of
SCI, it should be continued for 24 hours, whereas if MP is
initiated 3 to 8 hours after SCI it should be continued for 48
hours (31). The administration of MP is not given beyond 8 hours
from sel or to those with sel due to penetrating injuries as they
have shown no benefit and their use is associated with a higher
incidence of spinal and extraspinal infections (32,33). Because. of
some limitations of the NASCIS studies and that these findings have
not been consistently replicated, these specific recommendations
have not been universally adopted (34-36). Sygen (GM-I) is a
ganglioside that is present in high concentrations in the central
nervous system (CNS) and forms the major component of cell
membranes. An initial small study treating patients within 48 hours
of injury for an average of 26 days found greater mean recovery at
1 year including some .. , ' , . \' proved recovery in muscles with
no strength at entry of study (37). A subsequent large muIticenter
study reported. trend toward improvement in neurological recovery
in ASlA: ., : ' individuals at 26 weeks after being treated for 8
weeks. No nificant effect however was noted at the principal 26
weeks-in the total group of patients studied (38). Role of Surgery
Animal model evidence suggests that early decompressive SUrgery
leads to improved neurological recovery after SCI (39). However,
the role and timing of cervical decompression has not been
substantiated in patients with sel since most studies . have failed
to demonstrate any significant neurological recovery among patients
with complete or 'incomplete deficits (40). While there is no
conclusive evidence in humans with sel supporting the benefit of
"early" versus "late" surgery (40-42), select patients with
incomplete Sel, such as those with cervical facet fracture
dislocations, may experience improved neurological recovery if
early decompreSSion is' performed within 8 hours (43). In the
thoracic and thoracolumbar spine, surgery for patients with
complete injuries has not been shown to have significantly improved
neurological function after decompres.sion, while those with
incomplete injuries do benefit from surgical intervention (44). The
indication for emergent surgical treatment is progressive
neurological deterioration, due to compression or an expanding
epidural hematoma. Despite uncertainty regarding the appropriate
timing of surgery, the safety of early surgery after sel has been
validated, and is not TABLE 79-3. Glossary of Key Terms e i' t Key
muscle groups: Ten muscle groups that are tested as part of the
standardized spinal cord examination. Root Level MusdeGroup Root
Level Muscle Group CS Elbow flexors L2 Hip flexors C6 Wrist
extensors L3 Knee extensors C7 Elbow extensors L4 Ankle dorsi
flexors CB Long finger flexors L5 Long toe extensor T1 Small finger
abductors Sl Ankle plantarflexors Motor level: The most caudal key
muscle group that is graded 3/5 or greater with the segments
cephalad graded normal (5/5) strength. Motor index score:
Calculated by adding the muscle scores of each key muscle group; a
total score of 100 is possible. Sensory level: The most caudal
dermatome to have normal sensation for both pinprick/dull and light
touch on both sides. Sensory index score: Calculated by adding the
scores for each dermatome; a total score of 112 is possible for
each pinprick and light touch. Neurological level of injury (NU):
The most caudal level at which both motor and sensory modalities
are intact. Complete injury: The absence of sensory and motor
function in the lowest sacral segments. Incomplete injury:
Preservation of motor or sensory function below the neurological
level that includes the lowest sacral segments. Skeletal level: The
level at which, by radiological examination, the greatest vertebral
damage is found. Zone of partial preservation (ZPP): Used only with
complete injuries, refers to the dermatomes and myotomes caudal to
the neurological level that remain partially innervated. The most
caudal segment with some sensory or motor function defines the
extent of the ZPP. From Kirshblum se, O'Connor K. Levels of injury
and outcome in traumatic spinal cord injury, Phys Med Rehabil Clin
North Am 2000;11 :1-27, '. associated with greater than expected
complications (40,45). Early surgery may contribute to a shorter
rehabilitation length stay and a similar frequency of medical
complications than ,lOse undergoing surgery beyond 24 hours from
injury (46). NEUROLOGICAL ASSESSMENT The most accurate way to
document impairments in a person with a new SCI is by performing a
standardized neurological examination as endorsed by the
International Standards for Neurological Classification of Spinal
Cord Injury Patients (47). These standards provide basic
definitions of the most common terms used by clinicians in the
assessment of SCI and describe the neurological examination. Key
terms are defined in Table 79-3. The examination has two main
components, sensory and motor. The information from this
neurological examination is recorded on a standardized flow sheet
(Fig. 79-1), and helps determine the sensory and motor index
scores; the sensory, motor, and NU; the completeness of the injury;
and to classify the impairment. For the sensory examination, there
are 28 key dermatomes, each tested for pinprick and light touch on
each side of the 79: REHABILITATION OF SPINAL CORD INJURY 1719
body. A 3-point scale (range 0 to 2) is used, with the face as the
normal control point. Absent pinprick, a score 9f zero, is the
inability to distinguish between the sharp and dull edge of the
pin. Impaired pin sensation, a score of 1, is assignl,'!d when the
patient can distinguish between the sharp and dull edge of the pin,
but the pin is not felt as sharp as on the face. Normal or intact
sensation, a score of 2, is assigned only if the pin is felt as
sharp in the tested dermatome as when tested on the face. For light
touch, a cotton tip applicator is used, and scored as follows:
intact, same sensation as on the face; impaired, less than on the
face; and absent, no sensation. To test for deep anal sensation, a
rectal digital examination is performed. The patient is asked to
report any sensory awareness, touch, or pressure, with firm
pressure of the examiner's digit on the rectal wall. Deep anal
sensation is recorded as either present or absent. The required
elements of the, motor examination consist of strength grading of
ten key muscles bilaterally: five in the upper limb (C5-Tl
myotomes) and five in the lower limb (L2-S1) on each side of the
body (see Table 79-3). Testing is performed with the patient in the
supine position. is graded on a 6-point scale (range 0 to 5).
Voluntary anal contraction is tested by sensing contraction of the
external anal sphincter around the examiner's finger and graded as
either present or absent. STANDARD NEUROLOGICAL CLASSIFICATION OF
SPINAL CORD INJURY LIGHT PINMOTOR TOUCH PRICK R L KEY MUSCLES R l R
L C2 ,....., C3 t.... C4 Elbow flexors CS Wrist extensors C6 Elbow
extensors C7 Finger flexors (distal phalanx 01 middle linger) CB
SENSORY KEY SENSORY POINTS . . 0= absent t impaired 2= normal
/liT", not testable .. ' ..
... .0'" Finger abductors (little finger) !;,.::::1,. 0 = total
paralysis 1 = palpable or viSible contraction r"! 2 = active
movement, gravity eliminated ,.."" 3 = active movement, l,,,j
against gravity " . . 4 = active movement. r:1 against some
resistance 5 = active movement, :"'''1 against full resistance
,t:::J .' NT = not testable Ankle dorsiflexors Long toe extensors
Ankle plantar flexors T1 T5 T6 TT91 0 T11 T12 L4 lS Sl Any anal
sensation (Yes/No) [}[] = c:::J PIN PRICK SCORE (max: 112) TOTALS 0
+ 0 = c:=:J MOTOR SCORE 1---...... = c:::J LIGHT TOUCH SCORE (max:
112) t::J c=::::J Voluntary anal contraction (VeS/No) !!.S
(MAXIMUM) (50) (50) (100) (MAXIMUM) (56) (56) (56) (56)
NEUROLOGICAL R L COMPLETE OR INCOMPLETE? ZONE OF PARTIAL R LEVEL
SENSORY 0 0 JncompJste:::; Any sensory or motor functIOn in S4-S5
PRESERVATION SENSORY Cl 0 The. most caudal segmem MOTOR DD
Partially innerv.'ed segments MOTOR Cl 0 With normallvndion ASIA
IMPAIRMENT SCALE This form may be copied freely but should not be
altered without permiSSion from the American Spinal Injury
Association, Figure 79-1. American Spinal Injury Association (ASIA)
neurological flow sheet. L 1720 IV: SPECIFIC CONDITIONS TABLE 79-4.
Summary of the Steps in Classifying an Individual with a Spinal
Cord Injury 1. Perform sensory exam in 28 dermatomes bilaterally
for pinprick and light touch including the 54/5 dermatome and test
for anal sensation on rectal examination. 2. Determine sensory
level (right and left) and total sensory score. 3. Perform motor
exam in the 10 key muscle groups including voluntary anal
contraction on rectal examination. 4. Determine motor level (right
and left) and motor index score. 5. Determine the neurological
level of injury (NU). 6. Classify injury as complete or incomplete.
7. Categorize American 5pinallnjury Association (ASIA) Impairment
Scale (A through E). 8. Determine zone of partial preservation if
A51A A. From Kirshblum SC, Donovan WHo Neurologic assessment and
classification of traumatic spinal cord injury. In; Kirshblum se,
Campagnolo D, Delisa JE, eds. Spinal cord medicine. Philadelphia:
Lippincott Williams & Wilkins, 2002:82-95. The NU is the most
caudal level at which both motor and sensory modalities are intact
on both sides of the body. The motor and sensory levels are the
same in less than 50% of complete injuries, and the motor level may
be multiple levels below the sensory level at I-year postinjury
(48). In cases where there is no key muscle available (i.e.,
cervical levels at and above C4; T2-L1; and sacral levels S2-5),
the neurological level is that which corresponds to the sensory
level. The motor level and upper extremity motor index score better
reflect the degree of function as well as the severity of
impairment and disability, relative to the NU, after motor complete
tetraplegia (48). Table 79-4 lists a summary of the steps to be
followed in classifying an individual with an SCI (49). In 1982,
the American Spinal Injury Association (ASIA) first published
Standards for Neurological Classification of SCI, adopting the
Frankel Scale (50). These standards were refined with the Frankel
Scale being replaced in 1992 by the ASIA Impairment Scale (51),
revised in 1996 (52), and again in 2000 (53), with reprinting in
2002 (47). The ASIA Impairment Scale describes five categories of
SCI (Table 79-5). TABLE 79-5. American Spinal Injury Association
(ASIA) Impairment Scale A= Complete; No motor or sensory function
is preserved in the sacral segments 54-SS. B = Incomplete: Sensory
but not motor function preserved below the neurological level and
includes the sacral segments 54-SS. C = Motor function is preserved
below the neurological level, and more than half of the key muscles
below the neurological level have a muscle grade less than 3. D=
Incomplete: Motor function is preserved below the neurological
level, and at least half of key muscles below the neurological
level have a muscle grade of 3 or more. E= Normal; Motor and
sensory function are normal. Note: For an individual to receive a
grade of Cor D, he/she must be incomplete, that is, have sensory or
motor function in the sacral segments S455. In addition, the
individual must have either (a) voluntary anal sphincter
contraction or (b) sparing of motor function more than 3 levels
below the motor level. From American Spinal Injury Association.
Intemationa! standards for neuro!ogiC,1! ci.mification of spina!
cord injury. Chicago: A5IA, 2002. A complete injury is defined as
the absence of sensory motor function in the lowest sacral
segments, and injury defined as preservation of motor function or
SetlSaliciq. below the NU that includes the lowest sacral segmenli
(termed sacral sparing). Sacral sparing is tested by light lQUch.
and pin sensation at the anal mucocutaneous junction (54/5
dermatome), on both sides, as well as testing VOluntarY allal
contraction and deep anal sensation as part of the rectal examj..
nation. If any of these are present, whether intact or impaired .
individual has sacral sparing and therefore an inJury. There were
three main revisions to the standards in 2000 (53). First there was
clarification that for an individual to receive an ASIA
classification of "motor incomplete" (ASIA Cor D), the patient must
have either (a) voluntary anal sphincter contraction or (b) have
sensory sacral sparing with sparing of motor function more than
three levels below the motor level. Previously, an individual only
needed to have sparing more than two levels below the motor level.
Secondly, the zone ofpartial preservation (ZPP) should as the most
caudal segment with some sensory or motor function bilaterally,
rather than all areas spared. Lastly, in the 2000 revision, the
Functional Independence Measure (FJM) was eliminated from the
standards. Incomplete Spinal Cord Injury Syndromes Incomplete SCI
syndromes include central cord, BrownSequard, anterior cord, conus
medullaris, and cauda equina syndromes. The most common of the
incomplete syndromes is central cord syndrome (CCS), which applies
almost exclusively to cervical injuries and is characterized by
motor weakness in the upper extremities greater than the lower
extremities, in association with sacral sparing (54). Bladder
dysfunction and varying sensory loss below the level of the lesion
may also be present. CCS most frequently occurs in older persons
with cervical spondylosis who suffer a hyperextension injury, but
may occur in persons of any age and is associated with other
etiologies, predisposing factors, and injury mechanisms. The
postulated mechanism of injury involves compression of the cord
both anteriorly and posteriorly by degenerative changes of the bony
structures, with inward bulging of the flavum during hyperextension
in an already narrowed spmal canal (55). CCS usually has a
favorable prognosis. Recovery occurs earliest and to the greatest
extent in the lower extremities, fol lowed by bowel and bladder
function, proximal upper extremity, and then distal hand function.
Prognosis for functional recovery of ambulation, activities of
daily living bowel and bladder function is dependent upon the
patIent.s age (less than or greater than 50 years of age), with a
less ?ptlmistic prognosis in older patients relative to younger
patients (56-58). Older, newly injured individuals however, with a
classification of ASIA D tetraplegia, have a good prognosis for
recovery of independent ambulation (58). Bro'Wn-Seqllard syndrome
(BSS) involves a hemisection of spinal cord, consisting of
asymmetric paresis with hypalgeSia more marked on the less paretic
side and accounts for 2% to 4% of all traumatic seIs (59-61). In
the classic presentation of aSS, there is (a) ipsilateral loss of
all sensory modalities at the level of the lesion; (b) ipsilateral
flaccid paralysis at the level of the ( lesion; (c) ipsilateral
loss of position sense and vibration below the lesion; (d)
contralateral loss of pain and temperature beloW the lesion; and
(e) ipsilateral motor loss below the level of lesion. This is due
to the crossing of the spinothalamic tracts In the spinal cord, as
opposed to the corticospinal and dorsal columns that cross in the
brainstem. The pure form of B5S is rare; Brown-Sequard plus
syndrome (BSPS) is much more com(62) and refers to a relative
ipsilateral hemiplegia with a relative contralateral hemianalgesia.
Although BSS has traditionally been associated with knife injuries,
a variety of etiologies, including those that result incIosed
spinal injuries with or \\,ithout vertebral fractures may be the
cause (62,63). Recovery usually takes place in the ipsilateral
proximal extensors and then the distal flexors (64,65). Motor
recovery of any extremity having a pain/temperature sensory deficit
occurs before the opposite extremity and these patients may expect
functional gait recovery by 6 months. Overall, patients Irith BSS
have the greatest prognosis for functional outcome and potential
for ambulation of the incomplete syndromes, as 7 5 ~ ; ' to 90% of
patients ambulate independently at discharge from rehabilitation
and nearly 70% perform functional skills and ADLs independently
(60,62). Recovery of bowel and bladder function is also favorable.
The anterior cord syndrome involves a lesion affecting the anterior
two-thirds of the spinal cord while preserving the posterior
columns. It may occur with retropulsed disc or bone fragments (66),
direct injury to the anterior spinal cord, or with lesions of the
anterior spinal artery which provides the blood supply to the
anterior spinal cord (67). There is a variable loss of motor as
well as pinprick sensation with a relative preservation of light
touch, proprioception, and deep-pressure sensation. Patients
usually have a 10% to 20% chance of muscle recovery (68). CONUS
MEDULLARIS AND CAUDA EQUINA INJURIES The conus medullaris, which is
the terminal segment of the dult spinal cord, lies at the inferior
aspect of the L1 vertebrae. 'he segment above the conus medullaris
is termed the epiCOIIIIS, consisting of spinal cord segments L4-S1.
Lesions of the epiconus will affect the lower lumbar roots
supplying muscles of the lower part of the leg and foot, with
sparing of reflex function of sacral segments. The bulbocavernosus
reflex and micturition reflexes are preserved, representing an
upper motor neuron (UMN) or suprasacral lesion. Spasticity will
most likely develop in sacral innervated segments (toe flexors,
ankle plantarflexors, and hamstring muscles). Recovery is similar
to other UMN SCIs. Conus medullaris lesions affecting neural
segments 52 and below will present with lower motor neuron (LMN)
deficits of the anal sphincter and bladder due to damage of the
anterior horn cells of 52-54. Bladder and rectal reflexes are
diminished Or absent, depending on the exact level and extent of
the lesion. Motor strength in the legs and feet may remain intact
if the nerve roots (13-52) are not affected (Le., "root escape").
Injuries below the L1 vertebral level usually affect the cauda
equina or nerve rootlets supplying the lumbar and sacral segments
producing motor weakness and atrophy of the lower extremities
(L2-S2) with bowel and bladder involvement (52-54), 3.nd areflexia
of the ankle and plantar reflexes. Often the patIent may have
spared sensation in the perineum or lower extremities, but still
have paralysis. In cauda equina injuries there IS loss of anal and
bulbovernosus reflexes, as well as impotence. Cauda equina injuries
have a better prognosis for recove : ~ 1110st likely due to the
fact that the nerve roots are more resllIttlt to injury. Cauda
equina injuries may represent a 01fopraxia or axonotmesis and
demonstrate progressive re'ery over. a course of weeks and months.
As the cauda .'lIina rootlets arehistologically peripheral nerves,
regeneratIon can occur. Separation of cauda equina and conus
lesions in clinical practice is difficult, because some of the
clinical features of 79: REHABILITATION OF SPINAL CORD INJURY 1721
these lesions overlap. Pain is uncommon in conus lesions but is
frequently a complaint in cauda equina lesions. Sensory
abnormalities occur in a saddle distribution in conus lesions and,
if there is sparing, there is usually dissociated loss with a
greater loss of pain and temperature while sparing touch sensation.
In cauda equina lesions, sensory loss occurs more in a root
distribution and is not dissociated. THE FUNCTIONAL EVALUATION
Numerous attempts have been made to correlate impairment with
disability in SCI. The most conunonlyused disability measures in
SCI are the Functional Independence Measure (FIM), the Modified
Barthel Index (MBn, and the Quadriplegia Index of Function (QIF)
(69-71). A review of the scales and their reliability and validity
have been covered elsewhere (49, 72). The Craig Handicap and
Reporting Technique (CHART) is an excellent measure of handicap
(limitation of activity) for individuals with SCI (73,74). The FIM
is highly predictive of hours of assistance received by individuals
after discharge from inpatient rehabilitation, but the cognitive
domain may be inappropriate for use in 5CI (75,76). While the FIM
was added to the Standards in 1992, it was removed in the 2000
revisions (53). There is good correlation between a seven-item
short version and the total motor FIM for persons with SCI (77).
The QIF is more sensitive and a better indicator of motor recovery
as compared to the FIM, since it can reflect small gains in
function that parallel small strength gains (78). Another scale
described is the Capabilities of Upper Extremity instrument (CUE)
that measures upper extremity functional limitations in individuals
with tetraplegia (79). To more precisely measure physical
assistance and devices required for walking after 5CI, the Walking
Index for. Spinal Cord Injury (WISCI) has been developed and has
shown good validity and reliability (80,81). This is currently
being used in research studies. PROGNOSTICATING RECOVERY AFTER
TRAUMATIC SPINAL CORD INJURY Fundamental to predicting outcome is
the knowledge and skill of performing an accurate examination based
on the International Standards. The use of radiological and
neurophysiological tests can aid and supplement in the diagnosis
and prognosis for recovery (f$2-84). The keys to prognosticating
recovery from traumatic SCI depend on the initial level of injury,
the initial strength of the muscles, and most importantly, whether
the injury is complete or incomplete as determined by physical
examination. While the initial examination establishes a baseline,
testing over a period of days and at 72 hours is superior to a
single early examination (85). Vertebral displacement of less than
30% and age under 30 years at the time of injury is associated with
improved recovery (86). No correlation has been found between other
variables such as the degree of vertebral wedging or type of
fracture. The etiology of the injury only plays a role in
determining whether the injury is more likely to be neurologically
complete or not (87). The types of injuries that are more likely to
cause a complete injury include bilateral cervical facet
dislocation, thoracolumbar f1exionl rotation injuries, and
transcanal bull et locations. 1722 IV: SPECifiC CONDITIONS TABLE
79-6. Summary of Recovery in CompleteTetraplegia A. Patients
(300/0-80%) regain one mOtor level from 1 wk to 1 y. B. At 72 h to
1 wk, recovery of the next m()tor level to at least 3/5 at 1 y
depends on the motor level and initral . Approximately300/0-400k
offirst 0/5 mllsdes 70%-90% of muscles 1 or 2/5' '....., .....
Presence of sensation at that leVel increases chances ofrecovery c.
At 1 mo ... recovery at 1 y:. , . .' . Greater than 9S% of 1 or 2/5
fullsclesrec0vt:r to 3/5 50%...60% of first O/S muscles retover,to
}IS. '. .'" Approximately 25% of first O/S muscle,s recover to <
less than 10% of second 0/5 muscles recoverto l/5... . '
Approximately 1 % of seCond O/Srrtuscle,sfeeov.ers t031S D. The
initial strength of the muscle, IS a achieving antigravity strength
and its rate of recQvery.' .... . . E. The faster an initial 0/5
muscle starts torecQvksCime strength, the betterthe prognosis for
recovery. ' .....' . , .,' ..... . >" '.F. Mostupper extremity
recovery QCcurs duringthe first 6 /'nos; with the greatest rate of
change during the 0105. G. Mostpatients with some initial PClwer
plateau at .an earlier time .. and at a higher level than
patients,withn!:,m()tor power. Motc:>r . recoVery can continue,
with lesser seen In the secorid year, . espeCially for patients
with initially 0/5 strength. From DitunnoJ. Flanders A, Kirshblum
SC, et al. Predicting outcome in traumatic spinal cord injury. In:
Kirshblum SC, Campagnolo 0, Delisa jE, eds. , Spinal cord medicine.
Philadelphia: llppincott WUllams & Wilkins, 2002:108-122.
Complete Tetraplegia It has frequently been stated that patients
with complete cervical lesions recover one root level of function
(SS). Table 79-6 lists certain generalizations regarding recovery
patterns in patients with complete tetraplegia. The initial
strength of a muscle is a significant predictor of achieving
antigravity strength at the level caudal to the NU, as well as the
rate of recovering antigravity strength (S9,9O). The faster
recovery begins the greater the chance of recovering anti gravity
strength (91). Recovery at the C4 level to the CS level, both motor
and sensory, may be less and slower than at the lower cervical
segments, especially if there is initially no strength at the CS
myotome (76,92). Most upper extremity (UE) motor recovery occurs
during the first 6 months after injury; with the greatest rate of
change during the initial 3 months. Overall, the mean total motor
score improvement for persons with complete tetraplegia between 1
and 6 months postinjury is 6.6 4.7; between 1 month and 1 year, S.6
4.7; and during the. second year, 1.7 1.9 (90). In patients with no
motor strength at the first caudal level, recovery may continue for
up to 2 years after injury (92). Only a small percentage of
subjects have motor recovery below the first caudallevel from the
NU (90). . Incomplete Tetraplegia Persons with incomplete
tetraplegia have a better prognosis for recovery. UE motor recovery
is approximately twice as great as in complete tetraplegia, with
the potential for varying degrees of lower extremity (LE) motor
recovery and functional ambulation. For patients who are sensory
incomplete initially, the prognosis for motor recovery is more
favorable in those with sparing of pin sensation rather than light
touch sensation alone. Studies using the Frankel Scale found that
for Frankel B patients with pin sensation initially intact, the
prognosis for ombul'tion is 66% to89%, while fo, th"", with only
I:' intact it was 11% to 14% (93-95). The basis of a more favo'
outcome for pinprick sparing in the initially sensory ir. plete
patient may be explained by the close anatomical t,,: . tionship of
the motor tracts (mediated through the / .....cospinal tract),
which are just dorsal to the sensory (lateral spinothalamic tract)
carrying pain and tern' " fibers. A sensory incomplete motor
complete patient .' at 1 month may still regain LE muscle recovery
if bila ' sacral pin sensation is present, although the prognosis
for coI:' munity ambulation is poor (96). Functional and
neurolosiCil recovery is more favorable for patients with an
initial motoriit!: complete injury (93-98).,i For UE motor
recovery, the total motor score improvemerit from 1 month to 1 year
is 10.6 6.9, with improvement in second year of 1.1 2.3 (96). The
majority of functional ery is within the first 6 months after
injury and the early' appearance of motor function suggests a
better functional outcome (85,94,96,98). A greater percentage of
key muscles recover antigravity strength orbetteri,as well as
earlier, distal to the NU in individuals with motor incomplete
tetraplegia thanJIL those with motor complete injuries (99,100).
The mean LE motor recovery in incomplete tetraplegics is 13.5 7
from 1 month to 1 year and 1.8 3.1 in the second year (96). Motor
recovery in the UE and LE occur concurrently, rather than
sequentially. Complete Paraplegia Recovery from injuries resulting
in paraplegia has not been studied to the same degree of
tetraplegia. For persons complete injury with a NU at TS or above,
Waters and \ leagues found that none regained any LE motor
function. Tht: potential for LE motor recovery improves with lower
initial neurological levels of injury; 15% of patients with an NU
between T9-Tll and 55% of those with an initial NU below T12 gain
some recovery. Most movement gained is in the proximal LE
musculature (101). This improvement may represent recovery of
partially injured lumbar roots or "root escape" (102). Incomplete
Paraplegia Individuals with incomplete paraplegia have the best
prognosis for LE motor recovery and ambulation (103,104). Eighty
percent of individuals with incomplete paraplegia regain
antigravity hip flexors and knee extensors at 1 year. Individuals
with no LE strength at 1 month may still show significant return by
1 year. Conversion from Complete to Incomplete Status While it has
been reported that up to 10% of patients with an initial Frankel A
classification can progress to Frankel D or E, Maynard and
associates prospectively demonstrated that jects with this degree
of change had sustained closed head 111juries with cognitive
impairment and were incorrectly initially diagnosed as Frankel A
(105). Model systems data report that up to 16% of initially
neurological complete (Frankel or A) patients improve at least one
classification grade, from 1Il0 tial early examination to the
I-year follow-up, but only up . 5.8% improve to grade C and 3%
improve to grade D .. , Marino and colleagues reported a small
difference when using the Frankel and the ASIA Impairment scales
(106). . Between 4% to 10% of patients may undergo late
converS1()ll (after 30 days) from complete to incomplete status
which haS bet' n reported to occur years after injury (90,107,108).
Motor re-seems to be slightly improved and some, usually non-LE
recovery, may take place. the Effect of Reflexes fhe presence of
spinal shock may play a role in prognosis; fOr the same degree of
SCI the presence of spinal shock implies a lJ10re rapid evolution
of injury and a worse prognosis (109). In individual lesions,
especially in high level cervical SCI, the lJ10st distal sacral
reflexes, including the bulbocavernosus (BC) JIld anal wink, may
remain intact. The order that reflexes return in the postinjury
period may help prognosticate outcome (110). The lack of the BC
reflex ISJ-l roots) or the anal reflex (S2-4 roots) after the acute
period 124 to 72 hours) suggests injury to the conus medullaris or
cauda equina (Le., LMN injury). As such, prognosis regarding
recovery and also the potential use of rehabilitation intervention
(e.g., electrical stimulation) can be determined. The delaved
plantar response, which may be the first of all reflexes to return,
occurs within hours or days following SCI, and shows a high
correlation with complete injuries and a poor prognosis for LE
motor recovery and function (ambulation) {no,111). Magnetic
Resonance Imaging in Predicting Outcome Numerous studies have shown
a direct correlation between the appearance of the spinal cord on
MRI and the degree of functional deficit at the time of injury and
the capacity for neuro1 recovery. Overall, the findings on MRI that
correspond to a more severe initial injury and have a poor
prognosis for neurolOgical recovery include the presence and length
of hemorrhage, the length of spinal cord edema, and spinal cord
compression. An intramedullary hemorrhage equates with a severe
initial neurological deficit, most commonly ASIA A (complete)
injury on clinical examination, and carries a poor prognosis
(83,112-114). The location of the hemorrhage corresponds
anatomically to the level of the neurological injury. If no
hemorrhage is seen on initial MRI, these individuals usually have
an incomplete lesion by clinical exam and have a better prognosis
for motor and functional recovery. Cord edema alone is associated
with mild to moderate initial deficits. Cord edema that extends for
more than the span of one vertebral segment is associated with a
more severe initial injury than smaller areas of edema (114,115).
In the chronic stage after SCI, persons with persistent signal
Changes in their spinal cord on follow-up MRI exams demonstrrd
injury patients: effects of PES exercise. I ReIUlbiI Res Dev 1994;
31:50-"], BeDell KK Scremin AME, Pere\l KL, et al. Effects of
functional electrical stimulatiCJ)1-induced lower extremity cycling
on bone density of spinal cordinjured patient. Am' Phys Med
Rehabill996;75:29-34. :CS. Klose KJ, Ganz W, et al. Bone mineral
density after bicycle er,o:"metry training. Arch Phys Med RehabiI
1990;71:207-279. .lo:," Rdwbil199'J;80:1166. 391. Olenik LM. Laskin
H, Bumham R. et aL Efficacy of rOWing, backward wheeling and
isolated scapular retr,lctor exercise as remedial strength
activities 1750 IV: SPECIFIC CONDITIONS for wheelchair users:
application of electromyography. Paraplegia 1995;33: 148-152. 392.
Cums KA, Tyner TM, Zachary L, et al. Effect of a standard exercise
protocol on shoulder pain in long-term wheelchair users. Spinal
Cord 1999;37:421429. 393. Robinson MD, Hussey RW, Ha CY. Surgical
decompression of impingement in the weight bearing shoulder. Arch
Phys Med Rellabil 1993;74:324-327. 394. Gellman H, Chandler D,
Petrasek J, et al. Carpal tunnel syndrome in paraplegic patients. I
Bone It Surg (Am) 1988;70:517-519. 395. Davidoff G, Wemer R, Waring
W. Compressive mononeuropathies of the upper extremity in chronic
paraplegia. Paraplegia 1991;29:17-24. 396. Nemchausky BA, Ubilluz
RM. Upper extremity neuropathies in patients with spinal cord
injuries.J Spinal Cord Med 1995;18:95-97. 397. Kirshblum S, Druin
E, Planten K. Musculoskeletal conditions in chronic spinal cord
injury. Tap Spinal Cord Inj Rehabil1997;2:23-35. 398. Yerzierski
RP. Pain following spinal cord injury: the clinical problem and
experimental studies. Pain 1996;68: 185-194. 399. Siddahl pJ,
Taylor DA, McClelland JM. et al. Pain report and the relationship
of pain to physical factors in the first 6 months following spinal
cord injury. Pain 1999;81:187-199. . 400. Demirel G, Ylhnaz H,
Gencosmanoglu B, et al. Pain following spinal cord injury. Spinal
Cord 1998;36:25-28. 401. Davidoff G, Guarradni M, Roth E, et a!.
Trazodone hydrochloride in the treatment of dysesthetic pain in
traumatic myelopathy: a randomized, double-blind placebo-controlled
study. Pain 1987;2:151-161. 402. Putzke ID, Richard JS, DeVivo MJ.
Preceptors of pain 1 year post-spinal cord injury. I Spinal Cord
Med 2001;24(1):47-53. 403. Siddall PI, Loeser JO. Pain following
spinal cord injury. Spinal Cord 2001; 39:63-73. 404. Stormer S,
Gemer HJ, Gruninger W, et al. Chrome pain/dysaethesiae in spinal
cord injury patients: results of a multicentre study. Spinal Cord
1997; 35:446-455. 405. Turner JA, Cardenas DD, Warms CA, et al.
Chronic pain associated with spinal cord injuries: a community
survey. Arclz Phys Med Rehab 200182(4): 501-509. 406. Siddall pJ,
Yezierski RP, Loeser JD. Pain following spinal cord injury:
clinical features, prevalence, and taxonomy. IASP Newslell
200()-,3-7. 407. McAdoo DJ, Xu GY, Roback G, et a!. Changes in
amino add concentrations over time and space around an impact
injury and their diffusion through the rat spinal cord. Exp
Neuro/1999;159:538-544. 408. Eide PK, Stubhaug A, Stenehjem AE.
Central dysesthesia pain after traumatic spinal cord injury is
dependent on N-methyl-D-aspartate receptor activation. Neurosurgery
1995;37(6):1080-1087. 409. Hao JX, et at. Baclofen reverses the
hypersensitivity of dorsal horn wide dynamic ranged neurons to
mechanical stimulation after transient spinal cord ischemia:
implications for a tonic GABAergic inhibitory control of myelinated
fiber input. I NeurophysioI1992;68:392-396. 410. Ciom B, Meglio M,
Pentimalli L, et al. Spinal cord stimulation in the treatment of
paraplegic pain. I Neurosurg 1995;82:35-39. 411. Cole J, lIlis L,
Sedgwick E. Intractable pain in spinal cord injury is not relieved
by spinal cord stimulation. Paraplegia 1991;20:167-172. 412.
Vaarwerk I, Staal M. Spinal cord stimulation in chronic pain
syndromes. Spinal Cord 1998;36:671-682. 413. Oavis R, Lentini R.
Transcutaneous nerve stimulation for treatment of pain in patients
with spinal cord injury. Sllrg NeuroI1975;4:100-101. 414. Friedman
AH, Nashold SS Jr. OREZ lesions for relief of pain related to
spinal cord injury. f Neurosurg 1986;65(4):465-469. 415. To TP, Lim
TC, Hill sr, et at. Gabapentin for neuropathic pain following
spinal cord injury. Spinal Cord 2002;40(6):282-285. 416. Tai Q,
Kirshblum 5, Tai Q, Kirshblum se, Chen B, Millis S, et a!.
Treatment of neuropathic pain with gabapentin in spinal cord
injury. JSpinal Cord Med 2002;25:100-105. 417. Warms CA, Turner JA,
Marshall HM, et al. Treatments for chronic pain associated with
spinal cord injuries: many are tried, few are helpful. Clill I Pain
2OO2;18(3):1S4-163. . 418. Biering-Sorensen F, Bohr H. Bone mineral
content of the lumbar spine and lower extremities years after
spinal cord lesion. Paraplegia 1988;26:293-301. . 419. Kiralti B,
Smith A, Nauenberg T, et al. Bone mineral and geometric changes
through the femur with immobilization due to spinal cord injury. I
Rehabil Res Dev 2000;37:225-233. 420. Chow Y, Imnan C, Pollintine
P, et at. Ultrasound bone densitometry and dual energy x-ray
absorptiometry in patients with spinal cord injury: a cross
sectional study. Spinal Cord 1996;34:736-741. 421.
Needham-Shropshire B, Broton J, Klose K, et al. Evaluation of a
training program for persons with SCI paraplegia using the Parastep
1 ambulation system: .Part 3. Lack of effect on bone mineral
density. Arc" P"ys Med Rd,abil 1997;78:799-803. 422. Belanger M,
Stein R, Wheeler G, et al. Electrical stimulation: can it increase
muscle strength and reverse osteoporosiS in spinal cord injured
individuals. Arcll PI,y. Med Rellabil2000;81:1090-1098. 423. Mohr
T, Podenphant l, Biering-Sorensen F, et al. Increased bone mineral
drnsity after prolonged electrically induced cycle training of
paralyzt>d limbs in spinal cord injured man.
CalcifTisslllt1997;61:22-25. 424. Harktopp A, Murphy R, Mohr T, et
al. Bone fracture during electrical stimulation of the quadriceps
in a spinal cord injured subject. Arcll P1Iys Med
Re/IQ1>iI1998;79:1133-1136. 425. Nance P, Schryvers 0, Leslie W,
et al. Intravenous pamidronate bone density loss after acute spinal
cord injury. Arcll Phys Med 1999;80:243-251. 426. Luethi M, Zehnder
Y, Michel D, et al. Alendronate in the treatment loss after spinal
cord injury: preliminary data of a 2 year rartclorniz...t trolled
trial in 60 paraplegic men. J Bone Mill Res 2oo1;16[Suppl (abst). .
427. Ragnarsson KT, Sell GH. Lower extremity fractures after spinal
cord a retrospective study. Arch Phys Med
Rellilbi/1981;62:416--423. .-,,;:. 428. Frisbie JH. Fractures after
myelopathy. I Spitllll Cord Med 429. Garland DE. Pathologic
fractures and bone mineral density at the __., ... Spinal Cord Med
1999;22:335. 430. Freehafer AA. Limb fractures in spinal cord
injury. Arch Phys Med Reh4i . 1995;76:823-827. . 431. Rossier AB,
Foo 0, Shillito J, et al. Post-traumatic syringomyelia: clinical
electrophysiological studies, syrinx protein and results of
conservative and operative treatment. Brain 1985;108:439-461. 432.
El Masry W, Biyani A. Incidence, management, and outcome of
post.traumatic syringomyelia.J Neurol Nellr05l1rg Psychiatry
1996;60:141-146. 433. Perrouin-Verbe B, Lenne-Aurier K, Robert R,
et a!. Post-traumatic syringo. myelia and post-traumatic spinal
canal stenosis: a direct relationship: review of 75 patients with a
spinal cord injury. Spinal Cord 1998;36:137-141 434. Sett P,
Crockard H. The value of magnetic resonance imaging (MRI) in !he
follow-up management of spinal injury. Paraplegia 1991;29:396-410.
435. Backe HA, Beta RR, Mesgarzacfeh M; et'al;' Post-traumatic
spinal cord cysts evaluated by magnetic resonance imaging.
Paraplegia 1991;29:607-612. 436. Abel R, Gerner HJ, Smit C, et a!.
Residual deformity of the spinal canal in patients with traumatic
paraplegia and secondary changes of the spinal cord. Spinal Cord
1999;37:14-19. 437. Nogues MA, Gene R, Encabo H. Risk of sudden
death during sleep in sy_ ringomyelia and Syringobulbia. JNeurol
Nellros1trg Psycltiatry 1992;55:585589. 438. Wang D, Bodley R, Sett
P, et al. A clinical magnetic resonance imaging study of the
traumatised spinal cord more than 20 years following injury.
Paraple gia 1996;34:65-81. 439. Little lW, Robinson LR, Goldstein
B, et al. Electrophysiologic findings in posttraurnatic
syringomyelia: implications for clinical management. JAm Paro Soc
1992;15:44-52. 440. Nogues MA, Stalberg E. Electrodiagnostic
findings in syringomyelia. Muscle Nerve 1999;22:1653-1659. 441.
Sgouros S, Williams B. Management and outcome of posttraumatic sy
ringomyelia. JNeur05nrg 1996;85:197-205. 442. Dautheribes LW,
POintillart V, Gaujard E, et al. Mean term follow-up of a series of
posttraumatic syringomyelia patients after syringo-peritoneal
shunting. Paraplegia 1995;33:241-245. 443. Levi ADO, Sonntag VKH.
Management of posttraumatic syringomyelia using expansile
duraplasty. Spine 1998;23:128-132. 444. Batzdorf U, KJekamp J,
Johnson JP. A critical appraisal of syrinx cavity shunting
procedures. I NellrOSllrg 1998;89:382-388. 445. Grant R, Handley
OM, Lang 0, et al. MRI measurement of the syrinx size before and
after operation. JNeurol Neurosurg Psychiatry 1987;50:1685-1687.
446. Waters RL, Muccitelli LM. Tendon transfers to impro\'e
function of patienl5 with tetraplegia. In Kirshblum SC, Campagnolo
D, DeLisa JE, t>ds. Spilllll cord medicine. Philadelphia:
Lippincott Williams & Wilkins, 447. Pedretti LW. Occupational
performance: a model for practice in dysfunction. In: Pedretti LW,
ed. Occupational therapy practice skillsJor pltysrcal dysfunction,
4th ed. St. Louis: Mosby, 1996;3-12. 448. Mc Dowell CL, Moberg EA,
House JH. The Second International Conference on Surgical
Rehabilitation of the Upper Limb in Traumatic Quadriplegia. J Hand
Surg 1986;l1A:604--608. . 449. Moberg E, Freehafer AA, Lamb DK, et
al. International federation eties for surgery of the hand. A
report from the committee on spinal injunes 1980. Scalld f Relzabil
M,ti 1982;14:3-5. . 450. Moberg E. Surgical rehabilitation of the
upper limb in tetraplegia. ParaplfglQ 1990;28:330-334. 451.
Freehafer AA. Tendon transfers in patients with cervical spinal
cord injury. 1 Hand Surg 1991;16A:804-809. . 452. Johnson DL,
Gellman H, Waters RL, et al. Brachioradialis transfer for wnst
extension in tetraplegic patients who have fifth-cervical-level
neurological function. I BOlle IOi"t Sllrg 1996;78A:1063-1067. 453.
Raczka R, Braun R, Waters RI.. Posterior deltoid-to-triceps
transfer in quadriplegia. Cllin OrtllOp 1984;187:163-167. . 454.
Revol M, Briand E, Servant JM. Biceps-to-triceps transfer in
tetraplegIa. The medial route. JHallti SurS (Br)
1999;24(2):235-237. . . _ 455. Kuz lE, Van Heest AE, HouseJH.
Biceps-to-triceps transfer in tetrapleglcpa, tients: report of the
medial routing technique and follow-up of three case> Ha/ld
SlIrg (Am) 1999;24(1):161-172. .' 456. Moberg E. TIle 1l1'1','r
lilllb ill t,'tTllI'/C;;:ia. a "1/ClI' al'proacll" /() ';/IIXical
",lmMIIIl tiOIl. Stuttgart: G"orge Thieme, 1978. . . 457. Moberg
EA.TIu., present state ofsurgical rehabilitation for the upper Iim
td:aplegia. Parlll'i.'sia1987;25:351-356. . ce458. Rel"er TV,
Waters RI.. Long term follmv up of the :Yloberg key gnp pro dure./
f/1l1I,1 SlIr)11981>;11 A:724-728. 459. W.1t"r R. K, Grabofi 5,
et al. I3mchiomdi.11i, tn ne",r p,)lIids ionsgUS rgtendon transfer
for active lateral pinch in tht' ,Halla u 1985;lOA(3):385-391. I
f!tO House JH, Shannon M. Restoration of strong grasp and lateral
pinch in . tetraplegia: a comparison of two methods of thumb
control in each patient. I Hand Surg 1985;IOA(I):22-29. Gansel J,
Waters RL, Geilman H. Pronator teres to flexor digitorum profun>
dus transfer in quadriplegia. I BOlle loint Surg
1990;72A(3):427-432. . , la 113 SD, Gellman H, Waters R et at
Single-stage reconstruction of key . pinch and extension of the
elbow in tetraplegic patients. J Bone /t Surg
199-1;76(A):1451-1456. fttl. Gellman H, Kan D, Waters RL, et al.
Rerouting of the biceps brachii for paraIvlic supination
contracture of the forearm in tetraplegia due to trauma. / Borl 11
Surg 1994;76A:398-402. Gellman H. The hand and upper limb in
tetraplegia. Curr Or/hop 1991;5: 233-238. Waters RL, Stark LZ,
Gubemick I, et al. Electromyographic analysiS of brachioradialis to
flexor pollicis longus tendon transfer in quadriplegia. / Hand
Sltrg 1990;15A:335-339. -167. Pe(kham PH, Keith MW, Kilgore Kt, et
al. Implantable Neuroprosthesis Re,earch Group. Efficacy of an
implanted neuroprosthesis for restoring hand grasp in tetraplegia:
a multicenter study. Arch Phys Med Rehabil 2001;82: 1380-1388. Wl.
Mulcahay MJ, Beta RR, Smith BT, et al. Implanted functional
electrical stimulation hand system in adolescents with spinal
injuries: an evaluation. Arch Pity> Med Rehabill997;78:597-607.
WJ. Hobby J, Taylor PN, Esnouf J. Restoration of tetraplegic hand
function by use of the neurocontrol freehand system. JHand Surg
(Br) 2001;26:459-464. 4i\J. l