lower extremity sensory function in children with cerebral palsy
TRANSCRIPT
Lower extremity sensory function in children with cerebral palsy
JOHN F. MCLAUGHLIN1, STEVEN D. FELIX2, SOGOL NOWBAR3, ANNE FERREL1,
KRISTIE BJORNSON1 & ROSS M. HAYS1
1University of Washington, Washington, USA, 2Carillon Neurodevelopmental Clinic, Roanoke, VA, USA,3Idaho Pulmonary Associates, Boise, ID, USA
AbstractObjectives: (1) To determine the feasibility of qualitative sensory testing in the lower extremities (LE) of children withcerebral palsy (CP), especially spastic diplegia. (2) To determine if there is a detectable difference in qualitative LE sensoryfunction in children with CP compared to typical children. (3) To determine if dorsal rhizotomy results in detectable changesin LE sensory function in children with spastic diplegia.Design: Objectives 1 and 2: Prospective observational cohort study. Objective 3: Add-on to prospective interventional studies.Setting: Regional tertiary children’s hospital.Participants: Objectives 1 and 2: 62 children with CP and 65 typical children between 3–18 years of age. Objective 3:34 children with spastic diplegia.Interventions: Objectives 1 and 2: None. Objective 3: Dorsal rhizotomy.Main outcome measures: Pain, light touch, direction of scratch, vibration, toe position and knee position using standardqualitative techniques.Results: Objective 1: 32 (52%) children with CP and 55 (85%) typical children completed all items ( p¼ 0.09). Objective 2:Summary scores for separate LE sensory modalities were lower in children with CP for direction of scratch ( p<0.001), toeposition ( p¼ 0.01) and vibration sense ( p¼ 0.01). Objective 3: No changes of LE sensory function.Conclusions: LE sensory testing in young children with CP is feasible. There is a qualitative sensory deficit in this sample ofchildren with CP and specifically in children with spastic diplegia that is traditionally associated with dorsal column sensorymodalities. A conservative dorsal rhizotomy does not produce a measurable change in LE sensory function in this sample ofchildren with spastic diplegia.
Key words: Cerebral palsy, spastic diplegia, rhizotomy, sensation disorders, children.
Introduction
Cerebral palsy (CP) is a motor disability due to
brain injury or maldevelopment occurring in the
pre-natal, perinatal or post-natal periods [1]. In spite
of advances in perinatal care, the incidence of CP
has not decreased. The prevalence of children with
impairments secondary to CP remains constant. The
traditional emphasis on the motor component may
result in neglect of any coexisting sensory impair-
ment, especially in the lower extremities. Sensory
examinations are often excluded in clinical practice
to save time. It is unlikely that brain and concomitant
spinal cord injuries, such as are found in some
cases of CP [2,3], are solely limited to the motor
cortex and/or the motor portion of the spinal cord.
Testing with somatosensory evoked potential (SEP)
methodology has demonstrated abnormalities in
cortical SEP responses to stimulation of the lower
extremities of children with spastic diplegia [4].
Damage to central sensory pathways detected with
diffusion tensor imaging has recently been reported
in two children with spastic diplegia [5]. Normal
sensory function is a pre-requisite for unimpeded
motor performance. Sensory deficits may contribute
to or be responsible for some aspects of motor
dysfunction.
If sensory function is compromised in children
with CP, treatment programmes designed to address
such deficits need to be incorporated in the man-
agement programme. Researchers in the 1950s
and 1960s first documented that up to 70% of
Correspondence: John F. McLaughlin, MD, Children’s Hospital and Regional Medical Center, 4800 Sandpoint Way NE, M2-8, PO Box 5371,
Seattle, WA 98105-0371, USA. Tel: 206-987-2204. Fax: 206-987-3824. E-mail: [email protected]
Received for publication 12 December 2003. Accepted 1 March 2004.
Pediatric Rehabilitation, January 2005; 8(1): 45–52
ISSN 1363–8491 print/ISSN 1464–5270 online 45–52 � 2005 Taylor & Francis Group Ltd
DOI: 10.1080/13638490400011181
Dev
Neu
rore
habi
l Dow
nloa
ded
from
info
rmah
ealth
care
.com
by
Uni
vers
ity o
f A
lber
ta o
n 10
/25/
14Fo
r pe
rson
al u
se o
nly.
children with spastic hemiplegia have upper extrem-
ity sensory deficits [6–9]. The most commonly
affected sensory modalities are stereognosis, two-
point discrimination and position sense. Recent
studies of upper extremity sensory function in
children with hemiplegic CP have confirmed the
common presence of deficits in stereognosis and
two-point discrimination [10–13].
Researchers recognize that sensory stimuli influ-
ence motor acts [14]. Umansky [15] hypothesized
that, in the presence of decreased/absent afferent
input to the brain, motor learning and functional
body image formation may not develop completely.
As a result, those with sensory deficits will fail to
incorporate the involved body part into a functional
body image, resulting in limited use of that body
part. This hypothesis is supported by studies in
both animals and humans with congenital or acquired
sensory deficits, in which failure to use the deaf-
ferented limb, in time, results in a learned disuse
phenomenon which brings about a greater deficit of
motor capability in the affected limb [16–18].
Active movement is thought to be required for
creating sensory experiences from which adequate
sensory function and mature motor activities evolve
[19]. Umansky [15] further suggests that motor
development should not be regarded as solely
dependent on sensory function, but should be
considered a function of a combined integrative
pathway between sensory input and motor ability.
Unrecognized sensory deficits may explain much
of the variability in performance among children
with apparently similar motor deficits [13,20]. The
success of constraint-induced therapy may be influ-
enced by sensory function, which adds urgency to
accumulating data on sensory function in children
with CP [18].
A review of the literature reveals a large array of
data on quantitative sensory examinations in adults
[21–23]. These techniques do not appear feasible
for use in testing lower extremity function of young
children. Only modest amounts of data are available
on qualitative sensory examination in healthy young
children [24,25].
No data could be found regarding clinical exam-
ination of the lower extremity sensory function of
children with spastic diplegia in the literature. Some
children with spastic diplegia have abnormal cortical
SEPs that do not change after dorsal rhizotomy (DR)
in the presence of demonstrable electrophysiologic
changes induced at the spinal cord level consistent
with reduction of spasticity [4]. It is not clear how
SEPs relate to the individual modalities such as
vibration, pain, light touch, etc. that seem to have
functional correlation. A further understanding of
sensory function may be particularly important for
evaluating children with spastic diplegia undergoing
DR because of the potential for iatrogenic sensory
impairment [26,27]. In other research on DR
conducted by the team, careful sensory examination
was deemed essential [27].
The first aim of this study was to determine
whether it is feasible to perform qualitative LE
sensory examination in young children with and
without CP. The second aim of this study was to
determine whether there was a detectable difference
in qualitative LE sensory function between samples
of children with and without CP. Qualitative sensory
examination using techniques familiar to clinicians
who examine children was chosen in this study
because it is felt that quantitative or more elaborate
qualitative testing would be difficult for younger
children and impractical in clinical settings. In a
second study reported here, the aim was to use
the same qualitative LE sensory testing battery to
examine children with spastic diplegia before and
after DR to identify sensory changes induced by the
surgical procedure.
Methods
Sensory testing methods
The sensory examination consisted of light touch,
pain sensation, position sense, vibration sense
and direction of scratch. These tests were selected
because they had face validity for use with young
children. Reports supporting the validity and reli-
ability of knee position and direction of scratch
were identified in the literature (see below). The
other tests are familiar to clinicians who frequently
examine young children, cover the primary sensory
modalities and are described in most standard
textbooks, but validity and reliability data were not
found. Before each individual test, a description
of the test along with a practice run was performed.
The focus was on the sensory function in the lower
extremities. A forearm site was included for light
touch, pinprick and direction of scratch to facilitate
making a judgement if the child understood these
tasks. All tests were performed bilaterally and
repeated three times at each of the sites tested. The
children were tested in a sitting position and their
vision was obstructed either by having the child
close his/her eyes or by having the parent/guardian
obstruct the child’s view with a blind. All examina-
tions were performed by examiners who were not
blinded to the children’s clinical status.
Light touch was tested by stroking the child with a
cotton ball on the dorsal forearm (C7–8), mid-thigh
(L2–3), mid-shin (L3–4) and mid-dorsum of the foot
(L4–S1). These sites were chosen to represent major
sensory dermatomes with a minimum of difficulty
and time. The child was asked to say ‘yes’ when the
46 J. F. McLaughlin et al.
Dev
Neu
rore
habi
l Dow
nloa
ded
from
info
rmah
ealth
care
.com
by
Uni
vers
ity o
f A
lber
ta o
n 10
/25/
14Fo
r pe
rson
al u
se o
nly.
cotton ball was felt. Pain sensation was tested in
the same sites using the sharp and dull ends of
a safety pin. The child was asked to distinguish
between sharp and dull sensations. For both light
touch and pain sensation, control tests in which the
skin was not touched or pricked were administered
to monitor the child’s attention to the task. Position
sense was examined by moving the great toe in an
upward or downward direction. The children were
asked to identify the position of the toe as ‘up’ or
‘down’.
Vibration sense was tested by applying a 128Hz
fork to the medial malleolus, tibial tuberosity and
first metatarsal head. The child was asked to identify
when he/she could no longer detect a vibration. The
trial was classified as a ‘pass’ if the child reported
cessation of the stimulus at or shortly after the
examiner could no longer feel the stimulus.
Direction of scratch was performed as described
by Hankey and Edis [28] using the wooden end
of a cotton applicator. A 2 cm scratch in a randomly
varied upward or downward direction was applied at
the same locations used to test pain and light touch
(see above). If the child was unable to correctly
identify three out of three tries, a 5 cm stimulus was
performed. If the child again was unable to correctly
identify three out of three stimuli then a 10 cm
scratch was performed.
Position sense of the knee was tested using a
previously described technique [29]. The child was
asked to sit with his/her legs hanging freely over the
edge of the table (at 90� of knee flexion). A cuff with
attached straps was placed around the lower leg that
allowed passive elevation of the leg. The leg was
raised to 15, 30 and 45� and returned to its original
position. The child was asked to reproduce the angle
without visual input.
Feasibility/representative sample study
Study design/sample. The study was performed at
a single tertiary care children’s hospital. It was
designed as a prospective cross-sectional observa-
tional study comparing the LE sensory status of
children with CP to a comparison group without
CP. Children selected were between the ages of
3–18 years and able to comprehend and respond
verbally or non-verbally to the techniques used.
Children with a diagnosis of CP were selected from
specialty clinics. Comparison children attended the
general medical clinic and had no previous history
of sensory deficits or a medical condition known to
be associated with sensory deficits. All comparison
children were developing in a typical manner by
parent report. Informed consent was obtained for
enrollment and testing. The institutional review
board at CHRMC approved the study. All testing
in the feasibility and reliability study was carried out
by one of two trained examiners (SF and SN).
Sensory function before and after dorsal rhizotomy
This study was carried out as part of quality
improvement monitoring of the DR procedure.
The same sensory test battery described above was
used except that knee position was not tested. Lower
extremity test results were analysed. The examina-
tions were carried out by two of the authors ( JM &
RH) at baseline and between 12–24 months later in
34 children participating in a randomized clinical
trial (RCT) with institutional review board approval.
The results of the RCT have been published
previously and are not the subject of this report
[30]. Twenty-nine additional children with spastic
diplegia were tested by the authors prior to DR.
Statistical analysis
Each sensory stimulus was repeated three times
at each site to allow for possible distraction or
inadequate stimuli in one of the trials. The number
of accurate and inaccurate responses for each site
tested for a given sensory modality was recorded.
Two or three accurate responses were considered a
pass for a given site. Credit was given for one missed
trial occurring in any order because young children
are easily distractible and the qualitative method-
ology does not assure exactly equal stimuli at each
trial. The Pearson �2 test was used to analyse the
data site-by-site for each specific sensory modality.
If a child could not co-operate, refused or was
physically unable to complete a particular sensory
task, the response was coded as a null data point.
After performing the site-by-site comparison, the
passes and fails were summed for all anatomic sites
for each modality for each child. Null data points
were excluded. With the exception of toe position,
each child was then assigned a pass for each sensory
modality if no more than one anatomic site was
scored as a fail. For toe position, a pass was assigned
only if both sites were passed. The statistical com-
parison between groups was performed for each
sensory modality using the Pearson �2. Because of
the use of multiple �2 tests, the p-values should be
interpreted conservatively. This study has chosen
to consider p-values less than or equal to 0.01 as
statistically significant.
Post-hoc analyses
Delayed cognition. The impact of delayed cognition
on sensory function was evaluated in two ways using
the data from the feasibility/representative sample
study. The presence of delayed cognition in the
Sensory function in cerebral palsy 47
Dev
Neu
rore
habi
l Dow
nloa
ded
from
info
rmah
ealth
care
.com
by
Uni
vers
ity o
f A
lber
ta o
n 10
/25/
14Fo
r pe
rson
al u
se o
nly.
group of children with cerebral palsy was determined
by parent report supplemented by chart review
and clinical impression independent of the sensory
examination procedure. For the purposes of this
report, ‘delayed cognition’ includes mental retarda-
tion, learning disability and developmental delay.
Typical cognitive development was a selection
criterion for the comparison group and was also
determined by parental report supplemented by
chart review. Within the group of children with CP,
sensory function was compared between children
with and without delayed cognition. LE sensory
function of children with CP and typical cognitive
development was compared to the children with
typical development. The comparison was done for
the sensory modality performance as described in the
statistical analysis section above.
Spastic diplegia. To assemble a larger sample of
children with spastic diplegia, the sensory function
data for all 34 children in the feasibility/representa-
tive sample with the diagnosis of spastic diplegia,
34 children in the DR randomized clinical trial at
baseline and an additional 29 children with spastic
diplegia who were evaluated for DR were pooled and
analysed by sensory modality as described in the
statistical analysis section above. This pooled sample
contains a total of 97 children aged 3–18 years. The
definition of spastic diplegia for all children in this
pooled sample is that of Hagberg et al. [31] and
excludes children with detectable dystonia or ataxia.
Results
Feasibility/representative sample study
Clinical status and demographics. Seventy-five
children with CP and 75 comparison children were
approached for enrollment during scheduled clinic
visits. Thirteen of the children with CP and 10
comparison children declined participation due
to prior commitments or time limitation. Sixty-
two children with CP and 65 comparison children
completed the study. Forty-six children with CP
had typical cognitive development and 16 had some
degree of delayed cognition. The types of CP include
34 children with spastic diplegia, 14 children with
spastic quadriplegia, six with left hemiplegia, four
with right hemiplegia, three with athetoid quadriple-
gia and one with hypotonic CP. Definition of the
type of CP was made by direct examination and used
the taxonomy of Hagberg et al. [31]. The average age
of the children with CP was 9.1 years (SD 4.0), vs
7.8 years (SD 3.85) for the comparison children
( p¼ 0.06). There were 38 boys and 24 girls in the
CP group and 37 boys and 28 girls in the comparison
group.
Feasibility/reliability of lower extremity
sensory examination
. CP Group: Thirty-two (52%) of the children
with CP were able to successfully complete the
LE sensory examination with the youngest child
being 5 years of age. Of the remaining 30 children,
10 were physically unable to perform the knee
position exam secondary to contractures. Comple-
tion of all sensory items in the children with CP
was not achieved in any age group (Table I).
. Comparison Group: The youngest child to suc-
cessfully complete the sensory examination was
4 years 1 month. Fifty-five children (85%) in this
group were able to successfully complete the
sensory examination. The remaining 10 children
either refused a particular test or could not
co-operate with the exam. The most commonly
refused items of the exam included direction of
scratch or pinprick. Lack of co-operation appeared
to be mainly due to distractibility. All children
greater than 5 years of age passed all items on the
sensory examination (Table I). There is a trend
favouring more complete testing in the comparison
group that is not statistically significant ( p¼ 0.09).
Sensory function by modality. Toe position ( p¼ 0.01),
direction of scratch ( p<0.001) and vibration sense
( p¼ 0.01) were failed by more children with CP
than the comparison children (Table II). Pinprick
Table II. Sensory testing by modality: children with cerebral palsy
and comparison children who correctly identified sensory stimuli.
Modality
Cerebral
palsy* (%)
Comparison*
(%) �2 p
Light touch 39/45 (87) 62/65 (95) 2.7 0.16
Pinprick 36/47 (77) 57/61 (93) 3.9 0.02
Toe position 32/41 (78) 60/63 (95) 7.2 0.01
Vibration sense 31/46 (67) 55/62 (89) 7.4 0.01
Direction of scratch 15/45 (33) 47/63 (75) 18.3 <0.001
Knee position 19/28 (68) 44/56 (79) 1.14 0.3
*Numerator¼number of children who correctly identified thestimuli; Denominator¼number of children who cooperated forthe test procedure (see text). Total n¼62 for the cerebral palsygroup and n¼65 for the comparison group.
Table I. Children with cerebral palsy and comparison children
who could be successfully tested for sensory function by age.
Age (years) Cerebral palsy* (%) Comparison* (%)
3–5 1/12 (8) 10/20 (50)
6–8 9/17 (53) 16/16 (100)
9–11 8/12 (67) 11/11 (100)
12–14 7/10 (70) 13/13 (100)
15–18 7/11 (64) 5/5 (100)
Total 32/62 (52) 55/65 (85)
* Numerator¼number of children who understood andco-operated with the test battery; Denominator¼ total numberof children in age group in the study (see text).
48 J. F. McLaughlin et al.
Dev
Neu
rore
habi
l Dow
nloa
ded
from
info
rmah
ealth
care
.com
by
Uni
vers
ity o
f A
lber
ta o
n 10
/25/
14Fo
r pe
rson
al u
se o
nly.
approached significance ( p¼ 0.02). Light touch
( p¼ 0.16) and knee position ( p¼ 0.3) were not
different. Sensory data for each anatomic site did not
reveal any variation by site and are not presented.
Lower extremity sensory function after
dorsal rhizotomy
The children in the dorsal rhizotomy RCT were
mostly young (DR group: 6.1 (SD 3.0) years, range
2.9–14.3 years; PT Only group 6.8 (SD 4.3) years,
range 3.0–17.3 years). As can be seen by inspection
of Table III, fewer of these children were able to
complete the sensory examination. Toe position
was especially difficult. There was no evidence of
any loss of sensory function after DR in those who
did complete the task. Given the obvious lack of
change, no statistical analysis was performed.
Post-hoc analyses
Delayed cognition. Delayed cognitive function might
influence performance on this sensory test battery.
Within the CP group, children with delayed cogni-
tion appeared to perform slightly less well on some
modalities, but none reached statistical significance.
Children in the CP group with delayed cognition
were then excluded from the between-group analysis.
Significant differences between children with
CP and the comparison group were still present
for direction of scratch ( p<0.001), toe position
sense ( p¼ 0.01) and vibration sense ( p¼ 0.01). No
difference was noted in knee position ( p¼ 0.18),
pinprick ( p¼ 0.02) and light touch ( p¼ 0.13). Data
not shown.
Spastic diplegia/sensory modality. When all sites were
combined for each sensory modality in the pooled
spastic diplegia sample, pinprick ( p¼ 0.001) and
direction of scratch ( p¼ 0.002) were failed by more
children with spastic diplegia than the comparison
children (Table IV). Toe position ( p¼ 0.05)
approached significance. Light touch ( p¼ 1) and
vibration sense ( p¼ 0.46) were not different.
There is a relatively high rate of inability to
complete the sensory test battery compared with
the feasibility/representative sample data associated
with the inclusion of younger children from the
DR sample.
Discussion
Feasibility/representative sample study
The first goal of this study was to establish whether
qualitative sensory testing is feasible in young
children. The results of the study indicate that
sensory testing is feasible in children as young as
5 years in the CP group and 4 years 1 month in the
comparison group. Some children with CP in every
age group demonstrated difficulties in completing
all test items. Children in the comparison group were
uniformly able to complete the exam after 5 years
of age. No children under 4 years of age were
able to successfully complete the examination in
either group. A number of children with CP with
normal test results were noted to be slower in
responding to the stimuli when compared to the
comparison children and to be less sure of their
responses.
Table III. Lower extremity sensory testing by modality in a randomized trial of dorsal rhizotomy: percentage of children in the surgical and
PT only groups at baseline and after 12 months who correctly identified sensory stimuli.
Baseline** Follow-up (>12 months)**
Modality* DRþPT (n¼19) PT only (n¼15) DRþPT (n¼ 17) PT only (n¼10)
Light touch 16/16 (100%) 14/14 (100%) 10/10 (100%) 8/8 (100%)
Pinprick 12/13 (92%) 11/13 (85%) 9/9 (100%) 7/8 (83%)
Vibration sense 9/11 (82%) 13/13 (100%) 7/8 (88%) 7/7 (100%)
Toe position 3/3 (100%) 5/6 (83%) 9/9 (100%) 5/5 (100%)
Direction of scratch 5/5 (100%) 10/10 (100%) 7/8 (88%) 4/4 (100%)
*Knee position not tested.**Numerator¼number of children who correctly identified the stimuli; Denominator¼number of children who co-operated for the testprocedure (see text).
Table IV. Sensory testing by modality: children with spastic
diplegia and comparison sample of children without CP who
correctly identified sensory stimuli.
Modality
Spastic
diplegia* (%)
Comparison*
(%) �2 p
Light touch 77/82 (94) 62/65 (95) 0.16 1
Pinprick 47/67 (70) 57/61(93) 11.4 0.001
Toe position 38/46 (83) 60/63 (95) 4.67 0.05
Vibration sense 58/69 (84) 55/62 (89) 0.59 0.46
Direction of scratch 28/55 (51) 47/63 (75) 9.7 0.002
*Numerator¼number of children who correctly identifiedthe stimuli; Denominator¼number of children who cooperatedfor the test procedure (see text). Total n¼ 97 for pooled spasticdiplegia group. Total n¼ 65 for the comparison group.
Sensory function in cerebral palsy 49
Dev
Neu
rore
habi
l Dow
nloa
ded
from
info
rmah
ealth
care
.com
by
Uni
vers
ity o
f A
lber
ta o
n 10
/25/
14Fo
r pe
rson
al u
se o
nly.
The second goal of this study was to determine
if there are differences in sensory testing in children
with CP. Past studies have shown that such deficits
exist in the upper extremities of children with
hemiplegia. No studies have focused on lower
extremity sensation or have included a comparison
group. The results indicate that, in this sample, there
are differences in lower extremity sensory function
between children with CP and typical children.
Decreased accuracy among children with CP for toe
position and direction of scratch are most prominent
and may have the greatest clinical importance.
Inspection of the data at individual lower extremity
anatomic sites revealed no localization to specific
dermatomes (data not presented). This leads one to
believe that the deficits are central in origin rather
than due to injury to peripheral nerves or discrete
sensory tracts.
Where are the presumed lesions causing the
sensory deficits? Mediation of the most affected
modalities in this sample (vibration, toe position,
direction of scratch) is traditionally assigned to the
dorsal columns. Lesions of the dorsal columns may
not be sufficient to explain the loss of those sensory
modalities traditionally associated with these tracts.
Wall and Noordenbos [32] described three patients
with defined lesions of the spinal cord and concluded
that patients with dorsal column lesions do not
lose one or more of the classic primary modalities
of sensation. Rather, there is a loss of the ability
to carry out tasks where the individual being
tested must simultaneously analyse spatial and
temporal characteristics of the stimulus. Vierck [33]
further demonstrated that monkeys with cut dorsal
columns could differentiate between stationary and
moving stimuli but were severely impaired in their
ability to detect direction of movement. The deficit
noted in the patients may not be localized to the
dorsal columns or their direct projections in the
brain.
Other authors have found that cerebral lesions
may produce sensory impairments similar to those
described here. Fox and Klemperer [34] described
seven patients with known cerebral lesions that
produced impaired position and vibration sense.
They further described various dysesthesias of
vibration sense in patients with these cerebral lesions.
One of the older patients with mild spastic
diplegia volunteered that vibration sense in the
lower extremities was qualitatively different and less
clear compared to the upper extremities. This may
indicate a sensory lesion that is cerebral in origin.
Knee position was not different between groups.
This may be due to the difficulty both groups had in
understanding and performing the task. In addition,
it was cumbersome to perform and did not seem as
useful as other tests.
Sensory function after dorsal rhizotomy
No loss of lower extremity sensory function was
detected following DR. The improvement noted
(Table III) is most likely attributable to maturation,
since most of the younger children were at least
5 years old at the time of follow-up. The absence of
sensory loss is consistent with the adverse events
questionnaire data reported in the original study
which revealed only transient post-operative changes
that would not be detected by this test battery [30].
The dorsal root tissue transection percentage was
a conservative 34% (range 20–56%) in the RCT. It
was not anticipated that the electrophysiologic
testing procedure to select rootlets for sectioning
would result in a lower percentage of root tissue
sectioning than in other similar studies. A more
aggressive transection rate might lead to detectable
sensory changes. Given the independent evidence
from the adverse event monitoring that there were
no clinically detectable sensory losses associated with
DR, the pre- and post-testing presented here provide
some support for the reliability of the sensory test
battery that was used. On the other hand, one cannot
exclude the possibility that a more sensitive measure
such as SEP testing might have detected differences.
Post-hoc analyses
In this study, delayed cognition did not make a
difference in the ability of children with CP to
correctly identify sensory stimuli. It is important to
keep in mind the fact that children with cognitive
function below that of a typically developing 3 year
old were excluded a priori from the study because
they were unable to comprehend the sensory testing
protocol. Cognitive testing was not performed in this
study, but depended on parental report and existing
data. Parental report is known to be sufficiently
reliable to allow the type of categorical classification
used in this study [35]. These factors preclude
any conclusions regarding the possible association of
increased sensory dysfunction in children with more
severe intellectual impairment.
If the pooled sample of children with spastic
diplegia is representative of the larger population,
the presence of deficits in accurate perception of
pinprick and direction of scratch is common. When
compared to the total sample of children with CP,
the children with spastic diplegia have fewer deficits.
In particular, disturbance of vibration sense does
not appear to be associated with spastic diplegia.
The sample of children with CP other than spastic
diplegia is too small to allow any reliable conclusions
about the possibility of distinct patterns of sensory
deficits by type of spastic CP (e.g. quadriplegia,
hemiplegia).
50 J. F. McLaughlin et al.
Dev
Neu
rore
habi
l Dow
nloa
ded
from
info
rmah
ealth
care
.com
by
Uni
vers
ity o
f A
lber
ta o
n 10
/25/
14Fo
r pe
rson
al u
se o
nly.
Conclusion
Qualitative lower extremity sensory testing is feasible
in typical children at 4 years of age and in some
children with CP as young as 5 years of age. Children
with CP may have qualitative differences in sensory
testing most notably in functions traditionally related
to the dorsal columns. These differences in sensation
appear to be central in origin, consistent with
anatomic localization of the motor impairments in
CP. The data provide no evidence that a conservative
DR procedure is likely to produce a measurable
sensory deficit in children with spastic diplegia.
The data do not reveal any evidence that adaptive
cognitive ability less than age level but above that of a
3-year-old impairs ability to respond to qualitative
sensory testing as described here.
Acknowledgements
Susan Astley PhD provided consultation regarding
statistical questions. The authors thank the children
and families for their participation. The research was
supported in part by a grant from the National
Institute of Neurological Disease and Stroke #RO1-
NS27867 and from the United Cerebral Palsy
Research and Educational Foundation. These data
were presented in part at the Annual Meeting,
American Academy for Cerebral Palsy and
Developmental Medicine, San Antonio, TX, 1998,
USA.
No commercial party having a direct financial
interest in the results of the research supporting this
article has or will confer a benefit on the authors
or on any organization with which the authors are
associated.
References
1. Bax M. Terminology and classification of cerebral palsy
(Annotations). Developmental Medicine and Child Neurology
1964;6:295–307.
2. Bax M. The spinal cord in motor disorders (Editorial).
Developmental Medicine and Child Neurology 1988;30:709–
710.
3. Clancy RR, Sladky JT, Rorke LB. Hypoxic-ischemic spinal
cord injury following prenatal asphyxia. Annals of Neurology
1989;25:185–189.
4. Kundi M, Cahan L, Starr A. Somatosensory evoked potentials
in cerebral palsy after partial dorsal rhizotomy. Archives of
Neurology 1989;46:524–527.
5. Hoon AH, Lawrie WT, Melhem ER, Reinhardt EM, van
Zilj PCM, Solaiyappan AM, Jiang H, Johnston MV, Mori S.
Diffusion tensor imaging of periventricular leukomalacia shows
affected sensory cortex white matter pathways. Neurology
2002;59:752–756.
6. Tizard JPM, Paine RS, Crothers B. Disturbances in sensation
in children with hemiplegia. JAMA 1954;155:628–632.
7. Twitchell TE. Sensation and the motor deficit in cerebral palsy.
Clinical Orthopaedics 1966;46:55–61.
8. Tachdjian M, Minear WL. Sensory disturbances in the hands
of children with cerebral palsy. Journal of Bone & Joint Surgery
1958;40A:85–90.
9. Jones MH. Management of hemiplegic children with periph-
eral sensory loss. Pediatric Clinics of North America 1960;7:
765–775.
10. Van Heest AE, House J, PutnamM. Sensibility deficiencies in
the hands of children with spastic hemiplegia. Journal of Hand
Surgery [America] 1993;18A:2278–2281.
11. Lesny I, Stehlik A, Tomasek J, Tomankova A, Havlicek I.
Sensory disorders in cerebral palsy: two-point discrimination.
Developmental Medicine and Child Neurology 1993;35:
402–405.
12. Bolanos AA, Bleck EE, Firestone P, Young L. Comparison of
stereognosis and two-point discrimination testing of the hands
of children with cerebral palsy. Developmental Medicine and
Child Neurology 1989;31:371–376.
13. Krumlinde-Sundholm L, Eliasson AC. Comparing tests
of tactile sensibility: aspects relevant to testing children
with spastic hemiplegia. Developmental Medicine and Child
Neurology 2002;44:604–612.
14. Dannenbaum RM, Dykes RW. Sensory loss in the hand after
sensory stroke: therapeutic rationale. Archives of Physical
Medicine Rehabilitation 1988;69:833–839.
15. Umansky R. The hand sock, an artificial handicap to
prehension in infancy, and its relation to clinical disuse
phenomena. Pediatrics 1973;52:546–554.
16. Asanuma H, Arissian K. Experiments on the functional role
of peripheral input to motor cortex during voluntary move-
ments in the monkey. Journal of Neurophysiology 1984;52:
212–227.
17. Bairstow PJ, Laszlo JI. Kinaesthetic sensitivity to passive
movements and its relationship to motor development and
motor control. Developmental Medicine and Child
Neurology 1981;23:606–616.
18. Taub E, Uswatte G, Elbert T. New treatments in neuro-
rehabilitation founded on basic research. Nature Reviews.
Neuroscience 2002;3:228–236.
19. Taub E. Somatosensory deafferentiation research with
monkeys: implications for rehabilitation medicine. In:
Ince LP, editor. Behavioral psychology in rehabilitation
medicine: Clinical applications. Baltimore, MD: Williams
and Wilkins; 1980. p 371–401.
20. Gordon AM, Duff SV. Relation between clinical measures
and fine manipulative control in children with hemiplegic
cerebral palsy. Developmental Medicine and Neurology
1999;41:586–591.
21. Anonymous. Proceedings of a consensus development con-
ference on standardized measures in diabetic neuropathy.
Quantitative sensory testing. Muscle Nerve 1992;15:
1155–1157.
22. Redmond J, Cross D, Shahani BT. Variability of quantitative
sensory testing: implications for clinical practice. Henry Ford
Hospital Medical Journal 1990;38:62–67.
23. Goldberg JM, Lindblom U. Standardized method of deter-
mining vibratory perception thresholds for diagnosis and
screening in neurological investigation. Journal of Neurology,
Neurosurgery and Psychiatry 1979;49:793–803.
24. Thibault A, Forget R, Lambert J. Evaluation of cutaneous
and proprioceptive sensation in children: a reliability study.
Developmental Medicine and Child Neurology 1994;36:796–
812.
25. Cooper J, Majnemer A, Rosenblatt B, Birnbaum R. A
standardized sensory assessment for children of school-
age. Physical Occupational Therapy in Pediatrics
1993;13:61–80.
26. Montgomery P. A clinical report of long term outcomes
following selective posterior rhizotomy: implications for
Sensory function in cerebral palsy 51
Dev
Neu
rore
habi
l Dow
nloa
ded
from
info
rmah
ealth
care
.com
by
Uni
vers
ity o
f A
lber
ta o
n 10
/25/
14Fo
r pe
rson
al u
se o
nly.
selection, follow-up, and research. Physical Occupational
Therapy in Pediatrics 1992;12:69–87.
27. McLaughlin JF, Bjornson KF, Astley SJ, Hays RM,
Hoffinger SA, Armantrout EA, Roberts TR. The role of
selective dorsal rhizotomy in cerebral palsy: critical evaluation
of a prospective clinical series. Developmental Medicine and
Child Neurology 1994;36:755–769.
28. Hankey GH, Edis RH. The utility of testing tactile perception
of direction of scratch as a sensitive clinical sign of posterior
column dysfunction in spinal dysfunction in spinal cord
disorders. Journal of Neurology, Neurosurgery and Psychiatry
1989;52:395–398.
29. Skinner HB, Wyatt MP, Hodgdon JA, Conard DW,
Barrack RL. Effect of fatigue on joint position sense of
the knee. Journal of Orthopaedic Research 1986;4:112–
118.
30. McLaughlin JF, Bjornson KF, Astley SJ, Graubert C,
Hays RM, Roberts TS, Price R, Temkin N. Selective dorsal
rhizotomy: efficacy and safety in an investigator-masked
randomized clinical trial. Developmental Medicine and
Child Neurology 1998;40:220–232.
31. Hagberg B, Hagberg G, Olow I. The changing
panorama of cerebral palsy in Sweden IV. Epidemiological
trends 1959–978. Acta Paediatrica Scandinavica 1984;73:
433–440.
32. Wall PD, Noordenbos W. Sensory functions which remain
in man after complete transection of dorsal columns. Brain
1977;100:641–653.
33. Vierck CJ. Tactile movement detection and discrimination
following dorsal column lesions in monkeys. Experimental
Brain Research 1974;20:331–346.
34. Fox JC, Klemperer WW. Vibratory sensibility. Archives of
Neurology and Psychiatry 1942;48:622–645.
35. Ireton H, Glascoe FP. Assessing children’s development using
parents’ reports: the Child Development Inventory. Clinical
Pediatrics 1995;34:248–255.
52 J. F. McLaughlin et al.
Dev
Neu
rore
habi
l Dow
nloa
ded
from
info
rmah
ealth
care
.com
by
Uni
vers
ity o
f A
lber
ta o
n 10
/25/
14Fo
r pe
rson
al u
se o
nly.