Hand function of C6 and C7 tetraplegics 1-16 years following injury

Abstract

Retrospective cohort study.
To quantify the hand function of C6 and C7 tetraplegics 1 - 16 years after injury.
Patients were assessed in their homes.
Medical records of patients admitted to the Prince Henry Hospital Spinal Injuries Unit between 1984 and 1999 were used to identify all patients with C6 or C7 tetraplegia at 3 months post injury. Sixty-five patients (107 hands) were identified in this way. Forty-seven patients (81 hands) were located and agreed to partake in the study, and seven (nine hands) had died. Thus 81% of patients (83% of hands) still alive at follow-up were assessed.
Unilateral hand function was assessed with the Grasp and Release Test (GRT) and a 10 item Activities of Daily Living (ADL) Test. Prevalance and severity of contractures, lateral grasp (key grip) strength and extensibility of the extrinsic finger flexor muscles were also determined.
All hands except one had been managed without surgical intervention. The median number of ADL tasks successfully completed was 9/10 (interquartile range=8 - 10) and the median number of objects successfully manipulated in the GRT was 3/6 (IQ range=3 - 5). Lateral grasp was poor (74% and 75% of hands could not use a lateral grasp to move the paperweight or depress the fork in the GRT, respectively), and the prevalence of contractures was low (53% of hands had full passive range of motion).
In the long term, most C6 and C7 tetraplegics attain a high level of hand function despite poor lateral grasps.

Full text

Hand function of C6 and C7 tetraplegics 1 ± 16 years following injury
LA Harvey*
,1
, J Batty
2
, R Jones
3
and J Crosbie
1
1
School of Physiotherapy, University of Sydney, Australia;
2
Department of Physiotherapy, The Prince Henry
Hospital, Australia;
3
Rehabilitation and Spinal Medicine, The Prince Henry Hospital, Australia
Study design: Retrospective cohort study.
Objective: To quantify the hand function of C6 and C7 tetraplegics 1 ± 16 years after injury.
Setting: Patients were assessed in their homes.
Methods: Medical records of patients admitted to the Prince Henry Hospital Spinal Injuries
Unit between 1984 and 1999 were used to identify all patients with C6 or C7 tetraplegia at 3
months post injury. Sixty-®ve patients (107 ha nds) were identi®ed in this way. Forty-seven
patients (81 hands) were located and agreed to partake in the study, and seven (nine hands)
had died. Thus 81% of patients (83% of hands) still alive at follow-up were assessed.
Main outcome measures: Unilateral hand function was assessed with the Grasp and Release
Test (GRT) and a 10 item Activities of Daily Living (ADL) Test. Prevalance and severity of
contractures, lateral grasp (key grip) strength and extensibility of the extrinsic ®nger ¯exor
muscles were also determined.
Results: All hands except one had been managed without surgical intervention. The median
number of ADL tasks successfully completed was 9/10 (interquartile range=8 ± 10) and the
median number of objects successfully manipulated in the GRT was 3/6 (IQ range=3 ± 5).
Lateral grasp was poor (74% and 75% of hands could not use a lateral grasp to move the
paperweight or depress the fork in the GRT, respectively), and the prevalence of contractures
was low (53% of hands had full passive range of motion).
Conclusion: In the long term, most C6 and C7 tetrapleg ics attain a high level of hand
function despite poor lateral grasps.
Spinal Cord (2001) 39, 37 ± 43
Keywords: tetraplegia; upper limb function; hand; paralysis; rehabilitation
Intro duction
C6 and C7 tetraplegics have paralysis of ®nger and
thumb ¯exor muscles but retain voluntary control of
wrist extensor and sometimes wrist ¯exor and ®nger
extensor muscles.
1,2
Despite such extensive paralysis
these patients are generally able to hold and
manipulate objects.
3
An ability to eectively use the
hand is critical to independence and quality of life.
4,5
Consequently a great deal of attention is directed at
optimising hand function of tetraplegics.
In the past, C6 and C7 tetraplegics' hands were
managed with manual therapeutic regimes involving
the administration of regular pas sive moveme nts and
stretches and the early application of spl ints.
6±9
In
addition or alternatively, patients were provided with
orthotics and adaptive equipment.
1,7,10
In more recent
years surgery has been widely advocated as the
treatment of choice,
5,11 ± 14
though a large proportion
of patients are either unsui table for sur gery , do not
have access to these types of service, or opt for non-
surgical interventions. Neuroprostheses are also
occasionally used to enhance the hand function of
this group of patie nts,
4,15,16
though less comm only in
C7 tetrapl egics.
Surprisi ngly, there is little funda mental information
about the hand function of C6 and C7 tetraplegics.
Most studies either measure hand fun ction following
surgery
2,11,12,17 ± 19
or implantation of neuroprosth-
eses.
15,20,21
Alternatively, studies outline the principles
underlying dierent methods of conservative
1,7 ± 9,22
and s urgical management.
2,5,23,24
Whilst some studies
prov ide more quant itative i nformation about the ha nd
function of C6 and C7 t etraplegics, t hey recruit small
numbers of subjects on the basis of convenience and
consequently are susceptible to selection bias.
25
No
study has yet quanti®ed the level of hand function
attained by a large and representative cohor t of C6
and C7 tetraplegics.
Quantitative information about the hand function
typically attained by C6 and C7 tetraplegics is va luable
for two reasons. Firstly, it enables therapists, clinicians
*Correspondence: LA Harvey, School of Physiotherapy, University
of Sydney, P.O. Box 170, Lidcombe, NSW, 1825, Australia.
Spinal Cord (2001) 39, 37 ± 43
ã
2001 International Medical Society of Paraplegia All rights reserved 1362 ± 4393/01 $15.00
www.nature.com/sc

and s urgeons to better de®ne `good' and `poor'
outcomes. In this way, they are better equipped to
objectiv ely g auge the eectiveness of dierent inter-
ventions within the clinical setting. Secondly, informa-
tion abo ut the level of hand function typically attained
is important for predicting hand f unction of tetra -
plegics. The ability to foresee future hand function
enables therapists and clinicians to provide recently
injured pa tients and their long-term care providers
with accurate information about the leve l of hand
function that can be expected.
The purpose of t his study was to quant ify the hand
function of a large cohort of C6 and C7 tetr aplegics
long after injury. Speci®cal ly, the aims were to
determine the eectiveness of pal mar and lateral
grasps, and assess abilities to use the hand to perfo rm
useful and purposeful unilateral tasks. Subjects in the
cohort were managed conservatively (i e, without
surgery) and thus the results of this study can be
used to quantify the lev el of hand function that can be
expe cted in the absenc e of surgical intervention.
Methods
Subjects
Medical records of patients admitted to the Prince
Henry Hospital Spinal Injuries Unit in Sydney between
1984 and 1999 were examined. Patients were identi®ed
if during this period they had presented with a sudden
onset of tetraplegia following trauma, surgery or a
medical condition. All complete or incomplete C6 and
C7 tetraplegics with grade 2/5 or less strength in the
®nger ¯exor muscles at approximately 3 months after
the onset of tetraplegia were included in the study
(according to the American Spinal Injuries Associa-
tion's standards for neurological classi®cation
26
).
Strength at 3 months post-injury was used as a
selection criterion, rather than strength immediately
post-injury, to eliminate patients whose neurological
status improved markedly in the ®rst 3 months.
Patients who had died or been transferred elsewhere
within the ®rst 3 months after injury were also
excluded.
Sixty-®ve patients were identi®ed in this way. All
were male except eight. Both hands of 42 patients and
one hand of 2 3 patients met the inclusion criteria (107
hands in total). The median (interquartile range) age
at injury was 28 years (19 ± 37 years). Forty-seven of
the initial coh ort of 65 patients were locat ed and
agreed t o participate in the study . In 34 both hands
were suita ble for inclusion and in the other 1 3 only
one hand was suita ble for i nclusion (81 hands in total).
Seve n patie nts (nine hands) had died (median and
interquartile age at injury was 62 years and 34 ± 67
years, and median and interquartile age at death was
69 years and 42 ± 74 years). The remaining patients
had either moved overseas (six patients, 10 hands),
were unable to be located (one patient, one ha nd) or
declined to be involved (four patients, six hands). In
this way 81% of patients (83% of hands) alive at
foll ow-up were reviewe d. The project was approved by
the appro priate Ethics Committees and all participants
signed an informed consent sheet. Subjects were
assessed by one of two therapists at their place of
residence.
Measurement procedure
Strength of lateral and palmar grasp Initially
`strength' of the lateral and palmar grasps was
measured. Strength of the lateral grasp was measured
with a modi®ed pinch meter.
21,27
Subjects were tested
three times and the median score was recorded.
Strength of the palmar grasp was assessed by
determining the ability to hold dierent sized and
weighted cylinders in the palm of the hand with the
thumb adducted adjacent to the index ®nger (ie, the
thumb was not hooked around the cylinders).
Cylinders were 2.5, 5 or 7.5 cm in diameter and
weighted 0.1, 0.5, 1 or 2 kg, thus there were 12
dierent cylinders. The cylinders were placed in
subjects' hands by the assessor and subjects were
required to hold the cylinders upright for 10 s. Subjects
were scored out of 12 according to the number of
cylinders successfully held.
Passive range of motion Passive range of motion
(ROM) of the wrist, metacarpoph alangeal (MCP) an d
interphalangeal (IP) joints were measured with a
goniometer and rated according to the passive range
of motion of the least mobile joint. The convention
used was:
.
Full ROM, indicated full passive ra nge of motion in
wrist, MC P and IP joints.
.
Minimal loss of ROM, indicated loss of passive
range of motion in wrist, MCP or any IP joint of
158 or le ss.
.
Moderate loss of ROM, indicated loss of passive
range of motion in wrist, MCP or any IP joint of
15 ± 458.
.
Severe loss of ROM, indicated loss of passive range
of motion in wrist, MCP or any IP joint of 458 or
more.
Passive ¯exion of the MCP and IP joints was
measured with the wrist extended to reduce stretch on
the extrinsic ®nger and thumb extensor mus cles. In the
same way passive extension of the MCP and I P joints
was measured with t he wrist ¯exed to reduce stretch
on t he extrinsic ®nger and th umb ¯exor muscl es. In
this way, joint range of motion measures were not
overly in¯uenced by the extensibility of multi-articular
muscles such as the extrinsic ®nger ¯exor and extensor
muscles.
22,28
Extensibility of the extrinsic ®nger ¯exor muscles Mea-
sures of passive wrist extension with the MCP and IP
Hand function of tetraplegics
LA Harvey et al
38
Spinal Cord

joints maintained in extension were used to re¯ect the
extensibility of the extrinsic ®nger ¯exor muscles.
Provided the extrinsic ®nger ¯exor muscles are
suciently stretched over the MCP and IP joints of
the ®ngers, they, rather than other muscles and soft
tissue spanning the ¯exor aspect of the wrist, will
predominantly limit wrist extension.
29
In this way
limited passive wrist extension re¯ects limited extensi-
bility in the extrinsic ®nger ¯exor muscles and vice-
versa. Passive wrist extension was measured with a
device previously described and tested for its relia-
bility.
29
A 0.72 Nm wrist e xtension torque was applied
and wrist angle was measured from a goniometer
attached to the side of the measuring device. Zero
degrees indicated a neutral posit ion of the wrist and
limited extensibility of the extrinsic ®nger ¯exor
muscles.
Lateral and palmar grasp The eectiveness of lateral
and palmar grasps was assessed with the Grasp and
Release Test (GRT).
15,21,27
The GRT was designed
primarily to assess the eectiveness of lateral and
palmar grasps in C5 and C6 tetraplegics ®tted with
neuroprostheses. It assesses the ability to pick up,
move and release six objects of varying sizes, weights
and textures using a palmar or lateral grasp. Each
hand is tested and scored separately. The objects used
in the GRT include a peg, paperweight, fork, block,
can and videotape. Each object was cho sen to
represent one or more objects routi nely manipulated
for activities of daily living (ADL). For example, the
paperweight, can and videotape were included to
re¯ect subjects' abilities to hold large heavy objects
such as a glass or book. The fork was included to not
only assess subjects' ability to stab food but also their
ability to grasp a pen. The fork was attached to a
spring-loaded piston and subjects were required to
hold the `fork' and depress it downwards with a 4.4 N
vertical force. Subjects were not permitted to depress
the fork without grasping it between the thumb and
index ®nger.
The GRT was administered according to conven-
tional protocol
15,21,30
and initially subjects were given
a pre -test to determine which tasks they could and
could not complete. Subjects wer e not tested on tasks
that they were unable to perform in the pre-test.
Subj ects were given an opportunity to practise prior to
test ing and sucient r est was provided betwe en each
item to minimise fatigue. The objects were presented in
a random order and subjects were requir ed to use
either a palmar (for the block, video and can) or
lateral (for the peg, fork a nd paper weight) grasp to
move or depress each item. Subje cts were sc ored on
their ability to successfully move (or in the case of the
fork, depress) eac h of the objects as ma ny times as
possible in 30 s. In addition, a summed score (tallied
GRT score) for each subject was calculated by tallying
scores across ite ms. If a subject failed to move an item
in the pre-test they scored ze ro for that particular
item.
Ability to use hand in any way to perform functional
tasks An Activities of Daily Living Test (ADL test),
similar to ones recommended by others,
20,21,31 ± 33
was
also administered. It assessed subjects' ability to
independently perform 10 unilateral tasks, namely
using a fork, applying toothpaste to a toothbr ush,
answering a phone, writing with a pen, lifting a glass
and mug to the mouth and picking-up a biscuit,
toothbrush, computer disc and book. Unlike the
GRT, subjects were allowed to use any method to
perform each task (ie, they were not solely restricted
to a lateral or palmar grasp). In this way, the test
assessed the ability of subjects to use their hands in
any way to perform purpo seful tasks. However,
subjects were not allowed to use splints, their other
hand, another part of their body or adapted
equipment.
The validity, reliability and sensitivity of the GRT
and the ADL test have not been formally assessed.
Howe ver, both were initially designed fr om existing
hand asses sment tools
15,34
after extensive review of all
other available assessment tools (see reference
15
for
summ ary of problems associated with using either the
Sollerman or Jebsen tests for assessing tetra plegic
hand functi on). The GRT and ADL test utilise objects
commonly used in activities of daily living, so have
face and content validity. In addition, there i s some
evidence to indicate that the GRT and similar
activities of daily living tests are sensitive to change
in function, especially following i mplantation of
neuropros theses.
4,20,21,27,32,34
Data reduction and analysis
Median and interquartile ranges were derived for all
variables unless otherwise speci®ed. In subjects with
both hands included in the study, the results of each
hand were analyzed separately. This was necessary
because often subjects had asymmetrical lesions. In
order to avoid confusion arising from this method of
data analysis, most results are presented in terms of
hands rather than subjects.
The neurological status of hands at follow-up were
categorised into the following three groups:
(i) com plete and incomplete C6 lesion with less than
grade 3/5 str ength in the ®nger and/or thumb
¯exor musc les (C6 subj ects).
(ii) complete and incomplet e C7 lesion with less than
grade 3/5 str ength in the ®nger and/or thumb
¯exor musc les (C7 subj ects).
(iii) complete and incomplete C6 to C8 lesion with
grade 3/5 or more strength in the ®nger and/or
thumb ¯exor muscles (®nger/thumb ¯exor sub-
jects).
In addition, for descriptive pu rpose s, subjects'
hands were also classi®ed accord ing to Moberg's
modi®ed International Classi®catio n for Surgery of
the Hand in Tetraplegia (ICSHT
2,34
).
Hand function of tetraplegics
LA Harvey et al
39
Spinal Cord

Results
The median age at time of injury (and interquartile
range) of the 47 reviewed subjects (81 hands) was 29
years (19 ± 37 years; see Table 1). Subjects were
reviewed on average (SD) 8 years since injury (5
years). Table 1 provides information about the
neurological status of subjects' hands at follow-up
according to the classi®cations of ASIA and ICSHT.
At follow-up 62 hands (77%) had either complete or
incomplete C6 and C7 lesions with less than grade 3/5
strength in the ®nger and thumb ¯exor muscles. Whilst
no hand of any subject had more than grade 2/5
strength in the ®nger or thumb ¯exor muscles at 3
months post injury, 19 hands (23%) had grade 3/5 or
more strength in these muscles at follow-up. Only one
hand of a subject had deteriorated neurologically. This
subject's hand had motor and sensory loss consistent
with a complete C6 lesion at 3 months post injury (ie,
wrist extension strength was 3/5) but consistent with a
complete C5 lesion at follow-up (ie, wrist extension
strength was 3-/5). At the risk of inaccuracy but for the
purpose of simplicity his lesion was classi®ed as C6 for
all analyzes.
All subjects' hands but one were conse rvatively
managed with conventional occupational therapy and
phys iotherapy.
22
The majority of subjects wore splints
for between 2 and 4 months after injury but were not
receiving any therapy (from therapists or carers) at
foll ow-up. H owever, most state d tha t they themselves
moved or stretched the ir hands regularly for
therapeutic purposes. One sub ject with a bilateral C6
motor-complete lesion had undergon e tendon transfers
(brachioradialis to ¯exor pollicis longis and posterior
deltoid to triceps) in one extremity, 1 year after injury.
Whilst at follow -up his lateral grasp was re latively
strong (8.3 N), his hand function was not dissimilar to
his C6 counterparts (for example, he could successfully
manipulate nine of the 10 objects in the ADL test and
his tallied GRT score was 34). This subjects' data were
included in all analyses.
Passive range of motion Only 16 hands (of nine
subjects) had `moderate' or `severe' loss of passive
range of motion (see Table 1) and more than half the
hands (ie, 52%) had full passive range of motion.
Passive range of motion in the elbow and shoulder
were not measured, however it was noted that three
subjects (three hands) had severe co ntracture in one or
both of these joints preventing reaching to the face
and/or the testing table.
Strength of late ral and palmar grasp and extensibility of
the extrinsic ®nger ¯exor muscles Extensibility of the
extrinsic ®nger ¯exor muscles and `strength' of the
lateral and palmar grasps are presented in Table 1. Not
surprisingly, subjects with paralysis of the ®nger and/or
thumb ¯exor muscles had weaker lateral and palmar
Table 1 Characteristics of hands at follow-up including median (interquartile range) age at time of injury (years), time since
injury (years), classi®cation of hands according to ICSHT, lateral pinch strength (N), cylinder tally (number of cylinders held),
passive wrist extension with MCP and IP extension, and prevalence of contractures. Contractures prevented testing of the
extensibility of the extrinsic ®nger ¯exor muscles in 11 hands. For the purpose of analyses subjects' hands were classi®ed into
three groups, namely complete and incomplete C6 lesions with less than grade 3/5 strength in ®nger and thumb ¯exor muscles,
complete and incomplete C7 lesions with less than grade 3/5 strength in ®nger and thumb ¯exor muscles and complete and
incomplete C6 ± 8 lesions with less than grade 3/5 strength in ®nger and thumb ¯exor muscles
C6 with 5 grade 3/5
®nger/thumb ¯exion
(n=38)
C7 with 5 grade 3/5
®nger/thumb ¯exion
(n=24)
C6 ± C8 with 5 grade 3/5
®nger/thumb ¯exion
(n=19)
Age at time of injury 31 (19 ± 33) 24 (19 ± 31) 56 (29 ± 57)
Time since injury 10 (4 ± 14) 8 (4 ± 10) 6 (4 ± 12)
Classi®cation of hands
according to ICSHT
Cu.1=1
a
, O.2/3=5;
Cu.2/3=12
b
, O.4=1;
Cu.4=13
c
, O.5=2; Cu.5=4
O.2/3=1; Cu.4=5;
Cu.5=14; Cu.6=4
Cu.2/3=3; Cu.4=4;
Cu.5=5; Cu.6=5
d
,
Cu.7=2
Lateral pinch strength 3 (1.9 ± 5.8) 4 (2.5 ± 6.7) 17 (12.5 ± 30.8)
Cylinder tally (max.=12) 3 (2 ± 5) 5 (3 ± 6) 8 (6 ± 11)
ROM:
Full passive range 17 18 7
5158 loss 13 5 3
15 ± 468 loss 5 1 3
4458 loss 3 0 4
Wrist extension with MCP
and IP extension
85 (80 ± 90) 80 (75 ± 85) 75 (60 ± 75)
Tallied GRT score 35 (20 ± 53) 47 (34 ± 57) 70 (53 ± 100)
a
One hand had grade 3-/5 strength in the wrist extensor muscles but was classi®ed as ASIA C6 for simplicity (see text for
details).
b
One hand had been treated surgically (see text for details).
c
Two hands had severe elbow and/or shoulder contractures
restricting upper limb movement between the testing table and hand (see text for details).
d
One hand had severe elbow and/or
shoulder contractures restricting upper limb movement between the testing table and hand (see text for details)
Hand function of tetraplegics
LA Harvey et al
40
Spinal Cord

grasps than the C6 ± C8 subjects with grade 3/5 or
more strength in these muscles.
Lateral and palmar grasp The eectiveness of lateral
and palmar grasps was measured with the GRT.
Subjects that passed the pre-GRT rarely dropped or
failed to successfully grasp an item during testing
(median and interquartile failure rate was 0 and 0 ± 1
failures per tested item). Nearly all subjects could
successfully move the peg and block, regardless of
paralysis of the ®nger and thumb ¯exor muscles
(Figures 1 and 2). The hands of subjects with grade
3/5 or more strength in the ®nger and/or thumb ¯exor
muscles were more successful at manipulating the
paper weight, fork and video than the hands of
subjects with paralysis in these muscles. In contrast,
there was little dierence in the ability of subjects to
move the can.
The tallied GRT scores are provided in Table 1.
Subj ects with active ®nger or thumb ¯exion at follow -
up attained higher tallie d GR T scores than those
without.
Ability to use the hand in any way to perform functional
tasks The ADL test re¯ect s subjects' ability to use
their hands in any way to perform purposeful tasks.
Figure 3 shows the percentage of hands that could
successfully manipulate each of the 10 items. Subjects
had particular diculty with the fork, glass and pen.
Figure 4 shows the correlation between tallied GRT
scores and ADL test scores. There was a clear `ceiling'
eect with the ADL test, and a close but non-linear
correlation between the GRT and ADL test scores
(Spearman's rho=0.80).
Discussion
This study quanti®es the hand function of C6 ± C7
tetraplegics 1 ± 16 years after injury. All subjects had at
Figure 1 Percentage of hands that passed the initial pre-
GRT (ie, successfully moved items in the GRT at least once
with the required grasp). For the purposes of analyses
subjects' hands were classi®ed into three groups, namely
complete and incomplete C6 lesions with less than grade 3/5
strength in ®nger and thumb ¯exor muscles (C6 subjects),
complete and incomplete C7 lesions with less than grade 3/5
strength in ®nger and thumb ¯exor muscles (C7 subjects),
and complete and incomplete C6 ± C8 lesions with grade 3/5
or more strength in the ®nger and/or thumb ¯exor muscles
(®nger/thumb ¯exor subjects)
Figure 2 Median (and interquartile) number of successful
grasps and releases in 30 s of the items in the GRT. Only the
data of subjects that successfully passed the pre-test are
included. For the purposes of analyses subjects' hands were
classi®ed into three groups, namely complete and incomplete
C6 lesions with less than grade 3/5 strength in ®nger and
thumb ¯exor muscles (C6 subjects), complete and incomplete
C7 lesions with less than grade 3/5 strength in ®nger and
thumb ¯exor muscles (C7 subjects), and complete and
incomplete C6 ± C8 lesions with grade 3/5 or more strength
in the ®nger and/or thumb ¯exor muscles (®nger/thumb
¯exor subjects)
Figure 3 Percentage of hands that independently manipu-
lated the dierent items of the ADL test. For the purposes of
analyses subjects' hands were classi®ed into three groups,
namely complete and incomplete C6 lesions with less than
grade 3/5 strength in ®nger and thumb ¯exor muscles,
complete and incomplete C7 lesions with less than grade 3/5
strength in ®nger and thumb ¯exor muscles, and complete
and incomplete C6 ± C8 lesions with grade 3/5 or more
strength in the ®nger and/or thumb ¯exor muscles
Hand function of tetraplegics
LA Harvey et al
41
Spinal Cord

least grade 3/5 wrist extension and not more than
grade 2/5 ®nger or thumb ¯exion at 3 months post
injury and only one subject had undergone surgical
intervention. Previous studies ha ve also provided
descriptive information about the hand function of
C6 and C7 tetraplegics managed conservatively,
7,25
but
such studies have recruited small numbers of subjects
on the basis of convenience and have consequently
been susceptible to selection bias. Bias of this type was
minimised in this study by assessing 81% of a
consecutive series of eligible C6 and C7 tetraplegics
still alive at follow-up.
The ADL tes t asse ssed the ability of subjects to
perform important activitie s of daily living . The results
indicated that subjects attained high levels of hand
function without surgical interv ention and despite
extensive paralysis. For example, 71% of the C6
hands, 79% of the C7 hands and 8 9% of C6 ± C8
hands with active ®nger and/or thumb ¯exion could
independently perform at least eight of the 10 tasks.
Even hi gher levels of hand function would have been
attained if subjects were not restricted to using one
hand.
Often subjects used novel and ingenious ways to
perform the tasks of the ADL test wit hout utilising
lateral or palmar grasps.
25
For exa mple, subj ects
unable to us e a lateral grasp to pick up the book
often were able to pick it up between the second and
third ®ng er or rotate it over and cradle it in the palm
of the hand. Similarly su bjects unable to use a lateral
grasp to hold the fork or pen could weave these
utensils over and under ®ngers.
Currently there is no standardised o r accepted tool
for assessing tetraplegics' ability to use their hands in a
functional way,
1,6,9,15,3 5
though such a tool is urgently
needed. The ADL test used in this study was similar to
ones previously used to asse ss the eectiveness of
neur oprostheses,
20,21,31 ± 33
and while it provided
valuable information about subjects' abi lities to
perform important activities of daily living, there
were some diculties in dierentiating subtle differ-
ences in function of the more capable subjects. The se
problems could be somewhat ove rcome either by
changing the scoring system to account for differences
in the time taken to complete each task or by
including some more c hallenging manipulative ta sks.
The eectiveness of the lateral and palmar grasp
was measured with the GRT (the peg, paper weight
and fork items were used to assess la teral grasp and
the block, can and video it ems were used to assess
palmar grasp). Subjects with paralysis of the ®nger
and/or thumb ¯e xor m uscles were generally unable
to hold the fork or paperweight betwe en t he thumb
and index ®nger. Similarly they were unable to use a
palmar grasp to pick up and move the video, though
they had little dic ulty with the block and can
items. The few subj ects with paralysis of the ®nger
and/or thumb ¯exor muscles who successfully picked
up these items oft en enhanced their grasps by
elic iting spasticity in the ®nger and thumb ¯exor
muscles. Interest ingly, subjects with active ®nger and/
or thumb ¯ex ion also had diculty completing the
can task. This wa s perhaps because four sub jects
(seven hands) had un usual patterns of ne urol ogical
loss with active ®nger ¯exion but paralysis of the
®nger e xtensor muscles. These patients tended to
have c ontr actures of the IP joints and/or excessive
loss of extensibility i n the ®nger ¯exor muscles,
making it dicult for them to initially get their
®ngers around the can.
The incidence of contracture was low. Some subjects
with moderate or severe contractures at follow-up
attained high talli ed GRT and ADL test scores. One
notable subjec t with active ®nger and thumb ¯exion
had severe contractures in his hand yet independent ly
manipulated nine of the 10 items in the ADL test and
attained a tallied GRT score of 79.
Curr ently there is little consensus about optimal
management of the C6 and C7 tetraplegic hand.
Whilst surgery is widely advocated, its bene®ts have
not been clari®ed with good quality rand omised
clinical trials. Similarly, no splinting or manual
therapy protocol has been subjected to rig orous
evaluati on. This is in part due to the inherent
diculties associated with performing good quality
research in this area. The results of the present study
prov ide baseline information on the hand function of
C6 and C7 tetraplegics managed without surgery 1 ± 16
years after injury. This information can be used by
researchers to quantify the eectiveness of dierent
interventions. In addition, this information can be
used by clinicians to monitor outcomes and objectively
gaug e the eectiveness of dierent interventions on a
day-to-day basis.
Acknowledgements
We ackn owledge the ®nanc ial assistance prov ided by the
Motor Accident Author ity of NSW.
Figure 4 Correlation between tallied GRT score and
number of items successfully manipulated in the ADL test
Hand function of tetraplegics
LA Harvey et al
42
Spinal Cord

References
1 Sutton S. An overview of the management of the C6 quadriplegic
patient's hand: an occupational therapist's perspective. Br J Occ
Ther 1993; 56: 376 ± 380.
2 House JH. Reconstruction of t he thumb in tetraplegia following
spinal cord injury. Clin Orthop Rel Res 1985; 195: 117 ± 128.
3 Peckham PH, Marsolais E, Mortimer JT. Restoration of key grip
and release in the C 6 tetraplegic patient through functional
electrical stimu lation. J Hand Surg (Am) 1980; 5A: 462 ± 469.
4 Mulcahey MJ et al. Functional neuromuscular stimulation:
outcomes in young people with tetraplegia. JAmParaSoc
1994; 17: 20 ± 35.
5 Waters RL, Sie I, Gellman H, Tognella M. Functional hand
surgery following tetraplegia. Arch Phys Med Rehab 1996; 77:
86 ± 94.
6 Curtin M. The management of the C 6 quadriplegic patient's
hand. ( letter) Br J Occ Ther 1993; 56: 455.
7 DiPasquale-Lehnerz P. Orthotic intervention for development o f
hand function with C-6 quadriplegia. Am J Occ Ther 1994; 48:
138 ± 144.
8 Krajnik SR, Bridle MJ. Hand splinting in q uadriplegia: current
practice. Am J Occ Ther 1992; 46: 149 ± 156.
9 Curtin M. Development of a tetraplegic hand assessment and
splinting protocol. Paraplegia 1994; 32: 159 ± 169.
10 Nichols PJ, Peach SL, Haworth RJ, Ennis J. The value of ¯exor
hinge and splints. Pros Orth Int 1978; 2: 86 ± 94.
11 Gansel J, Wates R, Gellman H. Transfer of the pronator teres
tendon to the tendons of the ¯exor digi torum profundus in
tetraplegia. J Bone Joint Surg (Am) 1990; 72A: 427 ± 432.
12 House JH , Comadoll J, Dahl A. One-stage key pinch and release
with thumb carpal-metacarpal fusion in tetraplegia. J Hand Surg
(Am) 1992; 17A: 530 ± 538.
13 Freehafer AA. Tendon transfers in patients with cervical spinal
cord injury. J Hand Surg (Am) 1991; 16A: 804 ± 809.
14 Moberg E. Surgical rehabilitation of the upper limb in
tetraplegia. Paraplegia 1990; 28: 330 ± 334.
15 Wuolle KS et al. Development of a quantitative hand grasp and
release test for patients with tetraplegia using a hand
neuroprosthesis. J Hand Surg (Am) 1994; 19A: 209 ± 218.
16 Keith M et al. Tendon transfers and functional electrical
stimulationforrestorationofhandfunctioninspinalcord.J
Hand Surg (Am) 1996; 21A: 89 ± 99.
17 KellyCM,FreehaferAA,PeckmanPH,StrohK.Postoperative
results of o pponensplasty and ¯exor tendon transfer in pat ients
with spinal cord injury. J Hand Surg (Am) 1985; 10A: 890 ± 894.
18 Hentz VR, B rown M, Keoshian LA. Upper limb reconstruction
in quadrip legia: functional assessment and propose d treatment
modi®cations. J Hand Surg (Am) 1983; 8A: 119 ± 131.
19 Vanden Berghe A, Van Laere M, Hellings S, Ve rcauteren M.
Reconstruction of the upper extremity in tetraplegia: functional
assessment, surgical procedures and rehabilitation. Paraplegia
1991; 29: 103 ± 112.
20 Wijman CA et al. Funct ional evaluation of quadriplegic patients
using a hand neuroprosthesis. Arch Phys Med Rehab 1990; 71:
1053 ± 1057.
21 Mulcahey MJ et al. I mplanted functional electrical stimulation
hand system in adolescents with spinal injuries: an evaluation.
Arch Phys Med Rehab 1997; 78: 597 ± 607.
22 Harvey L. Principles o f conservative non-orthotic management
for a tenodesis grip. J Hand Ther 1996; 9: 238 ± 242.
23 Waters R, Moore KR, Grabo SR, Paris K. Brachioradialis to
¯exor pollicis longus tendon transfer for active lateral pinch in
the tetraplegic. Hand Surg (Am) 1985; 10A: 385 ± 391.
24 Moberg E. Surgical treatment for absent s ingle-hand grip and
elbow extension in quadriplegia. Principles and preliminary
experience. J Bone Joint Surg (Am) 1975; 57A: 196 ± 206.
25 Curtin M. An analysis of tetraplegic hand grips. Br J Occ Ther
1999; 62: 444 ± 450.
26 Ditunno JF, Young W, Donovan WH, Creasey G. The
international standards booklet for neurological and functional
classi®cation of spinal cord i njury. American Spinal Injuries
Association. Paraplegia 1994; 32: 70 ± 80.
27 Smith BT, Mulcahey MJ, Betz RR. Quantitative com parison of
grasp and release abilities with and without funct ional
neuromuscular stimulation in adolescents with tetraplegia.
Paraplegia 1996; 34: 16 ± 23.
28 Keenan MA et al. Results of transfer of the ¯exor digitorum
super®cialis te ndons to the ¯exor digitorum prof undus tendons in
adults with acquired spasticity of the hand. J Bone Joint Surg
(Am) 1987; 69A: 1127 ± 1132.
29 Harvey L, King M, Herbert R. Reliability of a tool for measuring
of long ®nger ¯exor muscles. J Hand Ther 1995; 7: 251 ± 254.
30 Neurocontrol Corporation. G rasp-Release Test Kit Manual.
Cleveland, Ohio, USA and Brussels, Belgium.
31 Saxena S, Nikolic S, Popovic D. An EMG-controlled grasping
system for tetraplegics. J Rehab Res Dev 1995; 32: 17 ± 24 .
32 StrohKC,VanDorenC.AnADLtesttoassesshandfunction
with tetraplegic patients. J Hand Ther 1994; 7: 47 ± 48.
33 Kilgore KL et al. An implanted upper- extremity neuroprosthesis:
follow-up of ®ve patients. J Bone Joint Surg (Am) 1997; 79A:
533 ± 541.
34 McDowell CL, Moberg EA, Smith AG. International c onference
on surgical rehabilitation of the upper limb in tetraplegia. J Bone
Joint Sur g (Am) 1979; 4A: 387 ± 390.
35 Lamb D. The current state of management of the upper limb in
tetraplegia. Paraplegia 1992; 30: 65 ± 67.
Hand function of tetraplegics
LA Harvey et al
43
Spinal Cord