1420
ORIGINAL ARTICLE
Break-Technique Handheld Dynamometry: Relation Between
Angular Velocity and Strength Measurements
Stephen P. Burns, MD, David E. Spanier, MD
ABSTRACT. Burns SP, Spanier DE. Break-technique handheld dynamometry: relation between angular velocity and
strength measurements. Arch Phys Med Rehabil 2005;86:
1420-6.
Objectives: To determine whether the muscle strength, as
measured with break-technique handheld dynamometry
(HHD), is dependent on the angular velocity achieved during
testing and to compare reliability at different angular velocities.
Design: Repeated-measures study. Participants underwent
HHD by using make-technique (isometric) and break-technique
(eccentric) dynamometry at 3 prespecified angular velocities. Elbow movement was recorded with an electrogoniometer.
Setting: Inpatient spinal cord injury unit.
Participants: Convenience sample of 20 persons with tetraplegia with weakness of elbow flexors or extensors.
Interventions: Not applicable.
Main Outcome Measures: Elbow angular velocity and
muscle strength recorded during HHD.
Results: With the break technique, angular velocities averaging 15°, 33°, and 55°/s produced 16%, 30%, and 51% greater
strength measurements, respectively, than velocities recorded
by using the make technique (all P⬍.006 for comparisons
between successive techniques). The intraclass correlation coefficient for intrarater reliability was .89 or greater for all
testing techniques.
Conclusions: Greater strength is recorded with faster angular velocities during HHD. Differences in angular velocity may
explain the wide range previously reported for break- versus
make-technique strength measurements. Variation in angular
velocity is a potential source of variability in serial HHD
strength measurements, and for this reason the make technique
may be preferable.
Key Words: Muscle spasticity; Rehabilitation; Reproducibility of results; Spinal cord injuries.
© 2005 by American Congress of Rehabilitation Medicine
and the American Academy of Physical Medicine and
Rehabilitation
ELIABLE AND PRECISE MEASUREMENTS of muscle
strength are essential for clinical care and research. AccuR
rate measurements allow clinicians to monitor strength im-
From the Spinal Cord Injury Service, Veterans Affairs Puget Sound Health Care
System (Burns); the Department of Rehabilitation Medicine, University of Washington (Burns, Spanier); and Harborview Injury Prevention and Research Center (Burns),
Seattle, WA.
Supported by the Department of Veterans Affairs (DVA). The views expressed in
this article are those of the authors and do not necessarily represent the views of the
DVA.
No commercial party having a direct financial interest in the results of the research
supporting this article has or will confer a benefit upon the author(s) or upon any
organization with which the author(s) is/are associated.
Correspondence to Stephen P. Burns, MD, SCI (128), VA Puget Sound Health
Care System, 1660 S Columbian Way, Seattle, WA 98108, e-mail:
spburns@u.washington.edu. Reprints are not available from the authors.
0003-9993/05/8607-9405$30.00/0
doi:10.1016/j.apmr.2004.12.041
Arch Phys Med Rehabil Vol 86, July 2005
provement with exercises or other prescribed treatments or to
monitor disease progression or development of secondary conditions. Manual muscle testing (MMT), the most widely used
method for measuring muscle strength, has many desirable
properties, such as ease of performance and the need for no
testing equipment. However, the 6-point (range, 0 –5) Medical
Research Council scale is ordinal, has only fair reliability when
assessed for individual muscles, and has limited utility in
discriminating between degrees of weakness for grade 4/5
strength.1,2 MMT also requires considerable training to achieve
this modest degree of reliability.3
Instrumented strength measurements produce continuous (ie,
nonordinal) measurements of muscle strength and in general
have better reliability than MMT.4,5 Most methods require
either large instruments, such as isokinetic dynamometers, or
placement of fixed gauges against which the patient exerts
maximal effort. This requirement makes these techniques
somewhat cumbersome in some clinical settings, and the
equipment cannot easily be configured to assess some clinically
relevant muscles. Handheld dynamometry (HHD) overcomes
some of these limitations by having the examiner stabilize the
force gauge, allowing multiple muscles to be tested at the
bedside with a single portable instrument. Prior studies6-8 have
confirmed excellent reliability with HHD under most conditions. Excellent reliability has been shown even with use of
inexperienced examiners.9-12
HHD is performed with 1 of 2 testing techniques: make or
break. The make technique requires the patient to exert a
maximal isometric contraction while the examiner holds the
dynamometer in a fixed position. The break technique, in
contrast, requires the examiner to overpower a maximal effort
by the patient, thereby producing a measurement of eccentric
muscle strength. Both methods produce strength measurements
that have excellent reliability, although the break technique
produces higher values.6-9,13 However, prior studies do not
agree about the magnitude of the difference between the 2
techniques. Some8,13 report that the break technique produced
3% to 5% greater strength measurements, whereas others6,7
report that the break technique produces 30% greater strength
measurements, and up to 70% greater strength measurements
in patients with stroke. In a previous study,9 we noted no
relation between the angular velocity achieved by using the
break technique and the strength recorded but that study had
limited power to detect this relationship. In vitro studies show
increased muscle strength (tension) with increased velocity of
muscle stretching, as shown with classic force-versus-velocity
curves representing muscle contractile properties.14 In contrast,
in vivo studies performed with isokinetic dynamometers show
at most a small increase in strength with eccentric contractions,
and strength does not increase with greater velocity of eccentric
contraction.15
The primary objective of the present study was to determine
the relation between strength measurements and the angular
velocity achieved during break-technique HHD. The hypothesis was that greater angular velocity would be associated with
higher strength measurements. Secondary objectives included
comparing the reliability of HHD performed at different angu-
BREAK-TECHNIQUE HANDHELD DYNAMOMETRY, Burns
1421
lar velocities, and determining whether spasticity would be
associated with proportionally greater break versus make
strength measurements.
METHODS
Participants
Study participants were inpatients on the spinal cord
injury (SCI) ward at VA Puget Sound Health Care System,
where they were undergoing hospitalization for annual medical evaluations, treatment of pressure ulcers, prescription of
new equipment, or additional physical and occupational
therapy. The study protocol received approval from institutional review boards at the University of Washington and
VA Puget Sound Health Care System, and written informed
consent was obtained from all participants. Inclusion criteria
for study participation included the following: history of
SCI (traumatic or nontraumatic etiology), and weakness of
either the elbow flexors or extensors with MMT grades of
3/5 or 4/5. Exclusion criteria included the following: upperlimb pathology that would preclude maximal strength testing (eg, recent upper-limb fracture), medical contraindications to maximal strength testing, or cognitive dysfunction
with inability to reliably follow commands. During the data
collection period, consecutive patients who were admitted to
the ward and met study criteria were offered participation.
One patient who participated had also participated in our
previous investigation of HHD.9
Participants underwent clinical examination by the senior
author (SPB) for muscle strength, sensation, reflexes, and spasticity. Muscle strength was initially assessed with MMT and
performed for the American Spinal Injury Association (ASIA)
key muscles, including elbow flexion and extension, by using
whole number muscle grades, as described in the ASIA Reference Manual.16 Classifications were assigned for motor levels and ASIA Impairment Scale.17 Deep tendon reflexes (DTR)
for the bilateral biceps and triceps brachii, assessed by tendon
reflex hammer, were graded on a 0 to 3⫹ scale (absent,
hypoactive, normal, hyperactive). Spasticity was assessed for
the elbow flexors and extensors with passive movement of the
upper limb and grading of the degree of resistance to movement using the Modified Ashworth Scale (MAS).18 Based on
the clinical examination, we selected a single muscle group
(either elbow flexors or extensors) unilaterally for HHD testing. We selected the muscle with greatest MAS score or, if
muscle tone was normal, the maximal DTR, provided that the
MMT score for the muscle was 3/5 or 4/5.
Strength Testing Protocol
Participants were positioned supine on a hospital bed, with
the posterior aspect of the arm contacting the bed and the
forearm in a vertical, gravity-eliminated position. An electrogoniometer was fabricated by using a CP-2FC rotary potentiometera with ⫾1% linearity. The electrogoniometer was placed
across the elbow joint and secured to the arm and forearm (fig
1), and a TSD-150 preamplified surface electromyography
electrodeb was placed over the muscle group undergoing testing. The electrogoniometer was calibrated with the elbow at
90° of flexion and at maximal extension (0°, or a positive angle
if the participant lacked full elbow extension). A universal
goniometer was used to determine elbow angles, with axis
alignment along the humerus and radius, with the fulcrum
centered over the lateral epicondyle.19 The electrogoniometer
was recalibrated at the start of each testing session. The electromyographic signal was used to assist the identification of
Fig 1. Setup for strength testing of biceps and data acquisition by
the HHD and electrogoniometer. During individual strength measurements, the examiner stabilized the proximal limb with 1 hand
and applied force to the distal limb through the HHD using the
opposite hand. The photo depicts the testing setup on a neurologically intact individual.
periods of HHD and electrogoniometer signals that occurred
during individual strength measurements, but the electromyographic signal was not otherwise analyzed. HHD was performed for either elbow flexion or extension by using a Chatillon CSD-200 dynamometer,c with the distal edge of the HHD
aligned with the radial or ulnar aspect of the radiocarpal joint.
The HHD has a continuous analog signal output that is proportional to the applied force, along with the conventional
digital display showing the peak force recorded during the test.
The examiner attempted to maintain the direction of force
application perpendicular to the forearm. The peak force recorded by the HHD was noted after each test. During HHD
testing, the analog outputs of the HHD, electrogoniometer, and
electromyography electrode were synchronously recorded by
means of an MP100WS interface and AcqKnowledge softwareb with sampling at 500Hz.
The testing protocol is in appendix 1. A single examiner
(DES) performed testing on each participant by using make
technique and 3 target angular velocities for the break technique. The examiner had relatively little experience performing
HHD, having occasionally tested patients during the preceding
3 years. We selected target angular velocities for break testing
so that the medium velocity would represent the velocity typically used by clinicians at our institution and that was used by
novice physical therapists in our previous study.9 The target
angular velocities were the following: high angular velocity
was defined as 90° to 100°/s, medium velocity as 30° to 40°/s,
and low velocity as 10° to 20°/s. During the first testing
session, the examiner tested the participant 8 times, twice by
using the make technique, and twice each at low, medium, and
high angular velocities by using the break technique. After a
5-minute rest, this procedure was repeated. Using scripted
instructions, participants were asked to exert a maximal effort
against the HHD when instructed by the examiner. They were
also told that the technique would vary between successive
trials, but in all cases they were to provide a maximal effort.
Although not depicted in figure 1, the examiner stabilized the
proximal limb with 1 hand while applying force through the
myometer with the other hand during each strength measurement. Consistent verbal encouragement was provided to parArch Phys Med Rehabil Vol 86, July 2005
1422
BREAK-TECHNIQUE HANDHELD DYNAMOMETRY, Burns
ticipants by the examiner during each measurement. The order
in which the 4 techniques were performed was randomized for
successive participants. Completion of all testing required approximately 30 minutes per participant. After testing of the first
participant, the investigators reviewed the electrogoniometer
and force data to confirm that the angular velocities were close
to the target ranges. The examiner did not view any subsequent
angular velocity data until testing was completed on all 20
participants.
Analysis
Electrogoniometric data were first reviewed graphically
along with the continuous HHD output and the electromyographic signal. For each test, we identified periods of peak
force application on the HHD signal tracing, and this was
confirmed by using the electromyographic signal. For breaktechnique trials, we calculated the angular velocity for the
portion of the force application during which the participant’s
muscle was overcome into an eccentric contraction. The joint
angle at which the peak force occurred and the total degrees
through which the limb was moved during the period of maximal muscle contraction was also recorded. Similar calculations were performed for make technique trials. In keeping
with our clinical practice and that of other investigators,20,21 we
used the highest strength measurement achieved from each
testing session with each technique in our primary analyses.
Other investigators have compared use of the maximum, mean,
or median for repeated HHD trials, or have compared the mean
to a single trial, and have shown similar reproducibility with
each method.10,11 We chose to repeat contractions only twice
per technique session because we wanted to minimize the
number of muscle contractions and resulting fatigue. We combined data from elbow flexor and extensor testing for most
analyses because mean strength was similar for the 2 muscle
groups. Strength as a proportion of predicted strength was
calculated for make technique tests from session 1 by using
previously published age- and sex-specific reference values for
the dominant extremity.22
Mean strengths and velocities were compared by using the
Friedman and Wilcoxon signed-rank tests. Break/make (B/M)
ratios were calculated by dividing the peak force obtained with
each break technique (low, medium, high) by the peak force
obtained with the make technique. The correlations between
the 3 B/M ratios and the DTR score for the muscle were
assessed with the Spearman correlation coefficient. The initial
analysis for strength versus angular velocity used the classification of the testing technique (make, low, medium, or high)
for the trial with the greatest strength value, whereas the
subsequent analysis used the angular velocity recorded by the
electrogoniometer for each of the 16 tests. By using the latter
data, Spearman correlation coefficients were calculated separately for each participant to assess for correlations between
strength and angular velocity. Additionally, the slope of the
strength versus angular velocity correlation was determined by
calculating a slope for each participant by using all 16 strength
and angular velocity measurements, normalizing strength data
for the mean of the participant’s 4 make-technique tests. Reliability of the measures was assessed with the intraclass correlation coefficient (ICC) 2-way mixed-effects model. We also
plotted differences versus means for strength measurements, as
described by Bland and Altman,23 to assess for systematic
differences across the range of strength values. All statistical
analyses were performed by using SPSS, version 10.0.5.d A P
value of less than .05 was considered significant.
Arch Phys Med Rehabil Vol 86, July 2005
Table 1: Angular Velocities for Each Testing Technique
Technique
Angular Velocity (°/s)
Make
Low break
Medium break
High break
1⫾5
15⫾0
33⫾15
55⫾26
NOTE. Values are mean ⫾ standard deviation (SD). P⬍.001 for make
vs low, low vs medium, and medium vs high comparisons.
RESULTS
Participants
The 20 participants included 19 men and 1 woman, all with
tetraplegia. The mean age of participants was 57⫾15 years, and
the mean duration of SCI was 19⫾21 years. The ASIA Impairment Scale scores were grades A, B, C, and D for 6, 4, 4,
and 6 participants, respectively. The motor level on the tested
side was C5, C6, and C7 for 11, 5, and 4 participants, respectively. The test muscles were the elbow flexors for 5 participants and the elbow extensors for 15 participants. MMT scores
for the test muscle were 3/5 for 7 participants and 4/5 for 13
participants. DTR scores were 0, 1⫹, and 3⫹ for 10, 6, and 4
participants, respectively, and MAS scores were 0, 1, 1⫹, and
2 for 16, 1, 1, and 2 participants, respectively. Participants with
motor-incomplete tetraplegia (ASIA grades C or D) had greater
DTR scores for the test muscle than did those with motorcomplete tetraplegia (ASIA grades A or B; P⫽.05 for difference).
Angular Velocity and Strength
The mean angular velocities achieved for each testing technique during all trials are in table 1. The angular velocities for
the 4 testing techniques differed significantly from each other
(P⬍.001). The mean velocities for low and medium velocity
break testing were close to the target velocities we had specified before the study, but high velocity testing (mean, 55°/s)
was lower than the 90° to 100°/s we had planned to achieve.
The examiner used a significantly greater (P⫽.008) angular
velocity during the initial high velocity trial (61°/s) than on the
final high velocity trial (51°/s), and there was a similar trend
with medium velocity testing. Over the course of testing the 20
participants, the examiner showed no trend in the angular
velocities used for any of the 4 testing techniques. The examiner moved the participant’s elbow through a mean ⫾ standard
deviation (SD) of 3°⫾2°, 16°⫾10°, 20°⫾14°, and 20°⫾13° for
make, low, medium, and high testing techniques, respectively.
For the 15 participants who underwent testing of elbow extensors, the peak force was recorded at elbow angles of 89°, 99°,
111°, and 114° for make, low, medium, and high testing
techniques, respectively. For the 5 participants who underwent
testing of elbow flexors, the respective elbow angles at peak
force were 83°, 68°, 60°, and 56°. However, during most tests,
force was relatively constant over a range of degrees of elbow
movement, without a pronounced peak force.
Mean strength for each technique and each testing session is
shown in figure 2 and table 2. The strength recorded for the
make technique in session 1 averaged 33%⫾18% of predicted
strength. Strength was significantly greater with increased velocity (make vs low, low vs medium, medium vs high) for all
comparisons in both testing sessions, with the exception of low
versus medium in the second testing session (P⫽.09; see table
2). Strength comparisons are also reported as mean B/M ratios
for the 3 break techniques (table 3). In both sessions, low-
1423
BREAK-TECHNIQUE HANDHELD DYNAMOMETRY, Burns
Table 3: B/M Ratios*
Ratio
Session 1
Session 2
Low B/M
Medium B/M
High B/M
1.16⫾0.16
1.30⫾0.26
1.51⫾0.46
1.16⫾0.21
1.33⫾0.33
1.43⫾0.36
NOTE. Values are mean ⫾ SD.
*P⫽.005 and P⫽.009 for low B/M vs medium B/M, and P⫽.003 and
P⫽.048 for medium B/M vs high B/M comparisons for sessions 1 and
2, respectively.
Fig 2. Mean strength for the 4 testing techniques, session 1. Error
bars represent the standard error of the mean. The differences
between strength with each of the 4 techniques were statistically
significant. See table 2.
velocity testing produced significantly lower B/M ratios than
medium-velocity testing, and medium-velocity testing resulted
in significantly lower ratios than high velocity testing. Comparing a technique between sessions 1 and 2, we found that
strength was lower in session 2 (range, 3.9% to 7.8% lower)
with significant decrements for medium (P⫽.04) and high
techniques (P⫽.007).
We had planned to analyze B/M ratios versus MAS scores.
Because only 4 participants had increased muscle tone (MAS
score, ⱖ1) in the test muscle, we instead assessed the relation
between B/M ratio and DTR scores for test muscles. The B/M
ratios were correlated (␳ range, .533–.788; all P⬍.016) with
DTR scores for all B/M ratios in session 1 and for medium and
high B/M ratios in session 2. Figure 3 shows the relation
between B/M ratio and DTR for medium velocity testing from
session 1. When the same analysis was performed by using
MAS scores, significant correlations were found for low-velocity B/M ratios in both sessions and medium-velocity B/M
ratios in the second session, and there was a trend (P⫽.051)
toward a correlation with high-velocity B/M ratio in the second
session. However, given the low number of participants with
abnormal MAS scores, these findings should be interpreted
with caution.
Representative strength and angular velocity data for a single
participant using all 16 tests recorded in the 2 sessions are in
figure 4. The mean correlation coefficient (Spearman ␳) for the
20 participants was .60, and 15 of 20 participants showed
statistically significant positive correlations. The mean slope
for the correlation between strength and angular velocity corresponded to an 8%⫾6% increase in strength for every 10°
increase in angular velocity.
Reliability of Testing Techniques
The ICC for each testing technique from session 1 to 2 is
shown in table 4, along with 95% confidence intervals (CIs) for
the coefficients. The ICCs for all testing techniques were .89 or
greater, with the lowest value for the make technique, but CIs
for all techniques overlapped. Bland-Altman plots of mean
strength versus difference in strength between sessions 1 and 2
showed no consistent relationship between the 2 variables for
any testing technique.
DISCUSSION
The findings confirmed our hypothesis that greater angular velocity during break technique HHD testing is associated with higher strength measurements. Break-technique
testing, performed at angular velocities averaging 15°, 33°,
and 55°/s, produced 16%, 30%, and 51% greater strength,
respectively, than isometric testing by using the make technique during the first session. Large differences were seen
between high-velocity break versus make testing for some
participants; for 4 participants, this difference exceeded
100%. Our prior study9 did not show an association between
angular velocity and strength measurements. However, the 2
examiners in that study did not intentionally perform break
testing over a wide range of angular velocities, and the
velocities used by the 2 examiners for testing were similar.
Table 2: Mean Strength for Each Testing Technique
Difference
Between
Sessions
Strength (kg)
Technique
Make
Low break
Medium break
High break
Session 1
Session 2
%
P
6.6⫾3.4
7.6⫾3.8
8.2⫾3.9
9.0⫾3.5
6.3⫾4.1
7.3⫾4.3
7.8⫾4.0
8.3⫾4.0
4.5
3.9
4.9
7.8
.09
.51
.04
.007
P
Comparison
Session 1
Session 2
Make vs low break
Low break vs medium break
Medium break vs high break
.003
.007
.006
.004
.09
.045
NOTE. Values are mean ⫾ SD or otherwise indicated.
Fig 3. B/M ratio for medium velocity testing (session 1) versus DTR
scores.
Arch Phys Med Rehabil Vol 86, July 2005
1424
BREAK-TECHNIQUE HANDHELD DYNAMOMETRY, Burns
Fig 4. Angular velocity versus strength for all measurements on a
single participant. ␳ⴝ80; the trend line indicates a 7% greater
strength measurement for every 10° increase in angular velocity.
Additionally, electrogoniometric data were obtained from
only a subset of participants.
This velocity dependence of break-technique strength
measurements may be an explanation for conflicting results
from prior HHD studies that compared strength by using
make and break techniques. Van der Ploeg and Oosterhuis13
reported B/M ratios of 1.03 for normative subjects and up to
1.11 for subjects with organic weakness. They concluded
that a ratio of greater than 1.2 supported the diagnosis of
functional weakness (eg, conversion disorder). Stratford and
Balsor8 reported B/M ratios of 1.05 to 1.06 in neurologically
intact subjects. These ratios were similar to those they
obtained using an isokinetic dynamometer set to perform
eccentric testing at an angular velocity of 2°/s. In contrast,
Bohannon6,7 reported B/M ratios of 1.3 in normative subjects and on the nonparetic side in patients with stroke. The
B/M ratios obtained in the current study with medium velocity testing and in our prior study9 agree most closely with
those reported by Bohannon. Other investigators did not
measure or describe the angular velocity that occurred with
break testing, and the description of break testing has varied
considerably between studies,6-8,11,13,21 so variability in
technique may account for the wide range previously reported for B/M ratios. Differences in break-technique angular velocity are a potential source of variability over serial
measurement. Although previous studies have shown high
interrater reliability for make and break techniques, study
factors (small sample sizes, short time interval between
testing sessions, concurrent training of examiners) may have
limited the variability in technique. Use of examiners who
trained separately could allow differences in the break technique to affect strength recordings. It may be more practical
to have examiners perform make technique testing because
reproduction of an isometric test should be simpler than
attempting to train examiners to produce a specific angular
velocity for break testing.
Upper motoneuron (UMN) weakness is a second factor that
may have contributed to the variability in B/M ratios in prior
studies because some of these have included subjects with UMN
disorders.7,13 We found higher B/M ratios in participants with
preserved DTRs in the test muscle. The presence of a DTR in a
muscle at or below the motor level is likely to be associated with
some degree of UMN weakness because their presence was more
common in patients with motor-incomplete (ASIA grades C, D)
Arch Phys Med Rehabil Vol 86, July 2005
injuries. The recovery of strength seen in the first few months
postinjury is in part because of collateral sprouting from intact
motor units innervating the muscle.24 For this reason, lower motoneuron weakness predominates in paretic muscles at or below
the motor level in persons with motor-complete tetraplegia.
Bohannon7 previously noted elevated B/M ratios in paretic
muscles in persons with UMN weakness. In that study of 22
patients with stroke, B/M ratios were 1.7 on the hemiparetic side
compared with 1.3 on the less involved side, and the side-to-side
difference in B/M ratios correlated positively with MAS scores.
The elevated ratio is presumably due to augmentation of breaktechnique strength measurements because break-technique testing
involves the same type of muscle stretching for assessment of
muscle tone with the MAS. In our previous study,9 we did not find
a correlation between MAS and B/M ratios, whereas in the present
study significant correlations were seen for some tests. The small
number of participants with elevated MAS grades was a limitation
in both studies. This relation could be investigated in an SCI
population with greater spasticity or in a different population. If
B/M ratios are associated with muscle spasticity, then the ratios
could be altered if a patient’s spasticity changes, either with
worsening as spasticity develops after onset of the neurologic
deficit or with improvement through use of antispasticity medications or neurolytic injection procedures. This hypothesis could be
confirmed by studying patients before and after initiation of antispasticity medications, such as intrathecal baclofen. As Bohannon7 has noted, it would be useful in future studies to measure
function, to determine which strength measurement technique
produces the highest correlation with function.
Our present findings are in agreement with prior observations
on differences between isometric and eccentric strength derived
from in vitro studies but are in conflict with most from in vivo
studies.14 With maximal electric stimulation of animal whole
muscle preparations, muscle tension during eccentric contraction
increases with the velocity of muscle stretch and can exceed
isometric tension by up to 100%.14 When maximal voluntary
contractions are performed in human experiments using isokinetic
dynamometers, eccentric strength only slightly exceeds isometric
strength.15 With the application of electric stimulation on intact
human muscles, the force-velocity relationship becomes more
similar to that of in vitro muscle preparations, with greater forces
recorded for faster eccentric contrations.25 These observations
have been interpreted as reflecting a neural force-regulating mechanism that prevents development of potentially injurious tension
with eccentric contractions, and evidence supports the presence of
separate neural control mechanisms for eccentric contractions.26
Control of force generation may be altered in patients with neurologic disorders, which may in part explain the B/M ratios seen
in this study and others.7
We also considered whether the greater strength at faster angular velocities could reflect the force required by the examiner to
accelerate the mass of the distal limb to the desired angular
velocity. We believe that, for several reasons, this is unlikely to
account for much of the strength difference with increased angular
velocity. First, the peak forces on the continuous myometer output
occurred after the limb had been accelerated to a relatively con-
Table 4: ICC for Testing Techniques, Session 1 Versus Session 2
Technique
ICC
95% CI
Make
Low break
Medium break
High break
.89
.93
.97
.97
.74–.96
.83–.97
.93–.99
.92–.99
1425
BREAK-TECHNIQUE HANDHELD DYNAMOMETRY, Burns
stant angular velocity and not during the periods corresponding to
maximal limb acceleration, and peak forces typically occurred
during a long period of relatively stable force. Also, the force
required to accelerate the forearm and hand to the target angular
velocities would be relatively small compared with the recorded
forces for most participants because limb acceleration was relatively slow and the force was applied at a distal point on the limb.
However, some force is required for limb acceleration, and for
participants with severe weakness this force could contribute
significantly to the increased B/M ratio at higher velocities.
Review of the elbow movement recorded by the electrogoniometer during strength testing revealed technical factors
that could lead to strength measurement errors. When performing break testing on the elbow extensors, the examiner
infrequently moved the elbow into excessive flexion
(⬎140°), producing a considerably greater strength measurement at the end range of flexion. The peak force recorded in those cases likely occurred as soft tissues of the
arm and forearm contacted each other. The examiner tended
to produce a greater than intended angular velocity on the
first trial of high-velocity break testing if the muscle was
relatively weak. Also, the examiner had difficulty achieving
the intended high velocity during break testing if the muscle
was strong. The examiner rarely produced excessive eccentric (break) movement when attempting to perform make
technique. This occurred much less frequently than in our
previous study, during which 25% of make trials resulted in
excessive break-type movement, which was defined as more
than 5° of eccentric movement.9 It is possible that the
examiner in the present study was more attentive to how the
make technique testing was performed because the intention
was to produce movement that was distinct from the low
(10°–15°/s) velocity break technique.
A number of factors should be considered when choosing
a testing technique. The present study showed excellent
intrarater reliability for the make technique and the 3 break
techniques, and prior studies have also shown high reliability for the 2 techniques.6-9,13 The small decline in strength
between testing sessions for some techniques was likely
secondary to muscle fatigue from repeated contractions.
However, as discussed earlier, if examiners were to perform
the break technique at velocities that differ, it could lead to
differences in the strength recorded. This source of variability has likely been minimized by study designs, but, in a
clinical setting or in a multisite study, it could potentially
affect reliability. If muscle spasticity is anticipated to fluctuate over time, then the make technique might be preferable
because changes in tone could change break-technique
strength to a greater degree than the make technique. In
cases in which the examiner is relatively weak or the patient
is relatively strong, it may be useful for the examiner to
confirm that examiner strength is not limiting the maximal
force recorded. This could be accomplished by attempting
an additional trial using the break technique, with successful
overpowering of the test muscle indicative of adequate
examiner strength, and inability to overpower the muscle
indicative that examiner strength may have limited the recorded strength. However, it is possible that an examiner
could have adequate strength to resist a maximal effort and
could thus obtain a measurement of maximal effort but
might not have enough strength to overpower the muscle, in
which case make testing would be preferable.
There are some additional limitations to our findings. This
study was performed in a population with tetraplegia, and
the findings may not generalize to other patient populations.
The participants had a relatively low level of muscle tone,
and similar findings regarding B/M ratios may not be present
in populations with pure lower motoneuron disorders or
greater spasticity. Our planned analysis of strength versus
muscle tone involved use of MAS scores, not DTR scores,
so these findings should be interpreted with caution. The
muscles tested were relatively weak, averaging 33% of
predicted strength. Lower strength was recorded during the
second session for medium- and high-velocity testing, which
likely was caused by muscle fatigue induced by multiple
maximal contractions. For this reason, we consider the results from session 1 to be more representative of what would
be obtained with conventional HHD testing, and we did not
calculate additional measures of repeatability for the 2 testing sessions. Finally, testing was performed by a single
examiner, who was given feedback on the velocities he
achieved for testing on the first participant. Data on testing
velocities are not available to examiners when performed in
a clinical setting with standard equipment.
CONCLUSIONS
Break-technique HHD produces higher strength measurements when it is performed at greater angular velocities in
persons with tetraplegia. The velocity dependence of the
strength measurements may explain some of the variability
previously reported when comparing strength measurements
with make and break techniques. Variability in angular
velocity during break testing is a potential source of error
when longitudinal strength measurements are performed.
Patients with signs of UMN weakness may show relatively
greater strength with the break versus the make technique,
and fluctuations in muscle spasticity could potentially alter
break-technique measurements more so than make-technique measurements. To minimize variability in strength
measurements related to differing angular velocities and
fluctuations in muscle tone, it may be preferable to perform
HHD by using the make technique.
APPENDIX 1: TESTING PROTOCOL
Session
Trial
1
1
Test Technique
Make
Low velocity break
Medium velocity break
High velocity break
Make
Low velocity break
Medium velocity break
High velocity break
2
5-minute rest
2
1
2
Make
Low velocity break
Medium velocity break
High velocity break
Make
Low velocity break
Medium velocity break
High velocity break
NOTE. For each participant, the techniques (make, low velocity,
medium velocity, high velocity) were performed in random order.
Arch Phys Med Rehabil Vol 86, July 2005
1426
BREAK-TECHNIQUE HANDHELD DYNAMOMETRY, Burns
References
1. Frese E, Brown M, Norton BJ. Clinical reliability of manual
muscle testing. Middle trapezius and gluteus medius muscles.
Phys Ther 1987;67:1072-6.
2. Dvir Z. Grade 4 in manual muscle testing: the problem with
submaximal strength assessment. Clin Rehabil 1997;11:36-41.
3. Mendell JR, Florence J. Manual muscle testing. Muscle Nerve
1990;13:S16-20.
4. Delitto A. Isokinetic dynamometry. Muscle Nerve 1990;
13(Suppl):S53-7.
5. Larsson B, Karlsson S, Eriksson M, Gerdle B. Test-retest reliability of EMG and peak torque during repetitive maximal concentric
knee extensions. J Electromyogr Kinesiol 2003;13:281-7.
6. Bohannon RW. Make tests and break tests of elbow flexor muscle
strength. Phys Ther 1988;68:193-4.
7. Bohannon RW. Make versus break tests for measuring elbow
flexor muscle force with a hand-held dynamometer in patient with
stroke. Physiother Can 1990;42:247-51.
8. Stratford PW, Balsor BE. A comparison of make and break tests
using a hand-held dynamometer and the Kin-Com. J Orthop
Sports Phys Ther 1994;19:28-32.
9. Burns SP, Breuninger A, Kaplan C, Marin H. Hand-held dynamometry in persons with tetraplegia: comparison of make- versus
break-testing techniques. Am J Phys Med Rehabil 2005;84:22-9.
10. Goonetilleke A, Modarres-Sadeghi H, Guiloff RJ. Accuracy, reproducibility, and variability of hand-held dynamometry in motor
neuron disease. J Neurol Neurosurg Psychiatry 1994;57:326-32.
11. Horvat M, Croce R, Roswal G. Intratester reliability of the Nicholas Manual Muscle Tester on individuals with intellectual disabilities by a tester having minimal experience. Arch Phys Med
Rehabil 1994;75:808-11.
12. Ottenbacher KJ, Branch LG, Ray L, Gonzales VA, Peek MK,
Hinman MR. The reliability of upper- and lower-extremity
strength testing in a community survey of older adults. Arch Phys
Med Rehabil 2002;83:1423-7.
13. van der Ploeg RJ, Oosterhuis HJ. The “make/break test” as a
diagnostic tool in functional weakness. J Neurol Neurosurg Psychiatry 1991;54:248-51.
14. Lombardi V, Piazzesi G. The contractile response during steady
state lengthening of stimulated frog muscle fibres. J Physiol
(Lond) 1990;431:141-71.
Arch Phys Med Rehabil Vol 86, July 2005
15. Seger JY, Thorstensson A. Electrically evoked eccentric and concentric torque-velocity relationships in human knee extensor muscles. Acta Physiol Scand 2000;169:63-9.
16. American Spinal Injury Association. Reference manual for the
standards for neurologic and functional classification of spinal
cord injury. Chicago: ASIA; 2003.
17. Marino RJ, Barros T, Biering-Sorenson F, et al. International
standards for neurological classifications of spinal cord injury.
J Spinal Cord Med 2002;25:S50-6.
18. Bohannon RW, Smith MB. Interrater reliability of a modified
Ashworth scale of muscle spasticity. Phys Ther 1987;67:206-7.
19. Norkin CC, White DJ. The elbow and forearm. In: Norkin CC,
White DJ, editors. Measurement of joint motion: a guide to
goniometry. 2nd ed. Philadelphia: FA Davis; 1995. p 67-77.
20. Merlini L, Mazzone ES, Solari A, Morandi L. Reliability of
hand-held dynamometry in spinal muscular atrophy. Muscle
Nerve 2002;26:64-70.
21. Phillips BA, Lo SK, Mastaglia FL. Muscle force measured using
“break” testing with a hand-held myometer in normal subjects
aged 20 to 69 years. Arch Phys Med Rehabil 2000;81:653-61.
22. Bohannon RW. Reference values for extremity muscle strength
obtained by hand-held dynamometry from adults aged 20 to 79
years. Arch Phys Med Rehabil 1997;78:26-32.
23. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;
1:307-10.
24. Marino RJ, Herbison GJ, Ditunno JF Jr. Peripheral sprouting as a
mechanism for recovery in the zone of injury in acute quadriplegia: a single-fiber EMG study. Muscle Nerve 1994;17:1466-8.
25. Dudley GA, Harris RT, Duvoisin MR, Hather BM, Buchanan P.
Effect of voluntary vs. artificial activation on the relationship of
muscle torque to speed. J Appl Physiol 1990;69:2215-21.
26. Enoka RM. Eccentric contractions require unique activation strategies by the nervous system. J Appl Physiol 1996;81:2339-46.
Suppliers
a. Midori America Corp, 2555 E Chapman Ave, Ste 400, Fullerton,
CA 92831.
b. BIOPAC Systems Inc, 42 Aero Camino, Goleta, CA 93117.
c. AMETEK, 37 N Valley Rd, Bldg 4, PO Box 1764, Paoli, PA
19301.
d. SPSS Inc, 233 S Wacker Dr, 11th Fl, Chicago, IL 60606.