Assessment of Quadriceps Muscle Performance

by Hand-Held, isometric, and Isokinetic

Dynamometry in Patients With Knee
Dysfunction
Mark F. Reinking, MS, PT, SCS, A TC '
Kelly Bockrath-Pugliese, MS, PT, ATC *
Teddy Worrell, EdD, PT, SCS, A T C ~
Randall I. Kegerreis, MS, PT, SCS, ATC4
Downloaded from www.jospt.org at on May 9, 2025. For personal use only. No other uses without permission.

Kristine Miller-Sayers, MS, PT, A TCS

lack Farr, MD6
Copyright © 1996 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.

A
ssessment of muscle One component of patient evaluation is muscle performance assessment. The purpose of
performance is one this study was: 1) to determine the difference and correlation between hand-held, isometric, and
component of evalua- isokinetic dynamometry test results in patients with knee dysfunction and 2) to determine the effect
tion in patients with of pain during such testing. Bilateral quadriceps strength in 23 subjects with unilateral knee
orthopaedic and neuro- dysfunction was tested using concentric and eccentric isokinetic dynamometry at 60"/sec, isometric
logic dysfi~nction(18). A need exists dynamometry, and hand-held dynamometry, both at 60" of knee flexion. Pain ratings were
for quantitative, objective measures of obtained before, during, and after each test. Statistical analysis revealed a significant difference
muscle performance that provide between involved and uninvolved limbs for eccentric isokinetic dynamometry (p = 0.002) and
documentation of progress and inter- hand-held dynamometry fp = 0.005); no difference was found between limbs for the concentric
Journal of Orthopaedic & Sports Physical Therapy®

vention efficacy. Manual muscle test- isokinetic and isometric dynamometry fp > 0.05). Mean percent deficits in quadriceps strength
ing is a widely used method for evalu- ranged from 1 I to I8%, with no significant difference found between testing modes. Pearson
ating muscle performance. However, product moment correlations ranged from 0.34 to 0.76 when comparing testing modes. No
research has shown that manual m u s significant difference existed in pain scores before, during, and after each mode of testing. It was
cle testing is subjective, especially at concluded that large variation existed between different testing modes, which results in different
the higher muscle test grades, and conclusions regarding the strength of the quadriceps in patients with knee dysfunction.
may not detect muscle performance Key Words: muscle strength, knee, methods
deficits (5,6,24). Other commonly
used clinical methods of instru- ' Assistant Professor, Clarke College, Program of Physical Therapy, 1550 Clarke Drive, Dubuque, IA 52001-
3198. At the time of this study, Mr. Reinking was a physical therapist, St. Francis Hospital and Health
mented muscle testing include hand- Centers, Indianapolis, IN
held, isokinetic, and isometric dyna- * Physical Therapist, St. Francis Hospital and Health Centers, Indianapolis, IN
mometry. All three methods are ' Assistant Professor, Director of Research, Kranned Graduate School of Physical Therapy, University of
reported to be reliable measures of Indianapolis, Indianapolis, IN
Manager and Physical Therapist, St. Francis Hospital and Heahh Centers, Indianapolis, IN
muscle performance (3-5,9,10,19,20). Physical Therapist, St. Francis Hospital and Health Centers, Indianapolis, IN
Clinically, hand-held dynamome- Orthopaedic Surgeon, Specialty Centers for Odhopedic and Rehabilitative Excellence, Indianapolis, IN
try is easy to use, low cost, and re-
quires minimal training. Conversely,
computerized isokinetic and isomet- netic measurements may be evaluat- ric and isokinetic dynamometry for
ric dynamometry is costly, requires ing similar characteristics of muscle knee extension range from 0.46 to
extensive training, and occupies large function (1,10,11,14). The coeffi- 0.86 in previous studies (1,7,8,10,12).
clinical space. Several studies have cients of determination (r2) describ The relationship between hand-held
indicated that isometric and isoki- ing the relationship between isomet- dynamometry and both isokinetic

Volume 24 Number 3 Seotember 1996 IOSPT

 Diagnosis N % 0 No pain
the edge of a treatment table and
1 Dull vague ache positioned in 60" of knee flexion u s
Anterior knee pain syndrome 8 35
ACL reconstruction 7 30
2 Slight persistent pain ing a standard goniometer. The
3 More than slight pain hand-held dynamometer was posi-
Patella realignment (tibia1tubercle) 5 22
4 Painful tioned two finger widths above the
Arthroscopic meniscectomy 2 9
5 Very painful
Patella tendinitis 1 4 lateral malleolus on the anterior
6 Unbearably painful
ACL = Anterior cruciate ligament. tibia. Subjects were asked to stabilize
TABLE 2. Talag pain scale used to assess pain. their pelvis by holding onto the edge
TABLE 1. Subjects' diagnoses. (Adapted from Talag (21), reproduced with of the treatment table. A make test
permission from the American Alliance for Health,
was used, in which the tester matches
Physical Education, Recreation and Dance, Reston,
and isometric dynamometry has also VA 220911. the muscle force generated by the
been reported to be correlated with subject, as contrasted with a break
coefficients of determination ranging test, in which the tester attempts to
from 0.32 to 0.72 (5,15). No studies the St. Francis Hospital and Health exceed the force generated by the
were located that compared hand- Centers Research Committee. S u b subject (23).
held, isokinetic, and isometric dyna- jects were excluded from the study if Four warm-up contractions were
mometry for a patient population. If any of the following criteria were performed, with subjects instructed
these testing modes are highly corre- met: I) neurological or neuromuscu- to gradually increase their knee ex-
lated and reveal similar findings re- lar pathology, 2) increased knee pain tension force over 3 seconds. Subjects
Downloaded from www.jospt.org at on May 9, 2025. For personal use only. No other uses without permission.

garding muscle function, then clini- with manual muscle testing, or were instructed to give approximately
cians may choose the less expensive 3) anterior cruciate ligament recon- 50% effort in the first three warm-ups
and timeconsuming devices for eval- struction less than 5 months postsur- and a maximal contraction on the
uating muscle performance. J3er-Y. fourth warm-up. Four maximal trials
Another issue that has been were then performed, with the peak
Copyright © 1996 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.

raised regarding muscle performance Testing Procedures force of the fourth contraction re-
is the possible confounding influence corded.
of pain during testing. Lysholm (13) Testing was performed during a
has reported that knee pain reduces scheduled physical therapy appoint- Isometric Dynamometer Test
quadriceps isokinetic torque values. ment. All testing was completed on
Deones et a1 (5) compared isokinetic the same day, with the order of test- Isometric knee extension peak
dynamometry and hand-held dyna- ing device randomly determined. The force was measured using the Kin-
mometry and concluded "we believe uninvolved limb was tested first, fol- Com 500H (Chattanooga Corp.,
that pain didn't influence our re- lowed by testing of the involved limb .
Chattanooga, TN) Bilateral testing
Journal of Orthopaedic & Sports Physical Therapy®

sults," but quantification of pain dur- for all three modes. Subjects were was performed with the uninvolved
ing testing was not reported. asked to rate their pain using a 0-6 limb tested first. Subjects were tested
scale (Table 2) (21) immediately be- at 60" of knee flexion as measured
METHODS fore, during, and immediately follow- using the dynamometer goniometer.
ing each test. For the warm-up, s u b Testing was then performed with the
Subjects jects rode a stationary bicycle for 5 subjects seated, using stabilization
minutes prior to the first test. A straps at the pelvis and over the ante-
Twenty-three subjects (17 males, timed Eiminute rest was given be- rior thigh. The dynamometer axis
six females) who were seen at St. tween testing modes. Subjects were was aligned with the axis of the knee,
Francis Outpatient Physical Therapy, informed that they could end a test- identified as a point on the lateral
Indianapolis, IN, participated in this ing session at any time because of femoral condyle 2.5 cm superior to
study. Subjects had unilateral knee knee pain or fatigue. the fibular head. Gravity correction
pathology, including both surgical was not used.
and nonsurgical cases (Table 1) with Hand-Held Dynamometer Test Four warm-up contractions were
a prescription for rehabilitation and performed, with subjects instructed
knee extensor muscle performance The hand-held dynamometer to gradually increase their knee ex-
testing. Subjects ranged in age from (Microfet, Draper, UT) was one in- tension force over each h e c o n d is+
15 to 54years (2 = 27.17 2 12.41 strument used to test bilateral knee metric bout. Subjects were instructed
years). All subjects were advised of extensor muscle performance. All to give approximately 50% effort in
the purpose and risks of the study, subjects were tested by the same the first three warm-ups and a maxi-
and then each subject completed an tester, and the uninvolved limb was mal contraction on the fourth warm-
informed consent form approved by .tested first. Subjects were seated at up. Three maximal trials were then

JOSPT Volume 24 Number 3 September 1996

 RESEARCH STUDY

performed, with the maximal peak ing) were analyzed using a repeated
force recorded. measures analysis of variance.

lsokinetic Dynamometer Test

Concentric and eccentric knee Reliability Study
extension peak force was measured
using the Kin-Com. Bilateral testing Intraclass correlation coefficients
was performed with the uninvolved TABLE 3. Reliability data (N = 10). ranged from 0.76 to 0.92, and stan-
limb tested first. Subjects were tested dard errors of measurement ranged
at 60°/sec isokinetic speed through from 4.3 to 76.71 N for all testing
70" of knee motion (10-80" of knee modes (Table 3). A significant differ-
reliability on the Kin-Com was deter- ence existed between testing trials for
flexion) as measured using the dyna-
mined using 10 normal subjects in isometric testing (p = 0.01).
mometer axis. Testing was performed
two sessions 1 week apart. Testing
with the subjects seated, using stabili-
procedure was as described in the Comparative Study
zation straps at the pelvis and over
isometric and isokinetic dynamome-
the anterior thigh. The dynamometer All subjects completed the study;
try testing sections.
axis was aligned with the axis of the no subjects met any of the exclusion-
knee, identified as a point on the
Downloaded from www.jospt.org at on May 9, 2025. For personal use only. No other uses without permission.

Statistical Analyses ary criteria. Dependent t tests re-

lateral femoral condyle 2.5 cm supe- vealed a significant difference be-
rior to the fibular head. Gravity cor- Intraclass correlation coefficients tween the involved and uninvolved
rection was not used. (ICC 2,l) (17) and standard error of limbs for eccentric isokinetic dyna-
Four concentric and eccentric measurement (SEM) (2) were used mometry (df = 45, t = 18.88, p <
Copyright © 1996 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.

warm-up contractions were per- to determine reliability for all testing .05) and hand-held dynamometry
formed, with subjects instructed to modes prior to initiation of this (df = 45, t = 14.96, p < .O5). No sig-
give approximately 50% effort in the study. Dependent t tests were used to nificant difference existed between
first three warm-ups and a maximal compare involved with uninvolved the involved and uninvolved limbs
contraction on the fourth warm-up. for concentric isokinetic dynamome-
limbs for each testing mode. Bonfer-
Maximal concentric and eccentric try and isometric dynamornetry (p >
roni correction (.05/4 = .0125) was
trials were then performed until 0.05). Average percent deficit be-
performed to adjust for multiple t
three reproducible force curves were tween the involved and uninvolved
tests. Percent deficits (involved -
obtained. Maximal concentric and limbs ranged from 11 to 18% for the
Journal of Orthopaedic & Sports Physical Therapy®

uninvolved/uninvolved X 100) were testing modes (Table 4). No signifi-

eccentric peak forces were recorded. calculated for all subjects and were
No visual feedback was provided to cant difference existed between per-
then compared across all testing cent deficits obtained from the differ-
the subject during the test. Standard- modes using a repeated measures
ized verbal instructions were given ent testing modes (p > .05). Testing
analysis of variance. Pearson product modes showed a large range of values
using the word "push" to begin the moment correlations were used to for patients, as indicated by the large
concentric contraction and the word determine the relationship between coefficient of variation values (Table
"resist" to begin the eccentric con- deficits for each testing mode. In ad- 4). Comparing the percent deficit
traction. dition, coefficient of variation was across all four testing conditions for
determined (SD/X X 100%). Pain each patient, great variation existed
Testing Reliability data (before, during, and after test- in the difference between limbs, de-
Reliability for hand-held dyna-
mometry was performed using 10
normal subjects prior to initiation of X SD Range
the study. Testing procedure was as
Dependent Variable
(%I
cv
described in the hand-held dyna- Hand-held dynamometer -1 1 18 -11 to t 1 6 -165
mometry testing section. Each subject Isometric -1 1 25 . -7 to t 4 1 -219
was tested twice within 2 hours with lsokinetic concentric -12 3 -75 to t 4 3 -260
lsokinetic eccentric -18 28 -73 to t 4 6 -155
the tester blinded to test results. Is*
kinetic and isometric dynamometry
TABLE 4. Percent deficit as determined by each testing device (p > 0.05).
 pending on the mode of testing. In metric dynamometry) to 0.76 (iso- the measurement error and improve
only seven of the 23 patients was metric to concentric isokinetic dyna- the intraclass correlation coefficients.
there agreement in which extremity mometry). The purpose of this study was to
was weaker or stronger. The sign (-) Analysis of the pain data showed determine the differences between
indicates that the injured quadriceps no significant difference between the various modes of muscle perfor-
produced a lower force, while the involved and uninvolved limbs be- mance assessment in a group of sub-
sign (+) indicates that the injured fore, during, and after each test (Ta- jects with knee dysfunction. All sub-
quadriceps produced a greater force ble 7). Also, no significant difference jects were physician-referred to
(Table 5). As presented in Table 6, was detected between pain ratings for physical therapy for rehabilitation
the Pearson product correlations each mode of testing. and muscle performance testing. Test
ranged from 0.34 (hand-held to iso- results showed that eccentric isoki-
DISCUSSION netic dynamometry and hand-held
dynamometry revealed a significant
HHD IS0 ' CON' ECC Regarding the reliability study, difference between the involved and
1. -27.9 4-5.7 t3.5 t7.6 intraclass correlation coefficients uninvolved quadriceps peak force. In
2. -13.4 t10.8 t7.1 -1.5 were generally acceptable with the contrast, involved and uninvolved
3. t6.2 -11.2 -8.6 -16.8 exception of eccentric isokinetic, concentric isokinetic force values and
4. -18.4 t10.6 -3.2 -11.1
5. -49.0 -69.9 -74.6 -73.5 which was .76. Tredinnick and Dun- isometric force values were not signif-
6. t9.8 -22.7 -28.6 -1 1.9 can (22) examined the reliability of icantly different for quadriceps peak
Downloaded from www.jospt.org at on May 9, 2025. For personal use only. No other uses without permission.

7. -27.2 -23.0 t18.4 -18.2 concentric and eccentric testing on force. All of the testing modes
8. -4.6 -2.5 -10.0 t3.8 the Kin-Com and reported an ICC of showed relatively similar average per-
9. -6.1 -7.4 -2.4 t12.3 3 9 for concentric torque and .47 for cent deficits between limbs for the
10. -25.8 -49.4 -54.4 -70.3
11. -46.1 -57.1 -73.8 -55.5 eccentric torque. Deones et al (5) group of patients tested, but a large
reported an ICC of .93 and a stan- amount of individual variation ex-
Copyright © 1996 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.

12. -1.2 -20.9 -5.0 -33.0

13. -8.6 -22.8 -19.8 t46.5 dard error of measurement of 15.6 N isted in percent deficit between test-
14. -42.6 t10.1 -18.0 -36.1 for knee extensor testing using the ing modes, as demonstrated in Table
15. t15.5 t5.8 t13.8 -25.5 hand-held dynamometer at the 60" 5. This illustrates the limited value of
16. -8.0 -17.9 -18.7 -22.5
17. 4-10.7 -15.6 t42.8 t13.0 position. In this study, the standard the mean without information con-
18. t6.6 -6.1 -31.2 -41.1 errors of measurement relative to the cerning the standard deviation, coef-
19. -7.5 t18.8 t18.6 -7.8 mean values of the testing mode ficient of variation, and range.
20. t7.3 -4.3 -2.6 - .9 ranged from 1.30 to 19.40%, which, Physical therapists are asked by
21. -3.2 -17.6 -47.4 -38.2 in our opinion, indicates relatively referring physicians to evaluate a pa-
22. -3.8 -14.6 -8.2 -23.1
small measurement error (Table 3). tient's readiness to return to work,
Journal of Orthopaedic & Sports Physical Therapy®

23. -17.9 t40.8 t34.0 -7.9

For isometric dynamometry, a signifi- sports, or activities of daily living. His-
HHD = Hand-held dvnamometer. cant increase ( p = 0.01) existed be- torically, measures of muscle perfor-
/SO = Isometric.
CON = lsokinetic concentric.
tween the first and second test, indi- mance have been a major part of
ECC = lsokinetic eccentric. cating a learning effect occurred. that determination. Depending on
TABLE 5. Percent deficit (injured - n o n i n j u r d Therefore, during isometric and ec- equipment and time available, differ-
noninjured x 100)as determined by each testing centric isokinetic testing, more prac- ent measures of muscle performance
mode. Negative sign indicates that the injured tice trials may be needed to decrease are employed. An underlying assump
extremity's quadriceps force was less than the
noninjured extremity's quadriceps force. Positive
sign indicates that the injured extremity's quadriceps g. .
force was greater than the noninjured extremity's 'fxtreki/~ondhion
quadriceps force. Testing Mode
NI/B NI/A
Hand-held dynamometer 0 1 0 2 0 1
Isometric 0 1 1 2 0 1
lsokinetic concentric 0 0 0 1 0 1
lsokinetic eccentric 0 1 1 2 2 1
NIIB = Noninjuredl~~.~ore testing.
HHD = Hand-held d\mrnometrr. I/B = lnjuredheiore testing.
/SO = Isometric. Nl/D = Noninjured/during testing.
CON = lsokinetic concentric. VD = lnjuredlduring testing
ECC = lsokinetic eccentric. NVA = Noninjuredlafter testing.
TABLE 6. Pearson correlation coeficients matrix for 'IA= Injuredlditer
percent deficit as determined by each testing mode. TABLE 7. Pain rating for each testing mode.

JOSPT Volume 24 Number 3 September 1996

 tion of this practice has been that with knee dysfunction, Deones et al pain ratings was found between test-
regardless of the mode of muscle as- (5) found that hand-held dynamome- ing modes.
sessment, similar differences between try at 0 and 60" did not detect a dif-
limbs will be found. In his review arti- ference between the involved and Limitations
cle on muscle performance assess- uninvolved quadriceps peak force,
ment, Sapega (16) suggested that a whereas concentric isokinetic dyna- Limitations in this study include
muscle performance deficit of 20% mometry at 60°/sec did detect a sig- the heterogeneous sample with re-
or greater in a limb is "probably ab- nificant difference. They raised the spect to diagnostic groups and age.
normal," and deficits ranging from issue of the confounding influence of However, this represents a patient
10 to 20% are "possibly abnormal." the strength of the tester as well as population and adds to the external
The results of this study demonstrate the lack of stabilization in hand-held validity of the study in our opinion.
that for each individual subject, the dynamometry. In contrast, this study The small sample size ( N = 23) is
percent deficit between the involved showed a significant difference be- also acknowledged, and it is recog-
and uninvolved quadriceps peak tween involved and uninvolved quad- nized that this may have impacted
force ranged from as much as -22 to riceps peak force using hand-held the large variability in the test results.
+46.5%, depending on the testing and eccentric isokinetic dynamometry Gravity correction was not used on
device. As was reported, only seven of but no significant difference using the Kin-Com, as it was not possible to
the 23 subjects had agreement in the concentric isokinetic dynamometry gravity correct the data obtained us-
sign of the percent deficit in all four and isometric dynamometry. It was ing hand-held dynamometry as well
Downloaded from www.jospt.org at on May 9, 2025. For personal use only. No other uses without permission.

tests. Certainly one would draw differ- theorized that the strength of the as gravity correction not commonly
ent conclusions regarding the status tester may be a confounding variable being used in clinical testing. In addi-
of the involved limb, depending on in the use of hand-held dynamome- tion, reliability data revealed that a
which testing device is used. try. A pilot study revealed a signifi- learning effect occurred between ses-
Kues et al (10) compared isomet- cant difference between testers of sions for the isometric testing mode.
Copyright © 1996 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.

ric peak torque, concentric peak different strength. Consequently, the

torque, and eccentric peak torque in Future Research
20 healthy female subjects. They re-
ported that the coefficient of deter- A need exists for future research
mination (r2) ranged from .73 to .94 No difference in in the area of muscle performance
assessment to clarify the difierences
between isometric, concentric, and
eccentric peak torque. These mea-
pain ratings was between various modes of testing.
sures were "moderately to highly cor- found bet ween Similar studies with larger clinical
related, suggesting that the measure- samples may decrease the effect of
Journal of Orthopaedic & Sports Physical Therapy®

ments obtained during different testing modes. subject variability and clarify relation-
maximal voluntary contractions may ships between testing modes.
be assessing similar components of
performance." Based on the small decision was made to use a single CONCLUSION
sample size in their study (N = 20) tester for the hand-held dynamome-
and subject variability, they identified ter testing. Differences between various
a coefficient of determination of .70 Deones et al (5) also suggested modes of muscle performance assess-
or greater as a strong correlation, a that pain may be a confounding vari- ment were examined in this study.
coefficient of determination from .50 able in a muscle performance deficit Results of this study demonstrate that
to .69 as a moderate correlation, and but stated that "no patients reported quadriceps testing by hand-held dyna-
a coefficient of determination less an increase in pain during or after mometry and eccentric isokinetic dy-
than .5 as a weak correlation. Our the testing procedure." However, namometry produced significant dif-
coeficients of determination ranged they failed to quantify pain ratings ferences between the involved and
from 0.12 to 0.58 between testing pre- and posttesting. As a secondary uninvolved quadriceps peak force,
modes in subjects with knee dysfunc- purpose of this study was to evaluate whereas quadriceps testing by con-
tion. Our coefficients of determina- the confounding effect of pain on centric isokinetic dynamometry and
tion were lower, indicating a weaker test results, pain ratings were assessed isometric dynamometry showed no
correlation between testing modes in before, during, and after each testing difference. In addition, large varia-
our study, which is supported by the mode. The results demonstrated that tion existed between different modes
variability in percent deficits between subjects' average pain ratings for all of testing for individual subjects and
different testing modes (Table 5). modes ranged from 0-2 on a 0-6 resulted in different conclusions re-
In a previous study using patients scale (Table 2). No difference in garding quadriceps muscle perfor-

Volume 24 Number 3 September 1996 JOSPT

 RESEARCH STUDY
- ., . ". - - - - --.- .- -

mance. No significant difference ex- Sci Sports Exerc 2 1:304 -307, 1989 Sapega AA: Muscle performance eval-
isted in pain ratings before, during, 8. Knapik JJ, Ramos MU: lsokinetic and uation in orthopaedic practice. J Bone
isometric torque relationships in the Joint Surg 72A(lO):1562- 1572, 1990
or after testing, and no differences
human body. Arch Phys Med Rehabil Shrout PE, Fleiss JL: Interclass correla-
were found in pain ratings between 61:64-67, 1980 tions: Uses in accessing rater reliability.
testing modes. JOSPT 9. Knapik JJ, Wright JE, Mawdsley RH, Psycho1 Bull 86(2):420-428, 1979
Braun J: Isometric, isotonic, isokinetic Stuberg WA, Metcalf WK: Reliability of
torque variations in four muscle groups quantitative muscle testing in healthy
REFERENCES through a range of a joint. Phys Ther children and in children with Duch-
63:938-947, 1983 enne muscular dystrophy using a hand-
1. Aniansson A, Gimbry G, Rundgren A:
10. Kues J, Rothstein JM, Lamb RL: The held dynamometer. Phys Ther 68(6):
lsometric and isokinetic quadriceps
relationship among knee extensor 977-98 1, 1988
muscle strength in 70-year-old men
torques produced during maximal vol- Sullivan SJ, Chesley A, Herbert G, Mc-
and women. Scand J Rehabil Med 12:
untary contractions under various test Faull S: The validity and reliability of
161-168, 1980
conditions. Phys Ther 74(7):674- 682, hand-held dynamometry in assessing
2. Baumgartner TA: Norm referenced
1994 isometric external rotator performance.
measurement: Reliability. In: Safrit MJ,
Woods TM (eds), Measurement Con- 11. Lankhorst GJ, Vande Stradt RJ, Van de J Orthop Sport Phys Ther 1O(6):2l3-
cepts in Physical Education and Exer- Korst JK: The relationships of functional 217, 1988
cise Science, pp 45-72. Champaign, IL: capacity, pain, and isometric and iso-
Surburg PK, Suomi R, Poppy WK: Va-
Human Kinetics Publishers, 1989 kinetic torque in osteoarthritis of the
lidity and reliability of hand-held dyna-
3. Bohannon RW: Test-retest reliability of knee. Scand / Rehabil Med 17:167-
172, 1985 mometer with two populations. J Or-
hand-held dynamometry during a sin- thop Sports Phys Ther 16:229-231,
12. Lord JP, Aitkens SG, McCrory MA, Bre-
Downloaded from www.jospt.org at on May 9, 2025. For personal use only. No other uses without permission.

gle session of strength assessment. Phys 1992

Ther 66(2):206-209, 1986 mauer EM: lsometric and isokinetic
measurement of hamstring and quadri- Talag T: Residual muscle soreness as
4. Bohannon RW, Andrews AW: Inter- influenced by concentric, eccentric,
rater reliability of hand-held dynamom- cep strength. Arch Phys Med Rehabil
73:324-330, 1992 and static contractions. Res Q 44:458-
etry. Phys Ther 67(6):93 1-933, 1987 469, 1973
5. Deones VL, Wiley SC, Worrell T: As- 13. Lysholm J: Comparison between pain
Copyright © 1996 Journal of Orthopaedic & Sports Physical Therapy®. All rights reserved.

sessment of quadriceps muscle perfor- and torque in an isokinetic strength test Tredinnick TJ, Duncan PW: The reli-
mance by a hand-held dynamometer. of knee extension. Arthroscopy 3(3): ability of measurements of concentric
J Orthop Sports Phys Ther 20(6):296- 182- 184, 1987 and eccentric loading. Phys Ther 68(5):
301, 1994 14. Otis JC, Godbold JH: Relationship of 656 - 659, 1988
6. Hayes KW, Falconer J: Reliability of isokinetic torque to isometric torque. Van der Ploeg RJO, Oosterhuis HIGH:
hand-held dynamometry and its rela- J Orthop Res 1:165-171, 1983 The "makebeak test" as a diagnostic
tionship with manual muscle testing in 15. Reed RL, Hartford RD, Yochum K, tool in functional weakness. J Neurol
patients with osteoarthritis in the knee. Pearlmutter L, Ruttinger AC, Moora- Neurosurg Psychiatry 54:248-25 1, 1991
J Orthop Sports Phys Ther 16(3):145- dian AD: A comparison of hand-held Wikholm JB, Bohannon RW: Hand-
149, 1992 isometric strength measurement with held dynamometer measurements:
7. Kannus P, Jarvinen M : Prediction of
Journal of Orthopaedic & Sports Physical Therapy®

isokinetic muscle strength measure- Tester strength makes a difference.

torque acceleration energy and power ment in the elderly. J Am Geriatr Soc J Orthop Sports Phys Ther 13(4):191-
of thigh muscles from peak torque. Med 4 1 53-56, 1993 198, 1991

JOSPT Volume 24 Number 3 September 1996