Genucom, KT-1000, and Stryker knee laxity
measuring device comparisons
Devicereproducibility and interdevice
asymptomatic subjects
CARL L. HIGHGENBOTEN,*† MD,
*
From Orthopaedic Consultants, Dallas, Texas, and ‡ North
ABSTRACT
Generally, three devices (the Genucom Knee Analysis
System, the MEDmetric KT-1000 Arthrometer, and the
Stryker Knee Laxity Tester), which have been pre-
sented in the literature, may be used to assess clinically
the amount of knee laxity in an objective manner. This
study compared the reproducibility of the anterior and
posterior knee laxity values for each and made direct
comparisons of the results obtained. Thirty asympto-
matic subjects were given a test-retest protocol on
both legs with all three devices. Devices were tested in
a counterbalanced order. Repeatability of test values
within devices was variable, but all were acceptable,
the lowest being r 0.74. Analysis of variance (AN-
=
OVA), and correlational analysis revealed that device-
specific anterior and posterior laxity values were pro-
duced. We concluded that each commercially available
knee laxity testing device can provide reproducible
quantitative measurements of knee laxity; however,
due to differences in device sensitivities and functional
design, numerical results from one device cannot be
generalized to another device.
Recent literature reports the use of various testing devices
in order to obtain objective knee laxity measurement. 14
These devices-the Stryker Knee Laxity Tester (Stryker,
Kalamazoo, MI), the MEDmetric KT-1000 Arthrometer
(MEDmetric Corporation, San Diego, CA), and the Genu-
t Address correspondence and repnnt requests to Carl L Highgenboten,
MD, Orthopaedic Consultants, 7777 Forest Lane, Building C, Suite 106, Dallas,
TX 75230
743
comparison in
ALLEN AND
JACKSON,‡ EdD,
NEIL B. MESKE,* MS
Texas State University, Denton,
Texas
com Knee Analysis System (FARO Medical Technologies
Inc., Champlain, NY)-can be used to test anterior-pusteriur
knee laxity. All three test devices and protocols are to a
large degree unique, but they have some common traits.
When a knee laxity measuring device is attached to the leg
that is to be tested, an anterior-posterior drawer test is
conducted and the anterior-posterior displacements are re-
corded in millimeters. One of the primary purposes of such
test devices is to aid in the diagnosis of disruptiun of the
anterior cruciate and/or posterior cruciate ligaments and
the determination of improvement from surgery attd treat-
ment. Markolf and AmStUtZ6 have stated that future devel-
opments in sports medicine should lead to htandardized
methods of quantifying knee stability.
The Stryker Knee Laxity Tester is operatiuttally the least
complicated of the test devices. It includes a patient pusi-
tioning seat, a force applicator, and a measuring instrument
for recording the millimeters of displacement resulting from
the anterior-posterior drawer test. Boniface et al.l measured
the displacements of uninjured knees and knees with ACL
disruption. Their tests were conducted with the knee joint
at a 20° flexion angle and 89 N of applied force. The injured
knees had a mean anterior displacement of 8.1 mm, and the
uninjured knees had a mean displacement of 2.5 mni.
The MEDmetric KT-1000 Arthrometer is a more complex
device than the Stryker Knee Laxity Tester and alluws a
more sensitive test analysis. It also measure5 attteriur-pus-
terior knee laxity displacements resulting trom atr anterior-
posterior drawer test. Daniel et a1.1 2 compared uninjured
knees and knees with ACL disruption using the arthrometer.
They also conducted their tests at a knee joint angle of 20°
of flexion and used 89 N of force. Their repurted mean
anterior displacement values were 13.0 man for ACL injured
knees and 5.65 mm for uninjured knees. Sherman et al.8
744
conducted a comparison of the KT-1000 and the University
of California at Los Angeles (UCLA) instrumented clinical
knee testing apparatus. The results indicated that both
devices, while providing device-specific data, were able to
correctly classify 90% to 95% of patients with disrupted
ACLs.
The Genucom Knee Analysis System, a computerized tool
for the evaluation of the knee, is the most complicated of
the test devices. It combines a computer, a biomechanical
digitizer, an electrogoniometer, a dynometer, and an inkjet
printer. It provides a variety of test protocols, which theo-
retically allows a total description of knee ligament stability.
Highgenboten and Jackson5 have reported that the device
gives reproducible test-retest results for a variety of tests.
Of particular relevance is an anterior-posterior drawer test
conducted on uninjured subjects. The knee joint angle was
30° of flexion and the force was 93.45 N. The mean anterior
displacement was 5.67 mm.
The purpose of the present investigation was to compare
the numerical results of anterior-posterior knee laxity test-
ing using the three test devices mentioned. Specifically,
reproducibility coefficients were estimated for each device
and a comparison of test results was made to determine if
significant differences existed, and if they did, to determine
their meaning.
METHODS
Subjects
Both knees of 30 subjects having no prior history of knee
injury abnormality provided a sample of 60 knees. Ages
or
of the 9 males and 21 females ranged from 16 to 44 years.
Each volunteer signed an informed consent that provided
information concerning the voluntary nature of their partic-
ipation, the purpose of the study, and the anticipated bene-
fits from its successful completion.
Procedures
Subjects received two complete tests on each leg. A complete
test using all three devices was performed on one leg and
then the opposite leg; then, each leg received a second set of
complete tests. Intertest reliability was estimated between
the two complete tests. During testing with each of the three
devices, three repeated measurements, or trials, were re-
corded. Intertrial reproducibility was estimated from the
three repeated measurements. The tests administered to
each subject were performed in succession on the same day.
The order of the devices used in testing was counterbalanced
across subjects, resulting in a Latin Square design, which
prevents order effects in the results. Thus, each subject was
given two complete tests using each of three devices on each
leg, with three repeated measurements per device during
each test session. Anterior and posterior displacements were
recorded in millimeters for each trial. Therefore, the inde-
pendent variables are &dquo;test device with three levels&dquo; and
&dquo;test administration with two levels.&dquo; The dependent vari-
ables are the anterior and posterior laxity measurements.
For the Stryker Knee Laxity Tester and the MEDmetric
KT-1000 Arthrometer, an anterior-posterior drawer test was
performed at 20° of flexion and 89 N (20 pounds) of force.
The Genucom test was also performed at 20°, but the system
requires a force of 93.45 N (21 pounds), which was the only
deviation in test protocol between devices. Standard testing
procedures for each test device, as provided by their vendors,
were followed. Detailed descriptions of test protocols using
the Stryker and the KT-1000 are available in the literature.’-
3, 5,
8 However, the Genucom protocol has not been described;
therefore, the essential steps in conducting a test on the
device are given as follows:
1. The subject’s permanent data file was established using
the system’s computer and stored on disk.
2. The subject’s test leg was secured with the system’s
thigh restraints, and an electrogoniometer was attached to
the lower leg.
3. Seven anatomical landmarks (tibial tubercle, tibial
crest, medial femoral condyle, lateral femoral condyle, me-
dial tibial plateau, lateral tibial plateau, and the inferior
pole of the patella) were physically marked and digitized,
and the distance between them recorded.
4. A soft tissue compensation test in three planes was
performed to compensate for possible movement that can
occur within the test apparatus without affecting the dis-
tance between the digitized bony landmarks.
5. The anterior-posterior drawer test was performed at
20°.
6. The subject’s data was stored permanently on disk.
7. A hard copy of each test was printed, from which data
for statistical analysis was obtained.
Data analysis
To estimate reproducibility, intraclass correlation proce-
dures, which incorporate an ANOVA model, were used to
determine the relationship between trials and if significant
differences existed between trials. A 3 (test device) by 2
(test) ANOVA with repeated measures on both factors was
conducted to determine if significant differences existed
between test devices and if the test device interacted with
repeated tests. Correlations were also calculated between the
test device results to determine their relationships.
RESULTS
Reproducibility
As indicated in the Procedures, three trials or repeated
measurements wereadministered within a test or device
session. Intraclass correlations indicated that all three de-
vices demonstrated excellent reproducibility for anterior and
posterior laxity values across the three trials within a ses-
sion. The values are given in Table 1. Trend analysis indi-
cated two significant trial effects, but the magnitude of the

TABLE 1
Descriptive statistics and reliability estimates of laxity values
TABLE 2
Correlation matrices of knee laxity test device results
a p < 0.0] .
trend
was not greater than 0.3 mm and was considered
clinically unimportant. Complete test or device sessions were
given on a test-retest basis. Table 1 provides the statistical
results relevant to the completed test session reproducibility
analysis. R was variable between devices but acceptable, the
lowest being the Stryker’s anterior laxity value of 0.74. The
trend analysis indicated no significant differences between
test sessions.
Device comparison
The 3 (test device) by 2 (test) ANOVA with repeated meas-
ures revealed a significant difference between the device
measures for anterior (F 2,118 62.77; P < 0.01) and
=
posterior (F 2,118 17.26; P < 0.01) laxity values. The test
=
administration and interaction effects were not significant.
A post hoc analysis (P 0.01) was conducted to examine
=
the device main effect. The two degrees of freedom were
partitioned within the ANOVA to provide two single degree
of freedom contrasts. These results indicated that the Gen-
ucom produces significantly higher anterior laxity values (F
1,59 31.3; P < 0.01) than the Stryker or the KT-1000. The
=
KT-1000 produced significantly higher anterior laxity values
(F 1,59 130.31; P < 0.01) than the Stryker. For posterior
=
laxity, the Stryker and Genucom values were not signifi-
cantly different, but they were significantly higher (F 1,59
=
45.7; P < 0.01) than the KT-1000 values. Table 2 shows
the results of a between-device correlation. While several of
the correlations were significant (P < 0.01), they tended to
range from moderate to extremely low values. These low to
moderate correlations, along with the ANOVA results, in-
dicated that the anterior and posterior values were specific
to the device used in the test.
745
DISCUSSION
Intradevice reproducibility
As noted by Sherman et al.’, in order to obtain reproducible
(i.e., repeatable numerically) quantitative data from any of
these knee laxity measuring devices, the test administrator
must become proficient in the correct use and application
of the device and in the testing protocol. A routine testing
protocol should be followed each time for each patient tested.
In this study, the test administrator spent in excess of 40
hours becoming proficient with each of the three testing
devices and the three test protocols. The manufacturers of
the Genucom System offer a testing and certification pro-
gram. The test administrator for this study has been rated
as excellent on the anterior-posterior drawer test by this
certification program. Hanten and Pace4found reproducible
results with the KT-1000 during repeated trials and repeated
testers. Based on our data in Table 1, we conclude that all
three devices can provide reproducible test results.
lnterdevice comparisons
The Genucom produces anterior laxity values higher than
either the KT-1000 or the Stryker, and the KT-1000 pro-
duces significantly higher anterior laxity values than the
Stryker. As Table 1 demonstrates, a higher degree of inter-
individual variability will be detected by the Genucom due
to the larger standard deviations it produces for knee laxity
measurements.
The posterior laxity values of the Genucom and the Stry-
ker were not significantly different from one another, but
each was significantly different from the posterior laxity
values obtained from the KT-1000. The reason for this
difference may be understood by looking at the functional
design of the three devices. The Genucom fixes the femur.
The Stryker attempts to restrict movement of the femur
with the use of a VELCRO (VELCRO USA Inc., Manches-
ter, NH) strap. The KT-1000 does not attempt to fix the
femur. Sherman et al.’ found in his study of the KT-1000
and another knee laxity measuring device that the KT-1000
demonstrated twice the magnitude in laxity values for both
normal and ACL deficient knees when a force of 89 N was
applied. However, the difference in magnitude decreased to
less than 2 mm when the standard force of 200 N was
applied with the other device. The only difference between
devices was the fact that the KT-1000 did not fix the femur,
while the other device did. They concluded that each device
produced device-specific measurements due to differences in
functional design.
Statistical analysis of the results of our study (Table 2)
indicates that each device produces anterior and posterior
laxity measurements that are peculiar for each device and
related to sensitivity and functional design of the device.
Clinical
significance
For the clinical practitioner, device-specific results do not
matter. The practitioner should be aware of the following:
746
1. The devices will provide reproducible results when
appropriate practice and test experience have been achieved.
2. Specific laxity values and results for one device cannot
be generalized to another.
3. Past studies have demonstrated the Stryker Knee Lax-
ity Tester and the MEDmetric KT-1000 will discriminate
between normal knees and ACL disrupted knees.1-3,8a
Currently, literature is not available supporting the ability
of the Genucom to accomplish that goal. Oliver and
Coughlin’ have demonstrated, however, that the Genucom
test results will have a high degree of agreement for the
anterior-posterior drawer test at 30° with clinical examina-
tion for subjects with an abnormal knee history. In addition,
the Genucom gives reproducible laxity values. Thus, it is
highly probable that it can differentiate between normal and
ACL disrupted knees.
A final point that should be mentioned in discussing these
instrumented knee laxity test devices is the cost factor. The
costs of these devices vary widely, with the Stryker at $900,
the KT-1000 at $2900, and the Genucom at $49,000. While
the Stryker and KT-1000 are relatively simple mechanical
devices with a small cost, the Genucom is a complex com-
puterized system which allows a full battery of knee laxity
tests for analysis of total knee ligament stability, including
varus-valgus, pivot shift, and recurvatum. Its computer ca-
pacity allows storage of patient data for subsequent compar-
ison of later tests to determine success of treatment and
rehabilitation.
Physicians must determine their needs in knee laxity
testing. If measurement of anterior-posterior laxity is all
that is needed, then the Stryker or KT-1000 will be accept-
able. However, if a more sophisticated analysis of knee
ligament status is needed, then the Genucom and its asso-
ciated expenses may be justified.
REFERENCES
1 Boniface JR, Fu FH, Ilkhanipour K. Objective anterior cruciate ligament
testing Orthopedics 9 391-393, 1986
2. Daniel DM, Malcom LL, Losse G, et al: Instrumented measurement of
anterior laxity of the knee. J Bone Joint Surg 67A: 720-726, 1985
3 Daniel DM, Stone ML, Sachs R, et al. Instrument measurement of anterior
knee laxity in patients with acute anterior cruciate ligament disruption. Am
J Sports Med 13 . 401-407, 1985
4 Hanten WP, Pace MB Reliability of measuring anterior laxity of the knee
joint using a knee ligament arthrometer. Phys Ther 67. 357-359, 1987
5 Highgenboten C, Jackson AW, Meske NB Measurement of knee laxity
using the Genucom analysis system Med Sci Sports Exerc 21. S81, 1989
6 Markolf KL, Amstutz HC. The clinical relevance of instrumented testing of
ACL insufficiency. Experience with the UCLA clinical knee testing appara-
tus Clin Orthop 223 198-207, 1987
7 Oliver JH, Coughlin LP. Objective knee evaluation using the Genucom
Knee Analysis System. Clinical implications Am J Sports Med 15 . 571-
578, 1987
8 Sherman OH, Markolf KL, Ferkel RD: Measurements of anterior laxity in
normal and anterior cruciate absent knees with two instrumented test
devices Clin Orthop 215
. 156-161, 1987