Journal of Orthopaedic & Sports Physical Therapy

2001;31(10) :588-597

Hop Tests as Predictors of Dynamic Knee

Stability
G. Kelley Fitzgerald, PhD, PT; O C 9
Scott M . Lephart, PhD, ATC?
)i Hye Hwang, MD, PhD)
Maj Robert S. Wainner, PhD, PT; OCS, ECS4

Single leg hop tests are commonly used as physical performance measures of function ingle leg hop tests are
and are also commonly used to evaluate progress in knee rehabilitation programs,
particularly for individuals recovering from anterior cruciate ligament injury or
reconstructive surgery. While there is some evidence that hop tests may show promise as a
predictive measure for identifying individuals who are at risk for recurrent dynamic
instability, further work is needed to clearly define the role of hop test measurements for this
purpose. The purposes of this clinical commentary are to review the research that has been
S commonly used as physi-
cal performance mea-
sures of function, partic-
ularly in individuals re-
covering from anterior cruciate
ligament (ACL)injury or surgical
Downloaded from www.jospt.org by 114.142.169.48 on 07/04/19. For personal use only.

done to establish hop tests as a physical performance measure of function, to discuss reconstruct~on~~.~2.~.s.~~.m.n:~~:~Ic
neuromuscular and biomechanical considerations related to hop performance and dynamic A variety of hop test procedures
knee stability, to discuss existing evidence that supports the potential for hop tests as a have been described, including
predictor of dynamic knee stability, and to discuss considerations for future studies that are single leg hop tests for distance or
designed to more clearly define the role of hop tests in predicting dynamic knee stability. ) time,m.-:.. " " q 5. hop and stop tests,22

Orthop Sports Phys Ther 200l;3 1:588-597. and vertical jump t e ~ t s . ~ - ~ ~ T C l i n i -

cians have used single leg hop
Key Words: anterior cruciate ligament, functional performance testing, hop tests
tests to assess their patients' lower
extremity muscular strength and
ability to perform tasks that chal-
J Orthop Sports Phys Ther 2001.31:588-597.

lenge knee stability. Single leg hop

tests are also commonly used to
evaluate progress in knee rehabili-
tation programs.
Previous hop test studies have
provided information on measure-
ment the relation-
ships between hop test measure-
Assistant professor, Department of Physical Therapy, School of Health and Rehabilitation Sciences, ments and other physical impair-
University of Pittsburgh, Pittsburgh, Pa. ments, such as muscle weak-
Director, Neuromuscular Research laboratory Musculoskeletal Research Center, Department of ness,".17."i27.%i passive joint
Orthopaedic Surgery, University of Pittsburgh, Pittsburgh, Pa. l a ~ i t y , ~and * . ~knee
~ joint proprio-
Assistant professor, Department of Physical Medicine and Rehabilitation, Sungkyunkwan University, ception defi~its;~." and the use of
College of Medicine, Samsung Medical Center, Seoul, Korea.
Research and practice coordinator, Wilford Hall United States Air Force Medical Center, lackland hop tests as indicators of function-
AFB, Tex. al performance capacity in pa-
The opinions or assertions contained herein are the private views of the authors and are not to be tients with ACL injllry~~li.l 32025.35
construed as official or as reflecting the views of the Department of the Air Force or the Department Hop tests have also been used in
of Defense.
Send correspondence to G.Kelley Fitzgerald, Department of Physical Therapy, School of Health and
studies as a clinical measure of
Rehabilitation Sciences, University of Pittsburgh, 6035 Forbes Tower, Pittsburgh, PA 15260. E-mail: progress in response to surgical or
kfitzger+ @pitt.edu rehabilitation interven-
 tions.1~1~iJ'~1~2R
However, no studies have adequately deficit. In general, the investigators found that using
established the predictive ability of hop tests to iden- a combination of single leg hop tests to detect a b
tify individuals who may have future problems with normal LSIs was more sensitive than using any one
dynamic knee stability as a result of knee injury or hop test in isolation. However, in both studies, a sig-
pathology. While there is some evidence that hop nificant number of subjects with ACL deficiency
tests may show promise as a predictive measure for demonstrated normal LSIs on the hop tests. Further-
identifjmg individuals who are at risk for recurrent more, it was unclear whether normal or abnormal
dynamic i n ~ t a b i l i t y , ~ ~further
. ~ ~ . ~ work
' is needed to LSIs correlated well with a patient's overall function-
clearly define the role of hop test measurements for al ability. In order for hop tests to be useful in evalu-
this purpose. This clinical commentary reviews the ating deficits in lower limb function, it would be im-
research that has been done to establish hop tests as portant to know how these tests correlate with other
a physical performance measure of function, discuss- measures of impairment and function, as well as how
es the evidence that exists to support the potential accurately they can predict which subjects are ready
for hop tests as a predictor of dynamic knee stability, to return to full activity and which subjects are at risk
and offers considerations for future studies that are for continued problems with functional instability.
designed to more clearly define the role of hop tests Hop tests have been used in studies to reflect
in predicting dynamic knee stability. changes in functional status in response to knee re-
habilitation p r o g r a m ~ ~ " ~well
~ . ~as
% surgical
~ inter-
HOP TESTS AS MEASURES OF FUNCTION AND vention~.~J~.~!'In most of these studies, the data indi-
cate that performance on hop tests generally im-
DISABILITY
proves concomitantly with improvements in other
Measurement Properties outcome measures that are used to reflect changes in
functional status in response to treatment. It would
Measurement reliability has been reported for vari- seem reasonable that hop tests could be used to re-
ous hop test procedures in nonimpaired subjects? flect changes in the patient's status in response to
as well as in subjects who have undergone ACL re- treatment. What is lacking, however, is information
constr~ction.~ In nonimpaired individuals, intraclass that would allow us to determine just how much
Downloaded from www.jospt.org by 114.142.169.48 on 07/04/19. For personal use only.

correlation coefficients (ICCs) were reported to vary change in hop test performance would constitute a
from 0.932 to 0.96j for the single hop for distance, clinically meaningful change in response to treat-
0.94' to 0.955 for the triple hop for distance, 0.90' to ment. To our knowledge, no reported studies have
0.963 for the triple crossover hop for distance, and established the responsiveness of hop test scores to
0.665 to 0.9P for the &meter timed hop test. In s u b change or the magnitude of change that would r e p
jects who have undergone ACL reconstruction, ICCs resent a minimal clinically important difference in
for the single hop for distance, the &meter timed the patient's functional status. Further work is need-
hop, and the vertical jump test were reported to vary ed to establish these important measurement charac-
from 0.88 to 0.97." Based on the available data, it a p teristics and clarify the role of hop tests as treatment
pears that hop test measurements demonstrate rea- outcome measurements.
J Orthop Sports Phys Ther 2001.31:588-597.

sonable reliability in both nonimpaired subjects and

those who have undergone ACL reconstruction. Relationship of Hop Test Scores With Other
A ratio of limb symmetry known as the limb sym- Measurements of Impairment and Disability
metry index (LSI) has been the most frequently re-
ported criterion to assess whether a hop test is nor- H@ tests and muscle perfmance Although hop
mal or a b n ~ r m a l . ~The
- l ~LSI expresses the distance test scores are sometimes believed to reflect lower ex-
or time recorded from the test of the involved limb tremity muscular strength, there are conflicting re-
to the test of the uninvolved limb as a percentage ports among studies that have investigated the rela-
(involved/uninvolved * 100 = LSI). Normal LSI val- tionship of hop test scores with quadriceps and ham-
ues have been reported as greater than or equal to string strength measurements in subjects with either
80%" and greater than or equal to 85%.' It should ACL injury or those who have undergone ACL re-
be clarified that these values were empirically estab- construc~on!i.!!.5.6' ;2H,Xi Table 1 summarizes reported
lished by noting that 90% of subjects without a hist* correlations between various hop test scores and
ry of ACL injury had LSIs of greater than or equal measures of muscle performance. Differences in
to 80%" and greater than or equal to 85%.s methodologies among studies that have investigated
Some investigators have used the LSI to determine the relationship between muscle strength and hop
the sensitivity and specificity of single leg hop tests test scores most likely contribute to conflicting re-
for detecting deficits in lower limb function in s u b sults. There were differences in strength testing pro-
jects with ACL defi~iency.~"z~ The underlying assump cedures, hop test procedures, and follow-up periods
tion in these studies was that detection of an abnor- in all of these studies. Despite these differences, it
mal LSI would indicate the presence of a functional appears that there is only a low to moderate relation-

J Orthop Sports Phys Ther .Volume 31. Number 10. October 2001
 TABLE 1. Reported correlations between hop test scores and measurements of muscle performance.
Citation Sample Muscle performance measure Correlations with hop tests
Wilk et aP6 Subjects with anterior cruciate ligament Peak isokinetic knee extension torque at Single hop = 0.41-0.62
reconstruction( N = 50) 180°s, 300°/s, 450% Crossover hop = 0.53-0.69
Timed hop = 0.45-0.60
Peak isokinetic knee flexion torque at Reported no correlations with hop
18O0/s, 30O0/s, 450% tests; correlation coefficients not
provided
Petschnig et atz6 Subjects with anterior cruciate ligament Peak isokinetic knee extension torque at Single hop = 0.45-0.51
reconstruction (N = 55) 1Sols Triple hop = 0.48-0.55
Vertical Jump = 0.01-0.51
Borsa et aI6 Subjects with anterior cruciate ligament Peak isometric knee extension torque at Single hop = 0.06
deficiency ( N = 29) 60" of knee flexion
Noyes et alz5 Subjects with anterior cruciate ligament Peak isokinetic knee extension torque at Single hop = 0.26-0.49
deficiency (N = 67) 6O0/s, 30O0/s
Peak isokinetic knee flexion torque at Single hop = 0.00432
180°/s, 300°/s
Risberg et alz8 Subjects with anterior cruciate ligament lsokinetic knee extension total work at Triple hop = 0.09-0.58
reconstruction( N = 60) 60"/s, 240'1s Stair hop = 0.08-0.47
lsokinetic knee flexion total work at 60'1 Triple hop = 0.04-0.58
s, 240% Stair hop = 0.06-0.53

ship between measurements of lower extremity mus- ship between hop test scores and self-report mea-
cular strength and performance on hop tests in indi- sures of knee function and disability in patients with
viduals with either ACL injury or those who have un- ACL injury o r those who have undergone ACL re-
dergone ACL reconstruction. This would imply that ~ ~ ~ J .4" 6summarizes reported
c o n ~ t r u c t i o n . ~ . ~ Table
there are other factors that would influence perfor- correlations between various hop test scores and self-
mance on hop tests in addition to an individual's lev- report measures of knee function. The relatively low
Downloaded from www.jospt.org by 114.142.169.48 on 07/04/19. For personal use only.

el of lower extremity muscular strength. correlations between hop test scores and self-report
Hop tests and passive anta'or knee joint laxity The measures of function may indicate that neither of
relationship between passive anterior knee joint laxi- these methods can stand alone as an adequate assess-
ty and hop test scores has been studied in patients ment of knee function. Each of these methods may
with ACL deficiency, as well as those who have un- capture different aspects of physical performance
' ~ ~ ~ 2~ ~
dergone ACL r e c o n s t r u c t i ~ n . Table summariz-
~~ and function, and both types of measurements may
es reported correlations between various hop test be needed to more completely describe the patient's
scores and measurements of anterior knee laxity. The status of function and disability at a given point in
available evidence indicates that passive knee joint time.
laxity does not appear to influence performance on
J Orthop Sports Phys Ther 2001.31:588-597.

single leg hop tests.

Hop tests and knee joint position sense Passive knee HOP TESTS AS A METHOD TO EVALUATE
joint position sense has been used as a measure to BIOMECHANICALAND NEUROMUSCULAR
identify proprioception deficits in subjects with ACL MECHANISMS OF DYNAMIC KNEE STABILITY
deficiency. Table 3 summarizes reported correlations
between hop test scores and measurements of knee The studies we have reviewed thus far have empha-
joint position sense. The relationship between passive sized the use of hop tests as clinical, physical perfor-
knee joint position sense and performance on single mance measures of knee function and stability. It is
leg hop tests is unclear at this time.'-"he conflicting also important to recognize that these same tests may
results, summarized in Table 3, may be in part a be useful in the laboratory setting to examine biome-
function of different methodologies to measure knee chanical and neuromuscular profiles in individuals
joint position sense. In either case, there does not with knee ligament injuries. Using hopping and
appear to be a strong correlation between knee joint landing activities would allow investigators to exam-
position sense and performance on the single hop ine lower extremity kinematic, kinetic, and EMG
for distance test. To our knowledge, there is no data characteristics under conditions where dynamic knee
comparing the relationships of other types of hop stability is significantly challenged. Information
test procedures, such as the crossover triple hop or gained from these studies might allow us to deter-
timed hop test, with passive knee joint position mine the differences between effective and ineffec-
sense. tive control strategies, which may, in turn, guide the
Hop tests and self-rpport measures of function and development of effective rehabilitation strategies for
disability A few studies have examined the relation- patients who have sustained knee ligament injuries

590 J Orthop Sports Phys Ther-Volume 31 .Number 10.0ctober 2001

 TABLE 2. Reported correlations between hop test scores and measurements of anterior knee joint laxity.
Citation Sample Laxity measurement Correlations with hop tests
Sernert et aln Subjects with anterior cruciate ligament reconstruc- KT-1000knee arthrometer Single hop = -0.08
tion ( N = 527) Manual Lachman test Single hop = 0.09
Risberg et alZ8 Subjects with anterior cruciate ligament reconstruc- KT-1000knee arthrometer Triple hop = -0.26-0.00
tion ( N = 60) Stair hop = -0.23-0.01
Eastlack et all1 Subjects with anterior cruciate ligament deficiency KT-2000knee arthrometer Reported no correlation with single hop,
(N= 45) crossover hop, triple hop, or timed hop
tests; correlation coefficients not provided

or have undergone surgical stabilization procedures. injury during performance of a single leg hop
Presently, there have been only a few studies in task.lCs1 Gauffin and Troppl%tudied bilateral kine-
which hop tests were used to compare the biome- matic, kinetic, and muscular activation patterns at
chanical and neuromuscular profiles between sub- the ankle, knee, and hip joints during a 1-legged
jects with nonimpaired knees and those with ACMe- jump in subjects who were chronically ACL deficient.
ficient knees or those who have undergone ACL re- Their results revealed similar involved and uninvol-
construction. ved limb hop scores; however, the movement pat-
Some investigators have described hop test perfor- terns and muscle activity patterns differed between
mance from a biomechanical perspective. Colby et the injured and noninjured limbs. The ACMeficient
all1 calculated stability indices using vertical ground limb demonstrated greater flexion angles for the hip
reaction force to measure the stabilizing characteris- and knee joints at touchdown. In addition there was
tics of subjects with nonimpaired, ACMeficient, and a reduction in EMG activity of the injured limb
ACLreconstructed knees during a 1-legged hop and quadriceps compared to the noninjured limb, while
a step down test. They were able to detect some hamstring EMG activity remained similar between
slight differences in vertical ground reaction force limbs. Gauffin and Tropp suggested that the in-
characteristics between the injured and noninjured creased hip and knee flexion, combined with a re-
limbs of both the ACMeficient and ACLrecon- duction in quadriceps activity, would improve the ca-
Downloaded from www.jospt.org by 114.142.169.48 on 07/04/19. For personal use only.

structed subjects. Ernst et all4'reported reduced knee pacity of the hamstrings to control anterior tibia1
extensor moments during take-off and landing shear during landing.
phases of a single leg vertical jump, as well as re- Gauffin and Tropp16 provide some information
duced summated lower extremity extensor moments about biomechanical and neuromuscular compensa-
during the landing phase of the vertical jump in sub- tory strategies for maintaining knee stability during
jects who underwent ACL reconstruction compared hop performance in subjects who apparently were
to matched control subjects without ligament injury. functioning well despite the loss of ACL function.
Ernst et all4 speculated that the differences in knee However, this study does not provide us with a com-
extensor moments could be explained by alterations parison of how biomechanical and neuromuscular
in muscle activity patterns; however, no lower extrem- control strategies may differ between those who can
J Orthop Sports Phys Ther 2001.31:588-597.

ity electromyography (EMG) data was provided to and cannot maintain knee stability during a challeng-
support this notion. Although the results of both ing activity such as the single leg hop. In contrast,
studies indicate that there are differences in the ki- Rudolph et al" compared lower extremity kinematic,
netic profiles of patients with ACL injury or those kinetic, and EMG activity during the single leg hop
who have undergone ACL reconstruction, the lack of in noninjured subjects with ACMeficient knees who
kinematic and EMG data makes it difficult to relate were successful in returning to high level physical ac-
these findings to mechanisms responsible for these tivity (classified as "copers") and subjects who were
differences. unsuccessful in returning to high level physical activi-
Other investigators have used a more comprehen- ty (classified as "noncopers"). The goal of their
sive approach, including lower extremity kinematic, study was to differentiate between successful and un-
kinetic, and EMG data, to describe the biomechani- successful compensatory strategies for maintaining
cal and neuromuscular profiles of subjects with ACL dynamic knee stability.

TABLE 3. Reported correlations between hop test scores and measurements of knee joint position sense.
Citation Sample Knee joint position sense measurement Correlations with hop tests
Carter et a19 Subjects with anterior cruciate liga- Reproduction of passive knee joint po- Reported no significant correlation with sin-
ment deficiency (N = 50) sition gle hop or figure-of-eight run; correlation
coefficients not provided
Borsa et a17 Subjects with anterior cruciate liga- Threshold detection of passive joint Single hop = -0.56-(-0.46)
ment deficiency (N = 29) movement

J Orthop Sports Phys Ther-Volume 51 .Number 10.0ctober 2001 59 1

 TABLE 4. Reported correlations between hop test scores and self-report measures of function.
Citation Sample Self-report measurement Correlations with hop tests
Noyes et alZi Subjects with anterior cruciate liga- Cincinnati Knee Scale (component Single hop = 0.03-0.28
ment deficiency (N= 67) scores)
Wilk et aIJb Subjects with anterior cruciate liga- Cincinnati Knee Scale (total score) Single hop = 0.48
ment reconstruction( N = 50) Crossover hop = 0.38
Timed hop = 0.31
Borsa et alb Subjects with anterior cruciate liga- Cincinnati Knee Scale (total score) Single hop = -0.1 1
ment deficiency (N= 29) Lysholm Knee Scale Single hop = 0.02
Sernert et all1 Subjects with anterior cruciate liga- International Knee Single hop = 0.28
ment reconstruction( N = 527) Documentation Committee
Evaluation System
Tegner Activity Scale Single hop = 0.25
Lvsholm Knee Scale Single hop = 0.36

Rudolph et alS1reported very little difference in cludes the analysis of kinematic, kinetic, and EMG
kinematic, kinetic, and EMG variables between c o p data to thoroughly describe the biomechanical and
ers and noninjured subjects. Their data suggested neuromuscular profiles of the stable and unstable
that the copers had a tendency to use more contri- knee. It is hoped that continued work in this area
bution from the ankle and less contribution from the would provide us with the ability to differentiate be-
hip in maintaining the support moment of both low- tween appropriate and inappropriate mechanisms for
er extremities when compared to control subjects. maintaining dynamic knee stability and that such in-
There were no differences in vertical ground reac- formation will result in refinement of clinical deci-
tion forces between the copers and control subjects. sion making and treatment planning.
In contrast, noncopers appeared to hop differently
from copers and control subjects in that they demon-
strated less knee flexion during ground contact, had HOP TESTS AS CLINICAL PREDICTORS OF
Downloaded from www.jospt.org by 114.142.169.48 on 07/04/19. For personal use only.

lower external knee moments, and had lower peak DYNAMIC KNEE STABILITY
vertical ground reaction forces. Noncopers also had
greater contributions from the hip and lesser contri- Predictive Versus Discriminative Characteristics of
butions from the knee to the support moment dur- Hop Tests
ing the transition from weight acceptance to hop
propulsion. In order for a test procedure to have predictive va-
It is not possible to make definitive conclusions at lidity, the measurements obtained from the test
this time regarding the biomechanical and neuro- should allow the examiner to predict conditions or
muscular control mechanisms involved in maintain- events that are likely to occur in the future with a
ing dynamic knee stability during hopping tasks. reasonable degree of accuracy.%'If we are to accept
J Orthop Sports Phys Ther 2001.31:588-597.

However, the studies cited above provide some inter- hop tests as valid measures to predict dynamic knee
esting insights. It may be that asymmetries in vertical stability, then we must have evidence that perfor-
ground reaction forces can be used to determine mance on hop tests at a given point in time will al-
whether or not individuals are adequately controlling low us to determine who will demonstrate adequate
knee stability during physically demanding tasks. Ob- dynamic knee stability and who will be at risk for ex-
serving that individuals use limited excursions of hip periencing episodes of knee instability in the future.
and knee flexion during hop testing may indicate The predictive ability of a test should not be con-
that they have not developed appropriate compensa- fused with the discriminative ability of a test. Where-
tory mechanisms for maintaining knee stability. as the predictive ability of a test implies that we can
There may be a need to design treatment programs predict future events based on measurements taken
that emphasize the role of hip and ankle neuromus- at a previous point in time, the discriminative ability
cular control strategies to assist in stabilizing the implies that we are able to discriminate, at a given
ACLcompromised knee. The studies cited above point in time, between groups of subjects who may
demonstrate the potential utility of hop test proce- or may not have some target condition (eg, dynamic
dures in differentiating between patients who are knee instability), based on the test results. The ability
able or unable to dynamically stabilize the knee, as of a test to discriminate between individuals with a
well as the potential use of hop tests as a method to given condition at a single point in time does not
explore biomechanical and neuromuscular control necessarily guarantee that the test will also have the
mechanisms in maintaining dynamic knee stability. ability to predict who will acquire the condition in
Gauffin and Tropp"' and Rudolph et a13' underscore the future.
the need to use a comprehensive assessment that in- A number of studies have examined the discrimi-

J Orthop Sports Phys Ther.Volume 31 .Number 10.0ctober 2001

 native ability of hop tests to identify individuals with
knee instability. As mentioned earlier, some studies
compared hop symmetry indices between nonim-
paired control subjects and individuals with ACMefi-
cient k n e e ~ . ~ ' ) 2Eastlack
" et allJ reported the useful-
u Phase

ness of the crossover hop test used in conjunction

with other measurement variables to discriminate be-
tween subjects with chronic ACL deficiency who
Return to
could not functionally compensate well for their inju-
ries (noncopers) and those who were known to suc-
cessfully return to premorbid levels of physical activi-

/ \
w
ty (copers). Hop test measurements in these studies Classification
were obtained at a set period of time and indicate

Q
that hop test scores provide some degree of ability to
determine who has a problem with knee instability at Successful Failed
the time of the evaluation. None of these studies,
however, provide evidence that the hop tests can ac-
tually predict which patients will have continued dif- Calculate Calculate
ficulty with dynamic knee instability in the future. Specificity Sensitivity
There is some evidence that single leg hop tests
may show promise as a tool to predict whether pa-
tients with ACMeficient knees can return to high
level physical activity following nonoperative rehabili-
tation without experiencing continued episodes of
knee instability. Fitzgerald et all5 described a deci- Operator
Characteristic Curve
sion-making scheme using a criterion-based selection
process to identify patients with ACLdeficient knees
Downloaded from www.jospt.org by 114.142.169.48 on 07/04/19. For personal use only.

who may be likely to successfully return to high level

physical activity following nonoperative rehabilitation,.
The selection criteria were based on self-report mea- Identify Cut-Point
For Predicting
sures of function and knee instability, as well as per- Success Versus
formance on a timed &meter single leg hop test. Failure
Twenty-eight subjects who met the criteria and at-
tempted to return to high level activity following re- FIGURE 1. A conceptual framework of an experimental design to examine
hop tests as clinical predictors of dynamic knee stability.
habilitation were followed over a &month period.
After this period, subjects were classified as success-
ful if they had returned to premorbid levels of activi-
some ability to predict future success in returning to
J Orthop Sports Phys Ther 2001.31:588-597.

ty without experiencing an episode of recurrent knee

high level activity with nonoperative rehabilitation in
instability (giving way at the knee). Subjects were
patients with ACL rupture. Further work, using larg-
considered to have failed rehabilitation if they expe-
er sample sizes and prospective experimental designs,
rienced at least l episode of recurrent instability on
is needed to clarify the role of hop tests as tools to
return to premorbid levels of activity. Fitzgerald et
predict future difficulties with dynamic knee stability.
all" compared pretraining hop test symmetry scores
between subjects who were successful and subjects
who ultimately failed rehabilitation at the &month A Conceptual Framework for Designing Future
follow-up. This comparison indicated that those who Studies To Determine the Use of Hop Tests as Clinical
had failed rehabilitation had significantly lower pre- Predictors of Dynamic Knee Stability
training timed hop test symmetry scores (85.4 + 4.1,
n = 6) than those who were successful (95.1 6.6, + Figure 1 illustrates a conceptual framework for an
n = 22). experimental design to examine hop tests as clinical
The relatively small number of subjects and the predictors of dynamic knee stability. For this design,
disparity in sample size between groups make it diffi- we will assume that all subjects wish to return to
cult to draw definitive conclusions regarding the abil- physically demanding levels of activity that would re-
ity of the timed hop test to predict the likelihood of quire a high degree of dynamic knee stability, such
future problems with recurrent knee instability, as jumping, landing, twisting, and cutting activities.
based on the data of Fitzgerald et all5 However, this The first step in this approach would be to obtain
study does demonstrate that hop test measurements hop test measurements prospectively, before subjects
taken early in the rehabilitation process may have attempted a return to premorbid levels of activity.

J Orthop Sports Phys Ther.Volume 31 -Number 10-October 2001

 Because hop tests are a relatively demanding type of
physical performance test, subjects would likely un-
dergo rehabilitation prior to performing the hop
tests. Subjects should meet specific clinical mile-
stones before the hop tests are administered to en-
sure that these tests can be performed safely. For ex-
ample, the criteria we use to determine when hop
tests can be administered in our clinic are: (1) full
knee motion has been obtained, (2) no extensor lag
on straight leg raising, (3) no joint effusion present,
(4) quadriceps strength on the involved limb is
equivalent to 80% of the uninvolved limb, and (5)
the patient can tolerate hopping on the involved 1 -SpeclRclty (falseposiiive rate)

limb without pain. When subjects meet these criteria, FIGURE 2. A hypothetical receiver operator characteristic curve to deter-
the hop tests are administered. mine an optimal cut-point of the hop symmetry index for predicting dy-
The next phase of the experiment would be a re- namic knee stability.
turn-to-activity phase. During this phase, subjects * indicates the point on the curve representing the optimal hypothetical
cut-point.
would be given the opportunity to return to premor-
bid levels of physical activity. An important considera-
tion in this phase is to ensure that subjects have ved hop test scores) in identifying successful versus
been given enough time to make the transition to unsuccessful subjects. For example, when using a
full premorbid levels of activity. This time frame may hop test symmetry index of 60%,we would say the
vary depending on the population being studied. For test is positive in identifying subjects who will be like-
example, patients attempting a nonoperative return ly to fail if the index is less than 60% and negative if
to activity may not require as much time as those their index was greater than 60%.The sensitivity for
who are attempting to return to activity following the hop test symmetry index would be the propor-
surgical treatment of their injuries. The point is that tion of subjects with an index below 60%who actual-
Downloaded from www.jospt.org by 114.142.169.48 on 07/04/19. For personal use only.

the investigators should define a reasonable time pe- ly failed. The specificity would be the proportion of
riod for the return-to-activity phase that would allow subjects who succeeded whose index was greater
them to determine whether or not subjects in the than 60%.The sensitivity and specificity would be
study successfully returned to premorbid levels of ac- calculated for each hop test symmetry index.
tivity. Once the sensitivity and specificity have been cal-
At the end of the return-to-activity phase, subjects culated for each hop test symmetry index, the cut-
would be classified as either having succeeded or point for the index that would best predict who will
having failed a return to premorbid activity. An oper- fail and who will succeed can be determined by gen-
ational definition of success versus failure must be es- erating a receiver operating characteristic curve."
tablished. For patients attempting a nonoperative re- This curve is generated by plotting the sensitivity on
J Orthop Sports Phys Ther 2001.31:588-597.

turn to activity after ligament injury, an example of the y-axis of the graph and 1 - the specificity (the
success might be the ability to return to premorbid false-positive rate) on the x-axis of the graph for
levels of physical activity over a specified time period each hop test symmetry index. The best hop test
without experiencing an episode of giving way at the symmetry index for predicting failure and success
knee. For patients who are returning to activity after would be the index with the best combination of
postoperative rehabilitation, the frequency of giving higher sensitivity with the lower false-positive rate.
way may be low, but success versus failure may be This corresponds to the point closest to the upper
judged based on the degree to which they return to left corner of the graph. Figure 2 is a schematic dia-
premorbid levels of activity over a specified time pe- gram illustrating a hypothetical best cut-point for the
riod. hop symmetry index.
Following classification of subjects into groups, the The above example is, admittedly, a simple expla-
postrehabilitation hop test measurements taken prior nation of designing a study that determines the abili-
to attempting a return to activity would be analyzed ty of hop tests to predict dynamic knee stability, and
to determine the degree to which these measure- execution of this type of design is easier said than
ments were able to predict which subjects were suc- done. Key elements of the design, such as assuring
cessful and which subjects failed at returning to pre- that subjects receive similar rehabilitation programs,
morbid activity levels. We will use hop test symmetry obtaining hop test measurements on all subjects pro-
indices to illustrate this analysis. spectively before subjects return to activity, determin-
The first step in the analysis is to determine the ing an adequate time period for the return-to-activity
sensitivity and specificity of various hop test symme- phase, and operationally defining success and failure
try indices (eg, 60%, 70%,80% of involved/uninvol- to return to activity, may be difficult to address in

J Orthop Sports Phys Ther-Volume 31 .Number 10.0ctober 2001

 some circumstances. However, we believe that this ing subjects who are functionally stable and return to
type of experimental design would be appropriate pre-injury performance levels, not simply those who
for providing evidence that hop tests can be used as return to the pre-injury activity i t ~ e l f . ~ ~ ~
a clinical measure to predict future problems with
dynamic knee stability. Issues Related to Sex Differences

CONSIDERATIONS FOR FUTURE RESEARCH Recently, several studies have revealed sex-specific
differences in both knee injury frequency and mech-
It is difficult to conclude that there are any defini- anisms contributing to neuromuscular control of
tive neuromuscular patterns consistently demonstrat- knee f u n c t i ~ n . ~It~is. critical
~ - ~ ~ that future research
ed during hop tests in subjects who have ACL defi- focus on sex-specific performance of landing and
ciencies or have undergone ACL reconstruction. hopping to distinguish unique characteristics contrib-
Much of the conflicting data available can be attrib- uting to functional stability in men and women. Oth-
uted to methodological inconsistencies among stud- er specific performance characteristics that need to
ies and the lack of comprehensive neuromuscular be considered include standardization of single leg
and biomechanical approaches to assessment of hop hop test procedures relative to landing techniques
test performance. These inconsistent and conflicting (stiff landing or soft landing), foot position, and hop
data clearly warrant more focused research to ascer- height and distance. These potentially confounding
tain neuromuscular control mechanisms promoting variables may influence strategies employed to
functional stability and successful performance of achieve functional stability and mask compensatory
hop tests and hopping-like activities. mechanisms critical to function.
Most importantly, however, for a true appreciation
lssues Related to Neuromuscular Control of movement strategies contributing to functional sta-
bility and superior hop test performance, the clini-
One of the first tasks for researchers in this area is cian needs data establishing the underlying mecha-
to more clearly differentiate between those individu- nisms that are compensatory for the ligamentdefi-
als with ACMeficient or ACLreconstructed knees cient and the ligament-reconstructed knee. This in-
Downloaded from www.jospt.org by 114.142.169.48 on 07/04/19. For personal use only.

who have sufficient neuromuscular control and bio- formation can only be made available if researchers
mechanical efficiency to promote functional stability strive to obtain a comprehensive assessment of the
and those who d o not. The process by which this neuromuscular and biomechanical factors that result
functional differentiation has been made is inconsis- in functional joint stability. Study methodologies
tent and, in most instances, has failed to consider all must include the synchronous analysis of muscle acti-
aspects of the compensatory nature of landing tasks vation patterns, joint kinematics, and kinetic vari-
and hop performance. Once these differentiating ables acting on the system while performing landing
mechanisms are determined, the utility of the hop tasks. To date, there are good studies demonstrating
test will be such that it will provide clinical objectives that the knee joint musculature plays an important
to guide the rehabilitation process. The concepts of role in normalizing knee joint stability. However,
J Orthop Sports Phys Ther 2001.31:588-597.

copers and noncopers put forth by Rudolph et al"."' without data defining hip and ankle joint kinematics
are a significant stride to this end. and the corresponding mechanism contributing to
However, the defining elements of copers and all joint movement patterns, we are left with describ-
noncopers still have shortcomings that need to be re- ing inconclusive compensatory mechanisms in these
fined. The differentiating variable employed in these patients. A comprehensive approach to assessment is
studies was the ability to return to all cutting and piv- critical to enhance the utility of clinically employed
oting sports and whether ACL reconstruction was hop tests.
elected. Although these elements are critical to the
definition of functional stability, there are considera- lssues Related to Comparison Groups
tions preventing clear delineation among subjects.
For example, some individuals may alter speed or Further complicating the clinical usefulness of the
movement strategies to maintain knee stability on re- hop test is the selection of a control for comparison
turn to cutting and pivoting sports; however, the al- of the ACMeficient or ACL-reconstructed knee. The
terations may compromise the ability to engage in ideal criterion standard clearly would be to have pre-
these activities at pre-injury levels of performance. In injury data and, thus, a prospective approach to eval-
contrast, others may acquire alterations in movement uation of functional level of the ACLinjured limb,
patterns that do not compromise their performance but this is often not practical. Therefore, clinicians
levels, and they can achieve a complete return to and researchers are faced with making bilateral com-
pre-injury levels of function. Thus, performance- parisons and the assumption that the uninvolved
based criteria that assess variables such as speed and limb represents the normal pre-injury status of the
precision of movement are necessary for distinguish- ACLinjured limb, or attempts to compare the per-

J Orthop Spow Phys Ther.Volume 31 .Number 10.0ctober 2001

 formance to a matched individual o r control group. 3. Barber SD, Noyes FR, Mangine RE, McCloskey JW, Hart-
Most agree that, from a neuromuscular and biome- man W. Quantitative assessment of functional limitations
in normal and anterior cruciate ligament-deficient knees.
chanical standpoint, there are limitations to using
Clin Orthop. 1990;255:204-2 14.
the contralateral l i m b for comparison because the 4. Blackburn JR, Morrissey MC. The relationship between
nonimpaired l i m b may undergo compensatory a d a p open and closed kinetic chain strength of the lower limb
tations for the insufficiency. Thus, most studies at- and jumping performance. ) Orthop Sports Phys Ther.
tempt t o compare the performance outcome to 1998;27:430-435.
some matched nonimpaired individual o r control 5. Bolgla LA, Keskula DR. Reliability of lower extremity
functional performance tests. J Orthop Sports Phys Ther.
group. 1997;26:13&142.
By its nature, control group matching poses some 6. Borsa PA, Lephart SM, lrrgang JJ. Comparison of perfor-
validity concerns that make the selection process cru- mance-based and patient-reported measures of function
cial. Given the variability o f sex, age, somatotype, and in anterior cruciate ligament-deficient individuals. J Or-
sport specificity, the task o f matching is subjective at thop Sports Phys Ther. 1998;28:392-399.
7. Borsa PA, Lephart SM, lrrgang JJ, Safran MR, Fu FH. The
best. A better alternative to subject matching seems
effects of joint position and direction of joint motion on
t o be matching to a large sample size o f a homoge- proprioceptive sensibility in anterior cruciate ligament-
nous group taking into consideration variables such deficient athletes. Am J Sports Med. 1997;25:336-340.
as sex, age, somatotype, and sport o r occupational 8. Brosky JAJr, Nitz AJ, Malone TR, Caborn DN, Rayens MK.
activity. T o date, there is limited information that lntrarater reliability of selected clinical outcome measures
would allow for a differentiation between what could following anterior cruciate ligament reconstruction. J Or-
thop Sports Phys Ther. 1999;29:39-48.
be considered normal performance versus pathologi-
9. Carter ND, Jenkinson TR, Wilson D, Jones DW, Torode
cal performance within the context o f these vari- AS. Joint position sense and rehabilitation in the anterior
ables, and this seems t o be a necessary starting point cruciate ligament deficient knee. Br J Sports Med. 1997;
for research t o identify comprehensive neuromuscu- 31 :2O9-212.
lar and biomechanical mechanisms mediating perfor- 10. Cerullo JF, Riemann BL, McCaw ST, Lephart SM. Com-
mance o f hopping and landing tasks. Once norma- parison of vertical ground reaction forces between males
and females during drop landings [abstract]. Med Sci
tive databases are established, the performance crite- Sports Exerc. 2000;32(suppI):268.
ria will be in place t o evaluate restoration o f normal 11. Colby SM, Hintermeister RA, Torry MR, Steadman JR.
Downloaded from www.jospt.org by 114.142.169.48 on 07/04/19. For personal use only.

performance and describe those pathological profiles Lower limb stability with ACL impairment. J Orthop
that include adaptive mechanisms contributing t o Sports Phys Ther. 1999;29:444-454.
functional stability in subjects with A C M e f i c i e n t and 12. Daniel D, Malcolm L, Stone ML, Perth H, Morgan J, Riehl
B. Quantification of knee instability and function. Con-
ACL-reconstructed knees.
temporary Orthopaedics. 1982;5:83-91.
13. Eastlack ME, Axe MJ, Snyder-Mackler L. Laxity, instability,
CONCLUSION and functional outcome after ACL injury: copers versus
noncopers. Med Sci Sports Exerc. 1999;3 1:2 10-2 15.
14. Ernst GP, Saliba E, Diduch DR, Hurwitz SR, Ball DW.
We believe that h o p test procedures may show
Lower extremity compensations following anterior cruci-
promise as clinical tools to predict dynamic knee sta- ate ligament reconstruction. Phys Ther. 2000;80:251-
bility and may also be useful as research tools to 260.
J Orthop Sports Phys Ther 2001.31:588-597.

broaden our understanding o f neuromuscular con- 15. Fitzgerald GK, Axe MJ, Snyder-Mackler L. A decision-
trol mechanisms required t o maintain dynamic knee making scheme for returning patients to high-level activ-
stability following injury. We have indicated through- ity with nonoperative treatment after anterior cruciate lig-
out this commentary, however, that a considerable ament rupture. Knee Surg Sports Traumatol Arthrosc.
2000;8:76-82.
amount o f information still needs t o be acquired to 16. Gauffin H, Tropp H. Altered movement and muscular-ac-
clearly delineate the role o f h o p tests as predictors o f tivation patterns during the one-legged jump in patients
dynamic knee stability. We hope that o u r commen- with an old anterior cruciate ligament rupture. Am J
tary will foster further work in this area and will Sports Med. 1992;20:182-192.
serve, in part, as a basis for designing future studies. 17. Greenberger HB, Paterno MV. Relationship of knee exten-
sor strength and hopping test performance in the assess-
ment of lower extremity function. J Orthop Sports Phys
Ther. 1995;22:202-206.
REFERENCES 18. Harner CD, Marks PH, Fu FH, lrrgang JJ, Silby MB, Men-
gato R. Anterior cruciate ligament reconstruction: endo-
Anderson C, Odensten M, Gillquist J. Knee function after scopic versus two-incision technique. Arthroscopy. 1994;
surgical or nonsurgical treatment of acute rupture of the 10:502-512.
anterior cruciate ligament: a randomized study with a 19. Howell SM, Deutsch ML. Comparison of endoscopic and
long-term follow-up period. Clin Orthop. 1991;264:255- two-incision techniques for reconstructing a torn anterior
263. cruciate ligament using hamstring tendons. Arthroscopy.
Bandy WD, Rusche KR, Tekulve FY. Reliability and sym- 1999;15:594-606.
metry for five unilateral functional tests of the lower ex- 20. ltoh H, Kurosaka M, Yoshiya S, lchihashi N, Mizuno K.
tremity. lsokinetics and Exercise Science. 1994;4:108- Evaluation of functional deficits determined by four dif-
111. ferent hop tests in patients with anterior cruciate ligament

J Orthop Sports Phys Ther-Volume 31 .Number 10.0ctober 2001

 deficiency. Knee Surg Sports Traumatol Arthrosc. 1998;6: 29. Rothstein JM, Echternach JL. Reliability and validity. In:
241-245. Rothstein JM, Echternack JL, eds. Primer on Measurement:
21. Jaeschke R, Guyatt G, Sackett DL. User's guides to the An introductory Guide to Measurement Issues. Alexan-
medical literature. Ill. How to use an article about a di- dria, Va: APTA Publications; 1993:59-95.
agnostic test: B. What are the results and will they help 30. Rozzi SL, Lephart SM, Gear WS, Fu FH. Knee joint laxity
me in caring for my patients?The Evidence-Based Medi- and neuromuscular characteristics of male and female
cine Working Group. JAMA. 1994;271:703-707. soccer and basketball players. Am] Sports Med. 1999;27:
22. Juris PM, Phillips EM, Dalpe C, Edwards C, Gotlin RS, 312-319.
Kane DJ. A dynamic test of lower extremity function fol- 31. Rudolph KS, Axe MJ, Snyder-Mackler L. Dynamic stability
lowing anterior cruciate ligament reconstruction and re- after ACL injury: who can hop? Knee Surg Sports Trau-
habilitation./ Orthop Sports Phys Ther. 1997;26:184-191. mat01 Arthrosc. 2000;8:262-269.
23. Lephart SM, Perrin DH, Fu FH, Gieck J, McCue FC. Re- 32. Rudolph KS, Eastlack ME, Axe MJ, Snyder-Mackler L.
lationship between selected physical characteristics and 1998 Basmajian Student Award Paper: Movement patterns
functional capacity in the ACL-deficient athlete. / Orthop after anterior cruciate ligament injury: a comparison of
Sports Phys Ther. 1992;16:174-181. patients who compensate well for the injury and those
24. Lephart SM, Perrin DH, Minger K, Fu FH. Functional per- who require operative stabilization. / Electromyogr Kine-
formance tests for the anterior cruciate ligament-insuffi- siol. 1998;8:349-362.
cient athlete. / Athletic Training. 1991;26:44-90. 33. Sernert N, Kartus J, Kohler K, et al. Analysis of subjective,
25. Noyes FR, Barber SD, Mangine RE. Abnormal lower limb objective, and functional examination tests after anterior
symmetry determined by function hop tests after anterior cruciate ligament reconstruction. A follow-up of 527 pa-
cruciate ligament rupture. Am / Sports Med. 1991;19: tients. Knee Surg Sports Traumatol Arthrosc. 1999;7:160-
513-518.
165.
26. Petschnig R, Baron R, Albrecht M. The relationship be-
tween isokinetic quadriceps strength test and hop tests for 34. Swanik CB, Lephart SM, Giraldo JL, DeMont RG, Fu FH.
distance and one-legged vertical jump test following an- Reactive muscle firing of anterior cruciate ligament-in-
terior cruciate ligament reconstruction. / Orthop Sports jured females during functional activities. / Athletic Train-
Phys Ther. 1998;28:23-3 1. ing. 1999;34:121-129.
27. Pincivero DM, Lephart SM, Karunakara RG. Relation be- 35. Tegner Y, Lysholm J, Lysholm M, Gillquist J. A perfor-
tween open and closed kinematic chain assessment of mance test to monitor rehabilitation and evaluate anterior
knee strength and functional performance. Clin I Sport cruciate ligament injuries. Am / Sports Med. 1986;14:
Med. 1997;7:11-16. 156-1 59.
28. Risberg MA, Holm I, Tjomsland 0, Ljunggren E, Ekeland 36. Wilk KE, Romaniello WT, Soscia SM, Arrigo CA, Andrews
A. Prospective study of changes in impairments and dis- JR. The relationship between subjective knee scores, iso-
Downloaded from www.jospt.org by 114.142.169.48 on 07/04/19. For personal use only.

abilities after anterior cruciate ligament reconstruction. / kinetic testing, and functional testing in the ACL-recon-
Orthop Sports Phys Ther. 1999;29:400-412. structed knee. / Orthop Sports Phys Ther. 1994;20:60-73.
J Orthop Sports Phys Ther 2001.31:588-597.

J Orthop Sports Phys Ther.Volume 31 .Number 10.0ctober 2001

 This article has been cited by:

1. Allison R. Toole, Matthew P. Ithurburn, Mitchell J. Rauh, Timothy E. Hewett, Mark V. Paterno, Laura C. Schmitt. 2017.
Young Athletes Cleared for Sports Participation After Anterior Cruciate Ligament Reconstruction: How Many Actually Meet
Recommended Return-to-Sport Criterion Cutoffs?. Journal of Orthopaedic & Sports Physical Therapy 47:11, 825-833. [Abstract]
[Full Text] [PDF] [PDF Plus]
2. Axe Michael J., Snyder-Mackler Lynn, Grieder Anna Shovestul, Miller Jeffrey, Dreger Melissa, Palmer Michael, Manal Tara Jo.
Knee 1-58. [Abstract] [Full Text] [PDF]
3. Manal Tara Jo, Grieder Anna Shovestul, Kist Bryan W.. The Knee: Physical Therapy Patient Management Using Current Evidence
1-83. [Abstract] [Full Text] [PDF]
4. Amir Curcio dos Reis, João Carlos Ferrari Correa, André Serra Bley, Nayra Deise dos Anjos Rabelo, Thiago Yukio Fukuda, Paulo
Roberto Garcia Lucareli. 2015. Kinematic and Kinetic Analysis of the Single-Leg Triple Hop Test in Women With and Without
Patellofemoral Pain. Journal of Orthopaedic & Sports Physical Therapy 45:10, 799-807. [Abstract] [Full Text] [PDF] [PDF Plus]
[Supplemental Material]
5. David Logerstedt, Stephanie Di Stasi, Hege Grindem, Andrew Lynch, Ingrid Eitzen, Lars Engebretsen, May Arna Risberg,
Michael J. Axe, Lynn Snyder-Mackler. 2014. Self-Reported Knee Function Can Identify Athletes Who Fail Return-to-Activity
Criteria up to 1 Year After Anterior Cruciate Ligament Reconstruction: A Delaware-Oslo ACL Cohort Study. Journal of
Orthopaedic & Sports Physical Therapy 44:12, 914-923. [Abstract] [Full Text] [PDF] [PDF Plus]
6. Laura C. Schmitt, Mark V. Paterno, Timothy E. Hewett. 2012. The Impact of Quadriceps Femoris Strength Asymmetry on
Functional Performance at Return to Sport Following Anterior Cruciate Ligament Reconstruction. Journal of Orthopaedic & Sports
Physical Therapy 42:9, 750-759. [Abstract] [Full Text] [PDF] [PDF Plus]
7. Douglas Adams, David Logerstedt, Airelle Hunter-Giordano, Michael J. Axe, Lynn Snyder-Mackler. 2012. Current Concepts
for Anterior Cruciate Ligament Reconstruction: A Criterion-Based Rehabilitation Progression. Journal of Orthopaedic & Sports
Physical Therapy 42:7, 601-614. [Abstract] [Full Text] [PDF] [PDF Plus] [Supplemental Material]
Downloaded from www.jospt.org by 114.142.169.48 on 07/04/19. For personal use only.

8. Mark Verstegen, Susan Falsone, Russell Orr, Steve Smith. 2012. Suggestions From the Field for Return to Sports Participation
Following Anterior Cruciate Ligament Reconstruction: American Football. Journal of Orthopaedic & Sports Physical Therapy 42:4,
337-344. [Abstract] [Full Text] [PDF] [PDF Plus] [Supplemental Material]
9. Kai Mithoefer, Karen Hambly, David Logerstedt, Margherita Ricci, Holly Silvers, Stefano Della Villa. 2012. Current Concepts for
Rehabilitation and Return to Sport After Knee Articular Cartilage Repair in the Athlete. Journal of Orthopaedic & Sports Physical
Therapy 42:3, 254-273. [Abstract] [Full Text] [PDF] [PDF Plus]
10. Nyland John. Therapeutic Strategies for Developing Neuromuscular Control in the Kinetic Chain . [Citation] [Full Text]
11. Mark V. Paterno, Laura C. Schmitt, Kevin R. Ford, Mitchell J. Rauh, Gregory D. Myer, Timothy E. Hewett. 2011. Effects of
Sex on Compensatory Landing Strategies Upon Return to Sport After Anterior Cruciate Ligament Reconstruction. Journal of
J Orthop Sports Phys Ther 2001.31:588-597.

Orthopaedic & Sports Physical Therapy 41:8, 553-559. [Abstract] [Full Text] [PDF] [PDF Plus]
12. David S. Logerstedt, Lynn Snyder-Mackler, Richard C. Ritter, Michael J. Axe, Joseph Godges. 2010. Knee Pain and Mobility
Impairments: Meniscal and Articular Cartilage Lesions. Journal of Orthopaedic & Sports Physical Therapy 40:6, A1-597. [Abstract]
[Full Text] [PDF] [PDF Plus]
13. David S. Logerstedt, Lynn Snyder-Mackler, Richard C. Ritter, Michael J. Axe, Joseph J. Godges. 2010. Knee Stability and
Movement Coordination Impairments: Knee Ligament Sprain. Journal of Orthopaedic & Sports Physical Therapy 40:4, A1-A37.
[Abstract] [Full Text] [PDF] [PDF Plus]
14. Hunter-Giordano Airelle O., Burlovich Erin, Manal Tara Jo. Rehabilitation Following Anterior Cruciate Ligament Reconstruction
1-36. [Abstract] [Full Text] [PDF]
15. Gregory D. Myer, Mark V. Paterno, Kevin R. Ford, Carmen E. Quatman, Timothy E. Hewett. 2006. Rehabilitation After Anterior
Cruciate Ligament Reconstruction: Criteria-Based Progression Through the Return-to-Sport Phase. Journal of Orthopaedic &
Sports Physical Therapy 36:6, 385-402. [Abstract] [PDF] [PDF Plus]