Am J Sports Med 2001 Kvist 72 82

8/17/2019 Am J Sports Med 2001 Kvist 72 82

1/12

http://ajs.sagepub.com/ Medicine

The American Journal of Sports

http://ajs.sagepub.com/content/29/1/72The online version of this article can be found at:

2001 29: 72Am J Sports Med

Joanna Kvist and Jan GillquistChain Exercises in Anterior Cruciate Ligament-Deficient Patients and Control Subjects

Sagittal Plane Knee Translation and Electromyographic Activity During Closed and Open Kinetic

Published by:

http://www.sagepublications.com

On behalf of:

American Orthopaedic Society for Sports Medicine

can be found at:The American Journal of Sports Medicine Additional services and information for

http://ajs.sagepub.com/cgi/alertsEmail Alerts:

http://ajs.sagepub.com/subscriptionsSubscriptions:

http://www.sagepub.com/journalsReprints.navReprints:

http://www.sagepub.com/journalsPermissions.navPermissions:

at CAPES on May 4, 2011ajs.sagepub.comDownloaded from

http://ajs.sagepub.com/http://ajs.sagepub.com/http://ajs.sagepub.com/content/29/1/72http://ajs.sagepub.com/content/29/1/72http://ajs.sagepub.com/content/29/1/72http://www.sagepublications.com/http://www.sagepublications.com/http://www.aossm.org/http://ajs.sagepub.com/cgi/alertshttp://ajs.sagepub.com/cgi/alertshttp://ajs.sagepub.com/subscriptionshttp://www.sagepub.com/journalsReprints.navhttp://www.sagepub.com/journalsReprints.navhttp://www.sagepub.com/journalsPermissions.navhttp://www.sagepub.com/journalsPermissions.navhttp://ajs.sagepub.com/http://ajs.sagepub.com/http://ajs.sagepub.com/http://www.sagepub.com/journalsPermissions.navhttp://www.sagepub.com/journalsReprints.navhttp://ajs.sagepub.com/subscriptionshttp://ajs.sagepub.com/cgi/alertshttp://www.aossm.org/http://www.sagepublications.com/http://ajs.sagepub.com/content/29/1/72http://ajs.sagepub.com/


2/12

Sagittal Plane Knee Translation and

Electromyographic Activity During Closedand Open Kinetic Chain Exercises inAnterior Cruciate Ligament-DeficientPatients and Control Subjects

Joanna Kvist,*†‡ RPT, PhD, and Jan Gillquist,* MD, PhD

From the Divisions of *Sports Medicine and † Physical Therapy, Department of Neuroscience and Locomotion, Faculty of Health Science, Linköping University, Linköping, Sweden

ABSTRACT

Using electrogoniometry and electromyography, wemeasured tibial translation and muscle activation in 12patients with unilateral anterior cruciate ligament injuryand in 12 control subjects. Measurements were madeduring an active extension exercise with 0-, 4-, and8-kg weights and during squats on two legs and on oneleg where the projection of the center of gravity wasplaced over, behind, and in front the feet. In the unin-

jured subjects, tibial translation increased with increas-ing load except during the squat with the center ofgravity behind the feet, which produced the smallesttranslation. For the active extension exercises, trans-lation was greater during eccentric activity. In the an-terior cruciate ligament-injured knees, all squats re-sulted in similar translation, which was smaller thanthat during the active extension exercise. The highestmuscle activation was seen during squats. Hamstringmuscle activity was low. Increased static laxity in theanterior cruciate ligament-deficient knee can be con-trolled during closed but not during open kinetic chain

exercises. Coactivation of the quadriceps and gastroc-nemius muscles seems to be important for knee sta-bility, whereas hamstring muscle coactivation was in-significant. To minimize sagittal translation duringnonoperative management of anterior cruciate liga-ment-deficient knees, closed kinetic chain exercises

are preferable to open kinetic chain exercises, andimportance should be attached to the spontaneouscoactivation of the quadriceps and gastrocnemiusmuscles.

Closed kinetic chain exercises have become increasinglypopular and strongly recommended for rehabilitation af-

ter an ACL injury because they are believed to be saferthan other exercises.6,12,23,28,30,40 In spite of this, Bynumet al.6 found only minor differences in instrumented kneelaxity measurements between patients who used openkinetic chain and closed kinetic chain exercises after ACLreconstruction, with the closed kinetic chain group having slightly less laxity.

The factors of importance to characterize an exercise asmore or less harmful for the injured knee are the musclecoactivation, shear forces, tibial translation, and ACLstrain associated with it. Coactivation of the lower limbmuscles is supposed to improve stability of the knee joint.2,15,22,28,32,38 It is accepted that closed kinetic chain

exercises lead to more quadriceps/hamstring muscle coac-tivation than do open kinetic chain exercises,23,38 but theresults regarding the amount of coactivation during openkinetic chain exercises are contradictory.9,15,19,27,31,32

Closed kinetic chain exercises seem to produce smalleranterior shear forces in healthy subjects and less anteriortibial translation in ACL-injured knees than do open ki-netic chain exercises.23,38,40 High anterior shear forcesand increased anterior tibial translation could be harmfulto the injured or reconstructed knee because the second-ary restraints or the ligament graft will be stretched. Inthe early phase of rehabilitation after an ACL reconstruc-tion, graft strain should be carefully controlled to promote

‡ Address correspondence and reprint requests to Joanna Kvist, RPT,PhD, Sports Medicine and Physical Therapy, Department of Neuroscience andLocomotion, Faculty of Health Sciences, Linköping University, SE-581 85Linköping, Sweden.

No author or related institution has received any financial benefit fromresearch in this study. See “Acknowledgments” for funding information.

0363-5465/101/2929-0072$02.00/0THE A MERICAN JOURNAL OF SPORTS MEDICINE, Vol. 29, No. 1

© 2001 American Orthopaedic Society for Sports Medicine

72


http://ajs.sagepub.com/http://ajs.sagepub.com/http://ajs.sagepub.com/http://ajs.sagepub.com/


3/12

healing because too little strain does not stimulate healing and too much strain leads to graft failure.39 Beynnon etal.4 have shown in normal knees that the ACL strain issimilar during open and closed kinetic chain exercises, butthere is no information about the effect of these exerciseson ACL-deficient knees. We have previously shown thatanterior tibial translation during an open kinetic chainexercise with maximal effort remains within the limits of translation seen during a 90-N Lachman test, both inuninjured and ACL-injured knees.1,18,19 Therefore, thereis no clear consensus about the effects of closed or openkinetic chain exercises on the ACL-deficient or recon-structed knee.

The purpose of this study was to investigate the tibialtranslation and EMG activity of the quadriceps, ham-string, and gastrocnemius muscles during an open kineticchain exercise (active knee extension) and three differentclosed kinetic chain exercises (squats where the projectionof the center of gravity was placed over, behind, and infront of the feet), in persons with unilateral ACL defi-ciency and uninjured control subjects. The specific aimswere to investigate the maximal tibial translation differ-ences 1) during exercises between injured and uninjuredlegs, 2) during four exercises performed with differentloads, and 3) between the four different exercises. In ad-dition, the tibial translation during the exercises was in-

vestigated in relation to the static laxity defined from the90-N Lachman test and at different flexion angles. Elec-tromyographic analysis was used in the ACL-deficientgroup to help explain differences in tibial translation.

MATERIALS AND METHODS

The study groups included 12 patients (4 women and 8men) with unilateral ACL injury verified by arthroscopicexamination or MRI, and 12 volunteers (6 women and 6men) without any knee joint injury. The mean age of the ACL-deficient patients was 28 years (range, 18 to 38) andthe mean age of the control subjects was 29 years (range,

18 to 35). The individual characteristics of all participantsare shown in Tables 1 and 2.

Measurements

Tibial Translation. A computerized goniometer linkage(CA-4000, OSI Inc., Hayward, California) was used to meas-ure the flexion angle and sagittal tibial translation. Thegoniometer has been described before.1,18,19,24,34–37,40 Thesystem is composed of three parts: the femoral and tibialframes and a rotation module. Three potentiometers in therotation module measure the relative rotations between thefemur and tibia. The potentiometer for sagittal motion,

mounted in the tibial frame, registers the difference in posi-tion between a spring-loaded patellar pad and the fixationpoint on the tibial tuberosity during knee motion. The sag-ittal plane direction is perpendicular to the tibial frame. Thepotentiometer registering knee extension-flexion wasaligned with an approximate knee flexion axis in the centerof the lateral femoral epicondyle and the alignment waschecked repeatedly during the examination. The flexion an-gle from the CA-4000 corresponds well with that measuredwith other equipment (unpublished data from our laborato-

TABLE 1Individual Characteristics of the ACL-Deficient Group

Sex Age Weight

(kg)Height

(cm)Injured

leg Examination

methodaConcominant

injuriesb Surgeryc

Time totest

(months)

Tegner level

Lysholmscore

Lachman (mm)

Preinjury At

exam-ination

ACL-deficient /uninjured

F 22 44 153 L ACI MCL partial, LM 34 3 6 85 11/10

M 24 75 180 L ACI MM MM 22 7 7 60 9/12M 32 88 187 L ACI LM LM 26 9 3 83 7/8M 30 77 181 L ACI 27 7 7 95 9/10M 31 73 170 L ACI 33 9 5 71 7/9M 25 70 183 R ACI LM 35 9 3 59 10/15M 38 61 164 R ACI 24 7 7 95 12/17M 27 87 170 R MRI 20 9 3 100 5/9M 37 81 179 R ACI MM, LM LM 24 9 6 99 7/14F 23 90 173 R MRI 17 9 3 90 8/14F 34 59 160 R ACI MM MM 35 3 3 73 9/13F 18 62 179 R MRI LM 21 9 2 100 8/12

a ACI, arthroscopy; MRI, magnetic resonance imaging.b MCL, medial collateral ligament; MM, medial meniscus; LM, lateral meniscus.c Surgery consisted of partial meniscectomy of the medial or lateral meniscus.

TABLE 2Individual Characteristics of the Control Group

Sex Age Weight

(kg)Length

(cm)Tegner

levelLachman (mm)

Right Left

M 34 70 183 4 8 8M 31 69 185 4 4 7M 28 68 183 4 5 7M 31 84 183 5 8 12F 29 63 175 5 7 9M 35 64 171 3 7 9F 29 58 167 4 11 10F 25 70 176 3 5 8F 23 64 170 4 8 10F 30 61 174 5 8 10F 18 51 160 10 9 12M 29 75 183 4 8 6

Vol. 29, No. 1, 2001 Translation and EMG Activity During Closed and Open Kinetic Chain Exercises 73




4/12

ry). The system was zeroed at the beginning of each test withthe subject lying on the examination table and the kneerelaxed in full extension. Data were sampled from the fourpotentiometers by a computer at a rate of 100 Hz per chan-nel. In this study, only the sagittal plane translation (in

millimeters) and the knee flexion angle (in degrees) wereanalyzed. The accuracy of the measurement system through-out a range of motion has been shown to be good comparedwith fluoroscopy.35

The tibial translation was calculated by subtracting thetibial position values during a passive extension (de-scribed under experimental protocol) from the position values during motion (Fig. 1). During passive extension,the tibia follows the path determined by joint geometry,including the femoral roll back, and falls back to theposterior limit of a passive envelope of knee motion due tothe gravity vector.5

EMG. The EMG activity of the vastus medialis, vastus

lateralis, hamstring, and gastrocnemius lateralis muscleswas recorded by surface electrodes. The muscles werelocated by palpation during a submaximal isometric con-traction, and the most prominent part of the muscle wasused for the placement of the electrodes. The electrodeplacement was verified through functional testing whileobserving the recording on the computer screen. The skin

was first dry shaved and then cleaned with 70% alcohol.On the skin above each muscle, two recording pregelledsilver-chloride electrodes (Blue sensor, M-00-S, Medi-cotest, Denmark; diameter of active part, 10 mm) wereplaced within 2 cm of each other (center-to-center dis-tance) and one ground electrode with amplifier was placedabout 10 cm from the measuring area. Signals were sam-pled at 2000 Hz by an MESPEC 4000 EMG unit system(MEGA Electronics Ltd., Kuopio, Finland). All signalswere amplified and analog-to-digital converted with 12-bitaccuracy in the signal range of 5000 V. An averagedEMG signal was formed by rectifying the raw signal andtaking the average of 12 signal readings taken at 1-msecintervals.

Data from the CA-4000 (knee flexion angle, tibial posi-tion in relation to the patella and femur, and anterior-posterior load from a force handle) and the four EMGchannels were acquired simultaneously in a personal com-puter. The output frequency of all signals was 83 Hz.

Experimental Protocol

The testing protocol was similar in the two groups, exceptthat the EMG measurements were performed on both legsin the ACL-deficient group only. In the control group, theright or left leg was tested first in a random order; in the ACL-deficient group, the uninjured leg was always tested

first.The subjects in the ACL-deficient group began with

three maximum isometric voluntary contractions each forknee extension, knee flexion, and ankle plantar flexion.For knee extension and flexion, the subjects were seatedwith the knee positioned and restrained by a strap at 40°and 130° of knee flexion, respectively. For ankle plantarflexion, the subjects were standing on one leg in an up-right position with balance support and instructed to riseon their toes as high as possible. The mean value during the period of 252 msec with the highest signal values foreach muscle was used to normalize the EMG values dur-ing the exercises.

In all subjects, an instrumented Lachman test was per-formed with the subject strapped to a special seat with theknee flexed to 20°. Tibial translation was recorded when aperpendicular, posterior-anterior directed load of morethan 140 N was applied by a force handle to a point 9 cmbelow the medial joint line. The total translation at 90 Nof force is reported.

To identify a reference line for the calculation of trans-lation during activity, a passive extension motion wasperformed from 100° to 0° of knee flexion with the subjectsitting on an examination table and his or her back sup-ported in a start position of 110° of hip flexion and 90° of knee flexion. The subjects relaxed and, with the heel sup-

Figure 1. Calculation of translation during the active exer-

cises. A, the filled circles represent the position of the tibia

relative to the patella/femur during the active exercise and

the open circles represents the same position during the

passive extension test (one mark for each fifth degree of

knee flexion angle). B, each marker represents the sagittal

tibial translation during the active exercise (subtracted value

of the tibial position during the passive extension from the

position values during active exercise). Negative translation

values imply posterior translation.

74 Kvist and Gillquist American Journal of Sports Medicine




5/12

ported by the examiner, the knee joint was first passivelyflexed to 100° and then extended to full extension, whileavoiding axial rotation of the tibia. The passive extensiontest has a good repeatability. The mean difference be-tween seven repetitions during a 1.5-hour long test ses-sion, with two to four other exercises in between, is 0.06 0.09 mm.17

The subjects then performed four repetitions of the fourexercises in a random order (Fig. 2). In the open kineticchain exercise (active knee extension), the start positionwas the same as for the passive extension test. The exer-cise was performed with 0-, 4-, and 8-kg weights appliedon the distal tibia.

Three closed kinetic chain exercises (squats) were done:1) a squat where the projection of the center of gravity wasapproximately over the feet, 2) a squat with the backresting against a wall and the feet 40 cm from the wall(the center of gravity in this exercise was behind the feet),and 3) a squat with the hands on a wall and the feet 1meter from the wall (the center of gravity in this exercisewas in front the feet). The load during the closed kineticchain exercises was varied by doing the exercises on one orboth legs.

Repeatability

The repeatability of the tibial translation measurementsthroughout a range of motion (0° to 90° to 0° at 5° intervalsfor a total of 37 angles compared) for the squat with centerof gravity over the feet was tested. Six subjects in thecontrol group (three right and three left legs) were ran-domly chosen to do the test on 2 different days. Using analysis of variance for repeated measures, we found no

significant differences between the measurements fromthe 2 days. The mean difference between days throughoutthe range of motion was 0.73 0.41 mm, with a mean 95%confidence interval at the different flexion angles between0.51 and 1.97.

The repeatability of the maximal translation betweenthe four repetitions in the right knee was analyzed in thecontrol group. Using analysis of variance for repeatedmeasures, we found no significant differences. The meandifference in the exercises between the four repetitionswas 0.07 0.08 mm ( N 9), with a mean individual 95%confidence interval between 0.29 and 0.43.

Data Analysis

The mean values from the four repetitions were used infurther calculations. The translation measurements at 20°of knee flexion in all exercises were normalized to thetranslation during a 90-N Lachman test for the same leg.In the ACL-deficient group, the translation at 20° of flex-ion in the injured leg was also normalized to the transla-tion of the uninjured knee during the 90-N Lachman test.

The EMG signals were normalized to signals during themaximum voluntary contraction test for the same leg. TheEMG activity of quadriceps muscle was calculated as themean of the vastus medialis and vastus lateralis muscles.

Statistics

A two-way analysis of variance for repeated measureswith contrast analysis was used for differences in maxi-mal translation and the flexion angle where the maximaltranslation occurred between the uninjured and injuredleg in the ACL-deficient group and between right and leftlegs in the control group. A two-way analysis of variance(2 10) between group design with contrast analysis wasused for differences in maximal translation and the flex-ion angle where the maximal translation occurred be-tween the control group and the uninjured leg in the ACL-deficient group.

Analysis of variance for repeated measures and Stu-dent’s t-test for dependent variables with Bonferroni cor-rection for multiple tests was used for comparisons be-tween the exercise subsets in the ACL-deficient group,regarding maximal translation and translation at 20° of knee flexion.

A three-way analysis of variance (4 2 11), betweenthe 4 exercises, within the 2 legs, and at 11 angles (10°,

20°, 30°, 50°, 70°, 90°, 70°, 50°, 30°, 20°, 10°), and theStudent’s t-test for dependent variables with Bonferronicorrection for multiple tests were used for comparisonbetween the four exercises regarding translation in thedifferent flexion angles in the ACL-deficient group.

The EMG data were analyzed at a range of motionbetween 10° and 90° of knee flexion. Analysis of variancefor repeated measures was done for differences in EMGbetween the legs and between the flexion angles, 10° to45°, 50° to 85°, 85° to 50°, and 45° to 10°, and between thedifferent exercises within the same leg.

Statistics were calculated using commercially availablesoftware (Statistica 4.3, StatSoft, Inc., Tulsa, Oklahoma).

RESULTS

Control Group

Maximal Translation. The translation recorded during the 90-N Lachman test (1.7 1.6 mm) was 19% greater inthe left leg than in the right. For the active exercises, nosignificant difference existed between the two legs. Themean values for the right and left legs were used in fur-ther statistical analysis.

In terms of the different loads used, in the open kineticchain exercise, the maximal translation increased 16%

between the 0-kg and the 4-kg load and 11% from the 4-kg to 8-kg load ( P 0.005). In the closed kinetic chain exer-cises, for squats with the center of gravity over the feet orin front of the feet, the increase in translation comparedwith that during active extension was similar (17% and22%, respectively), but there was no change in translationduring the squat with the center of gravity behind the feet(Table 3 and Fig. 3).

When comparing the different exercises at maximalload, the squat with the center of gravity over or in frontof the feet resulted in the same translation as the activeextension exercise, whereas the squat with the center of gravity behind the feet produced a 24% smaller transla-





6/12

tion ( P 0.005) than the other exercises (Table 3, Fig. 4).The squat with the center of gravity in front the feetproduced the largest translation of the closed chain exer-

cises ( P 0.005). Normalized Translation. All the exercise subsets pro-

duced translations within the limits seen during a 90-NLachman test (Table 3).

Flexion Angle Where Maximal Translation Occurred. In

the open chain exercises, the maximal translation oc-curred between 16° and 19° of knee flexion. In the three

Figure 2. The four active exercises. A, active knee extension. B, squat where the projection of the center of gravity was

approximately over the feet. C, squat where the projection of the center of gravity was approximately behind the feet. D, squat

where the projection of the center of gravity was approximately in front the feet.





7/12

closed kinetic chain exercises, the maximal translationoccurred at significantly higher angles (23° to 40°) com-pared with the open kinetic chain exercises, but therewere no significant differences between the three closedkinetic chain exercises.

Differences in Translation at Different Flexion Angles.In the open kinetic chain exercise, the translation washigher during the eccentric phase than during the con-centric phase. Between 40° and 10° of knee flexion, thetranslation was always smaller in the closed kineticchain exercises than in the open kinetic chain exercises. At 40° to 70°, a shift occurred so that at flexion anglesover 70°, the translation was larger in the closed kinetic

chain exercises, especially during the concentric phaseof the motion. The difference in the translation behaviorduring flexion and extension between the types of exer-cise was significant ( P 0.0007). During the closedkinetic chain exercises, the change in translation be-tween 30° and 90° of knee flexion was only half thatduring the open kinetic chain exercise (a change of 1.3mm per 20° interval during the closed kinetic chainexercise and 2.6 mm during the open kinetic chainexercise).

Figure 3. Maximal translation during the four exercises with

the different loads (mean SD). Open circles are for the

lowest load (0 kg in the active extension test and squat

performed on two legs in the squat tests), open squares are

for the intermediate load (4 kg in the active extension test),

and filled circles are for the highest load (8 kg in the active

extension test and squat performed on one leg in the squat

tests). C, control group; U, uninjured leg; I, injured leg; CG,

center of gravity. Asterisks indicate significant differences.

Figure 4. Maximal translation with maximal load for the fourexercises (mean SD): active extension with an 8-kg load

(filled circle), squat with the projection of the center of gravity

over the feet on 1 leg (open squares), squat with the projec-

tion of the center of gravity behind the feet on 1 leg (horizon-

tal line), and squat with the projection of the center of gravity

in front of the feet on 1 leg (open circles). Comparison within

groups for the control subjects, the uninjured leg in the ACL-

deficient subjects, and the injured leg in the ACL-deficient

subjects. a, significantly different from active extension with 8

kg; b, significantly different from squat with the center of

gravity over the feet; c, significantly different from squat with

the center of gravity behind the feet; d, significantly different

from squat with the center of gravity in front the feet.

TABLE 3Maximal Translation (in Millimeters) in the Control Group and the Uninjured and Injured Legs in the ACL-Deficient Group

(Mean and SD)

Exercise Control ACL-deficient group P Value

Uninjured leg Injured leg Difference ACLa C-ACLb

Lachman 8.2 1.7 8.4 1.9 11.9 2.8 3.5 2.3 0.0000 0.8754 Active extension

0 kg 5.6 1.9 7.9 3.5 10.1 3.1 2.2 1.9 0.0002 0.05904 kg 6.7 1.8 9.0 3.1 11.6 3.5 2.6 1.2 0.0000 0.04688 kg 7.5 2.0 10.3 3.5 12.4 3.4 2.1 2.2 0.0003 0.0140

SquatCenter of gravity over the feet

2 legs 5.9 2.1 7.1 3.6 8.9 2.9 1.8 1.6 0.0023 0.26261 leg 7.1 2.1 8.1 3.7 9.1 2.5 0.9 2.0 0.1059 0.3446

Center of gravity behind feet2 legs 5.8 1.5 6.8 3.3 9.0 2.7 2.1 1.6 0.0003 0.35521 leg 5.7 1.8 7.2 3.1 8.5 2.9 1.2 1.4 0.0276 0.1863

Center of gravity in front of feet2 legs 6.6 2.3 6.7 3.9 7.5 3.1 0.9 1.3 0.1300 0.98731 leg 8.5 2.1 8.8 4.1 9.8 3.0 1.1 1.8 0.0557 0.8174

a P value for difference between uninjured and injured leg in the ACL-deficient group.b P value for the difference between the control group and the uninjured leg in the ACL-deficient group.





8/12

ACL-Deficient Group

Maximal Translation. The translation in the uninjured

leg of the ACL-deficient patients was similar to that in thecontrol group in all exercises except during the activeextension tests performed with 4- and 8-kg loads, where it

was 27% larger ( P 0.05). In the ACL-injured leg, the

translation seen during the 90-N Lachman test (3.5

2.3mm) was 29% greater than that in the uninjured leg.During the active exercises, the maximal translation inthe injured legs was 15% to 24% greater ( P 0.05) than in

the uninjured legs in all but three exercise subsets (squatswith the center of gravity over the feet on one leg and

squats with the center of gravity in front of the feet onboth one and two legs).

In terms of the effect of different loads, the uninjured leg

had the same motion pattern as the control group in allexercises (Table 3, Fig. 3). The injured leg followed the

same pattern as the uninjured leg, with similar relativeincreases in translation with increased load, except during

the squats with the center of gravity behind and over thefeet, where translation did not increase with load.

When comparing the translation during maximal load

in the different exercises, the uninjured knees behavedsimilar to the control group (Table 3, Fig. 4). The closedkinetic chain exercises produced the same translation as

the active extension exercise except for the squat with thecenter of gravity behind the feet, where the translation

was 30% smaller ( P 0.05). In the injured leg, the trans-lation produced during the closed kinetic chain exerciseswas between 69% and 79% of that in the active extension

test ( P 0.005), with no differences between the individ-ual exercises.

Normalized Translation. In the uninjured leg, thetranslation limits were the same as those seen during the90-N Lachman test (Table 3). In the injured knee, the

translation measured during squats with the center of gravity over the feet on two legs, behind the feet on bothtwo legs and one leg, and in front of the feet on two legs

was only 70% of that seen during the Lachman test. In theactive extension exercises and the squats with the center

of gravity over and in front of the feet on one leg, thetranslation limits were the same as those during the Lach-man test. When comparing the translation during activity

in the injured knee with the 90-N Lachman test in theuninjured leg, only the active extension with 8 kg ex-

ceeded the envelope by 20%. Flexion Angle Where Maximal Translation Occurred. In

the uninjured leg, the maximal translation occurred at the

same flexion angles as in the control group. In the injuredleg the maximal translation occurred at the same angle asin the uninjured leg during the open kinetic chain exer-

cises but in significantly smaller angles (10°, P 0.05)during closed kinetic chain exercises except in the squatwith the center of gravity behind the feet on one leg.

Differences in Translation at Different Flexion Angles.

Both the uninjured and injured legs had the same motionpattern as the control group (Fig. 5).

EMG in the ACL-Deficient Group (Fig. 6)

No muscle activation was found during the Lachman testor the passive extension test.

Mean EMG Activity at 10°-90°-10° of Knee Flexion. Inthe uninjured leg, the EMG activity of the quadricepsmuscles increased (mean, 13% 2% units) with increasedload during all exercises except the squat with the centerof gravity in front of the feet. When comparing the exer-cises with the highest load in the uninjured leg, the activeextension test had less quadriceps muscle activity (mean,19% 17% maximal voluntary contraction [MVC]) thanthe closed kinetic chain exercises. The squats with thecenter of gravity over and behind the feet resulted ingreater EMG activity of the quadriceps muscle (33% 20% and 32% 21% MVC, respectively) than the squatwith the center of gravity in front of the feet (25% 23%MVC) ( P 0.05). The injured leg had the same EMGactivity pattern as the uninjured leg.

In the uninjured leg, the hamstring muscle EMG activ-ity increased with increased load (mean, 3% 3% units)

in all exercises except the squat with the center of gravityin front of the feet. When comparing the exercises with thehighest load in the uninjured leg, the activation of thehamstring muscles increased from 3% 1% during activeextension to 8% 7% during the squat with the center of gravity behind the feet and 9% 8% during the squatwith the center of gravity over the feet ( P 0.05). Thesquat with the center of gravity in front of the feet was amean of 5% 6% units lower in EMG activity than thesquat with the center of gravity over the feet ( P 0.05). Inthe injured leg, the activation pattern was the same as inthe uninjured leg except that the squat with the center of gravity in front of the feet resulted in less EMG activity

than did the other two squats ( P 0.05).In both legs, the EMG activity of the gastrocnemius

Figure 5. Tibial translation in the injured leg (mean) during

the range of motion between 0° and 90° of knee flexion

angle, in the four exercises with the highest load: active

extension with an 8-kg load (filled circles), squat with the

projection of the center of gravity over the feet on one leg

(open squares), squat with the projection of the center of

gravity behind the feet on one leg (plus), and squat with the

projection of the center of gravity in front of the feet on one

leg (open circles).





9/12

muscle remained the same with increased load during theopen kinetic chain exercises, but it increased a mean of 10% 2% units with increased load during the closedkinetic chain exercises. For the exercises with the highestload, in both legs the EMG activity of the gastrocnemiusmuscle was lower in the active extension exercise com-pared with all the closed kinetic chain exercises ( P

0.05), with no difference between the three closed kineticchain exercises.

EMG Activity at Flexion Angles of 10° to 45° and 50° to

85° in Both Concentric and Eccentric Phase. In both legs,in all exercises, the activation of the vastus medialis and vastus lateralis muscles was always significantly largerduring the concentric phase than during the eccentricphase. During the active extension test, the activation waslarger at the smaller knee flexion angles (10° to 45°); during the closed kinetic chain exercises, the activation was largerat the higher flexion angles (50° to 85°) ( P 0.05).

For the hamstring muscles, in both legs the activationlevel during the active extension exercise was the same at

the different flexion angles. Only small differences (mean,2% 1% MVC) were found between the different kneeflexion angles in the closed kinetic chain exercises, exceptin squats with the center of gravity over the feet, wherethe activation was a mean 5% 1% units greater during the concentric phase than during the eccentric phase( P 0.05).

The EMG activity of the gastrocnemius muscle during the active extension test was not significantly different atthe different flexion angles. During the closed kineticchain exercises, the activation pattern of the gastrocne-mius muscle was the same as for that of the quadricepsmuscles.

Summary of Results

•The injured legs produced greater tibial translationcompared with the contralateral uninjured leg during allexercises except for squats with the center of gravity overthe feet (on one leg) and in front of the feet (on both twolegs and one leg).

•In the uninjured legs, tibial translation increased withincreased load during the active extension and during thesquats with the center of gravity over the feet and in frontof the feet but remained the same during the squats withthe center of gravity behind the feet. In the injured leg, thetibial translation increased only during the active exten-sion and during the squat with the center of gravity infront of the feet.

•In the uninjured legs, the different closed kinetic chainexercises resulted in different amounts of tibial transla-tion, and only the squat with the center of gravity behindthe feet produced less translation than the other exercises.In the injured legs, all the closed kinetic chain exercises

produced similar translations, which were less than thetranslation produced during the open kinetic chainexercise.

•In the uninjured legs, all exercises resulted in trans-lation limits equal to those seen during the 90-N Lachmantest. In the injured legs, the limits of translation during the squats performed on two legs and during the squatwith the center of gravity behind the feet on one leg wereonly 70% of that seen during the Lachman test. The otherexercises resulted in translation equal to that seen during the Lachman test.

•The uninjured and injured legs followed the same pat-tern of tibial motion in the sagittal plane throughout the

Figure 6. Graph showing the EMG in the injured leg, of the quadriceps (mean of vastus medialis and vastus lateralis), the

gastrocnemius, and hamstring muscles during the open kinetic chain (active extension) test with an 8-kg load (AE8) and theclosed kinetic chain squat tests on one leg with the projection of the center of gravity (CG) over the feet (CG01), behind the feet

(CGB1), and in front of the feet (CGF1). The hamstring muscle activity is expressed as negative values. The four columns/groups

represent the EMG activity at angles of 10°to 45°and 50°to 85°of knee flexion (eccentric phase) and at angles of 85°to 50°

and 45°to 10°of knee flexion (concentric phase).





10/12

range of motion. The translation was always less during the closed kinetic chain exercises than during the openkinetic chain exercises between 40° and 10° of knee flex-ion, and at flexion angles over 70° the translations werelarger in the closed kinetic chain exercises.

•Similar muscle activation patterns were found in theuninjured and injured knees. The overall EMG activity

was higher during the closed kinetic chain exercises thanduring the open kinetic chain exercises. The quadricepsand gastrocnemius muscles acted simultaneously. The ac-tivity in the hamstring muscles was generally low.

DISCUSSION

The main result of this study was that translation in-creased with increasing load in the uninjured and normalknees except during the squats with the center of gravitybehind the feet, where it remained the same. The increasein translation indicates an increased recruitment of therestraining mechanism with increased load, in accordancewith other studies.1,18,19 Translation values during theactive extension exercises were higher during the eccen-tric phase than during the concentric phase, which is inline with our previous results.19 The different closed ki-netic chain exercises resulted in different amounts of tib-ial translation, and the squat with the center of gravitybehind the feet produced less translation than did theother exercises in all uninjured and normal knees, indi-cating that the restraining mechanism is engaged depend-ing on variations of the external load.

In the ACL-deficient knees, all closed kinetic chain ex-ercises produced less translation than the open kineticchain exercise, and there was no difference between theclosed kinetic chain exercises when performed with the

highest load. Activation of the quadriceps and gastrocne-mius muscles also increased with increasing load, exceptfor insignificant gastrocnemius muscle activity during theopen kinetic chain exercises and low quadriceps muscleactivation during the squat with the center of gravity infront of the feet. The hamstring muscle activation levelswere rather low, but increased with load similar to thequadriceps muscle activation pattern. The highest activa-tion levels for all muscles were seen during closed kineticchain exercises.

The knee joint has six degrees of freedom, and all mo-tions are combinations of rotations and translations. Lig-aments guide and limit the combined motions, and an

ACL rupture will affect primarily the anterior tibial trans-lation but also the axial rotation of the tibia.7 In thepresent study, we analyzed only sagittal translation anddescribed the motion pattern only two-dimensionally. Thismethod of analyzing tibial translation has been found tohave good accuracy and repeatability,1,17,18,34,35 but stan-dardized methods to evaluate axial rotations are lacking.

The amount of anterior tibial translation in normalknees depends on the muscle activation, joint compressionforces, geometry, and ligament restraints, primarily the ACL.7 After an ACL rupture, muscle activation and jointcompression forces become more important as joint stabi-lizers. In the ACL-deficient knees we found that closed

kinetic chain exercises caused less translation than didthe other exercises, which confirms the findings by Yack etal.40 The joint moments differed between the exercisesand, in spite of the smaller translation during the closedkinetic chain exercises, the largest quadriceps muscle ac-tivation was found during these exercises and occurred atlarge flexion angles. In contrast, in the open kinetic chain

exercises, the maximum quadriceps muscle activity oc-curred near extension. During the closed kinetic chainexercises, the tibia assumed an initial anterior positionand then remained within a rather narrow range. In con-trast, during the open kinetic chain exercises, the changein translation with change in knee flexion was muchlarger. Here the neutral tibial position during passivemotion was reached at 90° of flexion, when the quadricepsmuscle activity was low. This anterior displacement of thestarting position with compressive load on the joint isthought to be due to the joint geometry, including theposterior tilt of the tibial plateau, and has been describedbefore in vitro by Torzilli et al.33 and Li et al.21

Knee extension moments cause anterior shear forces onthe tibia at flexion angles between 0° and 70° during anopen kinetic chain exercise.3,25,29,38 During closed kineticchain exercises, posterior shear forces, which should pro-hibit anterior translation of the tibia, have been recordedthroughout the whole range of motion.38 Our results agreewith findings of higher joint compression forces that de-crease anteroposterior translation during a closed kineticchain exercise compared with an open kinetic chainexercise.38

Previous studies have analyzed the importance of ham-string muscle coactivation for knee stability.2,15,22,28,32,38

In agreement with other studies, we found low levels of hamstring muscle coactivation during the open kinetic

chain exercise but significantly more activation during squats with the center of gravity over and behind thefeet.19,23,38 Even so, the activation levels were generallylow, and it has been questioned if these levels can coun-teract the anterior shear forces generated by the quadri-ceps muscle.13 Torzilli et al.33 found in an in vitro exper-iment that not even a 100-N posterior force could reducethe anterior positioning that occurred after sectioning of the ACL when the knee was under a joint compressionload corresponding to weightbearing. In contrast to ourresults, Kellis and Baltopoulos14,16 found 30% to 40%hamstring muscle coactivation during an isokinetic quad-riceps muscle exercise. They calculated the hamstring mo-

ment to be about 29 Nm during concentric quadricepsmuscle action and 43 Nm during eccentric quadricepsmuscle action,15 which is only about 10% of the simulta-neous quadriceps moment. Other studies have presentedwidely varying figures for hamstring muscle coactivationduring open kinetic chain exercises, from 1.5%18 to 40%MVC,9,27,31,32 and during a squat it varies between 3%and 15%.10,11,13 Our registered hamstring muscle activa-tion was close to what can be found as cross talk8 and is inthe lower end of the range in other studies. It is thereforequestionable if this activation had importance for jointstability.

On the other hand, the gastrocnemius muscle activation





11/12

coincided with and was about 50% of the quadriceps mus-cle activation during the closed kinetic chain exercises andwas generally higher in the ACL-deficient leg, as has beenseen in another study.20 The gastrocnemius muscle hasbeen shown to pull the femur backward on the tibia.26

Limbird et al.22 found subnormal gastrocnemius andquadriceps muscle activation in the ACL-deficient knee,

which supports our suggestion of a parallel activation of these muscles. In the present study, this parallel activa-tion probably moved the tibia to a forward position, whichcould also be observed at the start of the squatting motion.It would also tend to increase joint compressive forces and,accordingly, we found no increase in translation in the ACL-deficient knees with increasing load in squats withthe center of gravity over and behind the feet, when bothmuscles were activated. Therefore, the simultaneous activation of the quadriceps and gastrocnemius musclesseems to represent an important mechanism for stabiliza-tion of the unstable knee in contrast to hamstring muscleactivation.

The results of this study indicate that patients cancontrol the increased static laxity of the ACL-deficientknee during closed kinetic chain exercises, as has beenreported in previous studies,37,40 but not during openkinetic chain exercises, especially not during the eccentricphase of the motion, in line with our previous results.19

For nonoperative treatment after an ACL injury, it isimportant to strengthen the spontaneous coactivation of quadriceps and gastrocnemius muscles, which will help tomaintain a relatively stable knee position because of theincreased joint compression forces. It is still unknown howdifferent performance of the closed kinetic chain exercisesinfluences knee kinematics after an ACL reconstructionand what effect these exercises might have on long-termknee stability.

CONCLUSIONS

1. Open kinetic chain exercises with increased load onthe knee joint cause increasing anterior tibial translation.

2. Closed kinetic chain exercises cause less translationin the ACL-injured knee than do open kinetic chainexercises.

3. In the ACL-deficient knee, the translation was thesame during the three different closed kinetic chain exer-cises, in contrast to the results in normal knees where thesquat with the center of gravity behind the feet caused theleast translation.

4. Hamstring muscle coactivation was low during theopen kinetic chain exercises but somewhat higher during the closed kinetic chain exercises. In contrast, the gastroc-nemius muscle activation was much higher, 50% of thequadriceps muscle activation, during the closed kineticchain exercises.

5. Joint compression and simultaneous quadriceps andgastrocnemius muscle activation seem to be importantmechanisms to increase knee stability.

ACKNOWLEDGMENTS

The authors thank Peter Rockborn, MD, PhD, Norrköping County Hospital, Sweden, for referring the ACL-deficientpatients. This study was supported by grants from TheSwedish Foundation for Health Care Sciences and AllergyResearch “Vårdalsstriftelsen,” Stockholm; the SwedishMedical Research Council (nr K1999–73x-012223– 03),Stockholm; the Swedish Centre for Research in Sports (nr103/98), Stockholm; and the “Doctor Svantessons Memo-rial Fund.”

REFERENCES

1. Antonopoulos J, Gillquist J: Anterior tibia translation related to isokineticconcentric quadriceps torques. Isokin Exerc Sci 6: 145–151, 1996

2. Baratta R, Solomonow M, Zhou B, et al: Muscular coactivation: The role ofthe antagonistic musculature in maintaining knee stability. Am J Sports Med 16: 113–122, 1988

3. Beynnon BD, Fleming BC, Johnson RJ, et al: Anterior cruciate ligamentstrain behavior during rehabilitation exercises in vivo. Am J Sports Med 23: 24–34, 1995

4. Beynnon BD, Johnson RJ, Fleming BC, et al: The strain behavior of the

anterior cruciate ligament during squatting and active flexion-extension. Acomparison of an open and closed kinetic chain exercise. Am J Sports Med 25: 823–829, 1997

5. Blankevoort L, Huiskes R, de Lange A: The envelopment of passive knee joint motion. J Biomech 21: 705–720, 1988

6. Bynum EB, Barrack RL, Alexander AH: Open versus closed chain kineticexercises after anterior cruciate ligament reconstruction. A prospectiverandomized study. Am J Sports Med 23: 401–406, 1995

7. Daniel DM, Akeson WH, O’Connor JJ: Knee Ligaments. Structure, Func- tion, Injury, and Repair. New York, Raven Press, 1990

8. De Luca CJ, Merletti R: Surface myoelectric signal cross-talk amongmuscles of the leg. Electroencephalogr Clin Neurophysiol 69: 568–575,1988

9. Draganich LF, Jaeger RJ, Kralj AR: Coactivation of the hamstrings andquadriceps during extension of the knee. J Bone Joint Surg 71A: 1075– 1081, 1989

10. Graham VL, Gehlsen GM, Edwards JA: Electromyographic evaluation ofclosed and open kinetic chain knee rehabilitation exercises. J Athletic

Training 28: 23–30, 199311. Gryzlo SM, Patek RM, Pink M, et al: Electromyographic analysis of knee

rehabilitation exercises. J Orthop Sports Phys Ther 20: 36–43, 199412. Henning CE, Lynch MA, Glick KR Jr: An in vivo strain gage study of

elongation of the anterior cruciate ligament. Am J Sports Med 13: 22–26,1985

13. Isear JA, Erickson JC, Worrell TW: EMG analysis of lower extremitymuscle recruitment patterns during an unloaded squat. Med Sci Sports Exerc 29: 532–539, 1997

14. Kellis E, Baltzopoulos V: Muscle activation differences between eccentricand concentric isokinetic exercise. Med Sci Sports Exerc 30: 1616–1623,1998

15. Kellis E, Baltzopoulos V: The effects of antagonist moment on the result-ant knee joint moment during isokinetic testing of the knee extensors. Eur J Appl Physiol 76: 253–259, 1997

16. Kellis E, Baltzopoulos V: The effects of normalization method on antag-onistic activity patterns during eccentric and concentric isokinetic kneeextension and flexion. J Electromyogr Kinesiol 6: 235–245, 1996

17. Kvist J: Knee motion pattern during activity, in individuals with anteriorcruciate ligament injury and uninjured controls. Doctoral Thesis, Faculty ofHealth Sciences, Linköping University, Linköping, Sweden, 2000

18. Kvist J, Gillquist J: Anterior tibial translation during eccentric, isokineticquadriceps work in healthy subjects. Scand J Med Sci Sports 9: 189–194,1999

19. Kvist J, Karlberg C, Gerdle B, et al: Knee motion pattern during differentisokinetic quadriceps torques in anterior cruciate ligament deficient indi-viduals. J Orthop Sports Phys Ther 31: in press, 2001

20. Lass P, Kaalund S, leFrevre S, et al: Muscle coordination following ruptureof the anterior cruciate ligament. Electromyographic studies of 14 patients.Acta Orthop Scand 62: 9–14, 1991

21. Li G, Rudy TW, Allen C, et al: Effect of combined axial compressive andanterior tibial loads on in situ forces in the anterior cruciate ligament: Aporcine study. J Orthop Res 16: 122–127, 1998

22. Limbird TJ, Shiavi R, Frazer M, et al: EMG profiles of knee joint muscu-lature during walking: Changes induced by anterior cruciate ligamentdeficiency. J Orthop Res 6: 630–638, 1988





12/12

23. Lutz GE, Palmitier RA, An KN, et al: Comparison of tibiofemoral jointforces during open-kinetic-chain and closed-kinetic-chain exercises.J Bone Joint Surg 75A: 732–739, 1993

24. Lysholm M, Messner K: Sagittal plane translation of the tibia in anterior

cruciate ligament-deficient knees during commonly used rehabilitationexercises. Scand J Med Sci Sports 5: 49–56, 1995

25. Nisell R: Mechanics of the knee. A study of joint and muscle loadwith clinical applications. Acta Orthop Scand (Suppl 216) 56: 4–42,1985

26. O’Connor J, Shercliff T, FitzPatrick D, et al: Mechanics of the knee, inDaniel DM, Akeson WH, O’Connor JJ (eds): Knee Ligaments. Struc- ture, Function, Injury, and Repair . New York, Raven Press, 1990, pp

201–23727. Osternig LR, Caster BL, James CR: Contralateral hamstrings (biceps

femoris) coactivation patterns and anterior cruciate ligament dysfunction.Med Sci Sports Exerc 27: 805–808, 1995

28. Palmitier RA, An KN, Scott SG, et al: Kinetic chain exercise in kneerehabilitation. Sports Med 11: 402–413, 1991

29. Renström P, Arms SW, Stanwyck TS, et al: Strain within the anterior

cruciate ligament during hamstring and quadriceps activity. Am J Sports Med 14: 83–87, 1986

30. Shelbourne KD, Nitz P: Accelerated rehabilitation after anterior cruciateligament reconstruction. Am J Sports Med 18: 292–299, 1990

31. Snow C, Cooper J, Quanbury A, et al: Antagonist cocontraction of kneeflexors during constant velocity muscle shortening and lengthening. J Electromyogr Kinesiol 3: 78–86, 1993

32. Solomonow M, Baratta R, Zhou BH, et al: The synergistic action of theanterior cruciate ligament and thigh muscles in maintaining joint stability.Am J Sports Med 15: 207–213, 1987

33. Torzilli PA, Deng X, Warren RF: The effect of joint-compressive load andquadriceps muscle force on knee motion in the intact and anterior cruciateligament-sectioned knee. Am J Sports Med 22: 105–112, 1994

34. Vergis A: Sagittal plane knee translation in healthy and ACL deficientsubjects. A methodological study in vivo with clinical implications. 1999Doctoral Thesis, Faculty of Health Sciences, Linköping University, Linkö-ping, Sweden

35. Vergis A, Gillquist J: Fluoroscopic validation of electroniometrically meas-ured femoral roll-glide motion in the knees of healthy and ACL deficientsubjects. Trans Orthop Res Soc 25: 486, 2000

36. Vergis A, Gillquist J: Sagittal plane translation of the knee during stairwalking. Comparison of healthy and anterior cruciate ligament-deficientsubjects. Am J Sports Med 26: 841–846, 1998

37. Vergis A, Hindriks M, Gillquist J: Sagittal plane translations of the knee inanterior cruciate deficient subjects and controls. Med Sci Sports Exerc 29: 1561–1566, 1997

38. Wilk KE, Escamilla RF, Fleisig GS, et al: A comparison of tibiofemoral jointforces and electromyographic activity during open and closed kineticchain exercises. Am J Sports Med 24: 518–527, 1996

39. Woo SL-Y, Chan SS, Yamaji T: Biomechanics of knee ligament healing,repair and reconstruction. J Biomech 30: 431–439, 1997

40. Yack HJ, Collins CE, Whieldon TJ: Comparison of closed and open kineticchain exercise in the anterior cruciate ligament-deficient knee.Am J Sports Med 21: 49–54, 1993



Documents

Am J Sports Med 2001 Kvist 72 82