Quadriceps Femoris and Hamstring Muscle Function in a Person With an Unstable Knee.

Knee stability during walking following anterior cruciate ligament (ACL) injury is the result of passive connective tissue tension, learned motor patterns, and muscular responses to mechanical stimuli.[1] Because the ACL is the primary connective tissue constraint to anterior translation of the tibia on the femur, increased tibiofemoral joint laxity was found with passive displacement tests[2] and isolated quadriceps femoris muscle contraction.[3] During gait, a person with an ACL-deficient knee may also respond to external forces by using muscles for stability.

There is evidence that muscular control of knee joint stability is impaired following an ACL injury.[4-6] Hamstring muscle contractions in response to postural perturbations, for example, were found to be slower,[4] postural sway during one-leg standing was found to be greater,[5] and the threshold to detect passive motion of the knee joint was found to be decreased[6] in subjects with ACL deficiencies compared with uninjured subjects. Disability in people following an ACL injury may be caused by changes in muscular control.[7]

Persistent muscle weakness has been attributed to patients' inability to voluntarily activate the muscles.[8] Researchers have described associations between the inability to control the activity of the quadriceps femoris muscles and pain,[8] joint effusion,[9] immobilization,[10] and altered joint receptor function.[7] We are not aware of any studies in which the influence of other potential sources of muscle inhibition, such as psychological and emotional aspects, are documented.

The evaluation of knee dysfunction in people with possible problems in motor control presents a challenge. Lorentzon et al[11] used tomography to measure thigh muscle cross-sectional area in conjunction with isokinetic testing in subjects with ACL deficiencies. They found a 5% decrease in quadriceps femoris muscle cross-sectional area and a 25% decrease in knee extensor peak torque at 30 [degrees] /s compared with the uninjured side. Lorentzon et al[11] concluded that limited activity of the quadriceps femoris muscle was the most important factor in producing the torque deficits. These authors, however, did not report whether the percentage of muscle deficit was gravity corrected. Therefore, their use of percentages should be viewed with caution. Snyder-Mackler et al[12] stimulated the femoral nerve during maximal voluntary isometric contractions in subjects with ACL injuries and in subjects without ACL injuries. They found that electrical stimulation of the femoral nerve produced an increase in torque above that obtained with a maximum volitional effort in the subjects with subacute ACL injuries compared with subjects without ACL injuries. In both protocols,[11,12] the maximum volitional activity of the knee extensor muscles was measured relative to the estimated maximum muscular force. The authors[11,12] provided potential methods for evaluating a patient's level of muscle inhibition.

The ability of physical therapy interventions to enhance force production in muscles of people with unstable knees remains controversial.[11] Whether muscle inhibition can be modified, in our view, would depend in part on the inhibitory mechanisms involved. Some researchers[11,12] have questioned the effectiveness of "strengthening" exercises for patients with muscle inhibition, assuming that muscle inhibition is a potentially unmodifiable and overwhelming mechanism of thigh muscle dysfunction. Other experts recommend muscle stimulation, electromyographic (EMG) biofeedback, and other methods to overcome muscle inhibition.[8] The purpose of our case report is to describe the evaluation, treatment, and short-term outcome for a patient with chronic instability of the knee during walking.

Case Description

Patient

The patient was a 34-year-old man who had bilateral knee pain and who reported "giving way" of the left knee as a consequence of a motor vehicle accident (MVA). There were no fractures as a consequence of the MVA. His left knee problems limited his ambulation to short distances. He had a previous history of left ACL deficiency from a soccer injury sustained 4 years prior to the MVA. The diagnosis by the orthopedic surgeon following the MVA was bilateral ACL-deficient knees and left knee osteoarthritis. Ten months after the MVA, he underwent a patellar tendon autograft reconstruction of his left ACL.

Eight months post-reconstruction, he was reviewed by a multidisciplinary group because of left knee instability. During walking, the left leg demonstrated observable anterior instability of the tibia on the femur, and the left tibia visibly subluxated anteriorly with every step. Although there was no known traumatic injury to the hamstring or quadriceps femoris muscles, the patient did not appear, based on our observations, to contract the thigh muscles in an appropriate sequence during gait. The consensus of the surgeons at rounds was that the patient was not a candidate for a revision of his ACL reconstruction and that the patient should seek physical therapy directed at muscular coordination.

Modified KT2000

The KT2000 Knee Ligament Arthrometer(*) has been used to quantify the passive displacement of the anterior tibial tubercle relative to the patella (representative of the femoral position) during the application of anterior and posterior forces to the tibia.[2] The patient was positioned supine with the knee supported at an angle of approximately 20 degrees, and the force-displacement measurement device was attached to the lower extremity. The analog output of the KT2000 was stored on a computer disk,([dagger]) and the force-displacement curve was used to quantify the displacement and stiffness of both knees.[13] We measured passive displacement at 135 N because this measure has been found to be the most sensitive in detecting differences between left and right sides.[14] The reliability of measurements obtained with the KT2000 was found to be .94 (intraclass correlation coefficient) for subjects with ACL-deficient knees.[15] The passive knee instability measure provided an indication of the integrity of the ACL reconstruction of the left knee. We hypothesized that the instability of the left knee during walking was related to inability of the reconstructed connective tissue to constrain anterior translation of the tibia on the femur, and we were interested in determining whether differences in passive constraints between the patient's 2 unstable knees could account for differences in stability during walking.

Force-displacement plots during anterior translation of the tibia demonstrated passive laxity of both knees (Fig. 1). At 135 N of force, the anterior displacement was 21 mm for the right knee and 22 mm for the left knee. The mean normal anterior displacement at 135 N of force has been reported to be 6 mm.[16]

[Figure 1 ILLUSTRATION OMITTED]

Isometric Torque Measurements

Inadequate torque production by the quadriceps femoris or hamstring muscles may be associated with loss of knee stability.[17] Isometric and isokinetic torque are commonly measured in the ACL-deficient knee.[18-20] The Cybex 340 dynamometer([double dagger]) was used to quantify maximum volitional isometric extension and flexion torques at 90 degrees of knee flexion intermittently throughout the treatment period. Measurement of isometric quadriceps femoris muscle contractions at 90 degrees of knee flexion was chosen because it is not believed to produce an anterior shear of the tibia on the femur.[21] Previous studies provide comparison measurements for subjects with ACL-reconstructed knees.[20] The reliability of torque measurements obtained with a Cybex isokinetic dynamometer has been studied in patients without impairments,[22] but the reliability of torque measurements obtained for patients with ACL deficiency is unknown. Initially, right knee extension and flexion isometric torques were 154 and 101 N.m, respectively. Left knee extension and flexion isometric torques were 80 and 78 N.m, respectively.

Assessment of Muscle Inhibition

The assessment of muscle inhibition, in our opinion, can theoretically allow for the discrimination between muscle weakness caused by changes within the muscle and muscle weakness caused by insufficient activation of a muscle. Muscle inhibition was estimated using the interpolated twitch torque technique.[23] Bipolar surface EMG electrodes were placed on the vastus lateralis muscle. Carbon-rubber surface stimulation electrodes were placed over the femoral nerve distal to the inguinal canal and over the distal mitt-quadriceps femoris muscle. The patient was positioned on a KinCom dynamometer (Kinematic Communicator 125 AP([sections])). Stimulation of 240 V for 0.8 milliseconds was applied with a Grass S88 muscle stimulator equipped with a subject isolation unit.([parallel]) Electrical stimulation was applied at rest (resting twitch torque) and during maximal isometric quadriceps femoris muscle contractions (interpolated twitch torque). Before testing, the patient was familiarized with the test situation, and a series of near-maximal practice contractions were performed. Three trials at 2 knee joint angles (90 [degrees] and 30 [degrees] of knee flexion) on both lower extremities were performed before and after the 12-week training period. Muscle inhibition (expressed as a percentage) was calculated by the following equation:

(interpolated twitch torque/resting twitch torque) x 100

The percentage of muscle inhibition prior to treatment was far greater (Table) than for a group of subjects without knee instability.[23] The mean muscle inhibition of the quadriceps femoris muscle determined for a group of 10 volunteers was found to be 10% at 30 degrees of knee flexion and 12% at 90 degrees.[23]

Table. Percentage of Muscle Inhibition at Knee Angles of 30 and 90 Degrees From Full Extension

                30 [degrees]         90 [degrees]
                Right         Left   Right   Left

Pretreatment    26             116   62      81
Posttreatment   23              60   54      63

Gait Analysis

Three-dimensional kinematics and kinetics as well as EMG activity of the left leg during walking were evaluated to assess patterns of muscle activity, joint angles, and forces and moments. A high-speed video digitizing system recorded the 3-dimensional motions of reflective markers placed on the patient's femur, tibia, and foot. Surface EMG electrodes were placed over the biceps femoris and vastus lateralis muscles of the patient's left thigh. Prior to further analysis, EMG data were filtered using a high-pass Butterworth filter with a cutoff of 25 Hz to remove the movement artifact. The patient walked at a self-selected pace without braces or walking aids along a 7-m walkway that was equipped with a Kistler force plate.(#) Ground reaction forces, lower-extremity joint angles, and joint moments (using an inverse dynamics model[24]) were calculated using Kintrak software.(**) Walking speed was determined as the average horizontal speed of the marker placed over the greater trochanter throughout the data collection. Selected gait variables were plotted graphically as the mean of 5 trials before treatment and after 12 weeks of training (Figs. 2-4). The repeatability of these methods has been studied previously in uninjured subjects but not in people similar to the patient we studied.[25] The coefficients of multiple correlation for the measures used in this study (moments about the knee, knee joint angle, and forces) were reported to range from .94 to .99.[25]

[Figures 2-4 ILLUSTRATION OMITTED]

The patient's primary goals for physical therapy were to decrease knee instability during walking and to improve his short-distance walking. Therefore, quantitative gait analysis results were compared with previously published graphics of the gait variables of uninjured subjects to determine the patient's compensation for knee instability during walking.[25,26] Hamstring muscle EMG recordings (Fig. 2) during gait confirmed our clinical observations that the hamstring muscle activity was not coordinated with the gait cycle. We believed that a successful physical therapy program would be associated with hamstring muscle activity at the initiation of the stance phase. In addition, this patient's measured ground reaction forces (Fig. 3), calculated moments at the knee (Fig. 4), and measured knee angle (Fig. 5) differed from those of uninjured subjects. Subjects without ACL impairments have 2 distinct peaks in the vertical ground reaction force during the stance phase of walking.[26] The anteroposterior ground reaction force of a hypothetical subject without impairment during walking contains a distinct anterior force peak that is approximately equal (and opposite in sign) to the posterior force peak. The mediolateral component of the ground reaction force of people without injuries often includes a small medial shear peak immediately after foot-strike. The knee angle during walking, previously reported to be between 0 and 25 degrees,[23] also was found to be much greater in this patient (Fig. 5). In the gait of subjects with ACL injuries, the resultant knee moments have been suggested to reflect a destabilizing mechanism.[25,26] Reduction of the knee extension moment is thought by some authors to decrease anterior translation of the tibia on the femur during walking.[25,26]

[Figure 5 ILLUSTRATION OMITTED]

Physical Therapy

The purpose of the therapeutic program was to improve the activity of the quadriceps femoris and hamstring muscles. Active exercises for the lower extremities were the primary focus during the training sessions. Our hypothesis was that, as the thigh muscle coordination improved, gait would also improve, as would stability of the tibiofemoral joint. We believed that the patient might learn to control the anterior translation of the tibia on the femur and transfer this skill to walking. The patient participated in 24 training sessions of approximately 2 hours' duration over a period of 12 weeks. Several variations on resisted hamstring muscle exercises were performed throughout the treatment period: unilateral knee flexion and extension in the prone position using a 2.2-kg weight attached to ankle, unilateral knee flexion with elastic resistance while in a sitting position, and isometric hamstring muscle exercises in a supine position with the knee flexed at 20 degrees. In addition, hamstring and quadriceps femoris muscle isometric co-contraction exercises were performed in sitting and standing positions with the knee at 20 degrees of flexion and with Myomed 432 bipolar surface EMG electrodes([dagger][dagger]) over the vastus medialis and biceps femoris muscles. The patient was encouraged to activate the hamstring muscle group prior to the quadriceps femoris muscle. The knee was flexed to 20 degrees because this is the knee flexion angle during the mid-stance phase of gait for people without injuries.[23]

The Shuttle modified less press([double dagger][double dagger]) was chosen as an exercise device because the knee joint angle could be controlled between 5 and 60 degrees (Fig. 6). Zero to 66 degrees of knee flexion is considered to be the normal range of motion for humans during walking and stair climbing.[1] We also believed that forces applied to the lower extremity could be controlled by the patient at a level at which the knee position could be maintained with this device. The apparatus also permitted easy setup for EMG biofeedback and visual feedback. The patient was in a supine position on the modified leg press and performed 1- and 2-legged presses while observing the position of his knee in a mirror and monitoring activity levels of the quadriceps femoris and hamstring muscles with EMG biofeedback (Fig. 6).

[Figure 6 ILLUSTRATION OMITTED]

The patient was encouraged to complete 3 sets of 10 repetitions of each exercise unless prevented by fatigue or pain. The exercises were progressed by increasing the number of repetitions or the resistance, depending on the patient's level of pain and perceived exertion. The patient was encouraged to contract the hamstring muscle prior to the quadriceps femoris muscles and to observe the activation levels and sequence of activation using the EMG biofeedback.

Gait training on a treadmill at 2 km/h with bilateral upper-extremity support and a mirror was used to encourage symmetrical lower-extremity motion and endurance. The patient was instructed to contract the hamstring muscles prior to the onset of weight bearing.

Outcomes

Patient's Report

The patient remarked that he was able to walk with increased confidence. He stated that his knee was more stable. The knee pain, however, was not resolved.

Isometric Torque Production

Maximal isometric torque production by the extensors and flexors of both knees increased throughout the 12-week training period (Fig. 7). Right knee extension isometric torque at 90 degrees of knee flexion increased by 119%, right knee flexion torque increased by 79%, left knee extension torque increased by 209%, and left knee flexion torque increased by 117%.

[Figure 7 ILLUSTRATION OMITTED]

Muscle Inhibition

Both the right and left knees demonstrated a decrease in muscle inhibition from pretraining to posttraining measurements (Table). The left knee had a greater decrease in muscle inhibition compared with the right knee at both 30 and 90 degrees of knee flexion.

Gait Analysis

The assessment of treatment outcome was based, in part, on comparisons with previously published graphics of the gait variables.[25,26] Our analysis also considered reports proposing theories of adaptive and maladaptive compensation patterns in subjects with ACL injuries.[1,27] The patient's average self-determined walking speed during gait assessment was 0.54 m/s prior to the training program and 0.67 m/s after the training program. Rectified and smoothed hamstring muscle EMG during gait (Fig. 2) demonstrated a distinct activation prior to heel-strike and during the initial portion of the stance phase at the posttraining assessment.

Ground reaction forces for the left lower extremity, averaged over 5 trials before and after treatment, are depicted in Figure 3. Our patient demonstrated some improvement in the separation of the deceleration phase and the acceleration phase of the vertical ground reaction force after treatment (Fig. 3A). The mean ([+ or -] standard deviation) anterior ground reaction force increased from the pretreatment assessment (0.043 [+ or -] 0.013 of body weight [BW]) to the posttreatment assessment (0.084 [+ or -] 0.007 BW). The mean posterior ground reaction force also increased from the pretreatment assessment (0.103 [+ or -] 0.025 BW) to the posttreatment assessment (0.124 [+ or -] 0.011 BW). The medial peak ground reaction force (Fig. 3C) was negligible before treatment, (0.002 [+ or -] 0.002 BW) but was distinct after treatment (0.020 [+ or -] 0.008 BW).

Mean left ankle and knee joint moments measured before and after treatment are illustrated in Figure 4. Ankle plantar-flexion moments were increased after treatment (0.078 [+ or -] 0.055 BW) compared with before treatment (0.028 [+ or -] 0.038 BW), indicating that more of the propulsive forces were generated through the ankle (Fig. 4A). The mean knee extension moment (Fig. 4B) was reduced in the posttraining trials (0.059 [+ or -] 0.033 BW) compared with the pretraining trials (0.125 [+ or -] 0.022 BW).

Mean knee joint angle measurements obtained during the stance phase of gait are illustrated in Figure 5. Pretreatment measurements showed that the knee joint flexion angle increased abnormally early before toe-off (50% of stance phase). After treatment, the knee joint angle was maintained for a longer duration during the stance phase (80% of stance phase).

Discussion

The patient described in this case report had bilateral ACL-deficient knees. Although the passive knee instability was comparable for both sides, only the left tibia displayed anterior displacement with each step during the stance phase of gait. The patient demonstrated instability of the left tibiofemoral joint during gait that had been progressive over time. The case history presented here describes an approach to the evaluation and treatment of a patient with muscle inhibition and gait abnormalities. We focused on inpatient measurements because we did not expect changes in function to occur during the period of the study.

Ideally, assessment of the causes of knee instability should lead to appropriate treatment procedures, but there does not appear to be a clear cause of the unilateral tibiofemoral instability in this case. Anterior translation of the tibia is known to be related to the anterior component of the ground reaction force. The quadriceps femoris muscle has been shown to produce an anteriorly directed force on the tibia when it contracts at between 30 and 0 degrees of knee flexion.[28] The adaptations of gait observed in our patient, that is, reduction in knee extension moment, decreased anterior ground reaction force, and increased knee flexion, all caused a decrease in anterior shear at the tibiofemoral joint, which should have resulted in a decreased anterior translation of the tibia relative to the femur. Despite these adaptations, instability during walking remained to some extent. Knee stability may also be affected by the contact surface geometry, including congruity, radius of curvature, and anteroposterior tilt, relative to the direction of the applied forces.[29] Radiographs were difficult to interpret in this regard, and the mechanics of surface geometry in our patient remain unclear.

Another possible mechanism for unilateral instability during walking may be motor control deficits. Disruption of walking mechanics caused by ACL injury may require alternative muscle activity patterns during locomotion. Left thigh muscle activity during gait appeared, based on observation, to be sporadic and not timed to the gait cycle. This observation was confirmed by EMG analysis. Peak knee extension moment, which occurred at a knee flexion angle of between 14 and 38 degrees, was unopposed by coordinated activity of the hamstring muscles. Over the 12-week training period, volitional muscle activity was improved, as demonstrated by isometric torque measurements and a decrease in muscle inhibition. Gait assessment demonstrated increased weight shifting onto the left lower extremity, activity of the hamstring muscles during touchdown, and a decreased net knee extensor moment. These findings suggested to us improved motor control. An observed decrease in knee extension moment after training as a result of hamstring muscle activity may have resulted in reduced anterior shear of the tibia on the femur. We believe that these findings warrant further investigation to determine whether the altered gait patterns typical of individuals with ACL deficiencies[1] may be a consequence of muscle inhibition and muscle weakness.

The relationship between knee flexor and extensor peak torque measurements and knee function during walking remains controversial. Some authors[30] have reported a relationship between torque and ratings of knee symptoms during activities obtained with a questionnaire, whereas other authors[31] have disputed a relationship between function and measured torque. In general, there is little evidence that a linear relationship exists. The peak isokinetic knee extensor and flexion torque deficits following ACL reconstruction have been found to have a wide range.[15] Our report documents a case in which initially the left knee extensors and flexors failed to regain the ability to produce a torque. The injured knee extensor torque was only 52% of that of the uninjured knee for this patient. Increasing the knee extensor torque without improving hamstring muscle activity during gait may increase instability in the standing position because of the unopposed active drawer effect on the tibia.[27]

The use of braces is another method of decreasing instability. Several brace designs were used by this patient, but they did not achieve the desired effect of stabilizing the anterior displacement of the tibia relative to the femur. The degree of anterior tibial subluxation during the stance phase of gait while wearing any of the braces was comparable to that observed without a brace. This observation would lead us to believe that the best stabilization effect was achieved by the onset of hamstring muscle activity prior to the onset of quadriceps femoris muscle activity at heel-strike during the stance phase of gait. There is evidence that certain individuals with ACL deficiencies may successfully compensate for ACL-related knee instability.[32] In some cases, the hamstring muscles assumed the role of joint stabilizer in patients with ACL deficiencies.[7] Ciccotti et al[33] described mechanisms where the vastus lateralis, biceps femoris, or tibialis anterior muscle may protect the unstable knee.

The patient remarked that he noticed improvements in his ability to walk on level ground and on stairs. Single-leg stance was considerably easier. He also reported that he believed the improvements in torque production carried over to everyday activities. The knee gave out less often, and he felt more confident in his ability to get around.

Despite the patient's reports of improved gait and our measurements of gait, force production, and muscle inhibition, the patient had continued complaints of pain and residual instability of the left knee. Tibiofemoral and patellofemoral joint osteoarthritis and meniscal degeneration are well-documented sequelae of the ACL injury that would have had an impact on the patient's pain and function.[34] Co-contraction of the quadriceps femoris and hamstring muscles may increase internal joint loads, whereas the calculated net internal moments are decreased. Consequently, pain may be increased by muscle co-contraction, and increased joint loads have been implicated in the onset and progression of osteoarthritis.[35]

Conclusion

A patient with unilateral observable knee instability during walking was the focus of this case report. Both of the patient's knees had comparable passive laxity due to ACL injury, but only one knee exhibited instability during walking. The protocol described in this case report was used to assess some of the potential mechanisms contributing to the unilateral instability, such as muscle activity, range of motion, and forces and moments calculated during walking. In addition, muscle torque and muscle inhibition were assessed. In view of the absence of reliability for the measures we used on the type of patient in our study, our observations must remain tentative. The physical therapy focused on methods to improve mechanisms associated with active joint stability. This case report indicates that selected gait variables, torque production, and muscle inhibition may be improved with treatment. Research is needed, however, to determine whether the treatment affects function and disability level.

Acknowledgments

The ability to carry out this case report was the result of collaboration and sharing of resources between departments and laboratories. We are very grateful for the eager cooperation of the McCaig Centre for Joint Injury and Arthritis Research. Equipment used for this project was generously provided by funding from The Western Orthopaedic Arthritis Research Foundation, ATCO & Canadian Utilities, and Trans Canada Pipeline.

(*) MEDmetric Carp, 7542 Trade St, San Diego, CA 92121-2412.

([dagger]) Dataq Instruments, B220-150 Springside Dr, Akron, OH 44333.

([double dagger]) Lumex, PO Box 9003, Ronkonkoma, NY 11779-0903.

([sections]) Chattanooga Group Inc, PO Box 489, Hixson, TN 37343.

([parallel]) Grass Instruments, Astromed Industrial Park, West Warwick, RI 02893.

(#) Kistler Instruments Carp, 75 John Glenn Dr, Amherst, NY 14228-2171.

(**) Motion Analysis Carp, 3617 Westwind Blvd, Santa Rosa, CA 95403.

([dagger][dagger]) Enraf-Nonius Delft, Rontgenweg 1, PO Box 453, 2600 AL Delft, the Netherlands.

([double dagger][double dagger]) Contemporary Design Carp, PO Box 5146, Glacier, WA 98241.

References

[1] Andriacchi TP. Dynamics of pathological motion: applied to the anterior cruciate deficient knee. J Biomech. 1990;23:99-105.

[2] Daniel DM, Stone ML, Sachs R, Malcom L. Instrumented measurement of anterior knee laxity in patients with acute anterior cruciate ligament disruption. Am J Sports Med. 1985;13:401-407.

[3] Yack HJ, Collins CE, Whieldon TJ. Comparison of closed and open kinetic chain exercise in the anterior cruciate ligament-deficient knee. Am J Sports Med. 1993;21:49-54.

[4] Beard DJ, Kyberd PJ, O'Connor JJ, et al. Reflex hamstring contraction latency in anterior cruciate ligament deficiency. J Orthop Res. 1994;12:219-228.

[5] Shiraishi M, Mizuta H, Kubota K, et al. Stabilometric assessment in the anterior cruciate ligament-reconstructed knee. Clin J Sports Med. 1996;6:32-39.

[6] Barrack RL, Skinner HB, Buckley SL. Proprioception in the anterior cruciate deficient knee. Am J Sports Med. 1989;17:1-6.

[7] Solomonow M, Baratta R, Zhou BH, et al. The synergistic action of the anterior cruciate ligament and thigh muscles in maintaining joint stability. Am J Sports Med. 1987;15:207-213.

[8] Morrissey MC. Reflex inhibition of thigh muscles in knee injury: causes and treatment. Sports Med. 1989;7:263-276.

[9] Kennedy JC, Alexander IJ, Hayes KC. Nerve supply of the human knee and its functional importance. Am J Sports Med. 1982;10:329-332.

[10] Wolf E, Magora A, Gonen B. Disuse atrophy of the quadriceps muscle. Electromyography. 1971;11:479-490.

[11] Lorentzon R, Elmqvist L-G, Sjostrom M, et al. Thigh musculature in relation to chronic anterior cruciate ligament tear: muscle size, morphology, and mechanical output before reconstruction. Am J Sports Med. 1989;17:423-429.

[12] Snyder-Mackler L, De Luca PF, Williams PR, et al. Reflex inhibition of the quadriceps femoris muscle after injury or reconstruction of the anterior cruciate ligament. J Bone Joint Surg Am. 1994;76:555-560.

[13] Maitland ME, Bell GD, Mohtadi NGH, Herzog W. Quantitative analysis of anterior cruciate ligament instability. Clinical Biomechanics. 1995;10:93-97.

[14] Stratford PW, Miseferi D, Ogilvie R, et al. Assessing the responsiveness of five KT1000 knee arthrometer measures used to evaluate anterior laxity at the knee joint. Clin J Sports Med. 1991;1:225-228.

[15] Maitland ME, Lowe R, Stewart S, et al. Does Cybex testing increase knee laxity after anterior cruciate ligament reconstructions? Am J Sports Med. 1993;21:690-695.

[16] Myrer JW, Schulthies SS, Fellingham GW. Relative and absolute reliability of the KT-2000 arthrometer for uninjured knees: testing at 67, 89, 134, and 178 N and manual maximum forces. Am J Sports Med. 1996;24:104-108.

[17] Kannus P, Jarvinen M. Age, overweight, sex, and knee stability: their relationship to the post-traumatic osteoarthrosis of the knee joint. Injury. 1988;19:105-108.

[18] Murray SM, Warren RF, Otis JC, et al. Torque-velocity relationships of the knee extensor and flexor muscles in individuals sustaining injuries of the anterior cruciate ligament. Am J Sports Med. 1984;12: 436-440.

[19] Kannus P, Latvala K, Jarvinen M. Thigh muscle strengths in the anterior cruciate ligament deficient knee: isokinetic and isometric long-term results. J Orthop Sports Phys Ther. 1987;9:223-227.

[20] Harter RA, Osternig LR, Standifer LW. Isokinetic evaluation of quadriceps and hamstrings symmetry following anterior cruciate ligament reconstruction. Arch Phys Med Rehabil. 1990;71:465-468.

[21] Smidt GL. Biomechanical analysis of knee flexion and extension. J Biomech. 1973;6:79-92.

[22] Mawdsley RH, Knapik JJ. Comparison of isokinetic measurements with test repetitions. Phys Ther. 1982;62:169-172.

[23] Suter E, Herzog W. Extent of muscle inhibition as a function of knee angle. Journal of Electromyography and Kinesiology. 1997;7:123-130.

[24] Bresler B, Frankel JP. The forces and moments in the leg during level walking. Trans Am Soc Mech Eng. 1950;72:27-36.

[25] Kadaba MP, Ramakrishnan HK, Wootten ME, et al. Repeatability of kinematic, kinetic, and electromyographic data in normal adult gait. J Orthop Res. 1989;7:849-860.

[26] Nigg BM. Measurement techniques. In: Nigg BM, Herzog W, eds. Biomechanics of the Musculo-Skeletal System. New York, NY: John Wiley & Sons Inc; 1994:222.

[27] Berchuck M, Andriacchi TP, Bach BR, Reider B. Gait adaptations by patients who have a deficient anterior cruciate ligament. J Bone Joint Surg Am. 1990;72:871-877.

[28] Grood ES, Suntay WJ, Noyes FR, Butler DL. Biomechanics of the knee-extension exercise: effect of cutting the anterior cruciate ligament. J Bone Joint Surg Am. 1984;66:725-734.

[29] Beynnon B, Yu J, Huston D, et al. A sagittal plane model of the knee and cruciate ligaments with application of a sensitivity analysis. J Biomech Eng. 1996;118:227-239.

[30] Wilk KE, Romaniello WT, Soscia SM, et al. The relationship between subjective knee scores, isokinetic testing, and functional testing in the ACL-reconstructed knee. J Orthop Sports Phys Ther. 1994;20:60-73.

[31] Lephart SM, Perrin DH, Fu F, et al. Relationship between selected physical characteristics and functional capacity in the anterior cruciate ligament-insufficient athlete. J Orthop Sports Phys Ther. 1992;16: 174-181.

[32] Casteleyn P-P, Handelberg F. Non-operative management of anterior cruciate ligament injuries in the general population. J Bone Joint Surg Br. 1996;78:446-451.

[33] Ciccotti MG, Kerlan RK, Perry J, Pink M. An electromyographic analysis of the knee during functional activities, II: the anterior cruciate ligament-deficient and -reconstructed profiles. Am J Sports Med. 1994;22:651-658.

[34] Daniel DM, Stone ML, Dobson BE, et al. Fate of the ACL-injured patient: a prospective outcome study. Am J Sports Med. 1994;22: 632- 644.

[35] Peyron JG. Is osteoarthritis a preventable disease? J Rheumatol. 1991; 18 (suppl) :2-3.

This article was submitted October 9, 1997, and was accepted August 6, 1998.

ME Maitland, PhD, PT, is Associate Professor and Physical Therapist, Sport Medicine Centre, University of Calgary, 2500 University Dr NW, Calgary, Alberta, Canada T2N 1N4 ([e-mail]). Address all correspondence to Dr Maitland.

SV Ajemian, is Customer Service Engineer, Motion Analysis Corp. Mr Ajemian was a student at the McCaig Centre for Joint Injury and Arthritis Research, University of Calgary, when this case report was written.

E Suter, PhD, is Adjunct Assistant Professor, Human Performance Laboratory, University of Calgary.