INJURY CLINIC
Sports Medicine II (6): 402-413, 1991 0112-1642/91/0006-0402/$06.00/0 © Adis International Limited. All rights reserved. SP01022.
Kinetic Chain Exercise in Knee Rehabilitation Randal A, Palmitier, Kai-Nan An, Steven G, Scott and Edmond Y.S Chao Department of Physical Medicine and Rehabilitation and Biomechanics Laboratory, Department of Orthopedics, Mayo Clinic, Rochester, Minnesota, USA
Contents 402 403 403 404 404 404 405 405 405 405 406 406 408 408 409 409
4/l
412
Summary
Summary I. The Evolution of ACL Rehabilitation 1.1 The 'Paradox' of Exercise 1.2 Weightbearing (Kinetic Chain) Exercise 2. The Origin of the Kinetic Chain Concept 2.1 Kinesiology 2.2 Terminology 3. The Theoretical Basis for Decreased ACL Strain 3.1 Hamstring Co-Contraction 3.2 Biomechanical Considerations 4. The Concurrent Shift 4.1 Specificity of Training 4.2 Effect of the Concurrent Shift on Other Joints 5. Clinical Application of the Kinetic Chain Concept 5.1 Effective Recruitment of the Kinetic Chain During Exercise 5.2 Applying Biomechanical Principles to the Leg Press 6. Implications for Treatment 7. Conclusion
Rehabilitation is recognised as a critical component in the treatment of the anterior cruciate ligament (ACL) injured athlete, and has been the subject of intense research over the past decade. As a result, sound scientific principles have been applied to this realm of sports medicine, and have improved the outcome of both surgical and nonsurgical treatment. Possibly the most intriguing of these principles is the use ofthe kinetic chain concept in exercise prescription following ACL reconstruction. The hip, knee, and ankle joints when taken together, comprise the lower extremity kinetic chain. Kinetic chain exercises like the squat recruit all 3 links in unison while exercises such as seated quadriceps extensions isolate one link of the chain. Biomechanical assessment with force diagrams reveals that ACL strain is reduced during kinetic chain exercise by virtue of the axial orientation of the applied load and muscular co-contraction. Additionally, kinetic chain exercise through recruitment of all hip, knee, and ankle extensors in synchrony takes advantage of specificity of training principles. More importantly, however, it is the only way to reproduce the concurrent shift of 'antagonistic' biarticular muscle groups that occurs during simultaneous hip, knee, and ankle extension. Incoordination of the concurrent shift fostered by exercising each muscle group in isolation may ultimately hamper complete recovery. Modifying present day leg press and isokinetic equipment will allow clinicians to make better
Kinetic Chain Exercise in Knee Rehabilitation
403
use of kinetic chain exercise and allow safe isokinetic testing of the ACL reconstructed knee. Reconstruction of the ACL with a strong well placed graft to restore joint kinematics, followed by scientifically sound rehabilitation to improve dynamic control of tibial translation, will improve the outcome after ACL injury.
'Closed kinetic chain' exercise is touted as the best way to exercise the knee after anterior cruciate ligament (ACL) reconstruction and, as a result, has become a component of most ACL rehabilitation protocols (Anderson & Lipscomb 1989; Depalma & Zelko 1986; Giove et a1. 1983; Paulos et a1. 1981; Seto et a1. 1989; Silfverskold et a1. 1988). A review of the literature, however, fails to uncover convincing evidence to support these techniques or even an adequate explanation of the concept. This paper attempts to clarify the kinetic chain concept and verify theoretically the effect such exercises have on the knee. An understanding of these principles should make exercise prescription easier and rehabilitation more effective.
1. The Evolution of ACL Rehabilitation Rupture of the anterior cruciate ligament is one of the most common and devastating injuries an athlete can sustain. The devastation arises from the fact that returning enough stability to the ACL-deficient knee to allow a return to athletics is extremely difficult. There is still controversy in the literature regarding the treatment of choice for the torn ACL. Expert opinions vary from acute repair (Clancey et a1. 1988; Fetto & Marshall 1980; Hawkins et a1. 1986; Kennedy et a1. 1974; Pattee et a1. 1989; Straub & Hunter 1988) to late reconstruction (Noyes et a1. 1983) to nonsurgical rehabilitation (Balkfors 1982; Chick & Jackson 1978; Giove et a1. 1983; Marshall et a1. 1979). Many agree that good rehabilitation is imperative (Balkfors 1982; Chick & Jackson 1978; Clancey et a1. 1982; DePalma & Zelko 1986; Giove et a1. 1983) and some surgeons feel it is just as important as any operative procedure (Paulos et a1. 1981). Identifying exactly what constitutes a good ACL rehabilitation programme is the problem. Literature on knee rehabilitation prior to 1980 lacked sci-
entific backing (Gollanick et a1. 1976; Ryan & Allman 1974; Smillie 1971; Yamamoto et a1. 1976; Yocum et a1. 1978). McDaniel and Dameron (1980) found that those athletes with ACL injuries who achieved significant thigh hypertrophy on the involved side as compared to the uninvolved side were more successful at returning to athletic participation. Although this demonstrated the general need for rehabilitation, it did not shed any light on the specifics of exercise regimens. l.l The 'Paradox of Exercise'
Paulos et a1. (1981) found that ACL strain was dramatically increased during the last 30° of knee extension. This increase in strain was felt to be sufficient to endanger the healing graft. No increase in ACL strain was seen during knee flexion against resistance. Similar findings had previously been documented in the biomechanics literature (Lindahl & Movin 1967; Smidt 1973) and have since been confirmed (Arms et a1. 1984; Grood et al. 1984; Henning et al. 1985; Kaufman 1988; Renstrom et al. 1986). Giove et al. (1983) placed a group of patients with ACL-deficient knees on a progressive resistance exercise programme to promote thigh hypertrophy. They used a programme described by Smillie (1971), but emphasised hamstring strengthening. They found that higher levels of sports participation were achieved by the patients whose hamstring strength was equal to or greater than their quadriceps strength. Since hamstring strength in the normal population is usually only two-thirds of quadriceps strength, their findings suggested that overstrengthening of the hamstrings could help compensate for the loss of the cruciate ligament. It soon became common practice in ACL injury rehabilitation to emphasise hamstring strengthening,
404
while allowing only cautious quadriceps strengthening. Since powerful quadriceps are critical for functional and athletic use of the lower extremity, attempts were made to identify quadriceps exercises that would place minimal strain on the ACL. Arms (1984) and Renstrom (1986) demonstrated that the quadriceps could be exercised isometrically at knee flexion angles from 60 to 90° without causing undue strain in the ACL. It was becoming apparent, however, that one of the most common complications after anterior cruciate ligament reconstruction was patellofemoral pain syndrome (Clancey et al. 1982; Hejgaard et al. 1984; Marshall et al. 1979; Johnson et al. 1984; Kieffer et al. 1984; Straub & Hunter 1988). This disorder is exacerbated by performing resistive quadriceps strengthening exercises in greater than 30° of knee flexion (Hungerford & Barry 1979), making it difficult to recommend training in the 60 to 90° flexion range. Patellofemoral pain can be prevented and treated by strengthening the quadriceps in the 0 to 30° knee flexion range (Bourne et al. 1988), but this is the very range of motion which places maximal strain on the anterior cruciate ligament. Thus, what Paulos had called the 'paradox of exercise' has emerged. 1.2 Weightbearing (Kinetic Chain) Exercise Grood et al. (1984) looked at forces in ACLdeficient cadaver knees during simulated knee extension exercise. They documented increased anterior tibial displacement in the last 30° of knee extension and felt that this displacement would stretch the secondary restraints in the ACL-deficient knee if knee extension exercises were done injudiciously. They suggested exercising in the upright posture to allow the 'forces of weightbearing' to decrease tibial translation during quadriceps contraction. In support of this hypothesis, Henning et al. (1985) published their study of ACL strain in vivo. The study was performed by placing a strain gauge on the ACL of volunteers and then measuring ACL strain during various lower extremity exercises. They found that isometric knee extension at 0 and
Sports Medicine 11 (6) 1991
22° produced 5 to 17 times more strain on the ACL than weightbearing exercises such as stationary biking, walking on level ground, jumping rope, or performing a half squat with one leg. It should be emphasised that the study had only 2 subjects, showed wide variability of ACL strain between the 2 subjects and made no reference to joint angle during weightbearing exercise. Despite this, it is considered a landmark study which showed, for the first time, that the quadriceps can be strengthened without jeopardising the ACL, if weightbearing, or so-called 'closed kinetic chain', exercises are used. A thorough explanation of what exactly 'closed kinetic chain exercise' is, and how it decreases ACL strain, is needed, however.
2. The Origin of the Kinetic Chain Concept 2.1 Kinesiology A review of kinesiology reveals that the 'kinetic chain' terminology was derived by Steindler (1973) from the closed kinematic and link concepts set forth by Reuleaux for mechanical engineering purposes (Gowitzke & Millner 1988). In the link concept, rigid overlapping segments are connected in series by pin joints. The system is considered closed if both ends are connected to an immoveable framework, thus preventing translation of either the proximal or distal joint centre. This creates a system where movement at one joint produces movement at all other joints in a predictable manner, and is called a closed kinematic chain. The extremities can be thought of as rigid, overlapping segments in series. Steindler proposed that, although a closed kinematic chain never occurs in the extremities, two types of kinetic chain exist under different limb loading conditions. He observed that when the foot or hand 'meet considerable resistance', muscle recruitment and joint motion differ from that seen when the foot or hand is completely free to move. He felt the difference was significant enough to warrant distinguishing the 2 conditions with separate terms. Specifically, an open kinetic chain exists when the peripheral joint of the extremity can move freely, such as when
405
Kinetic Chain Exercise in Knee Rehabilitation
waving the hand or moving the foot forward in the swing phase of gait. A closed kinetic chain exists whenever the foot or hand meet resistance, such as in the chin-up or rising from a squat. He hastened to point out that a true closed kinetic chain only exists during isometric exercise, since by definition neither the proximal nor distal segment can move in a closed system. 2.2 Terminology There are numerous exercises available for strengthening the muscles which cross the knee joint. After publication of the aforementioned articles on ACL strain, there has been a trend in the rehabilitation community to use the open and closed kinetic chain terminology to divide these exercises into 2 major groups: open kinetic chain exercises are those in which the foot is free to move (seated leg extensions); and closed kinetic chain exercises are those in which the foot meets resistance (squats and leg presses). Indeed, there is a difference in ACL strain during the 2 types of exercise, but using the open and closed chain terminology to make the distinction is inaccurate and confusing. One could argue that neither exercise is a closed kinetic chain since either the proximal or distal segment moves in each situation. Conversely, one could argue that they are both closed kinetic chains, since the peripheral limb meets resistance in each situation. Applying accurate terms to these 2 exercise situations requires an understanding of how lower extremity kinetics are altered during weightbearing exercise.
3. The Theoretical Basis for Decreased ACL Strain
Fig. 1. Leji: anterior tibial translation as a consequence of isolated quadriceps contraction. Righi: stabilisation of tibia through hamstring co-contraction.
knee extension seems at first to be paradoxical, since this muscle group is classified as a primary knee flexor. Also, although authors have shown that co-contraction of the hamstrings does occur during nonweightbearing exercise (Baratta et al. 1988; Draganich et al. 1989; Kaufman 1988; Renstrom et al. 1986; Solomonow et al. 1987), Draganich (1989) demonstrated that this contraction is quite small (10 to 20% of maximal) and Renstrom (1986) demonstrated that it is relatively ineffective in decreasing ACL strain. During weightbearing exercise like the squat, however, more significant hamstring contraction is induced. Since the hamstrings are a biarticular muscle and also function as a strong hip extensor, forceful hamstring contraction is induced to stabilise the hip flexor moment. High tension in this muscle group then has a secondary effect on the knee. The reduction in knee shear that results can be illustrated with simple force diagrams. 3.2 Biomechanical Considerations
3.1 Hamstring Co-Contraction From a theoretical standpoint, the decrease in ACL strain seen during a weightbearing exercise has been explained by the fact that the hamstrings are active. Co-contraction of this muscle group helps neutralise the tendency of the quadriceps to cause anterior tibial translation, as shown diagramati cally in figure I. Hamstring contraction during
Figure 2 is a force diagram of the seated knee extension exercise. The system has been simplified such that only 3 fundamental forces are acting on the tibia: the applied force (A), the quadriceps muscle force (M), and the joint reaction force (R). If the tibia is in equilibrium, the line of application of all 3 forces must pass through a common point (d). Therefore, if the direction and point of appli-
406
Fig. 2. Force diagram. A = applied force; M = quadriceps muscle force; R =joint reaction force; f = knee instant centre of rotation; a,b,c = triangle formed by force vectors in equilibrium: used to determine length of unknown vector, R; d = common point of intersection of 3 force system in equilibrium; e---+g = shear component of R; e---+f = normal component of R.
cation of M and A are known, the direction of R can be determined. Vector R can then be resolved into its components, e-f (compression) and e-g (shear). The shear force is directed posteriorly, indicating that ifit were not for soft tissue constraint, the tibia would tend to translate anteriorly. Butler (1980) showed that 86% of this soft tissue constraint is provided by the ACL. Figures 3a-c show how variation in the location and orientation of the applied force influences the knee shear force. When A is applied proximally on the tibia, as in 3b, knee shear is dramatically decreased. This has been verified experimentally (Jurist & Otis 1985). Figure 3c shows the effect of changing the orientation of the applied force. If the force is applied with a more axial orientation, the shear component of the joint reaction force is, once again, smaller. Figure 3d shows the dramatic decrease in joint
Sports Medicine 11 (6) 1991
shear that occurs with hamstring co-contraction. In this situation, simultaneous contraction of the quadriceps, MI, and hamstrings, M2, produces a net force vector M3. As can be seen, the joint reaction force, R, caused by this net force, has a much smaller shear component than that seen during isolated quadriceps contraction. Kinetic chain exercise induces hamstring cocontraction. The reason for this is illustrated in figure 4. Consider a force A applied to the distal tibia. Since the hip and knee are unconstrained, the force creates a flexion moment at both joints. The hamstrings contract to stabilise the hip and the quadriceps contract to stabilise the knee. Activity in the biarticular hamstring muscle, induced for hip stability, helps neutralise the tendency of the quadriceps to cause anterior tibial translation. A similar phenomenon occurs at the ankle joint if the force A is applied to the bottom of the foot. Tension in the soleus, which would be induced to stabilise the ankle flexion moment, will have an indirect effect on knee shear. During exercise, the lower extremity kinetic chain is recruited when the hip, knee, and ankle are unconstrained and the force is applied axially, causing triarticular flexion. Therefore, knee shear forces are reduced by the mechanisms illustrated in both 3c and 3d. During seated leg extensions, the force is applied perpendicular to the tibia and causes a flexion moment at the knee only. Such joint isolation during exercise does not take advantage of the secondary stabilising effect of other muscles in the lower extremity kinetic chain. In other words, the kinetic chain is not exploited. These exercises should simply be called joint isolation exercises, while exercises such as the squat should simply be called kinetic chain exercises. Further description with the open and closed chain terminology is not necessary.
4. The Concurrent Shift 4.1 Specificity of Training Decreased shear at the knee is not the only advantage to the use of kinetic chain exercise in rehabilitation. Walking, running, jumping, climbing
Kinetic Chain Exercise in Knee Rehabilitation
407
and rising from a chair are all activities which exploit the kinetic chain. The lower extremity is usually used in this manner and should, therefore, be rehabilitated in this manner. Specificity of training is an accepted concept in rehabilitation and sports medicine (Fahey 1973; Harris 1984). It involves improving the strength and coordination of functional, or sport-specific, activities with exercises that approximate the desired activity. The progression from slow unresisted activity to quicker movement against resistance theoretically results in central nervous system engram patterning. Throughout this process, the emphasis is on precision of movement. Repetitive deviation from the correct movement pattern can result in substitution patterning. A substitution pattern is an incorrect and inefficient muscular recruitment pattern that can hinder performance and lead to injury. The importance of this concept in lower extremity rehabilitation becomes evident when the
kinesiology of kinetic chain exercise is considered. Consider the simultaneous hip and knee extension that occurs when rising from a squat. The rectus femoris and the hamstrings are both active. As the hip extends, the rectus femoris lengthens while the hamstrings shorten, but as the knee extends the rectus femoris shortens as the hamstrings lengthen. The result at the muscular level might be called a pseudoisometric contraction, due to simultaneous concentric and eccentric contractions at opposite ends of each muscle. This so-called concurrent shift (Steindler 1973) and the resultant pseudoisometric contraction cannot be reproduced with isolation exercises. Specificity of training must also be observed to ensure carryover of strength gains to functional activities on the playing field. Research has shown that isometric, isotonic and isokinetic strength gains are specific to the type of contraction used during training (Atha 1981). The same may hold true for
c
B M
D
s
s M,
M
A
,, 1 ,1
I 1 ,I
,
..
,, ,
I
.:' t', 'I
,, /
,,
,, ,, ,
" ,,
Fig. 3. Force diagrams demonstrating altered joint reaction force. In each example, C represents the compression component of the joint reaction force and S represents the shear component of the joint reaction force. A. Knee extension exercise with force applied on distal tibia. B. Knee extension with force applied proximally. C. Axial orientation of applied load. D. Quadriceps and hamstring co-contraction.
408
the unique type of muscular contraction that occurs in the hamstrings and rectus femoris during the concurrent shift. It has also been demonstrated that using a combination of isometric, concentric and eccentric contractions in an exercise routine results in larger strength gains than anyone alone (Atha 1981). All 3 types of contraction occur during each repetition of a kinetic chain exercise. Neural adaptation is another benefit of strength training that will be affected by the type of exercise chosen. The literature has repeatedly shown that strength gains seen during the first 4 weeks of training are primarily due to neural adaptation, not muscular hypertrophy (Dons et al. 1979; Fahey & Brown 1973; Kanehisa & Miyashita 1983; Thorstensson et al. 1976). Proposed mechanisms of this effect include: enhancement of the ability to maximally recruit agonist muscles, improved inhibition of antagonist muscles, improved synchronisation of motor unit recruitment or a combination thereof (Jones et al. 1986). Studies have shown that integrated EMG and synchronisation of contraction decrease with immobilisation and detraining (Jones et al. 1986). Since the neural adaptation that
Sports Medicine 11 (6) 1991
occurs in the lower extremity will differ dramatically as a function of whether the kinetic chain is recruited or the joint is isolated, lower extremity rehabilitation protocols must incorporate kinetic chain exercise early. 4.2 Effect of the Concurrent Shift on Other Joints Kicking is the only activity that requires powerful isolated quadriceps contraction. Therefore, strengthening the thigh with isolation exercises has no practical application for most patients. Repetitive recruitment of the quadriceps in isolation may be contraindicated since it causes abnormal engram patterning. Poor control or incoordination of the concurrent shift during kinetic chain activity will affect forces at the tibiofemoral, patellofemoral, lumbosacral and intervertebral joints, since the muscles involved cross both the knee and hip. The effect of the concurrent shift at joints other than the tibiofemoral joint raises some interesting questions. Imprecision of the concurrent shift may increase patellofemoral joint forces, and, therefore, may be a factor in anterior knee pain. Treatment of this disorder has traditionally included only joint isolation exercises (quad sets, and short arc quadriceps extensions), which may explain why so many of these cases are refractory to treatment. The effect of the hamstrings on pelvic tilt and, therefore, on lumbosacral and intervertebral joint forces is well known. Therefore, imprecision of the concurrent shift may also be a contributing factor in low back pain. Kinetic chain exercise may have a role in the treatment and prevention of patellofemoral pain and low back pain.
5. Clinical Application of the Kinetic Chain Concept
Fig. 4. Diagramatic representation of the induction of muscular co-contraction due to flexion moments caused by the applied force. A = applied force; rH = hip moment arm; rK = knee moment arm.
The use of kinetic chain exercise during anterior cruciate ligament rehabilitation makes sense from the theoretical standpoint. Unfortunately, scepticism still exists in the rehabilitation community. As a result, most ACL rehabilitation protocols emphasise joint isolation exercise early on and only
Kinetic Chain Exercise in Knee Rehabilitation
allow kinetic chain exercise in the late phases (Anderson & Lipscomb 1989; DePalma & Zelko 1986; Giove et al. 1983; Paulos et al. 1981; Seto et al. 1989). This appears to be a reversal of what should be done. The persistent use of joint isolation exercise in ACL rehabilitation may be partially responsible for the fact that most postreconstruction failures are due to stretching of the graft over time (Johnson et al. 1984). Similarly, follow-up studies of patients with ACL-deficient knees have shown that varus or valgus laxity occurs over time (Butler et al. 1980; McDaniel & Dameron 1980; Giove et al. 1983; Hejgaard et al. 1984). Repetitive loading of the secondary restraints during isolated knee extension exercise may be a contributing factor. In addition, isokinetic machines, which are universally used for strength tests, isolate the knee joint. The subjects are forced to maximally contract the quadriceps without the protection of hamstring co-contraction. Solomon ow and co-workers (1987) showed that a protective reflex arc exists between the ACL and hamstrings. With 'overloading' of the ACL, there is a reflex contraction of the hamstrings to protect the ligament. This reflex was recruited during isokinetic knee extension in their study. It appears that isokinetic testing may be contraindicated in ACL reconstructed knees (Paulos et al. 1981; Solomonow et al. 1987), and may excessively load the secondary restraints in ACL-deficient knees (Grood et al. 1984). The implications of such overloading in normal knees needs further research. 5.1 Effective Recruitment of the Kinetic Chain During Exercise Merely having the foot meet 'considerable resistance' does not guarantee the neutralisation of knee shear, however. Moments at the hip and knee will vary depending on the type of exercise performed. Figure 5 illustrates 2 weightbearing exercises that do not neutralise knee shear. In figure Sa, the intersection of the force lines ofbodyweight and the support force from the wall define the point through which the ground reaction force line must pass. As can be seen, the knee moment created by
409
A
B
Fs
Fig. 5. Hamstring co-contraction is minimal in these two exercise situations. A. The intersection of force line S (support force) and W (bodyweight) defines a point through which the force line of the ground reaction force must pass, producing a large knee moment, fK, and a small hip moment, rHo B. Arm support can easily stabilise the hip flexion moment due to its large moment arm fH2 compared to the moment arm of the ground reaction force, rH 1.
this force is large while the hip moment is small. In figure 5b, the large moment arm of the upper extremity support force allows the arm to easily overcome the hip flexion moment. 5.2 Applying Biomechanical Principles to the Leg Press Although power squats are difficult to perform when the patient is weak and in pain, the leg press, if done correctly, takes full advantage of the kinetic chain while simultaneously providing stability to the patient, decreasing strain on the low back, allowing exercise with less than bodyweight and allowing exercise of each leg independently. In order to accomplish this, however, certain modifications to the traditional leg press must be made. First, the machine must allow full hip extension. Traditional leg press machines allow, at best, -45 of full hip extension. The hamstrings, therefore, are not appropriately recruited, and the kinetic chain is not fully exploited. Full hip extension can be achieved by placing the patient in a supine position. If full hip and knee flexion and extension occur, the concurrent shift will be repro0
410
a
b
Fig. 6. Illustration of the line of force application relative to the tibia, hip, and knee for various leg press exercise situations. (In 6a, the resistance is applied horizontally via a pulley system.)
duced and appropriate hamstring recruitment will be assured. Second, the foot plate should move in an arc. An arc of motion will maintain the line of force application axial to the tibia, as was previously shown in figure 3c. If, on the other hand, the foot plate slides horizontally, or on a 4SO angle (figs 6a and b), a large component of the applied force is directed perpendicular to the tibia as flexion angles increase. The knee shear in this case will increase approaching the isolation exercise situation.
Sports Medicine 11 (6) /99/
An arc of motion also improves hamstring recruitment. The magnitude of hamstring contraction increases directly with hip moment arm. A moment arm is equal to the perpendicular distance from the joint centre of rotation to the line of force application. Therefore, only a small hip flexion moment is produced in the sliding foot plate configurations, since the line of force application passes near the hip throughout the range of motion (figs 6a and b). In addition, as the weight is lowered, the knee moment increases dramatically. The imbalance in muscular recruitment that results increases shear forces at the knee. Allowing the foot plate to move in an arc, gradually moves the line of force application away from the hip, increasing the hip flexion moment and decreasing the knee moment (fig. 6c). In the ideal situation the two moments will be nearly equivalent, thus neutralising anteroposterior shear. Finally, the foot plate must be fixed perpendicular to the frontal plane of the hip during this arc of motion, as in figure 6c. This can be accomplished with a 4-bar linkage. Fixation of the footplate in this plane once again helps maintain the line of force application axial to the tibia, but also capitalises on the knee extension moment that the soleus is capable of producing. Soleus activity, as stated earlier, helps neutralise the knee shear force. If the foot plate is allowed to move freely on the lever arm, plantarflexion merely rotates the plate, and the potential knee extensor moment is lost. Adding these simple modifications to the traditionalleg press apparatus comes very close to reproducing the squat. During the parallel squat, the trunk inclines forward as the weight is lowered, thus increasing the hip moment (fig. 7). This trunk flexion simultaneously decreases the knee flexion moment by moving the weight and/or centre of gravity forward. As a result of this decrease in knee moment arm, patellofemoral compression forces and knee shear forces remain low. This movement pattern occurs with all functional activities, including stair climbing, rising from a chair, jumping etc. In each case, the centre of gravity is shifted forward, increasing hip moment and decreasing knee moment.
Kinetic Chain Exercise in Knee Rehabilitation
Reproducing this movement pattern in the leg press would make it safe and functional for lower extremity rehabilitation. Similar modification of the isokinetic lever arm attachment will allow safe and functional isokinetic testing of the lower extremity. Development and biomechanical testing of such an apparatus is in progress at the Biomechanics Laboratory, Mayo Clinic, Rochester, MN. Until adequate equipment is developed, kinetic chain exercises such as multiangled isometric leg press, power squats, stair climbing and stationary biking should be emphasised in rehabilitation protocols. The rehabilitation team must, however, be aware of the biomechanical principles discussed herein if such exercises are expected to neutralise knee shear forces.
Fig.7. Relationship of the line offorce application to the hip and knee during performance of the squat. rH '" hip moment arm; rK '" knee moment arm.
411
6. Implications for Treatment The theory behind kinetic chain exercise and its effect on the knee would suggest that, although the ACL is the primary restraint to anterior tibial translation in the cadaver (Butler et al. 1980), it is not a check rein to translation under dynamic conditions. During activity, muscular co-contraction should neutralise the shear at the knee. The anterior and posterior cruciate ligaments then act as passive guides to the combined rolling and gliding motion of the femoral condyles on the tibial plateau as described in Mueller's 4-bar linkage model (Mueller 1983). In such situation, absolute strain on the ACL measured during maximal kinetic chain knee extension exercise would be no higher than the strain measured during active knee range of motion. Suggesting that the anterior cruciate ligament does not function as a 'check rein' is not meant to detract from its role in knee function. Indeed, the passive guiding function of the cruciates is critical to normal joint kinematics. Gerber and Matter (1983) demonstrated that disruption of the ACL alters knee kinematics such that dramatic changes in knee instant centre of rotation are seen with simple, unresisted, active range of motion. These changes produce abnormal loading of the joint surfaces which could theoretically lead to meniscal injury and degenerative changes. In addition, abrupt shifting of the joint centre of rotation during weightbearing may be a cause of instability and give-way symptoms. These biomechanical and kinesiological considerations make formulation of a treatment plan for the ACL injured knee less mysterious. Rehabilitation is not capable of returning normal joint kinematics to an ACL-deficient knee, but surgical reconstruction followed by a nonspecific and illadvised rehabilitation protocol is also doomed to failure. It would seem that a combined surgical/ rehabilitative effort is the key to success. Considering the 'surgical versus nonsurgical' debate that has permeated the literature to date, the confusion and generally dismal long term treatment results seen are not surprising.
a
Sports Medicine 11 (6) 1991
412
7. Conclusion Despite the fact that isolated knee extension exercise has repeatedly been shown to cause excessive strain on the ACL, many knee rehabilitation protocols continue to include or even emphasise these exercises. Biomechanical theory suggests that kinetic chain exercises neutralise knee shear forces through the induction of muscular co-contraction. Such exercise is, therefore, critical for successful rehabilitation of the ACL-reconstructed or ACL-deficient knee, but it must be appropriately prescribed. Although further research is needed in this area, the data presently available indicate that isolated knee extension exercises are contraindicated in these patients and that appropriately prescribed kinetic chain exercise can be instituted very early, if not immediately, after surgery. Kinetic chain exercise does more than stabilise the knee joint, however. The concurrent shift that occurs during such exercise causes unique muscular contraction and intricate muscular interactions that cannot be reproduced with joint isolation exercises. As such, specificity of training becomes a significant factor. Emphasising kinetic chain exercise prior to returning any lower extremity injured patient to strenuous weightbearing activity has theoretical merit. The incorporation of kinetic chain exercise in the rehabilitation protocols of other musculoskeletal disorders such as patellofemoral pain and low back pain should be considered. In addition, exercises that isolate ·specific joints in the lower extremity kinetic chain do not prepare the patient for weightbearing activity and may in fact predispose the patient to injury secondary to inappropriate muscular and neuromuscular training. Development of versatile equipment that safely and effectively recruits the kinetic chain will promote the incorporation of kinetic chain exercises early in rehabilitation protocols and testing procedures. Once this is accomplished, the treatment of such common musculoskeletal problems as anterior knee pain, ACL injury and low back pain may prove to be less frustrating.
Acknowledgements We would like to acknowledge the Department of Physical Medicine and Rehabilitation, Mayo Clinic, Rochester, MN, for funding this work, and Mr Fred Schultz, who had an integral part in the development of the biomechanically sound leg press apparatus described in this article.
References Anderson AF, Lipscomb AB. Analysis of rehabilitation techniques after anterior cruciate reconstruction. American Journal of Sports Medicine 17 (2): 154-160, 1989 Arms SW, Pope MH, Johnson RJ, et al. The biomechanics of anterior cruciate rehabilitation and reconstruction. American JournalofSports Medicine 12 (I): 8-18,1984 Atha J. Strengthening muscle. Exercise and Sports Science Reviews 9: 1-73, 1981 Balkfors B. The course of knee ligament injuries. Acta Orthopaedica Scandinavica 53 (Suppl. 198): 179, 1982 Baratta R, Solomonow M, Zhou BH, et al. Muscular coactivation: the role of the antagonist musculature in maintaining knee stability. American Journal of Sports Medicine 16 (2): 113-122, 1988 Bourne MH, Hazel WA, Scott SG, et al. Anterior knee pain. Mayo Clinic Proceedings 63 (42): 491, 1988 Butler OL, Noyes FR, Grood ES. Ligamentous restraints to anterior-posterior drawer in the human knee: a biomechanical study. Journal of Bone and Joint Surgery 62-A: 259-270, 1980 Chick RR, Jackson OW. Tears of the anterior cruciate ligament in young athletes. Journal of Bone and Joint Surgery 60-A: 970-973, 1978 Oancey WG, Nelson OA, Reider B, et al. Anterior cruciate ligament reconstruction using one-third of the patellar tendon ligament augmented by extra-articular tendon transfers. Journal of Bone and Joint Surgery 64-A (3): 352-359, 1982 Oancey WG, Ray JM, Zolten PJ. Acute tears ofthe anterior cruciate ligament - surgical vs. conservative treatment. Journal of Bone and Joint Surgery 70-A (10): 1483-1388, 1988 DePalma BF, Zelko RR. Knee rehabilitation after anterior cruciate ligament injury or surgery. Athl Training, Fall 1986 Dons B, Bollerup K, Bonde-Petersen F, et al. The effect of weightlifting exercise related to muscle fibre composition and muscle cross-sectional area in humans. European Journal of Applied Physiology 40: 95-106, 1979 Draganich LF, Jaeger RJ, Knalj AR. Coactivation of the hamstrings and quadriceps during extension of the knee. Journal of Bone and Joint Surgery 71-A (7): 1075-1080, 1989 Fahey TO. Physiologic adaptation to conditioning. In Fahey TO (Ed.) Athletic training: principles and practice, pp. 57-78, Mayfield Publishing, Palo Alto, CA, 1986 Fahey TO, Brown CH. The effects of an anabolic steroid on the strength, body composition and endurance of college males when accompanied by a weight training program. Medicine and Science in Sports 5: 272-276, 1973 Fetto JF, Marshall JL. The natural history and diagnosis of anterior cruciate ligament insufficiency. Clinical Orthopedics 147: 29-38, 1980 Gerber C, Matter P. Biomechanical analysis ofthe knee after rupture of the anterior cruciate ligament and its primary repair: an instant-centre analysis of function. Journal of Bone and Joint Surgery 65: 391, 1983 Giove T, Miller S, Kent B, et al. Nonoperative treatment of the
Kinetic Chain Exercise in Knee Rehabilitation
torn anterior cruciate ligament. Journal of Bone and Joint Surgery 65-A (2): 184-192, 1983 Gollanick PD, Erickson E, Harmark T, et al. Rehabilitation of the knee following surgery. Medicine and Science in Sports 8: 133-134, 1976 Gowitzke BA, Millner S. Scientific bases of human movement, pp. 1-3, Williams and Wilkins, Baltimore, 1988 Grood ES, Suntay WJ, Noyes FR, et al. Biomechanics of knee extension exercise. Journal of Bone and Joint Surgery 66-A (5): 725-734, 1984 Harris FA. Facilitation techniques and technology adjuncts in therapeutic exercise. In Basmajian JV (Ed.) Therapeutic exercise, 9th ed., pp. 110-178, Williams and Wilkins, Baltimore, 1984 Hawkins RJ, Misamore JW, Merritt TR. Followup of the acute nonoperated isolated anterior cruciate ligament tear. American Journal of Sports Medicine 14: 205-210, 1986 Hejgaard N, Sandberg H, Haad A, et al. The course of differently treated isolated ruptures of the anterior cruciate ligament as observed by prospective stress radiography. Clinical Orthopedics 182: 236-241, 1984 Henning CE, Lych MA, Glick JR. An in vivo strain gauge study of elongation ofthe anterior cruciate ligament. American Journal of Sports Medicine 13 (I): 22-26, 1985 Hungerford DS, Barry M. Biomechanics of the patellofemoraljoint. Clinical Orthopedics 144: 9-15,1979 Johnson RJ, Erikson E, Haggmark T, et al. Five to ten year follow-up evaluation after reconstruction of the anterior cruciate ligament. Clinical Orthopedics 183: 122-140, 1984 Jones NL, McCartney N, McComas AI. Human muscle power, Human Kinetics, Champaign, IL, 1986 Jurist KA, Otis Je. Anteroposterior tibiofemoral displacements during isometric extension efforts. American Journal of Sports Medicine 13 (4): 254-258, 1985 Kanehisa H, Miyashita M. Effect of isometric and isokinetic muscle training on static strength and dynamic power. European Journal of Applied Physiology 52: 104-106, 1983 Kaufman KR. Mathematical model of knee forces during exercise. PhD thesis, North Dakota State University, 1988 Kieffer DA, Curnow RJ, Southwell RB, et al. Anterior cruciate ligament arthroplasty. American Journal of Sports Medicine 12: 301, 1984 Kennedy JC, Weinberg HW, Andrews AS. The anatomy and function of the anterior cruciate ligament. Journal of Bone and Joint Surgery A (2): 223-235, 1974 Lindahl 0, Movin A. The mechanics of the knee joint. Acta Orthopedica Scandinavica 38: 226-234, 1967 Marshall JL, Warren RF, Wickiewicz VB, et al. The anterior cruciate ligament - the technique of repair and reconstruction. Clinical Orthopedics 143: 97, 1979 McDaniel WJ, Dameron TD. Untreated rupture of the anterior cruciate ligament: a followup study. Journal of Bone and Joint Surgery 62-A (5): 696-705, 1980
413
Mueller W. The knee: form, function and ligament reconstruction, Berlin Heidelberg New Y orlc, 1983 Noyes F, Movar PA, Matthews DS, et al. The symptomatic anterior cruciate deficient knee, Part I: The long term functional disability in athletically active individuals. Journal of Bone and Joint Surgery 65-A (2): 154-162, 1983 Pattee GA, Fox JM, Pizzo WD, et al. Four to ten year followup of uureconstructed anterior cruciate ligament tears. American Journal of Sports Medicine 17 (3): 430-435, 1989 Paulos L, Noyes FR, Grood E, et al. Knee rehabilitation after anterior cruci;lte ligament reconstruction and repair. American Journal of Sports Medicine 9 (3): 140-149, 1981 Renstrom P, Stanwyck TS, Johnson RJ, et al. Strain within the anterior cruciate ligament during a hamstring and quadriceps activity. American Journal of Sports Medicine 14 (I): 83-87, 1986 Ryan AI, Allman FL. Rehabilitation of the knee following anterior cruciate ligament injury. In Sports medicine, pp. 541548, New York Academy Press, New York, 1974 Seto JL, Brewster CE, Lombardo SJ, et al. Rehabilitation of the knee after anterior cruciate ligament reconstruction. Journal of Orthopedics and Sports Physical Therapy 8-18, 1989 Silfverskold JP, Steadman JR, Higgins RW, et al. Rehabilitation of the anterior cruciate ligament in the athlete. Sports Medicine 6: 308-319, 1988 Smidt GL. Biomechanical analysis of knee flexion and extension. Journal of Biomechanics 6: 79-92, 1973 Smillie IS. Injuries of the knee joint, 4th ed., Churchill Livingstone, Edinburgh, 1971 Solomonow M, Baratta R, Zhou BH, et al. The synergistic action of the anterior cruciate ligament and thigh muscles in maintaining joint stability. American Journal of Sports Medicine 15 (3) 207-213, 1987 Steindler A. Kinesiology of the human body under normal and pathological conditions, p. 63, Charles e. Thomas, Springfield IL, 1973 Straub T, Hunter R. Acute anterior cruciate ligamcent repair. Clinical Orthopedics 227: 238-250, 1988 Thorstensson A, Karlsson J, Viitasalo JHT, et al. The effect of strength training of EMG of skeletal muscle. Acta Physiologica Scandinavica 98: 232-236, 1976 Yamamoto SK, Hartman CW, Feagan JA, et al. Functional rehabilitation of the knee: a preliminary study. American Journal of Sports Medicine 6: 288-293, 1976 Yocum LA, Bohman DC, Noble HB, et al. The deranged knee, restoration of function - a protocol for rehabilitation of the injured knee. American Journal of Sports Medicine 6: 51-53, 1978 Correspondence and reprints: Randal A. Palmitier, START Physical Medicine Clinic, 1910 E. Southern Ave, Tempe, AZ 85282, USA.
Sports Medicine II (6): 402-413, 1991 0112-1642/91/0006-0402/$06.00/0 © Adis International Limited. All rights reserved. SP01022.
Kinetic Chain Exercise in Knee Rehabilitation Randal A, Palmitier, Kai-Nan An, Steven G, Scott and Edmond Y.S Chao Department of Physical Medicine and Rehabilitation and Biomechanics Laboratory, Department of Orthopedics, Mayo Clinic, Rochester, Minnesota, USA
Contents 402 403 403 404 404 404 405 405 405 405 406 406 408 408 409 409
4/l
412
Summary
Summary I. The Evolution of ACL Rehabilitation 1.1 The 'Paradox' of Exercise 1.2 Weightbearing (Kinetic Chain) Exercise 2. The Origin of the Kinetic Chain Concept 2.1 Kinesiology 2.2 Terminology 3. The Theoretical Basis for Decreased ACL Strain 3.1 Hamstring Co-Contraction 3.2 Biomechanical Considerations 4. The Concurrent Shift 4.1 Specificity of Training 4.2 Effect of the Concurrent Shift on Other Joints 5. Clinical Application of the Kinetic Chain Concept 5.1 Effective Recruitment of the Kinetic Chain During Exercise 5.2 Applying Biomechanical Principles to the Leg Press 6. Implications for Treatment 7. Conclusion
Rehabilitation is recognised as a critical component in the treatment of the anterior cruciate ligament (ACL) injured athlete, and has been the subject of intense research over the past decade. As a result, sound scientific principles have been applied to this realm of sports medicine, and have improved the outcome of both surgical and nonsurgical treatment. Possibly the most intriguing of these principles is the use ofthe kinetic chain concept in exercise prescription following ACL reconstruction. The hip, knee, and ankle joints when taken together, comprise the lower extremity kinetic chain. Kinetic chain exercises like the squat recruit all 3 links in unison while exercises such as seated quadriceps extensions isolate one link of the chain. Biomechanical assessment with force diagrams reveals that ACL strain is reduced during kinetic chain exercise by virtue of the axial orientation of the applied load and muscular co-contraction. Additionally, kinetic chain exercise through recruitment of all hip, knee, and ankle extensors in synchrony takes advantage of specificity of training principles. More importantly, however, it is the only way to reproduce the concurrent shift of 'antagonistic' biarticular muscle groups that occurs during simultaneous hip, knee, and ankle extension. Incoordination of the concurrent shift fostered by exercising each muscle group in isolation may ultimately hamper complete recovery. Modifying present day leg press and isokinetic equipment will allow clinicians to make better
Kinetic Chain Exercise in Knee Rehabilitation
403
use of kinetic chain exercise and allow safe isokinetic testing of the ACL reconstructed knee. Reconstruction of the ACL with a strong well placed graft to restore joint kinematics, followed by scientifically sound rehabilitation to improve dynamic control of tibial translation, will improve the outcome after ACL injury.
'Closed kinetic chain' exercise is touted as the best way to exercise the knee after anterior cruciate ligament (ACL) reconstruction and, as a result, has become a component of most ACL rehabilitation protocols (Anderson & Lipscomb 1989; Depalma & Zelko 1986; Giove et a1. 1983; Paulos et a1. 1981; Seto et a1. 1989; Silfverskold et a1. 1988). A review of the literature, however, fails to uncover convincing evidence to support these techniques or even an adequate explanation of the concept. This paper attempts to clarify the kinetic chain concept and verify theoretically the effect such exercises have on the knee. An understanding of these principles should make exercise prescription easier and rehabilitation more effective.
1. The Evolution of ACL Rehabilitation Rupture of the anterior cruciate ligament is one of the most common and devastating injuries an athlete can sustain. The devastation arises from the fact that returning enough stability to the ACL-deficient knee to allow a return to athletics is extremely difficult. There is still controversy in the literature regarding the treatment of choice for the torn ACL. Expert opinions vary from acute repair (Clancey et a1. 1988; Fetto & Marshall 1980; Hawkins et a1. 1986; Kennedy et a1. 1974; Pattee et a1. 1989; Straub & Hunter 1988) to late reconstruction (Noyes et a1. 1983) to nonsurgical rehabilitation (Balkfors 1982; Chick & Jackson 1978; Giove et a1. 1983; Marshall et a1. 1979). Many agree that good rehabilitation is imperative (Balkfors 1982; Chick & Jackson 1978; Clancey et a1. 1982; DePalma & Zelko 1986; Giove et a1. 1983) and some surgeons feel it is just as important as any operative procedure (Paulos et a1. 1981). Identifying exactly what constitutes a good ACL rehabilitation programme is the problem. Literature on knee rehabilitation prior to 1980 lacked sci-
entific backing (Gollanick et a1. 1976; Ryan & Allman 1974; Smillie 1971; Yamamoto et a1. 1976; Yocum et a1. 1978). McDaniel and Dameron (1980) found that those athletes with ACL injuries who achieved significant thigh hypertrophy on the involved side as compared to the uninvolved side were more successful at returning to athletic participation. Although this demonstrated the general need for rehabilitation, it did not shed any light on the specifics of exercise regimens. l.l The 'Paradox of Exercise'
Paulos et a1. (1981) found that ACL strain was dramatically increased during the last 30° of knee extension. This increase in strain was felt to be sufficient to endanger the healing graft. No increase in ACL strain was seen during knee flexion against resistance. Similar findings had previously been documented in the biomechanics literature (Lindahl & Movin 1967; Smidt 1973) and have since been confirmed (Arms et a1. 1984; Grood et al. 1984; Henning et al. 1985; Kaufman 1988; Renstrom et al. 1986). Giove et al. (1983) placed a group of patients with ACL-deficient knees on a progressive resistance exercise programme to promote thigh hypertrophy. They used a programme described by Smillie (1971), but emphasised hamstring strengthening. They found that higher levels of sports participation were achieved by the patients whose hamstring strength was equal to or greater than their quadriceps strength. Since hamstring strength in the normal population is usually only two-thirds of quadriceps strength, their findings suggested that overstrengthening of the hamstrings could help compensate for the loss of the cruciate ligament. It soon became common practice in ACL injury rehabilitation to emphasise hamstring strengthening,
404
while allowing only cautious quadriceps strengthening. Since powerful quadriceps are critical for functional and athletic use of the lower extremity, attempts were made to identify quadriceps exercises that would place minimal strain on the ACL. Arms (1984) and Renstrom (1986) demonstrated that the quadriceps could be exercised isometrically at knee flexion angles from 60 to 90° without causing undue strain in the ACL. It was becoming apparent, however, that one of the most common complications after anterior cruciate ligament reconstruction was patellofemoral pain syndrome (Clancey et al. 1982; Hejgaard et al. 1984; Marshall et al. 1979; Johnson et al. 1984; Kieffer et al. 1984; Straub & Hunter 1988). This disorder is exacerbated by performing resistive quadriceps strengthening exercises in greater than 30° of knee flexion (Hungerford & Barry 1979), making it difficult to recommend training in the 60 to 90° flexion range. Patellofemoral pain can be prevented and treated by strengthening the quadriceps in the 0 to 30° knee flexion range (Bourne et al. 1988), but this is the very range of motion which places maximal strain on the anterior cruciate ligament. Thus, what Paulos had called the 'paradox of exercise' has emerged. 1.2 Weightbearing (Kinetic Chain) Exercise Grood et al. (1984) looked at forces in ACLdeficient cadaver knees during simulated knee extension exercise. They documented increased anterior tibial displacement in the last 30° of knee extension and felt that this displacement would stretch the secondary restraints in the ACL-deficient knee if knee extension exercises were done injudiciously. They suggested exercising in the upright posture to allow the 'forces of weightbearing' to decrease tibial translation during quadriceps contraction. In support of this hypothesis, Henning et al. (1985) published their study of ACL strain in vivo. The study was performed by placing a strain gauge on the ACL of volunteers and then measuring ACL strain during various lower extremity exercises. They found that isometric knee extension at 0 and
Sports Medicine 11 (6) 1991
22° produced 5 to 17 times more strain on the ACL than weightbearing exercises such as stationary biking, walking on level ground, jumping rope, or performing a half squat with one leg. It should be emphasised that the study had only 2 subjects, showed wide variability of ACL strain between the 2 subjects and made no reference to joint angle during weightbearing exercise. Despite this, it is considered a landmark study which showed, for the first time, that the quadriceps can be strengthened without jeopardising the ACL, if weightbearing, or so-called 'closed kinetic chain', exercises are used. A thorough explanation of what exactly 'closed kinetic chain exercise' is, and how it decreases ACL strain, is needed, however.
2. The Origin of the Kinetic Chain Concept 2.1 Kinesiology A review of kinesiology reveals that the 'kinetic chain' terminology was derived by Steindler (1973) from the closed kinematic and link concepts set forth by Reuleaux for mechanical engineering purposes (Gowitzke & Millner 1988). In the link concept, rigid overlapping segments are connected in series by pin joints. The system is considered closed if both ends are connected to an immoveable framework, thus preventing translation of either the proximal or distal joint centre. This creates a system where movement at one joint produces movement at all other joints in a predictable manner, and is called a closed kinematic chain. The extremities can be thought of as rigid, overlapping segments in series. Steindler proposed that, although a closed kinematic chain never occurs in the extremities, two types of kinetic chain exist under different limb loading conditions. He observed that when the foot or hand 'meet considerable resistance', muscle recruitment and joint motion differ from that seen when the foot or hand is completely free to move. He felt the difference was significant enough to warrant distinguishing the 2 conditions with separate terms. Specifically, an open kinetic chain exists when the peripheral joint of the extremity can move freely, such as when
405
Kinetic Chain Exercise in Knee Rehabilitation
waving the hand or moving the foot forward in the swing phase of gait. A closed kinetic chain exists whenever the foot or hand meet resistance, such as in the chin-up or rising from a squat. He hastened to point out that a true closed kinetic chain only exists during isometric exercise, since by definition neither the proximal nor distal segment can move in a closed system. 2.2 Terminology There are numerous exercises available for strengthening the muscles which cross the knee joint. After publication of the aforementioned articles on ACL strain, there has been a trend in the rehabilitation community to use the open and closed kinetic chain terminology to divide these exercises into 2 major groups: open kinetic chain exercises are those in which the foot is free to move (seated leg extensions); and closed kinetic chain exercises are those in which the foot meets resistance (squats and leg presses). Indeed, there is a difference in ACL strain during the 2 types of exercise, but using the open and closed chain terminology to make the distinction is inaccurate and confusing. One could argue that neither exercise is a closed kinetic chain since either the proximal or distal segment moves in each situation. Conversely, one could argue that they are both closed kinetic chains, since the peripheral limb meets resistance in each situation. Applying accurate terms to these 2 exercise situations requires an understanding of how lower extremity kinetics are altered during weightbearing exercise.
3. The Theoretical Basis for Decreased ACL Strain
Fig. 1. Leji: anterior tibial translation as a consequence of isolated quadriceps contraction. Righi: stabilisation of tibia through hamstring co-contraction.
knee extension seems at first to be paradoxical, since this muscle group is classified as a primary knee flexor. Also, although authors have shown that co-contraction of the hamstrings does occur during nonweightbearing exercise (Baratta et al. 1988; Draganich et al. 1989; Kaufman 1988; Renstrom et al. 1986; Solomonow et al. 1987), Draganich (1989) demonstrated that this contraction is quite small (10 to 20% of maximal) and Renstrom (1986) demonstrated that it is relatively ineffective in decreasing ACL strain. During weightbearing exercise like the squat, however, more significant hamstring contraction is induced. Since the hamstrings are a biarticular muscle and also function as a strong hip extensor, forceful hamstring contraction is induced to stabilise the hip flexor moment. High tension in this muscle group then has a secondary effect on the knee. The reduction in knee shear that results can be illustrated with simple force diagrams. 3.2 Biomechanical Considerations
3.1 Hamstring Co-Contraction From a theoretical standpoint, the decrease in ACL strain seen during a weightbearing exercise has been explained by the fact that the hamstrings are active. Co-contraction of this muscle group helps neutralise the tendency of the quadriceps to cause anterior tibial translation, as shown diagramati cally in figure I. Hamstring contraction during
Figure 2 is a force diagram of the seated knee extension exercise. The system has been simplified such that only 3 fundamental forces are acting on the tibia: the applied force (A), the quadriceps muscle force (M), and the joint reaction force (R). If the tibia is in equilibrium, the line of application of all 3 forces must pass through a common point (d). Therefore, if the direction and point of appli-
406
Fig. 2. Force diagram. A = applied force; M = quadriceps muscle force; R =joint reaction force; f = knee instant centre of rotation; a,b,c = triangle formed by force vectors in equilibrium: used to determine length of unknown vector, R; d = common point of intersection of 3 force system in equilibrium; e---+g = shear component of R; e---+f = normal component of R.
cation of M and A are known, the direction of R can be determined. Vector R can then be resolved into its components, e-f (compression) and e-g (shear). The shear force is directed posteriorly, indicating that ifit were not for soft tissue constraint, the tibia would tend to translate anteriorly. Butler (1980) showed that 86% of this soft tissue constraint is provided by the ACL. Figures 3a-c show how variation in the location and orientation of the applied force influences the knee shear force. When A is applied proximally on the tibia, as in 3b, knee shear is dramatically decreased. This has been verified experimentally (Jurist & Otis 1985). Figure 3c shows the effect of changing the orientation of the applied force. If the force is applied with a more axial orientation, the shear component of the joint reaction force is, once again, smaller. Figure 3d shows the dramatic decrease in joint
Sports Medicine 11 (6) 1991
shear that occurs with hamstring co-contraction. In this situation, simultaneous contraction of the quadriceps, MI, and hamstrings, M2, produces a net force vector M3. As can be seen, the joint reaction force, R, caused by this net force, has a much smaller shear component than that seen during isolated quadriceps contraction. Kinetic chain exercise induces hamstring cocontraction. The reason for this is illustrated in figure 4. Consider a force A applied to the distal tibia. Since the hip and knee are unconstrained, the force creates a flexion moment at both joints. The hamstrings contract to stabilise the hip and the quadriceps contract to stabilise the knee. Activity in the biarticular hamstring muscle, induced for hip stability, helps neutralise the tendency of the quadriceps to cause anterior tibial translation. A similar phenomenon occurs at the ankle joint if the force A is applied to the bottom of the foot. Tension in the soleus, which would be induced to stabilise the ankle flexion moment, will have an indirect effect on knee shear. During exercise, the lower extremity kinetic chain is recruited when the hip, knee, and ankle are unconstrained and the force is applied axially, causing triarticular flexion. Therefore, knee shear forces are reduced by the mechanisms illustrated in both 3c and 3d. During seated leg extensions, the force is applied perpendicular to the tibia and causes a flexion moment at the knee only. Such joint isolation during exercise does not take advantage of the secondary stabilising effect of other muscles in the lower extremity kinetic chain. In other words, the kinetic chain is not exploited. These exercises should simply be called joint isolation exercises, while exercises such as the squat should simply be called kinetic chain exercises. Further description with the open and closed chain terminology is not necessary.
4. The Concurrent Shift 4.1 Specificity of Training Decreased shear at the knee is not the only advantage to the use of kinetic chain exercise in rehabilitation. Walking, running, jumping, climbing
Kinetic Chain Exercise in Knee Rehabilitation
407
and rising from a chair are all activities which exploit the kinetic chain. The lower extremity is usually used in this manner and should, therefore, be rehabilitated in this manner. Specificity of training is an accepted concept in rehabilitation and sports medicine (Fahey 1973; Harris 1984). It involves improving the strength and coordination of functional, or sport-specific, activities with exercises that approximate the desired activity. The progression from slow unresisted activity to quicker movement against resistance theoretically results in central nervous system engram patterning. Throughout this process, the emphasis is on precision of movement. Repetitive deviation from the correct movement pattern can result in substitution patterning. A substitution pattern is an incorrect and inefficient muscular recruitment pattern that can hinder performance and lead to injury. The importance of this concept in lower extremity rehabilitation becomes evident when the
kinesiology of kinetic chain exercise is considered. Consider the simultaneous hip and knee extension that occurs when rising from a squat. The rectus femoris and the hamstrings are both active. As the hip extends, the rectus femoris lengthens while the hamstrings shorten, but as the knee extends the rectus femoris shortens as the hamstrings lengthen. The result at the muscular level might be called a pseudoisometric contraction, due to simultaneous concentric and eccentric contractions at opposite ends of each muscle. This so-called concurrent shift (Steindler 1973) and the resultant pseudoisometric contraction cannot be reproduced with isolation exercises. Specificity of training must also be observed to ensure carryover of strength gains to functional activities on the playing field. Research has shown that isometric, isotonic and isokinetic strength gains are specific to the type of contraction used during training (Atha 1981). The same may hold true for
c
B M
D
s
s M,
M
A
,, 1 ,1
I 1 ,I
,
..
,, ,
I
.:' t', 'I
,, /
,,
,, ,, ,
" ,,
Fig. 3. Force diagrams demonstrating altered joint reaction force. In each example, C represents the compression component of the joint reaction force and S represents the shear component of the joint reaction force. A. Knee extension exercise with force applied on distal tibia. B. Knee extension with force applied proximally. C. Axial orientation of applied load. D. Quadriceps and hamstring co-contraction.
408
the unique type of muscular contraction that occurs in the hamstrings and rectus femoris during the concurrent shift. It has also been demonstrated that using a combination of isometric, concentric and eccentric contractions in an exercise routine results in larger strength gains than anyone alone (Atha 1981). All 3 types of contraction occur during each repetition of a kinetic chain exercise. Neural adaptation is another benefit of strength training that will be affected by the type of exercise chosen. The literature has repeatedly shown that strength gains seen during the first 4 weeks of training are primarily due to neural adaptation, not muscular hypertrophy (Dons et al. 1979; Fahey & Brown 1973; Kanehisa & Miyashita 1983; Thorstensson et al. 1976). Proposed mechanisms of this effect include: enhancement of the ability to maximally recruit agonist muscles, improved inhibition of antagonist muscles, improved synchronisation of motor unit recruitment or a combination thereof (Jones et al. 1986). Studies have shown that integrated EMG and synchronisation of contraction decrease with immobilisation and detraining (Jones et al. 1986). Since the neural adaptation that
Sports Medicine 11 (6) 1991
occurs in the lower extremity will differ dramatically as a function of whether the kinetic chain is recruited or the joint is isolated, lower extremity rehabilitation protocols must incorporate kinetic chain exercise early. 4.2 Effect of the Concurrent Shift on Other Joints Kicking is the only activity that requires powerful isolated quadriceps contraction. Therefore, strengthening the thigh with isolation exercises has no practical application for most patients. Repetitive recruitment of the quadriceps in isolation may be contraindicated since it causes abnormal engram patterning. Poor control or incoordination of the concurrent shift during kinetic chain activity will affect forces at the tibiofemoral, patellofemoral, lumbosacral and intervertebral joints, since the muscles involved cross both the knee and hip. The effect of the concurrent shift at joints other than the tibiofemoral joint raises some interesting questions. Imprecision of the concurrent shift may increase patellofemoral joint forces, and, therefore, may be a factor in anterior knee pain. Treatment of this disorder has traditionally included only joint isolation exercises (quad sets, and short arc quadriceps extensions), which may explain why so many of these cases are refractory to treatment. The effect of the hamstrings on pelvic tilt and, therefore, on lumbosacral and intervertebral joint forces is well known. Therefore, imprecision of the concurrent shift may also be a contributing factor in low back pain. Kinetic chain exercise may have a role in the treatment and prevention of patellofemoral pain and low back pain.
5. Clinical Application of the Kinetic Chain Concept
Fig. 4. Diagramatic representation of the induction of muscular co-contraction due to flexion moments caused by the applied force. A = applied force; rH = hip moment arm; rK = knee moment arm.
The use of kinetic chain exercise during anterior cruciate ligament rehabilitation makes sense from the theoretical standpoint. Unfortunately, scepticism still exists in the rehabilitation community. As a result, most ACL rehabilitation protocols emphasise joint isolation exercise early on and only
Kinetic Chain Exercise in Knee Rehabilitation
allow kinetic chain exercise in the late phases (Anderson & Lipscomb 1989; DePalma & Zelko 1986; Giove et al. 1983; Paulos et al. 1981; Seto et al. 1989). This appears to be a reversal of what should be done. The persistent use of joint isolation exercise in ACL rehabilitation may be partially responsible for the fact that most postreconstruction failures are due to stretching of the graft over time (Johnson et al. 1984). Similarly, follow-up studies of patients with ACL-deficient knees have shown that varus or valgus laxity occurs over time (Butler et al. 1980; McDaniel & Dameron 1980; Giove et al. 1983; Hejgaard et al. 1984). Repetitive loading of the secondary restraints during isolated knee extension exercise may be a contributing factor. In addition, isokinetic machines, which are universally used for strength tests, isolate the knee joint. The subjects are forced to maximally contract the quadriceps without the protection of hamstring co-contraction. Solomon ow and co-workers (1987) showed that a protective reflex arc exists between the ACL and hamstrings. With 'overloading' of the ACL, there is a reflex contraction of the hamstrings to protect the ligament. This reflex was recruited during isokinetic knee extension in their study. It appears that isokinetic testing may be contraindicated in ACL reconstructed knees (Paulos et al. 1981; Solomonow et al. 1987), and may excessively load the secondary restraints in ACL-deficient knees (Grood et al. 1984). The implications of such overloading in normal knees needs further research. 5.1 Effective Recruitment of the Kinetic Chain During Exercise Merely having the foot meet 'considerable resistance' does not guarantee the neutralisation of knee shear, however. Moments at the hip and knee will vary depending on the type of exercise performed. Figure 5 illustrates 2 weightbearing exercises that do not neutralise knee shear. In figure Sa, the intersection of the force lines ofbodyweight and the support force from the wall define the point through which the ground reaction force line must pass. As can be seen, the knee moment created by
409
A
B
Fs
Fig. 5. Hamstring co-contraction is minimal in these two exercise situations. A. The intersection of force line S (support force) and W (bodyweight) defines a point through which the force line of the ground reaction force must pass, producing a large knee moment, fK, and a small hip moment, rHo B. Arm support can easily stabilise the hip flexion moment due to its large moment arm fH2 compared to the moment arm of the ground reaction force, rH 1.
this force is large while the hip moment is small. In figure 5b, the large moment arm of the upper extremity support force allows the arm to easily overcome the hip flexion moment. 5.2 Applying Biomechanical Principles to the Leg Press Although power squats are difficult to perform when the patient is weak and in pain, the leg press, if done correctly, takes full advantage of the kinetic chain while simultaneously providing stability to the patient, decreasing strain on the low back, allowing exercise with less than bodyweight and allowing exercise of each leg independently. In order to accomplish this, however, certain modifications to the traditional leg press must be made. First, the machine must allow full hip extension. Traditional leg press machines allow, at best, -45 of full hip extension. The hamstrings, therefore, are not appropriately recruited, and the kinetic chain is not fully exploited. Full hip extension can be achieved by placing the patient in a supine position. If full hip and knee flexion and extension occur, the concurrent shift will be repro0
410
a
b
Fig. 6. Illustration of the line of force application relative to the tibia, hip, and knee for various leg press exercise situations. (In 6a, the resistance is applied horizontally via a pulley system.)
duced and appropriate hamstring recruitment will be assured. Second, the foot plate should move in an arc. An arc of motion will maintain the line of force application axial to the tibia, as was previously shown in figure 3c. If, on the other hand, the foot plate slides horizontally, or on a 4SO angle (figs 6a and b), a large component of the applied force is directed perpendicular to the tibia as flexion angles increase. The knee shear in this case will increase approaching the isolation exercise situation.
Sports Medicine 11 (6) /99/
An arc of motion also improves hamstring recruitment. The magnitude of hamstring contraction increases directly with hip moment arm. A moment arm is equal to the perpendicular distance from the joint centre of rotation to the line of force application. Therefore, only a small hip flexion moment is produced in the sliding foot plate configurations, since the line of force application passes near the hip throughout the range of motion (figs 6a and b). In addition, as the weight is lowered, the knee moment increases dramatically. The imbalance in muscular recruitment that results increases shear forces at the knee. Allowing the foot plate to move in an arc, gradually moves the line of force application away from the hip, increasing the hip flexion moment and decreasing the knee moment (fig. 6c). In the ideal situation the two moments will be nearly equivalent, thus neutralising anteroposterior shear. Finally, the foot plate must be fixed perpendicular to the frontal plane of the hip during this arc of motion, as in figure 6c. This can be accomplished with a 4-bar linkage. Fixation of the footplate in this plane once again helps maintain the line of force application axial to the tibia, but also capitalises on the knee extension moment that the soleus is capable of producing. Soleus activity, as stated earlier, helps neutralise the knee shear force. If the foot plate is allowed to move freely on the lever arm, plantarflexion merely rotates the plate, and the potential knee extensor moment is lost. Adding these simple modifications to the traditionalleg press apparatus comes very close to reproducing the squat. During the parallel squat, the trunk inclines forward as the weight is lowered, thus increasing the hip moment (fig. 7). This trunk flexion simultaneously decreases the knee flexion moment by moving the weight and/or centre of gravity forward. As a result of this decrease in knee moment arm, patellofemoral compression forces and knee shear forces remain low. This movement pattern occurs with all functional activities, including stair climbing, rising from a chair, jumping etc. In each case, the centre of gravity is shifted forward, increasing hip moment and decreasing knee moment.
Kinetic Chain Exercise in Knee Rehabilitation
Reproducing this movement pattern in the leg press would make it safe and functional for lower extremity rehabilitation. Similar modification of the isokinetic lever arm attachment will allow safe and functional isokinetic testing of the lower extremity. Development and biomechanical testing of such an apparatus is in progress at the Biomechanics Laboratory, Mayo Clinic, Rochester, MN. Until adequate equipment is developed, kinetic chain exercises such as multiangled isometric leg press, power squats, stair climbing and stationary biking should be emphasised in rehabilitation protocols. The rehabilitation team must, however, be aware of the biomechanical principles discussed herein if such exercises are expected to neutralise knee shear forces.
Fig.7. Relationship of the line offorce application to the hip and knee during performance of the squat. rH '" hip moment arm; rK '" knee moment arm.
411
6. Implications for Treatment The theory behind kinetic chain exercise and its effect on the knee would suggest that, although the ACL is the primary restraint to anterior tibial translation in the cadaver (Butler et al. 1980), it is not a check rein to translation under dynamic conditions. During activity, muscular co-contraction should neutralise the shear at the knee. The anterior and posterior cruciate ligaments then act as passive guides to the combined rolling and gliding motion of the femoral condyles on the tibial plateau as described in Mueller's 4-bar linkage model (Mueller 1983). In such situation, absolute strain on the ACL measured during maximal kinetic chain knee extension exercise would be no higher than the strain measured during active knee range of motion. Suggesting that the anterior cruciate ligament does not function as a 'check rein' is not meant to detract from its role in knee function. Indeed, the passive guiding function of the cruciates is critical to normal joint kinematics. Gerber and Matter (1983) demonstrated that disruption of the ACL alters knee kinematics such that dramatic changes in knee instant centre of rotation are seen with simple, unresisted, active range of motion. These changes produce abnormal loading of the joint surfaces which could theoretically lead to meniscal injury and degenerative changes. In addition, abrupt shifting of the joint centre of rotation during weightbearing may be a cause of instability and give-way symptoms. These biomechanical and kinesiological considerations make formulation of a treatment plan for the ACL injured knee less mysterious. Rehabilitation is not capable of returning normal joint kinematics to an ACL-deficient knee, but surgical reconstruction followed by a nonspecific and illadvised rehabilitation protocol is also doomed to failure. It would seem that a combined surgical/ rehabilitative effort is the key to success. Considering the 'surgical versus nonsurgical' debate that has permeated the literature to date, the confusion and generally dismal long term treatment results seen are not surprising.
a
Sports Medicine 11 (6) 1991
412
7. Conclusion Despite the fact that isolated knee extension exercise has repeatedly been shown to cause excessive strain on the ACL, many knee rehabilitation protocols continue to include or even emphasise these exercises. Biomechanical theory suggests that kinetic chain exercises neutralise knee shear forces through the induction of muscular co-contraction. Such exercise is, therefore, critical for successful rehabilitation of the ACL-reconstructed or ACL-deficient knee, but it must be appropriately prescribed. Although further research is needed in this area, the data presently available indicate that isolated knee extension exercises are contraindicated in these patients and that appropriately prescribed kinetic chain exercise can be instituted very early, if not immediately, after surgery. Kinetic chain exercise does more than stabilise the knee joint, however. The concurrent shift that occurs during such exercise causes unique muscular contraction and intricate muscular interactions that cannot be reproduced with joint isolation exercises. As such, specificity of training becomes a significant factor. Emphasising kinetic chain exercise prior to returning any lower extremity injured patient to strenuous weightbearing activity has theoretical merit. The incorporation of kinetic chain exercise in the rehabilitation protocols of other musculoskeletal disorders such as patellofemoral pain and low back pain should be considered. In addition, exercises that isolate ·specific joints in the lower extremity kinetic chain do not prepare the patient for weightbearing activity and may in fact predispose the patient to injury secondary to inappropriate muscular and neuromuscular training. Development of versatile equipment that safely and effectively recruits the kinetic chain will promote the incorporation of kinetic chain exercises early in rehabilitation protocols and testing procedures. Once this is accomplished, the treatment of such common musculoskeletal problems as anterior knee pain, ACL injury and low back pain may prove to be less frustrating.
Acknowledgements We would like to acknowledge the Department of Physical Medicine and Rehabilitation, Mayo Clinic, Rochester, MN, for funding this work, and Mr Fred Schultz, who had an integral part in the development of the biomechanically sound leg press apparatus described in this article.
References Anderson AF, Lipscomb AB. Analysis of rehabilitation techniques after anterior cruciate reconstruction. American Journal of Sports Medicine 17 (2): 154-160, 1989 Arms SW, Pope MH, Johnson RJ, et al. The biomechanics of anterior cruciate rehabilitation and reconstruction. American JournalofSports Medicine 12 (I): 8-18,1984 Atha J. Strengthening muscle. Exercise and Sports Science Reviews 9: 1-73, 1981 Balkfors B. The course of knee ligament injuries. Acta Orthopaedica Scandinavica 53 (Suppl. 198): 179, 1982 Baratta R, Solomonow M, Zhou BH, et al. Muscular coactivation: the role of the antagonist musculature in maintaining knee stability. American Journal of Sports Medicine 16 (2): 113-122, 1988 Bourne MH, Hazel WA, Scott SG, et al. Anterior knee pain. Mayo Clinic Proceedings 63 (42): 491, 1988 Butler OL, Noyes FR, Grood ES. Ligamentous restraints to anterior-posterior drawer in the human knee: a biomechanical study. Journal of Bone and Joint Surgery 62-A: 259-270, 1980 Chick RR, Jackson OW. Tears of the anterior cruciate ligament in young athletes. Journal of Bone and Joint Surgery 60-A: 970-973, 1978 Oancey WG, Nelson OA, Reider B, et al. Anterior cruciate ligament reconstruction using one-third of the patellar tendon ligament augmented by extra-articular tendon transfers. Journal of Bone and Joint Surgery 64-A (3): 352-359, 1982 Oancey WG, Ray JM, Zolten PJ. Acute tears ofthe anterior cruciate ligament - surgical vs. conservative treatment. Journal of Bone and Joint Surgery 70-A (10): 1483-1388, 1988 DePalma BF, Zelko RR. Knee rehabilitation after anterior cruciate ligament injury or surgery. Athl Training, Fall 1986 Dons B, Bollerup K, Bonde-Petersen F, et al. The effect of weightlifting exercise related to muscle fibre composition and muscle cross-sectional area in humans. European Journal of Applied Physiology 40: 95-106, 1979 Draganich LF, Jaeger RJ, Knalj AR. Coactivation of the hamstrings and quadriceps during extension of the knee. Journal of Bone and Joint Surgery 71-A (7): 1075-1080, 1989 Fahey TO. Physiologic adaptation to conditioning. In Fahey TO (Ed.) Athletic training: principles and practice, pp. 57-78, Mayfield Publishing, Palo Alto, CA, 1986 Fahey TO, Brown CH. The effects of an anabolic steroid on the strength, body composition and endurance of college males when accompanied by a weight training program. Medicine and Science in Sports 5: 272-276, 1973 Fetto JF, Marshall JL. The natural history and diagnosis of anterior cruciate ligament insufficiency. Clinical Orthopedics 147: 29-38, 1980 Gerber C, Matter P. Biomechanical analysis ofthe knee after rupture of the anterior cruciate ligament and its primary repair: an instant-centre analysis of function. Journal of Bone and Joint Surgery 65: 391, 1983 Giove T, Miller S, Kent B, et al. Nonoperative treatment of the
Kinetic Chain Exercise in Knee Rehabilitation
torn anterior cruciate ligament. Journal of Bone and Joint Surgery 65-A (2): 184-192, 1983 Gollanick PD, Erickson E, Harmark T, et al. Rehabilitation of the knee following surgery. Medicine and Science in Sports 8: 133-134, 1976 Gowitzke BA, Millner S. Scientific bases of human movement, pp. 1-3, Williams and Wilkins, Baltimore, 1988 Grood ES, Suntay WJ, Noyes FR, et al. Biomechanics of knee extension exercise. Journal of Bone and Joint Surgery 66-A (5): 725-734, 1984 Harris FA. Facilitation techniques and technology adjuncts in therapeutic exercise. In Basmajian JV (Ed.) Therapeutic exercise, 9th ed., pp. 110-178, Williams and Wilkins, Baltimore, 1984 Hawkins RJ, Misamore JW, Merritt TR. Followup of the acute nonoperated isolated anterior cruciate ligament tear. American Journal of Sports Medicine 14: 205-210, 1986 Hejgaard N, Sandberg H, Haad A, et al. The course of differently treated isolated ruptures of the anterior cruciate ligament as observed by prospective stress radiography. Clinical Orthopedics 182: 236-241, 1984 Henning CE, Lych MA, Glick JR. An in vivo strain gauge study of elongation ofthe anterior cruciate ligament. American Journal of Sports Medicine 13 (I): 22-26, 1985 Hungerford DS, Barry M. Biomechanics of the patellofemoraljoint. Clinical Orthopedics 144: 9-15,1979 Johnson RJ, Erikson E, Haggmark T, et al. Five to ten year follow-up evaluation after reconstruction of the anterior cruciate ligament. Clinical Orthopedics 183: 122-140, 1984 Jones NL, McCartney N, McComas AI. Human muscle power, Human Kinetics, Champaign, IL, 1986 Jurist KA, Otis Je. Anteroposterior tibiofemoral displacements during isometric extension efforts. American Journal of Sports Medicine 13 (4): 254-258, 1985 Kanehisa H, Miyashita M. Effect of isometric and isokinetic muscle training on static strength and dynamic power. European Journal of Applied Physiology 52: 104-106, 1983 Kaufman KR. Mathematical model of knee forces during exercise. PhD thesis, North Dakota State University, 1988 Kieffer DA, Curnow RJ, Southwell RB, et al. Anterior cruciate ligament arthroplasty. American Journal of Sports Medicine 12: 301, 1984 Kennedy JC, Weinberg HW, Andrews AS. The anatomy and function of the anterior cruciate ligament. Journal of Bone and Joint Surgery A (2): 223-235, 1974 Lindahl 0, Movin A. The mechanics of the knee joint. Acta Orthopedica Scandinavica 38: 226-234, 1967 Marshall JL, Warren RF, Wickiewicz VB, et al. The anterior cruciate ligament - the technique of repair and reconstruction. Clinical Orthopedics 143: 97, 1979 McDaniel WJ, Dameron TD. Untreated rupture of the anterior cruciate ligament: a followup study. Journal of Bone and Joint Surgery 62-A (5): 696-705, 1980
413
Mueller W. The knee: form, function and ligament reconstruction, Berlin Heidelberg New Y orlc, 1983 Noyes F, Movar PA, Matthews DS, et al. The symptomatic anterior cruciate deficient knee, Part I: The long term functional disability in athletically active individuals. Journal of Bone and Joint Surgery 65-A (2): 154-162, 1983 Pattee GA, Fox JM, Pizzo WD, et al. Four to ten year followup of uureconstructed anterior cruciate ligament tears. American Journal of Sports Medicine 17 (3): 430-435, 1989 Paulos L, Noyes FR, Grood E, et al. Knee rehabilitation after anterior cruci;lte ligament reconstruction and repair. American Journal of Sports Medicine 9 (3): 140-149, 1981 Renstrom P, Stanwyck TS, Johnson RJ, et al. Strain within the anterior cruciate ligament during a hamstring and quadriceps activity. American Journal of Sports Medicine 14 (I): 83-87, 1986 Ryan AI, Allman FL. Rehabilitation of the knee following anterior cruciate ligament injury. In Sports medicine, pp. 541548, New York Academy Press, New York, 1974 Seto JL, Brewster CE, Lombardo SJ, et al. Rehabilitation of the knee after anterior cruciate ligament reconstruction. Journal of Orthopedics and Sports Physical Therapy 8-18, 1989 Silfverskold JP, Steadman JR, Higgins RW, et al. Rehabilitation of the anterior cruciate ligament in the athlete. Sports Medicine 6: 308-319, 1988 Smidt GL. Biomechanical analysis of knee flexion and extension. Journal of Biomechanics 6: 79-92, 1973 Smillie IS. Injuries of the knee joint, 4th ed., Churchill Livingstone, Edinburgh, 1971 Solomonow M, Baratta R, Zhou BH, et al. The synergistic action of the anterior cruciate ligament and thigh muscles in maintaining joint stability. American Journal of Sports Medicine 15 (3) 207-213, 1987 Steindler A. Kinesiology of the human body under normal and pathological conditions, p. 63, Charles e. Thomas, Springfield IL, 1973 Straub T, Hunter R. Acute anterior cruciate ligamcent repair. Clinical Orthopedics 227: 238-250, 1988 Thorstensson A, Karlsson J, Viitasalo JHT, et al. The effect of strength training of EMG of skeletal muscle. Acta Physiologica Scandinavica 98: 232-236, 1976 Yamamoto SK, Hartman CW, Feagan JA, et al. Functional rehabilitation of the knee: a preliminary study. American Journal of Sports Medicine 6: 288-293, 1976 Yocum LA, Bohman DC, Noble HB, et al. The deranged knee, restoration of function - a protocol for rehabilitation of the injured knee. American Journal of Sports Medicine 6: 51-53, 1978 Correspondence and reprints: Randal A. Palmitier, START Physical Medicine Clinic, 1910 E. Southern Ave, Tempe, AZ 85282, USA.
