Anterior Cruciate Ligament Injury Rehabilitation in Athletes

INJURY CLINIC Sports Mad. 1996 Jut; 22 (I): 54-64 0112-1642/96/0007 -0054/S05.5O/0

© Adis International limited. All rights reserved

Anterior Cruciate Ligament Injury Rehabilitation in Athletes Biomechanical Considerations

Bruce D. Beynnon and Robert J. Johnson

Department of Orthopaedics and Rehabilitation, McClure Musculoskeletal Research Center, Stafford Hall, University of Vermont, Burlington, Vermont, USA

Contents Summary , , , , , , , , , , , , , , , " ","',',",',' 54

55 56 58 58 59 61 61 62 62

1, The Anterior Cruciate Ligament (ACL) and In Vivo Rehabilitation Exercises 2, ACL Disruption and the Neuromuscular System 3, The Bone-Patellar Tendon-Bone Graft In Vivo 4, Functional Knee Braces ' , , , , , , , , 5, Healing ACL Grafts in Animals , , , , , 6, Rehabilitation and Healing ACL Grafts

6,1 Retrospective Studies 6,2 Prospective Studies,

7, Conclusions, , , , , , , , ,

Summary Postoperative rehabilitation is a major factor in the success of an anterior cruciate ligament (ACL) reconstruction procedure. Clinical investigations ofpatients after ACL reconstruction have shown that immobilisation of the knee, or restricted motion without muscle contraction, leads to undesired outcomes for the articular, ligamentous, and musculature structures that surround the knee.

Early joint motion is beneficial for; reducing pain, capsular contractions, articular cartilage, and for minimising scar formation that limit joint motion. These findings, combined with graft materials that have biomechanical properties similar to the normal ACL, and adequate fixation strength, have led many to recommend aggressive rehabilitation programmes that involve contraction of the dominant quadriceps muscles.

Recently, a prospective, randomised study of rehabilitation following ACL reconstruction has presented evidence that a closed kinetic chain exercise programme (foot fixed against a resistance) results in anterior-posterior knee laxity values that are similar to the contralateral normal knee. Also, open kinetic chain exercises (foot not fixed against a resistance) result in increased anterior-posterior knee laxity compared with the normal knee. Criteria must be observed because the relationship between rehabilitation exercises and the healing response of an ACL graft is unknown at present.

Biomechanical studies of healing ACL grafts performed in animals have shown that the graft requires a long time to revascularise and heal, and that the

ACL Injury Rehabilitation in Athletes 55

biomechanical behaviour of the graft never returns to normal. Functional knee braces provide a protective strain-shielding effect on the ACL when anterior shear loads and internal torques are applied to the knee in the non-weight-bearing condition. However, the strain shielding effect of functional braces decrease as the magnitude of anterior shear and internal torque applied to the knee increase. Future studies should strive to determine the actual loads transmitted across the knee and ACL graft strain during various rehabilitation exercises and relate these to the healing response of the knee and graft.

In a recent review, Buckwalter stated 'for at least 250 years physicians have recognised that loading and movement of musculoskeletal tissues caused by physical activity alters these tissues and the structures they form, and may affect healing; yet the appropriate role of activity in the treatment of injuries has been a subject of controversy.'! I] This statement certainly pertains to healing anterior cruciate ligament (ACL) grafts in that there is currently little information available in the literature regarding the optimal means to rehabilitate a knee after ACL reconstruction. 15 years ago, rehabilitation programmes included immobilisation of the leg for 6 weeks or more after an ACL reconstruction procedure while inflammation diminished and the ACL graft healed.f2·3]

Biomechanical studies performed on animals have documented the adverse effect of knee immobilisation on the articular cartilage, ligaments, capsular structures, leg musculature and periarticular bone.!4-IO] This led some to advocate early motion rehabilitation programmes that included guarded early motion in a knee brace,15] or continuous passive knee motion l II] immediately after ACL reconstruction. Other studies indicate that early mobilisation of healing ACL grafts, including immediate full weight-bearing activities such as walking, is possible without endangering the healing tissues.l 12-14]

Immobilisation after ACL reconstruction results in undesirable effects on the reconstructed knee; however, it is currently unknown how much activity will promote adequate rehabilitation of an injured knee without permanently elongating the graft producing graft failure, or resulting in failure of graft fixation . This paper reviews the current literature concerning the effects of rehabilitation

© Adis International Limited. All rights reserved.

programmes on the biomechanical behaviour of ACL grafts.

1. The Anterior Cruciate Ligament (ACL) and In Vivo Rehabilitation Exercises

Our group has measured the strain behaviour of the anteromedial aspect of the normal ACL with the goal of describing the effect of muscle forces and joint position on ACL strains.l 15] This work has provided an objective database for prospective investigations of the effect of rehabilitation exercises on healing ACL grafts. We have measured ACL strain with the Hall effect strain transducer and more recently with the differential variable reluctance transducer. Study participants were patient volunteers with normal ACLs undergoing arthroscopic meniscal surgery or diagnostic procedures under local anaesthesia. After the routine surgical procedure, the differential variable reluctance transducer was implanted into the ACL (fig. 1). ACL strain values varied with knee position and muscle contraction for simultaneous contraction of the quadriceps and hamstrings muscles, isometric quadriceps contraction, and active flexion/extension of the leg. Simultaneous quadriceps and hamstrings contraction at 15° of knee flexion created an increase in ACL strain, but the ACL remained unstrained at 30°, 60° and 90° of knee flexion (fig. 2). Isometric contraction of the quadriceps at 15° and 30° of knee flexion results in a significant increase in ACL strain values, but at 60° and 90° of flexion this activity does not change ACL strain compared with the relaxed muscle condition (fig. 3). ACL strain values increased with active motion of the knee from a flexed to an extended position (fig. 4).

Sports Med. 1996 Jul: 22 (l)

56

Addition of a 45N weight to the foot during active flexion/extension produced a significant increase in ACL strain at 10° and 20° of flexion relative to the same activity without weights applied. ACL strain values did not increase with isometric contraction of the hamstrings (fig. 5).

During early healing of an ACL graft, weightbearing (closed kinetic chain) exercises have been advocated because the compressive joint load produced by body weight is thought to force congruent articular surfaces together and protect a healing ACL graft. Exercises that involve contraction of the dominant quadriceps in the non-weight-bearing knee (open kinetic chain) are thought to be antagonistic to a healing graft and are usually prescribed after adequate graft healing has occurred.

Many investigators have claimed that closed kinetic chain exercises are safer and can be used earlier following ACL reconstruction than open kinetic chain activities.[16-21 1 Yet no valid evidence exists to support this conclusion.

Posterior cruciate

HEST

Fig. 1. An anterior view of the knee, with the anterior cruciate and posterior cruciate ligaments exposed. Shown is the orientation of the Hall effect strain transducer (HEST) prior to implantation on the anteromedial aspect of the anterior cruciate ligament. Recent investigations have used the differential variable reluctance transducer (DVRT) to measure the strain behaviour of the anterior cruciate ligament (from Beynnon et al.,psI with permission).

© Adis International Limited. All rights reseNed.

6 5 4

l 3 c 2 '§

1 iii 0

- 1 - 2 - 3

15 30

Beynnon & Johnson

• Relaxed (n = 8) o Maximum (n = 8)

60 90

Knee flexion angle (' )

Fig. 2. Anterior cruciate ligament (ACL) strain values produced by simultaneous contraction of the leg musculature at 15°, 30°, 60° and 90° of knee flexion angle. The ACL strain values depended on both knee flexion angle and the level of muscle activity, There was a significant increase in ligament strain relative to the relaxed muscle condition at 15° of flexion, ACL strain values did not change with simultaneous contraction of the muscles at 30°,60° and 90° of flexion (from Beynnon et al.,1151 with permission) .

Exercises that develop low ACL strain values and would be safe for a healing ACL graft include:

• isometric hamstrings contraction at all knee flexion angles

• contraction of the dominant quadriceps muscles with the knee flexed at 60° or greater (isometric quadriceps and simultaneous quadriceps and hamstrings contraction exercises)

• active flexion of the reconstructed knee between 40° and 90° of flexion.

These data are summarised in table I, and are rank ordered according to the different peak strain value developed during the rehabilitation exercise.

2. ACL Disruption and the Neuromuscular System

As early as 1826, Bell observed that the sensory nerves of the dorsal spinal roots were involved with the sensation of movement and position of the lower extremitiesp21 80 years later the term 'proprioception' was introduced by Sherrington to characterise an individual's ability to sense the movement and position of a limb through the nerve receptors in the joint capsule, ligaments, tendons and musclesP31 Proprioception has since been evaluated with many different techniques; and as a

Sports Med. 1996 Jul; 22 (1)

ACL Injury Rehabilitation in Athletes

consequence the literature is unclear regarding the effect of ACL disruption on knee proprioception.

Barrett[24] evaluated proprioception by measuring joint position sense (JPS) using a visual analogue of knee position in space, and reported that ACL disruption changed the individual's ability to approximate knee joint position. When evaluated by the threshold to detection of joint motion (kinesthesia), proprioception was altered by ACL disruption according to Barrack et al.[25] In contrast, we studied individuals with acute, isolated, ACL tears that were arthroscopically documented and found that, with individuals in a standing position, and either a passively or actively set index angle, JPS was not changed by ACL disruptionJ26] These findings are in agreement with the reports of Corrigan et al.[27] and Klein et al.!28] who made similar

measurements of JPS with individuals in a seated position. These studies indicate that a torn ACL does not affect the ability to reposition the knee to a predetermined indexed position. It may be that more sensitive measurement techniques need to be developed to assess the effect of compromised ACL mechanoreceptor activity on knee joint proprioception.

It has been suggested that disruption of the ACL causes an adverse change in the closed loop efferent reflex pathway of the leg musculature. This has led to the speculation that the leg muscles cannot produce an adequate reflex contraction in response

6 5 4

~ 3 c: 2 .~

in 1 0

- 1 - 2 - 3

15 30

• Relaxed (n = 8) o 15N'm(n = 8) D30N ' m(n=5)

60 90

Knee Ilexion angle (' )

Fig. 3. Anterior cruciate ligament strain (ACL) values produced by isometric quadriceps contraction at 15°, 30°, 60° and 90° of knee flexion. The ACL strain values were dependent on both knee flexion angle and the level of muscle activity (from Beynnon et al.,115J with permission).


6 5 4

~ 3 c 2 .~

1 in 0

- 1 - 2 -3

0

- Normal (n = g) ---- 45N Boot (n = 9) --. Flexion +-- Extension

10 20 30 40 50 60 70 80 90 100

Knee flexion angle n

57

Fig. 4. Anterior cruciate ligament (ACL) strain values produced by active flexion-extension of the lower leg and the same activity with a 45N weight boot attached to the foot. For both exercises, ACL strain values depended upon both knee flexion angle and the muscle activity level. The ACL strain values increased as the knee was brought into an extended position. There was significantly greater ACL strain for active motion of the knee with the 45N weight at 10° and 20° of flexion compared with the same activity without weights (from Beynnon et al.,115J with permission).

to an instability episode, resulting in an inhibition of the stabilising effect of the leg muscles on the knee, possibly leading to subluxation of the joint. The efferent reflex pathway has been evaluated by measuring the latency of knee muscle contraction (the time delay between knee joint perturbation and muscle activation).

The latency of reflex contraction for the biceps femoris muscle in ACL-deficient individuals is 15 msec longer than that of the contralateral normal knee, according to Beard and coworkers.[29] In contrast, Jennings and Seedholm[30] reported no difference in the latency of reflex contraction of the biceps femoris between normal and ACL-deficient knees. We also found no significant difference in latency of reflex contraction between the ACL deficient and contralateral normal limbs for the vastus lateralis, gastrocnemius and biceps femoris musclespll One possible explanation for the difference between our findings and those of Beard et al.[291 is that we studied individuals who adapted well to their ACL injury, did not have a detectable pivot shift exam and did not experience episodes of joint instability. It may be that the longer latency of reflex muscle contraction reported by Beard et alp91 was produced not only by the disrupted ACL, but also caused by the permanent deformation of

Sports Med. 1996 Jul; 22(1)

58

6 5 4

l 3 c: 2

'~ 1 (jj 0

- 1 - 2 - 3

15 30

• Relaxed (n = 8) D - l0N ·m(n=8)

60 90 Knee flexion angle (')

Fig. 5. Anterior cruciate ligament strain values produced by isometric hamstrings muscle contraction at 15°, 30°, 60° and 90° of flexion. At all 8 positions there was no change in ligament strain with contraction of the hamstrings muscles compared with the relaxed condition (from Beynnon et al.psi with permission).

the secondary structures that function as restraints in the absence of the ACL and by an increase in anterior knee joint laxity. This hypothesis is supported by the work of Wojtys and Huston[321 who reported that the muscle timing and recruitment order that occurs in response to an anterior tibial perturbation are related to the amount of anterior tibial laxity and vary with the amount of time from the initial ACL injury.

3. The Bone-Patellar Tendon-Bone Graft In Vivo

The deleterious increase in anterior laxity of the knee joint that occurs after ACL reconstruction (anterior translation of the tibia relative to the femur) have been attributed to the temporal changes in the material behaviour (strength and stiffness) of the graft that occur during remodellingJ331 However, we used the Hall effect transducer to measure the changes in the length of the midsubstance portion of the bone-patellar tendon-bone (B-PT-B) graft immediately after its fixation, while the knee was moved through passive flexion/extension motion. This demonstrated that the length pattern of the graft was dissimilar to the normal anterior cruciate in that it changed during the initial 20 cycles of passive knee motion. This phenomenon was defined as the cyclic response of the graft (fig. 6).

In some individuals, the length of the graft increased as a result of the initial passive motion cycles while it decreased in others. With the knee in


Beynnon & Johnson

extension, the increase in graft length produced a maximum of a Imm increase in anterior laxity of the knee. This increase can occur immediately after reconstruction of the ACL in response to only a few cycles of passive flexion-extension motion. Also, changes in the length of the graft occur after fixation at loads far less than the ultimate failure load of the graft or the failure strength of graft fixation. The findings indicate that the cyclic response of the graft, and not just the structural properties of the graft or the ultimate failure load of graft fixation, should be considered during rehabilitation.

Our previous investigations of the biomechanical behaviour of the normal ACL and B-PT-B graft in humans raises concern regarding current aggressive or accelerated rehabilitation programmes. Strains are imparted to a healing ligament during rehabilitation programmes that include immediate full range of motion, weight-bearing as tolerated, unlimited quadriceps activity, and return to unrestricted activity 4 to 6 months fOllowing ACL reconstruction. There is no current information about which strains are deleterious to a healing graft or how much strain is required for adequate graft healing. However, clinicians developing postoperative rehabilitation programmes should understand that these grafts are strained and aggressive rehabilitation exercises may damage healing tissues. Despite claims that aggressive and accelerated rehabilitation are not harmful, no randomised prospective trials of aggressive versus conservative rehabilitation techniques have been conducted.

4. Functional Knee Braces

To determine whether a functional brace can protect a healing ACL graft, we evaluated the effect of functional braces on ACL strain in humans with normal knee ligaments and normal muscle activityJ351 We determined that none of the 7 functional braces we tested produced a deleterious increase in ACL strain. This finding contradicts our earlier investigation of functional braces performed on human cadaversJ361 Two of the 7 functional braces provided a protective strain shielding effect on the ACL when anterior directed shear loads were ap-


ACL Injury Rehabilitation in Athletes 59

Table I. A comparison of rehabilitation exercises based on the peak anterior cruciate ligament strain values developed during the exercise (from Beynnon et al.,[151 with permission)

Activity

Iso quads contraction at 15° (to 30 N· m of extension torque)

Active ROM with a 45N weight boot

Lachman test (150N of anterior shear load at 300 )a,b

Active ROMa,c

Simultaneous quads and hams contraction at 15°

Iso quads contraction at 30° (to 30 N • m of extension torque)a

Anterior drawer test (150N of anterior shear load at 900 )a,d

Iso hams contraction at 15° (to -10 N· m of flexion torque)

Simultaneous quads and hams contraction at 30° Passive ROMa,.

Iso quads contraction at 60° (to 30 N • m of extension torque)

Iso quads contraction at 90° (to 30 N • m of extension torque)



Iso hams contraction at 30°, 60° and 90° (to -10 N· m of flexion torque)

Peak strain (%)

4.4

3,8

3,7

2,8

2.8

2.7

1.8

0.6

0.4

0,1

0,0

0,0

0.0

0.0

0.0

No. of participants

8

9

10

18

8 18

10

8

8 10

8

18

8

8

5 a Data from this study were combined with data from a previous investigation,

b Anterior shear load applied to the tibia with the knee at 30° of flexion,

c Flexion-extension of the knee performed by the participant.

d Anterior shear load applied to the tibia with the knee at 90° of flexion,

e Flexion-extension of a participant's knee performed by the investigator,

Abbreviations: hams = hamstrings; iso = isometric; quads = quadriceps; ROM = range of motion.

plied to the knee in the non-weight-bearing state. It is important to point out, however, that the strain shielding effect of the brace decreased as the magnitude of the anterior-directed shear load increased. Some of the functional braces produced a protective strain shielding effect on the ACL for internal torque of the lower leg.

5. Healing ACL Grafts in Animals

The structural and material properties of healing ACL grafts in humans are unknown at present. To gain insight into the healing response of ACL grafts, animal investigations have been performed. Since animals cannot comply with a postoperative rehabilitation programme, the results cannot be extrapolated directly to humans to determine the effect of rehabilitation on healing grafts. However, they do provide insight into the biomechanical behaviour of the graft during healing. These animal studies of ACL reconstruction have shown that the ACL graft loses a large percentage of its ultimate failure strength and linear stiffness in the first 6


weeks and then these properties slowly improve but never approach that of the normal ACL.l37]

Investigations of ACL reconstruction with an iliotibial tract autograft for periods up to 1 year after reconstruction have shown that the ultimate failure load values range from 23 to 40% of the normal control ACL, while the stiffness is 45% of the normal ACL.(33] For the patellar tendon autograft, the ultimate failure load values have been reported to range from 11 to 50% of the control ACL, while the stiffness has been reported to range from 13 to 57% of the normal ACL a year or more post -reconstruction. (33,38]

The graft healing period is also associated with an increase of anterior translation of the tibia relative to the femur. Investigations performed in animals have revealed that anterior-posterior joint laxity after ACL reconstruction with a patellar tendon autograft ranges from 156 to 269% of the contralateral normal knee at I year or longer after reconstructionP3,38] It is unclear why ACL grafts lose their strength, become more compliant and have

Sports Med, 1996 Jul; 22 (1)

60

--- 20'

~ 6 -80'

~

~ Ol

4 Q)

2 -£; '0 Q) 0 (/J c 0 -2 Co (/J

~ -4 .!'! C3 -6 '" ()

2 4 6 8 10 12 14 16 18 20

Flexion cycle number

Fig. 6. Immediately after fixation, the bone-patellar tendon-bone graft undergoes a cyclic response with passive flexion-extension motion of the knee. This was characterised by determining the percentage change in the length of the Hall Effect Transducer (relative to the length for the first cycle of passive motion at 200 and 800 of flexion) and then plotted for the first 20 cycles. The 95% confidence intervals are plotted about the mean cyclic response values, demonstrating that by the twentieth cycle of passive motion there was no significant difference between the cyclic response values and the zero value, or reproducible behaviour. This demonstrates that a graft undergoes a cyclic response immediately after fixation and suggests that seating of the graft should be perlormed just prior to its final fixation if reproducible behaviour is to be obtained (from Beynnon et al.l34) with permission).

increased anterior-posterior knee laxity during graft healing. It has been postulated that revascularisation and remodelling must be important. It may be that successful reconstruction of the ACL depends on many factors such as:

• graft material • graft cross-sectional area • intra-articular positioning of the graft

• graft setting tension • initial graft strain values developed at the time

of graft fixation . This hypothesis was tested by Jackson and col

leagues who investigated an ideally placed and oriented autogenous ACL graft.[39) In this investigation, the ACL was devitalised and devascularised through multiple in situ freeze-thaw cycles. After a 26-week healing period, there was no significant difference between the healing ACL and the contralateral normal ACL in anterior-posterior knee laxity, ultimate failure load, linear stiffness, elastic

modulus, strain to maximum stress or maximum

© Adis Internotional Limited. All rights reserved.

Beynnon & Johnson

failure stress. These findings led the investigators to postulate that loss of ACL graft structural properties and increases in anterior knee laxity (documented by animal investigations) may not be caused by the natural process of devascularisation and remodelling of the graft, but rather by improper positioning of the graft and initial graft strain conditions at the time of fixation.

The increase in anterior knee laxity that occurs during graft healing is a concern. We performed an investigation to determine the relationship between anterior-posterior knee laxity and the structural properties of autogenous patellar tendon grafts used to replace the ACL in dogsJ40) There was a significant inverse correlation between anterior-posterior knee laxity and the failure strength of the ACL graft at 30° (full extension for the dog), 60° and 90° of knee flexion (fig. 7; top) . Similarly, there was a significant inverse correlation between anteriorposterior knee laxity and the linear stiffness of the ACL graft at 60° and 90° of knee flexion (fig. 7; bottom). The increase in anterior knee laxity was consistently associated with a decrease in the failure load and linear stiffness values of the ACL graft. This study indicates that knee laxity measurements that reveal an increase in anterior displacement of the tibia relative to the femur during ACL graft healing may indicate that the graft has weakened structural properties.

The mechanism that produced the inverse relationship between knee laxity and the structural properties of the graft is unclear. Furthermore, it is unclear why there was a significant increase in anterior translation of the tibia relative to the femur compared with the contralateral normal control knees. It may be that we were unable to control the initial graft strain accurately at the time of surgical fixation. It is also possible that the increase in anterior knee laxity was caused by excessive graft strain at the time of fixation and during healing. This could have led to permanent elongation of the graft or inhibited the healing process. However, the effect of initial graft strain on the kinematic behaviour of the knee and the healing response of the ACL graft is currently unknown.

Sports Med. 1996 Jul; 22 (l)


300 0

o

200

o 100

(ij E o O+------,-------r------.-----~-

~ 0 .2 U 2 iii c: 8 ! 300 i!' 'x m ...J

200

100

10 20 30 40

Ultimate load (reconstruction/normal) [%]

o

o

O+------,------.-------r------r-o 10 20 30 40

Stiffness (reconstruction/normal) [%]

Fig. 7. A significant inverse relationship was found between anterior-posterior knee laxity and the structural properties [load (top) and stiffness (bottom)] of autogenous patellar tendon grafts used to reconstruction the anterior cruciate ligament in dogs. A significant inverse correlation was found at 90° of flexion between knee laxity, and the ultimate failure load of the patellar tendon graft (r = -0.62). Likewise, a significant inverse correlation was found between knee laxity and the stiffness of the patellar tendon graft (r = -0.58) [from Beynnon et al.,140) with permission].

6. Rehabilitation and Healing ACL Grafts

6.1 Retrospective Studies

Shelbourne and Nitz[12] performed a retrospective study of rehabilitation after ACL reconstruction. They reported that an aggressive rehabilitation programme (including full knee extension, immediate walking with full weight-bearing and return to light sports by 8 weeks) was more effective than a conservative rehabilitation protocol. However, in the aggressive and conservative reha-


61

bilitation groups, 25% of the individuals had greater than a 3mm increase in anterior knee laxity for the reconstructed knee compared with that of the opposite normal side. This increase is of concern because it is greater than the side-to-side differences in knee laxity for individuals with normal knees.

Glasgow and colleagues(4)] performed a retrospective study of early (2 to 6 months), versus late (7 to 14 months) return to vigorous cutting activities on the outcome after ACL reconstruction. At an average follow up of 46 months, early return to activity did not predispose individuals to knee reinjury, different manual maximum KT-l 000 measures of anterior knee laxity, or less satisfactory subjective measures of knee function.

Barber-Westin and Noyes[42] performed a retrospective study of the effect of rehabilitation after ACL reconstruction. Their rehabilitation programme consisted of 4 phases. The first phase was an assisted ambulatory phase that involved immediate continuous passive knee motion and partial weight bearing with crutch support until postoperative weeks 7 to 9. The second was from postoperative weeks 9 to 16 and was considered the early strength training phase. The third phase lasted from postoperative weeks 16 to 52 and was considered the intensive strength training phase. The fourth phase was return to sport activity. Of those who underwent ACL reconstruction with a B-PT-B allograft, 54% had greater than a 3mm increase in anterior knee laxity as measured with the KT-I 000 knee arthrometer. Approximately 50% presented with abnormal anterior knee displacements within the first postoperative year, while the other half presented with abnormal displacements between the first and fourth postoperative year. Of those who underwent ACL reconstruction with a B-PT-B allograft combined with an iliotibial band extraarticular procedure, 28% had greater than a 3mm increase in anterior knee laxity. Again, 50% presented with abnormal knee laxity in the first year, while the other half went on to abnormal laxity between the first and fourth postoperative years. This investigation demonstrates the need for fol-


62

low-up measurements over a minimum of 2 to 4 years before conclusions can be made about the effect of different rehabilitation programmes on the knee.

Findings from retrospective studies of rehabilitation programmes should be interpreted cautiously, since a priori inclusion-exclusion admission criteria cannot be applied. This can produce a susceptibility bias that comes from comparing the results of rehabilitation programmes between groups of individuals that differ prognostic ally. For example, comparing individuals with multiple ligament injuries (i.e. ACL and medial collateral ligament), or ACL combined with meniscal injury, with those who have an isolated ACL injury would produce a susceptibility bias. The results of retrospective investigations of rehabilitation may also be confounded by a performance bias arising from changes in the level of skill in performing a surgical procedure or rehabilitative programme. Finally, a transfer bias may occur when comparisons are made to an historic rehabilitation group that might have had a substantially longer follow-up time period.

6.2 Prospective Studies

Noyes and associates! II) performed a prospective, randomised study of immediate versus delayed motion rehabilitation after ACL reconstruction. Participants in the immediate motion programme underwent continuous passive motion on the second postoperative day, while those in the delayed motion group had their knees braced at 10° of flexion and began continuous passive motion on the seventh postoperative day. The 2 rehabilitation programmes were similar in all other aspects. No difference between the immediate and delayed motion programmes was reported with regard to joint effusion, haemarthrosis, soft tissue swelling, f1exionextension limits of the knee, use of pain medications or the time of hospital stay. Continuous passive knee motion immediately after ACL reconstruction did not lead to an increase in anterior knee laxity.

Rosen and coworkers!431 also performed a prospective, randomised investigation of rehabilitation after arthroscopically assisted ACL recon-


Beynnon & Johnson

struction with a bone-patellar tendon-bone autograft. They reported that continuous passive knee motion during the first month after ACL reconstruction compared with early active motion produced a similar range of joint motion and KT-l 000 measurements of anterior-posterior displacement of the tibia relative to the femur.

In a prospective, randomised investigation, Richmond and colleagues!441 compared the effects of continuous passive knee motion 4 versus 14 days after arthroscopically assisted ACL reconstruction with a B-PT-B autograft. They reported no difference between the groups in range of knee motion or lower limb girth measurements.

Bynum and coworkers,!45] who conducted a prospective, randomised investigation of a closed kinetic chain versus an open kinetic chain rehabilitation programme, reported that the former resulted in more normal KT-IOOO arthrometer measurements of anterior-posterior knee laxity, less patellofemoral pain, earlier return to normal activities of daily living and sports, and more satisfied patients.

7. Conclusions

Important advances have been made with regard to what is known about the effect of ACL disruption, reconstruction and rehabilitation on the biomechanical behaviour of the healing ACL graft and knee. Exercises that strain the ACL and would be safe for a healing ACL graft include isometric hamstrings contraction at all knee flexion angles and contraction of the dominant quadriceps muscles with the knee flexed at 60° or greater (isometric quadriceps, simultaneous quadriceps and hamstrings contraction, and active flexion-extension between 40° and 90° of flexion). Disruption of the ACL does not appear to alter joint position sense and does not affect the closed-loop efferent reflex pathway of the leg musculature (the time required to illicit a muscle reaction in response to a perturbation).

Functional knee braces provide a protective strain-shielding effect on the ACL when anterior shear loads and internal torques are applied to the knee in the non-weight-bearing state (commonly



thought to be 'injury-mechanism' loadings). However, the strain-shielding effect of functional braces appears to decrease as the magnitude of anterior shear, or internal torque applied to the knee increase. At present, it is difficult to relate the loads applied to the knee to those experienced during common sporting events and to determine whether or not knee braces can protect the knee, an ACL graft, or prevent intra-articular injury.

A recent prospective, randomised study of rehabilitation following ACL reconstruction has presented evidence that a closed kinetic chain exercise programme (foot fixed against a resistance) results in anterior-posterior knee laxity values that are similar to the normal contralateral knee.(46) In contrast, open kinetic chain exercises (foot not fixed against a resistance) were reported to produce an increase in anterior-posterior knee laxity in comparison to normal.l46) Future studies are needed to determine the effect of ACL graft strains on its healing response.

References I. Buckwalter JA. Activity vs rest in the treatment of bone, soft

tissue and joint injuries. Iowa Orthop J 1995; 15: 29-42 2. Paulos L. Noyes FR, Grood ES, et al. Knee rehabilitation after

anterior cruciate ligament reconstruction and repair. Am J Sports Med 1981; 9: 140-9

3. Steadman JR. Rehabilitation of acute injuries of the anterior cruciate ligament. Clin Orthop 1983; 172: 129-32

4. Akeson WHo Amiel D, Woo SL-Y. Immobility effects on synovial joints. The pathomechanics of joint contracture. Biorheology 1980; 17: 9S-11O

S. Haggmark T, Erickson E. Cylinder or mobile cast brace after knee ligament surgery: a clinical analysis and morphological and enzymatic study of changes in quadriceps muscle. Am J Sports Med 1979; 7: 48-56

6. Jozsa L, Jarvinen M, Kannus P, et al. Fine structural changes in the articular cartilage of the rat's knee following short-term immobilization in various positions: a scanning electron microscopical study. Int Orthop 1987; II: 129-33

7. Jozsa L, Reffy A, Jarvinen M, et al. Cortical and trabecular osteopenia after immobilization - a quantitative histological study in rats . Int Orthop 1988; 12: 169-72

8. Jozsa L, Thoring J. Jarvinen M, et al. Quantitative alterations in intramuscular connective tissue following immobilization: an experimental study in rat calf muscle. Exp Mol Pathol 1988; 49: 267-78

9. Kennedy Je. Symposium: current concepts in the management of knee instability. Contemp Orthop 1982; 5: 59-78

10. Noyes FR. Functional properties of knee ligaments and alterations induced by immobilization. Clin Orthop Rel Res 1977; 123: 210-42

© Adis International limited. All rights reserved.

63

II. Noyes FR, Mangine RE, Barber S. Early knee motion after open and arthroscopic anterior cruciate ligament reconstruction. Am J Sports Med 1987; 15: 149-60

12. Shelbourne KD, Nitz P. Accelerated rehabilitation after ACL reconstruction. Am J Sports Med 1990; 18: 292-9

13. MacDonald PB, Hedden D, Pacin 0, et al. Effects of accelerated rehabilitation programs after anterior cruciate ligament reconstruction with combined semitendinosus-gracilis autograft and a ligament augmentation device. Am J Sports Med 1995; 23: 588-92

14. Shelbourne DK, Klootwyk TE, Wilckens JH, et al. Ligament stability two to six years after anterior cruciate ligament reconstruction with autogenous patellar tendon graft and participation in accelerated rehabilitation program. Am J Sports Med 1995; 23: 575-9

IS . Beynnon BD, Fleming BC, Johnson RJ, et al. Anterior cruciate ligament strain behavior during rehabilitation exercises invivo. Am J Sports Med 1995; 23: 24-34

16. Draganich LF, Vahey Jw. An in-vitro study of anterior cruciate ligament strain induced by quadriceps and hamstrings forces. J Orthop Res 1990; 8: 57 -63

17. Henning CE, Lynch MA, Glick KR. An in-vivo strain gage study of elongation of the anterior cruciate ligament. Am J Sports Med 1985; 13: 22-6

18. Jurist KA, Otis Je. Anteroposterior tibiofemoral displacements during isometric extension efforts. The role of external load and knee flexion angle. Am J Sports Med 1985; 13: 254-8

19. Lutz GE, Palmitier RA, An KN, et al. Comparison oftibiofemoral forces during open-kinetic chain and closed-kinetic chain exercises. J Bone Joint Surg Am 1993; 75: 732-9

20. Shoemaker SC, Markolf KL. Effects of joint load on the stiffness and laxity of ligament deficient knees. J Bone Joint Surg Am 1985; 67: 136-46

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

22. Bell e. On the nervous circle which connects the voluntary muscles with the brain. Philos Trans R Soc Lond B 1826; 116: 163-73

23. Sherrington CS. The integrative action of the nervous system. London: Constable, 1906

24. Barrett DS. Proprioception and function after anterior cruciate reconstruction. J Bone Joint Surg Br 1991; 73 : 833-7

2S. Barrack RL, Skinner HB, Buckley SL. Proprioception in the anterior cruicate deficient knee. Am J Sports Med 1989; 17: 1-6

26. Good L, Beynnon BD, Gottlieb DJ, et al. Joint position sense is not changed after ACL disruption [abstract). Trans Orthop Res Soc 1995; 20 (I): 95

27. Corrigan JP, Cashman WF, Brady MP. Proprioception in the cruciate deficient knee. J Bone Joint Surg Br 1992; 74: 247-50

28. Klein BP, Blaha JD, Simons W. Anterior cruciate deficient knees do not have altered proprioception [abstract]. Orthop Trans 1992; 16: 539

29. Beard DJ, Kyberd PJ, Fergusson CM, et al. Proprioception after rupture of the anterior cruciate ligament. An object indication for the need for surgery? J Bone Joint Surg Br 1993; 7S: 311-S

30. Jennings AG, Seedholm BB. Proprioception in the knee and reflex hamstrings contraction. J Bone Joint Surg Br 1994; 76: 491-4

31. Beynnon BD, Gottlieb D, Elmqvist L-G, et al. The latency of reflex muscle contraction in normal and anterior cruciate ligament deficient subjects. Orthop Trans 1994; 8 (2): 450-1


64

32. Wojtys EM, Huston LJ. Neuromuscular performance in normal and anterior cruciate ligament deficient lower extremities. Am J Sports Med 1994; 22: 89-104

33. Newton PO, Horibe S, Woo SL-Y. Experimental studies of anterior cruciate ligament autografts and allografts. In: Daniel OM, Akeson WH, O'Conner JJ, editors. Knee ligaments: structure, function, injury and repair. New York: Raven Press, 1990: 389-9

34. Beynnon BO, Johnson RJ, Fleming BC, et al. The measurement of elongation of anterior cruciate grafts in-vivo. J Bone Joint Surg Am 1994; 76: 520-31

35. Beynnon BO, Pope MH, Wertheimer CM, et a!. The effect of functional knee-braces on strain on the anterior cruciate ligament in-vivo. J Bone Joint Surg Am 1992; 74: 1298-312

36. Arms SW, Oonnermeyer D, Renstrom P, et a!. The effect of knee braces on anterior cruciate ligament strain [abstract). Trans Orthop Res Soc 1987; 12: 245

37. Butler DL, Grood ES, Noyes FR, et a!. Mechanical properties of primate vascularized vs non vascularized patellar tendon grafts: changes over time. J Orthop Res 1989; 7: 68-79

38. Ballock RT, Woo SL-Y, Lyon RM, et a!. Use of patellar tendon autograft for anterior cruciate ligament reconstruction in the rabbit: a long term histological and biomecanical study. J Orthop Res 1989; 7: 474-85

39. Jackson DW, Grood ES, Cohn BT, et al. The effect of in-situ freezing of the anterior cruciate ligament. J Bone Joint Surg Am 1991; 73: 201-13

40. Beynnon BD, Johnson RJ, Tohyama H, et al. The relationship between anterior-posterior knee laxity and the structural prop-


Beynnon & Johnson

erties of the patellar tendon graft. A study in canines. Am J Sports Med 1994; 22: 812-20

41. Glasgow SG, Gabriel Jp, Sapega AA, et a!. The effect of early versus late return to vigorous activities on the outcome of anterior cruciate reconstruction. Am J Sports Med 1993; 21: 243-8

42. Barber-Westin SD, Noyes FR. The effect of rehabilitation and return to activity on anterior-posterior knee displacements after anterior cruciate ligament reconstruction . Am J Sports Med 1993; 21: 264-70

43. Rosen MA, Jackson OW, Atwell EA. The efficacy of continuous passive motion in the rehabilitation of anterior cruciate ligament reconstruction. Am J Sports Med 1992; 20: 122-7

44. Richmond JC, Gladstone J, MacGillivray J. Continuous passive motion after arthroscopically assisted anterior cruciate ligament reconstruction: comparison of short-versus long-term use. J Arthroscop ReI Res 1991; 7: 39-44

45. Bynum EB, Barrack RL, Alexander AH. Open versus closed kinetic chain exercises after anterior cruciate ligament reconstruction. Am J Sports Med 1995; 23: 401-6

Correspondence and reprints: Dr Bruce D. Beynnon, Department of Orthopaedics and Rehabilitation, Stafford Hall, Room 438A, University of Vermont, Burlington, VT 05405-0084, USA. e-mail: [email protected]


Documents

Anterior Cruciate Ligament Injury Rehabilitation in Athletes