gait biomechanics are not normal after anterior cruciate ......gait biomechanics are not normal...

Clinically Relevant

Gait biomechanics are not normal after anterior cruciate ligament reconstruction and accelerated rehabilitation

PAUL DEVITA, TIBOR HORTOBAGYI, and JASON BARRIER

Depclrrmet~r of Exercise and Sport Science, East Carolina U~riversity, Greenville. NC 27858

ABSTRACT

DEVITA. P.. T. HORTOBAGYI, and I. BARRIER. Gait biomechanics are not normal after anterior cruciate ligament reconstruction and accelerated rehabilitation. Med. Sci. Sports Exerc.. Vol. 30, No. 10. pp. 1481-1488. 1998. Purpose: Accelerated rehabilitation for anterior cruciate ligament (ACL) injury and reconstruction surgery is designed to return injured people to athletic activities in approximately 6 months. The small amount of empirical data on this population suggests, however, that the torque at the knee joint may not return until 29, months after surgery during walking and even longer during running. Although the rehabilitation has ended and individuals have returned to preinjury activities. gait mechanics appear to be abnormal at the end of accelerated programs. The purpose of this study was to compare lower extremity joint kinematics. kinetics, and energetics between individuals having undergone ACL reconstruction and accelerated rehabilitation and healthy individuals. Methods: Eight ACL-injured and 22 healthy subjects were tested. Injured subjects were tested 3 wk and 6 months (the end of rehabilitation) rfter surgery. Ground reaction force and kinematic data were combined with inverse dynamics to predict snginal plane joint toques and powers from which angular impulse and work were derived. Results: The difference in all kinematic variables between the two tests for the ACL group avenged 38% (all P < 0.05 ). The kinematics were not. different between the ACL group after rehabilitation and healthy subjects. Angular impulses and work averaged 100% difference for all joints (all P < 0.05) between tests for the ACL group. Afrer rehabilitation. the differences between injured and healthy groups in angular impulse and work at both the hip and knee remained luge and averaged 52% (all P < 0.05). Conclusions: Results indicated that after reconstruction surgery and accelerated rehabilitation for ACL injury. humans walk with normal kinematic patterns but continue to use altered joint torque and power patterns. Key Words: ACL INJURY, KNEE. WALKING. RECONSTRUCTION SURGERY, BIOMECHANICS

Steven 1. McCaw, Ph.0 De 1. of HPER 51!0 111. State Uavewty Normal. 11 61790-5120

I njury to the anterior crbciate ligament (ACL) produces distinct changes in lower extremity kinetics and energetics during gait. Whereas healthy individuals yalk

with an extensor torque at the knee between 10 and 45% of the stance phase (13,14,41), individuals with recent ACL deticiency walk with an extensor torque lasting nearly the entire stance phase (13). Over the next several years the extensor torque at the knee becomes reduced in ACL-deficient individuals and in many cases becomes a flexor torque t 1,4,5,28). Along with this adaptation, the extensor torque at the hip in early stance eventually increases to approximately 1.5 times the value seen in healthy individuals (4,28).

Gait adaptations in individuals with ACL injury who have reconstruction surgery are less clear for two reasons. First, no comprehensive gait analyses have been conducted on this

01 95-913 1/98/3010- I48 153.0010 LIEDICINE & SCIENCE IN SPORTS & EXERCISE, :opyright Q 1998 by the American College of Spons Medicine

Submitted for publication January 1998. Accepted for publication April 1998.

population and, second, the large variation in surgical and rehabilitation procedures and patient characteristics and their compliance with rehabilitation limit the generalizabil- ity of these potential results. One principle factor in the recovery from ACL injury and surgery is the rehabilitation protocol. During the past 10 yr there has been a shift from the standard, conservative rehabilitation program (7.39) to a more aggressive, accelerated protocol (9,20,26,30,3 1). Ac- celerated programs are designed to return the patient to full activity at 4-6 months after surgery, approximately twice as fast as the conservative approach. Accelerated programs generally emphasize full knee extension and immediate weight bearing on the first postoperative day, closed kinetic chain exercises by the third week, and easy running about the fifth week after surgery. Patients return to full athletic participation at about 6 months when the strength and range of motion (ROM) in the involved limb are similar to these measures in the contralateral limb.

Despite the acceptable strength and ROM measures at the end of the rehabilitation, there is evidence that the knee joint does not function normally during the complex actions of

gait. Timoney et al. (38) reported that at 10 months after surgcry using an autogenous, bon~-pi\tellar tendon-bone protocol individuals with ACL reconstruction walked with a 64% lower extensor torque at the knce in midstance compared with healthy controls. Their results showed an even lager reduction in knee torque at this late date than DeVita et al. ( 13) showed at 5 wk after the same surgery. Hurwitz et al. (21) identified similar magnitudes in the peak knee torque in midstance between ACL-reconstructed patients and healthy controls at 22 months after the same surgery. All subjects had completed accelerated rehabilitation in these studies. Although it appears that patients with ACL reconstruction may regain normal peak torque at the knee between 10 and 22 months after surgery, there is no empirical evidence showing that the power and work produced by the knee muscles and the torq~ie, power, and work produced by the hip and ankle muscles also return to normal during gait. Indeed, each of these variables was shown to be several-fold different than normal in ACL-reconstructed individuals while running at 3-4 rnonths after surgery ( 17).

Since the evidence suggests that the return of normal gait biomechanics takes much longer than 6 months if they return at all. the question arises as to the nature of the biomechanical characteristics used by patients during gait at the completion of an accelerated rehabilitation program. The purpose of this study was to compare gait biomechanics between individuals having undergone ACL reconstruction and accelerated rehabilitation and healthy individuals. It was hypothesized that although the rehabilitation program was completed. the gait biomechanics would be different between the two groups. It was reasoned that a detailed analysis of these variables might provide information that can be used to improve rehabilitation programs and the subsequent long- term prognosis for people with surgically reconstructed ACLs.

METHODS

Subjects. Eight ACL-injured subjects (mean age. 20.3 yr; mean mass, 74.3 kg), three females and five males, volunteered for the study. All subjects had complete rupture of the ACL and had never had a previous knee injury. Five subjects were collegiate athletes, one participated in high school athletics, and two were recreational athletes. There was no other ligamentous injury in any subject although there was some cartilage damage in each case. All subjects had arthroscopically assisted, endoscopic, bone-patellar tendon-bone reconstruction using the central one-third of the patellar tendon. All subjects completed an accelerated rehabilitation program similar to that described by Shelbourne and Nitz (3 1). Emphasis was given to achieving early knee extension and weight bearing. Active muscle contraction was used to increase the range of knee joint motion. Harn- strings and quadriceps strengthening with closed kinetic chain exercises were started during the second to third week using an exercise bicycle and stair stepper. The subjects started jogging between the fourth and fifth weeks. The entire rehabilitation program lasted 6 months. Twenty-two

healthy subjects. I4 males and Y females. also volunteered for the control group (mc:~n ugc. 21.7 yr; rncan mass. 73.7 kg). Control subjects had no history of lower limb pathol- ogy. All subjects gave written informed consent according to University and American College of Sports Medicine policy.

Instruments. A force platform (AMTI, Watertown. MA, Model LG6-4-2000) located in the center of a 20-m level walkway was used to measure vertical and horizontal ground reaction forces and the mediolateral moment at 1000 Hz. Sagittal plane kinematics were obtained with a video camera (SONY Corp., Park Ridge, NJ. Model SSC M350) and video cassette recorder (JVC, Elmwood Park. NJ, Model HRS5100U) sampling at 60 Hz.

~ r o t o c o l . The testing protocol was essentially identical between the injured and control groups except that the injured group was tested twice. The injured subjects were tested 3 wk after surgery (within a few days of free walking without assistance) and 6 rnonths after surgery (the end of the rehabilitation program). The control subjects were tested on one occasion.

Circumference measures of the upper thigh, knee. ankle. and metatarsal heads were obtained for the ACL-~njured limb or the right limb in the control subjects. along with body weight. Reflective markers were then placed on the lateral side of the shoe at the head of the fifth metatarsbl and heel. and on surface locations over the lateral malleolus, lateral femoral condyle. greater trochanter, and shoulder on each tested side. Subjects practiced walking along the walkway until they were comfortable and relaxed. A s~arting point was then identified so that the subject would contact the force platform in a normal stride with the appropriate limb. Eight trials of data were collected for each subject at each test. Trials were recollected if the subject altered his or her stride to contact the force platform.

Data reduction. Video records were digitized with the Peak Performance (Englewood, CO) system. Reflective markers on the subject and one on the force platform were digitized throughout the swing phase before contacting the platform and stance phase on the platform ( 1 I). Markers in four additional frames before and after these phases were also digitized to improve the accuracy of the data near the ends of the cycle.

High frequency error was removed from the digitized position coordinates with a second-order Butterwonh digital filter. Cut-off frequencies were determined with the method proposed by Winter (40) and they averaged approximately 7 Hz. The resultant position data were then interpolated to 200 Hz. Angular position was calculated for the hip, knee, and ankle joints with zero degrees being the anatomical position and positive values representing hyperextended hip and knee joints and plantarflexed ankle joint. Joint angular velocity was calculated with finite differences from the angular position.

The lower extremity was modeled as a rigid, linked seg- ment system. Magnitude of the segmental masses and location of the mass centers along with their moments of inertia were estimated from the position data using a mathematical

1482 Official Journal of the American College of Sports Medicine

model ( I 8 ) . segmental masses reported by Dernpster (10). and the individual subject's anthropometric data. Center of pressure was calculated from the ground reaction forces and the mediolateral moment on the platform. Center of pressure was expressed in the kinematic reference frame based on the digitized location of the force platform. Inverse dynamics using linear and angular Newtonian equations of motion were used to calculate the joint reaction forces and torques at each joint throughout the gait cycle. A support torque was calculated as the sum of the three joint torques (41). The support torque quantities the total torque output from the extremity and provides a quantitative assessment of the support and propulsive effort of the musculature in the entire limb. It also provides a total torque value against which each of the individual joint torques can be compared to assess the relative contribution of the individual joints to the performance. All joint torques were expressed as positive values for extensor or plantarflexor torques. Joint powers were calculated as the product of the joint torques and angular velocities. Positive power indicated that the observed joint torques and angular velocities were in the same direction and the torque performed work to increase the energy of the skeletal system. Negative power indicated that the observed joint torques and angular velocities were in the opposite direction and the torque performed work to decrease the energy of the skeletal system.

Data analysis. ROM during the swing phase and average position during the stance phase were derived from each position curve. Extensor (or plantarflexor) angular impulses during the entire stance phase (support and ankle) or during the initial half of stance (hip and knee) were derived from the torque curves. Angular impulse is the area under a torque curve and extensor or plantarflexor angular impulses represent the areas under the positive phases of the torque curves in this study. Angular impulse quantifies the total contribution of a joint torque toward producing movement and it provides a better comparison than peak torque values. Gait changes can be adaptations in either the length of time a particular muscle group produces torque (14) or in the magnitude of torque produced by the muscle group (1). Angular impulse will account for either of these adaptations whereas peak values will not.

Positive work during the first half of stance (40 to -70% of gait cycle) was calculated from the hip power curves and negative and positive work during this time (-45 to -55% and -55 to -65% of gait cycle) were calculated from the knee power curves. Negative and positive work throughout stance (40 to -83% and -83% to 100% of gait cycle) were derived from the ankle power curves. These work phases correspond to phases HI. K1. K2, A l , and A2 for the hip. knee, and ankle joints from Winter (41). Positive and negative work at each joint were calculated as the area under the respective positive and negative portions of the power curves. Positive work indicated that the torque was produced through concentric muscle contraction that generated mechanical energy and contributed to propelling the individual. Negative work indicated that the torque was produced through eccentric muscle contraction that absorbed

GAIT AFTER ACL RECONSTRUCTION AND REHABlUTATION

3 weeks - - - - 6 months - Hea l thy . . . . . 25r

d 2 Knee

Sw~ng Stance

Figure I-Joint position curves from the three groups. Zew is an erect. standing position (foot nt YO0 to 14) and negative values indicate flexion or ankle dorsiflexion. At 3 wk after surgery the ACL-injured subjects walked with about 10" more flexion at each joint compared with that at 6 months after surgery. Gait kinematics were nearly identical between ACL-injured and healthy subjects after 6 months of rehabilitation.

I

mechanical energy and contributed to decelerating the individual.

These variables were entered into a two-way, mixed factor. ANOVA (23) with knee condition (injured vs hqalthy) and time (first vs second test; control group had first test only) as the factors. Two direct comparisons were conducted: I) first versus second test in the ACL-injured group and 2) ACL-injured. second test versus healthy controls. first test. All statistical decisions were based on an alpha level of 0.05.

RESULTS

The angular kinematics were different between the two tests for the injured group (Fig. 1, Table I). ROM during swing was 10% smaller at the hip, 39% larger at the knee, and 32% larger at the ankle (all P < 0.05) at 6 months after surgery compared with 3 wk after surgery. Stance phase kinematics were also improved over the rehabilitation period and all lower extremity joints were flexed (or dorsi- flexed) less during stance at 6 months compared with 3 wk after surgery. The average position of the hip, knee. and ankle joints were 48. 35, and 63% less flexed (all P < 0.05) during stance 6 months after surgery compared with 3 wk after surgery. The angular kinematics showed excellent recovery at 6 months after surgery and were nearly identical to those of the healthy controls. None of the kinematic variables were statistically different between the injured group at 6 months after surgery and the control group (all P > .0.05).

The joint torques during stance were also much different between the two tests for the injured group (Figs. 2 and 3).

Medicine 8. Science in Sports & Exercise 1483

TABLE I Mean values and SOs across groups lor k~nemat~c variables. -- - -- -- --

ACL-6 Variable ACL-3 wlc months Healthv

Hip ROM in swing Average pos. ~n stance

Knee ROM in swing Average pos. in stance

Ankle ROM in swing 20.9 (1 0.6) a 27.5 (4.1) 30.9 (5.7) Average pos. in stance -9.8 (6.7) a -3.6 (6.9) -5.1 (3.7)

All values ~n degrees; a: ACL: 3 wk vs. 6 months. P < 0.05.

The extensor angular impulse at the hip and plantarflexor angular impulse at the ankle were 38% and 35% smaller (both P < 0.05) at 6 months after surgery compared with 3 wk al'ter surgery. In contrast, the extensor angular impulse at the knee during the first half of stance was 20% larger (P < 0.05) after 6 months of rehabilitation compared with 3 wk after surgery. The combined torque output at a11 three joints was hurnrnarized in the support torque. The support torque and angular impulse were unusually large soon after surgery but decreased 49% ( P < 0.05) at 6 months after surgery.

The ACL-injured subjects showed a partial recovery in their stance phase joint torques after rehabilitation. but the torques remained different than the torques in the control

3 weeks - - - - 6 montns - Healthy

Ankle

Stance

Figure >Joint torque curves from the three groups. Positive values indicate extensor or plantartlexor torque. At 3 wk after surgery, there were large extensor (plantarflexor) torques at the hip and ankle and a low extensor torque at the knee during stance. The total support torque, the sum of the joint torques. was highly extensor and was necessary to support the injured subjects who were more flexed than healthy individuals. The joint torques improved after rehabilitation but only the ankle torque return to a normal value. The extensor torques at the hip and knee were still larger than and smaller than normal at the 6-month test.

group. Although the extensor angular impulse at the hip was reduced from the 3-wk level, it remained 37% larger (P < 0.05) than the control group value. The extensor angular impulse at the knee during the initial half of stance also showed only a partial recovery at 6 months after surgery. The impulse was only 57% (P < 0.05) of the corresponding impulse in healthy controls. The improvement in ankle joint kinetics was statistically complete at 6 months after surgery, and the difference between the plantartlexor impulse at this time and the value for the control group (17%) was not significant. Although two individual joints had different torques in the injured group at 6 months compared with those in the healthy controls, the combined changes in the individual joint torques produced a support torque thit was statistically identical to the support torque in the controls. The difference in angular impulse from the support torque between groups was only 9% and not statistically significant. t

The joint power and work characteristics were distinctly different between 3 wk and 6 months after surgery in the ACL-injured subjects (Figs. 4 and 5). The positive work at the hip in the first half of stance decreased 44% ( P < 0.05) after 6 months of rehabilitation compared with the 3-wk value. The amount of negative and positive work done at the knee joint during the first half of stance changed during the initial 6 months following surgery. The negative work increased 2508 and the positive work increased 267% (both P < 0.05) during this time. There was a 43% decrease ( P < 0.05) in the amount of negative work done at the ankle joint during stance between the 3-wk and the 6-month tests. The increase in positive work done at the ankle (8%) in the ACL group was not significant.

As in the joint torques, the joint powers showed only a partial recovery after accelerated rehabilitation. The reduction of work done at the hip at 6 months compared with that at 3 wk after surgery still did not reduce the hip work to the level seen in control subjects. The work at the hip was 77% larger (P < 0.05) in the ACL-injured subjects at 6 months after surgery compared with the healthy subjects. The negative work done at the knee did not recover after 6 months to the level observed in the healthy subjects. This negative work was only 53% ( P < 0.05) of the negative work done by the control group. The positive work done at the knee at 6 months also did not reach the level observed in the controls. The positive work. in the injured subjects at 6 months was only 44% (P < 0.05) of the work observed in


Knee - 1 5 , 1

i Ankle

/ 50 E / 540 0 4 30 a i 1 20

healthy subjects. The negative work done at the ankle joint did recover after 6 months to the level observed in the healthy subjects. The difference between the 6-month ob- servation and the control value at the ankle (1%) was statistically nonsignificant (P > 0.05).

DISCUSSION

Numerous studies have evaluated ACL-injured individuals at various times after reconstruction surgery and rehabilitation. The majority of these investigations tested iso-

3 weeks - - - - 6 montns - Heal thy . . .

Swlng Stance

Figure 4-Joint power curves from the three tests. Positive and negative values indicate concentric contraction along with energy gener- ation and eccentric contraction along with energy absorption by the muscles. At 3 wk after surgery, the hip extensors produced relatively high power and work while the power and work produced at the knee were negligible. The hip and knee joint powers improved after rehabilitation but were still not at the levels seen in healthy individuals. The ankle power was normal after rehabilitation.

GAIT AFTER ACL RECONSTRUCTION AND REHABILITATION

Figure 3--hngular impulse values from the three groups. Values are means 2 SD. a) ACL at 3 wk vs ACL at 6 months, P < 0.05: b) ACL at 6 months vs Healthy, P < 0.05.

Lated knee joint function using. for example, isokinetic devices for muscle strength measures (29.33-35.39) and arthrometers (e.g., KT- 1000) for joint stability measures (19,25.30.31). These studies provided essential functional assessments of the knee itself and of the surgical and r ~ h a - bilitation procedures applied to the knee. Few. if any, studies, however, have provided comprehensive evaluations of the biomechanics of the entire extremity during daily human movements even though injury to a single knee ligament causes neuromuscular adaptations throughout the lower extremity (4,11). In addition, although a reconstructed and rehabilitated knee may have acceptable muscle strength and stability. it is not clear that this knee would function normally during daily activities, particularly at the end of an accelerated rehabilitation protocol. The present work inves- tigated biomechanical adaptations throughout the lower extremity during the daily activity of walking and provided a comprehensive, functional analysis of an injured limb after ACL reconstruction and rehabilitation,

The general movement pattern used in gait, as evaluated through the joint kinematics. was restored to normal after 6 months of accelerated rehabilitation. The initial response to reconstruction surgery was to walk with approximately LOo more flexion at all joints, particularly during stance (Fig. 1). After rehabilitation, the joint position curves were nearly identical between the injured and control subjects and none of the comparisons was significant. The ACL-injured subjects returned to a more erect and normal style of wallung after 6 months of rehabilitation compared with that at the start of the rehabilitation. Gait kinematics during stair climbing in ACL-reconstructed subjects also returned to normal after 6 months of accelerated rehabilitation from the same surgery as presently used (24).

Although the kinematics returned to normal after rehabilitation, the causative joint torques and powers made only a partial recovery. The joint torques improved at all three joints; however, only the torque at the ankle joint returned to normal (Figs. 2 and 3).

Medicine B. Science in Sports & Exercise 1485

I HIP - Positive Work ! I

1 25 1 1 1 i

Figure 5-Work values from the thrcc groups. Valucs are means 2 SO. a) 4CL at 3 wk vs ACL at 6 months. P < 11.05; b) ACL at 6 months vs Healthy, P c 0.05.

4

Knee -Negative Work I ) Knee - Poaitlve Work I

The torque at the knee was improved at 6 months. particularly in midstance. but i t was still different than the knee torque in he~l:hy subjects. Three weeks after surgery, ACL- reconstructed subjects used a continuous extensor torque throughout stance, unlike healthy individuals who switched to a tlexor torque in midstance (Fig. 2, (13,14,41)), After rehabilitation, the ACL-injured subjects showed the proper phasic relationships and had alternating flexor, extensor, flexor. then extensor knee torques during stance, but they still had a decrement in the magnitude of the extensor torque and angular impulse in early stance compared with that in the healthy controls. Kowalk et al. (24) reported a similar reduction in knee torque for the same population while ascending stairs. As described by Kowalk et al.. it is rea- sonable to observe reduced knee extensor torque at 6 months after surgery since the procedure harvested the central one- third of the patellar tendon. Other studies have also observed reduced knee extensor torque in isokinetic testing after reconstruction surgery (34,35,37). The reduced torque presently observed produced less negative and positive work during stance compared with that in healthy subjects. The reduction in negative work is in agreement with Juris et al. (22) who reported a decreased "force absorption" capability in the involved quadriceps during hopping and jumping at 6 months after ACL surgery and rehabilitation and with De- Vita et al. (12) who showed reduced work in completely rehabilitated subjects during running.

The excessive extensor torque at the hip in stance shortly after surgery was reduced through rehabilitation but not to the level observed in healthy subjects. The atypical hip

I I_- !- -

torque after rehabilitation was seen not as an increase in peak torque compared with healthy controls but as an extensor torque lasting longer into the stance phase. This longer lasting torque enabled the hip extensors to produce more angular impulse and work for the ACL-injured and rehabilitated subjects in the first half of stance c~mpared with that in the healthy controls. In this study hip torque and power differed between the ACL-rehabilitated and healthy groups as a function of time and area under the curves but not in peak magnitudes. We chose to compare angular impulse and not peak torque because it is the torque applied over a period of time that changes joint angular velocity and ultimately produces skillful human movement. The assessment of peak values may not provide as much insight about the total effect or contribution of the muscles at a particular joint since they provide only a momentary assessment of the dynamic joint torque.

The increased output of the hip extensors in the rehabilitated subjects performed two important functions. The extensor torque provided a lager proportion of the total support torque during the initial half of stance compared with that in healthy subjects. The angular impulse generated by the hip extensors was 65% of the support angular impulse during the first half of stance in the rehabilitated, ACL- injured subjects compared with 40% in the healthy subjects. The hip extensors also produced most of the positive work used to generate forward velocity during the first half of stance. After rehabilitation, the ACL-injured subjects main- tained forward progression with a much greater reliance on the hip extensors compared with healthy controls. In both

-.1

I Ankle - Nugatlve Work I I


Ankle - Poaitlve Work ! i

c k n e e ex- ; Iclisors PTOCILICO~ a net pohitivc ;111iount 01' work while the

,\nhlc plantartlcxors produccd a net ncgutivc amount of wo1.k during the first half 01' alarice (Fig. 4). After A C L rcconstruction and rehabilitation, the hip exlcnsors produced 83% of this work compared with only 56% i n the hcultliy individuals. The incrcucd output ol' tho hip extrn- scxs during walking is in agreement with h ip extensor function during stair ascent i n XCL-reconstructcd patients (24). These subjects used the hip extensors to produce a greater proportion o f the total work done to l i f t thcrnsclves up the ll ight o f stairs. Therefore, at 6 months after A C L reconstruction and accelerated rehabilitation. the hip extensors pmvide more vertical support and forward progression during the t i n t half o f stance compared wi th the contributiun seen i n healthy individuals. The results describing joint torques and powers at the hip and knee indicate that although the ACL-injured subjects completed an accelerated rchahilitatioti program and have returned to competitive athlaics. they do not use normal kinetics and energetics during walking.

The suppon torque was nearly identical between the ACL-rehnbilitated and healthy subjects. After 6 months o f rel~abilitation. the ACL-injured subjects werc able to walk with less flexion at all joints compared wi th that at 3 w k after surgery. This more erect posture reduced the external flexor torques at each o f the joints and enabled the subjects to walk with less total extensor effort.

The reduced knee torque and increased hip torque after rehabilitation cotlipared wi th healthy subjects were probably a result o f adaptations i n force production by the quadriceps and hamstrings muscles. Many studies have shown that ACL-deficient subjects have reduced quadriceps and increased hamstrings EMG during gait (6,15.17,37,) and that these people use either a reduced or absent extensor torque at the knee and an increased extensor torque at the h ip ! I.J.S.13,28). The present differences i n the knee and h ip torques between the ACL-reconstructed and rehabilitated and healthy subjects suggest that the injured subjects also had reduced quitdriceps and increased hamstrings force during the stance phase of walking. I t is wel l documented that maximal quadriceps force is reduced soon after A C L reconstruction surgery (34,3539) and several years after surgery (7.37). The present results suggest that subjects w i th A C L reconstruction and accelerated rehabilitation produce less quadriceps force during gait compared wi th that i n healthy individuals even though only moderate levels o f force are required.

The alterations i n gait kinematics and kinetics during the rehabilitation period suggest the load o n the repaired liga-

REFERENCES '{I- 1. ANDR~ACCHI. T. and D. BIRAC. Functional testing in the anterior

\ cruciate 11gamrn1-deticient knee. Clin. Orthop. Rrl. Res. 288:Q- 47. 1993.

1. ARVIDSON, I.. E. ERICKSSON, T. HAGGMARK, and R. JOHNSON. [SO-

kinetic thigh muscle strength after Ligament reconstruction in the knee joint: ~su l t s From a 5- to 10-yew follow-up after reconstruc-

GAIT AiTER ACL RECONSTRUCTION AND REHABILITATION

ment during gait increased over this period. Ge:~emlly. A C L Streas i s greater with increased knee and hip extension o r quadriccps force iuld i t is less with increased hamstrings force (8.27.42.43). Thc observed adaptations of increased knee ill id hip extension and extensor angular impulse at the knee along with decreased extensor angular impulse at the hip all support the idea that the load on the A C L during gait increased during the first 6 months after surgery. The in- cre:ised load d id not seem to reach normal level, however. since the extensor irnpulses at the knee and hip were less than and greater than normal. respectively. The present results suggest that after A C L rcconstruction surgery and accelerated rehabi1it:ition. the stress on the A C L during gait is different than A C L stress i n healthy individuals. The adaptations may provide increased protection for the repaired and rehabilitated knee (3.17.361.

There are several important delimitations to this study. Biomechanical adaptations to A C L injury and reconstruction surgery are dependent on several important factors including patient characteristics. surgical procedure. ' and rehabilitation protocol ( 16). The present study was delimited to physically active subjects with presumably relatively good neuromuscular-skeletal systems. The subjects under- went arthroscopically assisted, endoscopic, bone-patellar tendon-bone reconstructiori using the central one-third o f the patellar tendon. The subjects completed an accelefated rehabilitation protocol and they were highly compliant w i th the protocol. The present results may apply only to ACL- injured individuals w i th identical or, at least, similar characteristics. t

Within these delimitations, the results indicated that after reconstruction surgery and accelerated rehabilitation fo r A C L injury. humans walk wi th normal kinematic patterns but use altered jo int torque and power patterns. Rehabili- tated, ACL-injured individuals use a larger extensor torque at the hip that produces more work and a smaller extensor torque at the knee that produces less work than healthy individuals. The altered kinetics used during gait might contribute to the increased rate o f knee symptoms observed i n later years in this population.

The authors gratefully acknowledge the contributions of Tally Lassiter, M.D., Jim Bazluki, A.T.C.. and Mike Hanley, A.T.C. in recruiting injured subiects. We thank Mike Tony, Ph.D., Jeffrey Money. M.S.. Kathryn Glover. M.S., David Speroni. M.S. and John Lochmann, M.S. for their contributions to the data reduction pro- cess.

This research was supported by the Office of Research and Graduate Studies. East Carolina University.

Address for correspondence: Paul DeVita, Ph.D., Dept. of Exer- cise and Sport Science. East Carolina University. Greenville. NC 27858. E-mail: [email protected].

tion of the anterior cruciate ligament in the knee joint. Inr. J. Sports Med 27-11. 1981.

3. BARATTA. R.. M. SOLOMONOW. B. ZHOU, D. LETSON, R. CHUINARD. /54! . and R. D'AMBROSIA. MUSCUI~ coxtivation: the role of antagonist I ig musculature in maintaining knee stability. Am. J. SpotTs Med.

16: I 13-122, 1988. 4. BERCHUCK, M., T. ANDRIACCHI, B. BACH, and B. BIDER. Gait

Medicine 8 Science in Sports 8. Exercise 1487

adnptatttrnh by patients uho have a Jct'icicnt anterior cruciate I~g;ililciil. J. Rotrc Jorrrt SII~A? 771-A:X7 1-877. IOW.

5. BIR,\c. D.. T. ANORI,\CCIII. and B. B~(.II. Time relatctl ch;tngcs f'ollowt~ig ACL rupture. Proc. 37tlr ,\trtr. Mcctirr.~ Orrlrop. Rcs. Soc. 37:231, 1991.

6. BHANCII T.. R. HUMER. and M. D ~ N A ~ ~ I . Dyt~amic EMG analysis of anterior cruciate deficient legs with and without bracing during cutting. t\m. J. Sponr Med. 17:35-41. 1980.

7. B R E W S ~ R . C.. D. MOYNW, and F. Joee. Rehabilitation for antcrior cruciate reconstruction. J. Orrhop. Sports Phys. Ther. 5: 121-126. 1983.

8. COLLINS. J. and J. O'CONNER. Muscle-ligament interactions at the Lnce during walking. PIT)^. Inst. Merlr. Dtg. ISSN: 054-41 19: 11-18, 1991.

9. DECARLO. M.. K. SHELBOURNE. J. MCCAHR~LL. and A. RETIG. Tra-

y ditional versus accelerated rehabilitation following ACL recon- qtruction: a one-year follow-up. J. Orrhop. Sports P h y . Tlrer. 15:309-316, 1992.

IQ. DBMPSTER. W. Space requirements o f the seated operator. In: bVA D C T~~c l~n ica l Reporr 55- 159. OH: Wright-Patterson Air Force Baac. 1959 pp. 55-159.

I I. DEVITA. P. Selection of a standard convention for analyzing gait data based on the analysis of relevant biomechanical factors. J. Bionreclt. 2750 1-508, 1994.

12. DEVITA. P.. P. B. HU~TER, and W. SKELLY. Effects of a functional hnee brace on the biomechanics of running. Med. Sci. Sports Exerc. 14797-806, 1992.

. 13. DEVITA. P.. T. HORTOB,\(;YI. J. BARRIER. et 31. Gait adaptations ! before and after antenor cruciate ligament reconstruction surgery.

Med. Scr. Sports Exerc. 19.853-859, 1997. '

14. DEVITA. P.. M. TORRY. K. GLOVER, and D. SPERONI. A functional knee brace alters joint torque and power patterns during walking and running. J. Biorneclr. 29:583-588. 1996.

15. EL~~QIJIST. L.. R. LORENIZEN. C. JOHANSSON. et al. Does a torn

; anterior cruciate ligament lead to change in the central nervous - drive of the knee extensors? Eur. J. Appl. Physiol. 58:103-107, 1988.

16. FIWXK. C. and D. JACKSON. The science of reconstruction o f the anterior cruciate ligament. J. Bone Joinr Surg. 79-A: 1556-1576. 1997.

17. GRABINER. M. and G. WEIKER. Anterior cruciate ligament injury and hamstrings coactlvation. Clin. Biomeclr. 8215-219, 1993.

18. HANAVAN. E. A mathematical model of the human body. In: AMRL Teclmicul Documenra~ Repon. Ohio: Wright-Patterson Air Force Base, 1964. pp. 64-102.

19. HEIER. K.. D. MACK. J. MOSELY, R. PAINE, and I. BOCELL. An analysis of anterior cruciate ligament reconstruction in middle- aged patients. Am. J. Spom Med. 25:527-532, 1997.

20. HOWELL. S. and M. TAYLOR. Brace-free rehabilitation, with early return to activity. for knees reconstructed with a double-looped semitendinosus and gracilis gnft. J. Bone Joint Surg. 78A:514- 823. 1996.

21. HURWITZ. T.. T. ANDRIACCHI. P. KOPINSKI. C. BUSH-JOSEPH. and B.

3 BACH. Knee function following antenor cruciate ligament reconstruction. Proc. 20th Ann. Meeting Am. Soc. Biomech. 20:39-40. 1996.

22. JURIS. P.. E. PHILLIPS. C. DALPE, C. EDWARDS, R. GOTLIN. and D. LNE. A dynamic test of lower extremity function following anterior cruciate ligament reconstruction and rehabilitation. J. Or- thop. Spons Phys. Ther. 26: 184-19 1. 1997.

23. KEPPEL. G. Design and Analysis: A Researcher's Huncll?ook. Englewood Cliffs. NJ: Prentice-Hall. 1981. pp. 127-142.

24. KOWALK. D.. J. DUNCAN. F. MCCUE. and C. VAUGHN. Anterior cruciate ligament reconstruction and joint dynamics during stair climbing. Med. Sci. Sports Exerc. 29: 1406-14 13, 1997.

75. Novss. F. and S. BARII~~R-Wlmln. A co~nparison of rcsults in chronic anterior cruciatc ligarncnt rupturc of anhroscopic;illy assisted autogenous patellar tendon reconstruction. Am. ./. Sl~orts Med. 25:460-471. 1997.

26. NOYES. F.. R. MANGINB. iind S. BARBBH. Early knee motion after open and arthroscopic anterior cruciate ligament reconstruction. Am. J. Sl~orrs Med. 15: 149-159, 1987.

27. PANDY. M. and K. SHELBURNE. Dependence of cruciate ligament loading on muscle forces and external load. J. Biomech. 30: 10 15- 1024, 1997.

28. PATEL, R.. T. HURWITZ. T. ANDR~ACCHI. C. BUSH-JOSEPH, and B. BACH. Mechanisms by which patients with anterior cruciate ligament deficiency generate the quadriceps avoidance gait. Proc. 2lst Ann. Meerirrg Am. Soc. Biomech. ?1:2 10-2 1 1, 1997.

19. SHELBOURNE K. and D. FOULK. Timing of surgery in acute anterior cruciate ligament tears on the return o f quadriceps muscle strength after reconstruction using an autogenous patellar tendon gnft. Am. J. Sports Med. 23:686-689. 1995.

30. SHELBOURNE. K. M. KLOOTWYK. J. WILCKENS, and M. DECARLO. Ligament stability two to six years after anterior cruciate ligament reconstruction with autogenous patellar tendon graft and participation in accelerated rehabilitation program. Am. J. Spans Mecl. 23575-579, 1995.

3 1. SHELBOURNE, K. and P. NITZ. Accelerated rehabilitation aftcr anterior cruciate !igament reconstruction. Am. J. Sporrs Met/. IS: 292-299. 1990.

33,. SINKJAER. T. and L. ARENOT-NIELSEN. Knee stability and musclc coordination in patients with anterior cruciate ligament injuries: an electromyographic approach. J. Electrortt\.o. Kines. 1 :?OY -2 17. 1991.

33. SNYDER-MACKLER. L., S. BINOER-MACLEOD, and P. WILLIA~IS.'F~-

tigability of human quadriceps femoris muscle following anterior cruciate ligament reconstruction. Med. Sci. Spons Ererc. 5 7 8 3 - 789. 1993.

34. SNYDER-MACKLER, L., A. DELITTO. S. BAILEY, and S. STR,\~KA. Strength of the quadriceps fernoris muscle and functional recovery after reconstruction of the anterior cruciate ligament. J. Bone Joitrr Strrg. 77-A: 1 166-1 173, 1995.

35. SNYDER-MACKLER. L.. P. DELUCA. P. WILLIAMS. M. EASTLACK. and A BARTOLOZZI. Reflex inhibition o f the quadriceps femoris muscle after injury or reconstruction of the anterior cruciate ligament. J. Bone Joinr Surg. 76-A:555-560, 1994.

36. SOLOMONOW M., R. BARATTA. B. ZJIOU. et al. The synergistic action ? of the anterior cruciate ligament and thigh muscles in maintaining

j ' j ' '' joint stability. Am. J. Sports Med. 15:207-213. 1987. 37. Treo~+ J. and T. MCH. A biomechanical analysis o f anterior

cruciate ligament reconstruction with the patellar tendon: a two- year follow-up. Am. J. Sports Med. 16:332-335. 1988.

38. TIMONEY, I.. W. INMAN. P. QUESADA, et al. Return o f normal gait 'i 2 ;'i patterns after anterior cruciate ligament reconstruction. ' '

Am. J. Sporrs Med. 21387-889, 1993. 39. VEGSO. J.. S. GENUARIO, and J. TORG. Maintenance of hamstring

strength following knee surgery. Med. Sci. Sports Exerc. 17:376- 379, 1985.

40. WINTER. D. Biomechanics and Moror Control of Human Move- ment. New York: John Wiiey & Sons. 1990. pp. 41-13.

41. W t m ~ , D. Biomechanical motor patterns i n normal walking. J. Motor Behav. 15:302-329, 1983.

42. YASUDA, K. and T. SASAKI. Muscle exercise after anterior cruciate ligament reconstruction. Clin. Onhop. Rel. Res. 220266-174. 1987.

43. YASUDA, K. and T. SASAKI. Exercise after anterior cruciate ligament reconstruction. Clin. Orthop. Rel. Res. 210:275-283, 1987.


gait biomechanics are not normal after anterior cruciate ......gait biomechanics are not normal...

Documents