Motion Alterations After Anterior Cruciate Ligament Reconstruction: Comparison of the Injured and Uninjured Lower Limbs During a Single-Legged Jump

Participants

A total of 11 men (age = 23.3 ± 3.8 years, mass = 81.2 ± 17.0 kg) who underwent surgical reconstruction of unilateral isolated ACL tears with BPTB (n = 6) or hamstrings tendon (n = 5) grafts participated in the study. They took part in the experiment 7.3 ± 1.1 months (range = 6–8.5 months) after surgery, when they were identified as able to return to sport activities using the following criteria: no pain during activities of daily living or rehabilitation sessions, no episodes of “giving way,” full range of motion at the knee joint, and in the last stage of the rehabilitation program. The rehabilitation protocols had a similar number of training sessions with similar content (ie, electrotherapy, neuromuscular reeducation, strengthening exercises, and functional training). Participants had no history of surgery or traumatic injury to the contralateral limb or of injury to the IL after the surgery. All participants provided written informed consent, and the study was approved by the Ethical Committee of the Université de Lyon – Université Claude Bernard.

Testing Procedures

The squat jump was chosen because this task is similar to movements performed in sport activities.¹⁸ Upper limb motion was forbidden so that we could focus on lower limb motion. No countermovement was allowed, thereby eliminating the stretch-shortening cycle effect, which enabled us to analyze the capacity to generate force during the concentric muscular-contraction phase. Participants performed warm-up and jump-training sessions to become familiar with the task and minimize possible sources of bias, such as countermovement. Subsequently, each participant performed 6 maximal single-legged squat jumps: 3 jumps on the IL and 3 jumps on the UL in randomized order. The initial position was the preferred position they chose, and they were instructed to jump as high as possible without downward movement. They were told to keep their hands on their hips for the duration of the jump to avoid back movements.

Instrumentation

Reflective markers were placed on the following anatomic landmarks according to the anthropometric data of Winter¹⁹: fifth metatarsophalangeal joint, lateral malleolus, lateral femoral condyle, greater trochanter, and acromion process. Participants jumped with the foot on a force platform (Advanced Mechanical Technology, Inc, Watertown, MA) with a frequency of 1000 Hz while they were filmed in the sagittal plane with a 100-Hz camcorder (model UI-2220SE; Imaging Development Systems, Obersulm, Germany).

Data Analysis

Both kinematic and kinetic data of the push-off phase of a single-legged squat jump were analyzed. The goal was to assess the recovery in terms of performance after ACL reconstruction. Kinetic and kinematic data were smoothed with a zero-lag, fourth order, low-pass Butterworth filter with cutoff frequencies of 15 Hz and 10 Hz, respectively. Kinetic data were down-sampled to 100 Hz. A 4 rigid-segments model (foot, shank, thigh, and upper body [ie, head, arms, and trunk]) was obtained from the digitalization of the center of the landmarks. The mean and standard deviation (SD) of the vertical ground reaction force (Rz) was determined over the first second in which the participant held the initial position. After this first second when Rz decreased more than 3 SDs below body weight, the trial was discontinued. The beginning of the push-off phase (POP) corresponded to the instant when, after the first second, Rz increased more than 2 SDs above body weight (adapted from Vanrenterghem et al²⁰). Next, the kinematic and kinetic data were synchronized. For this purpose, we determined the end of the POP for the kinematics and kinetics, which corresponded to the last frame when the foot was in contact with the ground and the last time sample before Rz decreased to zero, respectively. For further analysis, the best of the 3 trials, based on the maximal height reached during flight, was selected. The jump height was calculated by the raising of the body's center of mass between the standing position and the position at the apex of the jump. The location of the center of mass was determined using anthropometric data from Winter,¹⁹ and joint angles (hip, knee, and ankle) were calculated from the absolute coordinates of the landmarks. The net moments of the hip, knee, and ankle joints during the POP were calculated using a standard inverse-dynamic procedure. The joint power was equal to the product of the joint moment and the angular joint velocity during the POP. The values of moment and power were normalized by participant body mass. Maximal moments and powers corresponded to the maximal value of the moment and the power for the hip, knee, and ankle joints during the POP. At maximal joint power, the values of the joint moment and the angular joint velocity were reported. Total joint moment and power corresponded to the sum of each maximal moment¹⁶ and power of the lower limb joints. Positive moments and powers were defined as joint extension.

Statistical Analyses

All data were analyzed using a 2-factor analysis-of-variance model with lower extremity (IL, UL) as a repeated factor and group (BPTB graft, hamstrings tendon graft) as a between-subjects factor. The α level was set at .05. Effect sizes (Glass Δ) were calculated and reported when we found differences.²¹ Statistical analyses were conducted using the distribution free software R (version R.2.7.2; R Foundation for Statistical Computing, Vienna, Austria).

Joint Kinematics

After ACL reconstruction, knee-joint extension was less at takeoff in the IL than the UL. We think that a phenomenon of “overprotection” of the knee joint develops after surgery on the IL. Psychological factors, such as the fear of reinjury, have been shown to limit the extent of functional recovery after ACL reconstruction^12,22 and, therefore, may limit knee-joint extension. During dynamic movements and given the anatomic constraint,²³ antagonist muscles play the role of “brakes” against the movement to protect the joint. In this study, we speculate that early contraction of the hamstrings muscles slowed knee-joint extension during the performance of a single-legged jump. We also can suggest that this phenomenon could be related to the deficit in proprioception observed in the knee joint after ACL injury and surgery.^24,25 Katayama et al²⁵ described a correlation between reduced joint position sense and decreased jumping performance after ACL injury, whereas MacDonald et al²⁴ observed a higher threshold to detect passive motion of the knee joint in the IL than UL after ACL reconstruction. The reduced knee extension in the IL affects the ankle joint by limiting plantar flexion. The limitation of both the extension of the knee joint and ankle plantar flexion could decrease the vertical energy generation during the POP after ACL reconstruction, leading to a decrease in jumping performance in the IL compared with the UL.

Joint Kinetics

We observed a 14% deficit in total moment but no difference for maximal moments of the hip, knee, and ankle joints between the IL and UL. This result is related to the insignificant difference observed for each maximal joint moment. However, summing the 3 maximal moments amplifies the difference and leads to significance. Given that a moment is mainly the result of muscular forces, we suggest that the global muscular recovery of the lower limb is not complete in the IL after ACL reconstruction. These results contrast with the findings of Ernst et al,¹⁶ who reported equal total moments, decreased maximal knee moment, and no difference for the maximal hip and ankle moments between the IL and UL after ACL reconstruction. The nonstandardization of the testing procedures in their study¹⁶ limits the comparison with our results. Their participants performed a single-legged jump and used their natural jumping techniques with no restriction on countermovement or use of the upper limbs. Motion of the upper limbs can affect the mechanical output of the lower limb joints^26–30 and thereby modify jump performance by increasing the control of balance and changing the orientation of the body at takeoff.³¹ Moreover, Ernst et al¹⁶ assessed the stretch-shortening cycle, whereas we focused only on the concentric contraction during the POP of a jump. These differences in methods and testing procedures could explain the variability observed in the results between our studies.

Webster et al³² analyzed energy dissipation during eccentric contraction through the landing of a vertical jump. They found equal total moments and decreased maximal knee-flexion moment between the IL and UL and no effect of surgical technique (BPTB or hamstrings tendon grafts) on these variables. The difference in the mode of muscular contraction could explain the disparity between their results³² and ours.

We observed a decrease of 35% in total power that can be explained by the 34% decrease in maximal ankle power between the IL and the UL. By dividing the maximal ankle power into the joint moment and the angular velocity, we demonstrated that the deficit in maximal ankle power does not come from an alteration in joint moment but from a decrease in the angular velocity of ankle plantar flexion. Our results showed that maximal hip and knee power was not different between legs. Our findings contrast with those of Castanharo et al,¹⁷ who observed a decrease in the maximal knee power and similar maximal hip power during a bilateral squat jump. However, they did not measure the total power or maximal ankle power. The different outcomes between their study¹⁷ and ours could be related to the nature of the activity performed. Indeed, different motion strategies are adopted for double-legged and single-legged jumps.³³ Yeow et al³³ noted a greater hip and knee contribution to dissipate energy during a double-legged landing, whereas the hip and ankle dissipated most of the energy during a single-legged landing. Moreover, most ACL injuries occur during landing.³⁴ It is then that the fear of reinjury could limit the use of the knee joint and decrease knee power at landing.

In our study, kinematic and kinetic data were not different between ACL reconstruction with BPTB and hamstrings tendon grafts. These results are supported by researchers who have shown no differences among graft types used in ACL ligamentoplasty, especially for functional and kinematic outcomes.^1,35–37 Nevertheless, our results cannot be generalized due to the small sample sizes of the 2 subgroups.

The originality of our work is in the way we analyzed maximal joint power; dividing the power into the joint moment and the joint angular velocity allows more accurate identification of the deficit. Rehabilitation, therefore, could be individualized to optimize recovery after ACL ligamentoplasty.