Drop-Landing Performance and Knee-Extension Strength After Anterior Cruciate Ligament Reconstruction

This investigation was a descriptive laboratory study, and all measures were completed during a single testing session.

Twenty-two participants with ACLR and 24 healthy control participants enrolled in this study (Table 1). Individuals were assigned to the control group if they had no history of substantial lower extremity injury and no lower extremity injury within the 6 weeks before the study. Volunteers with ACLR were included if they had a history of primary, unilateral, uncomplicated ACLR with no substantial chondral resurfacing at least 9 months before enrolling. In addition, all ACLR participants had returned to recreational activity after clearance from a physician. All participants provided written informed consent, and the study was approved by the Institutional Review Board for Health Sciences Research at the University of Virginia.

After enrollment, participants completed patient-reported outcome measures, including the 2000 International Knee Documentation Committee (IKDC) Subjective Knee Evaluation Form,²⁴  Lower Extremity Functional Scale,²⁵  Tegner Activity Level Scale,²⁶  and a visual analog scale for knee pain during the previous 24 hours and after performing 10 consecutive body-weight–resisted squats with their hands on their hips. Next, they completed bilateral normalized knee-extension maximal voluntary isometric contraction (MVIC) testing followed by the LESS.

Participants were secured in a Biodex multimode dynamometer (System 3; Biodex Medical Systems Inc, Shirley, NY) with the hips and knees at 90° of flexion. Data were digitized at 125 Hz (model MP150; BIOPAC Systems Inc, Santa Barbara, CA). To expedite the testing procedure, the left limb was tested first for all participants, regardless of history of injury. Participants were secured to the Biodex dynamometer using a waist strap, and the testing limb was secured to the dynamometer arm 2 cm above the calcaneus. Next, they familiarized themselves with the equipment and the isometric knee-extension task, during which they were instructed to contract once at 25%, 50%, and 75% and twice at 100% of perceived maximum ability. Participants rested for at least 1 minute before data collection. During all knee-extension MVIC contractions, we provided constant oral encouragement, such as “keep going” and “push harder,” until a torque plateau was maintained for at least 2 seconds, and we instructed participants to maintain good seated posture.²⁷  Trials in which participants did not achieve a stable torque plateau or maintain good testing posture were discarded, and replacement trials were completed. Participants rested at least 1 minute between test contractions to prevent fatigue. After completion, the same procedures were conducted on the contralateral limb. Knee-extension MVIC torque was normalized to body mass (×Nm/kg) to allow for comparison among participants.

Drop-landing procedures were consistent with the original description of the LESS.²²  Participants jumped from a 30-cm-tall box onto a target located 50% of their own heights away from the box. They were instructed to jump out horizontally to the landing zone and, immediately after landing, to attempt a maximum vertical jump. Two standard handheld camcorders (HF R400 HD Flash Camcorder; Canon USA, Inc, Melville, NY) captured the drop-landing task for evaluation at a later time using the LESS. Both cameras were 48 in (121.92 cm) from the ground and secured to free-standing tripods, with 1 camera 136 in (345.44 cm) lateral to the landing zone and the other camera 136 in (345.44 cm) anterior to the landing zone. Before we recorded trials, participants practiced until they were comfortable with the task. They rested 30 seconds between trials. A mean score from 3 jump landings was used to calculate the LESS score for each participant.²²  We assessed the involved limb for the ACLR group and the self-identified nondominant limb for the control group.

Means and standard deviations were calculated for all demographic and outcome variables. We used independent-samples t tests to determine group differences in demographics, excluding sex (for which a Fisher exact test was used), and a Mann-Whitney U test to compare LESS scores between groups. Paired-samples t tests were performed to assess the between-limbs differences in normalized knee-extension MVIC torque within each group. Cohen d effect sizes and associated 95% confidence intervals were also calculated for the between-groups differences in LESS scores. Spearman ρ correlations were used to quantify the relationship between LESS score and both bilateral normalized knee-extension MVIC torque and participant-reported outcome measures. We chose Spearman ρ correlations because LESS scores are not truly continuous variables, so a nonparametric correlation was the more conservative approach. The α level was set at ≤.05. Statistical analysis was performed using SPSS statistical software (version 20.0; IBM Corporation, Armonk, NY).

Participants with ACLR exhibited greater LESS scores than healthy control participants (Cohen d = 1.45, 95% confidence interval = 0.80, 2.10; Figure 1). They also exhibited a between-limbs difference for normalized knee-extension MVIC torque (involved = 2.50 ± 0.84 Nm/kg, uninvolved = 2.92 ± 0.65 Nm/kg; t₂₁ = 3.49, P = .002), whereas healthy control participants did not exhibit a between-limbs difference (dominant = 2.90 ± 0.59 Nm/kg, nondominant = 2.77 ± 0.50 Nm/kg; t₂₃ = 1.52, P = .14). For all participants, more landing errors were moderately related to lower IKDC scores (Table 2). For participants with ACLR, more landing errors were moderately related to lower current Tegner scores (Table 2) and lower involved-limb normalized knee-extension MVIC torque (Figure 2). Landing errors by LESS item for each group are shown in Figures 3 and 4.

Easy, time-effective assessment tools that can be used clinically with little equipment are invaluable when attempting to monitor a patient's progress throughout rehabilitation and develop a plan for return to physical activity after ACLR. We used the LESS to investigate the effect of ACLR on the quality of lower extremity biomechanics while landing. We confirmed our hypothesis that participants with ACLR would have higher LESS scores, indicating more landing errors. The ACLR participants committed an average of 6.0 ± 3.6 errors, which is consistent with the findings of a previous study²³  involving participants with a similar activity level and time since surgery. In addition, we found that higher LESS scores were related to quadriceps weakness and a lower physical activity level in participants with ACLR. These observations suggested that after ACLR, individuals may experience altered movement patterns that persist beyond rehabilitation and return to physical activity. Monitoring landing biomechanics may help provide quantifiable milestones for rehabilitation that can aid in making return-to-activity decisions.

Paterno et al¹⁹  reported that individuals with ACLR were at nearly 6 times greater risk of subsequent ACL injury to the previously injured knee or the contralateral limb than healthy individuals at a 2-year follow up. In addition, these individuals are thought to be at greater risk of early-onset tibiofemoral and patellofemoral joint OA.²¹  Currently, clinicians use a wide variety of quantitative and nonquantitative assessment tools throughout rehabilitation to monitor patient progress and readiness to return to activity; however, injury rates continue to rise.²⁸  The LESS has been used widely as a prospective indicator of knee-injury risk in healthy control individuals; however, altered lower extremity biomechanics have consistently been reported during functional tasks at various follow-up points after ACLR.^17,29,30  In healthy individuals, a LESS score greater than 6 has been related to greater injury risk, but a similar threshold has not been established in individuals with injury histories.²²  When applying the threshold value of a LESS score greater than 6.0 to our sample, we determined that 10 of 22 participants with ACLR exhibited potentially harmful lower extremity movement patterns, whereas only 4 of 24 control participants scored greater than 6.0. Further investigation may be warranted to test the hypothesis of whether the LESS score can predict subsequent injury, such as graft failure, patellofemoral pain, or joint degeneration, in people with ACLR.

When looking at which errors were committed most commonly in each group, control participants most often landed with reduced knee flexion (20.8%) and increased knee valgus (20.8%) at initial contact, coupled with increased knee-valgus displacement (33.3%) and trunk-flexion displacement (25.0%; Figure 3). These errors are consistent with previously reported risk factors for primary ACL injury; however, the low overall LESS scores reported in this group indicated a potentially low risk of knee injury.²²  Participants with ACLR commonly exhibited reduced knee (72.7%) and hip (45.5%) flexion and increased knee valgus (40.9%), forward trunk flexion (54.5%), and lateral trunk flexion (40.9%) at initial contact (Figure 4). In addition, participants with ACLR had a high likelihood of increased knee-valgus displacement (54.5%) throughout landing. Increases in knee-valgus displacement³¹  and lateral trunk flexion³²  have been linked to increased loading of the ACL, whereas decreases in knee flexion and hip flexion at the point of initial contact have been linked to increased overall knee-joint loading during functional activities.^11,30  The combination of these risk factors for acute and potential long-term knee-joint injury highlights the detrimental effect of returning to physical activity if optimal neuromuscular control and lower extremity movement patterns are not restored after ACLR.

Evidence supports the hypothesis that lower extremity biomechanics change after ACLR. For example, researchers^17,29,30  have reported that external knee-flexion moment and peak knee-flexion angle during the loading phase of jump landings were less for participants with ACLR than for healthy control participants, suggesting an alteration in landing technique and force attenuation. These observations reflect a more knee-extended landing position with a smaller contribution of the knee extensors to combat the forces imposed on the knee joint during this high-loading task. Similar to other researchers who have used more sophisticated motion-analysis techniques, we found that the asymmetrical quadriceps strength driven by persistent reductions in strength in the involved limb is related to alterations in lower extremity movement patterns.^30,33  In this case, individuals experiencing weakness of the involved limb after ACLR were more likely to have higher LESS scores, indicating potentially harmful movement patterns associated with knee-injury risk (Figure 2). The return of normal bilateral quadriceps strength after ACLR represents an important and often frustrating clinical rehabilitation goal. Persistent reductions in quadriceps strength are thought to have major negative implications for knee-joint health after injury due to the inability to dynamically absorb forces at the joint during functional movement.³⁴  The observation that quadriceps strength of the involved limb within the ACLR group was related to LESS score (accounting for 27.4% of the variance in LESS score) provides further evidence that quadriceps weakness has implications for normal movement patterns while highlighting simple intervention strategies that may reduce subsequent injury risk.

Compared with uninjured individuals, patients commonly have reduced knee function³⁵  and physical activity levels^7,36  after ACLR. These changes in patient-reported function may persist for several years after the initial procedure despite the patient's receiving clearance to return to activity. In addition, researchers^37–39  have shown a strong relationship among persistent quadriceps dysfunction, functional performance, and self-reported knee function in patients with ACLR. In our investigation, participants with ACLR reported activity levels similar to those of the control group (Table 1); however, within the ACLR group, those reporting lower physical activity levels were more likely to exhibit potentially harmful landing biomechanics (Table 2). In addition, a relationship was not present in the ACLR group alone, but when all participants were considered together, those with lower self-reported knee function also had potentially harmful landing biomechanics. These findings, combined with observations by other researchers, represent an interesting and potentially clinically valuable interaction among poor patient-reported function, quadriceps dysfunction, and functional performance that reinforces the importance of a multifactorial approach to functional assessment before return to activity after ACLR.

Our study had several limitations that should be considered when interpreting the results. These data were part of a larger cross-sectional investigation. Given the limitations in this type of study design, we were not able to achieve as much experimental control as a prospective design would have allowed. We recruited a relatively homogeneous sample of patients based on demographics and physical activity level; however; the participants with ACLR had diverse experiences with rehabilitation after surgery and a wide range of times since surgery (31.5 ± 23.5 months). In addition, we were limited to assessment of isometric knee-extension strength at 90° of knee flexion. Despite the relationship between involved limb strength and LESS score in the ACLR group, future investigations should focus on more dynamic modes of strength assessment that are not limited to dynamometers.

Participants with ACLR displayed more errors while landing. Landing errors were negatively correlated with knee-extension strength, suggesting that weaker participants had more landing errors. Persistent quadriceps weakness commonly associated with ACLR may be related to poorer-quality lower extremity movement during dynamic tasks.