1 INTRODUCTION

Following anterior cruciate ligament (ACL) reconstruction a variety of measures including pain, range of motion, self-reported outcome measures, and functional performance tests are used to assess rehabilitation progression and to determine readiness for return to activity or sport.1-5 Benchmarks for rehabilitation progression are partially informed by achieving a defined level of symmetry between limbs (e.g., 80%–90% limb symmetry index [LSI]) for functional performance tests such as quadriceps strength or single-leg hop performance.6, 7 Individuals who achieve more symmetrical function before surgery and early in the rehabilitation process tend to have more favorable clinical outcomes.8-10 Greater levels of asymmetry are known to negatively impact patient self-reported function at later stages of rehabilitation (6–12 months).9, 11-15 This is concerning since individuals who have deficits at 6 months are two times more likely to continue to have deficits 1–2 years following ACL reconstruction14 and are at 4–5 times greater risk of reinjury.1, 3 Better understanding factors that contribute to decreased function in earlier stages of rehabilitation can help clinicians better prioritize treatment options.

While a variety of factors influence rehabilitation progression or delay reaching clinical milestones (e.g., pain, swelling, range of motion, graft type),16-19 limb dominance is not thought to negatively impact rehabilitation outcomes following ACL reconstruction.20-22 A limitation of these studies was that the dominant limb was defined as the limb used to kick a ball. While decisions regarding ACL rehabilitation progression or return to sport may be informed by kicking performance tests,23 other more traditional measures focus on strength and hop performance. The preferred limb to kick a ball and the preferred limb to perform a single-leg vertical jump are often not in agreement.24, 25 Healthy individuals have a preferred limb for unilateral jumping and hopping tasks and typically demonstrate better performance on that limb.24, 25 It is not clear how injury to the preferred or nonpreferred jumping limb impacts patient function during intermediate stages of rehabilitation when patients are starting to perform more advanced rehabilitation. The purpose of this study was to determine if postoperative strength and functional outcomes differ between individuals who injured their preferred or nonpreferred jumping limb and to determine if these same outcomes differ between individuals who injured their preferred or nonpreferred limb used to kick a ball. It was hypothesized that individuals who injured their preferred jumping limb would demonstrate higher self-reported functional status and have more symmetrical quadriceps and single-leg forward hop LSI values compared to individuals that injured their nonpreferred jumping limb. We further hypothesized there would be no differences in clinical outcomes when dichotomizing groups based on the more traditional method of the preferred limb used for kicking a ball. In an effort to better direct rehabilitation strategies, a secondary purpose was to determine the association of quadriceps strength and single-leg forward hop performance with patient self-reported function during the intermediate stage (3–6 months) of ACL rehabilitation. We hypothesized that quadriceps LSI values would be more strongly associated with patient-reported outcome measures than single-leg forward hop LSI values.

2 METHODS

Study approval was granted by the Institutional Review Board at Creighton University (IRB: 636798, 857046, 928791) and was a secondary analysis of existing data. Forty participants with a history of a unilateral ACL reconstruction were included in this analysis (Table 1). Participants were included if they were 13–35 years old and had a history of primary unilateral ACL reconstruction in the past 3–6 months (meniscus or other ligament pathology permitted). A timeline of 3–6 months post-ACL reconstruction was selected since initial impairments (e.g., ROM, swelling) have resolved for most individuals and they are starting to transition to more advanced exercises and activities to prepare for return to sport. Participants were excluded if they had a concomitant knee injury such as cartilage defect >2 cm or fracture, spine or lower extremity injury/surgery (except ACL) within the past year, contralateral knee surgery, current pregnancy, or inability to give consent or understand procedures of the experiment. All participants completed an approved informed consent form (assent/parental permission for minors), standardized health history form, self-reported forms related to preinjury physical activity level (Tegner Activity Scale), and patient reported measures of function (IKDC Subjective). Participants also answered two questions specific to the preferred limb to (1) perform a single-leg vertical jump and (2) kick a ball. If they were unable to determine limb preference, they were allowed to practice the task to determine preference. Before data collection height (cm), body mass (kg), and leg length (cm) were obtained (Table 1).

Quadriceps strength testing [Color figure can be viewed at wileyonlinelibrary.com] Table 1. Participant demographics Sex Male 12, Female 28 Age 20.0 ± 4.6 years Height 174.2 ± 8.3 cm Mass 71.2 ± 12.7 kg Time since surgery 5.3 ± 0.8 months IKDC subjective 78.1% ± 14.4% Tegner activity scale (preinjury) 8.3 ± 1.4 Graft type 33 hamstring tendon graft 4 bone-patellar-bone graft 3 other graft types (allograft) Injured limb based on preferred limb to kick ball 19 injured preferred limb to kick ball 21 injured nonpreferred limb to kick ball Injured limb based on preferred limb for single leg vertical jump 21 injured preferred limb for single leg vertical jump 19 injured nonpreferred limb for single leg vertical jump 2.1 Quadriceps strength testing

Quadriceps isometric strength was assessed using an electromechanical dynamometer (Biodex System 3; Computer Sports Medicine Inc.) which has good to excellent intersession reliability (ICC 0.94–0.98).26 Participants were positioned using standardized procedures with their hips and knees flexed at 90° with straps placed across the chest and waist (Figure 1).26-29 The dynamometer was interfaced with a data acquisition system (MP150; Biopac Systems, Inc.) with torque data sampled at 2000 Hz. Participants performed a progressive warm up consisting of submaximal and maximal isometric contractions. Loud verbal encouragement and visual feedback (90% and 100% of maximum torque obtained during warm up) was provided to ensure maximal effort.30 Following the warm-up, participants performed six maximal isometric contractions with a 30 s break between trials. The average peak torque (Nm) from the two highest trials was used for data analysis. The uninvolved limb was tested first followed by measures repeated on the involved limb. LSI values were calculated and expressed as a percentage (LSI = Involved/Uninvolved × 100).

Single leg forward hop [Color figure can be viewed at wileyonlinelibrary.com] 2.2 Single-leg forward hop

Participants performed a single-leg forward hop (Figure 2) in which they jumped for maximal distance and a controlled landing on one limb.2, 9 Single-leg forward hop has excellent between session reliability (ICC 0.92–0.95) and minimal detectable change of 8% for the LSI.31 Six successful trials were completed on each limb. A 30 seconds rest was given between trials. Distance hopped was measured in centimeters from the tip of the toe at the start of the jump to the back of the heel at landing. Trials were successful if the participant stuck the landing without the opposite leg or hand touching the ground, maintained that position for 2 seconds, and had no additional hops upon landing or sliding of the heel.31-33 Consistent with strength testing, the uninvolved limb was tested first followed by involved limb. Average of the furthest two jumps were used for data analysis and LSI values were calculated (LSI = Involved/Uninvolved × 100).

2.3 Statistical analysis

Statistical analyses were performed using SPSS version 26.0 (SPSS Inc.). Significance for all analyses was set at a priori with an alpha level ≤0.05. Primary outcome measures included, IKDC scores, quadriceps LSI values, and single-leg forward hop LSI values. Participants were placed into one of two groups based on if they tore the ACL in their preferred jumping limb or nonpreferred jumping limb. To allow comparison between other studies participants were also dichotomized based on injury to the preferred limb or nonpreferred limb to kick a ball. Independent t tests were performed to determine differences in primary outcome measures between groups. Effect sizes (Cohen's d) were calculated to interpret the magnitude of differences between groups with effect sizes interpreted as follows: small (0.20), moderate (0.50), and large (0.80). Pearson product-moment correlation coefficients were used to determine relationships between outcome measures and stepwise linear regression was used to determine the ability of quadriceps LSI values and single-leg forward hop LSI values to predict IKDC scores. Predictor variables that were significantly correlated with IKDC scores were entered into the model. Correlations were qualified as good to excellent (r ≥ 0.76), moderate to good (r = 0.50–0.75), fair (r = 0.25–0.50), and little to no relationship (r ≤ 0.25).34

3 RESULTS

There were no significant differences between groups when using injury to the preferred or nonpreferred limb to kick a ball for any of the outcome variables (Table 2). When using injury to the preferred or nonpreferred jumping limb to dichotomize groups there were significant differences (p = 0.05, d = 0.77) in the single-leg forward hop LSI values (Table 2). Individuals who injured their nonpreferred jumping limb (81.1% ± 19.5%) demonstrated significantly lower single-leg forward hop LSI values compared to individuals who injured their preferred jumping limb (94.1% ± 12.6%). No significance differences between groups based on injury to the preferred jumping limb were found for IKDC values (p = 0.31, d = 0.33;) or for quadriceps LSI values (p = 0.26, d = 0.32).

Table 2. Study results based on preferred limb to kick a ball or preferred limb for jumping Injured limb based on preference to kick ball Injured limb based on preference to jump Dominant Nondominant p d Dominant Nondominant p d IKDC 78.1% ± 13.8% 78.1% ± 15.2% 0.99 0.01 80.4% ± 14.0% 75.7% ± 14.7% 0.31 0.33 Quadriceps LSI 78.3% ± 17.4% 80.8% ± 21.8% 0.69 −0.13 82.6% ± 21.2% 76.3% ± 17.6% 0.31 0.32 Single leg forward hop LSI 87.0% ± 18.4% 89.6% ± 16.2 0.70 −0.15 94.1% ± 12.6% 81.1% ± 19.5% 0.05 0.77

There was a significant correlation between all outcome variables (Table 3) and both quadriceps LSI and single-leg forward hop LSI were entered into the regression model. Quadriceps LSI had the strongest association with IKDC scores and explained 62% of the variance in IKDC scores (Table 4). Single-leg forward hop LSI added to the predictive model (10.5%) explaining a total of 73% of the variance in IKDC scores.

Table 3. Correlations between patient-reported and performance outcome measures IKDC Quadriceps LSI IKDC – Quadriceps LSI 0.790* – Single leg forward hop LSI 0.778* 0.685* Table 4. Regression models to predict IKDC Scores at intermediate stages of ACL rehabilitation (3–6 months postsurgery) Model Variable Standardized β p R R 2 R 2 change SE 1 Constant <0.001 0.790 0.624 0.624 0.086 Quadriceps LSI 0.790 2 Constant <0.001 0.854 0.729 0.105 0.074 Quadriceps LSI 0.485 Single leg forward hop LSI 0.445 4 DISCUSSION

The primary aim of this study was to determine if postoperative strength and functional outcomes differ between individuals who injure their preferred or nonpreferred jumping limb and to determine if these same outcomes differ between individuals who injured their preferred or nonpreferred limb used to kick a ball. Following ACL reconstruction, when the nonpreferred jumping limb was injured, LSI values for the single-leg forward hop were significantly less than values for individuals who injured their preferred jumping limb. This finding was consistent with the original hypothesis. Self-reported function (IKDC scores) and quadriceps LSI values were not impacted by injury to the preferred jumping limb which was inconsistent with the original hypothesis. There were no significant differences between groups based on injury to the preferred limb to kick a ball for any of the outcome variables and confirmed the original hypothesis. Both quadriceps LSI and single-leg forward hop LSI explained a substantial amount (73%) of the variance in IKDC scores in the intermediate phase of ACL rehabilitation (3–6 months).

4.1 Jumping preference influences LSI scores

Individuals who injured their nonpreferred jumping limb demonstrated significantly lower single-leg forward hop LSI values compared to individuals who injured their preferred jumping limb. These results contrast with previous studies that indicated limb dominance did not influence single-leg forward hop symmetry21 or landing biomechanics.22 Differences between studies may be explained by how limb dominance was defined, specifically the preferred limb used to kick a ball was utilized in previous studies.21, 22 Limb preference in the current study was defined as the limb in which participants would use to jump off of for maximal height (i.e., single-leg vertical jump). For comparison to other studies, participants also provided information in context to limb preference for kicking a ball, which is the more traditional method to determine limb dominance.20-22 There was no difference in study outcomes when the preferred limb was defined as the limb to kick a ball. These results are largely in agreement with previous studies which suggest that leg dominance, defined as limb used to kick a ball, does not have a significant impact on strength,20 hop symmetry,21 landing biomechanics,22 short-term functional outcomes,20 or return to sport rates.20 Clinically, results from the current study suggest that it may not be possible to broadly define a dominant limb for the lower extremity based on a single task (e.g., kicking a ball).

4.2 Preference is determined by the task

Decisions regarding ACL rehabilitation progression or return to sport are not traditionally informed by kicking performance tests,23 thus operationally defining limb dominance by the limb used to kick a ball may have less relevance outside of kicking related sports such as soccer or rugby. Limb dominance or more accurately stated, limb preference, is dependent upon the task being performed. The Waterloo Footedness Questionnaire can be used to define limb dominance and questions such as the limb used to kick a ball, jump off, pick up marbles, smooth sand, sto​mp out fire, push shovel into the ground, preferred to balance on, first to go down stairs, limb felt to be more powerful, or limb in front for a sprint are used to define limb dominance.25, 35 For individuals who are right handed, the right lower limb will be chosen 80%–100% of the time for tasks that are considered manipulative in nature (stepping on bug, writing word, kicking ball).25, 36 A study assessing postural control found differences between limbs when using the preferred limb to kick a ball to define the dominant limb.37 Postural control is a manipulative task, thus using kicking limb to define dominance should be used in that instance. Jumping is not considered a manipulative task and may explain why using the preferred limb to kick a ball may not result in between group differences. Using a definition for limb dominance that uses a manipulative task (e.g., kicking a ball) as a reference is not ideal when considering performance for a single-leg hopping task that is not manipulative in nature. Since participants in the current study were not performing lower extremity tasks that were considered manipulative in nature it would seem intuitive to determine limb preference in a manner that is more consistent with the task being performed. When using the preferred jumping limb (a nonmanipulative task), to determine limb dominance, differences were found between limbs for single-leg forward hop LSI values. Interpretation of postoperative outcomes following ACL reconstruction that consider limb dominance or preference should be specific to the task completed. Previous research has shown the limb for a single-leg vertical jump and limb to kick a ball often are not in agreement.24, 25, 38 Since healthy individuals demonstrate better hop test performance on their preferred jumping limb24, 25 hop limb preference may be a more relevant factor to consider following ACL reconstruction.

4.3 Strength and self-reported outcome measures

Obtaining symmetrical quadriceps strength is an important clinical benchmark following ACL reconstruction.39, 40 While performance before injury and limb preference may impact post-injury LSI values the results of this study suggest that limb preference does not negatively impact quadriceps LSI values or patient self-reported function as assessed with the IKDC subjective form. Although individuals who injured their nonpreferred jumping limb had on average 6.3% lower quadriceps LSI values and 4.7% lower IKDC scores, the differences from those who injured their preferred jumping limb were not statistically significant and the effect size was small to moderate (Table 2). The magnitude of differences was even lower when examining groups based on limb preference to kick a ball. While outcomes for quadriceps LSI and IKDC scores were not different based on the injured limb preference for a task (kicking ball, jumping) it is possible that preference for a jumping or kicking task may not be in agreement with the muscle strength before injury. Since preinjury strength or single-leg forward hop test data were not available in the current study it was not possible to determine the limb with better performance or perceived to be stronger before injury. Similar to limb preference being task specific for single-leg forward hop outcomes it is plausible that strength should be defined in a similar manner (e.g., stronger limb). Although healthy individuals have relatively symmetrical strength,40 studies of elite athletes suggest nearly half (53%) have a quadriceps strength LSI value <90%.37 This suggests the potential for a substantially different strength profile between limbs. Conceptually, if the injured limb had lower performance before injury, relative to the uninjured limb, this difference would only be of greater magnitude after injury. Future studies would need to track outcomes both before and after injury to better determine if the limb which demonstrates higher performance capacity (i.e., stronger, greater hop distance) better recovers symmetrical performance after ACL reconstruction compared to if the lower performance limb is injured.

4.4 Outcomes to predict self-reported function

More symmetrical quadriceps strength and hop test performance are known to positively influence patient self-reported function during later stages of ACL rehabilitation (6–12 months).9, 11-15, 41 Results from the current study suggest that higher quadriceps LSI values and single-leg forward hop values explain 73% of the variance in IKDC scores at more intermediate stages of ACL rehabilitation (3–6 months postsurgery). While previous studies42, 43 have shown lower values for correlation coefficients, differences may be due to examination of change scores over time42 or examination of strength measures earlier (2 months postsurgery) in the rehabilitation process.23 Despite subtle differences between studies, clinical translation of these results suggests that initial emphasis should be placed on achieving more symmetrical quadriceps strength followed by achieving more symmetrical hop test performance. Symmetrical and sufficiently strong quadriceps are known to more positively contribute to patient function, compared to single-leg forward hop test performance, as patients are transitioning back to activity.15 An emphasis for prioritizing strength before initiating hopping activities has been suggested in clinical practice guidelines which indicate individuals should have at least 80% quadriceps LSI before beginning hopping and jogging activities.7, 18 While data from the current study was used clinically to inform rehabilitation progression, specific decisions by individual clinicians regarding hopping or jogging progressions were not monitored.

4.5 Limitations and future directions

A limitation of this study was that we did not have extended follow up to determine if timelines to resolve clinically meaningful impairments in performance measures and patient self-reported function differ between groups or to determine if impairments continued to persist during later stages of rehabilitation. It is possible that with extended follow up that LSI values would tend to reach more of an equilibrium regardless of the injured limb preference for task due to either improvements in involved limb function or a relative decline of contralateral limb performance.44 Additionally, a variety of factors (e.g., pain, swelling, range of motion, graft type) are known to influence rehabilitation progression or delay reaching clinical milestones.16-19 Future studies should consider a more longitudinal assessment which would provide researchers and clinicians more comprehensive insights into the influence of limb preference on postoperative outcomes and the timeline for reaching clinical milestones. A second limitation was that only a single hop test was performed versus a more robust battery of hop tests. Hop tests including the triple hop,13, 45 cross-over hop,9 6 m timed hop,9 30 s side hop,46 and single-leg vertical jump46, 47 have all been shown to individually contribute to patient outcomes following ACL reconstruction. While there is a degree of redundancy for hop test performance in a forward direction,48 more complex hops (e.g., cross-over hop, timed side hop) may result in different outcomes between groups based on preferred limb jumping injury or may contribute to patient self-reported function to a different degree. Currently, there is no gold standard battery of functional assessments to determine return to sport readiness following ACL reconstruction, but a commonality across most test batteries is inclusion of quadriceps strength and single-leg forward hop performance.1-3 Clinicians should consider a robust battery of hop tests, performed in different directions (forward, lateral, vertical), and include assessment of both performance (hop distance) as well as quality of movement48 to gain comprehensive insights into patient function. A third limitation was that we could not provide greater insights into unique patient characteristics such as age, graft type, or specific sport specialization due to a smaller sample size, especially when compared to other studies with a greater number of participants (n = 88–139).12, 13, 40 Regarding graft type, there were a substantially greater number of participants in this study with hamstring tendon autografts versus bone-patellar tendon-bone or other (e.g., allograft) graft, but there was a nearly even distribution across graft types when dichotomizing participants based on preferred kicking or jumping limb. Graft type is known to influence quadriceps- and hamstring strength-related outcomes differently49 and individuals with bone-patellar-bone autografts are known to be slower at reaching clinical milestones.16 Due to the lower number of bone-patellar tendon-bone and other graft types it was not possible to conduct adequately powered, specific analyses. Additionally, sport specialization may influence outcomes. Specifically, a soccer player may be more likely to kick a ball with their right leg, but a basketball player may be more likely to prefer to jump from their left leg for a right-handed layup. Previous research has shown that the preferred limb to perform a single-leg vertical jump or to kick a ball may not be in agreement,24, 25, 38 but clinically meaningful differences between limbs are not thought to be influenced by sport.50 While the health history form used in this study did allow identification of a primary sport, participants often listed a variety of secondary sports. Anecdotally, it was not uncommon for a participant in this study to be a high school basketball player, but also play soccer. Thus, it was not possible to conduct an adequately powered, sport specific analysis. Despite these limitations, this study does provide clinicians with insights that suggest single-leg forward hop test outcomes within 3–6 months of ACL reconstruction may be partially influenced by if the patient injured their preferred or nonpreferred jumping limb. This study further highlights the substantial need to restore appropriate quadriceps strength, then hop performance, in the intermediate phases of rehabilitation as these measures have significant impact on patient self-reported function. Clinicians should consider how pre-injury limb preference to perform activity may impact post-injury outcomes, especially when comparing between limb performance (i.e., LSI values).

ACKNOWLEDGEMENTS

Portions of research reported in this publication was supported by the National Institute of General Medicine Sciences of the National Institutes of Health under award number 5P20GM109090 (PI Stergiou), NASA Nebraska Space Grant Consortium, and Creighton University School of Pharmacy and Health Professions Student Research Fellowship Grant.

CONFLICT OF INTERESTS

The authors declare that there are no conflict of interests.

AUTHOR CONTRIBUTIONS

Substantial contributions to the conception or design of the work: Brooke Farmer, Dillon Anderson, Dimitrios Katsavelis, Jennifer J. Bagwell, Kimberly A. Turman, and Terry L. Grindstaff. Acquisition, analysis, or interpretation of data: Brooke Farmer, Dillon Anderson, and Terry L. Grindstaff. Drafting the work: Brooke Farmer, Dillon Anderson, and Terry L. Grindstaff. Critical revisions for important intellectual content: Dimitrios Katsavelis, Jennifer J. Bagwell, Kimberly A. Turman, and Terry L. Grindstaff. All authors have read and approved the final submitted manuscript and are in agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

ETHICS STATEMENT

Granted by the Institutional Review Board at Creighton University (IRB 636798, 857046, 928791). Written informed consent was obtained from each participant before participation.