1 INTRODUCTION

Anterior cruciate ligament (ACL) injuries are common in athletes,1 and surgical reconstruction is typically recommended to allow sports participation without knee instability.2 Patients with ACL reconstruction (ACLR) typically have high expectations for full recovery, including the resumption of pre-injury sport activities,3 but as many as 45% fail to return to competitive sport.4 Physical impairments, such as quadriceps weakness,5 and psychological factors, such as high fear of re-injury and poor self-efficacy,6, 7 can deter a patient from returning to sport participation following ACLR. If a patient does resume sport participation following ACLR, the risk of reinjury is high8 despite recent research highlighting the importance ensuring that patients pass return to sport criteria before receiving clearance for sport participation.9-11 The return to sport outcomes emphasize the need to adequately prepare patients with ACLR physically and psychologically for sports participation.

Standard ACLR rehabilitation includes exercises focused on resolving physical impairments after surgery (e.g. knee range of motion, lower extremity strength and neuromuscular control).12 Between 6 and 12 months postoperative readiness for return to sport is assessed with clinical assessments such as hop tests, quadriceps strength testing and drop landing movement assessment to determine whether recovery is sufficient to safely allow resumption of sport activity.9-11, 13 At this time, there is a lack of standardized assessment components for return to sport or agreement on the criteria for determining readiness for sport participation.14 Furthermore, although assessment of psychological readiness for sport has been recommended,15 it is not performed as frequently. Poor correlation between physical and psychological measures suggests that these are separate factors that cannot be used as surrogates for each other.16, 17

Physical impairments5 and fear of reinjury18, 19 can persist at the end of ACLR rehabilitation. Moreover, patients that complete rehabilitation and pass return to sport criteria often receive little direction on how to continue the return to sport continuum without supervision.17 Advanced training programs with more sport-like activities have been used to bridge the end of standard rehabilitation and the transition into full athletic competition.20-22 These programs usually target the 6–12 months postoperative time period and generally include lower extremity strengthening and neuromuscular training (agility, perturbations and/or plyometrics) at an increased intensity level compared to standard rehabilitation.16, 23 Research has shown that potential benefits of advanced training include improved function,22-24 improved passing rate on return to sport criteria consisting of physical impairment and psychological measures,16 and possibly prevention of secondary (re)-injury22, 25 Even without an explicit psychological intervention, previous studies have shown that patients who complete advanced training have improved overall psychological readiness for sport participation26 and improved self-efficacy for knee-related activities.24 To our knowledge, no study has directly compared physical and psychological outcomes in patients that did or did not participate in advanced training during the return to sport phase following ACLR.

This study compared readiness to return to sport based on completion of advanced training after ACLR. We hypothesized that patients who completed advanced training after ACLR would show better psychological readiness for sport and passing of return to sport criteria than those who did not complete advanced training over a similar time interval.

2 MATERIALS AND METHODS 2.1 Study overview

This prospective cohort study (level of evidence 2) included two groups of patients with ACLR: one group participated in advanced training (TRAINING) and the other did not (NoTRAINING). Baseline and follow-up testing were conducted, and the time points coincided with pre- and post-training for TRAINING and a similar time interval for NoTRAINING (Figure 1). Testing consisted of demographic information, psychological measures and hop testing.

Study enrollment

2.2 Participants

Patients with ACLR who completed standard rehabilitation and received physician clearance to initiate a return to sport between August 2016 and November 2019 were eligible to participate in this study. Inclusion criteria were: (1) age 14–30 years at the time of surgery; (2) received primary ACLR; (3) participated in competitive sports that require cutting, jumping or pivoting before injury; (4) stated an intention to return to pre-injury activity level at the time of study participation; (5) ≥5 months postoperative; (6) symmetrical knee extension and knee flexion within 5 degrees of the contralateral knee; (7) able to hop vertically on the surgical leg without knee instability or pain; (8) trace effusion or less; and (9) surgeon approval for participating in advanced training. Exclusion criteria included previous lower extremity surgery or additional surgical procedures that altered postoperative rehabilitation (i.e., concomitant meniscal or ligamentous repair or cartilage stimulating procedure). Study criteria were meant to identify patients with good potential to return to sport after ACLR.

All patients were offered the opportunity to participate in a voluntary, cash-based (USD300) advanced training program offered at TRIA Orthopedic Center. Patients that enrolled in advanced training (TRAINING) were matched by age (±1 years) and sex to patients that did not enroll in advanced training (NoTRAINING). Patients in NoTRAINING were asked to avoid participation in other supervised advanced training programs but did not otherwise have any restrictions on their sports participation. All participants gave written consent or assent (minors) for this study on a form approved by the institutional review board at our facility.

2.3 Advanced training

The advanced training program has been described previously.16 Training is held in a group format and consists of two sessions per week for five weeks (10 total sessions), with a maximum of 8 participants who start and finish training collectively. Each training session lasts two hours and starts with a dynamic warm-up, followed by the exercises prescribed for that session, and ends with a cool down. Advanced training includes plyometric exercises, dynamic knee stability exercises, lower extremity and core strengthening exercises and agility drills (Table 1). A pre-determined exercise progression is followed unless modifications are deemed necessary to ensure safety for individual patients. Generally, exercises are progressed from double leg, stable surface and slow movement to single leg, unstable surface and quick movements. Perturbations are incorporated via unstable surface (Airex pad, BOSUball) to challenge dynamic knee stability. The program is staffed with one physical therapist and one certified athletic trainer in order to provide each participant with adequate feedback on movement patterns during exercise.

Table 1. Example components of the advanced training program Category Type Example Plyometrics Lateral Single leg lateral hop Anterior Triple hop Tuck Double tuck jump Lunge scissor Scissor jump unilateral weight Rotatory 90 degree hops on foam Strength Trunk Side plank Quadriceps Lunge with shoulder press Hamstring Single leg bridge on BOSU Agility Foot speed Ladder drills Cutting Cone drills Reaction Reaction ball 2.4 Demographic information

Demographic information collected at baseline included age, sex, body mass index, injury type, graft type, time from surgery to baseline, and the primary reason for electing to participate or not participate in advanced training. Injury type was categorized as “contact” or “Noncontact” based on patient recall of whether or not there was contact with another player at the time of injury. Autograft type was categorized as “hamstrings tendon” or “bone-patellar tendon-bone.” Activity level was quantified at baseline and follow-up with the Marx Activity Rating Scale, which includes four items pertaining to cutting, jumping, deceleration and running.27 Items are scored from 0 (less than 1 time in a month) to 4 (4 or more times in a week) for a total score of 16 points, and higher scores indicate a higher activity level. Lastly, patients were asked to list the primary preinjury sport or activity they planned to resume.

2.5 Psychological measures

The primary psychological measure was the Anterior Cruciate Ligament Return to Sport after Injury (ACL-RSI) scale, which determines general psychological readiness for sport participation.28 The questionnaire contains 12 items that cover the domains of emotions, confidence in performance, and risk appraisal. Items are scored 0–100 points and the total is divided by 12, giving a score range from 0 (a completely negative psychological outlook) to 100 (a completely positive psychological outlook). The ACL-RSI has acceptable reliability, validity and test-retest reliability,28, 29 and ACL-RSI scoreshave distinguished athletes who return to their pre-injury level of sport after ACLR.7, 30 An ACL-RSI score of 70 points was considered to indicate an appropriate level of psychological readiness for sport asprevious research identified a mean score of 70.6 ± 19.4 at 6 months post-surgery for those that returned to sport by 2 years post-surgery.31

Secondary measures focused on kinesiophobia, self-efficacy for activities involving the knee, and intensity of fear related to a task or situation specific to each patient. Previous research has shown that the level of kinesiophobia is higher and level of knee activity self-efficacy is lower in patients that do not return to sport following ACL reconstruction.18, 32-34 In addition, the intensity of fear is highest for activities and situations that are specific to each patient.26 Kinesiophobia, or pain-related fear of movement/reinjury, was assessed with the 11-item Tampa Scale for Kinesiophobia (TSK-11).35 Items are scored from 1 (strongly disagree) to 4 (strongly agree) and summed to create a total score that ranges from 11 to 44 points, with higher scores indicating greater kinesiophobia. TSK-11 has been shown to be relevant to the ACLR population.36 Knee Activity Self-Efficacy (KASE) was assessed with a 10-item questionnaire that measures self-efficacy, or confidence, for performing activities involving the knee such as hopping on the injured leg, changing directions while running, and avoiding new knee injury.24 Items are scored from 0 (strongly disagree) to 10 (strongly agree) and summed to create a total score that ranges from 0 to 100 where higher scores indicate greater self-efficacy in knee related activities. KASE has shown good test-retest reliability.24 Patients were also asked to identify the top 3 tasks or situations for which they were most afraid after injury and rate the intensity of fear for each task on a scale of 0 (no fear) to 10 (greatest fear imaginable). The primary (top-ranked) fear-evoking task or situation and corresponding fear intensity was used for analysis.

2.6 Hop testing

A single-leg hop testing battery consisted of single hop for distance, triple hop for distance, crossover triple hop for distance, and six meter timed hop.37 Patients performed three trials of each hop test on the nonsurgical limb followed by the surgical limb, and the best performance for each hop test was recorded. Limb symmetry was calculated as [(surgical limb performance/nonsurgical limb performance) × 100] for distance measures and [(nonsurgical limb performance/surgical limb performance) × 100] for the timed measure. This hop testing battery has been shown to be reliable and valid for patients following ACLR38 and is recommended to assess functional readiness to return to sports following ACLR.9, 10, 15

It is recommended that patients achieve 90% limb symmetry on all tests in the battery as criteria for returning to sport.9, 10, 13

2.7 Statistical analysis

Sample size was estimated from ACL-RSI scores and the proportion of patients meeting return to sport criteria in a previous study on advanced training after ACLR.16 For the calculation, pre-training values were used to represent NoTRAINING, and post-training values were used to represent TRAINING. An online calculator (https://www.stat.ubc.ca/~rollin/stats/ssize/) with requirements of 80% power and alpha level 0.05 determined a sample size between 19 and 24 participants per group was necessary.

Statistical analysis was performed with SPSS Statistics version 26 (IBM Corp). Descriptive statistics were generated for all variables. Demographic variables were compared between TRAINING and NoTRAINING using independent sample t-test or Chi-square test, as appropriate. Statistical significance was set at p < .05 for all analyses.

The primary outcomes were ACL-RSI score and proportion of patients meeting the return to sport criteria at follow-up. Group differences in ACL-RSI score at follow-up were analyzed with a general linear model that included baseline score as a co-variate. Readiness for return to sport was based on meeting the pre-defined criteria for ACL-RSI score (≥70 points) and hop testing limb symmetry (≥90% on each test in the battery). The proportion of patients meeting the return to sport criteria at baseline and follow-up was compared between groups with Chi square statistic. Secondary outcomes of TSK-11 score, KASE score, and fear intensity for the primary fear-evoking task or situation were analyzed with separate general linear models with baseline score as a co-variate.

3 RESULTS

A total of 46 patients with ACLR (n = 23 in both groups) participated in the study (Figure 1). Participants in TRAINING and NoTRAINING were similar across all baseline demographic variables (Table 2). The groups were comprised of more females (60%), Noncontact injuries (74%) and reconstruction with patellar tendon autograft (70%). Most of the participants in TRAINING (n = 21, 91%) stated they enrolled in the program because of a desire to return to sport, while one participant cited the close driving distance and another cited the relatively low cost of the program. Participants in NoTRAINING elected not to enroll in the advanced training program because of schedule constraints related to the time of day when training sessions were held (n = 7) or the proximity to the start of the competitive season (n = 3), receiving rehabilitation at another facility (n = 5), driving distance (n = 6) or cost (n = 2). Table 3 lists the primary preinjury sport or activity for patients in both groups.

Table 2. Demographic information. Continuous variables are presented as mean (SD) Variable TRAINING (n = 23) NoTRAINING (n = 23) Age (years) 19.2 (5.4) 19.5 (5.3) Sex (n) Male 9 9 Female 14 14 Injury type (n) Contact 6 6 Noncontact 17 17 Graft type (n) Hamstrings 7 8 Patellar tendon 16 15 Body mass index (kg/m2) 23.9 (2.1) 23.5 (4.5) Surgery to baseline (days) 181.8 (17.2) 185.6 (18.4) Marx activity Baseline 15.2 (1.7) 14.0 (2.6) Rating (points) Follow-up 15.4 (1.6) 13.2 (3.6) Table 3. Primary preinjury sport or activity TRAINING NoTRAINING Soccer 8 2 Basketball 4 5 Hockey 2 5 Lacrosse 1 1 Football 2 0 Skiing/snowboarding 0 4 Cricket 1 1 Dance 0 3 Crossfit 1 0 Volleyball 2 0 Baseball 1 1 Cheer 1 0 Gymnastics 0 1

ACL-RSI score at follow-up was not significantly different between groups (p > .05, Table 4). Post-hoc testing with the groups combined showed a significant increase in ACL-RSI score from baseline to follow-up (p = .018). The proportion of patients who passed the return to sport criteria was not significantly different between groups at baseline (3/23 in TRAINING vs. 7/23 in NoTRAINING; p = .153) or follow-up (12/23 in TRAINING vs. 10/23 in NoTRAINING; p = .555). Further analysis showed that while the groups did not significantly differ on the distribution of return to sport criteria met at baseline (Figure 2, p = .334), significant group differences were found at follow-up (p = .011, Figure 3). Specifically, more patients in TRAINING than NoTRAINING met hop testing criteria only (n = 7 vs. n = 1, respectively) and, more patients in NoTRAINING than TRAINING failed to meet any criteria at all (n = 6 vs. n = 0, respectively). Inspection of change in criteria met from baseline to follow-up for each patient showed that 5 patients in TRAINING and 12 patients in NoTRAINING did not improve on the number of criteria met (stayed the same or declined). Finally, the single-leg hop test limb symmetry index at follow-up was higher in TRAINING compared to NoTRAINING (p = .015, Table 5).

Table 4. Psychological questionnaire scores Questionnaire Time Point Group TRAINING NoTRAINING ACL-RSI (points) Baseline 62.0 (18.1) 62.2 (23.9) Follow-up 79.3 (13.5) 77.1 (19.0) TSK-11 (points) Baseline 20.4 (4.4) 18.7 (4.4) Follow-up 17.3 (4.7) 17.0 (4.1) KASE (points) Baseline 81.1 (13.4) 83.7 (11.1) Follow-up 92.7 (6.2)* 89.1 (8.0) Fear intensity for the primary fear-evoking task or situation (points) Baseline 6.7 (2.1) 6.1 (2.3) Follow-up 4.7 (2.6) 4.0 (2.3)

Distribution of return to sport criteria met at baseline in (A) TRAINING and (B) NoTRAINING groups are not significantly different (p > .05)

Distribution of return to sport criteria met at follow up in (A) TRAINING and (B) NoTRAINING groups is significantly different (p < .05)

Table 5. Limb symmetry indices for each single-leg hop test in the hop test battery Group Time point TRAINING NoTRAINING Single hop (%) Baseline 92.5 (5.4) 91.1 (11.1) Follow-up 95.1 (5.1)a 89.0 (14.6) Triple hop (%) Baseline 93.5 (6.5) 88.8 (11.5) Follow-up 95.4 (3.6) 90.3 (9.9) Crossover triple hop (%) Baseline 95.3 (5.8) 90.6 (14.3) Follow-up 95.4 (4.3) 92.5 (8.2) Timed hop (%) Baseline 93.4 (8.6) 90.6 (13.9) Follow-up 95.6 (4.5) 92.2 (10.1)

Scores on the secondary psychological measures can be found in Table 2. KASE score at follow-up was significantly higher in TRAINING compared to NoTRAINING (p = .006). No significant group differences were found in follow-up TSK-11 score (p = .477) or fear intensity for the primary fear-evoking task or situation (p = .562). Post-hoc testing with both groups combined showed a significant decrease in TSK-11 score and fear intensity for the primary fear evoking task or situation from baseline to follow-up (p < .001).

4 DISCUSSION

This study compared readiness to return to sport participation between patients that did or did not complete advanced training during the return to sport phase of ACLR rehabilitation. Our hypothesis was not fully supported in that there were no group differences in overall psychological readiness for sport, kinesiophobia or passing rate on return to sport criteria at follow-up. However, those who completed advanced training showed higher self-efficacy for knee activities and better progression toward meeting return to sport criteria at follow-up, including a higher proportion meeting the hop test criterion and no patients failing to meet at least one criterion. Additionally, almost half (12/23) of the patients in NoTRAINING failed to show progression toward meeting the return to sport criteria over the 6 week period compared to less than 25% (5/23) in TRAINING. Together these findings indicate that advanced training after ACLR enhances readiness for sport participation by improving confidence and hop performance.

Both groups demonstrated higher ACL-RSI scores at follow-up than baseline, indicating improved overall psychological readiness for sport. This pattern is consistent with previous studies that have shown improvement in ACL-RSI scores with advanced training 16 and over time.30, 31 At follow-up, 70% of the patients in both groups met the ACL-RSI return to sport criterion. Moreover, mean ACL-RSI scores in both groups were comparable to uninjured athletes (80 points)39 and surpassed mean scores reported for patients who do not return to sports or experienced reinjury.40 Thus, mean ACL-RSI scores in both groups indicated acceptable psychological readiness for sport at follow-up, and improvement after advanced training may reflect a natural time course in the ACLR population, especially those that are motivated to return to sport.

Secondary psychological measures in this study provide additional insight into psychological changes that occur during advanced training. Self-efficacy for activities involving the knee was higher in TRAINING than NoTRAINING at follow-up. A previous study on plyometric training in an earlier timeframe after ACLR also found improved knee activity self-efficacy post-training.24 This contrasts to a study that showed incorporating four strategies to enhance self-efficacy during rehabilitation (understanding, maturity, persistence, and coping) did not result in better knee symptoms, quality of life or self-efficacy than control.41 Bandura42 defined self-efficacy as the judgement of a person's potential ability to carry out a task and suggested it stems from performance attainment, imitation and modeling, verbal and social persuasion and judgements of physiological states. The structure of the group-format advanced training program in the current study provided an opportunity to address each of these factors. As self-efficacy is an important factor for the rehabilitation outcome after ACLR,43 strategies to enhance low self-efficacy, such as advanced training programs, maybe necessary for satisfactory outcomes after ACLR.44 Although a statistically significant difference was seen in self-efficacy, it is unknown if this difference is clinically significant, as the minimum clinically important difference for the outcome is not known.

In contrast, advanced training did not show a benefit to kinesiophobia or individual situational fears. This agrees with findings of no change in TSK-11 scores after plyometric training administered at an earlier timeframe post ACLR,24 and persistent injury-related fears after advanced training.26 However, as both groups combined to show improvement over time, the advanced training exercises may not specifically target fear. Individuals who demonstrate higher kinesiophobia or individual fears may need a separate intervention. Further research is needed to determine alternative ACLR rehabilitation approaches to address injury-related fear. It is possible that a graded exposure treatment approach, which identifies and hierarchically exposes patients to their specific fear, might be needed to address injury-related fear as it has shown benefit for other fears and phobias.45

The goal of ACLR rehabilitation should be for patients to meet both physical and psychological return to sport criteria upon discharge to optimize the return to sport outcome.46 Psychological readiness for sport encompasses a variety of constructs, including fear of reinjury6 and self-efficacy.43 High fear of reinjury and low self-efficacy are not desirable as they can lead to poor function,47 but low fear of reinjury and high self-efficacy, coupled with unresolved knee impairments, may also be undesirable if it leads to premature sport participation that increases secondary injury risk. Clinicians should be aware of this “duality of psychological readiness”48 and its relationship to knee impairments and biological healing time frames to help patients make an appropriate risk appraisal about initiating sport participation. Also, patients who demonstrate inconsistency between physical and psychological readiness for sport may require additional supervised rehabilitation to safely return to sport. It appears that advanced training facilitates progress among physical, psychological or both criteria as 80% of patients in TRAINING showed improvement towards meeting the return to sport criteria compared to only 50% in NoTRAINING.

In both groups, baseline group hop test limb symmetry index scores were high, which is consistent with recommended clinical practice guidelines15 and reflected adequate preparation for advanced training. However, the proportion of patients meeting return to sport criteria at follow-up was only 52% in TRAINING and 43% in NoTRAINING, which failed to reach statistical significance. At face value, this result suggests the advanced training program did not serve its intended purpose of preparing patients with ACLR for sports participation. However, further analysis of the criterion met showed that all patients in TRAINING met at least one criterion, while 26% of the patients in NoTRAINING failed to meet at least one criterion. This result supports the advanced training program as facilitating better progress on readiness for return to the sport than no training. In addition, more patients in TRAINING than NoTRAINING met the hop testing criterion, which supports the specificity of exercises geared toward improved physical performance that may help reduce risk of reinjury.9

Advanced training is not a standard part of ACLR rehabilitation and often insurance benefits for rehabilitation expire before the knee is ready to partake in this type of high-impact exercise. The advanced training program described in this study is offered in a group format to be cost-effective, and can be undertaken when patients are appropriately ready. Even so, barriers reported by the NoTRAINING group for participating in advanced training included time constraints and cost. Although not an outcome of this study, we observed separation in Marx Activity Rating scores and higher variability (standard deviation) in NoTRAINING at follow-up that suggests activity level may be starting to fall for some patients who do not receive continued exercise supervision. Future research might explore methods of provid