Association of the Psychological Response to the ACL-SPORTS Training Program and Self-reported Function at 2 Years After Anterior Cruciate Ligament Reconstruction

Abstract

Background:

Psychological readiness to return to sport has emerged as an important factor associated with outcomes after anterior cruciate ligament reconstruction (ACLR). Psychological factors are potentially modifiable during the course of rehabilitation and improving them may lead to better outcomes.

Purpose:

To determine whether athletes with a positive psychological response after participation in a neuromuscular training and second injury prevention program had better self-reported function and activity outcomes compared to athletes who did not have a meaningful change.

Study Design:

Secondary analysis of a prospective clinical trial; level of evidence, 3

Methods:

Following ACLR and completion of formal rehabilitation, 66 level I/II athletes completed the following self-report measures at enrollment (pre-training): Anterior Cruciate Ligament-Return to Sport after Injury scale (ACL-RSI), International Knee Documentation Committee Subjective Knee Form (IKDC), and the five subscales of the Knee injury and Osteoarthritis Outcome Score (KOOS). Subjects completed these measures after ten sessions of agility, plyometric, and progressive strength training, and one and two years after ACLR. Subjects who displayed an increase in ACL-RSI score from pre-training to post-training that exceeded the MCID (≥10) were defined as having a positive psychological response (responder) to training while those who did not were defined as nonresponders. A mixed-model analysis of variance was used to determine if group differences in IKDC and KOOS scores existed over the four time points.

Results:

The responders reported better self-reported function compared to the nonresponders regardless of time (IKDC (p=.001), KOOS-Sport (p=.014), KOOS-Pain (p=.007), and KOOS-Symptoms (p=.002), but not for KOOS-QOL (p=.078)). Seventy-seven percent of responders and 67% of nonresponders had returned to their previous level of sport by 1 year after ACLR (p=.358). Eighty-two percent of responders and 78% of nonresponders had returned to their previous level of sport by 2 years from ACLR (p=.668).

Conclusion:

Fifty-nine percent of the athletes in this study displayed a meaningful improvement in psychological outlook over the course of the training program. Responders demonstrated persistently better self-reported function at post-training and one and two years after ACLR, but there were no between group differences in return to sport rates.

Clinical Relevance:

How patients respond psychologically to treatment programs incorporating progressive agility, plyometric, and strength training may affect subsequent function and activity.

Introduction

Upwards of 200,000 anterior cruciate ligament (ACL) ruptures occur each year in the United States.²⁴ Anterior cruciate ligament reconstruction (ACLR) is the gold standard treatment for athletes involved in cutting, jumping, and pivoting sports with the goal of returning the athlete to their prior level of sport. However, recent data suggests that less than two-thirds of athletes that undergo ACLR return to their prior level of sport despite good knee function.^7,9 Psychological factors, such as fear of reinjury, confidence, self-efficacy, and psychological readiness to return to sport have emerged as potential barriers for athletes attempting to return to sport.^{4,5,8,16,18,21,22} Webster et al developed the ACL-Return to Sport after Injury Scale (ACL-RSI) which measures psychological readiness to return to sport.²⁹ Psychological readiness to return to sport encompasses three psychological responses (emotions, confidence, and risk appraisal) specifically related to returning to sport. Cross sectional studies have found that athletes who were able to return to sport have higher ACL-RSI scores.^4,19 Prospective studies also suggest that higher ACL-RSI scores, indicating a more positive psychological outlook, are predictive of returning to sport and return to competition.^8,20

Transitioning back to sport can be especially fearful for athletes after ACLR.² This is also the time when athletes are weaned from consistent rehabilitation and focus is placed on an independent program. The ACL Specialized Post-Operative Return to Sport (ACL-SPORTS) secondary injury prevention program was designed to bridge the gap between the time when formal physical therapy traditionally ends and the time an athlete returns to sport.³⁰ While this program was not specifically designed to address psychological factors, such as fear of reinjury and confidence, it does apply the concepts of graded activity and graded exposure that have been investigated with other patient populations. For example, graded exposure is effective in reducing disability, pain, and catastrophizing in patients with chronic low back pain.²³ The ACL-SPORTS secondary injury prevention program therefore has the potential to improve psychological readiness to return to sport.

No published studies to our knowledge have investigated how psychological factors change over the course of a secondary injury prevention program. Furthermore, it is unknown how a change in psychological outlook during the time an athlete is transitioning back to sport is related to functional and activity related outcomes. Therefore, the purpose of this study was to 1) determine if athletes who demonstrate a an improvement in psychological readiness to return to sport have better self-reported function after the program, and 1 year and 2 years after ACLR than those who do not, 2) determine if athletes who demonstrate an improvement in psychological readiness to return to sport have better activity related outcomes 1 year and 2 years after ACLR than those who do not. We hypothesized that athletes with an improvement in psychological readiness to return to sport would demonstrate better outcomes immediately after the program and 1 year and 2 years after ACLR compared to athletes who did not demonstrate a meaningful positive response.

Methods

Participants

Sixty-six level I/II athletes between the ages of 13 and 55 (participated in cutting, pivoting, jumping sports for at least 50 hours per year)¹⁵ after primary ACLR were included in this secondary analysis of a prospective clinical trial. All athletes were part of a randomized controlled trial investigating the effect of a secondary injury prevention program with or without the addition of a specialized neuromuscular training program.³⁰ This clinical trial was approved by the institutional review board at our institution and all subjects completed informed consent. Athletes with a pre-training ACL-RSI score ≥ 91were excluded from this secondary analysis due to the mathematical impossibility of achieving a meaningful increase at least 10 (i.e., ceiling effect).

Following ACLR, all subjects had undergone post-operative rehabilitation and had to meet the following criteria prior to enrollment: at least 80% quadriceps strength index (measured isometrically on an isokinetic dynamometer), full knee range of motion compared to the uninjured side, minimal effusion present (trace or less)²⁸, ability to hop on one leg without pain, and had started a running progression. Subjects were excluded from enrollment in this trial if they had a previous ACL or other significant lower extremity injury, a concomitant grade III ligament injury, or an osteochondral defect ≥ 1cm². For more information regarding the enrollment process of this trial see Capin et al and Arundale et al.^10,11

Training Program

Full details of the protocol have been published.³⁰ All subjects in the trial underwent 10 sessions of progressive strengthening, plyometric training, and agility training. Proper form (encouraging greater knee flexion and reducing lower extremity valgus) was encouraged throughout the sessions during the plyometric and agility exercises. Plyometric and agility exercises were gradually progressed in quantity and complexity over the course of the 10 sessions. For example, single legged plyometric exercises were performed over ground during the first 3 sessions. Sessions 4–10 incorporated a hurdle that increased in height over the sessions. For agilities, linear movements were introduced first and then progressed to multidirectional movements with the athlete completing movements related specifically to their individual sport and utilizing a ball/equipment specific to their sport.

Self-Reported Measures

All subjects completed the ACL-RSI at enrollment (pre-training) and also after the training program (post-training).²⁹ This scale includes 12 questions and measures an athlete’s psychological readiness to return to sport, which encompasses emotions (including fear of reinjury), confidence, and risk appraisal. The ACL-RSI is scored on a scale from 0–100 with higher scores indicating a more positive psychological outlook in terms of returning to sport (i.e., less fear of reinjury, more confidence). Subjects in the present study were dichotomized into 2 groups based on their ACL-RSI score. Subjects who displayed an increase in ACL-RSI score from pre-training to post-training that equaled or exceeded 10 points were defined as having a positive psychological response (responder) to training while those who did not were defined as nonresponders. An increase of 10 or greater was chosen to define the groups based on face validity and known group validity due to the fact that there are no established minimal clinically important difference (MCID) scores for the ACL-RSI. Face validity was based on expert consensus from our research group who has extensive clinical and research experience in the ACL population. We determined an increase of 10 reflected a meaningful improvement based on a mean 1 point increase on each of the twelve questions on the ACL-RSI and that Langford et al examined’s study of ACL-RSI scores 3 and 6 months after ACLR found that the group of athletes that returned to competition at one year scored 9 points higher on the ACL-RSI and eleven points higher at 6 months.²⁰

In addition to the ACL-RSI, all subjects completed the International Knee Documentation Committee Subjective Knee Form (IKDC)¹⁵ and the 5 subscales of the Knee injury and Osteoarthritis Outcome Score (KOOS)²⁶ at pre-training, post-training, and one and two years after ACLR. The IKDC is a validated instrument that is used for patients with various knee conditions and includes questions about symptoms, sports and daily activities, as well as current knee function.¹⁷ The KOOS was designed to evaluate short-term and long-term outcomes in subjects after knee injury and knee osteoarthritis. It includes 5 subscales: 1) KOOS pain, 2) KOOS symptoms, 3) KOOS activities of daily living (KOOS-ADL), 4) KOOS Sport and Recreation, 5) KOOS quality of life (KOOS-QOL).²⁷ The KOOS-ADL was not included in this analysis due to a high ceiling effect in this population. Finally, subjects were asked at one year after ACLR if they have returned to their previous level of sport activity.

Statistical Analysis

All statistical analyses were performed using SPSS Version 24.0 (IBM Corp, Armonk, NY). A mixed model analysis of variance was performed for the IKDC and each subscale of the KOOS to determine the interaction of group (responder vs. nonresponder) by time (pre-training, post-training, 1 year, 2 years), and the main effects of group and time. Independent t-tests and chi-square were used to determine differences between groups in timing and demographics. The percentages of athletes who returned to their previous level of sport were reported by group. A p-value ≤ .05 was determined a priori to denote statistically significant differences between groups.

Results

Responder vs. Nonresponder Analyses

Thirty-nine athletes (59%) demonstrated an increase in ACL-RSI ≥ 10 and formed the responder group (pretraining score= 56.7 ± 19.9; post-training score=79.0 ± 17.7), while twenty-seven athletes had a change < 10 and formed the nonresponder group (pretraining score= 59.9 ± 17.7; post-training score=58.4 ± 18.2). There were no significant group differences in sex (p=.453), age (p=.478), body mass index (p=.060), weeks from surgery to pre-training (p=.469), weeks from surgery to post-training (p=.594), preinjury sport level (p=.130), or preinjury competitive level (p=.504)(Table 1).

Table 1:

Demographics and Timing between Responders and Nonresponders

	Responder (N=39)	Nonresponder (N=27)	p-value
Sex	21 women, 18 men	12 women, 15 men	0.453
Age (years)	20.7 ± 7.6	23.0 ± 8.8	0.478
BMI (kg/m2)	25.9 ± 7.8	26.1 ± 3.0	0.06
Weeks from surgery to pre-training	24.6 ± 8.1	22.0 ± 7.4	0.469
Weeks from surgery to post-training	31.9 ± 8.2	29.3 ± 7.8	0.594
ACL-RSI pre-training (mean ± SD)	56.7 ± 19.9	59.9 ± 13.2	0.497
ACL-RSI post-training (mean ± SD)	79.0 ± 17.7	58.4 ± 13.4	<0.001
Number of athletes that received perturbation training	21 (54%)	15 (55%)	0.891
Number of athletes by Competitive level			0.504
Intramural/Recreational	10	9
Club level	8	16
School Sponsored	9	13
Professional	0	1
Number of level I/II athletes	33/39 (85%)	26/27 (96%)	0.13

There was a significant main effect of group for IKDC, KOOS-Sport, KOOS-Pain, and KOOS-Symptoms (p≤.014), but not for KOOS-QOL (p=.078). The responder group demonstrated better self-reported function regardless of time. There was a significant group x time interaction for IKDC and KOOS-Quality of Life (Table 2). Both group’s IKDC scores improved significantly from pre-training to post-training (p<.001) and post-training to 1 year (p≤.002). The responders displayed significantly higher IKDC scores than the nonresponders at post-training (p=.001) and at 2 years (p=.001) but not at pre-training. For KOOS-QOL only the responders improved significantly from pre to post-training (p<.001), post-training to 1 year (p<.001) and from 1 year to 2 years (p=.010). Nonresponders improved significantly only from post-training to 1 year (p<.001). The responders displayed significantly higher KOOS-QOL scores at 2 years (p=.015). One year after ACLR there was no significant group difference in the percentage of athletes who returned to their previous levels of sport (responders: 30/39, 77%; nonresponders: 18/27, 67%, p=.358). At 2 years there was no significant difference in those who returned to their previous levels of sport (responder: 32/39, 82%; nonresponders: 21/27, 78%, p=.668).

Table 2:

Self-Reported Functional measures at pre-training, post-training, and 1 year between Responders and Nonresponders.

Self-reported Measure	Group	Pre-Training	Post-Training	1 Year	2 Year	ME Time	ME Group	Group × Time Int
IKDC	Responder	78.9 ± 8.6	89.0 ± 7.5	94.2 ± 7.9	97.1 ± 7.4	<.001	0.001	0.031*
	Non-responder	76.8 ± 8.1	81.9 ± 9.0	91.0 ± 9.4	87.6 ± 14.9
KOOS-Sport	Responder	78.1 ± 15.4	91.7 ± 10.7	95.0 ± 8.4	94.4 ± 13.1	<.001	0.014	0.117
	Non-responder	77.6 ± 13.8	82.2 ± 13.1	91.0 ± 9.8	85.7 ± 22.2
KOOS-QOL	Responder	54.8 ± 15.7	70.0 ± 21.7	82.7 ± 15.3	88.8 ± 15.8	<.001	0.078	0.020*
	Non-responder	57.9 ± 13.2	62.7 ± 13.4	75.7 ± 15.1	78.2 ± 18.2
KOOS-Symptoms	Responder	86.4 ± 9.7	88.5 ± 10.9	91.9 ± 8.7	93.3 ± 8.3	<.001	0.002	0.373
	Non-responder	81.2 ± 10.3	81.5 ± 11.4	87.8 ±11.3	84.7 ± 13.5
KOOS-Pain	Responder	91.6 ± 6.4	94.7 ± 5.4	97.0 ± 3.9	98.3 ± 4.8	<.001	0.007	0.248
	Non-responder	89.5 ± 8.2	92.2 ± 7.6	94.6 ± 5.1	93.0 ± 8.3

Discussion

The purpose of this study was to determine if athletes who demonstrated a positive psychological response after participation in the program had better self-reported function and activity related outcomes than those who did not. Our secondary purpose was to determine if athletes who demonstrate an improvement in psychological readiness to return to sport have better activity related outcomes 1 year and 2 years after ACLR than those who do not. Our first hypothesis was supported. The responder group demonstrated better self-reported function (IKDC and KOOS-QOL) immediately after participating in the training that persisted 1 and two years after ACLR when compared to the nonresponders. Our second hypothesis was not supported, however, the responder group had 10% more athletes return to sport level by 1 year or 2 years after ACLR. It is possible that this study was under powered to detect difference in return to sport level. Findings from this study offer important considerations for future research. Examining the effectiveness of a supervised neuromuscular training versus no intervention, and/or an intervention focused purely on targeting psychological factors will help clarify the psychological benefits of this program.

A key finding of this study is that the responder group demonstrated persistent better self-reported function up to 2 years after ACLR compared to the nonresponder group. The responder group’s mean IKDC, KOOS-QOL, and KOOS-Sport and Recreation scores were approximately 10 points higher than the nonresponder group at 2 years. These data suggest that those who improve psychological outlooks throughout the course of rehabilitation may report better self-reported function compared to those who do not improve their psychologic outlook. It is important to note that this analysis does not allow us to determine the direction of the relationship between psychological outlook and self-reported function. Perhaps our responder group’s improvement in self-reported function from pre to post-training led to an improved psychological outlook. There were group by time interactions for the IKDC. This scale encompasses questions regarding symptoms, function, and sport activities. Both the responder group and the nonresponder group improved on the IKDC from pre-training to post-training, and from post-training to 1 year. However, the responder group demonstrated significantly higher IKDC scores at post-training compared to the nonresponder group. With a score of 89.0% on the IKDC at post-training, the responder group almost achieved 90%, our clinical cutoff for return-to-sport. The nonresponder group mean IKDC score was well below 90% (81.9%) at post-training and did not achieve greater than 90% until 1 year after ACLR. Thus, it took the nonresponder group longer to achieve adequate self-reported knee function (i.e. ≥90%). While there was not a significant group x time interaction for the KOOS-Sport and Recreation, we see the same pattern of improvement between groups. The responder group’s mean score surpassed 90% (92%) while the nonresponder group’s mean score was well below 90% (82%) at post-training. At 1 year, the nonresponder group caught up to the responder group, achieving a KOOS-Sport and Recreation mean score greater than 90% (91%). The pattern of change scores for the IKDC and KOOS-Sport and Recreation has both beneficial and potentially deleterious implications for the responder group. From a positive perspective, a 90% or better score on a self-reported functional measures is often used as one criteria needed to allow an athlete to return to sport.^1,14 The responders approached 90% on the IKDC (89.0%) and exceeded 90% (91.7%) on the KOOS-Sport and Recreation at the post-training timepoint. From a negative perspective, higher self-reported function and a more positive psychological outlook for return to sport could lead to a sooner return to sport and perhaps increase second ACL injury risk.

The relationship between self-reported function and different psychological constructs has been well established.^3,6,25 However, the direction of this relationship is still not clear. In the current study both groups demonstrated improvements in self-reported function (IKDC, KOOS-Sport) from pre to post-training with the responder group demonstrating higher self-reported function at post-training. It is plausible that a greater improvement in self-reported function led to a greater improvement in psychological outlook. It is also plausible that the IKDC and KOOS-Sport are influenced by psychological outlook. Future research should elucidate the relationship between psychological outlook and self-reported function.

Group differences were not present in the number of athletes that returned to sport level at 1 or 2 years after ACLR. This indicates that many of the nonresponders had returned to sport level despite reporting deficits in function, sport specific activities, and knee related quality of life 2 years after ACLR. Though not statistically significant, however, 10% less athletes in the nonresponder group returned to theirs pre-injury sport level compared to the responder group. Additionally, a meta-analysis of sixty-nine studies found that only 65% of athletes returned to their pre-injury sport level.⁷ A higher percentage of athletes in the current study’s responder group returned to their pre-injury sport level (77%), while the nonresponder group’s percentage (67%) was consistent with the meta-analysis.

On average, the athletes in our cohort displayed similar ACL-RSI scores to other studies around this time frame. Langford et al examined psychological readiness to return to sports at 6 months post-ACLR.²⁰ Athletes that returned to competitive sport by 12 months had mean ACL-RSI scores of 63, while athletes that did not return to competition had a mean of 52. Ardern et al examined ACL-RSI scores at 4 months after ACLR finding that athletes who returned to sport by 1 year had a mean of 57, while athletes who did not return to sport had a mean of 40.⁸ Using a receiver operating curve analysis, the authors found a cutoff score of 56 at 4 months after ACLR was best at discriminating the athletes that are able to return to sport versus those that are not able. In the present study, the responders displayed very similar scores (mean of 57) to the nonresponders (mean of 60) at pre-training and both group’s scores were around the cut off score determined by Ardern et al. Our entire cohort therefore displayed psychological readiness scores that could potentially benefit from improvement. In fact, there were nineteen athletes in the current study will ACL-RSI scores of 70 or greater at pretraining. One may think that these athletes, with already high ACL-RSI scores, may not have room for improvement. However, eleven athletes with pretraining ACL-RSI scores of 70 or greater displayed at least a 10 point increase from pretraining to post training (responder group), while eight athletes with scores of 70 or greater did not display a 10 point increase (nonresponder group). This finding suggests that even athletes with scores of 70 or more on the ACL-RSI can still continue to improve their psychological outlook.

To our knowledge, this is the first study to examine if changes in psychological outlook after completing a secondary injury prevention program are associated with improved self-reported function after ACLR and with continued better function. However, the relationship between psychological factors and functional outcomes has been investigated cros-sectionally.^12,13 A recent clinical review indicates that self-reported function and fear of reinjury are associated during the late stages of rehabilitation.¹⁶ Inconsistent findings have been found during the earlier stages of rehabilitation, when physical impairments may contribute more to functional deficits.²² Future research should continue to evaluate how specific interventions affect psychological outlook given the association between psychological factors and outcomes after ACLR.

There are limitations to this study. First, there are currently no established minimal detectable change (MDC) or minimally clinically important difference (MCID) scores established for the ACL-RSI on the original 0–100 scale. Only one study examined test-retest reliability and calculated minimal detectable change scores.¹⁹ This study adopted a 0–10 scale as opposed to the original 0–100 scale and is therefore inappropriate to use. We therefore dichotomized our groups based on known groups validity and face validity. Future research should attempt to calculate true MDC and also MCID scores for the ACL-RSI. Second, based on the study’s design and lack of control group we are not able to ascertain whether having an improvement in psychological readiness led to better self-reported function or if an improvement in function led to psychological readiness scores. Future research should examine interventions directly related to improving psychological readiness to return to sport.

Conclusions

In conclusion, 59% of the athletes in this study demonstrated a positive psychological response by our definition to a neuromuscular training and secondary ACL injury prevention program. These athletes categorized as responders demonstrated better self-reported function immediately following the program and at 1 and 2 years after ACLR when compared to the nonresponders. Significant group differences in return to sport level were not found. Our findings suggest that those who improve psychological outlook over the course of rehabilitation may report increased function compared to those who do not, suggesting interventions targeting psychological factors may be warranted to improve rehabilitation outcomes.

What is known about the subject:

Psychological readiness is associated with activity related outcomes following ACLR.

What this study adds to existing knowledge:

Few studies have explored the relationship between psychological factors and outcomes (self-report and activity) during the time when athletes transition back to sport. Additionally, no studies to date have examined whether an improvement in psychological outlook during a specialized neuromuscular training and secondary injury prevention program is associated with better outcomes post-training, and at one and two years after ACLR.

Abbreviations:

IKDC

International Knee Documentation Committee Subjective Knee Form

KOOS

Knee injury and Osteoarthritis Outcome Score

QOL

quality of life

ME

main effect

Int

interaction