Abstract
Introduction:
Anterior cruciate ligament (ACL) revision patients report lower outcome scores on validated knee questionnaires post-operatively compared to primary ACL cohorts. In a previously active population, it is unclear if patient-reported outcomes (PROs) are associated with returning to activity (RTA) or vary by sport-participation level (high-level, recreational athletes).
Hypotheses:
Individual RTA will be associated with improved outcomes (i.e., decreased knee symptoms, pain, function) as measured through validated PROs. Recreational participants will report lower PROs compared to high-level and be less likely to RTA.
Study Design:
Prospective cohort study, Level 2
Methods:
There were 862 patients who underwent a revision ACL reconstruction (rACLR) and self-reported physical activity at any level pre-operatively. Those who did not RTA (nRTA) reported no activity two years post-revision. Baseline data included demographics, surgical history and characteristics, and PROs: International Knee Documentation Committee (IKDC) questionnaire, Marx activity score, Knee Outcomes and Osteoarthritis Score (KOOS) and the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC). A binary indicator identified patients with same/better PROs versus worse outcomes compared to baseline, quantifying the magnitude of change in each direction, respectively. Multivariable regression models were used to evaluate risk factors for nRTA, the association of two-year PROS following revision surgery by RTA, and interaction with activity level.
Results:
At two years post-operatively, approximately 15% did nRTA, with current smokers (aOR=3.3, P=0.001), females (aOR=2.9, P<0.001), recreational participants (aOR=2.0, P=0.02), and previous medial meniscal excision (aOR=1.9; P=0.013), having higher odds of not returning. Non-returners were significantly associated having worse PROs at two years across multivariate models. No differences in PROs at two years were seen between participation level.
Conclusion:
Recreational-level participants were twice as likely to nRTA compared to those participating at higher levels. Among a previously active cohort, nRTA was a significant predictor of lower PROs after rACLR, yet approximately 20% of ACL revision patients who did RTA reported lower outcome scores. Most rACLR patients who are active at baseline improved over time; however, patients who report worse outcomes at two years have a clinically meaningful decline across all PROs.
Keywords: revision anterior cruciate ligament, return to activity, recreational athletes
Introduction
A return to one’s pre-injury activity level is often a primary goal following recovery from a primary (ACLR) or revision anterior cruciate ligament reconstruction (rACLR) for active individuals.8 Surgical technique and post-surgical rehabilitation are similar between primary and rACLRs, but the literature consistently reports that patients with rACLR experience higher graft failure rates, lower patient-reported outcomes (PROs), and lower rates of returning to activity or sport.2–7, 11, 12, 17 Given the limited quality of evidence available from retrospective studies in conjunction with the complexity of rACLR injuries, it is difficult to determine the reason(s) for this discrepancy.8
In 2006, a prospective cohort, the Multicenter ACL Revision Study (MARS) was designed to understand the indicators of worse PROs following revision surgery.20 Published results consistently show improved PROs from baseline along with decreased activity levels, using the entire cohort to inform the average improvement over time. We sought to better isolate the magnitude of change, separately, for patients who report improvement versus decline in self-reported pain and function two years after rACLR. Known predictors of lower PROs scores include primary reconstruction graft choice, previous meniscal repair and articular cartilage injury.12–20 In our previous study, rACLR patients who reported sport participation at two years had significantly higher PROs compared those with no sport participation.18 However, the complete cohort was analyzed for this prior study, including patients who were not active at baseline. In an effort to better understand the clinical prognosis for active populations who may have goals to return to sport or physical activity, this study provides a focused sub-analysis of the MARS population who was active at baseline. Additionally, no previous studies have sought to analyze the effect of activity participation level (i.e., recreational, high school, college-level) on PROs in a generally active rACLR cohort.
Therefore, the purpose of this study was to determine: (i) the difference in injury profiles of previously active MARS patients who RTA and did nRTA at two years; (ii) to what degree nRTA is associated with two-year PROs; and 3) if this varies by activity level (recreational activity vs high-level activity).
Methods
Cohort Design
The MARS group is a collaboration of 83 sports-medicine fellowship-trained surgeons who represent an approximately 50:50 mix of practitioners from private and academic sites (N = 52 sites). After specific site institutional review board approval, study enrollment occurred from 2006–2011, of which 1205 patients undergoing a rACLR consented to partake. Failure of the previous ACLR was determined through either an arthroscopic surgery, orthopedic clinical examination, or magnetic resonance imaging that has been previously described elsewhere.12 Exclusion criteria included patients who presented with prior knee infection, multi-ligament reconstruction, complex regional pain syndrome or arthrofibrosis.
Data Sources and Measurement
Once informed consent was obtained, all patients completed a questionnaire to collect demographic information, activity participation, injury mechanism, comorbidities, and knee injury history.13, 15 A series of previously-validated PROs were completed by each patient that measured general and knee-specific outcomes at the time of revision: International Knee Documentation Committee (IKDC) questionnaire; Marx Activity Rating scale; Knee injury and Osteoarthritis Outcome Score (KOOS); and the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC), which was calculated from the KOOS questionnaire.14, 24, 25 Additionally, surgeons completed a questionnaire including physical examination findings, surgical procedures and implants, arthroscopic findings, and the management of any current meniscal or chondral damage to the injured knee.12
Completed data forms were mailed to the data coordinating center by each participating site. Data were abstracted from both the patient and surgeon’s questionnaire through TeleForm software (OpenText; Waterloo, Canada) using optical character recognition. Abstracted data were verified and transferred to a master database. Multiple quality control checks were performed prior to data analyses.
Patient Follow-up
Patients completed the same questionnaire pre-operatively (i.e., time of rACLR surgery) and at their two-year follow-up. Questionnaires were returned via mail and, additionally, study participants were contacted by phone to determine if any successive surgeries were performed on either knee since their initial ACLR. Operative reports were obtained when possible to verify subsequent injury and treatment.
Sub-analysis of a cohort that was physically active prior to rACLR
Given that some of the cohort was not active at baseline, only patients who reported being physically active in the two year’s prior to their rACLR on the baseline questionnaire were included in this analysis. Baseline activity levels were determined from the following questions: “What sport have you participated in the most over the last two years?” and “What second sport have you participated in the most over the last two years?” from the patient questionnaire administered.18 All patients who did not report playing either a primary or secondary sport at baseline, by selecting “none” on the questionnaire were excluded (N=133).
Self-Reported RTA & Activity Levels
Patients who reported taking part in activities at the time of their rACLR indicated participation in the following activities: basketball, baseball/softball, football, gymnastics, skiing, soccer, volleyball and “other”. Reported activities in the “other” category” included: cycling, cheerleading, dancing, frisbee, hockey, lacrosse, martial arts, roller skating, rugby, tennis, track and field, and trampolining.20 Participants who self-reported no primary or secondary activity on the two-year follow-up questionnaires, and additionally left the “other” indicator field blank, were classified as nRTA.
Two-year sport participation was defined as the highest level reported from the question: “What is the highest or most advanced level you achieved in that sport in the last two years?” Options included: none, recreational, amateur (team or club), high school, NCAA DI/non DI, semi-professional, professional. Recreational patients were defined as patients who checked that box, while those indicating a higher level of sport or activity participation above recreational were considered high-level patients.
Statistical Analysis
Continuous variables were summarized as percentiles (i.e., 25th, 50th, and 75th) with categorical variables as frequencies and percentages to describe characteristics of the study sample. Multivariable logistic regression analysis was used to determine risk factors that increased the odds of not returning to activity in the two years following revision ACLR surgery. Separate multivariable regression analyses were used to examine the association between each two-year outcome subscale score (KOOS subscales of Symptoms, Pain, Sports/Recreation, Activities of Daily Living [ADLs], and Quality of Life [QoL], WOMAC subscales [Stiffness, Pain, ADLs], IKDC and MARX) and RTA status. Adjusted odds ratios (aOR) with 95% confidence intervals (CI) are reported.
Initial analyses of the data showed that some patients did not improve from their baseline measures to the two-year mark. While prior work reports average change across the cohort from baseline to two year post-operative follow-up, we sought to better understand the magnitude of change, separately, for patients who reported improved compared to worse outcomes. To qualify improvement versus decline in PROs, a binary “change score” variable was created for each PRO instrument. Patients were classified as “improved/same” if their absolute two-year PRO scores were better or unchanged as compared to baseline while those who decreased were classified as having a “worse outcome.” For each analysis, baseline: age (years), sex (female/male), two year and baseline activity participation level (high-level/recreational), educational level (1–20), smoking status (never, former, current), body mass index (BMI), revision number (1, ≥ 2), time from most recent primary ACL surgery (years), self-reported mechanism of injury [traumatic vs non-traumatic], PRO improvement status [improved-same/decreased), previous injury characteristics: graft type [allograft, autograft, both and other], MCL injury (yes/no), articular cartilage injury (yes/no), medial & lateral meniscal repair (none, excision, repair [stable] & repair [unstable]) and current injury characteristics (graft choice, meniscal injury (medial & lateral), and articular cartilage injuries (medial femoral condyle, lateral femoral condyle, medial tibia plateau, lateral tibia plateau, trochlea, patella [all yes/no)) were controlled for in the model. Each continuous variable was tested for a nonlinear relationship and linearity was confirmed (P < 0.05). Categorical variables were fitted per their degrees of freedom (n-1). Given the restricted cohort, some previous reported categorical variables with low counts (less than five) were reduced to improve the stability of the models from previous MARS analyses. Lastly, to determine if the relationship between each PRO and nRTA differed by participation level, an interaction term was included in each model. If significant (P < 0.05) the specific PRO model was stratified by participation level (i.e., high-level and recreational).
Previous reports have identified minimal clinically important differences (MCIDs) in the PROMS used: eleven points for the IKDC,9 eight to ten points for the KOOS22 and WOMAC,24, 25 as well as two points for the MARX activity scale.26 All statistical analyses were performed in STATA 15 (StataCorp LLC, College Station, TX) and R (version 3.6.1).
Results
Study Population
Approximately 72% (N=862/1205) of the original MARS cohort was included in the sub-analysis. Eleven percent of excluded patients were not “active” at baseline (133/1205) and an additional 17% (210/1205) were lost to follow-up. As expected, patients in the “active” group sub-analysis were slightly younger (27.3 vs 28.0 years old) with a lower BMI (25.5 vs 26.1 kg/m2) and had higher MARX’s scores at baseline (MARX: 10.5 vs 9.5) when compared to the original MARS cohort. Additionally, this sub-analysis of “active” patients had a more equal ratio of males to females (original: 58:42; current: 55:45) with no significant differences in primary ACL injury characteristics. However, differences in current injury (i.e., pre-ACL revision) characteristics included fewer patients reporting a non-traumatic ACL graft injury (original: 69:31; current: 65:35), with a lower proportion of medial femoral condyle (41.0 vs 43.5%), medial tibial plateau (9.3 vs 10.9%), and trochlea (18.9 vs 20.5%) articular cartilage injuries.
In our sub-analysis of “active” patients at baseline, 15.2% (131/862) of those who underwent revision ACL surgery reported no activity participation at two years following surgery. Descriptive characteristics of those who did and did not report RTA are reported in Table 1. The sub-cohort was on average 27.3 years old, comprised of 55% males, 42% recreational patients with an average time of 5.6 years from their previous ACL reconstruction.
Table 1:
Demographic, previous surgical and current injury characteristics of the cohort by RTA statusa (N=862)
| Returned to activity n=731 | Did not return n=131 | ||||
|---|---|---|---|---|---|
|
|
|||||
| n | (%) | n | (%) | p-value | |
|
| |||||
| Demographics | |||||
| Age (years); (Mean ± SD) | 27.1 (10.1) | 28.8 (9.8) | 0.06 | ||
| BMI (kg/m2); (Mean ± SD) | 25.2 (3.8) | 26.8 (5.1) | 0.0001 | ||
| Education level (0–20); (Mean ± SD) | 14.5 (2.9) | 14.6 (2.9) | 0.64 | ||
| Baseline MARX score (0–16); (Mean ± SD) | 10.9 (5.3) | 7.8 (5.6) | <0.001 | ||
| Gender | <0.001 | ||||
| Female | 303 | (41.5) | 81 | (61.8) | |
| Male | 428 | (58.5) | 50 | (38.2) | |
| Ethnicity | 0.345 | ||||
| White | 621 | (85.0) | 116 | (88.5) | |
| Other, non-white | 110 | (15.0) | 15 | (11.5) | |
| Smoking status | <0.001 | ||||
| Never | 605 | (82.8) | 91 | (69.5) | |
| Current | 42 | (5.8) | 22 | (16.8) | |
| Quit | 75 | (10.4) | 18 | (13.7) | |
| NA | 9 | (1.0) | 0 | (0) | |
| Baseline level of participation | <0.0001 | ||||
| High-level patient | 447 | (61.1) | 47 | (35.9) | |
| Recreational patient | 282 | (38.6) | 84 | (64.1) | |
| NA | 2 | (0.3) | 0 | (0) | |
| Baseline activity participation | <0.0001 | ||||
| Multi-activity | 595 | (81.4) | 86 | (65.6) | |
| Single activity | 134 | (18.3) | 45 | (34.4) | |
| NA | 2 | (0.3) | 0 | (0) | |
| Primary activity at baseline | 0.011 | ||||
| Baseball/Softball | 46 | (6.4) | 13 | (9.9) | |
| Basketball | 121 | (16.5) | 19 | (14.5) | |
| Football | 74 | (10.1) | 8 | (6.1) | |
| Gymnastics | 12 | (1.6) | 1 | (.8) | |
| Skiing | 60 | (8.2) | 6 | (4.6) | |
| Soccer | 139 | (19.0) | 18 | (13.7) | |
| Volleyball | 35 | (4.8) | 16 | (12.2) | |
| Other | 244 | (33.4) | 50 | (38.2) | |
| Previous Injury Characteristics | |||||
| Previous Graft Choice | 0.227 | ||||
| Allograft | 160 | (21.8) | 30 | (22.9) | |
| Autograft | 504 | (69.0) | 86 | (65.6) | |
| Other/Unknown graft type | 67 | (9.2) | 14 | (10.7) | |
| NA | 0 | (0) | 1 | (0.8) | |
| Medial Meniscal Repair | <0.001 | ||||
| None | 482 | (65.9) | 66 | (50.4) | |
| Excision | 185 | (25.3) | 60 | (45.8) | |
| Repair (Stable/Unstable) | 64 | (8.8) | 4 | (3.0) | |
| NA | 0 | (0) | 1 | (0.8) | |
| Lateral Meniscal Repair | 0.259 | ||||
| No | 593 | (81.1) | 107 | (81.7) | |
| Excision | 106 | (14.5) | 19 | (14.5) | |
| Repair (Stable/Unstable) | 32 | (4.4) | 4 | (3.0) | |
| NA | 0 | (0) | 1 | (0.8) | |
| Previous Articular Cartilage Repair | 0.409 | ||||
| No | 654 | (89.5) | 114 | (87.0) | |
| Yes | 77 | (10.5) | 17 | (13.0) | |
| Current Injury Characteristics | |||||
| Time since last surgery (years); (Mean ± SD) | 5.5 (5.6) | 6.1 (5.4) | 0.278 | ||
| Mechanism of Injury | <0.001 | ||||
| Non-traumatic | 203 | (27.8) | 60 | (45.8) | |
| Traumatic | 527 | (72.1) | 70 | (53.4) | |
| NA | 1 | (0.1) | 1 | (0.8) | |
| Number of Previous ACL Reconstructions | 0.012 | ||||
| One | 662 | (90.6) | 109 | (83.2) | |
| Two or more | 69 | (9.4) | 22 | (16.8) | |
| Current Graft Choice | 0.078 | ||||
| Allograft | 342 | (46.7) | 71 | (54.2) | |
| Autograft | 363 | (49.7) | 56 | (42.8) | |
| Other/unknown graft type | 26 | (3.6) | 3 | (2.2) | |
| NA | 0 | (0) | 1 | (0.8) | |
| Graft Change from Previous ACLR | 0.163 | ||||
| Same Graft Type & Graft Source | 142 | (19.4) | 18 | (13.8) | |
| Same Graft Type, Different Graft Source | 171 | (23.4) | 35 | (26.7) | |
| Same Graft Source, Different Graft Type | 206 | (28.2) | 37 | (28.2) | |
| Different Graft Type & Graft Source | 212 | (29.0) | 40 | (30.5) | |
| NA | 0 | (0) | 1 | (0.8) | |
| Medial Meniscus Injury | 0.374 | ||||
| No | 399 | (54.6) | 77 | (58.8) | |
| Yes | 332 | (45.4) | 54 | (41.2) | |
| Lateral Meniscus Injury | 0.091 | ||||
| No | 457 | (62.5) | 92 | (70.2) | |
| Yes | 274 | (37.5) | 39 | (29.8) | |
| Medial Collateral Ligament Injury | 0.787 | ||||
| No | 680 | (93.0) | 121 | (92.4) | |
| Yes | 51 | (7.0) | 10 | (7.6) | |
| # of Articular Cartilage Injuries (0–9) | 0.026 | ||||
| None | 237 | (32.4) | 27 | (20.6) | |
| One | 204 | (27.9) | 44 | (33.6) | |
| ≥ Two | 290 | (39.7) | 60 | (45.8) | |
Excluded patients who did not play any sports at baseline; ACLR = Anterior cruciate ligament reconstruction, NA = Missing data
Multiple differences were noted between patients based on their RTA status. Patients who did nRTA had a significantly higher BMI (difference = 1.6 kg/m2; P < 0.001) and reported a lower MARX score by three points (P < 0.001). Also, approximately 20% more females reported not returning to activity (P < 0.001) and these patients were slightly older (28.8 ± 9.8 years compared to 27.1 ± 10.1 years, P = 0.06). Of those who did not return, 16.8% were current smokers compared with 5.8% of those returning (P < 0.001). Recreationally active patients represented 64.1% of those who did not return and 39.0% of those who did RTA (P < 0.001). Undergoing a previous medial meniscus excision represented 45.8% of those who did not return compared to 25.3% of those who returned to activity (P < 0.001). Among current injury characteristics, no difference was seen in the time since their last surgery (P = 0.278) and those who did not return had a significantly higher proportion of patients who self-reported a non-traumatic ACL graft re-injury mechanism (P < 0.001).
Two-Year PRO Scores & RTA Status
On average, PROs scores in both groups improved from baseline to the two-year mark following revision surgery (Figure 1).
Figure 1:
Percent of Patients Returning to Activity & Two-Year Patient Improvement Status on PROs (N=862)
ADL = Activities of Daily Living; IKDC = International Knee Documentation Committee; KOOS = Knee Injury and Osteoarthritis Outcome Score; PRO = Patient Reported Outcome Measures; QoL = Quality of Life; RTA-N = Self-Reported RTA (No); RTA-Y= Self-Reported RTA (Yes); Sport/Rec = Sports & Recreation; WOMAC = Western Ontario and McMaster Universities Osteoarthritis Index
However, not all patients who RTA improved on their PROs at the two-year mark. Across PROs except the MARX, 13.5 to 27.0% of patients reported decreased PROs at two years. On the KOOS, WOMAC and IKDC scales, 11.2 to 25.6% of patients who RTA reported a decrease in PRO while 22.1 to 35.3% of those who did nRTA reported a decrease in PROs (Table 2).
Table 2:
Median Differences in Two-Year PROs Scores by RTA Status and Patient Improvement Status (IQR)
| KOOS (−100–100) | WOMAC (−100–100) | IKDC | MARX | |||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Symptoms | Pain | Sport/Rec | ADL | QoL | Stiffness | Pain | ADL | (−100–100) | (−16–16) | |||||||||||
|
|
||||||||||||||||||||
| Overall | ||||||||||||||||||||
| 7 | (−4, 21) | 8 | (0, 22) | 25 | (5, 45) | 6 | (0, 18) | 19 | (6, 38) | 0 | (0, 25) | 5 | (0, 15) | 6 | (0, 18) | 20 | (8, 33) | −3 | (−7, 0) | |
| RTA-Y | 11 | (−4, 21) | 8 | (0, 22) | 25 | (5, 45) | 6 | (0, 18) | 25 | (6, 44) | 0 | (0, 25) | 5 | (0, 15) | 6 | (0, 18) | 22 | (9, 36) | −2 | (−6, 0) |
| RTA-N | 4 | (−4, 14) | 6 | (−6, 19) | 20 | (−5, 40) | 6 | (0, 19) | 12 | (0, 31) | 0 | (−13, 25) | 5 | (−5, 20) | 6 | (0, 19) | 10 | (−1, 25) | −4 | (−9, 0) |
| Improved/Same | ||||||||||||||||||||
| 14 | (7, 29) | 14 | (6, 25) | 30 | (15, 48) | 9 | (2, 22) | 25 | (13, 44) | 13 | (0, 25) | 10 | (0, 20) | 9 | (2, 22) | 24 | (13, 37) | 1 | (0, 4) | |
| RTA-Y | 14 | (7, 29) | 14 | (6, 25) | 30 | (15, 50) | 9 | (2, 22) | 32 | (13, 44) | 13 | (0, 25) | 10 | (0, 20) | 9 | (2, 22) | 24 | (14, 37) | 1 | (0, 4) |
| RTA-N | 11 | (4, 21) | 14 | (6, 25) | 30 | (15, 45) | 10 | (3, 22) | 19 | (6, 38) | 13 | (0, 25) | 10 | (5, 20) | 10 | (2, 22) | 19 | (8, 30) | 1 | (0, 3) |
| Decreased | ||||||||||||||||||||
| −11 | (21, −7) | −8 | (−14, −3) | −15 | (−25, −5) | −5 | (−13, −2) | −13 | (−22, −6) | −19 | (−25, −13) | −10 | (15, −5) | −5 | (−13, −2) | −8 | (−16, −4) | −5 | (−9, −3) | |
| RTA-Y | −11 | (−7, −4) | −8 | (14, −3) | −15 | (−30, 5) | −4 | (−10, −1) | −13 | (−25, −6) | −13 | (−25, −13) | −10 | (−15, −5) | −4 | (−10, −2) | −8 | (−16, −5) | −5 | (−8, −3) |
| RTA-N | −11 | (−21, −4) | −8 | (−11, −6) | −15 | (−25, −5) | −9 | (−19, −3) | −13 | (−19, −13) | −25 | (−38, −13) | −14 | (−15, −10) | −9 | (−19, −3) | −10 | (−16, −4) | −8 | (11, −4) |
Improved/Same = 2 Year PRO score increased or stayed the same from baseline (i.e., a positive median difference indicates that the 2-yr outcome score was better/improved compared to the baseline score), Decreased = 2 Year PRO score decreased from baseline (i.e., A negative median difference indicates that the 2-year outcome score was worse compared to the baseline score)
Minimal clinically important differences (MCIDs) in the PROMS used: eleven points for the IKDC,9 eight to ten points for the KOOS22 and WOMAC,24, 25 and two points for the MARX activity scale.26
ADL = Activities of Daily Living; IKDC = International Knee Documentation Committee; IQR = Interquartile range; KOOS = Knee Injury and Osteoarthritis Outcome Score; PRO = Patient Reported Outcome Measures; QoL = Quality of Life; RTA-N = Self-Reported RTA (No); RTA-Y = Self-Reported RTA (Yes); Sport/Rec = Sports & Recreation; WOMAC = Western Ontario and McMaster Universities Osteoarthritis Index
While the overall differences in scores between those who RTA and do not demonstrate worse outcomes overall, stratifying the cohort by improvement versus decline yields differences beyond MCID thresholds across all PROs. Significant differences in the proportions of those who did not improve were found between groups on the IKDC and KOOS Pain, Sports and Recreation, and Quality of Life subscales. On the MARX score, a significant difference between groups found 40% of those who returned to activity reported an increase or maintained their pre-injury activity level compared with 32% of those not returning (P = 0.017).
An individual’s RTA is significantly associated with 2-year outcome scores across all PRO tools, even after controlling for patients who reported worse PRO outcomes compared to baseline (Supplemental Tables 1–4). Not RTA is associated with lower PRO scores across all questionnaires; however, only the decline in MARX score yielded a MCID in the fully adjusted models (−3.2, 95% CI −4.4, −2.1). After adjusting for RTA status, the magnitude of change for patients who report worse outcomes compared to baseline is clinically significant. A patient with a worse outcome reported has, on average, a 28 point lower IKDC score, more than 21 to 37 points lower on each KOOS subscale, 16 to 31 points lower on WOMAC subscales, and six to seven points lower on MARX for high-level versus recreational athletes, respectively (all p<0.001). The combination of nRTA among patients reporting worse PRO scores reduces predicted 2-year post-operative function and pain by an additional three-to-six points.
Predictors of Not Returning to Activity
Those who did nRTA had significantly higher odds of having increased BMI, being female, participating in a recreational sport, and smoking (Figure 2).
Figure 2:
Significant Predicators of Not Returning to Activity in Two Years following Revision ACLR
ACLR = Anterior Cruciate Ligament Revision; kg = kilograms; m2 = meters squared; KSREC = KOOS Sports and Recreation; MOI = self-reported mechanism of injury; PRO = Patient Reported Outcome Measures
Recreational patients had significantly higher odds of nRTA within the two years following rACLR compared to high level patients (aOR = 2.03, 95% CI: 1.14 to 3.61, P = 0.016). As expected, the higher the MARX score at baseline, the lower the odds of nRTA (aOR = 0.89, 95% CI: 0.84 to 0.94, P < 0.001). Additionally, females were 2.9 times more likely to report nRTA compared to males in the two years following rACLR surgery (95% CI: 1.77 to 4.77, P < 0.001). Current smokers were found to be 3.3 times more likely to nRTA following rACLR surgery compared to those who reported never smoking (aOR = 3.34, 95% CI: 1.62 to 6.88, P = 0.001).
Only one previous injury characteristic was found to significantly impact RTA status. Patients who underwent a previous medial meniscal excision were two times more likely not to RTA compared with those who did not have a previous medial meniscal excision (aOR = 1.94, 95% CI: 1.15 to 3.27, P = 0.013). Interesting, those who reported not returning to activity had significantly higher odds of self-reporting a non-traumatic versus traumatic mechanism of injury to their ACL graft (aOR = 1.74, 95% CI: 1.07 to 2.85, P = 0.026).
Discussion
As RTA following clearance from rehabilitation is a common priority, we sought to answer three questions previously unanswered using a large, prospective cohort of only previously active patients undergoing a rACLR: 1) Are there differences in pre-revision injury profiles among those who do and do not RTA, 2) how does activity return correlate with two-year PROs scores, and 3) does this differ between recreational and high-level patients? Findings show that much of the cohort reported a return to being active in the two years following their rACLR. However, RTA within two years following surgery is not necessarily associated with patient-reported improvement in knee function and pain two years later among patients who were active at baseline. Most interesting, approximately 20% of patients who RTA had worse outcomes reported at two years.
From our previous work examining activity (i.e., sports) participation following rACLR in the complete cohort, we found patients involved in either multiple or single activities reported higher PROs at two years.18 After restricting the cohort to patients indicating participation in at least one type of activity at baseline and who had complete follow-up data, we found 85% returned to activity at two years. This “active” cohort was similar to the full cohort, however, it was slightly younger and more active as expected. Our results were comparable to those from the FAST cohort, where 75% of patients returned to running with approximately 60% returning to their same pre-injury sport.23 In addition, these results fall within the range (56–100%) of ACLR RTA status.8 Our categorization of “active,” was a broad definition which included all patients indicating a sport or specific activity on the patient questionnaire. Given the diverse range (e.g., dancing and aerobics to trampoline, Olympic sports, martial arts and walking), we therefore considered our population have a general definition of being physically active. This could account for differences in RTA when compared to other studies. However, the significant decrease in MARX scores of those who did not return to activity bolsters our confidence in patients being appropriately classified. These results represent a patient’s RTA relative to their participation before and after revision surgery, not whether they changed participation level or returned to their primary sport which should be examined in future studies.
Previous work from the complete cohort has shown an improvement in two-year outcomes on the IKDC, KOOS, and WOMAC instruments yet decreased MARX scores compared to baseline.17–19 This leads us to consider the paradox: Do patients stop being active due to the condition of their knee (function/pain), or is the knee structurally sound, but they cease activity in an attempt to protect the knee? Given the two time points of PRO collection, we cannot extrapolate when scores begin to decline. In order to answer this question, researchers need to consider adding shorter time intervals for PRO collection: post-revision ACL surgery (e.g., three and six months), post-surgical clearance by physical therapy, and at the one-year follow-up to better understand trends in PROs.
When examining cohort-level scores, we can say that the majority of patients RTA, but not all who RTA had improved outcomes. These results were found across questionnaires, as anywhere from 16% to 25% of patients reported a decrease in their two-year scores from baseline, whereas only approximately 40% of the patients improved on the MARX. This was unexpected, as most of the previous work has shown consistent improvement in scores when examining the complete cohort.
Examining the group-level direction of score change (i.e., same/improved versus worse) in combination with their self-reported RTA status revealed some novel findings. On average, if MARX scores are excluded, approximately 69% of patients returned to activity with improved scores indicating the continued effectiveness of rACLR. However, in those who returned to activity, on average 20% reported decreased scores that were clinically meaningful, averaging between nine and twenty-two points from baseline. Further investigations are needed to determine why these patients elected to RTA with lower self-reported PRO scores. Much has been published about specific demographics, previous and current injury characteristics, and surgical factors that can influence two-year outcomes in primary ACL reconstruction patients.1, 5, 7, 10, 13, 17, 19 Yet, the literature is sparse on what happens following ACL revision surgery, during rehabilitation phase, and through the two-year mark post-ACL revision surgery. For example, functional brace use have been previously found to influence two-year PROs scores in the cohort.16 Moving forward, this is a critical area of rehabilitation research in rACLR patients that needs to be considered what allows individuals to remain healthy and continuously active further out from their surgical timepoint. The availability of five-year follow-up data from this cohort can begin to answer these questions.
It should be noted that 15% of those who did not RTA were previously active at baseline. Patient specific profiles for not returning included having a higher BMI at baseline, a current smoker, recreational patients, gender (females), lower baseline MARX scores, previous medial meniscus excision, and self-reported non-traumatic mechanism of ACL graft injury. Changes in questionnaire scores other than the KOOS sports/recreation and MARX (as expected) did not predict RTA status. Most importantly each of the other PROs scales were not statistically associated with RTA status when accounting for all other variables in the model including their binary PRO change status (better/worse). Therefore, if patients return for a follow-up appointment and report being active, physicians should not assume this is associated with a healthy and highly functional knee.
Eleven percent of the cohort reported improved scores with nRTA while approximately four percent reported decreased scores and nRTA. We have previously theorized patients may be opting to not return to their activity following revision surgery in order to protect the integrity of their repaired knee.13 If so, this is only a small proportion of the cohort. Emerging evidence from the primary ACL reconstruction literature has found the increasing influence of psychological factors on returning to sport with fear of re-injury, lack of confidence in knee repair, depression and lack of motivation for not returning.21 These factors may be even more influential in post-revision ACLR recovery, particularly for those who have successful revisions [from a surgical viewpoint] but cease activity.
Those who did not RTA and reported worse scores at two years only represented an even smaller proportion of the cohort (4%). However, the magnitude of the PRO score reduction is nearly three-fold the threshold for MCID across all PRO measures. In these patients, the current structural integrity of the knee may be compromised and if these patients are either not expecting to RTA after surgery or unable due to their preceding injuries, non-operative treatment may be more beneficial for their presenting injury pattern. This leads us to consider the question of how many times can an ACL injury be successfully reconstructed with good outcomes both from a surgical and patients point of view? Physicians should continue to monitor expectations for RTA and its level among patients who present with these risk factors during their intake consultations.
In the cohort, recreational athletes represents a greater proportion of the overall active population but previous studies typically include only specific active populations (e.g., collegiate athletes, high-school populations or recreationally active individuals), not together. Overall, baseline recreational patients were two times as likely to report nRTA compared with high-level patients at two years. However, recreational patients did not have significantly lower PROs at two years compared with high-level patients except on the KOOS Sports/Recreation and Marx scores. Even though recreational level patients have lower odds of returning, this is not associated with lower patient reported outcome scores. Recreational patients scored two points lower on the KOOS sports and recreational scale compared with high-level patients, which is not clinically meaningful. On the Marx, high-level patients who did not RTA had a higher rate of decreasing activity levels than recreational patients who did nRTA. For example, high-level female patients who did nRTA following ACLR reported on average four points lower MARX scores compared to males, who had three points lower MARX scores. Our results were logical, as high-level patients would be expected to have more active participation hours within their activities at baseline than recreational patients. Given that patients become less active as they age, we expected these results in high-level active patients compared to recreational patients.13 These results demonstrated that baseline participation level only influences activity levels at two years, not knee health and function. Nonetheless, due to the small sample sizes in high participation levels, it is unknown if categories within high-level patients have different injury profiles (i.e., high school vs collegiate, semi-pro vs pro).
A key limitation in the study was the lack of information available as to why participants did nRTA, as this question was not asked on the two year post-operative questionnaires. Second, we used a proxy, sport participation, as an indicator for activity since no direct measurement of physical activity was available. Given the associated KOOS Sports/Recreation and MARX activity levels we are confident our approach provided a reliable estimate of patient self-reported activity levels. Further, we do not know if or when patients were cleared to RTA after rACLR. Last, at the two year follow-up assessment, no physical exam or imaging data were collected. Strengths of the study include the cohort being the largest prospective cohort study examining rACLR using validated PROs at each follow-up period, with additional planned assessments at six and ten years post-surgery. Given the initial study design, we have a broad representation of varying levels of activity participation and active patients in the cohort, which increases the generalizability of these results to all active patients undergoing rACLR.
Conclusion
Among previously active patients, 85% undergoing rACLR return to activity at their previous pre-operative activity level without significant impact on most two-year PROs. Recreational-level activity participants at time of revision surgery were twice as likely to nRTA compared to those participating in higher-level activities. Overall, rACLR leads to improved function and decreased pain scores for a majority of those who do RTA. However, sports medicine specialists should not assume that all patients who RTA will have higher PROs two years; up to 20% of patients who RTA reported a clinically meaningful decline in PRO score. This patient group may be one of the driving factors for the long-term osteoarthritis changes seen in rACLR patients ten-years after surgery. Patients with known risk factors for nRTA after rACLR should be counseled on realistic expectations for post-surgical outcomes and encouraged to explore alternatives. Future work is needed to better understand why some patients who RTA after rACLR do not have improved PROs at two-year follow-up.
What is known about the subject:
Prior researchers report patients undergoing an ACL revision surgery have increased patient reported outcome scores from baseline, yet not to the same levels as those undergoing a primary ACL reconstruction. Previous graft choice and demographic characteristics have been found to influence this association.
What this study adds to existing knowledge:
This study identifies risk factors for nRTA limited to a previously active population. Patients who report worse scores at two years have a clinically meaningful decline across all PROs, yet active patients at baseline still participate in activities two years after surgery. Results suggest activity participation is not necessarily associated with a healthy and highly functional knee. Changes in two-year PROs do not vary between recreational and high-level participants.
Acknowledgements:
This project was funded by grant No. 5R01-AR060846 from the National Institutes of Health/National Institute of Arthritis and Musculoskeletal and Skin Diseases.
We express our appreciation to the late Barton Mann, PHD (AOSSM, Rosemont, IL USA), Timothy M. Hosea, MD (University Orthopaedic Associates LLC, Princeton, NJ USA), and Allen F. Anderson, MD (Tennessee Orthopaedic Alliance, Nashville, TN USA) whose contribution to this work was of great significance.
We would also like to thank Jack T. Andrish, MD (Cleveland Clinic, Cleveland, OH USA), John D. Campbell, MD (Bridger Orthopedic and Sports Medicine, Bozeman, MT USA) and Diane L. Dahm, MD (Mayo Clinic, Rochester, MN USA) for their effort and leadership on this project. All are enjoying a well-deserved and happy retirement after many years of dedication to the advancement of orthopedics.
Glossary
- rACLR
revision ACLR
- ACLR
primary
- nRTA
not returning to activity
Contributor Information
John P. Bigouette, Slocum Research & Education Foundation, Eugene, OR USA.
Erin C. Owen, Slocum Research & Education Foundation, Eugene, OR USA.
Brett Brick A. Lantz, Slocum Research & Education Foundation, Eugene, OR USA.
Rudolf G. Hoellrich, Slocum Research & Education Foundation, Eugene, OR USA.
Rick W. Wright, Vanderbilt University, Nashville, TN USA.
Laura J. Huston, Vanderbilt University, Nashville, TN USA.
Amanda K. Haas, Washington University in St. Louis, St. Louis, MO USA.
Christina R. Allen, Yale University, New Haven, CT USA.
Daniel E. Cooper, W.B. Carrell Memorial Clinic, Dallas, TX USA.
Thomas M. DeBerardino, The San Antonio Orthopaedic Group, San Antonio, TX USA.
Warren R. Dunn, Texas Orthopedic Hospital, Houston, TX USA.
Kurt P. Spindler, Cleveland Clinic, Cleveland, OH USA.
Michael J. Stuart, Mayo Clinic, Rochester, MN USA.
John P. Albright, University of Iowa Hospitals and Clinics, Iowa City, IA USA.
Annunziato (Ned) Amendola, Duke University, Durham, NC USA.
Christopher C. Annunziata, Commonwealth Orthopaedics & Rehabilitation, Arlington, VA USA.
Robert A. Arciero, University of Connecticut Health Center, Farmington, CT USA.
Bernard R. Bach, Jr, Rush University Medical Center, Chicago, IL USA.
Champ L. Baker, III, The Hughston Clinic, Columbus, GA USA.
Arthur R. Bartolozzi, 3B Orthopaedics, University of Pennsylvania Health System, Philadelphia, PA USA.
Keith M. Baumgarten, Orthopedic Institute, Sioux Falls, SD USA.
Jeffery R. Bechler, University Orthopaedic Associates LLC, Princeton, NJ USA.
Jeffrey H. Berg, Town Center Orthopaedic Associates, Reston, VA USA.
Geoffrey A. Bernas, State University of New York at Buffalo, Buffalo, NY.
Stephen F. Brockmeier, University of Virginia, Charlottesville, VA USA.
Robert H. Brophy, Washington University in St. Louis, St. Louis, MO USA.
Charles A. Bush-Joseph, Rush University Medical Center, Chicago, IL USA.
J. Brad Butler V, Orthopedic and Fracture Clinic, Portland, OR USA.
James L. Carey, University of Pennsylvania, Philadelphia, PA USA.
James E. Carpenter, University of Michigan, Ann Arbor, MI USA.
Brian J. Cole, Rush University Medical Center, Chicago, IL USA.
Jonathan M. Cooper, HealthPartners Specialty Center, St. Paul, MN USA.
Charles L. Cox, Vanderbilt University, Nashville, TN USA.
R. Alexander Creighton, University of North Carolina Medical Center, Chapel Hill, NC USA.
Tal S. David, Synergy Specialists Medical Group, San Diego, CA USA.
David C. Flanigan, The Ohio State University, Columbus, OH USA.
Robert W. Frederick, The Rothman Institute/Thomas Jefferson University, Philadelphia, PA USA.
Theodore J. Ganley, Children’s Hospital of Philadelphia, Philadelphia, PA USA.
Elizabeth A. Garofoli, Washington University in St. Louis, St. Louis, MO USA
Charles J. Gatt, Jr, University Orthopaedic Associates LLC, Princeton, NJ USA.
Steven R. Gecha, Princeton Orthopaedic Associates, Princeton, NJ USA.
James Robert Giffin, Fowler Kennedy Sport Medicine Clinic, University of Western Ontario, London Ontario, Canada.
Sharon L. Hame, David Geffen School of Medicine at UCLA, Los Angeles, CA USA.
Jo A. Hannafin, Hospital for Special Surgery, New York, NY USA.
Christopher D. Harner, University of Texas Health Center, Houston, TX USA.
Norman Lindsay Harris, Jr, Grand River Health in Rifle, CO USA.
Keith S. Hechtman, UHZ Sports Medicine Institute, Coral Gables, FL USA.
Elliott B. Hershman, Lenox Hill Hospital, New York, NY USA.
David C. Johnson, National Sports Medicine Institute, Leesburg, VA USA.
Timothy S. Johnson, National Sports Medicine Institute, Leesburg, VA USA.
Morgan H. Jones, Cleveland Clinic, Cleveland, OH USA.
Christopher C. Kaeding, The Ohio State University, Columbus, OH USA.
Ganesh V. Kamath, University of North Carolina Medical Center, Chapel Hill, NC USA.
Thomas E. Klootwyk, Methodist Sports Medicine, Indianapolis, IN USA.
Bruce A. Levy, Mayo Clinic Rochester, MN USA.
C. Benjamin Ma, University of California, San Francisco, CA USA.
G. Peter Maiers, II, Methodist Sports Medicine Center, Indianapolis, IN USA.
Robert G. Marx, Hospital for Special Surgery, New York, NY USA.
Matthew J. Matava, Washington University in St. Louis, St. Louis, MO USA.
Gregory M. Mathien, Knoxville Orthopaedic Clinic, Knoxville, TN USA.
David R. McAllister, David Geffen School of Medicine at UCLA, Los Angeles, CA USA.
Eric C. McCarty, University of Colorado Denver School of Medicine, Denver, CO USA.
Robert G. McCormack, University of British Columbia/Fraser Health Authority, British Columbia, Canada.
Bruce S. Miller, University of Michigan, Ann Arbor, MI USA.
Carl W. Nissen, Connecticut Children’s Medical Center, Hartford, CT USA.
Daniel F. O’Neill, Littleton Regional Healthcare, Littleton, NH USA.
Brett D. Owens, Warren Alpert Medical School, Brown University, Providence, RI USA.
Richard D. Parker, Cleveland Clinic, Cleveland, OH USA.
Mark L. Purnell, Aspen Orthopedic Associates, Aspen, CO USA.
Arun J. Ramappa, Beth Israel Deaconess Medical Center, Boston, MA USA.
Michael A. Rauh, State University of New York at Buffalo, Buffalo, NY USA.
Arthur C. Rettig, Methodist Sports Medicine, Indianapolis, IN USA.
Jon K. Sekiya, University of Michigan, Ann Arbor, MI USA.
Kevin G. Shea, Intermountain Orthopaedics, Boise, ID USA.
Orrin H. Sherman, NYU Hospital for Joint Diseases, New York, NY USA.
James R. Slauterbeck, University of South Alabama, Mobile, AL USA.
Matthew V. Smith, Washington University in St. Louis, St. Louis, MO USA.
Jeffrey T. Spang, University of North Carolina Medical Center, Chapel Hill, NC USA.
LTC Steven J. Svoboda, Keller Army Community Hospital, United States Military Academy, West Point, NY USA.
Timothy N. Taft, University of North Carolina Medical Center, Chapel Hill, NC USA.
Joachim J. Tenuta, Albany Medical Center, Albany, NY USA.
Edwin M. Tingstad, Inland Orthopaedic Surgery and Sports Medicine Clinic, Pullman, WA USA.
Armando F. Vidal, University of Colorado Denver School of Medicine, Denver, CO USA.
Darius G. Viskontas, Royal Columbian Hospital, New Westminster, BC Canada.
Richard A. White, Fitzgibbon’s Hospital, Marshall, MO USA.
James S. Williams, Jr, Cleveland Clinic, Euclid, OH USA.
Michelle L. Wolcott, University of Colorado Denver School of Medicine, Denver, CO USA.
Brian R. Wolf, University of Iowa Hospitals and Clinics, Iowa City, IA USA.
James J. York, Orthopaedic and Sports Medicine Center, LLC, Pasedena, MD.
References
- 1.The Development and Early to Midterm Findings of the Multicenter Revision Anterior Cruciate Ligament Study. J Knee Surg. 2016;29(7):528–532. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Meniscal and Articular Cartilage Predictors of Clinical Outcome After Revision Anterior Cruciate Ligament Reconstruction. Am J Sports Med. 2016;44(7):1671–1679. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Brophy RH, Haas AK, Huston LJ, Nwosu SK, Wright RW. Association of Meniscal Status, Lower Extremity Alignment, and Body Mass Index With Chondrosis at Revision Anterior Cruciate Ligament Reconstruction. Am J Sports Med. 2015;43(7):1616. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Chen JL, Allen CR, Stephens TE, et al. Differences in mechanisms of failure, intraoperative findings, and surgical characteristics between single- and multiple-revision ACL reconstructions. Am J Sports Med. 2013;41(7):1571–1578. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Cooper DE, Dunn WR, Huston LJ, et al. Physiologic Preoperative Knee Hyperextension Is a Predictor of Failure in an Anterior Cruciate Ligament Revision Cohort: A Report From the MARS Group. Am J Sports Med. 2018:363546518777732. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Denti M, Vetere DL, Bait C, Schönhuber H, Melegati G, Volpi P. Revision Anterior Cruciate Ligament Reconstruction: Causes of Failure, Surgical Technique, and Clinical Results. The American Journal of Sports Medicine. 2008;36(10):1896–1902. [DOI] [PubMed] [Google Scholar]
- 7.Ding DY, Zhang AL, Allen CR, et al. Subsequent Surgery After Revision Anterior Cruciate Ligament Reconstruction: Rates and Risk Factors From a Multicenter Cohort. Am J Sports Med. 2017;45(9):2068–2076. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Glogovac G, Schumaier AP, Grawe BM. Return to Sport Following Revision Anterior Cruciate Ligament Reconstruction in Athletes: A Systematic Review. Arthroscopy. 2019;35(7):2222–2230. [DOI] [PubMed] [Google Scholar]
- 9.Irrgang JJ, Anderson AF, Boland AL, et al. Responsiveness of the international knee documentation committee subjective knee form. Am J Sports Med. 2006;34(10):1567–1573. [DOI] [PubMed] [Google Scholar]
- 10.Magnussen RA, Borchers JR, Pedroza AD, et al. Risk Factors and Predictors of Significant Chondral Surface Change From Primary to Revision Anterior Cruciate Ligament Reconstruction: A MOON and MARS Cohort Study. Am J Sports Med. 2018;46(3):557–564. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Mahmoud SS, Odak S, Coogan S, McNicholas MJ. A prospective study to assess the outcomes of revision anterior cruciate ligament reconstruction. Int Orthop. 2014;38(7):1489–1494. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Group MARS. Effect of graft choice on the outcome of tevision anterior cruciate ligament teconstruction in the multicenter ACL revision study (MARS) cohort. Am J Sports Med. 2014;42(10):2301–2310. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Group MARS. Factors influencing graft choice in revision anterior cruciate ligament reconstruction in the MARS Group. J Knee Surg. 2016;29(6):458–463. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Group MARS. Meniscal and articular cartilage predictors of clinical outcome after revision anterior cruciate ligament reconstruction. Am J Sports Med. 2016;44(7):1671–1679. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Group MARS. Radiographic findings in revision anterior cruciate ligament reconstructions from the Mars cohort. J Knee Surg. 2013;26(4):239. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Group MARS. Rehabilitation predictors of clinical outcome following revision ACL reconstruction in the MARS cohort. J Bone Joint Surg Am. 2019;101(9):779–786. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Group MARS, Allen CR Anderson AF, et al. Surgical predictors of clinical outcomes after revision anterior cruciate ligament reconstruction. Am J Sports Med. 2017;45(11):2586–2594. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Group MARS, Bigouette JP Owen EC, et al. Relationship Between Sports Participation After Revision Anterior Cruciate Ligament Reconstruction and 2-Year Patient-Reported Outcome Measures. Am J Sports Med. 2019;47(9):2056–2066. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Group MARS, Wright RW Huston LJ, et al. Predictors of Patient-Reported Outcomes at 2 Years After Revision Anterior Cruciate Ligament Reconstruction. Am J Sports Med. 2019;47(10):2394–2401. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Group MARS, Wright RW Huston LJ, et al. Descriptive epidemiology of the Multicenter ACL Revision Study (MARS) cohort. Am J Sports Med. 2010;38(10):1979–1986. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Nwachukwu BU, Patel BH, Lu Y, Allen AA, Williams RJ, 3rd. Anterior Cruciate Ligament Repair Outcomes: An Updated Systematic Review of Recent Literature. Arthroscopy. 2019;35(7):2233–2247. [DOI] [PubMed] [Google Scholar]
- 22.Roos EM, Lohmander LS. The knee injury and osteoarthritis outcome score (KOOS): from joint injury to osteoarthritis. Health Qual Life Outcomes. 2003;1:64. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Sadeqi M, Klouche S, Bohu Y, Herman S, Lefevre N, Gerometta A. Progression of the Psychological ACL-RSI Score and Return to Sport After Anterior Cruciate Ligament Reconstruction: A Prospective 2-Year Follow-up Study From the French Prospective Anterior Cruciate Ligament Reconstruction Cohort Study (FAST). Orthop J Sports Med. 2018;6(12):2325967118812819. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Wright RW. Knee injury outcomes measures. J Am Acad Orthop Surg. 2009;17(1):31–39. [DOI] [PubMed] [Google Scholar]
- 25.Wright RW. Knee sports injury outcome measures. J Knee Surg. 2005;18(1):69–72. [DOI] [PubMed] [Google Scholar]
- 26.Wright RW, Spindler KP, Huston LJ, et al. Revision ACL reconstruction outcomes - MOON cohort. J Knee Surg. 2011;24(4):289–294. [DOI] [PMC free article] [PubMed] [Google Scholar]