External validation of the Norwegian anterior cruciate ligament reconstruction revision prediction model using patients from the STABILITY 1 Trial

PURPOSE

A machine learning-based anterior cruciate ligament (ACL) revision prediction model has been developed using Norwegian Knee Ligament Register (NKLR) data, but lacks external validation outside Scandinavia. This study aimed to assess the external validity of the NKLR model (https://swastvedt.shinyapps.io/calculator_rev/) using the STABILITY 1 randomized clinical trial (RCT) data set. The hypothesis was that model performance would be similar.

METHODS

The NKLR Cox Lasso model was selected for external validation owing to its superior performance in the original study. STABILITY 1 patients with all five predictors required by the Cox Lasso model were included. The STABILITY 1 RCT was a prospective study which randomized patients to receive either a hamstring tendon autograft (HT) alone or HT plus a lateral extra-articular tenodesis (LET). Since all patients in the STABILITY 1 trial received HT ± LET, three configurations were tested: 1: all patients coded as HT, 2: HT + LET group coded as bone-patellar tendon-bone (BPTB) autograft, 3: HT + LET group coded as unknown/other graft choice. Model performance was assessed via concordance and calibration.

RESULTS

In total, 591/618 (95.6%) STABILITY 1 patients were eligible for inclusion, with 39 undergoing revisions within 2 years (6.6%). Model performance was best when patients receiving HT + LET were coded as BPTB. Concordance was similar to the original NKLR prediction model for 1- and 2-year revision prediction (STABILITY: 0.71; NKLR: 0.68-0.69). Concordance 95% confidence interval (CI) ranged from 0.63 to 0.79. The model was well calibrated for 1-year prediction while the 2-year prediction demonstrated evidence of miscalibration.

CONCLUSION

When patients in STABILITY 1 who received HT + LET were coded as BPTB in the NKLR prediction model, concordance was similar to the index study. However, due to a wide 95% CI, the true performance of the prediction model with this Canadian and European cohort is unclear and a larger data set is required to definitively determine the external validity. Further, better calibration for 1-year predictions aligns with general prediction modelling challenges over longer periods. While not a large enough sample size to elicit the true accuracy and external validity of the prediction model when applied to North American patients, this analysis provides more support for the notion that HT plus LET performs similarly to BPTB reconstruction. In addition, despite the wide confidence interval, this study suggests optimism regarding the accuracy of the model when applied outside of Scandinavia.

LEVEL OF EVIDENCE

Level 3, cohort study.

Navigating the 'new normal': what guidelines exist for postpartum return to physical activity and sport? A scoping review

OBJECTIVE

Women are often advised to return to activity (RTA) as early as 6 weeks postpartum, despite undergoing significant physical, physiological and psychological changes. Our objective was to examine existing evidence and clinical practice guidelines to navigate a safe and successful RTA or return to sport (RTS) postpartum.

METHODS

We searched CINAHL, Embase, Medline, PsycINFO and SPORTDiscus and included any secondary studies with recommendations or guidelines for RTA or RTS postpartum. Grey literature and primary sources were excluded. Four reviewers independently screened titles and abstracts, followed by full-text review for eligibility, with conflicts resolved by a third-party reviewer. One reviewer extracted data, which was cross-referenced by another reviewer.

RESULTS

5851 studies were screened, and 33 were included in this scoping review. Most studies stated that RTS postpartum can begin once 'medically safe', around 6 weeks postpartum, but this term was generally left undefined. In addition, most studies recommended engaging in 150 min of moderate-vigorous physical activity per week after 6 weeks postpartum, but the type of exercise recommended was often non-specific.

CONCLUSION

A lack of consistent, evidence-based guidelines exist for RTA or RTS postpartum. Multiple evidence gaps require additional research to inform patient and activity specific guidelines for a safe and successful RTA or RTS postpartum.

The Effect of Staged Versus Usual Care Physiotherapy on Knee Function Following Anterior Cruciate Ligament Reconstruction

CONTEXT

The long duration and high cost of anterior cruciate ligament reconstruction (ACLR) rehabilitation can pose barriers to completing rehabilitation, the latter stages of which progress to demanding sport-specific exercises critical for a safe return to sport. A staged approach shifting in-person physiotherapy sessions to later months of recovery may ensure patients undergo the sport-specific portion of ACLR rehabilitation. Design/Objective: To compare postoperative outcomes of knee function in patients participating in a staged ACLR physiotherapy program to patients participating in usual care physiotherapy through a randomized controlled trial.

METHODS

One hundred sixty-two patients were randomized to participate in staged (n = 80) or usual care physiotherapy (n = 82) following ACLR and assessed preoperatively and postoperatively at 2 weeks, 6 weeks, 3 months, and 6 months. The staged group completed the ACLR rehabilitation protocol at home for the first 3 months, followed by usual care in-person sessions. The usual care group completed in-person sessions for their entire rehabilitation. Outcome measures included the Lower Extremity Functional Scale, International Knee Documentation Committee Questionnaire, pain, range of motion, strength, and hop testing.

RESULTS

There were no statistically significant between-group differences in measures of knee function at 6 months postoperative. Patients in the usual care group reported significantly higher International Knee Documentation Committee scores at 3 months postoperative (mean difference = 5.8; 95% confidence interval,  1.3 to 10.4; P = .01).

CONCLUSION

A staged approach to ACLR rehabilitation does not appear to impede knee function at 6 months postoperative but may result in worse patient reported outcomes at early follow-ups.

A bifactor model supports unidimensionality of the International Knee Documentation Committee Subjective Knee Form in young active patients with anterior cruciate ligament tears: a retrospective analysis of a randomized controlled trial

BACKGROUND

The International Knee Documentation Committee Subjective Knee Form (IKDC) is the most highly recommended patient reported outcome measure for assessing patients with anterior cruciate ligament (ACL) injuries and those undergoing ACL reconstruction (ACLR) surgery. The IKDC was developed as a unidimensional instrument for a variety of knee conditions. Structural validity, which determines how an instrument is scored, has not been definitively confirmed for the IKDC in respondents with ACL injuries, and in fact an alternative two-factor/subscale structure has been proposed in this population. The purpose of this study was to determine the most appropriate structure and scoring system for the IKDC in young active patients following ACL injury.

METHODS

In total, 618 young patients deemed at high risk of graft rupture were randomized into the Stability 1 trial. Of the trial participants, 606 patients (98%) completed a baseline IKDC questionnaire used for this analysis. A cross sectional retrospective secondary data analysis of the Stability 1 baseline IKDC data was completed to assess the structural validity of the IKDC using exploratory and confirmatory factor analyses. Factor analyses were used to test model fit of the intended one-factor structure, a two-factor structure, and alternative four-factor and bifactor structures (i.e., a combination of a unidimensional factor with additional specific factors) of the IKDC, in a dataset of young active ACL patients.

RESULTS

The simple one-factor and two-factor structures of the IKDC displayed inadequate fit in our dataset of young ACL patients. A bifactor model provided the best fit. This model contains one general factor that is substantially associated with all items, plus four secondary, more specific content factors (symptoms, activity level, activities of daily living, and sport) with generally weaker associations to subsets of items. Although the single-factor model did not provide unambiguous support to unidimensionality of the IKDC based on fit indices, the bifactor model supports unidimensionality of the IKDC when covariance between items with similar linguistic structure, response options, or content are acknowledged.

CONCLUSIONS

Overall, findings of a bifactor model with evidence of a reliable general factor well defined by all items lends support to continue interpreting and scoring this instrument as unidimensional. This should be confirmed in other samples. Clinically, based on these findings, the IKDC can be represented by a single score for young active patients with ACL tears. A more nuanced interpretation would also consider secondary factors such as sport and activity level.

TRIAL REGISTRATION

The Stability 1 trial for which these data were collected was registered on ClinicalTrial.gov (NCT02018354).

No increase in adverse events with lateral extra-articular tenodesis augmentation of anterior cruciate ligament reconstruction - Results from the stability randomized trial

OBJECTIVES

Results from the Stability Study suggest that adding a lateral extra-articular tenodesis (LET) to a hamstring tendon autograft reduces the rate of anterior cruciate ligament reconstruction (ACLR) failure in high-risk patients. The purpose of this study is to report adverse events over the 2-year follow-up period and compare groups (ACLR alone vs. ACLR + LET).

METHODS

Stability is a randomized clinical trial comparing hamstring tendon ACLR with and without LET. Patients aged 14-25 years with an ACL deficient knee were included. Patients were followed and adverse events documented (type, actions taken, resolution) with visits at 3, 6, 12, and 24 months postoperatively. Adverse events were categorized as none, minor medical, minor surgical, contralateral ACL rupture, or graft rupture. Patient reported outcome measures (PROMs) collected at each visit included the Knee Injury and Osteoarthritis Outcome Score (KOOS), International Knee Documentation Committee Score (IKDC), and ACL Quality of Life Questionnaire (ACL-QOL).

RESULTS

In total, 618 patients were randomized (mean age 18.9 years, 302 (49%) male). Forty-five patients (7%) suffered graft rupture; 34 (11%) in the ACLR group compared to 11 (4%) in the ACLR + LET group (RRR = 0.67, 95% CI 0.36 to 0.83, p < 0.001). There were no differences in effusion or infection rates between groups. The ACLR + LET group experienced an increased number of hardware removals (10 vs. 4). Overall, the rate of minor medical events (11%), minor surgical events (7%), and ipsilateral or contralateral ACL tears (10%) were low considering the high-risk patient profile. Increasing severity of adverse events was associated with lower PROMs at 24 months post-operative. Patients in the ACLR + LET group reported greater degree of pain at 3 months only. There were no clinically significant differences in range of motion between groups.

CONCLUSIONS

The addition of LET to hamstring tendon autograft ACLR in young patients at high risk of re-injury resulted in a statistically significant reduction in graft rupture. While the addition of LET may increase rates of hardware irritation, there was no significant increase in overall rates of minor medical adverse events, minor surgical events, or overall re-operation rates. The concerns regarding complications associated with a LET did not materialize in this study.

LEVEL OF EVIDENCE

Level I.

Development and Preliminary Validation of the KOOS-ACL: A Short Form Version of the KOOS for Young Active Patients With ACL Tears

BACKGROUND

The Knee injury and Osteoarthritis Outcomes Score (KOOS) is a widely used region-specific outcome measure for assessing patients of all ages with a variety of knee conditions. Use of the KOOS for young active patients with anterior cruciate ligament (ACL) tear has been called into question regarding its relevance and interpretability for this specific population. Furthermore, the KOOS does not have adequate structural validity for use in high-functioning patients with ACL deficiency.

PURPOSE

To develop a condition-specific short form version of the KOOS that is appropriate for the young active population with ACL deficiency: the KOOS-ACL.

STUDY DESIGN

Cohort study (diagnosis); Level of evidence, 2.

METHODS

A baseline data set of 618 young patients (≤25 years old) with ACL tears was divided into development and validation samples. Exploratory factor analyses were conducted in the development sample to identify the underlying factor structure and to reduce the number of items based on statistical and conceptual indicators. Confirmatory factor analyses were conducted to check fit indices of the proposed KOOS-ACL model in both samples. Psychometric properties of the KOOS-ACL were assessed using the same data set, expanded to include patient data from 5 time points (baseline and postoperative 3, 6, 12, and 24 months). Internal consistency reliability, structural validity, convergent validity, responsiveness to change, floor/ceiling effects, and detection of treatment effects between surgical interventions (ACL reconstruction alone vs ACL reconstruction + lateral extra-articular tenodesis) were assessed.

RESULTS

A 2-factor structure was deemed most appropriate for the KOOS-ACL. Of 42 items, 30 were removed from the full-length KOOS. The final KOOS-ACL model showed acceptable internal consistency reliability (α = .79-.90), structural validity (comparative fit index and Tucker-Lewis index = 0.98-0.99; root mean square error of approximation and standardized root mean square residual = 0.04-0.07), convergent validity (Spearman correlation with International Knee Documentation Committee subjective knee form = 0.61-0.83), and responsiveness across time (significant small to large effects; P < .05).

CONCLUSION

The new KOOS-ACL questionnaire contains 12 items and 2 subscales-Function (8 items) and Sport (4 items)-relevant to young active patients with an ACL tear. Use of this short form would reduce patient burden by more than two-thirds; it provides improved structural validity as compared with the full-length KOOS for our population of interest; and it demonstrates adequate psychometric properties in our sample of young active patients undergoing ACL reconstruction.

External Validation of the KOOS-ACL in the MOON Group Cohort of Young Athletes Followed for 10 Postoperative Years

BACKGROUND

The Knee injury and Osteoarthritis Outcome Score-Anterior Cruciate Ligament (KOOS-ACL) is a short form version of the KOOS, developed to target populations of young active patients with ACL tears. The KOOS-ACL consists of 2 subscales: Function (8 items) and Sport (4 items). The KOOS-ACL was developed and validated using data from the Stability 1 study from baseline to postoperative 2 years.

PURPOSE

To validate the KOOS-ACL in an external sample of patients matching the outcome's target population.

STUDY DESIGN

Cohort study (diagnosis); Level of evidence, 1.

METHODS

The Multicenter Orthopaedic Outcomes Network group cohort of 839 patients aged 14 to 22 years who tore their ACLs while playing sports was used to assess internal consistency reliability, structural validity, convergent validity, responsiveness to change, and floor/ceiling effects of the KOOS-ACL at 4 time points: baseline and postoperative 2, 6, and 10 years. Detection of treatment effects between graft type (hamstring tendon vs bone-patellar tendon-bone) were also compared between the full-length KOOS and KOOS-ACL.

RESULTS

The KOOS-ACL demonstrated acceptable internal consistency reliability (α = .82-.89), structural validity (Tucker-Lewis index and comparative fit index = 0.98-0.99; standardized root mean square residual and root mean square error of approximation = 0.04-0.07), convergent validity (Spearman correlation with International Knee Documentation Committee subjective knee form = 0.66-0.85; Western Ontario and McMaster Universities Osteoarthritis Index function = 0.84-0.95), and responsiveness to change across time (large effect sizes from baseline to postoperative 2 years; d = 0.94 [Function] and d = 1.54 [Sport]). Stable scores and significant ceiling effects were seen from 2 to 10 years. No significant differences in KOOS or KOOS-ACL scores were detected between patients with different graft types.

CONCLUSION

The KOOS-ACL shows improved structural validity when compared with the full-length KOOS and adequate psychometric properties in a large external sample of high school and college athletes. This strengthens the argument to use the KOOS-ACL to assess young active patients with ACL tears in clinical research and practice.

Meniscal repair at the time of primary ACLR does not negatively influence short term knee stability, graft rupture rates, or patient-reported outcome measures: the STABILITY experience

PURPOSE

To assess how meniscal repair and excision impact short term patient-reported outcome measures (PROMs), knee stability, and early graft rupture rates following primary hamstring anterior cruciate ligament reconstruction (ACLR) with or without lateral extra-articular tenodesis (LET) in a group of young active patients where meniscal repair is commonly advocated.

METHODS

Six hundred and eighteen patients under 25 years of age at high-risk of graft failure following ACLR were recruited to the Stability 1 study. Multivariable regression models were developed to identify statistically and clinically significant surgical and demographic predictors of Knee Injury and Osteoarthritis Outcome Score (KOOS), International Knee Documentation Committee Subjective Knee Form (IKDC), ACL Quality of Life Questionnaire (ACL-QOL) and Marx Activity Rating Scale (MARS) scores. Chi-Square tests of independence were used to explore the association between meniscal status (torn, not torn), meniscal treatment (excision or repair), graft rupture, and rotatory knee laxity.

RESULTS

Medial meniscus repair was associated with worse outcomes on the KOOS (β = -1.32, 95% CI: -1.57 to -1.10, p = 0.003), IKDC (β = -1.66, 95% CI: -1.53 to -1.02, p = 0.031) and ACL-QOL (β = -1.25, 95% CI: -1.61 to 1.02, p = n.s.). However, these associations indicated small, clinically insignificant changes based on reported measures of clinical relevance. Other important predictors of post-operative PROMs included age, sex, and baseline scores. Medial meniscus excision and lateral meniscus treatment (repair or excision) did not have an important influence on PROMs. There was no significant association between meniscal treatment and graft rupture or rotatory knee laxity.

CONCLUSION

While repairing the medial meniscus may result in a small reduction in PROM scores at two-year follow-up, these differences are not likely to be important to patients or clinicians. Any surgical morbidity associated with meniscal repair appears negligible in terms of PROMs. Meniscal repair does not affect rotatory laxity or graft failure rates in the short term. Therefore, meniscal repair should likely be maintained as the standard of care for concomitant meniscal tears with ACLR.

LEVEL OF EVIDENCE

III.

The Knee Injury and Osteoarthritis Outcome Score Does Not Have Adequate Structural Validity for Use With Young, Active Patients With ACL Tears

BACKGROUND

The Knee Injury and Osteoarthritis Outcome Score (KOOS) is well known and commonly used to assess young, active patients with ACL injuries. However, this application of the outcome measure has been called into question. There is currently no evidence supporting the structural validity of the KOOS for this patient population. Structural validity refers to whether a questionnaire meant to provide scores on different subscales behaves as intended in the populations of interest. Structural validity should be assessed for all questionnaire measures with multiple items or subscales.

QUESTIONS/PURPOSES

Does the KOOS demonstrate adequate structural validity in young, active patients with ACL tears, when evaluated using (1) exploratory and (2) confirmatory factor analyses?

METHODS

Between January 2014 and March 2017, 1033 patients were screened for eligibility in the Stability 1 randomized controlled trial from nine centers in Canada and Europe. Patients were eligible if they had an ACL deficient knee, were between 14 and 25 years old, and were thought to be at higher risk of reinjury based on the presence of two or more of the following factors: participation in pivoting sports, presence of a Grade 2 pivot shift or greater, generalized ligamentous laxity (Beighton score of 4 or greater), or genu recurvatum greater than 10°. Based on this criteria, 367 patients were ineligible and another 48 declined to participate. In total, 618 patients were randomized into the trial. Of the trial participants, 98% (605 of 618) of patients had complete baseline KOOS questionnaire data available for this analysis. Based on study inclusion criteria, the baseline KOOS data from the Stability 1 trial represents an appropriate sample to investigate the structural validity of the KOOS, specifically for the young, active ACL deficient population.A cross sectional retrospective secondary data analysis of the Stability 1 baseline KOOS data was completed to assess the structural validity of the KOOS using exploratory and confirmatory factor analyses. Exploratory factor analysis investigates how all questionnaire items group together based on their conceptual similarity in a specific sample. Confirmatory factor analysis is similar but used often in a second stage to test and confirm a proposed structure of the subscales. These methods were used to assess the established five-factor structure of the KOOS (symptoms [seven items], pain [nine items], activities of daily living [17 items], sport and recreation [five items], and quality of life [four items]) in young active patients with ACL tears. Incremental posthoc modifications, such as correlating questionnaire items or moving items to different subscales, were made to the model structure until adequate fit was achieved. Model fit was assessed using chi-square, root mean square error of approximation (RMSEA) and an associated 90% confidence interval, comparative fit index (CFI), Tucker-Lewis index (TLI), as well as standardized root mean square residual (SRMR). Adequate fit was defined as a CFI and TLI > 0.9, and RMSEA and SRMR < 0.08.

RESULTS

Structural validity of the KOOS was not confirmed when evaluated using (1) exploratory or (2) confirmatory factor analyses. The exploratory factor analysis, where the 42 KOOS items were allowed to group naturally, did not reflect adequate fit for a five-factor model (TLI = 0.828). Similarly, the confirmatory factor analysis used to investigate the KOOS structure as it was originally developed, revealed inadequate fit in our sample (RMSEA = 0.088 [90% CI 0.086 to 0.091]). Our analysis suggested a modified four-factor structure may be more appropriate in young, active ACL deficient patients; however, the final version presented here is not appropriate for clinical use because of the number and nature of post-hoc modifications required to reach adequate fit indices.

CONCLUSION

The established five-factor structure of the KOOS did not hold true in our sample of young, active patients undergoing ACL reconstruction, indicating poor structural validity.

CLINICAL RELEVANCE

We question the utility and interpretability of KOOS subscale scores for young, active patients with ACL tears with the current form of the KOOS. A modified version of the KOOS, adjusted for this patient population, is needed to better reflect and interpret the outcomes and recovery trajectory in this high-functioning group. A separate analysis with a defined a priori development plan would be needed to create a valid alternative.

Same knee, different goals: patients and surgeons have different priorities related to ACL reconstruction

PURPOSE

The priorities of patients should be shared by those treating them. Patients and surgeons are likely to have different priorities surrounding anterior cruciate ligament reconstruction (ACLR), with implications for shared decision-making and patient education. The optimal surgical approach for ACLR is constantly evolving, and the magnitude of treatment effect necessary for evidence to change surgical practice is unknown. The aim of this study was to determine (1) the priorities of surgeons and patients when making decisions regarding ACLR and (2) the magnitude of reduction in ACLR graft failure risk that orthopaedic surgeons require before changing practice.

METHODS

This study followed a cross-sectional survey design. Three distinct electronic surveys were administered to pre-operative ACLR patients, post-operative ACLR patients, and orthopaedic surgeons. Patients and surgeons were asked about the importance of various outcomes and considerations pertaining to ACLR. Surgeons were asked scenario-based questions regarding changing practice for ACLR based on new research.

RESULTS

Surgeons were more likely to prioritize outcomes related to the surgical knee itself, whereas patients were more likely to prioritize outcomes related to their daily lifestyle and activities. Knee instability and risk of re-injury were unanimous top priorities among all three groups. A mean relative risk reduction in ACLR graft failure of about 50% was required by orthopaedic surgeons to change practice regardless of the type of change, or patient risk profile.

CONCLUSION

There are discrepancies between the priorities of surgeons and patients, and orthopaedic surgeons appear resistant to changing practice for ACLR.

LEVEL OF EVIDENCE

IV.

Validation of a Risk Calculator to Personalize Graft Choice and Reduce Rupture Rates for Anterior Cruciate Ligament Reconstruction

BACKGROUND

Anterior cruciate ligament reconstructions (ACLRs) fail at an alarmingly high rate in young active individuals. The Multicenter Orthopaedic Outcomes Network (MOON) knee group has developed an autograft risk calculator that uses patient characteristics and lifestyle to predict the probability of graft rupture if the surgeon uses a hamstring tendon (HT) or a bone-patellar tendon-bone (BPTB) graft to reconstruct the ligament. If validated, this risk calculator can be used during the shared decision-making process to make optimal ACLR autograft choices and reduce rupture rates. The STABILITY 1 randomized clinical trial offers a large, rigorously collected data set of similar young active patients who received HT autograft with or without lateral extra-articular tenodesis (LET) for ACLR.

PURPOSE/HYPOTHESIS

The purpose was to validate the ACLR graft rupture risk calculator in a large external data set and to investigate the utility of BPTB and LET for ACLR. We hypothesized that the risk calculator would maintain adequate discriminative ability and calibration in the external STABILITY 1 data set when compared with the initial MOON development data set.

STUDY DESIGN

Cohort study (diagnosis); Level of evidence, 1.

METHODS

The model predictors for the risk calculator include age, sex, body mass index, sport played at the time of injury, Marx Activity Score, preoperative knee laxity, and graft type. The STABILITY 1 trial data set was used for external validation. Discriminative ability, calibration, and diagnostic test validity of the model were assessed. Finally, predictor strength in the initial and validation samples was compared.

RESULTS

The model showed acceptable discriminative ability (area under the curve = 0.73), calibration (Brier score = 0.07), and specificity (85.3%) to detect patients who will experience a graft rupture. Age, high-grade preoperative knee laxity, and graft type were significant predictors of graft rupture in young active patients. BPTB and the addition of LET to HT were protective against graft rupture versus HT autograft alone.

CONCLUSION

The MOON risk calculator is a valid predictor of ACLR graft rupture and is appropriate for clinical practice. This study provides evidence supporting the idea that isolated HT autografts should be avoided for young active patients undergoing ACLR.

REGISTRATION

NCT00463099 (MOON); NCT02018354 (STABILITY 1) (ClinicalTrials.gov identifiers).