Early ACLR and Risk and Timing of Secondary Meniscal Injury Compared With Delayed ACLR or Nonoperative Treatment: A Time-to-Event Analysis Using Machine Learning

Systematic review of machine-learning models in orthopaedic trauma

Aims

Machine-learning (ML) prediction models in orthopaedic trauma hold great promise in assisting clinicians in various tasks, such as personalized risk stratification. However, an overview of current applications and critical appraisal to peer-reviewed guidelines is lacking. The objectives of this study are to 1) provide an overview of current ML prediction models in orthopaedic trauma; 2) evaluate the completeness of reporting following the Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD) statement; and 3) assess the risk of bias following the Prediction model Risk Of Bias Assessment Tool (PROBAST) tool.

Methods

A systematic search screening 3,252 studies identified 45 ML-based prediction models in orthopaedic trauma up to January 2023. The TRIPOD statement assessed transparent reporting and the PROBAST tool the risk of bias.

Results

A total of 40 studies reported on training and internal validation; four studies performed both development and external validation, and one study performed only external validation. The most commonly reported outcomes were mortality (33%, 15/45) and length of hospital stay (9%, 4/45), and the majority of prediction models were developed in the hip fracture population (60%, 27/45). The overall median completeness for the TRIPOD statement was 62% (interquartile range 30 to 81%). The overall risk of bias in the PROBAST tool was low in 24% (11/45), high in 69% (31/45), and unclear in 7% (3/45) of the studies. High risk of bias was mainly due to analysis domain concerns including small datasets with low number of outcomes, complete-case analysis in case of missing data, and no reporting of performance measures.

Conclusion

The results of this study showed that despite a myriad of potential clinically useful applications, a substantial part of ML studies in orthopaedic trauma lack transparent reporting, and are at high risk of bias. These problems must be resolved by following established guidelines to instil confidence in ML models among patients and clinicians. Otherwise, there will remain a sizeable gap between the development of ML prediction models and their clinical application in our day-to-day orthopaedic trauma practice.

Graft failure within 2 years of isolated anterior cruciate ligament reconstruction is associated with increased risk of secondary meniscus tears

PURPOSE

A debilitating complication following anterior cruciate ligament reconstruction is a secondary meniscus tear. Currently, the literature is mixed regarding the risk factors associated with the incidence of secondary meniscus tears. The aim of this study was to investigate risk factors associated with meniscus tears following an isolated primary anterior cruciate ligament reconstruction. ACL graft failure was hypothesized to be the strongest risk factor for secondary meniscal injury occurrence.

METHODS

A retrospective cohort analysis was performed using the PearlDiver Database. Patients with a primary anterior cruciate ligament reconstruction were identified in the database. Patients with concomitant knee ligament injury or meniscus injury present at the time the index procedure were excluded. Patients were grouped to those who had a secondary meniscus tear within 2 years following anterior cruciate ligament reconstruction and those who did not. Univariate analysis and multivariable regression analysis was conducted to identify significant risk factors for a secondary meniscus tear.

RESULTS

There were 25,622 patients meeting criteria for inclusion in this study. Within 2 years from the primary anterior cruciate ligament reconstruction, there were 1,781 patients (7.0%) that experienced a meniscus tear. Graft failure had the highest odds of having a postoperative meniscus tear within 2 years (OR: 4.1; CI 3.5-4.8; p < 0.002). Additional significant risk factors included tobacco use (OR: 2.0; CI 1.0-3.1; p < 0.001), increased Charlson Comorbidity Index (OR: 1.2; CI 1.1-1.4), male gender (OR: 1.1; CI 1.1-1.2; p < 0.001), obesity (OR: 1.1; CI 1.1-1.2; p < 0.001), delayed surgery (OR:1.1; CI 1.1-1.2; p < 0.002), and patients age 30 and older (OR: 1.0; CI 1.0-1.0; p < 0.001).

CONCLUSIONS

This study found that anterior cruciate ligament graft failure is the strongest predictor of post-operative meniscus tears. Other risk factors, including tobacco use, increased CCI, male gender, obesity, delayed surgery, and age 30 and older, were established, with several being modifiable. Therefore, targeted preoperative optimization of modifiable risk factors and postoperative protocols may reduce the risk of secondary meniscus tears.

LEVEL OF EVIDENCE

Level III, prognostic trial.

IL-6 induces periostin production in human ACL remnants: a possible mechanism causing post-traumatic osteoarthritis

OBJECTIVE

Perostin (POSTN) and IL-6 consistently elevated after ACL injury, and ACL has been proposed as the major source of POSTN. However, there is a lack of evidence whether IL-6 induces ACL remnants to produce POSTN. This study aimed to investigate the effect of IL-6 on POSTN production in ACL fibroblasts, which may help us understand more about the mechanism of PTOA after ACL injury and ACL reconstruction.

METHODS

ACL remnants were harvested from 27 patients undergoing ACL reconstruction. Quantitative real-time polymerase chain reaction (PCR) was performed to examine the POSTN gene expression of ACL fibroblasts after treatment of different concentrations of IL-6. The POSTN protein production of ACL fibroblasts was determined using western blot analysis. The blockers of possible signaling pathways, including PI3K/Akt, Ras/MAPK, and JAK/STAT pathways, were added to test whether the effect of IL-6 on ACL fibroblast could be attenuated. ACL fibroblast and chondrocyte co-culture was carried out to determine the influence of ACL and IL-6 on chondrocytes.

RESULTS

Quantitative real-time PCR showed that IL-6 time-dependently and dose-dependently increased POSTN gene expression of ACL fibroblast. Western blot analysis also revealed that IL-6 dose-dependently induced POSTN protein production. Regarding the chronicity of ACL injury, the POSTN protein production was comparable between ACL remnants which were derived within 3 months of injury and at least 6 months after injury. PI3K/Akt blockers could attenuate the effect of IL-6 on ACL remnants, whereas Ras/MAPK and JAK/STAT did not decrease POSTN production. The coexistence of ACL and IL-6 induced more MMP-13 and ADAMTS-4 by chondrocytes.

CONCLUSIONS

IL-6 induced ACL remnants to produce POSTN. This effect could be attenuated by the PI3K/Akt blocker. Coexistence of IL-6 and ACL remnants may accelerate post-traumatic arthritis.

Optimal Timing of Anterior Cruciate Ligament Reconstruction in Patients With Anterior Cruciate Ligament Tear: A Systematic Review and Meta-analysis

IMPORTANCE

The timing of surgery has been regarded as a key factor in anterior cruciate ligament reconstruction (ACLR), and early vs delayed ACLR remains a controversial topic.

OBJECTIVE

To synthesize up-to-date published data from randomized clinical trials (RCTs) comparing early vs elective delayed ACLR for patients with ACL deficiency, in terms of clinical outcomes and complications.

DATA SOURCES

The PubMed, Cochrane Library, and Web of Science databases were systematically searched until September 9, 2022.

STUDY SELECTION

All published RCTs comparing clinical and functional outcomes and complications associated with early ACLR vs elective delayed ACLR.

DATA EXTRACTION AND SYNTHESIS

Two reviewers independently extracted relevant data and assessed the methodological quality following the PRISMA guidelines.

MAIN OUTCOMES AND MEASURES

Due to the clinical heterogeneity, the random-effects model was preferred. The primary outcomes were functional outcomes and complications. The Mantel-Haenszel test was used to evaluate dichotomous variables and the inverse variance method was used to assess continuous variables.

RESULTS

This meta-analysis included 972 participants in 11 RCTs stratified by follow-up duration. The following factors did not differ between early and delayed ACLR: operative time (mean difference, 4.97; 95% CI, -0.68 to 10.61; P = .08), retear (OR, 1.52; 95% CI, 0.52-4.43; P = .44), and infection (OR, 3.80; 95% CI, 0.77-18.79; P = .10). There were also no differences between groups in range of motion, knee laxity, International Knee Documentation Committee (IKDC rating scale), and Tegner score. IKDC score (mean difference, 2.77; 95% CI, 1.89-3.66; P < .001), and Lysholm score at 2-year follow-up (mean difference, 2.61; 95% CI, 0.74-4.48; P = .006) significantly differed between early and delayed ACLR. In addition, the timing of surgery was redefined in the included RCTs and subgroup analyses were performed, which validated the robustness of the principal results.

CONCLUSION AND RELEVANCE

This systematic review and meta-analysis found that early ACLR was not superior to delayed ACLR in terms of most factors analyzed, except for IKDC and Lysholm scores. This information should be available to patients with ACL deficiency and clinicians as part of the shared decision-making process of treatment selection.