Dataset Information

Predicting anterior cruciate ligament failure load with T₂* relaxometry and machine learning as a prospective imaging biomarker for revision surgery.

ABSTRACT: Non-invasive methods to document healing anterior cruciate ligament (ACL) structural properties could potentially identify patients at risk for revision surgery. The objective was to evaluate machine learning models to predict ACL failure load from magnetic resonance images (MRI) and to determine if those predictions were related to revision surgery incidence. It was hypothesized that the optimal model would demonstrate a lower mean absolute error (MAE) than the benchmark linear regression model, and that patients with a lower estimated failure load would have higher revision incidence 2 years post-surgery. Support vector machine, random forest, AdaBoost, XGBoost, and linear regression models were trained using MRI T₂* relaxometry and ACL tensile testing data from minipigs (n = 65). The lowest MAE model was used to estimate ACL failure load for surgical patients at 9 months post-surgery (n = 46) and dichotomized into low and high score groups via Youden's J statistic to compare revision incidence. Significance was set at alpha = 0.05. The random forest model decreased the failure load MAE by 55% (Wilcoxon signed-rank test: p = 0.01) versus the benchmark. The low score group had a higher revision incidence (21% vs. 5%; Chi-square test: p = 0.09). ACL structural property estimates via MRI may provide a biomarker for clinical decision making.

SUBMITTER: Flannery SW

PROVIDER: S-EPMC9981601 | biostudies-literature | 2023 Mar

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Predicting anterior cruciate ligament failure load with T<sub>2</sub>* relaxometry and machine learning as a prospective imaging biomarker for revision surgery.

Flannery Sean W SW Beveridge Jillian E JE Proffen Benedikt L BL Walsh Edward G EG Kramer Dennis E DE Murray Martha M MM Kiapour Ata M AM Fleming Braden C BC

Scientific reports 20230302 1

Non-invasive methods to document healing anterior cruciate ligament (ACL) structural properties could potentially identify patients at risk for revision surgery. The objective was to evaluate machine learning models to predict ACL failure load from magnetic resonance images (MRI) and to determine if those predictions were related to revision surgery incidence. It was hypothesized that the optimal model would demonstrate a lower mean absolute error (MAE) than the benchmark linear regression model ...[more]

PMID: 36864112

Similar Datasets

Project description:PurposeExternal validation of machine learning predictive models is achieved through evaluation of model performance on different groups of patients than were used for algorithm development. This important step is uncommonly performed, inhibiting clinical translation of newly developed models. Machine learning analysis of the Norwegian Knee Ligament Register (NKLR) recently led to the development of a tool capable of estimating the risk of anterior cruciate ligament (ACL) revision ( https://swastvedt.shinyapps.io/calculator_rev/ ). The purpose of this study was to determine the external validity of the NKLR model by assessing algorithm performance when applied to patients from the Danish Knee Ligament Registry (DKLR).MethodsThe primary outcome measure of the NKLR model was probability of revision ACL reconstruction within 1, 2, and/or 5 years. For external validation, all DKLR patients with complete data for the five variables required for NKLR prediction were included. The five variables included graft choice, femur fixation device, KOOS QOL score at surgery, years from injury to surgery, and age at surgery. Predicted revision probabilities were calculated for all DKLR patients. The model performance was assessed using the same metrics as the NKLR study: concordance and calibration.ResultsIn total, 10,922 DKLR patients were included for analysis. Average follow-up time or time-to-revision was 8.4 (± 4.3) years and overall revision rate was 6.9%. Surgical technique trends (i.e., graft choice and fixation devices) and injury characteristics (i.e., concomitant meniscus and cartilage pathology) were dissimilar between registries. The model produced similar concordance when applied to the DKLR population compared to the original NKLR test data (DKLR: 0.68; NKLR: 0.68-0.69). Calibration was poorer for the DKLR population at one and five years post primary surgery but similar to the NKLR at two years.ConclusionThe NKLR machine learning algorithm demonstrated similar performance when applied to patients from the DKLR, suggesting that it is valid for application outside of the initial patient population. This represents the first machine learning model for predicting revision ACL reconstruction that has been externally validated. Clinicians can use this in-clinic calculator to estimate revision risk at a patient specific level when discussing outcome expectations pre-operatively. While encouraging, it should be noted that the performance of the model on patients undergoing ACL reconstruction outside of Scandinavia remains unknown.Level of evidenceIII.

Project description:BackgroundRevision anterior cruciate ligament (ACL) reconstruction is believed to have an inferior outcome compared with primary ACL reconstruction. The available literature on the outcome of revision ACL reconstruction is sparse compared with that for primary reconstruction. The purpose of this systematic review was to test the hypothesis that the outcome of revision ACL reconstruction compares unfavorably with the historical outcome of primary ACL reconstruction.MethodsA systematic review of studies evaluating the outcome of revision ACL reconstructions with a minimum of two years of follow-up was performed. Pooled data were collected when appropriate and a mixed-effect-model meta-analysis was performed for important outcome measures that were reported in several studies (objective graft failure, Lysholm score, International Knee Documentation Committee [IKDC] subjective score, and IKDC objective score). Objective failure was defined as repeat revision, a side-to-side difference of >5 mm measured with use of a KT1000 arthrometer, or a pivot-shift grade of 2+ or 3+.ResultsTwenty-one studies were included, and 863 of the 1004 patients in these studies had a minimum of two years of follow-up and were analyzed. The pooled mean age of the patients at the time of the revision procedure was 30.6 years, and 66% were male. Objective failure occurred in 13.7% ± 2.7% of the patients (95% confidence interval, 8.0% to 19.4%). The mean Lysholm score in 491 patients was 82.1 ± 3.3 (95% confidence interval, 74.6 to 89.5) according to a mixed-model meta-analysis. The mean IKDC subjective score in 202 patients was 74.8 ± 4.4 (95% confidence interval, 62.5 to 87.0).ConclusionsRevision ACL reconstruction resulted in a worse outcome compared with primary ACL reconstruction. Patient-reported outcome scores were inferior to previously published results of primary ACL reconstruction, but these differences may not be clinically important. A dramatically elevated failure rate was noted after revision ACL reconstruction; this rate was nearly three to four times the failure rate in prospective series of primary ACL reconstructions.

Project description:PurposeGiven the paucity of literature on the re-revision of ACL, the current study was undertaken. The purpose of this systematic review was to synthesise and qualitatively assess the currently available evidence in the literature regarding the re-revision of ACL reconstruction (rrACLR).MethodsA systematic review was conducted based on the PRISMA guidelines. The following search terms were used in the title, abstract and keywords fields: "ACL" or "anterior cruciate ligament" AND "revision" or "multiple" or "repeat". The outcome data extracted from the studies were the Lysholm score, Subjective IKDC, Marx Score, Tegner, Marx Score, KOOS score, radiological changes and the rate of return to sports. Complications, failures and/or revision surgery were also analysed.ResultsThe cohort consisted of 295 patients [191 (64.7%) men and 104 (35.3%) women] with a mean age of 29.9 ± 2.8 years (range 14-58 years) from 10 studies. The mean postoperative follow-up (reported in all studies except one) was 66.9 ± 44.7 months (range 13-230.4 months). Associated injuries were 103 (34.9%) medial meniscus tears, 57 (19.3%) lateral meniscus tears, 14 (4.7%) combined medial plus lateral meniscus tears, 11 (3.7%) meniscal tears (not specified), 252 (85.4%) cartilage lesions, 6 (2.0%) medial collateral ligament injury and 2 (0.7%) lateral collateral ligament injuries. In 47 (15.9%) patients an extra-articular plasty was performed for the anterolateral ligament. In all studies that reported pre- and post-operative IKDC (subjective and objective) and Lysholm score, there was a significant improvement compared to the pre-operative value (p < 0.05). At the final follow-up, laxity measured with KT-1000 was found to be 2.2 ± 0.6 mm. 31 (10.5%) out of 295 patients returned to their pre-injury activity level. A total of 19 (6.4%) re-ruptures were found, while only 4 (1.4%) complications (all minors) were reported, out of which 2 (0.7%) were superficial infections, 1 (0.3%) cyclops lesion and 1 (0.3%) flexion loss.ConclusionMultiple revisions of anterior cruciate ligament reconstruction allow acceptable clinical results and a good degree of knee stability with a low rate of subsequent new re-ruptures but the possibility of regaining pre-injury sports activity is poor; whenever possible, it is preferred to revise the ligament in one stage. This surgery remains a challenge for orthopaedic surgeons and many doubts persist regarding the ideal grafts, additional extra-articular procedures and techniques to use.Level of evidenceIV.Study registrationPROSPERO-CRD42022352164 ( https://www.crd.york.ac.uk/prospero/ ).

Dataset Information

Predicting anterior cruciate ligament failure load with T₂* relaxometry and machine learning as a prospective imaging biomarker for revision surgery.

Publications

Predicting anterior cruciate ligament failure load with T<sub>2</sub>* relaxometry and machine learning as a prospective imaging biomarker for revision surgery.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets

Dataset Information

Predicting anterior cruciate ligament failure load with T2* relaxometry and machine learning as a prospective imaging biomarker for revision surgery.

Publications

Predicting anterior cruciate ligament failure load with T<sub>2</sub>* relaxometry and machine learning as a prospective imaging biomarker for revision surgery.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets

Predicting anterior cruciate ligament failure load with T₂* relaxometry and machine learning as a prospective imaging biomarker for revision surgery.