1
|
Zhan H, Kang X, Zhang X, Zhang Y, Wang Y, Yang J, Zhang K, Han J, Feng Z, Zhang L, Wu M, Xia Y, Jiang J. Machine Learning Models Reliably Predict Clinical Outcomes in Medial Patellofemoral Ligament Reconstruction. Arthroscopy 2024:S0749-8063(24)00556-5. [PMID: 39128684 DOI: 10.1016/j.arthro.2024.07.028] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 03/05/2024] [Revised: 07/17/2024] [Accepted: 07/17/2024] [Indexed: 08/13/2024]
Abstract
PURPOSE To develop the machine learning model to predict clinical outcomes following MPFLR and identify the important predictive indicators. METHODS This study included patients who underwent MPFLR from January 2018 to December 2022. The exclusion criteria were as follows: 1) concurrent bony procedures, 2) history of other knee surgeries, and 3) follow-up period of less than 12 months. Forty-two predictive models were constructed for seven clinical outcomes (failure to achieve MCID of clinical scores, return to pre-injury sports, pivoting sports, and recurrent instability) using six machine learning algorithms (Random Forest, Logistic Regression, Support Vector Machine, Decision Tree, implemented multilayer perceptron, and K-nearest neighbor). The performance of the model was evaluated using metrics such as the area under the receiver operating characteristic curve (AUC), accuracy, specificity, and sensitivity. Additionally, Shapley Additive Explanation summary plot was employed to identify the important predictive factors of the best-performing model. RESULTS A total of 218 patients met criteria. For the best-performing models in predicting failure to achieve the MCID for Lysholm, IKDC, Kujala, and Tegner scores, the AUCs and accuracies were 0.884 (good) and 87.3%, 0.859 (good) and 86.2%, 0.969 (excellent) and 97.0%, and 0.760 (fair) and 76.8%, respectively; 0.952 (excellent) and 95.2% for return to pre-injury sports; 0.756 (fair) and 75.4% for return to pivoting sports; and 0.943 (excellent) and 94.9% for recurrent instability. Low preoperative Tegner score, shorter time to surgery, and absence of severe trochlear dysplasia were significant predictors for return to pre-injury sports, while absence of severe trochlear dysplasia and patellar alta were significant predictors for return to pivoting sports. Older age, female sex, and low preoperative Lysholm score were highly predictive of recurrent instability. CONCLUSION The predictive models developed using machine learning algorithms can reliably forecast the clinical outcomes of MPFLR, particularly demonstrating excellent performance in predicting recurrent instability. LEVEL OF EVIDENCE Level III, case-control study.
Collapse
Affiliation(s)
- Hongwei Zhan
- Department of Sports Medicine, Honghui Hospital, Xi'an Jiaotong University, Xi'an, Shaanxi, China
| | - Xin Kang
- Department of Sports Medicine, Honghui Hospital, Xi'an Jiaotong University, Xi'an, Shaanxi, China
| | - Xiaobo Zhang
- Department of Sports Medicine, Honghui Hospital, Xi'an Jiaotong University, Xi'an, Shaanxi, China
| | - Yuji Zhang
- Lanzhou University Second Hospital, Orthopaedic Clinical Research Center of Gansu Province, Intelligent Orthopaedic Industry Technology Center of Gansu Province, Lanzhou Gansu, China
| | - Yanming Wang
- Second School of Clinical Medicine, Lanzhou University, Lanzhou, Gansu, China
| | - Jing Yang
- Second School of Clinical Medicine, Lanzhou University, Lanzhou, Gansu, China
| | - Kun Zhang
- Department of Sports Medicine, Honghui Hospital, Xi'an Jiaotong University, Xi'an, Shaanxi, China
| | - Jingjing Han
- Second School of Clinical Medicine, Lanzhou University, Lanzhou, Gansu, China
| | - Zhiwei Feng
- Lanzhou University Second Hospital, Orthopaedic Clinical Research Center of Gansu Province, Intelligent Orthopaedic Industry Technology Center of Gansu Province, Lanzhou Gansu, China
| | - Liang Zhang
- Department of Sports Medicine, Honghui Hospital, Xi'an Jiaotong University, Xi'an, Shaanxi, China
| | - Meng Wu
- Lanzhou University Second Hospital, Orthopaedic Clinical Research Center of Gansu Province, Intelligent Orthopaedic Industry Technology Center of Gansu Province, Lanzhou Gansu, China
| | - Yayi Xia
- Lanzhou University Second Hospital, Orthopaedic Clinical Research Center of Gansu Province, Intelligent Orthopaedic Industry Technology Center of Gansu Province, Lanzhou Gansu, China
| | - Jin Jiang
- Lanzhou University Second Hospital, Orthopaedic Clinical Research Center of Gansu Province, Intelligent Orthopaedic Industry Technology Center of Gansu Province, Lanzhou Gansu, China.
| |
Collapse
|
2
|
Beitler BG, Sieberer J, Islam W, McDonald C, Yu K, Tommasini SM, Fulkerson JP. The Morphologic Patella Entry Point Into the Proximal Trochlea Is More Lateral in Recurrent Dislocators Than Controls as Measured by Entry Point-Trochlear Groove Angle. Arthroscopy 2024:S0749-8063(24)00308-6. [PMID: 38697328 DOI: 10.1016/j.arthro.2024.04.013] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 10/14/2023] [Revised: 03/31/2024] [Accepted: 04/09/2024] [Indexed: 05/04/2024]
Abstract
PURPOSE To create a metric for evaluating the degree of laterality of the patella's entry into the trochlea, the entry point-trochlear groove (EP-TG) angle, and to evaluate if this laterality is associated with recurrent patella instability. METHODS The time frame of the study was January 2020 to February 2023. The inclusion criteria were patients treated by the senior author (J.P.F.) (with the exception of 2 patients who were treated by another provider at the institution who was aware of the study) who have been diagnosed with recurrent atraumatic patellar dislocations. Controls without knee pathology were selected from the New Mexico Decedent Imaging Database (NMDID). Simpleware ScanIP was used to create 3-dimensional (3D) models of the distal femurs from computed tomography scans. Anteroposterior images of these 3D models were uploaded to a custom EP-TG angle measuring tool. Three measurers used the tool to measure the EP-TG angle of the distal femurs. RESULTS Twenty-eight patients were included for the recurrent dislocator group. Twenty-four decedents from the NMDID were selected for the control group, each with a left or right knee chosen randomly for measurement. A 1-sided Mann-Whitney U test, used to evaluate whether the recurrent dislocators had higher EP-TG angle values, yielded a P value <.001, demonstrating a high level of significance. A Bayesian mixed-effect model, used to determine how different the EP-TG angles are between the 2 groups, gave a posterior predictive interval of [11.93°, 19.12°] for the EP-TG angle shift of dislocators. The intraclass correlation coefficient was 0.648. CONCLUSIONS The morphologic entry point of the patella into the proximal trochlea is more lateral in recurrent patella dislocators than in controls. This increased laterality can be measured by the EP-TG angle, which may be useful information for optimizing treatment of recurrent patella instability. LEVEL OF EVIDENCE Level III, case control study.
Collapse
Affiliation(s)
- Brian G Beitler
- Department of Orthopedics, Icahn School of Medicine at Mount Sinai, New York, New York, U.S.A..
| | - Johannes Sieberer
- Department of Orthopaedics and Rehabilitation, Yale School of Medicine, New Haven, Connecticut, U.S.A
| | - Wasif Islam
- Department of Orthopaedics and Rehabilitation, Yale School of Medicine, New Haven, Connecticut, U.S.A
| | - Curtis McDonald
- Department of Statistics and Data Science, Yale University, New Haven, Connecticut, U.S.A
| | - Kristin Yu
- Department of Orthopedic Surgery, Mayo Clinic Orthopedic Surgery, Rochester, Minnesota, U.S.A
| | - Steven M Tommasini
- Department of Orthopaedics and Rehabilitation, Yale School of Medicine, New Haven, Connecticut, U.S.A
| | - John P Fulkerson
- Department of Orthopaedics and Rehabilitation, Yale School of Medicine, New Haven, Connecticut, U.S.A
| |
Collapse
|
3
|
Yan M, Liang T, Zhao H, Bi Y, Wang T, Yu T, Zhang Y. Model Properties and Clinical Application in the Finite Element Analysis of Knee Joint: A Review. Orthop Surg 2024; 16:289-302. [PMID: 38174410 PMCID: PMC10834231 DOI: 10.1111/os.13980] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 08/22/2023] [Revised: 11/21/2023] [Accepted: 12/04/2023] [Indexed: 01/05/2024] Open
Abstract
The knee is the most complex joint in the human body, including bony structures like the femur, tibia, fibula, and patella, and soft tissues like menisci, ligaments, muscles, and tendons. Complex anatomical structures of the knee joint make it difficult to conduct precise biomechanical research and explore the mechanism of movement and injury. The finite element model (FEM), as an important engineering analysis technique, has been widely used in many fields of bioengineering research. The FEM has advantages in the biomechanical analysis of objects with complex structures. Researchers can use this technology to construct a human knee joint model and perform biomechanical analysis on it. At the same time, finite element analysis can effectively evaluate variables such as stress, strain, displacement, and rotation, helping to predict injury mechanisms and optimize surgical techniques, which make up for the shortcomings of traditional biomechanics experimental research. However, few papers introduce what material properties should be selected for each anatomic structure of knee FEM to meet different research purposes. Based on previous finite element studies of the knee joint, this paper summarizes various modeling strategies and applications, serving as a reference for constructing knee joint models and research design.
Collapse
Affiliation(s)
- Mingyue Yan
- Department of Orthopedics, The Affiliated Hospital of Qingdao University, Qingdao, China
- Institute of Sports Medicine and Health, Qingdao University, Qingdao, China
| | - Ting Liang
- Department of Orthopedics, The Affiliated Hospital of Qingdao University, Qingdao, China
- Institute of Sports Medicine and Health, Qingdao University, Qingdao, China
| | - Haibo Zhao
- Department of Orthopedics, The Affiliated Hospital of Qingdao University, Qingdao, China
- Institute of Sports Medicine and Health, Qingdao University, Qingdao, China
| | - Yanchi Bi
- Department of Orthopedics, The Affiliated Hospital of Qingdao University, Qingdao, China
- Institute of Sports Medicine and Health, Qingdao University, Qingdao, China
| | - Tianrui Wang
- Department of Orthopedics, The Affiliated Hospital of Qingdao University, Qingdao, China
| | - Tengbo Yu
- Institute of Sports Medicine and Health, Qingdao University, Qingdao, China
- Department of Orthopedic Surgery, Qingdao Hospital, University of Health and Rehabilitation Sciences (Qingdao Municipal Hospital), Qingdao, China
| | - Yingze Zhang
- Department of Orthopedics, The Third Hospital of Hebei Medical University, Shijiazhuang, China
| |
Collapse
|
4
|
Ryan PC, Ross BJ, Stamm MA, Sherman WF, Heard WMR, Mulcahey MK. Concomitant Tibial Tubercle Osteotomy Reduces the Risk of Revision Surgery After Medial Patellofemoral Ligament Reconstruction for the Treatment of Patellar Instability. Arthroscopy 2023; 39:2037-2045.e1. [PMID: 36804459 DOI: 10.1016/j.arthro.2023.02.006] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 06/24/2022] [Revised: 01/23/2023] [Accepted: 02/02/2023] [Indexed: 02/23/2023]
Abstract
PURPOSE To compare postoperative complication rates between patients who underwent medial patellofemoral ligament reconstruction (MPFLR) and those who underwent MPFLR with tibial tubercle osteotomy (TTO) in a large-scale study. A secondary goal was to investigate demographic associations with outcomes. METHODS Patients who underwent MPFLR (n = 3,480) or MPFLR-TTO (n = 615) for patellar instability were identified in the PearlDiver database. Rates of surgery for infection, procedures for knee stiffness, patellar fracture, and revision MPFLR within 2 years postoperatively were compared using multivariable logistic regression. Demographic associations with outcomes were also investigated. RESULTS The MPFLR-TTO cohort exhibited a significantly lower rate of revision surgery at 2 years (0.8% vs 1.9%; odds ratio [OR], 0.33; 95% confidence interval [CI], 0.10-0.80; P = .036) when compared with the MPFLR group. Independent of index procedure, patients younger than 21 years had significantly lower rates of requiring procedures for knee stiffness (OR, 0.35; 95% CI, 0.22-0.54; P < .001) and any complication at 2 years (OR, 0.59; 95% CI, 0.45-0.78; P < .001) when compared with older patients. Male patients displayed a significantly lower rate of requiring procedures for knee stiffness at 2 years than female patients (OR, 0.46; 95% CI, 0.25-0.78; P = .007). Tobacco use was associated with a significantly higher rate of postoperative infection at 2 years (OR, 2.35; 95% CI, 1.00-5.38; P = .046). CONCLUSIONS The MPFLR cohort exhibited higher rates of revision surgery at 2 years compared with the MPFLR-TTO cohort. Patient age under 21 years was associated with lower rates of any complication and requiring procedures for knee stiffness, male sex was associated with a lower rate of requiring procedures for knee stiffness, and tobacco use was associated with a higher rate of surgery for postoperative infection. This information can assist surgeons when counseling patients before these procedures are performed. LEVEL OF EVIDENCE Level III, retrospective, comparative prognostic trial.
Collapse
Affiliation(s)
- Patrick C Ryan
- Tulane University School of Medicine, New Orleans, Louisiana, U.S.A
| | - Bailey J Ross
- Department of Orthopaedic Surgery, Emory School of Medicine, Atlanta, Georgia, U.S.A
| | - Michaela A Stamm
- Department of Orthopaedic Surgery, Tulane University School of Medicine, New Orleans, Louisiana, U.S.A
| | - William F Sherman
- Department of Orthopaedic Surgery, Tulane University School of Medicine, New Orleans, Louisiana, U.S.A
| | - Wendell M R Heard
- Department of Orthopaedic Surgery, Tulane University School of Medicine, New Orleans, Louisiana, U.S.A
| | - Mary K Mulcahey
- Department of Orthopaedic Surgery, Tulane University School of Medicine, New Orleans, Louisiana, U.S.A..
| |
Collapse
|
5
|
Elias JJ, Cosgarea AJ, Tanaka MJ. Adding Tibial Tuberosity Medialization to Medial Patellofemoral Ligament Reconstruction Reduces Lateral Patellar Maltracking During Multidirectional Motion in a Computational Simulation Model. Arthrosc Sports Med Rehabil 2023; 5:100753. [PMID: 37645404 PMCID: PMC10461214 DOI: 10.1016/j.asmr.2023.100753] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/21/2022] [Accepted: 05/29/2023] [Indexed: 08/31/2023] Open
Abstract
Purpose To determine whether adding tibial tuberosity medialization to medial patellofemoral ligament (MPFL) reconstruction reduces lateral patellar maltracking during a dynamic multidirectional activity and to investigate when medial patellofemoral contact pressures are elevated during daily activities, such as squatting. Methods Seven computational models representing knees with patellar instability, including lateral patellar maltracking, were evaluated following simulated MPFL reconstruction (bisect offset index > .75). Tibial tuberosity medialization was added to MPFL reconstruction for each model. Patellar tracking during multidirectional motion was evaluated by simulating pivot landing. Analysis of pivoting focused on early flexion (5° to 40°). Patellofemoral contact pressures during daily function were evaluated by simulating knee squatting. Data were analyzed with paired comparisons between MPFL reconstruction with and without tuberosity medialization. Results The patella dislocated during pivoting for 2 models with an isolated MPFL reconstruction and for 1 model including tibial tuberosity medialization. Adding tibial tuberosity medialization to MPFL reconstruction significantly decreased bisect offset index by ∼0.1 from 5° to 40° (P < .03). For knee squatting, medializing the tibial tuberosity significantly increased maximum medial contract pressure by ∼0.5 MPa from 30° to 85° (P < .05) but did not significantly influence maximum lateral pressure. Conclusions In this study of simulated multidirectional motion, MPFL reconstruction did not sufficiently constrain the patella for some knees. Adding tibial tuberosity medialization to MPFL reconstruction in these models reduced lateral patellar maltracking during multidirectional motion but increased pressure applied to medial cartilage during squatting. Clinical Relevance After establishing the influence of tibial tuberosity medialization on patellar maltracking for an idealized population, as was done in the current study, future simulation studies can be performed to better determine the anatomical characteristics of patients for whom tibial tuberosity medialization is needed to reduce the risk of postoperative patellar maltracking.
Collapse
Affiliation(s)
- John J. Elias
- Department of Health Sciences, Cleveland Clinic Akron General, Cleveland, Ohio, U.S.A
| | - Andrew J. Cosgarea
- Department of Orthopaedic Surgery, Johns Hopkins University, Baltimore, Maryland, U.S.A
| | - Miho J. Tanaka
- Department of Orthopaedic Surgery, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts, U.S.A
| |
Collapse
|