PDF herunterladen
J Knee Surg 2025; 38(08): 422-431
DOI: 10.1055/a-2525-4622
Original Article

Prediction of the Serial Alignment Change after Opening-Wedge High Tibial Osteotomy Based on Coronal Plane Alignment of the Knee Using Machine Learning Algorithm

Joon Hee Cho
1   Department of Orthopedic Surgery, Seoul National University College of Medicine, Seoul National University Bundang Hospital, Bundang-gu, Seongnam-si, Gyeonggi-do, South Korea
,
Hee Seung Nam
1   Department of Orthopedic Surgery, Seoul National University College of Medicine, Seoul National University Bundang Hospital, Bundang-gu, Seongnam-si, Gyeonggi-do, South Korea
,
Seong Yun Park
1   Department of Orthopedic Surgery, Seoul National University College of Medicine, Seoul National University Bundang Hospital, Bundang-gu, Seongnam-si, Gyeonggi-do, South Korea
,
Jade Pei Yuik Ho
1   Department of Orthopedic Surgery, Seoul National University College of Medicine, Seoul National University Bundang Hospital, Bundang-gu, Seongnam-si, Gyeonggi-do, South Korea
,
Yong Seuk Lee
1   Department of Orthopedic Surgery, Seoul National University College of Medicine, Seoul National University Bundang Hospital, Bundang-gu, Seongnam-si, Gyeonggi-do, South Korea
› Institutsangaben
Funding This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT), number 2021R1A2C1092657.
› Weitere Informationen
   Lizenzen und Reprints

Abstract

Categorization of alignment into phenotypes can be useful for predicting and analyzing postoperative alignment changes after opening-wedge high tibial osteotomy (OWHTO). The purposes of this study were to (1) develop a machine learning model for the predicting the Coronal Plane Alignment of the Knee (CPAK) phenotypes of final alignment after OWHTO, and (2) analyze predictive factors for final alignment phenotypes. Data were retrospectively collected from 163 knees that underwent OWHTO between March 2014 and December 2019. Each data were assessed at three time points: preoperatively, at 3 months postoperatively, and the final follow-up. Constitutional alignment was also evaluated. Machine learning models were developed using two independent feature sets consisting of serial radiologic parameters and CPAK phenotypes. The area under the receiver–operating characteristic curve (AUC) was used as a primary metric to determine the best model. To evaluate the feature importance, Shapley additive explanation (SHAP) analysis was also performed on the best model. Multilayer perceptron (MLP) was the best prediction model, with the highest AUC of 0.867 based on radiologic parameters and 0.783 based on CPAK phenotypes. Joint line obliquity (JLO) at 3 months postoperatively was the most important factor among the radiologic parameters for predicting the final CPAK phenotypes. The features of constitutional and preoperative alignments also contributed, although the features of alignments at 3 months postoperatively were the highest contributing predictors. In conclusion, the developed machine learning models of the MLP showed excellent performance in predicting the final CPAK phenotypes after OWHTO. Postoperative JLO was the most important radiologic parameter for predicting the final alignment. The combination of features of the constitutional, preoperative, and postoperative periods enabled high accuracy and performance in predicting the final alignment.

A retrospective cohort study with the level of evidence as level III.

Keywords

knee - high tibial osteotomy - CPAK classification - machine learning - prediction

Ethical Approval

Institutional Review Board approval was obtained before performing the study (B-2302-811-108).


Consent to Participate

Every author consented to participate.


Consent for Publication

Every author consented for publication.


Authors' Contributions

J.H.C. and Y.S.L. participated in study design and drafted the manuscript. J.H.C., S.Y.P., and J.P.Y.H. performed the statistical analysis. J.H.C., S.Y.P., H.S.N., and J.P.Y.H. collected and interpreted the data. Y.S.L. conceived of the study, participated in coordination, and helped to draft the manuscript. All authors read and approved the final manuscript.


Availability of Data and Materials

The data that support the findings of this study are available from the corresponding author upon reasonable request.




Publikationsverlauf

Eingereicht: 13. November 2024

Angenommen: 26. Januar 2025

Accepted Manuscript online:
27. Januar 2025

Artikel online veröffentlicht:
21. Februar 2025

© 2025. Thieme. All rights reserved.

Thieme Medical Publishers, Inc.
333 Seventh Avenue, 18th Floor, New York, NY 10001, USA

 

  • References

  • 1 Agneskirchner JD, Hurschler C, Wrann CD, Lobenhoffer P. The effects of valgus medial opening wedge high tibial osteotomy on articular cartilage pressure of the knee: a biomechanical study. Arthroscopy 2007; 23 (08) 852-861
    CrossrefPubMedSuche in Google Scholar
  • 2 El-Azab HM, Morgenstern M, Ahrens P, Schuster T, Imhoff AB, Lorenz SG. Limb alignment after open-wedge high tibial osteotomy and its effect on the clinical outcome. Orthopedics 2011; 34 (10) e622-e628
    PubMedSuche in Google Scholar
  • 3 Bito H, Takeuchi R, Kumagai K. et al. A predictive factor for acquiring an ideal lower limb realignment after opening-wedge high tibial osteotomy. Knee Surg Sports Traumatol Arthrosc 2009; 17 (04) 382-389
    CrossrefPubMedSuche in Google Scholar
  • 4 Lee YS, Lee BK, Kwon JH. et al. Serial assessment of weight-bearing lower extremity alignment radiographs after open-wedge high tibial osteotomy. Arthroscopy 2014; 30 (03) 319-325
    CrossrefPubMedSuche in Google Scholar
  • 5 MacDessi SJ, Griffiths-Jones W, Harris IA, Bellemans J, Chen DB. Coronal Plane Alignment of the Knee (CPAK) classification. Bone Joint J 2021; 103-B (02) 329-337
    CrossrefPubMedSuche in Google Scholar
  • 6 Lee OS, Lee SH, Lee YS. Does coronal knee and ankle alignment affect recurrence of the varus deformity after high tibial osteotomy?. Knee Surg Relat Res 2018; 30 (04) 311-318
    CrossrefPubMedSuche in Google Scholar
  • 7 Samuel AL. Some studies in machine learning using the game of checkers. II—Recent progress. IBM J Res Develop 1967; 11 (06) 601-617
    CrossrefPubMedSuche in Google Scholar
  • 8 Jeong HW, Kim M, Choi HG, Park SY, Lee YS. Development of a machine learning model to predict lateral hinge fractures by analyzing patient factors before open wedge high tibial osteotomy. Knee Surg Sports Traumatol Arthrosc 2023; 31 (08) 3070-3078
    CrossrefPubMedSuche in Google Scholar
  • 9 Fontana MA, Lyman S, Sarker GK, Padgett DE, MacLean CH. Can machine learning algorithms predict which patients will achieve minimally clinically important differences from total joint arthroplasty?. Clin Orthop Relat Res 2019; 477 (06) 1267-1279
    CrossrefPubMedSuche in Google Scholar
  • 10 Kim JE, Kim DH, Lee JI. et al. Difference of preoperative varus-valgus stress radiograph is effective for the correction accuracy in the preoperative planning during open-wedge high tibial osteotomy. Knee Surg Sports Traumatol Arthrosc 2021; 29 (04) 1035-1044
    CrossrefPubMedSuche in Google Scholar
  • 11 Hsu CE, Chen CP, Wang SP, Huang JT, Tong KM, Huang KC. Validation and modification of the Coronal Plane Alignment of the Knee classification in the Asian population. Bone Jt Open 2022; 3 (03) 211-217
    CrossrefPubMedSuche in Google Scholar
  • 12 Lee YS, Moon GH. Comparative analysis of osteotomy accuracy between the conventional and devised technique using a protective cutting system in medial open-wedge high tibial osteotomy. J Orthop Sci 2015; 20 (01) 129-136
    CrossrefPubMedSuche in Google Scholar
  • 13 Seo H, Lim D, Jang YW. et al. Distribution of the force in the knee joint during daily activities after open wedge high tibial osteotomy: a rationale for the proper postoperative management. J Knee Surg 2020; 33 (02) 158-166
    Thieme ConnectPubMedSuche in Google Scholar
  • 14 Lee OS, Lee ES, Lee YS. Disparity between preoperative target correction amount and postoperative correction amount in open wedge high tibial osteotomy. Knee Surg Relat Res 2019; 31 (02) 126-131
    CrossrefPubMedSuche in Google Scholar
  • 15 Lundberg SM, Lee SI. A unified approach to interpreting model predictions. Adv Neural Inf Process Syst 2017; 30: 4765-4774
    PubMedSuche in Google Scholar
  • 16 Kang BY, Lee DK, Kim HS, Wang JH. How to achieve an optimal alignment in medial opening wedge high tibial osteotomy?. Knee Surg Relat Res 2022; 34 (01) 3
    CrossrefPubMedSuche in Google Scholar
  • 17 Mehrjerd A, Rezaei H, Eslami S, Ratna MB, Khadem Ghaebi N. Internal validation and comparison of predictive models to determine success rate of infertility treatments: a retrospective study of 2485 cycles. Sci Rep 2022; 12 (01) 7216
    CrossrefPubMedSuche in Google Scholar
  • 18 Wang CW, Kuo CY, Chen CH, Hsieh YH, Su EC. Predicting clinical pregnancy using clinical features and machine learning algorithms in in vitro fertilization. PLoS One 2022; 17 (06) e0267554
    CrossrefPubMedSuche in Google Scholar
  • 19 Wu Q, Nasoz F, Jung J, Bhattarai B, Han MV. Machine learning approaches for fracture risk assessment: a comparative analysis of genomic and phenotypic data in 5130 older men. Calcif Tissue Int 2020; 107 (04) 353-361
    CrossrefPubMedSuche in Google Scholar
  • 20 Akamatsu Y, Kumagai K, Kobayashi H, Tsuji M, Saito T. Effect of increased coronal inclination of the tibial plateau after opening-wedge high tibial osteotomy. Arthroscopy 2018; 34 (07) 2158-2169.e2
    CrossrefPubMedSuche in Google Scholar
  • 21 Lee DK, Wang JH, Won Y. et al. Preoperative latent medial laxity and correction angle are crucial factors for overcorrection in medial open-wedge high tibial osteotomy. Knee Surg Sports Traumatol Arthrosc 2020; 28 (05) 1411-1418
    CrossrefPubMedSuche in Google Scholar
  • 22 Ogawa H, Matsumoto K, Ogawa T, Takeuchi K, Akiyama H. Preoperative varus laxity correlates with overcorrection in medial opening wedge high tibial osteotomy. Arch Orthop Trauma Surg 2016; 136 (10) 1337-1342
    CrossrefPubMedSuche in Google Scholar
  • 23 Choi NH. Editorial Commentary: Knee preoperative medial laxity may result in overcorrection or varus recurrence after open-wedge high tibial osteotomy. Arthroscopy 2022; 38 (05) 1555-1556
    CrossrefPubMedSuche in Google Scholar
  • 24 Song JH, Bin SI, Kim JM, Lee BS. What is an acceptable limit of joint-line obliquity after medial open wedge high tibial osteotomy? Analysis based on midterm results. Am J Sports Med 2020; 48 (12) 3028-3035
    CrossrefPubMedSuche in Google Scholar
  • 25 Sohn S, Koh IJ, Kim MS, In Y. Risk factors and preventive strategy for excessive coronal inclination of tibial plateau following medial opening-wedge high tibial osteotomy. Arch Orthop Trauma Surg 2022; 142 (04) 561-569
    CrossrefPubMedSuche in Google Scholar
  • 26 Kim JS, Lim JK, Choi HG. et al. Excessively increased joint-line obliquity after medial opening-wedge high tibial osteotomy is associated with inferior radiologic and clinical outcomes: what is permissible joint-line obliquity?. Arthroscopy 2022; 38 (06) 1904-1915
    CrossrefPubMedSuche in Google Scholar
  • 27 Ji HM, Han J, Jin DS, Seo H, Won YY. Kinematically aligned TKA can align knee joint line to horizontal. Knee Surg Sports Traumatol Arthrosc 2016; 24 (08) 2436-2441
    CrossrefPubMedSuche in Google Scholar
  • 28 Park SB, Lee YS. Association of the joint line orientation angle of the contralateral limb with the alignment change of the unilateral and bilateral opening-wedge high tibial osteotomy. Knee Surg Sports Traumatol Arthrosc 2023; 31 (04) 1593-1602
    CrossrefPubMedSuche in Google Scholar