Validation of a model-based measurement of the minimum insert thickness of knee prostheses: a retrieval study

E A van IJsseldijk; M K Harman; J Luetzner; E R Valstar; B C Stoel; R G H H Nelissen; B L Kaptein

doi:10.1302/2046-3758.310.2000304

Validation of a model-based measurement of the minimum insert thickness of knee prostheses: a retrieval study

Bone Joint Res. 2014 Oct;3(10):289-96. doi: 10.1302/2046-3758.310.2000304.

Authors

E A van IJsseldijk¹, M K Harman², J Luetzner³, E R Valstar¹, B C Stoel⁴, R G H H Nelissen¹, B L Kaptein¹

Affiliations

¹ Leiden University Medical Center, Department of Orthopaedics, Biomechanics and Imaging Group, PO 9600, 2300 RC, Leiden, The Netherlands.
² Clemson University, Department of Bioengineering, 301 Rhodes Engineering Research Center, Clemson, 29634, USA.
³ University Hospital Carl Gustav Carus, Department of Orthopaedic and Trauma Surgery, TU Dresden, Fetscherstrasse 74, 01307 Dresden, Germany.
⁴ Leiden University Medical Center, Department of Radiology, Div. of Image Processing, PO 9600, 2300 RC, Leiden, The Netherlands.

Abstract

Introduction: Wear of polyethylene inserts plays an important role in failure of total knee replacement and can be monitored in vivo by measuring the minimum joint space width in anteroposterior radiographs. The objective of this retrospective cross-sectional study was to compare the accuracy and precision of a new model-based method with the conventional method by analysing the difference between the minimum joint space width measurements and the actual thickness of retrieved polyethylene tibial inserts.

Method: Before revision, the minimum joint space width values and their locations on the insert were measured in 15 fully weight-bearing radiographs. These measurements were compared with the actual minimum thickness values and locations of the retrieved tibial inserts after revision.

Results: The mean error in the model-based minimum joint space width measurement was significantly smaller than the conventional method for medial condyles (0.50 vs 0.94 mm, p < 0.01) and for lateral condyles (0.06 vs 0.34 mm, p = 0.02). The precision (standard deviation of the error) of the methods was similar (0.84 vs 0.79 mm medially and both 0.46 mm laterally). The distance between the true minimum joint space width locations and the locations from the model-based measurements was less than 10 mm in the medial direction in 12 cases and less in the lateral direction in 13 cases.

Conclusion: The model-based minimum joint space width measurement method is more accurate than the conventional measurement with the same precision. Cite this article: Bone Joint Res 2014;3:289-96.

Keywords: Knee replacement; Measurement accuracy; Model-based wear measurements; Polyethylene wear; mSJW.

Grants and funding

Funding for this study was provided by the Dutch Arthritis Foundation (no. 08-1-303), faculty start-up funds from the Department of Bioengineering at Clemson University and by the Anna Fonds foundation (no. O2013/05). Prof Dr Harman acknowledges payment for educational presentations from DJO Global to Clemson University that is not related to this paper. Dr Med Luetzner reports grants to the University Hospital Carl Gustav Carus from Aesculap, Mathys, Smith&Nephew, Stryker and Zimmer, and also acknowledges payment for lectures from Aesculap and Mathys, neither of which is related to this paper.