Elsevier

Journal of Biomechanics

Volume 48, Issue 5, 18 March 2015, Pages 832-841
Journal of Biomechanics

A survey of micro-finite element analysis for clinical assessment of bone strength: The first decade

https://doi.org/10.1016/j.jbiomech.2014.12.024Get rights and content

Abstract

Micro-Finite Element (micro-FE) analysis is now widely used in biomedical research as a tool to derive bone mechanical properties as they relate to its microstructure. With the development of in vivo high-resolution peripheral quantitative CT (HR-pQCT) scanners, it can now be applied to analyze bone in-vivo in the peripheral skeleton. In this survey, the results of several experimental and clinical studies are summarized that addressed the feasibility of this approach to predict bone strength in-vivo. Specific questions that will be addressed are: how accurate are strength predictions based on micro-FE; how reproducible are the results; and, is it a better predictor of bone fracture risk than DXA based measures? Based on results of experimental studies, it is first concluded that micro-FE based on HR-pQCT images can accurately predict the strength of the distal radius during a fall on the outstretched hand using either linear elastic analysis, implementing a ‘Pistoia criterion’ or similar criterion in combination with an ‘effective’ Young’s modulus or using non-linear analyses. When evaluating results of clinical reproducibility studies, it is concluded that for single-center studies, errors at the radius are less than 4.4% and 3.7% and at the tibia less than 3.6% and 2.3% for stiffness and strength, respectively. In multicenter trials, however, these errors can be increased by some 1.8% and 1.4% for stiffness and strength, respectively. Finally, based on the results of large cohort studies, it is concluded that micro-FE calculated stiffness better separates cases from controls than bone density parameters for subjects with fragility fractures at any site, but not for subjects with only radius fractures. In this latter case, however, combinations of micro-FE derived parameters can significantly improve the separation.

Section snippets

Rik Huiskes and micro-FE: A historical perspective (by Bert van Rietbergen)

The title of this paper is loosely based on that of a paper that Rik Huiskes wrote some 30 years ago, at the time that finite element analyses became a well-accepted tool in the field of biomechanics (Huiskes and Chao, 1983). Ten years later, this paper was followed by a second one: “From structure to process, from organ to cell: recent developments of FE-analysis in orthopaedic biomechanics” (Huiskes and Hollister, 1993). Even though micro- finite element analyses had only just been developed

The methodology and the issues

Due to the fact that only one type of HR-pQCT machine has been used so far, clinical application of micro-FE analysis is a relatively well standardized technique. In most of the clinical studies, a standard image processing workflow is used and the segmented image obtained in this way is used as the geometry of the micro-FE model. The micro-FE analysis itself, however, is less standardized. Different software packages and different procedures have been used, and different output parameters are

Image generation and processing

With the standard workflow, a 9 mm region is scanned at the distal radius or tibia at an isotropic resolution of 82 μm, thus generating a stack of 110 slices. The scan region is chosen at a specific distance from the joint surface as measured from a scout view and all other image settings are standardized as well. The image processing procedure is also standardized, although over time optimized versions have been developed. With the most common workflow, the grey-level image obtained from the

Micro-FE modeling

Typically, micro-FE models are generated directly from the segmented images using a voxel conversion approach. With this approach, voxels representing bone tissue are converted to equally sized brick elements whereas voxels representing the bone marrow are ignored. For typical HR-pQCT scans, the voxel conversion approach will generate models in the range of 1–4 million elements for the radius and in the range of 4–9 million elements for the tibia. This large number complicates the use of

How accurate are the results?

The accuracy of the micro-FE approach to analyze bone stiffness and strength has been investigated in a number of ex-vivo experimental studies, although all of these studies focused on the radius only. The set-up of all these studies is more-or-less the same: a large number of cadaver bones is collected and scanned with the HR-pQCT device using the same settings as used in clinical studies. Following, the bones are compressed to failure to measure their strength and, if possible, their

How reproducible are the results?

Whereas the studies on cadaver bones provide some information on reproducibility of strength predictions, the actual reproducibility in a clinical setting is dependent on more factors. Such factors include misalignment errors, movement artifacts, (cross-) calibration errors and errors related to differences in image analyses by different operators. In this section, several studies that investigated the reproducibility of micro-FE results in a clinical setting are summarized.

MacNeil and Boyd

Is it better than DXA?

One of the major findings in all validations studies is that micro-FE results better predict bone failure load measured in an experiment than any DXA or other bone density-based parameter. In the original paper by Pistoia et al. (2004) a coefficient of determination of 0.66 was reported when comparing the micro-FE estimated and measured failure loads, whereas those for BMC and aBMD were much less (0.48 and 0.31, respectively). In the study by Varga et al. (2010), these differences were less

Alternative techniques and emerging developments

In this survey, only results obtained with the most common micro-FE approach, that uses the voxel conversion technique, were discussed. Other FE approaches have been tested as well, some of them even in clinical studies. Among these are techniques that can significantly reduce the computer time needed to solve the FE problem by representing the actual architecture by a simpler one built of plate and beam elements (Liu et al., 2012). Also, recently, FE approaches incorporating cohesive elements

Conclusion and recommendations

Whereas this review made clear that much information about the accuracy and reproducibility of the micro-FE technique now is available, it also made clear that there are several important technical issues that need to be addressed. One of the major issues with accuracy will be the calibration of the models. Presently, the tissue Young’s modulus is used as a sort of calibration parameter. As a result of this, using different segmentation settings, using different boundary conditions, or using a

Disclosures

Bert van Rietbergen is a consultant for Scanco Medical AG.

References (70)

  • J.A. MacNeil et al.

    Accuracy of high-resolution peripheral quantitative computed tomography for measurement of bone quality

    Med. Eng. Phys

    (2007)
  • J.A. Macneil et al.

    Bone strength at the distal radius can be estimated from high-resolution peripheral quantitative computed tomography and the finite element method

    Bone

    (2008)
  • J.A. MacNeil et al.

    Improved reproducibility of high-resolution peripheral quantitative computed tomography for measurement of bone quality

    Med. Eng. Phys

    (2008)
  • T.L. Mueller et al.

    Computational finite element bone mechanics accurately predicts mechanical competence in the human radius of an elderly population

    Bone

    (2011)
  • L. Mulder et al.

    Determination of vertebral and femoral trabecular morphology and stiffness using a flat-panel C-arm-based CT approach

    Bone

    (2012)
  • G.L. Niebur et al.

    High-resolution finite element models with tissue strength asymmetry accurately predict failure of trabecular bone

    J. Biomech.

    (2000)
  • J.B. Pialat et al.

    Local topological analysis at the distal radius by HR-pQCT: application to in vivo bone microarchitecture and fracture assessment in the OFELY study

    Bone

    (2012)
  • W. Pistoia et al.

    Estimation of distal radius failure load with micro-finite element analysis models based on three-dimensional peripheral quantitative computed tomography images

    Bone

    (2002)
  • W. Pistoia et al.

    Image-based micro-finite-element modeling for improved distal radius strength diagnosis: moving from bench to bedside

    J. Clin. Densitom.

    (2004)
  • J.D. Schipilow et al.

    Bone micro-architecture, estimated bone strength, and the muscle-bone interaction in elite athletes: an HR-pQCT study

    Bone

    (2013)
  • D. Ulrich et al.

    Finite element analysis of trabecular bone structure: a comparison of image-based meshing techniques

    J. Biomech.

    (1998)
  • A. Ural et al.

    A new fracture assessment approach coupling HR-pQCT imaging and fracture mechanics-based finite element modeling

    J. Biomech.

    (2013)
  • B. van Rietbergen et al.

    A new method to determine trabecular bone elastic properties and loading using micromechanical finite-element models

    J. Biomech.

    (1995)
  • P. Varga et al.

    HR-pQCT based FE analysis of the most distal radius section provides an improved prediction of Colles’ fracture load in vitro

    Bone

    (2010)
  • P. Varga et al.

    Assessment of volume fraction and fabric in the distal radius using HR-pQCT

    Bone

    (2009)
  • N. Vilayphiou et al.

    Finite element analysis performed on radius and tibia HR-pQCT images and fragility fractures at all sites in postmenopausal women

    Bone

    (2010)
  • Adams M.F., Bayraktar H.H., Keaveny T.M., Papadopoulos P., 2004. Ultrascalable implicit finite element analyses in...
  • P. Arbenz et al.

    A scalable multi-level preconditioner for matrix-free µ-finite element analysis of human bone structures

    Int. J. Numer. Methods Eng.

    (2008)
  • C. Bekas et al.

    Extreme scalability challenges in micro-finite element simulations of human bone

    Concurrency Comput.: Pract. Exp.

    (2010)
  • S. Boutroy et al.

    In vivo assessment of trabecular bone microarchitecture by high-resolution peripheral quantitative computed tomography

    J. Clin. Endocrinol. Metab.

    (2005)
  • S. Boutroy et al.

    Finite element analysis based on in vivo HR-pQCT images of the distal radius is associated with wrist fracture in postmenopausal women

    J. Bone Miner. Res.

    (2008)
  • A.J. Burghardt et al.

    A longitudinal HR-pQCT study of alendronate treatment in postmenopausal women with low bone density: relations among density, cortical and trabecular microarchitecture, biomechanics, and bone turnover

    J. Bone Miner. Res.

    (2010)
  • A.J. Burghardt et al.

    Multicenter precision of cortical and trabecular bone quality measures assessed by high-resolution peripheral quantitative computed tomography

    J. Bone Miner. Res.

    (2013)
  • A.C. Cheung et al.

    Reproducibility of trabecular structure analysis using flat-panel volume computed tomography

    Skelet. Radiol.

    (2009)
  • T. Chevalley et al.

    Fractures in healthy females followed from childhood to early adulthood are associated with later menarcheal age and with impaired bone microstructure at peak bone mass

    J. Clin. Endocrinol. Metab.

    (2012)
  • Cited by (71)

    View all citing articles on Scopus
    View full text