Elsevier

Morphologie

Volume 103, Issue 343, December 2019, Pages 180-186
Morphologie

Original article
Predicting bone strength from CT data: Clinical applications

https://doi.org/10.1016/j.morpho.2019.09.007Get rights and content

Highlights

  • On average the strength predicted with finite element models (QCT-FE) based on computed tomography is 7% more accurate that that predicted with areal bone mineral density from Dual X-ray Absorptiometry (DXA-aBMD), the current standard of care, both in term of laboratory validation on cadaver bones and in terms of stratification accuracy on clinical cohorts of fractured and non-fractured women.

  • This improved accuracy makes QCT-FE superior to DXA-aBMD in clinical research and in clinical trials, where the its use can cut in half the number of patients to be enrolled to get the same statistical power..

  • For routine clinical use to decide who to treat with antiresorptive drugs, QCT-FE is more accurate but less cost-effective than DXA-aBMD, at least when the decision is on first line treatment like bisphosphonates.

  • The ability to predict skeletal strength from medical imaging is now opening a number of other applications, for example in paediatrics and oncology.

Summary

In this review we summarise over 15 years of research and development around the prediction of whole bones strength from Computed Tomography data, with particular reference to the prediction of the risk of hip fracture in osteoporotic patients. We briefly discuss the theoretical background, and then provide a summary of the laboratory and clinical validation of these modelling technologies. We then discuss the three current clinical applications: in clinical research, in clinical trials, and in clinical practice. On average the strength predicted with finite element models (QCT-FE) based on computed tomography is 7% more accurate that that predicted with areal bone mineral density from Dual X-ray Absorptiometry (DXA-aBMD), the current standard of care, both in term of laboratory validation on cadaver bones and in terms of stratification accuracy on clinical cohorts of fractured and non-fractured women. This improved accuracy makes QCT-FE superior to DXA-aBMD in clinical research and in clinical trials, where the its use can cut in half the number of patients to be enrolled to get the same statistical power. For routine clinical use to decide who to treat with antiresorptive drugs, QCT-FE is more accurate but less cost-effective than DXA-aBMD, at least when the decision is on first line treatment like bisphosphonates. But the ability to predict skeletal strength from medical imaging is now opening a number of other applications, for example in paediatrics and oncology.

Résumé

Dans cette revue, nous résumons plus de 15 ans de recherche et de développement autour de la prédiction de la résistance des os entiers à partir des données de la tomodensitométrie (CT), avec une référence particulière à la prédiction du risque de fracture de la hanche chez les patients ostéoporotiques. Nous discutons brièvement du contexte théorique, puis nous fournissons un résumé de la validation en laboratoire et clinique de ces technologies de modélisation. Nous discutons ensuite des trois applications cliniques actuelles : en recherche clinique, en essais cliniques et en pratique clinique. En moyenne, la force prévue par les modèles d’éléments finis (QCT-FE) basés sur la tomographie par ordinateur est 7 % plus précise que celle prévue par la densité minérale osseuse surfacique de l’absorbtiométrie biphotonique (DXA-aBMD), la norme clinique actuelle, tant en termes de validation en laboratoire sur les os de cadavres que de stratification sur des cohortes cliniques de femmes fracturées et non fracturées. Cette précision améliorée semble indiquer que la QCT-FE puisse être la supérieure à la DXA-aBMD en recherche clinique et dans les essais cliniques où son utilisation peut réduire de moitié le nombre de patients à recruter pour obtenir la même puissance statistique. Pour un usage clinique de routine afin de décider qui traiter avec des médicaments antirésorptifs, la QCT-FE est plus précise mais moins rentable que la DXA-aBMD, du moins lorsque la décision est sur le traitement de première ligne comme les bisphosphonates. Mais la capacité de prédire la résistance squelettique à partir de l’imagerie médicale ouvre maintenant un certain nombre d’autres applications, par exemple en pédiatrie et en oncologie.

Introduction

The bones forming our skeleton fracture when they are exposed to abnormal loads, or when their biomechanical competence is compromised. 77% of all unintentional injuries that occur annually in the United States are to the musculoskeletal system [1]. In total, 8.9 millions of bone fractures are associated every year to osteoporosis, worldwide [2]. Because the increased propensity to fall and overload, as well as the reduction of biomechanical competence of the skeleton are both associated with ageing, projections of prevalence are all quite concerning, because of the ageing population [3].

It is thus very important to develop reliable methods that can estimate the biomechanical strength of specific bones in the skeleton to defined loading conditions, and from those derive the risk of bone fracture associated to the reduced biomechanical competence.

In this review paper we provide a description of the most advanced technologies used for the non-invasive prediction of bone strength, and of the clinical applications that such technologies are finding.

Section snippets

Predicting bone strength

The intensity of the force required to fracture a human bone, when such bone is loaded in a given direction, is function of three biophysical determinants: the bone geometry, the biomechanical properties of the tissues forming the bone, and the loading condition.

Accuracy of bone strength predictions

Bone fracture due to overloading can occur at any anatomical site. Those associated to a reduced biomechanical competence (fragility fractures) occur most commonly at wrist, followed by the ankle, spine and hip. The hip fracture is the one involving the most severe effects, and thus it is the most studied; here below we provide quantifications of accuracy for such fracture.

The standard of care accepted in most countries uses as predictor of the risk of hip fracture the areal bone mineral

Use of QCT-FE in clinical research

On the basis of the results summarised in the previous sections, QCT-FE can predict whole bone strength as measured on cadaver bones with an accuracy around 85%, while DXA-aBMD shows an accuracy of only 77%–78%, 7 points less. On the Sheffield Cohort QCT-FE strength shows a stratification accuracy of 82%, DXA-aBMD only 75%.

Thus, any time it is clinically justified to perform a clinical CT, QCT-FE can provide a fairly accurate estimate of the whole bone biomechanical strength. This has enabled

Use of QCT-FE in clinical trials

Clinical trials of antiresorptive drugs are particularly challenging: the primary endpoint, bone fractures occur relatively rarely, and the observational time frame must be at least five years. Thus, it is commonly accepted as the use of a biomarker as a surrogate of such long-term clinical endpoint: the most common is DXA-aBMD, but some studies now use instead bone strength predicted with QCT-FE.

When the stratification accuracy of QCT-FE models as measured on the Sheffield cohort was used to

Use of QCT-FE in the clinical practice

The only routine use for QCT-FE which has been explored in term of cost-benefit is that of prognostic biomarker in the treatment planning of primary osteoporosis [29]. Here the analysis concluded that the method could become cost-effective, when compared to DXA-aBMD, only when it was used on a subset of “difficult” cases, and when the price-point for a QCT-FE was sufficiently low (US$ 100). This study, based on the UK healthcare costs, substantially confirmed the findings of another similar

Discussion

From this brief review of some of the recent literature, we can conclude that it is possible to predict with excellent accuracy the biomechanical strength of bones using finite element models informed by Quantitative Computer Tomography. These patient-specific models are now being used for clinical research, to improve clinical trials of new treatments, and in a few cases also in the routine clinical practice.

The simplicity, robustness, and reliability of this patient-specific modelling method

Disclosure of interest

This study was partially supported by the European Union H2020 grant STriTuVaD: “In Silico Trial for Tuberculosis Vaccine Development (grant ID 777123). The authors declare that they do not have any financial or personal relationships with other people or organisations that could have inappropriately influenced this study.

References (42)

  • K. Strømsøe et al.

    Bending strength of the femur in relation to non-invasive bone mineral assessment

    J Biomech

    (1995)
  • X.G. Cheng

    Assessment of the strength of proximal femur in vitro: relationship to femoral bone mineral density and femoral geometry

    Bone

    (1997)
  • E. Dall’Ara et al.

    A nonlinear QCT-based finite element model validation study for the human femur tested in two configurations in vitro

    Bone

    (2013)
  • E. Schileo et al.

    To what extent can linear finite element models of human femora predict failure under stance and fall loading configurations?

    J Biomech

    (2014)
  • M. Viceconti et al.

    Are spontaneous fractures possible? An example of clinical application for personalised, multiscale neuro-musculo-skeletal modelling

    J Biomech

    (2012)
  • X. Li et al.

    Developing CT based computational models of pediatric femurs

    J Biomech

    (2015)
  • H.-J. Kim

    Biomechanical advantages of robot-assisted pedicle screw fixation in posterior lumbar interbody fusion compared with freehand technique in a prospective randomized controlled trial-perspective for patient-specific finite element analysis

    Spine J

    (2017)
  • P. Zysset

    Comparison of proximal femur and vertebral body strength improvements in the FREEDOM trial using an alternative finite element methodology

    Bone

    (2015)
  • A. Sternheim

    Pathological fracture risk assessment in patients with femoral metastases using CT-based finite element methods. A retrospective clinical study

    Bone

    (2018)
  • Y. Kawabata et al.

    The risk assessment of pathological fracture in the proximal femur using a CT-based finite element method

    J Orthop Sci

    (2017)
  • United States Bone and Joint Initiative: The Burden of Musculoskeletal Diseases in the United States (BMUS), Third...
  • Cited by (6)

    View full text