0
Research Papers

# Finite Element Modeling of Resurfacing Hip Prosthesis: Estimation of Accuracy Through Experimental Validation

[+] Author and Article Information

Laboratorio di Tecnologia Medica, Istituto Ortopedico Rizzoli, Via di Barbiano 1/10, 40136 Bologna, Italytaddei@tecno.ior.it

Saulo Martelli, Luca Cristofolini

Laboratorio di Tecnologia Medica, Istituto Ortopedico Rizzoli, Via di Barbiano 1/10, 40136 Bologna, Italy; Facoltà di Ingegneria, Università degli Studi di Bologna, Viale Risorgimento 2, 40136 Bologna, Italy

Harinderjit Singh Gill

Nuffield Department of Orthopaedic Surgery, University of Oxford, Nuffield Orthopaedic Centre, Oxford, UK

Marco Viceconti

Laboratorio di Tecnologia Medica, Istituto Ortopedico Rizzoli, Via di Barbiano 1/10, 40136 Bologna, Italy

1

Corresponding author.

J Biomech Eng 132(2), 021002 (Jan 05, 2010) (11 pages) doi:10.1115/1.4000065 History: Received August 08, 2008; Revised July 16, 2009; Posted September 01, 2009; Published January 05, 2010; Online January 05, 2010

## Abstract

Metal-on-metal hip resurfacing is becoming increasingly popular, and a number of new devices have been recently introduced that, in the short term, appear to have satisfactory outcome but many questions are still open on the biomechanics of the resurfaced femur. This could be investigated by means of finite element analysis, but, in order to be effective in discerning potential critical conditions, the accuracy of the models’ predictions should be assessed. The major goal of this study was to validate, through a combined experimental-numerical study, a finite element modeling procedure for the simulation of resurfaced femurs. In addition, a preliminary biomechanical analysis of the changes induced in the femoral neck biomechanics by the presence of the device was performed, under a physiologic range of hip joint reaction directions. For this purpose, in vitro tests and a finite element model based on the same specimen were developed using a cadaver femur. The study focused on the Conserve Plus, one of the most common contemporary resurfacing designs. Five loading configurations were identified to correspond to the extremes of physiological directions for the hip joint. The agreement between experimental measurements and numerical predictions was good both in the prediction of the femoral strains $(R2>0.9)$, and in the prosthesis micromotions $(error<20 μm)$, giving confidence in the model predictions. The preliminary biomechanical analysis indicated that the strains in the femoral neck are moderately affected by the presence of the prosthesis, apart from localized strain increments that can be considerable, always predicted near the stem. Low micromotions and contact pressure were predicted, suggesting a good stability of the prosthesis. The model accuracy was good in the prediction of the femoral strains and moderately good in the prediction of the bone-prosthesis micromovements. Although the investigated loading conditions were not completely physiological, the preliminary biomechanical analysis showed relatively small changes for the proximal femur after implantation. This validated model can support realistic simulations to examine physiological load configurations and the effects of variations in prosthesis design and implantation technique.

<>

## Figures

Figure 1

From the left: Picture of the intact specimen with outline of the prosthesis on the testing machine. The position of the strain gauges is indicated. Strain gauges labels refer to their placement in terms of anatomical aspect (A: anterior; L: lateral; M: medial; and P: posterior) and the vertical level (H: nearest to femoral head; N: femoral neck; 1: metaphysis; and 5: diaphysis). The in vitro test setup for the neutral position is shown (the other loading configurations were similar, where a wedge was placed under the femur to obtained the desired tilt angles). The system of cross-rails (a) to eliminate horizontal force components is visible proximally. The femur lies on a plate fixed on the loading cell (b). The three aluminum rods, which were potted and CT-scanned together with the bone specimen, are visible distally. On the right, from top to bottom: A detail of the proximal femur showing locations of proximal strain gages. Schematic of a femur showing the five extreme directions of the resultant hip joint force. The medioposterior view of the proximal implanted femur showing the position of the two displacement transducers (LVDT type) measuring gap opening, in the most medial location, and shear motion on the posterior side.

Figure 2

Solid model of the implanted femur (left). The cement is supposed to fill the gap between the prosthesis and the reamed femur. On the right, the finite element model of the same femur. A set of element is hidden to visualize the internal mesh. The ROI where comparison between strain levels in the intact and implanted femur was performed is indicated with dashed line.

Figure 3

Experimentally measured strain in the implanted femur as a fraction of the strain in the intact condition for load configuration 4 (0 deg abduction and 18 deg flexion). A value of 100% indicates no alteration; larger than 100% indicates strain increase; and lower than 100% indicates stress shielding. The bars correspond to the maximum and minimum principal strain (average±standard deviation for five loading repetitions). Error bars are hardly visible, because strain measurements were highly repeatable. Additionally, the change in principal strain direction (variation of the angle between the largest principal strain, and the long axis of the femur) are reported. Only values were strain exceeded 100με are reported, to exclude large % variations that lack of any biomechanical relevance.

Figure 4

Linear regression (including results from the strain gauge LH) between predicted strains (με) on the y-axis and experimental measurements (με) on the x-axis in the LC4 load configuration (the one experimentally tested)

Figure 5

The maximum principal strain (ε1) is plotted as this is the best predictor of the risk of failure for the bone tissue (76). All maps refer to the LC4 loading configuration, which produced the highest superficial strains. On the first row, the strain differences map on the superficial ROI region calculated as the difference between the ε1 in the implanted and in the intact femur. On the second row the ε1 map in the same region.

Figure 6

Changes of the strain field in the inner ROI volume when the femur is loaded by the LC5 loading scheme, which produced the highest differences

Figure 7

The cumulative frequency distribution of strain changes in the implanted femur with respect to the intact. Averaging absolute values of variations in tensile and compressive strains through all five loading configurations.

Figure 8

Von Mises stresses (MPa) in the prosthesis and the principal tensile stress in the cement. Both maps are referred to the LC4 loading configuration for which the peak stresses were predicted.

Figure 9

On the left the sliding micromotions map and on the right the gap-opening map. Both maps have been predicted for the LC4 load configuration, which produced the highest bone-prosthesis micromotions.

## Discussions

Some tools below are only available to our subscribers or users with an online account.

### Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related Proceedings Articles
Related eBook Content
Topic Collections