Bone Strain in the Patella

Overview

Patellofemoral complications, specifically patellar fracture and anterior knee pain, remain a cause for total knee replacement revision. Subject-specific finite element (FE) models developed from image data can capture the anatomy and bone material property distribution. This case study highlights the development of subject-specific FE analyses to evaluate patellofemoral mechanics and the distribution of bone strain.

Characteristics:

  • Segmentation of subject-specific anatomy from MR and CT images using ScanIP
  • Mapped bone material properties using +FE module
  • FE-based evaluation of patellofemoral mechanics and bone strains using Abaqus®

Thanks to

Center for Orthopaedic Biomechanics, University of Denver: 
C.K. Fitzpatrick • P.J. Rullkoetter • P.J. Laz

Image Processing

The structures of the knee were segmented from magnetic resonance (MR) and computed tomography (CT) images using ScanIP. Femoral, tibial and patellar bone and cartilage were reconstructed from MR data. Patellar bone geometry was extracted and material properties were assigned to individual elements based on density from the CT data. The mesh of the patellar bone was developed with a common element set for soft tissue attachment in both the natural and implanted models.

Mesh Generation

Subject-specific models were evaluated in the natural, TKR implanted and unresurfaced conditions during a simulated deep knee bend. The natural model included bone and cartilage, while the implanted model included a size-matched patellar button and femoral and tibial components. The unresurfaced model evaluated the natural patellar bone and cartilage articulating against a femoral component. Two-dimensional fiber-reinforced membranes represented the extensor mechanism and patellofemoral ligaments. A load of 1000 N was distributed amongst the heads of the quadriceps muscle proportional to their physiological cross-sectional areas.

Simulation

Contact mechanics and the distribution of bone strain were evaluated through the flexion cycle. Analyses were performed in Abaqus/Explicit. Peak compressive principal strains in the natural specimens were primarily in the softer cancellous bone underlying the contact patch in the proximal portion of the patella, while in the implanted specimens it was focused medially, around the component pegs and at the distal nose of the patella. Differences in the highly strained volumes between the natural and implanted conditions were negligible in early flexion, but statistically significant in deeper flexion.