Computed tomography analysis of radiostereometric data to determine flexion axes after total joint replacement: Application to the elbow joint

A Ericson, Henrik Olivecrona, André Stark, Marilyn E. Noz, Gerald Q. Maguire, M. P. Zeleznik, T ArndtJournal of Biomechanics 2010 · DOI

TL;DR

This paper presents a novel method for accurately visualizing and analyzing the in vivo motion of elbow joints after total replacement surgery. By fusing radiostereometric analysis (RSA) data with 3D computed tomography (CT) scans, researchers can precisely map the joint's flexion axes relative to the implanted prosthesis, paving the way for better implant design and surgical technique evaluation.

What Problem Does This Paper Address?

Elbow joint replacement surgery aims to restore function to severely damaged elbows. However, existing elbow prostheses have shown higher complication rates and loosening compared to hip and knee replacements. A key factor contributing to these issues is thought to be non-physiological joint kinematics, which may not accurately replicate the natural elbow's motion. To improve implant design and surgical outcomes, it's crucial to understand how these prostheses move within the body (in vivo kinematics). While radiostereometric analysis (RSA) is a powerful tool for tracking in vivo motion, it struggles to precisely relate these movements to anatomical landmarks or the prosthesis itself. This is particularly challenging after joint replacement, where the prosthesis can obscure natural anatomical features. This paper addresses this gap by developing a method to accurately align RSA kinematic data with detailed 3D CT scans of the joint and prosthesis.

Methods

The core of this research lies in the fusion of two imaging techniques: Radiostereometric Analysis (RSA) and 3D Computed Tomography (CT).

  • Radiostereometric Analysis (RSA): This technique involves implanting tiny, radio-opaque tantalum markers into the bone around the joint. A series of X-ray images are taken from different angles as the joint moves. By analyzing the precise positions of these markers in the X-ray images, researchers can reconstruct the 3D motion of the bone segments and, importantly, determine the joint's instantaneous axes of rotation (flexion axes).
  • 3D Computed Tomography (CT): CT scans provide detailed, high-resolution 3D images of the bone and any implanted prostheses. This allows for precise visualization and measurement of anatomical structures and implant geometry.

The novel aspect of this study is the fusion of RSA and CT data. This is achieved by:

  1. Registration: The positions of the tantalum markers identified in the RSA data are precisely mapped onto the corresponding locations in the 3D CT volume. This is done using a validated 3D volume fusion tool and a rigid body transformation, ensuring that the spatial relationship between the markers and the bone is maintained.
  2. Prosthesis-Based Reference System: A key innovation is the development of a method to define a stable reference system directly on the implanted prosthesis within the 3D CT volume. This involves marking specific points on the prosthesis (e.g., the joint surface for a transverse axis, or the stem for a longitudinal axis) using a 3D isosurface landmarking mode. This allows for the creation of prosthetic axes that are independent of surrounding bone anatomy.
  3. Visualization: Once the RSA kinematic data (flexion axes) are transformed into the CT coordinate system and aligned with the prosthesis-based reference system, the motion axes can be visualized directly in relation to the 3D CT model of the prosthesis.

Datasets

The study utilized data from six patients who had previously undergone total elbow joint replacement. These patients had received either a Kudo prosthesis or a Capitello-condylar prosthesis. The data included:

  • Radiostereometric analysis (RSA) data from a previous study on the same patients, which provided the in vivo flexion axes.
  • 3D Computed Tomography (CT) scans of the elbow joints, which provided detailed anatomical and prosthetic information.

No new benchmarks were introduced; rather, the study leveraged existing RSA data and combined it with new CT data and analysis methods.

Key Insights and Ideas

  • Bridging the Gap: The primary insight is that by combining RSA and CT, researchers can overcome the limitations of each individual technique. CT provides the detailed anatomical and prosthetic context, while RSA provides the precise in vivo kinematic data.
  • Prosthesis as a Reference: Establishing a reference system directly on the prosthesis is crucial. This allows for consistent comparison of motion across different patients and studies, regardless of individual anatomical variations or the presence of surgical artefacts.
  • Visualizing Motion in Context: The ability to visualize the flexion axes directly superimposed on the 3D CT model of the prosthesis is a significant advancement. This allows for a much more intuitive and accurate understanding of how the implant is functioning within the joint.
  • Reliability of Prosthetic Marking: The study demonstrates that a reliable prosthetic reference system can be established using the proposed CT-based marking method, with good repeatability for both longitudinal and transverse axes.

Key Results

  • Accurate Visualization: The fusion of RSA and CT data successfully enabled the visualization of elbow flexion axes in relation to the prostheses in the 3D CT volume.
  • Reliable Prosthetic Axis Definition: The method for marking prosthetic axes showed good reliability. For the longitudinal axis, differences between two repeated positioning repetitions were less than 0.7°. For the transverse axis, differences were less than 2.0° (and less than 0.8° in four out of six patients).
  • Precise Motion Axis Determination: The study confirmed that the method allows for accurate determination of the position of the flexion axes after total joint replacement in vivo. Differences in repeated positioning of the longitudinal axis were less than 0.6°, and for the transverse axis, they were less than 1.4° (or less than 0.6° in four out of six patients).

Limitations and Future Work

  • Artefacts: While the study notes that prostheses could be visualized with "very few and minor artefacts" in CT, metal artefacts can still pose challenges for precise landmark identification on the prosthesis.
  • Cadaver Limitations: The authors acknowledge that studies on cadaver elbows are limited by the absence of intact muscles and ligaments, making them less directly comparable to in vivo studies.
  • Future Work: The paper suggests that this combined RSA and CT approach is highly relevant for the early assessment of new prosthesis designs and for enabling precise comparisons of implant function between patients and studies. Further research could explore the application of this method to other joint replacements and investigate the biomechanical implications of different flexion axis positions on implant longevity and patient outcomes.

Summary updated

BibTeX
@article{ericson2010computed,
  author = {A Ericson and Henrik Olivecrona and André Stark and Marilyn E. Noz and Gerald Q. Maguire and M. P. Zeleznik and T Arndt},
  title = {{Computed tomography analysis of radiostereometric data to determine flexion axes after total joint replacement: Application to the elbow joint}},
  journal = {Journal of Biomechanics},
  year = {2010},
  doi = {10.1016/j.jbiomech.2010.03.016},
}
Figure