Quantifying the variability of glenoid geometry in intact and osteoarthritic shoulders

Open Access
Devries, Charlotte De
Graduate Program:
Mechanical Engineering
Doctor of Philosophy
Document Type:
Date of Defense:
June 10, 2014
Committee Members:
  • Matthew B Parkinson, Dissertation Advisor
  • Matthew B Parkinson, Committee Chair
  • Mary I Frecker, Committee Member
  • Stephen Jacob Piazza, Committee Member
  • April D Armstrong, Special Member
  • Glenoid
  • Total Shoulder Arthroplasty
  • Implant Design
  • Principal Component Analysis
  • Design for Human Variability
The purpose of this project is to identify sources of variability within the glenoid of the scapula to assist in implant placement and design for Total Shoulder Arthroplasty (TSA). The glenoid is the socket component of the ball and socket connection of the shoulder joint. When a person has osteoarthritis, the glenoid can wear down, resulting in a loss of mobility, pain, or even causing the shoulder to dislocate. To restore range of motion, and reduce or eliminate pain, surgeons perform TSA. This surgery involves inserting a metal implant in the humerus, and a plastic implant in the glenoid. If the glenoid implant is improperly placed, it can wear out or potentially become dislocated within a ten year period. In some cases, the full range of motion is not returned to the patient. Due to the limited visual input during surgery, it can be difficult to obtain the correct angle of implantation. By studying the variability of healthy glenoids, it may be possible to estimate the original healthy geometry of osteoarthritic shoulders. The objective of this research was to model intact and osteoarthritic glenoids. The model needed to be robust to the observed variability, and of sufficient resolution, such that it informed operative and design decisions. Achieving this required the quantification of variability in landmark locations and relevant bone geometry in both intact shoulders and those exhibiting osteoarthritis. Additionally, the surface geometry of the glenoid vault was modeled. This required the application of existing mathematical and statistical modeling approaches, including geometric fitting, radial basis functions, and principal component analysis. The landmark identification process represented the glenoid in new manner. The work was validated against existing approaches and CT scans from 42 patients. For this research, x-ray computed tomography (CT) scans of healthy scapulae were provided. The scapulae were isolated in Osirix, and exported as meshes. The meshes were cleaned using Netfabb and Meshlab. The code created to process and analyze these meshes was written in Matlab. Each scapula was oriented based on landmarks found on the bone; the center of the glenoid, the inferior-most part, and the connection of the medial border and the spine of the scapula. The concave face of the glenoid was isolated, and a sphere was fit to each glenoid. The sphere fitting provides the inclination and version angles of the glenoid with respect to the landmarks. Additionally, the optimal angle for maximum implant peg room was identified. A range of information on shoulder geometries can assist with preoperative planning, as well as implant design. As there is a limited supply of CT scans, potential shoulder geometries were synthesized. This was done by placing landmarks on the existing glenoid meshes, such that they provided enough information to represent the geometry, while being consistent across each glenoid. Principal component analysis was used to quantify the variability of shape across the glenoid landmarks. These landmarks were transformed into glenoid meshes using radial basis functions. The process of creation of these shoulder geometries may possibly be useful for the study of other joints. The models created will help surgeons and engineers to understand the effects of osteoarthritis on bone geometry, as well as the range of variability present in healthy shoulders.