JSES - 2026-04-09 - Case Reports; Journal Article
Bone Densities in Type E3 Glenoids: Quantitative Assessments and Topographical Distributions.
Klarer M, Fleet CT, Giles JW, Athwal GS, Johnson JA
Topics
Key Takeaway
In Favard E3 glenoids, peak cortical and trabecular bone density concentrates in the peripheral superior-anterior quadrant (rings 8–9), while the peripheral inferior-posterior quadrant shows the lowest density (both p<0.001).
Summary Depth
Choose how much analysis to show on this article page.
Summary
This study quantified regional bone density variations across Favard E3 glenoids using CT-derived analysis mapped through a novel 40-segment 3D grid (4 quadrants × 10 concentric rings, 18mm depth). In 28 patients (mean age 78), mixed repeated-measures MANOVA demonstrated peak cortical and trabecular density in the peripheral superior-anterior region and minimum density in the peripheral inferior-posterior region (p<0.001). The asymmetric density distribution only partially aligns with the global erosion pattern described in the Favard E3 classification, suggesting preferential joint loading in the superior-anterior zone.
Key Limitation
The sample of 28 patients is small and skewed female (82%), limiting generalizability and statistical power to detect density differences across subregions within the E3 classification.
Original Abstract
BACKGROUND
Favard E3 type glenoids are characterized by superior and global erosion patterns and can be successfully treated with reverse total shoulder arthroplasty. However, with severe bone loss, implant loosening remains a concern. Since little is known about the regional density variations of Favard type glenoids, this study analyzed E3 types from central to peripheral regions of the glenoid. The study worked with the following hypotheses: First, that regional variations of cortical as well as trabecular bone densities of E3 type glenoids exist; second, that a novel three-dimensional orientational grid can assess and show significant variability in topographical density distributions from central to peripheral regions of the glenoid; third, that the density distributions of E3 types correlate with their erosion pattern.
METHODS
This study used computer tomography scans of 28 patients (five males, 23 females, mean age 78) that were classified as E3 type glenoids by an experienced shoulder surgeon. A novel three-dimensional grid of 40 different segments, combining quadrants (superior, anterior, posterior, and inferior) as well as 10 concentric rings with a single depth layer of 18mm into the glenoid vault, mapped the glenoid bone density variations in fine detail. A mixed repeated-measures multiple analysis of variance was conducted to assess the variations in bone density between each quadrant, ring, and bone type (either cortical or trabecular).
RESULTS
The results showed that both cortical as well as trabecular bone densities peaked in the periphery (i.e., in rings 8 and 9) of the superior-anterior glenoid regions (P <0.001) and conversely showed low densities in the periphery of inferior-posterior glenoid regions (P <0.001).
CONCLUSION
Greater bone densities in the periphery of the superior-anterior regions in type E3 glenoids suggest that joint loading occurs preferentially in these areas. These density results therefore only partially coincide with the Favard E3 type erosion description. The density data may also help optimize future implant designs that leverage these high-density regions. Anchoring augmented baseplates in adequately dense bone at time-zero has shown to decrease implant loosening, minimizing the likelihood of surgical revisions by improving fixation and reducing micromotion. Also, the three-dimensional grid can be used to revisit other Favard subtypes to compare and correlate densities to gain further insight into the pathophysiology of cuff tear arthropathy.