JSES International - 2026-05-01 - Journal Article
Does humeral head size predict the lateralization required to preserve near-anatomic posterosuperior rotator cuff length in reverse shoulder arthroplasty?
Lante E, Blakeney WG, Bauer S
Topics
Key Takeaway
Humeral head size correlates with required graft thickness for near-anatomic lateralization in rTSA at r=0.93, with the Frankle-type configuration best reproducing native posterosuperior cuff muscle length.
Summary Depth
Choose how much analysis to show on this article page.
Summary
This computational study asked whether native humeral head size (HHS) could serve as a patient-specific anatomical reference to guide glenoid baseplate lateralization in rTSA. Using a validated statistical shape model applied to 83 preoperative CT scans, five implantation strategies from Grammont-type medialization to Frankle-type lateralization were modeled and correlated with posterosuperior cuff muscle-tendon lengths. HHS correlated with required graft thickness at r=0.93 and with infraspinatus length at r=0.88; the Frankle-type configuration most closely reproduced near-anatomic muscle lengths, with implantation strategy producing Kendall W of 0.9–0.97 across configurations.
Key Limitation
The study is purely geometric and computational with no clinical, functional, or intraoperative validation, so the assumption that preserving near-anatomic muscle-tendon length translates to improved postoperative rotator cuff function remains unproven.
Original Abstract
BACKGROUND
Reverse total shoulder arthroplasty (rTSA) lacks an anatomy-based framework for individualized implant planning. We hypothesized that native humeral head size (HHS) could serve as a patient-specific anatomical reference to guide lateralization in rTSA. The purpose of this study was to determine whether HHS predicts the amount of baseplate lateralization required to preserve near-anatomic posterosuperior rotator cuff muscle length, and whether muscle lengths can be reliably estimated in deformed joints using a statistical shape model (SSM).
METHODS
In this computational study, 83 pre-operative computed tomography scans were analyzed using a validated SSM to estimate rotator cuff muscle-tendon lengths. Five rTSA implantation strategies were modeled while controlling implant geometry, ranging from a medialized Grammont-type configuration to lateralized Frankle-type and hybrid strategies. Best-fit humeral head size was correlated with patient-specific lateralization quantified by overlap-based graft thickness and with posterosuperior cuff muscle length.
RESULTS
Best-fit humeral head size demonstrated an excellent correlation with graft thickness required for near-anatomic overlap (r = 0.93) and strong correlations with posterosuperior cuff muscle length, particularly the infraspinatus (r = 0.88), as well as the subscapularis (r = 0.76) and teres minor (r = 0.62). Among implantation strategies, the Frankle-type configuration most closely reproduced near-anatomic graft thickness and posterosuperior cuff muscle length. Implantation strategy produced large to very large differences in baseline muscle-tendon length across configurations (Kendall W ≈ 0.9-0.97). The SSM reliably reconstructed muscle-tendon geometry across a broad spectrum of degenerative pathology.
CONCLUSION
HHS predicts the lateralization required to preserve near-anatomic posterosuperior cuff muscle length in rTSA. A Frankle-type lateralization strategy most closely reproduced the overlap-based anatomic reference, while alternative strategies produced systematic deviations in muscle length. These findings support HHS as an anatomy-based, patient-specific guide for lateralization planning. This framework is geometric and conceptual in nature and requires future clinical validation.