Abstract
Introduction
Inadequate stability of the baseplate is a leading cause of revision within reverse total shoulder arthroplasty (rTSA). Micromotion between baseplate and bone is commonly used as a pre-clinical indicator for clinical stability (ASTM F2028-14). Finite element analysis (FEA) has been shown to accurately predict baseplate-bone micromotion, but results may be critically dependent on several modeling assumptions. Here, FEA was used to assess the impact of key modeling assumptions related to screw-bone interactions on various rTSA configurations.
Methods
FEA with Ansys ver. 16 was used to simulate a fixation experiment. Baseplates of two different sizes (25mm and 28mm diameter), each with a central screw and four peripheral screws, were virtually implanted in a synthetic bone block. Each baseplate was analyzed using 1.5mm and 3.5mm superior-inferior (SI) offsets of the glenosphere center, as well as using four (‘4S’) and two (‘2S’) peripheral screws. A clinically relevant loading of 756N was applied in compression as well as in inferior-to-superior shear direction through the glenosphere (Figure 1A, 1B).
Screw-bone block interactions were modeled in three different ways: (1) Threads were defeatured from the peripheral screws, which were bonded to the bone block (b-nt); (2) Threads were modeled, while still assuming bonded contact (b-t); (3) Threads were modeled, with frictional contact between threads-bone block (f-t). Micromotion results (Figure 1C) from all 24 simulations (3 screw-bone interactions × 2 baseplate diameters × 2 SI offsets × 2 screw configurations) were compared.
Results
Across all 24 configurations, the f-t screw-bone interaction resulted in increased micromotion relative to the corresponding bonded simulations. Differences between the two bonded simulations varied among configurations (Figure 2).
Screw configuration: For all baseplate diameters, SI offsets, and screw-bone interactions, the 4S configuration had less micromotion (7–20%) than the corresponding 2S configuration (Figure 2).
SI Offset: For all baseplate diameters, screw configurations, and screw-bone interactions, the 1.5mm SI offset configuration predicted higher micromotion than the corresponding 3.5mm SI offset configuration (increase of 5–18%), except for the 25mm baseplate in b-nt condition (12–19% decrease) (Figure 3A).
Baseplate diameter: For all screw configurations and SI offsets, the f-t modeling assumption resulted in decreased micromotion (5–12%) for the 28mm baseplate as compared to the 25mm baseplate. This trend was reversed for select screw configurations and SI offsets for the other two (b-nt, b-t) modeling assumptions (Figure 3B).
Discussion
This study highlights the importance of FEA model fidelity (the level of rigor with which the screw-bone interface is modeled) on evaluating differential performance between rTSA baseplate configurations within a single design family. Three different levels of rigor were considered, based on whether or not the screw threads were explicitly modeled, and on the level of friction allowed between the screw and the bone block. Results highlight that answers to basic questions on relative baseplate performance (e.g. is a 25mm or 28mm baseplate more stable?) are sensitive to these assumptions, and require adequate model validation. Increased care should be taken when conducting evaluations across multiple device families, when additional variables (e.g. screw pitch/torque) are present that could confound analyses.
