Abstract
Purpose:
The optimal degree of conformity between the glenoid and humeral components in cemented total shoulder arthroplasty (TSA) has not been established. Glenoid component stability is thought to be at risk due to the “rocking-horse” phenomenom, which, can lead to increased micromotion and loosening in response to humeral head edge loading. The goal of this biomechanical study is to investigate the influence of glenohumeral mismatch on bone-implant interface micromotion in a cemented glenoid implant model.
Methods:
Twenty-Five cemented glenoid components (Affiniti, Tornier, Inc., Bloomington, MN, USA) were implanted in polyurethane foam biomechanics testing blocks. Five glenoid sizes, 40 mm, 44 mm, 48 mm, 52 mm and 55 mm (n = 5 per glenoid size), were cyclically tested according to ASTM Standard F-2028-08. A 44 mm humeral head (Affiniti, Tornier, Inc., Bloomington, MN, USA) was positioned centrally within the glenoid fixed to a materials testing frame (MTS Mini-Bionix II, Eden Prairie, MN, USA). Phase I testing (n = 3 per glenoid size) involved a subluxation test for determination of the humeral head translation distance which would be used for phase II cyclic testing. During cyclic loading, the humeral head was translated ± distance for 50,000 cycles at a frequency of 2 Hz, simulating approximately 5 years of device use. Glenoid compression, distraction, and superior-inferior glenoid translation were measured throughout testing via two differential variable reluctance transducers.
Results:
Humeral head translation distance was identified as 0.55 mm, 1.09 mm, 2.32 mm, 3.82 mm, and 4.73 mm for each glenoid size, respectively (Figure 1). No significant difference was noted in 40 mm glenoids between cycle 1 and 50,000 for all parameters evaluated during testing (p > 0.05) (Figure 2). Conversely, a significant decrease in superior-inferior translation was present for 44 mm between cycle 1 and 50,000 (p = 0.010) (Figure 3). When analyzing all data from the first two smallest glenoid sizes, glenoid compression and translation both showed significantly increased micromotion with 40 mm glenoid sizes compared with the 44 mm glenoid size (p = 0.010 and p = 0.002, respectively). No significant difference was found with respect to glenoid distraction (p = 0.136).
Conclusion:
The first phase of mechanical testing established the subluxation displacement of the humeral head against the glenoid for each prosthetic mismatch couple, which was larger for couples with greater glenohumeral mismatch. During cyclic testing, this displacement distance was covered in the same amount of time leading to differences in humeral head velocity and resultant stresses seen at the implant-cement-foam interfaces. A smaller mismatch in glenohumeral radius may lead to greater stress with shorter humeral translation compared to greater mismatch allowing for larger translations with lower resultant stresses. Data from our study will provide further clarification on the importance of glenohumeral mismatch on implant stability. Further studies are warranted to fully evaluate the impact and optimal amount radial mismatch for a clinical setting.
