Glenohumeral mismatch affects micromotion of cemented glenoid components in total shoulder arthroplasty

The "rocking horse" phenomenon is considered the main cause of glenoid component loosening by eccentric loading of the glenoid rim. This study aimed to investigate the influence of increasing glenohumeral implant mismatch on bone-implant interface micromotion in a cemented all-polyethylene pegged glenoid biomechanical model.

Five glenoid sizes, 40 mm, 44 mm, 48 mm, 52 mm, and 56 mm, representing +2 mm, +6 mm, +10 mm, +14 mm, and +18 mm glenohumeral mismatch, respectively, were cyclically loaded according to ASTM Standard F2028-08 at a constant frequency of 2 Hz to a size-dependent humeral head subluxation translation. Additional glenoid components were cyclically loaded to their subluxation translations at a constant humeral head rate of 4.4 mm/s. Component micromotion was characterized as compression, distraction, and superior-inferior translation measured by differential variable reluctance transducers.

During constant frequency tests, 52-mm and 56-mm glenoids were unable to complete cyclic testing because of catastrophic failure of the glenoid-implant interface and permanent glenoid deformation, probably due to increasing severity of testing parameters. When tested at a constant humeral head speed, 48-mm, 52-mm, and 56-mm glenoids had significantly increased glenoid distraction and glenoid translation at cycle 50,000 compared with cycle 1. Distraction and translation measurements for 52-mm and 56-mm glenoids were significantly greater compared with 40-mm, 44-mm, and 48-mm glenoids at 50,000 cycles.

In a biomechanical model, optimal glenohumeral mismatch in cemented pegged glenoid implants is multifactorial and has not been definitively established. However, our data suggest that a radial mismatch of less than +10 mm may decrease the risk of glenoid micromotion.