A study of the micromovement of pegged and keeled glenoid components compared using radiostereometric analysis

In a prospective, randomized study between 2000 and 2004, 20 patients with primary osteoarthritis of the shoulder had a total shoulder arthroplasty with radiostereometric analysis, 10 with keeled and 10 with pegged glenoid components. The relative movement of the glenoid component with respect to the scapula was measured over a 24-month period. Three keeled and five pegged glenoids needed reaming for erosion. The largest translations occurred along the longitudinal axis (mean of 1.35 mm for keeled eroded components) (P = .017 for keeled vs pegged components and P = .013 for eroded vs non-eroded components). Both of the other translation axes showed no significant differences. The highest maximum total point movement at 24 months was 2.57 mm for keeled eroded components and 1.64 mm for pegged eroded components (P = .029 for keeled vs pegged components and P = .023 for eroded vs non-eroded components). The largest rotation was anteversion, with mean values of 5.5 degrees for keeled eroded components and 4.8 degrees for pegged eroded components (P = .658 for keeled vs pegged components and P = .90 for eroded vs non-eroded components). The mean varus tilt was 4.5 degrees for keeled eroded components compared with 2.3 degrees for pegged eroded components (P = .004 for keeled vs pegged components and P = .016 for eroded vs non-eroded components), and finally, anterior-posterior rotation mean values were 3.5 degrees for keeled eroded components and 1.1 degrees for pegged eroded components (P = .022 for keeled vs pegged components and P = .04 for eroded vs non-eroded components). In conclusion, whereas all components moved, radiostereometric analysis revealed increased migration with keeled components, exacerbated by glenoid erosion. Furthermore, a distinctive pattern of migration was identified over the 2-year period.