• HYPOTHESIS
    • Loosening of the glenoid component is a common complication in shoulder arthroplasty. We proposed to radiographically and clinically investigate the outcomes of keeled and pegged glenoid components to test the hypothesis that these implants would have equivalent performance.
  • MATERIALS AND METHODS
    • One hundred patients undergoing primary shoulder arthroplasty for osteoarthritis were reviewed. Fifty patients had pegged glenoid components, and fifty had keeled components. Standardized radiographs were taken. Preoperative films classified the severity of degenerative change. The postoperative films were analyzed for radiolucent lines and shift in component position-at-risk signs for loosening.
  • RESULTS
    • Mean radiographic follow-up was 51.3 months (range, 24-90 months) for the keeled group and 45.7 months (range, 27-98 months) for the pegged group. There were no differences in the degree of preoperative arthritic change between groups. Both groups had significant improvement in motion and pain (P < .001) postoperatively, with no differences in clinical outcome between cohorts (P > or = .20). Initial radiographs showed no radiolucencies around the pegs in 46 implants or around the keel in 49 components. At final follow-up, 10 glenoid implants were found to be at risk for loosening, 6 (12%) in the pegged group and 4 (8%) in the keeled group (P = .74).
  • CONCLUSIONS
    • Initial postoperative radiographs with pegged and keeled components show a low rate of radiolucent lines. These radiolucencies develop over time. However, there is no difference in clinical or radiographic outcomes between pegged and keeled components at intermediate-term follow-up.