Effect of glenoid cementation on total shoulder arthroplasty for degenerative arthritis of the shoulder: a review of the New Zealand National Joint Registry.

BACKGROUND

Despite the lack of literature showing improved results compared with cemented designs, uncemented glenoid components are still commonly used in total shoulder arthroplasty (TSA). Most studies comparing cemented with uncemented glenoids involve small numbers or include patients with inflammatory arthritis.

METHODS

New Zealand National Joint Registry data was used to compare the outcomes of uncemented and cemented glenoids in TSA performed for degenerative arthritis. Measured variables were the revision rate and the Oxford Shoulder Score (OSS).

RESULTS

Data were retrieved on 1596 patients, with a mean follow-up 3.5 years (range 2-10.7 years), 1065 of whom had a cemented glenoid. There were no significant differences in any preoperative factors between the 2 groups. The revision rate for uncemented glenoids was 4.4 times higher than for cemented glenoids (1.92 vs. 0.44 revisions per 100 component-years, P < .001). Age <55 years was an independent risk factor for revision (P < .001). The most common reason for revision was rotator cuff wear (35.5%) in the uncemented glenoids and loosening (36.3%) in the cemented glenoids. The difference in the mean OSS between the 2 groups was less than 1 point at 6 months (P = .109) and at 5 years (P = .377).

CONCLUSION

Uncemented glenoids had a markedly higher revision rate. Patients aged <55 years have the highest revision rate regardless of glenoid fixation method. The higher revision rate in the uncemented glenoid group persisted when the effect of young age was corrected for. There was no clinically or statistically significant difference in the OSS results for clinical outcome between the two groups.

LEVEL OF EVIDENCE

Level III, retrospective cohort, treatment study.