Cemented versus uncemented fixation of second-generation Trabecular Metal glenoid components: minimum 5-year outcomes

Total shoulder arthroplasty (TSA) with second-generation Trabecular Metal™ implants (Zimmer, Warsaw, IN, USA) has shown good short-term outcomes. Differences in outcomes between cemented and uncemented fixation are unknown. This study compared the clinical, radiographic, and patient-rated outcomes of TSA with cemented vs. uncemented TM glenoids at minimum 5-year follow-up. Patients who underwent anatomic TSA with second-generation TM glenoid components for primary osteoarthritis were identified for minimum 5-year follow-up. The patients were divided into 2 groups: cemented and uncemented glenoid fixation. Outcome measures included implant survival, patient-rated outcome scores (Patient-Reported Outcomes Measurement Information System [PROMIS] and American Shoulder and Elbow Surgeons scores), shoulder range of motion, and radiographic analysis. Findings were compared between groups. The study included 55 shoulders: 27 in the cemented group (21 with full radiographic follow-up) and 28 in the uncemented group (22 with full radiographic follow-up). Both groups had similar follow-up times (6.6 years in cemented group vs. 6.7 years in uncemented group, P = .60). Moreover, the groups did not differ significantly in sex composition, age at the time of surgery, or preoperative Walch glenoid grade distribution. No patients required revision surgery. The 2 groups had similar preoperative range of motion, but patients in the uncemented group had greater follow-up forward flexion (P = .03), external rotation (P < .01), and lateral elevation (P = .03) than did patients in the cemented group. PROMIS scores were not significantly different between groups. American Shoulder and Elbow Surgeons scores were similar (89.8 in cemented group vs. 94.1 in uncemented group, P = .21). Mid-term radiographs showed a metal debris rate of 24% in the cemented group and 27% in the uncemented group. Although these values were not significantly different (P = .90), the frequency of mild metal debris (grade 1-2), when present, was greater in the uncemented group (grade 2 in 6 shoulders) than in the cemented group (grade 1 in 4 and grade 2 in 1, P = .02). There was a greater presence of mild (grade 1) radiolucent lines in the uncemented group (64%) than in the cemented group (29%, P < .01). No glenoid had evidence of loosening (defined by a change in position or radiolucent lines > 2 mm). The presence of metal debris and radiolucent lines did not have a significant effect on clinical outcomes