Boileau and colleagues7 randomized 39 patients (40 shoulders) to either a cemented all-polyethylene glenoid or a cementless metal-backed glenoid component. Although the metal-backed glenoid components had a significantly lower rate of radiolucent lines, the metal-backed glenoids had a significantly higher rate of loosening. The authors subsequently abandoned use of uncemented metal-backed glenoid components. Taunton and colleagues8 reviewed 83 TSAs with a metal-backed bone ingrowth glenoid component. In 74 cases, the preoperative diagnosis was primary osteoarthritis. Mean clinical follow-up was 9.5 years. During follow-up, there were improvements in pain, forward elevation, and external rotation. Radiographic glenoid loosening was noted in 33 shoulders; 9 required revision for glenoid loosening. Both series demonstrated a high rate of revisions for cementless glenoid components.
Similar revision difficulties were noted by Montoya and colleagues.9 In their series of 65 TSAs performed for primary osteoarthritis, a cementless glenoid component was implanted. There were significant improvements in Constant scores, forward flexion, external rotation, and abduction but also an 11.3% revision rate noted at 68 months (mean follow-up). Glenoid revisions were required predominantly in patients with eccentric preoperative glenoid morphology. Lawrence and colleagues10 used a cementless ingrowth glenoid component in 21 shoulder arthroplasties performed for glenoid bone loss (13) or revision (8). They noted a high rate of revisions but good outcomes for the cases not revised. In both studies, there was a high rate of revision for glenoid loosening but also a tendency for revisions to be correlated with more challenging clinical applications.
Wirth and colleagues11 followed 44 TSAs using a minimally cemented ingrowth glenoid component. There were significant improvements in ASES scores, Simple Shoulder Test scores, and visual analog scale pain ratings. No revisions for glenoid loosening were noted. The implants were thought to provide durable outcomes at a mean follow-up of 4 years. These results were similar to those appreciated in the present study. In both series, the revision rate was much lower than described in the literature, and there were predictable improvements in pain and active motion.
Our study had several limitations: small number of patients, no comparison group, and relatively short follow-up. More long-term data are needed to appropriately compare cemented and uncemented glenoid components. In addition, it is difficult to compare our group of patients with those described in the literature, as the implants used differ. Despite these limitations, our data suggest that tantalum ingrowth glenoid components provide predictable and sustainable outcomes in TSA. With longer-term follow-up, tantalum ingrowth glenoids may demonstrate a durable and reliable alternative to cemented glenoid components.