Longitudinal observational study of total shoulder replacements with cement: fifteen to twenty-year follow-up

Abstract

Background: There is little information about the long-term longitudinal outcomes of total shoulder arthroplasty. Moreover, long-term data on a single shoulder replacement system are lacking. The aim of this study was to analyze, at different time points during a follow-up period of fifteen to twenty years, the clinical and radiographic outcomes of patients treated with a third-generation total shoulder arthroplasty.

Methods: Sixty-three total shoulder arthroplasties were performed in fifty-eight patients. Two patients were lost to follow-up, and sixteen patients died during the follow-up period, leaving a cohort of forty-five total shoulder arthroplasties in forty patients with a mean age of sixty-four years. Follow-up examinations were carried out at six months, one year, two years, three to four years, five to eight years, nine to fourteen years, and fifteen years or more. Preoperatively and at each follow-up visit, the Constant score as well as shoulder flexion and external rotation were assessed. Radiographs were analyzed for loosening and migration of the components at all time points.

Results: There was a substantial improvement of all clinical parameters from baseline to the latest follow-up evaluation. Clinical results reached a plateau at one year, remained stable without substantial worsening for eight years, and then remained as good as the six-month results until fifteen years postoperatively. Compared with the outcomes at the six-month follow-up examination, the results were worse beyond fifteen years. The clinical outcome was not influenced by sex, age, hand dominance, glenoid morphology, glenoid loosening, or upward migration of the humeral head. Thirty-three (73%) of the forty-five shoulders had radiographic evidence of glenoid loosening at the time of final follow-up. Fourteen shoulders (31%) were revised: a soft-tissue revision was performed in one of them, and implant revision was necessary in thirteen.

Conclusions: Although there is a significant and longitudinal improvement in shoulder function and pain relief after total shoulder replacement, long-term follow-up of fifteen to twenty years demonstrated a high revision rate in this cohort.