BACKGROUND The projected increase in revision shoulder arthroplasty has increased interest in the outcomes of these procedures. Glenoid component removal and conversion to a hemiarthroplasty (HA) is an option for aseptic glenoid loosening after anatomic total shoulder arthroplasty (aTSA). METHODS We identified patients who had undergone revision shoulder arthroplasty over a 15-year period. 17 patients met inclusion and exclusion criteria, and a retrospective chart review was conducted for pre-surgical and operative data. We contacted patients at a mean follow-up of 70 months from revision surgery for implant survival, reoperations and functional outcomes scores. RESULTS Implant survival was estimated to be 88% at 2 years and 67% at 5 years. Mean ASES score for surviving implants was 58 ± 22. Mean SANE score was 54 ± 24, and mean VAS pain score was 3.5 ± 2.8. Mean SF-12 Mental and Physical scores were 46 ± 15 and 38 ± 10, respectively. Five patients (50% of those with surviving implants) reported being either very satisfied or satisfied with the status of their shoulder. There were complications in 6 patients (35%) and 5 patients (29%) required reoperation. CONCLUSIONS HA following failed aTSA due to glenoid loosening produced modest clinical results and satisfaction rates. Reverse arthroplasty may be a more reliable treatment strategy in this patient population.