Grammont's reverse shoulder prosthesis for rotator cuff arthropathy. A retrospective study of 32 cases.

The authors have retrospectively studied a series of 32 reverse shoulder prostheses implanted in 30 patients by the same surgeon between 1992 and 2000. The mean age was 71 years; 26 patients were female, 4 were male. Thirteen patients (14 prostheses) were clinically and radiologically examined at follow-up; 9 patients (10 prostheses) were questioned by telephone and their radiological records were studied. With a mean follow-up of 31 months, 92 % (22/24) were entirely satisfied with the operation. The mean Constant score for the 14 shoulders which were clinically evaluated at follow-up was 60/100. There was one failure, related with glenoid loosening. These findings are in line with the good short-term results previously reported with this prosthesis. Glenoid notching is a well-known problem with the reversed prosthesis; it may lead to implant failure. We noted such an image in 50 % of our patients. In several cases however, the radiological finding was more suggestive of osteophytic formation than of real bony erosion, an observation that has not been reported before. The image remained stable over time and did not lead to glenoid loosening within the time limits of the study. Nevertheless, notching remains a concern with respect to the long-term survival of these implants which should therefore, in our opinion, be used only in elderly patients. Whether improved technique or design modification can address this issue still has to be established.