Postoperative proximal migration in total shoulder arthroplasty. Incidence and significance

A retrospective review of 131 Neer total shoulder arthroplasties in 111 patients performed between 1974 and 1986 was undertaken to identify factors associated with proximal humeral migration. The influence of proximal migration on results was evaluated to clarify the indications for prosthetic constraint. The average follow-up period was 55 months (range: 24-124 mo). The results showed proximal migration in 22% of patients (29 shoulders). The migration was progressive in all cases, without corresponding increases in pain. Major rotator cuff tears were present in 21% of patients with a normal glenohumeral relationship and in 24% of the patients with proximal migration. Limited elevation was noted in patients with proximal migration, but all patients had improved range of motion and decreased postoperative pain. The cause of postoperative proximal migration is believed to be secondary to a combination of factors and not rotator cuff disruption alone. These data do not support the routine use of a constrained prosthesis for irreparable rotator cuff tears.