Functional results of hemi- and total shoulder arthroplasty according to diagnosis and patient age at surgery.

Background and purpose - There is a lack of information on any associations between the functional outcome and age and diagnosis in patients who have undergone shoulder arthroplasty. We therefore evaluated the functional outcome in "young" and "old" patients treated with either hemiarthroplasty (HA) or total shoulder arthroplasty (TSA) with diverse diagnoses. Patients and methods - The functional results of 496 primary shoulder arthroplasties were analyzed using the Constant score (age- and sex-adjusted) and subjective satisfaction. Patients ≤55 years of age at surgery were defined as "young. Diagnoses were primary osteoarthritis (n = 339), posttraumatic osteoarthritis (n = 78), cuff tear arthropathy (n = 36), avascular necrosis (n = 30), and rheumatoid arthritis (n = 13). Mean length of follow-up was 4 (2-14) years. Results - 70% of the TSA patients were very satisfied with the postoperative result, as compared to 39% after HA. The Constant score and patient satisfaction were similar in the "young" and "old" groups. Pain relief was better in the "old" group. The mean improvement in the Constant score after cuff tear arthropathy (22 points) was inferior to that for primary osteoarthritis (36 points), avascular necrosis (34 points), and rheumatoid arthritis (37 points). Inferior mean Constant scores were also seen for posttraumatic osteoarthritis (29 points) compared to primary osteoarthritis (36 points). 63% of patients with primary osteoarthritis were very satisfied, as compared to only 36% of the patients with posttraumatic osteoarthritis. Interpretation - Shoulder arthroplasty is successful in the medium term for different glenohumeral diseases, irrespective of patient age at surgery. However, the appropriate treatment method for cuff tear and posttraumatic conditions of the shoulder remains to be found, particularly in young patients.