Registry of shoulder arthroplasty - the Scottish experience

Recognising that timely dissemination of information in the orthopaedic community was important and in the absence of any national guidelines for shoulder arthroplasty, the Scottish shoulder arthroplasty registry, a voluntary registry, was started in 1996. The goals of the registry were to assess contemporary practice, provide a benchmark against which surgeons could compare their practice, identify risk factors for a poor outcome, and to improve outcomes through continuous feedback to the participating surgeons.

A standardised proforma was used to collect information on the diagnostic and demographic data, type of procedure performed, type of implant used, any associated procedures performed in conjunction with the arthroplasty, and peri-operative complications. Postoperative pain, activity and patient satisfaction were assessed annually using another standardised proforma.

Twenty surgeons have contributed to the register and 451 shoulder arthroplasties were registered over a 5-year period. Of patients, 23.2% were male and 76.8% female. The mean age was 65 years (range, 37-90 years). Shoulder arthroplasty was commonly performed for rheumatoid arthritis followed by trauma, osteoarthritis and avascular necrosis of the humeral head. Overall, 397 (88%) patients had a hemi-arthroplasty and 54 (12%) had a total shoulder replacement. Of the 54 cases that had a glenoid replacement, 28 were performed for inflammatory arthritis, 21 for osteoarthritis and 5 were for revisions. The humeral component was cemented in 204 (45%) cases, 160 of whom had a shoulder replacement for trauma. The glenoid component was cemented in 48 (89%) cases. Cross referencing our data with the figures of the actual number of shoulder arthroplasties performed, however, indicated that our registry at best collected only 53% of all the shoulder arthroplasties performed in Scotland annually.

The value of a joint registry is dependent on the accuracy and completeness of the data entered. Our registry, therefore, fails as an implant registry. We believe that compliance for data registration can only be ensured if dedicated data collection staff are employed to co-ordinate the data collection and collation process.