Copeland Shoulder: Indications for Surface Replacement

Primary and secondary arthritis of the shoulder is the commonest indication. The prosthesis has been used for osteoarthritis, rheumatoid arthritis, avascular necrosis, cuff arthropathy, instability arthropathy, post trauma arthritis, post infective arthritis, arthritis secondary to glenoid dysplasia and ephiphysial dysplasia. It is not intended for use in fresh fractures.

The results of surface replacement, as in any other shoulder replacement, depends on the indications and diagnosis.   The best results are achieved in osteoarthritis with an intact cuff and the worst results in cuff tear arthropathy and post-traumatic arthritis.  It can be used in moderate to severe degrees of erosion of the humeral head as the humeral head is often bone grafted at the same time as the surface replacement.  Once the humeral head shaper has been used if there is more than 60% apposition between the undersurface of the trial prosthesis and bone then this would be suitable for surface replacement, ie up to 40% of the humeral head may be replaced by bone graft.

Hemi or Total

There is no long term randomised study concerning whether a total or hemi-arthroplasty should be used.  The most recent study by Gartsman (American JBJS) indicates the early results are better using a total shoulder replacement from the point of view of pain relief.  This would appear to be self-evident from first principals and hemi-arthroplasty results at the hip. However, the long-term failure rate of shoulder replacement relates to glenoid loosening failure and wear [5].  Therefore the comparison must always be made between glenoid erosion by a humeral hemi-arthroplasty related to late glenoid loosening because of HDP wear.  Of the 29 revision procedures performed in our unit over the past nine years only three have been conversions of hemiarthroplasties to total shoulder arthroplasties. The trade off must always be comparing minimal decrease in pain relief compared to very probable long-term failure of glenoid and late revision surgery.  This must be taken into account if the prosthesis is used in younger and younger patients.

Early in our unit nearly all joints had a total replacement.  The only occasion a glenoid was not used was in those where no glenoid was possible, i.e. too little bone was present to gain fixation, demonstrated by erosion beyond the base of the corocoid.  Now because of reports of rocking horse glenoid etc, nearly all procedures are hemi-arthroplasties. Only if there is non-concentric erosion is this an indication for a total arthroplasty. However, when a hemi-arthroplasty is done the glenoid is not ignored, it is drilled and either a fascial interposition arthroplasty made or just drilling of the osteochondral hard bone to encourage fibrocartilagenous secondary formation. Our low conversion rate from hemi-arthroplasty to total is possibly due to this and due to the anatomically-sized humeral head. If a normal modular prosthesis is used as a hemi-arthroplasty then the surgeon tends to err on the side of making the head diameter too small therefore the possibility of point glenoid erosion is high.  With the surface replacement system the tendancy is to err towards a larger head size and there may be rim erosion rather than point erosion.