Aim: The aim of our study was to review the massive rotator cuff tears that were repaired arthroscopically and evaluate the clinical results in respect to repair integrity as well as the effect on the progression of osteoarthritis.

Materials and Methods: We reviewed 56 (39 male,17 female) consecutive patients who underwent arthroscopic repair of their massive rotator cuff tears using biodegradable anchors by the senior author. The technique relies on the suspension bridge principal as described by S.Burkhart (1997). The mean age was 68.2 years (30–86) with most patients between 75 and 85 years. The mean follow up was 31months (24–41). Objective evaluation was done using the Constant score and subjective with the Oxford score. Osteoarthritis was investigated with radiographs and repair integrity with ultrasound. SPSS 16 for Windows was used for the statistical analysis of out results.

Results: 93% of the patients had good (11%) or excellent results according to the Oxford score and 91% had Constant score over 75. The improvement in the scores was significant statistically in all parameters (p=0.000, p for strength=0.001). Thirteen patients had postoperative OA, but this was not correlated with the results or the improvement and 11/13 had excellent Oxford scores and Constant scores over75. Seventeen patients had a re-tear which was not correlated with the results or the improvement and 15/17 had excellent Oxford scores and Constant scores over 75. Seven patients had both osteoarthritis and retear but again improvement and results were not affected.

Conclusion: Arthroscopic repair of massive rotator cuff tears has excellent clinical results regardless of the development of osteoarthritis or the repair integrity and should be the first line of treatment.

We have reviewed 29 patients (30 hips) who had undergone revision total hip arthroplasty using a Freeman metal-backed acetabular component and acetabular impaction allografting. The mean follow-up was for 15.3 years (12 to 17).

Five patients (5 hips) died with the prosthesis in situ and four (4 hips) were lost to follow-up. Twelve hips had failed and in the remaining nine there were minor symptoms. The mean time to failure requiring further surgery was nine years. Excluding patients who were lost to follow-up or had died, 72% of the hips were radiologically loose at the last review. The commonest pattern in those requiring revision was failure of the reinforcement ring in adduction with remodelling of the medial wall.

Of the nine patients who had not undergone revision, one with bilateral replacements had no current radiographs and only three of the remaining seven replacements had no radiological signs of loosening.

The short-term results for this technique have been reported to be satisfactory, but in the long term they are not. The factors associated with failure include the design of the prosthesis, which has been implicated in disappointing long-term results when used in primary arthroplasty, but not with the frequency of failure found in this series. It seems that the reliance on peripheral screw fixation over a bed of allograft without bridging the graft does not provide sufficient stability to allow incorporation of the graft.

Twenty-nine patients (30 hips) with uncemented acetabular impaction allografting contained behind a metal backed component screwed to the pelvis at revision hip arthroplasty were reviewed at 12 to17 years (average 15.3 years) follow up. Five patients had died with the prosthesis in situ, 4 patients were lost to follow up, 13 patients had failed requiring further revision (only one failed prior to 5 years) and 9 survivors were minimally asymptomatic. The mean time to failure warranting further surgery was 9 years.

Analysis of available serial radiographs (24 cases) demonstrated signs of loosening (migration, progressive radiolucent lines, screw breakage) in 54% of the latest radiographs of all cases. Removing those lost to follow up or deceased, 72% were radiologically loose (in the intact asymptomatic group 57% could be defined as loose). Additionally, in 70% of the cases the acetabular component tended to fail in a varus manner as the medial wall remodelled.

These results indicate as previously published, short term results for this technique are satisfactory but in the long term they are not. Factors associated with this include the pressfit nature of the polyethylene liner which has been implicated in disappointing long term results for this prosthesis in primary applications, but not of the level of failure encountered in the current series. This experience suggests that the reliance on screw fixation over a bed of allograft in the absence of cement supplementation does not provide sufficient stability for reliable bone graft incorporation.

The method reported above should be abandoned.