We have reviewed the results in 34 patients (39 operations) following simple excision of the trapezium for osteoarthritis of the basal joint of the thumb. The average age at operation was 57 years and the average follow-up was 6 years. All the patients were graded clinically and radiologically and were asked their opinion of the procedure. There was dramatic relief of pain following this procedure. Stability of the thumb was not compromised. When compared to the unoperated side, thumb length, thumb abduction and first web span were similar. There was a reduction in pinch strength (operated 8.1 k.p.a., non-operated 9.6 k.p.a.) and grip strength (operated 15.5 k.p.a., non-operated 19.5 k.p.a.) and an increase in MIP extension (operated 5.4 degrees, non-operated 2.9 degrees) following this procedure but the differences were not statistically significant. 11 patients (32%) had scar hyperaesthesia on testing but this was a clinical problem in two patients only (5%). Simple excision of the trapezium is a satisfactory procedure for the majority of patients with this disorder, but has a long post-operative rehabilitation period.

We studied serial CT scans of 45 arthritic shoulders (34 rheumatoid, 11 osteoarthritic) and 19 normal shoulders, making measurements at three levels on axial images. The maximum anteroposterior diameter of the glenoid was increased in rheumatoid glenoids at the upper and middle levels by 6 mm and in osteoarthritic glenoids at all levels by 5 to 8 mm as compared with normal. In rheumatoid cases, nearly half the available surface ofthe glenoid was ofunsupported bone, mainly posteriorly at the upper and middle levels. In osteoarthritic glenoids, the best supported bone was anterior at the upper level and central at the middle and lower levels. The depth of the rheumatoid glenoid was reduced by a mean of 6 mm at the upper and middle levels and by 3 mm at the lower level. This inclined the surface of the glenoid superiorly. The depth at the middle level in osteoarthritis was reduced by a mean of 5 mm, suggesting central protrusion. Osteoarthritic glenoids were retroverted by a mean of 12.5#{176}, but of rheumatoid glenoids two-thirds were retroverted (mean 15.1#{176}) and one-third anteverted (mean 8.2#{176}). Our findings have important implications for the planning and placement of the glenoid component of total shoulder replacements; CT can provide useful information.

We present an approach to the shoulder which is easily performed, offers unrivalled exposure for the accurate placement of prostheses or reconstruction of complicated fractures and is easily closed. We have used this approach since 1977 and report our results in 128 patients.