Between September 1993 and September 1996, we performed 34 Kudo 5 total elbow replacements in 31 rheumatoid patients. All 22 surviving patients were reviewed at a mean of 11.9 years (10 to 14). Their mean age was 56 years (37 to 78) at the time of operation. All had Larsen grade IV or V rheumatoid changes on X-ray. Nine (three bilateral replacements and six unilateral) had died from unrelated causes. One who had died before ten years underwent revision for dislocation.

Of the 22 total elbow replacements reviewed six had required revision, four for aseptic loosening (one humeral and three ulnar) and two for infection. Post-operatively, one patient had neuropraxia of the ulnar nerve and one of the radial nerve. Two patients had valgus tilting of the ulnar component.

With revision as the endpoint, the mean survival time for the prosthesis was 11.3 years (95% confidence interval (10 to 13) and the estimated survival of the prosthesis at 12 years according to Kaplan-Meier survival analysis was 74% (95% confidence interval 0.53 to 0.91).

Of the 16 surviving implants, ten were free from pain, four had mild pain and two moderate. The mean arc of flexion/extension of the elbow was 106° (65° to 130°) with pronation/supination of 90° (30° to 150°) with the joint at 90° of flexion. The mean Mayo elbow performance score was 82 (60 to 100) with five excellent, ten good and one fair result.

Good long-term results can be expected using the Kudo 5 total elbow replacement in patients with rheumatoid disease, with a low incidence of loosening of the components.

The posterior capsule is variously incised and excised during total hip replacement (THR). There is no consensus on the direction of the capsulotomy and the need to repair the posterior capsule. The objective of this study was to determine the orientation of the collagen fibres and nerves in the posterior hip capsule in patients undergoing THR.

Specimens from five patients with osteoarthrosis of the hip (with no fixed deformity) were obtained and fixed in 10% neutral buffered formalin. Sutures were placed to mark the head and trochanteric end before excising. A standard posterior approach was used. The samples were examined and reported by a pathologist. Samples were processed overnight in a VIP5 automatic tissue processor and embedded in paraffin wax, preserving the location of the suture sites on embedding. Sections were cut at 5 Ïm and routinely stained with haematoxylin and eosin. The van Gieson stain was used for collagen fibres. Nerve fibres were highlighted using immunohistochemistry for S100 protein and blood vessels using an antibody to CD34.

The collagen bundles seen were predominantly parallel to the axis of the specimen. Dispersed within the collagen bundles were small vascular leashes that were parallel with the collagen fibres. The S100 staining revealed that these were neurovascular leashes, with small nerves running alongside the vessels and the collagen. Nerves that separate from the vessels were likely to serve proprioceptive and nociceptive functions.

The direction of the capsulotomy during THR by posterior approach has been traditionally perpendicular to the direction of the capsular fibres. However, if possible, capsulotomy along the orientation of the collagen fibres may be advantageous. As this study demonstrates, it will result in less damage to the capsular collagen fibres, blood vessels and nerves resulting in better capsular repair and healing, and better conservation of pro-prioceptive and nociceptive functions.