We reviewed the results of 14 total hip replacements in patients with juvenile chronic arthritis. The mean age at operation was 16 years (range 12 to 22 years); follow-up was from four to 11 years (mean 8.5 years). Postoperatively pain relief was sustained in all but one hip, while movement generally remained significantly restricted. No hip has as yet required a revision operation, although eight hips (57%) show radiological changes suggestive of impending failure. All patients had severe polyarticular involvement with associated restriction of locomotor activity. Potential causes contributing to loosening such as continuing diaphyseal bone growth and increased immunocompetence in adolescence are discussed.

We describe a method of internal fixation for occipito-cervical fusion utilising a standard "small fragment" T-plate bent and fixed to the skull with three screws. The lower end of the plate is screwed and wired firmly to the spine of the axis. Of 14 patients so treated, 12 fused, one died and one failed to unite to the skull. Of eight with cord signs, seven remitted or improved and one died.

We present a study of 30 fusion operations in 26 rheumatoid arthritics with cervical spine instability. Atlanto-axial instability was present in 15, of whom 12 were fused; three had cord involvement and all made a partial or complete recovery following fusion. Cranial settling necessitated cranio-cervical fusion in four patients; all fused, and one with myelopathy was relieved. Subaxial instability required fusion in seven cases; two postoperative deaths followed the only two anterior interbody fusions. Posterior fusion was successful in the other five, with remission of neurological compromise in the three with myelopathy and one with radiculopathy. We conclude that neurological compromise in an unstable but mobile rheumatoid cervical spine can usually be brought to remission by immobilisation alone, so decompressive procedures are unnecessary in the first instance.

We report seven cases of patients with seropositive rheumatoid arthritis in whom involvement of the thoracic and lumbar vertebrae occurred. Histological corroboration is presented in four. Pathological lesions comprised various combinations of paravertebral joint erosions, erosive discitis, anteroposterior and rotatory instability, major lumbar nerve root compression, and vertebral collapse. Specific radiological features are presented, enabling a distinction to be made between pure degenerative spondylosis and rheumatoid spondylitis. We submit that subcervical rheumatoid spondylitis is commoner than is generally believed, though less common than rheumatoid involvement of the cervical spine.

A high percentage of hips in patients with rheumatoid arthritis presenting for total joint replacement have protrusio deformity with extremely thin medial acetabular walls. Biomechanical principles suggest that, to prevent inward migration of the acetabular component, the prosthesis should be positioned laterally near the acetabular rim. To reinforce the medial wall, the femoral head is fashioned into a solid graft which is anchored against the medial wall as a plug, minimising the amount of cement used and making the use of wire mesh and restraining rings about the acetabulum unnecessary. A series of nine operations on seven patients is presented. Serial radiographs, tomograms and scintigrams taken after operation suggest satisfactory incorporation of the grafts, and no complications have been encountered in a follow-up period ranging from three months to two years.