Between January 1984 and January 1986, 131 porous-coated anatomic total hip replacements were performed in 119 consecutive patients. Of these, 100 patients (110 hips) who were alive at a minimum of 18 years after replacement were included in the study. The mean age of the patients at surgery was 48.4 years and that of surviving patients at the latest follow-up was 67.8 years. The mean duration of the clinical and radiological follow-up was 19.4 years (18 to 20).

The mean Harris hip score initially improved from 55 points before to 95 points at two years after operation, but subsequently decreased to 91 points after six years, 87 points after 11.2 years, and 85 points after 19.4 years. At the final follow-up, 23 hips (18%) of the entire cohort and 21% of survivors had undergone revision because of loosening or osteolysis of the acetabular component and eight hips (6%) of the entire cohort and 7% of survivors for loosening or osteolysis of the femoral component. Only four femoral components (4%) were revised for isolated aseptic loosening without osteolysis and two (2%) for recurrent dislocation. On the basis of these long-term results, the porous-coated anatomic femoral component survived for a minimum of 18 years after operation while the acetabular component was less durable. The findings identify the principles of uncemented joint replacement which can be applied to current practice.

1. Hemiarthroplasty is a method of dealing with painful deformities of advanced osteoarthritis and rheumatoid arthritis of the knee.

2. The indications and contra-indications for this procedure are discussed. Careful selection of patients is essential.

3. The technique of operation and management after operation are described.

4. The results of such a procedure, as done by one surgeon, are given. Good results have been obtained in 80 per cent of the osteoarthritic knees and in 69 per cent of the rheumatoid knees.

5. The complications, place of associated operations and value of revision procedures are discussed.

1. A complete replacement arthroplasty of the hip is described in which both components are inherently stable.

2. The arthroplasty does not require the use of acrylic cement.

3. It is appropriate for the treatment of the severely arthritic hip in which arthrodesis is not indicated, and for the mobilisation of two stiff and painful hips at any age.

4. It can be performed on both sides at the same time.

5. It produces a stable, pain-free and mobile joint in a high proportion of cases, and has appeared not to deteriorate over periods of up to four years.

During open reduction of an irreducible anterior dislocation of a total hip replacement with an Oxinium femoral head, it was observed that the head had been significantly damaged. Gross and scanning electron microscopic examination revealed cracking, gouging, and delamination of the surface. Because of the risk which this poses for damaging the polyethylene acetabular liner, it is strongly recommended that patients with this type of prosthetic head be carefully monitored after a dislocation.

Impacted morcellised bone allograft and a Charnley stem was used to revise 59 loose femoral components in 57 consecutive patients. Femoral bone loss was rated as Endo-Klinik grade 2 in nine patients, grade 3 in 41, and grade 4 in nine. The immediate postoperative radiographs and those taken at the most recent follow-up were compared for radiolucencies, subsidence and incorporation of the graft. One patient was lost to follow-up and two were not available for radiological analysis.

The mean clinical follow-up in 58 procedures was 56.7 months (24 to 144) and the mean radiological review of 56 reconstructions was 54.4 months (24 to 144). An intraoperative femoral fracture occurred in one patient (1.7%) and was successfully treated by strut grafting and cerclage wiring. Extrusion of cement through perforations or incomplete hoop fractures was detected in the postoperative radiographs of ten procedures (17%); none of these patients sustained a complete fracture. Three patients had dislocations (5%) and two (3.5%) developed painful subsidence of the stem which required a further revision. The latest follow-up radiographs in 56 reconstructions showed a well fixed stem and radiological healing of the graft in 52 (93%), and definite loosening in four (7%). Of these four, two were revised again and two were asymptomatic after a follow-up of 120 months each. The mean subsidence in the 52 successful revisions was 0.38 mm (0 to 4). Impaction allografting with a Charnley stem restored bone stock and provided adequate fixation of the stem in 93% of the hips. There was a low rate of rerevision (3.5%) and a low incidence of intraoperative and postoperative complications.

We have reviewed 202 consecutive primary hip replacements using a Freeman cemented femoral component after a mean period of follow-up of 64 months (23 to 113). There was only one revision for aseptic loosening. Retention of the femoral neck may act to reduce the torsional and shear forces at the implant-cement interface and may provide a seal against the migration of polyethylene-laden joint fluid in the potential joint space. The cemented Freeman femoral component with retention of the femoral neck was successful for up to nine years.

We report the clinical outcome and survival of 100 total hip arthroplasties with Link RS cementless hip prostheses after a mean of 63 months (51 to 93). The average preoperative Merle d'Aubigne hip score was 9.4 points and the average score at the latest follow-up was 15.3 points. Thirteen prostheses have been revised and seven await revision. Survivorship analysis based on intention to revise showed only 68% survival at seven years (95% CI 53% to 81%). The survivals of femoral and acetabular components verified to be loose at surgery were 78% and 89% respectively at seven years. A number of other patients have pain and it seems likely that there will be substantial reduction in these survival figures in the future.

Sixty-five total hip arthroplastics were reinserted after sepsis around the hip, positive cultures being obtained from fifty-six. Although 65 per cent of patients still have their implant in position, only sixteen of sixty-five (25 per cent) show an excellent or good result on a Harris rating. Twenty-three of sixty-five (35 per cent) subsequently required an excision arthroplasty. The indications and contraindications for this procedure are given.