Severe acute respiratory syndrome (SARS) is a newly described infectious disease caused by the SARS coronavirus which attacks the immune system and pulmonary epithelium. It is treated with regular high doses of corticosteroids. Our aim was to determine the relationship between the dosage of steroids and the number and distribution of osteonecrotic lesions in patients treated with steroids during the SARS epidemic in Beijing, China in 2003.

We identified 114 patients for inclusion in the study. Of these, 43 with osteonecrosis received a significantly higher cumulative and peak methylprednisolone-equivalent dose than 71 patients with no osteonecrosis identified by MRI. We confirmed that the number of osteonecrotic lesions was directly related to the dosage of steroids and that a very high dose, a peak dose of more than 200 mg or a cumulative methylprednisolone-equivalent dose of more than 4000 mg, is a significant risk factor for multifocal osteonecrosis with both epiphyseal and diaphyseal lesions. Patients with diaphyseal osteonecrosis received a significantly higher cumulative methylprednisolone-equivalent dose than those with epiphyseal osteonecrosis. Multifocal osteonecrosis should be suspected if a patient is diagnosed with osteonecrosis in the shaft of a long bone.

Although alumina has been used in orthopaedic surgery since the 1970s, the long-term clinical results of zirconia have not been well documented in vivo. We studied hips with these two different ceramics during the same period and with a minimum follow-up of ten years. Because the size of the alumina and zirconia heads was different, hips with 32 mm alumina heads and those with 28 mm zirconia heads were compared with control hips with stainless-steel heads of the same size. Our aim was to compare the two ceramics.

There was an increased linear rate of penetration of the femoral heads into the liner between years five and 12 for the zirconia and the stainless-steel groups. This was severe in the zirconia group (0.4 mm/year compared with 0.13 mm/year for the stainless-steel group). During the same 12-year period there was, however, no significant change in the rate of wear in the alumina group (0.07 mm/year). The mean wear at the most recent follow-up was 1360 mm³ for the 28 mm zirconia group, 683 mm³ for the 28 mm stainless-steel group, 755 mm³ for the 32 mm alumina group and 1314 mm³ for the 32 mm stainless-steel group. The monoclinic content rose on the surface of three zirconia heads which were retrieved at revision. This change was associated with an increase in the surface roughness. A change in the roundness with an increase in the sphericity deviation was also observed both in the articular and non-articular parts of the femoral heads. The increase in rate of wear in the zirconia group was only evident after eight years and may be linked to a long-term biodegradation of zirconia in vivo, associated with the altered roughness and roundness which was observed on the retrieved heads.

We tested the accuracy of MRI for the precise quantification of the volume of osteonecrosis in 30 hips (stage III). The values were compared with direct anatomical measurements of the femoral heads obtained after total hip replacement. When the area of osteonecrosis was determined visually, and manually outlined on each slice, the accuracy of the measurement of volume was satisfactory, and the mean absolute deviation between MRI and anatomical measurements was similar to that between two MRI data sets. For ten of the hips which were measured by MRI, both before and after collapse, the volume did not appear to change significantly. Our findings suggest that the volume of osteonecrosis can be determined with accuracy by MRI, both before and after collapse.

The risk of articular penetration during tibial nailing is well known, but the incidence of unrecognised damage to joint cartilage has not been described. We have identified this complication in the treatment of tibial fractures, described the anatomical structures at risk and examined the most appropriate site of entry for tibial nailing in relation to the shape of the bone, the design of the nail and the surgical approach.

We studied the relationship between the intra-articular structures of the knee and the entry point used for nailing in 54 tibiae from cadavers. The results showed that the safe zone in some bones is smaller than the size of standard reamers and the proximal part of some nails. The structures at risk are the anterior horns of the medial and lateral menisci, the anterior part of the medial and lateral plateaux and the ligamentum transversum. This was confirmed by observations made after nailing 12 pairs of cadaver knees. A retrospective radiological analysis of 30 patients who had undergone tibial nailing identified eight at risk according to the entry point and the size of the nail. Unrecognised articular penetration and damage during surgery were confirmed in four.

Although intramedullary nailing has been shown to be a successful method for treating fractures of the tibia, one of the most common problems after bony union is pain in the knee. Unrecognised intra-articular injury of the knee may be one cause of this.

We have evaluated bone-marrow activity in the proximal femur of patients with corticosteroid-induced osteonecrosis and compared it with that of patients with osteonecrosis related to sickle-cell disease and with a control group without osteonecrosis. Bone marrow was obtained by puncture of the femoral head outside the area of necrosis and in the intertrochanteric region. The activity of stromal cells was assessed by culturing fibroblast colony-forming units (FCFUs).

We found a decrease in the number of FCFUs outside the area of osteonecrosis in the upper end of the femur of patients with corticosteroid-induced osteonecrosis compared with the other groups.

We suggest that glucocorticosteroids may also have an adverse effect on bone by decreasing the number of progenitors. The possible relevance of this finding to osteonecrosis is discussed.

Plain radiographs show only two dimensions of a three-dimensional object. On anteroposterior and lateral radiographs an implant may appear to be safely within the head of the femur although surface penetration has occurred.

We have attempted to identify this complication in the treatment of fractures of the femoral neck and have analysed the position of a screw or pin in the femoral head and neck on the basis of orthogonal frontal and lateral radiographs. A retrospective analysis of 60 cases of osteosynthesis of fractures of the femoral neck confirmed the risk of non-recognition of articular penetration or breaking of the cortex of the neck during surgery.

Unrecognised screw penetration of the hip was observed in 8% and of the posterior part of the neck in 10%. The risk differs according to the type of fracture: it is greater in the coxa valga produced by Garden-I fractures of the femoral neck.

From 1984 to 1988 we implanted 127 massive allografts irradiated with a dose of 25,000 grays. These were reviewed at a minimum follow-up of three years to determine the effect of irradiation on infection, the complications and the functional result. No bacteriological infection was seen in the 44 patients who had allografts for revision of joint arthroplasty or for a tumour with no adjuvant therapy. For the 83 patients who also had chemotherapy or radiotherapy or both for a bone tumour, the rate of infection was 13%. The major mechanical complications were nonunion in seven grafts (5.5%) and fracture in eight (6%). These rates do not differ greatly from those reported for non-irradiated grafts. Our results suggest that irradiation, which remains the most convenient and acceptable method of sterilisation, does not jeopardize the clinical results.

In ten patients with sickle-cell disease, we used a new technique of cement injection for the treatment of 16 painful hips with a radiographic crescent line or flattening of the articular surface due to avascular necrosis. The necrotic bone and overlying cartilage are elevated by the injection to restore the sphericity of the femoral head. Five days after the operation, full weight-bearing was allowed with the help of crutches for three weeks. The time in hospital averaged eight days; the average blood loss was 100 ml. There was early pain relief and postoperative radiographs showed improvement in the shape of the femoral head. At a mean follow-up of 5 years (3 to 7), 14 of the 16 hips were still improved although some gave slight pain. Only two hips had required revision to total hip arthroplasty, at one year and two years respectively. The increasing longevity of patients with sickle-cell disease means that avascular necrosis will be an increasing problem. Total hip replacement has a poor prognosis because of the risks of infection, high blood loss, and early loosening. Cement injection does not have these problems and allows for earlier, more conservative surgery.

We investigated the possible use of acrylic cement containing chemotherapeutic drugs in the treatment of malignant lesions in bone. The diffusion of methotrexate (MTX) from methylpolymethacrylate implants was studied in vitro: polymerisation of the cement did not destroy the drug; liberation began immediately and about 10% was released by 18 hours. Some release continued for as long as six months. In vivo experiments on rats with induced osteosarcoma showed that MTX in cement had both local and general effects which were dependent on the dosage. A series of 17 large dogs with spontaneous osteosarcoma were then treated by local resection and cement containing MTX. General chemotherapeutic effects were detectable from 2 hours to 5 days, survival was increased and local recurrence was reduced, but there were four cases of delayed wound healing. Preliminary studies in human patients confirm the possibility that this method of local chemotherapy could be a useful addition to the treatment of malignant tumours of bone.