Aim: To compare the outcomes of autologous versus no drainage in patients undergoing primary hip and knee arthroplasty.

Methods: Retrospective review of patients undergoing primary total hip or knee arthroplasty within our unit was carried out. There were 4 groups, each consisting of 20 patients. These were; total hip arthroplasty with and without the use of autologous drains, and total knee arthroplasty with and without the use of autologous drains. Outcome measures were- requirement for homologous (banked) blood, post-operative drop in haemoglobin and length of stay in hospital.

Results: For the knee arthroplasty groups, the homologous transfusion rate was 25% in non-drain patients compared with 5% in patients with drains. The average drop in Hb was 3.3 g/dl in the no-drain group compared with 2.7 in the drain group. The average length of stay was 6.35 days in the no-drain group compared with 6.75 days in the drain group. For the hip arthroplasty groups, the homologous transfusion rate was 10% in the non-drain group compared with 15% in the drain group. The average drop in Hb was 3.9 in the no drain group compared with 3.3 in the drain group. The average length of stay was 6.8 days in the no-drain group compared with 6.9 days in the drain group.

Conclusion: These results demonstrate improvements in the transfusion requirements as well as post-operative Hb drop in knee arthroplasty, but not hip arthroplasty when autologous re-transfusion drains were used. The length of stay was not significantly altered by the usage of drains. Based on these results, we would therefore recommend the use of autologous drains in primary total knee arthroplasty but not in primary total hip arthroplasty.

A retrospective review of our prospectively collected database was undertaken to determine the functional and oncologic outcome following combined pelvic allograft and total hip arthroplasty (THA) reconstruction of large pelvic bone defects following tumour resection. There were twenty-four patients with a minimum followup of fifteen months. The complication rate following hemipel-vic allograft and THA reconstruction of resection Types I+II and I+II+III was high, but when successful this reconstruction resulted in reasonable functional outcome. In comparison, the functional outcome after allograft and THA reconstruction of isolated Type II acetabular resections was better and more predictable.

Resection of large pelvic bone tumours often results in segmental defects with pelvic discontinuity and loss of the acetabulum. We reviewed the functional and oncologic outcomes following pelvic allograft and total hip arthroplasty (THA) reconstruction.

Reconstruction of large pelvic defects including the acetabulum using hemipelvic allograft and THA is associated with high complication rates, however when successful provides reasonable function. In comparison, the outcomes of allograft and THA for acetabular defects alone are better and more predictable.

A retrospective review of our prospectively collected database was undertaken. Minimum followup was fifteen months (15–167). Nineteen patients were hemipel-vic resections (twelve Type I+II and seven Type I+II+III, eleven cases including partial sacral resection) reconstructed by hemipelvic allograft and THA. Five patients had Type II acetabular resections, reconstructed with structural allograft, roof ring and THA.

Osteosarcoma and chondrosarcoma were the most frequent tumours. All patients required walking aids. In the hemipelvic group there were two early deaths (peri-operative haemorrhage and aplastic anaemia). In seven patients (37%) the allograft remained intact without infection but three required revision THA for loosening. For these seven patients the functional outcome scores were TESS 64%, MSTS87 17/35 and MSTS93 of 45% (mean fifty-two months.). There were nine cases of deep infection (47%) with three patients maintaining a functional implant. The nineteenth patient was revised following allograft fracture.

In the Type II acetabular group, three patients had no complications, and two patients dislocated. The average scores were TESS 78%, MSTS87 21/35 and MSTS93 64% (mean fifty-five months).

Introduction (Statement of purpose): Majority of the hips that are borderline on ultrasound progress to normal development subsequently, making the use of routine radiographs in follow up unnecessary. We present our experience in the last 5 years at the Musgrove Park Hospital in the management of borderline DDH

Materials and Methods: We studied 1452 patients who underwent an ultrasound examination for suspected DDH at Musgrove Park Hospital between January 1998 and December 2003. Ultrasound examination is performed in babies at a high risk for DDH or those who have abnormal hips on clinical examination at birth.42 babies were diagnosed to have dislocated or dislocatable hips and were treated with a harness. 239 babies, who had borderline dysplasia, had a repeat ultrasound at 6 weeks. Those with persistent borderline dysplasia had a radiographic and clinical examination at 6 months

Results: 60 patients were reported as borderline on follow-up ultrasound and underwent radiographs at 6 months.49 cases had normal radiographs and were asymptomatic.3 patients had mild dysplasia and were followed up for 18 months before being discharged as normal.3 patients were lost to follow up.4 cases presented late and had to undergo surgical procedures

Conclusion: No patients having borderline dysplasia on ultrasound developed symptomatic hip dysplasia. Routine radiographs are probably unnecessary in the follow-up of babies with borderline dysplasia on ultrasound except Graf 2c stages, which are important to recognise. Selective ultrasound screening is likely to fail in picking up some cases in the population (0.016%)

Introduction and Aims: Resection of large pelvic bone tumors often results in segmental defects with pelvic discontinuity and loss of the acetabulum. We reviewed the functional and oncologic outcomes following pelvic allograft and total hip arthroplasty (THA) reconstruction.

Method: A retrospective review of our prospectively collected database was undertaken. Minimum follow-up was 15 months (range 15–167 months). Nineteen patients were hemipelvic resections (12 Type I+II and seven Type I+II+III, 11 of these cases included partial sacral resection) reconstructed by hemipelvic allograft and THA. In comparison, five patients had Type II acetabular resections, reconstructed with structural allograft, roof ring and THA. Functional outcome was assessed by the Toronto Extremity Salvage score (TESS) and the Musculoskeletal Tumor Society scores (MSTS87 and MSTS93).

Results: Osteosarcoma and chondrosarcoma were the most frequent tumors. All patients required walking aids. In the hemipelvic group there were two early deaths (peri-operative haemorrhage and aplastic anaemia). In seven patients (37%), the allograft remained intact without infection but three required revision THA for component loosening. For these seven patients, the functional outcome scores were TESS 64%, MSTS87 17/35 and MSTS93 45% (mean follow-up 52 months). There were nine cases of deep infection (47%) with three patients maintaining a functional implant with antibiotic suppression. Of the remaining six patients with infection, four patients required hindquarter amputation, one patient required allograft removal and the allograft fragmented in the remaining patient. The 19th patient was revised following allograft fracture. Five patients sustained at least one allograft fracture.

In the Type II acetabular group, three patients had no complications, and two patients sustained dislocations. The average scores were TESS 78%, MSTS87 21/35 and MSTS93 64% (mean follow-up 55 months).

Conclusion: Reconstruction of large pelvic defects including the acetabulum using hemipelvic allograft and THA is associated with high complication rates, however when successful provides reasonable function. In comparison, the functional outcome after allograft and THA reconstruction of isolated Type II acetabular resections was better and more predictable.

Splitting fractures of the humeral head are rare; part of the humeral head dislocates and the unfractured part remains attached to the shaft. We report eight cases in young patients. In five the diagnosis was made at presentation: three had minimal internal fixation using a superior subacromial approach, one had a closed reduction and one a primary prosthetic replacement. All five patients regained excellent function with no avascular necrosis at two years. In three the injury was initially unrecognised; two developed a painless bony ankylosis and one is awaiting hemiarthroplasty.

It is important to obtain the three trauma radiographic views to diagnose these unusual fractures reliably. CT delineates the configuration of the fracture. In young patients open reduction and internal fixation seems preferable to replacement of the humeral head, since we have shown that the head is potentially viable.

In paralytic lesions in which the triceps surae is the only active muscle in the leg, elongation or division of the tendo calcaneus alone may not be enough to prevent recurrence of equinus deformity. In 10 patients (13 limbs) with this pattern of muscle activity, equinus deformity was treated by hemitransplantation of the tendo calcaneus. At follow-up, which was beyond the end of growth in seven limbs, there was no recurrence of deformity in nine. In three of the four failures, a technical fault may have caused loss of activity in the transplanted part of the tendon. The two-stage operation described is recommended in the management of this pattern of paralytic deformity.