We describe a management strategy for upper- and lower-limb fractures with associated arterial injury and report the results in 113 cases treated over a period of 18 years. Primary amputation was performed in 23 patients and of those who underwent primary vascular repair, 27 needed secondary amputation, two-thirds of them within a week of the injury. Of those requiring secondary amputation, 51.8% had ischaemia exceeding six hours, 81.4% had severe soft-tissue injury and 85.2% had type III open fractures. The patients whose limbs had been salvaged were followed up for an average of 5.6 years. The eventual outcome depended on the severity of the fracture, the degree of soft-tissue damage, the length of the ischaemic period, the severity of neurological involvement, and the presence of associated major injuries. There was a 30% incidence of long-term disability in the salvaged limbs, largely due to poor recovery of neurological function. Prompt recognition of such combined injuries is vital and requires a high index of suspicion in patients with multiple injuries and with certain fracture patterns. We recommend a multidisciplinary approach, liberal use of pre-operative angiography in upper-limb injuries and selective use of intra-operative angiography in lower-limb injuries. Stable external or internal fixation of the fractures and re-establishment of limb perfusion are urgent surgical priorities to reduce the period of ischaemia which is critical for successful limb salvage.

In an eight-year period we treated 51 cases of vascular injury associated with fractures and/or dislocations or soft-tissue injuries of the limbs. We relied on a clinical diagnosis and immediate exploration of blood vessels rather than the time-consuming procedure of arteriography. All patients were operated on by the orthopaedic residents on duty and not by vascular surgeons. Only 17 (33%) were repaired within six hours of injury. Limb viability with good function was obtained in 38. Complications included six deaths, four amputations, two renal failures and delayed occlusion in one case.

We measured pressures in the anterior and deep posterior compartments continuously for up to 72 hours in 20 patients with closed fractures of the tibial shaft treated primarily in plaster casts. All were examined independently after periods of three to 14 months. Pressures above 40 mmHg occurred in seven (35%) and above 30 mmHg in 14 (70%). No patient had the symptoms of compartment syndrome during monitoring. Abnormalities at review did not correlate with the maximum consecutive time periods during which the compartment pressures were raised. Thus, in the absence of symptoms the monitored pressures did not relate to outcome. Routine monitoring in this type of patient is therefore of doubtful benefit.

We report a prospective study of the influence of various factors on the six-month mortality of 531 patients with subcapital hip fractures. We performed univariate and multivariate analyses on the 403 patients treated surgically. The most significant predictors of the six-month mortality were dementia, postoperative chest infection, malignant neoplasia, old age and deep-wound infection, in that order. A simple test of mental ability was the most significant prognostic indicator and this test should be included in future studies of the management of hip fractures in the elderly.

We studied prospectively a consecutive series of 765 patients with proximal femoral fractures to determine if the time interval between injury and surgery influenced the outcome. Patients in whom surgery had been delayed for medical reasons were excluded. We divided the patients into four groups depending on the delay to surgery. Analysis of pre- and postoperative characteristics showed the groups to be similar. Mortality in the four groups was not significantly different but morbidity was increased by delay, particularly with regard to the incidence of pressure sores.

Forty patients with acute complete rupture of the calcaneal tendon were managed conservatively and randomly allocated to treatment groups using either cast immobilisation for eight weeks, or cast immobilisation for three weeks, followed by controlled early mobilisation in a Sheffield splint. The splint is an ankle-foot orthosis which holds the ankle in 15 degrees of plantar flexion, but allows some movement at the metatarsophalangeal joints. It is removed to allow controlled movement during physiotherapy. Patients treated with the splint regained mobility significantly more quickly (p less than 0.001) and preferred the splint to the plaster cast. The range of dorsiflexion at the ankle improved more rapidly after treatment in the splint (p less than 0.001), and patients were able to return to normal activities sooner. Recovery of the power of plantar flexion was similar in the two treatment groups, and no patient had excessive lengthening of the tendon. One re-rupture occurred in each group.

Of 55 patients with Duchenne muscular dystrophy offered surgical stabilisation of the spine, 32 accepted and 23 refused. We compared both groups pre-operatively and at six-month intervals in respect of survival, forced vital capacity, peak expiratory flow rate and severity of scoliosis. In the nonoperated patients, the forced vital capacity deteriorated by a mean of 8% per annum; in the operated group it remained static for 36 months and diminished slightly thereafter. Spinal stabilisation resulted in an improvement in the peak expiratory flow rate which was maintained for up to five years. In the nonoperated patients the scoliosis progressed from a mean of 37 degrees to a mean of 89 degrees at five years; in the stabilised spines it was improved from a mean of 47 degrees to a mean 34 degrees at five years. There was significantly improved survival in the patients who had undergone spinal stabilisation.

From 1981 to 1986 we treated 413 patients for acute spinal-cord injuries. We reviewed 356 patients followed for a minimum of two years of whom 71 (20%) developed heterotopic ossification around one or more joints. Heterotopic ossification occurred more often in male patients (23%) than in female (10%), and was most frequent in the 20- to 30-year age group. It was also more common after injuries of the lower cervical or thoracic spine than after those of the lumbar spine. Patients with severe neurological deficits (Frankel grades A and B) showed significantly more heterotopic ossification but there was no correlation with the number or severity of associated head and limb injuries. Serum calcium levels did not change significantly in either group for 30 weeks after injury, but the erythrocyte sedimentation rate and the alkaline phosphatase level were significantly increased at six weeks in patients with heterotopic ossification.

We report 13 patients with missed bilateral facet dislocation of the lower cervical spine who subsequently developed severe spinal-cord involvement. There were more women and the patients were older than in most groups with spinal injury. The commonest cause was a fall, and paralysis appeared from six to 48 hours after injury in most patients. Ten patients made some recovery after late reduction but three remained totally paralysed. We discuss the pathogenesis of the late cord lesion and the reasons for delay in diagnosis.

We report three patients with neurological deterioration after the reduction of cervical spine dislocation. In each case compression of the spinal cord by disc material was demonstrated by magnetic resonance imaging and recovery occurred following removal of the compressing disc. A separate radiological study suggested that disproportionate narrowing of the disc space implies disc extrusion; reduction of a dislocation showing this sign may produce or exacerbate neurological deterioration.

We recovered 23 meniscal bearings from 18 failed bicompartmental Oxford knee prostheses. They had been implanted for one to nine years. The minimum thickness of the retrieved bearings was measured and compared with the thickness of 25 unused bearings. The mean penetration rate, calculated by two methods, was either 0.043 or 0.026 mm per annum. This compares with 0.19 mm per annum reported for the Charnley hip. The use of a fully congruous meniscal bearing prosthesis can reduce wear in knee arthroplasty to a very low rate.

We studied the effect of a layer of cement placed under the tibial component of Freeman-Samuelson total knee prostheses with a metal back and an 80 mm intramedullary stem, using roentgen stereophotogrammetry to measure the migration of the tibial component during one year in 13 uncemented and 16 cemented knees. The addition of cement produced a significant reduction in migration at one year, from a mean of 1.5 mm to one of 0.5 mm (p less than 0.01), including a significant reduction in pure subsidence. One year postoperatively the clinical results were similar between the groups, but, at three years, one uncemented knee had required revision.

We determined the cumulative survival rates, in arthroplasty of the knee, of three designs of tibial component, using a change of position on standard radiographs or revision for aseptic loosening as criteria of failure. The average migration of each of the three designs in the first postoperative year is known from roentgen stereophotogrammetric analysis reported by other authors. The ranking order of the components as judged by cumulative survival is the same as that determined by early migration. This finding supports the view that the measurement of early migration can predict late aseptic loosening and therefore that such measurements are clinically of value.

We report the results of 75 consecutive primary press-fit Kinemax arthroplasties with an average follow-up of 14 months (three to 28). We reviewed 26 cemented and 49 non-cemented tibial components implanted into 72 patients (30 men and 42 women, median age 70 years). At the latest follow-up the overall evaluation (Hospital for Special Surgery knee rating scale) for cemented cases was excellent in 54%, good in 42% and poor in 4%. No cemented prosthesis loosened. Of the non-cemented cases 55% were excellent, 23% good, and 2% fair. Ten tibial implants (20%) loosened and required revision. Residual pain marred the result in 24% of the non-cemented prostheses and in 4% of the cemented group. We do not recommend the press-fit, smooth-surfaced Kinemax prosthesis for non-cemented use.

Proprioception was measured in the knees of 20 subjects with instability of the anterior cruciate ligament and compared with 17 age-matched control subjects. There was diminished position sense and threshold for movement detection in the injured patients compared with the control group. The proprioceptive deficit recorded from the injured knee showed a significant correlation with the hamstring/quadriceps power ratio recorded from the injured leg.

We implanted 203 smooth-stemmed femoral components before January 1988. The femoral component used was anatomically shaped, fluted and made of titanium. Thirty-two hips were revised due to mid-thigh pain, and the femoral implant was found to be loose in all. In the 157 patients with a two-year follow-up, the Merle d'Aubigne and pain scores for completely cementless arthroplasties were similar to those for hybrid prostheses (cemented acetabular cup and cementless femoral stem). Of the 145 cases with two-year radiographic follow-up, 59 had extensive radiolucencies and 22 were unstable. The five-year cumulative survival rate was 77%. Implantation of this stem should be restricted to patients in whom cement fixation is contra-indicated.

Retrospective review of 730 consecutive primary uncemented and cemented total hip arthroplasties revealed 19 intra-operative hoop-stress fractures of the femoral neck. These were incomplete, linear, and minimally displaced. Management was by cerclage wiring (12), bone graft and cerclage (two), further impaction (two), and the use of cement (three), with no change from our standard postoperative management and rehabilitation. Eighteen patients had excellent or good results with an average Harris hip score of 93. Radiographically, all but one patient had Engh stability-fixation scores consistent with stable bone ingrowth. We conclude that hoop-stress fractures of the proximal femur, properly managed, do not detract from the results of total hip arthroplasty.

We have determined the quality-adjusted-life-years and cost of such in several types of hip fracture and various treatment options. Operative treatment proved more cost-effective than other methods for displaced subcapital fractures and for extracapsular fractures. For undisplaced subcapital fractures conservative treatment was almost as cost-effective as surgery.

In 78 consecutive patients, we performed fine-needle aspiration of the hip before revision surgery. At the revision operation biopsies were taken from the capsule and joint tissues. The aspirate and the homogenised soft-tissue specimens were cultured using a radiometric technique. Fifteen hips proved to be infected and 63 sterile on culture of the tissues taken during revision surgery; 94% of the aspiration results were correct. The sensitivity of the technique was 87% and the specificity 95%. Fine-needle aspiration combined with a radiometric culture technique is a simple and reliable method of diagnosing infection of hip prostheses.

In a population-based study we identified a cohort of 282 women, whose children had been treated for osteoarthritis of the hip. The incidence of hip fracture in these mothers was 0.54 (95% CI 0.35 to 0.83) of that in the whole population of their age and sex. This finding could be explained by a genetic factor for increased rigidity of subchondral bone, making osteoarthritis more likely and osteoporosis less likely.

We perfused 16 human femora with a 50% barium sulphate suspension and studied the intra-osseous vessels by microfocal radiography and histology. There were few anastomoses between the vessels of the greater trochanter and those of the adjacent cancellous bone of the shaft. Ischaemia of the trochanter may contribute to nonunion after trochanteric osteotomy.

The transiliac method of leg lengthening uses a modification of Salter's innominate osteotomy. The bone graft increases the length of the hemipelvis distal to the sacro-iliac joint. Leg-length inequality in 23 patients was treated by this method with an average gain in length of 2.8 cm (2.0 to 3.5). Apart from one residual femoral nerve palsy there were no notable complications. The facility to redirect the acetabulum allowed by the technique, may be useful in cases of potential hip instability or acetabular dysplasia.

Ossification of the ligamentum flavum and secondary spinal-cord compression were produced experimentally in mice by implanting bone morphogenetic protein (BMP) in the lumbar extradural space. The ligamentum flavum became hypertrophied and ossified, and protruded into the spinal canal. The thickness of the ossified ligament increased gradually with time, leading to compression and deformation of the spinal cord which showed various degrees of degeneration. Demyelination occurred in the posterior and lateral white columns and neuronal loss or chromatolysis in the grey matter. The pathological findings in the experimental animals closely resemble those found in the human disease and suggest that BMP may be a causative factor of ossification of the ligamentum flavum in man. This experimental model may be useful for the study of myelopathy caused by gradual spinal-cord compression.

We cultured human osteoblasts from trabecular bone explants and confirmed their phenotype by alkaline phosphatase assay, increased cyclic adenosine monophosphate production in response to prostaglandin E2 and radiographic micro-analysis of nodules of calcification. The osteoblasts were seeded on to demineralised human bone fragments and examined at ten-day intervals over a 50-day period by scanning electron microscopy. During this time the bank bone became progressively repopulated by the cultured osteoblasts. This system may offer a means of graft enhancement in elective orthopaedic and maxillofacial surgery by delivery of cultured autologous human osteoblasts to bone defects.

We compared fibrin sealant, polydioxanone (PDS) pins and Kirschner wires in the fixation of osteochondral fractures in rabbit knees. Standardised osteochondral fractures of the right medial femoral condyle were made in 56 adult New Zealand white rabbits. There were equal groups of control knees, and those which had Kirschner-wire, fibrin-sealant or PDS-pin fixation. No external immobilisation was used. One animal from each group was killed at two, three and four weeks. The remaining rabbits were killed at six weeks. A fracture which healed with less than 1 mm of displacement was considered a success. There was successful healing in 29% of the control group, in all of the Kirschner-wire group, in 50% of the fibrin-sealant group, and in 86% of the PDS-pin group. The use of PDS pins appears to be a reliable alternative to the use of metal in the fixation of osteochondral fractures in rabbits.

We used 11 cadaver elbows and a three-dimensional electromagnetic tracking device to record elbow movements before and after implantation of a 'loose-hinged' elbow prosthesis (modified Coonrad). During simulated active motion there was a maximum of 2.7 degrees (+/- 1.5 degrees) varus/valgus laxity in the cadaver joints. This increased slightly after total elbow arthroplasty to 3.8 degrees (+/- 1.4 degrees). These values are lower than those recorded for the cadaver joints and for the prostheses at the limits of their varus/valgus displacements, indicating that both behave as 'semi-constrained' joints under physiological conditions. They suggest that the muscles absorb some of the forces and moments that in a constrained prosthesis would be transferred to the prosthesis-bone interface.

We reviewed 15 patients with an arthrodesis of the elbow using an AO compression-plate technique, after an average follow-up of 24 months. The most common indication was an open, infected high-energy injury with associated bone loss. Arthrodesis was successful in all but one patient in whom severe deep infection necessitated amputation. Eight patients were treated with the metal partly exposed in an infected wound. After removal of the metal, all wounds healed secondarily and none had clinical or radiographic signs of sepsis at latest review. Compression-plate arthrodesis of the elbow is a generally applicable method that can be used even in cases of severe bone loss. There appears to be greater certainty of union than with other techniques, and no increased risk of subsequent fracture.

We report overuse injuries in 14 elbows of ten elite young gymnasts. In 12 elbows of eight patients aged 11 to 15 years there was a spectrum of radiological abnormalities including widening of the olecranon physis and fragmentation of the epiphysis. The radiographs were compared with those of normal elbows matched for age and sex. The radiograph appearances were very similar to those of the Osgood-Schlatter lesion of the tibial tuberosity. Two older boys, aged 18 and 19, had stress fractures through the olecranon growth plate. Conservative management was successful in all except one case of stress fracture which required internal fixation.

We performed 20 de la Caffiniere trapeziometacarpal arthroplasties for osteoarthritis occurring only at this joint and reviewed all patients after periods of up to ten years. Eighteen arthroplasties were satisfactory postoperatively, although all 20 patients had a satisfactory range of motion and only one experienced pain after surgery such that it impeded normal function. Failure occurred in two patients and was due to overreaming of the trapezium during surgery and a traumatic dislocation. A radiolucency between the prosthesis and bone was observed in one arthroplasty, although this patient was asymptomatic. We recommend arthroplasty using the de la Caffiniere prosthesis as a satisfactory method of treatment for the osteoarthritic trapeziometacarpal joint.