It has been suggested that Perthes’ disease is more prevalent in urban areas, and that the risk increases with deprivation. We present the findings of a preliminary analysis of Perthes’ disease in Northern Ireland, which is shown to have one of the highest national annual rates of incidence in the world (11.6 per 100 000). Of the 313 children diagnosed over a seven-year period, 311 were allocated to the enumeration districts of the 1991 census, thus allowing the incidence to be calculated using both spatial and non-spatial aggregation. The cases were grouped according to the size of the settlement from highly urbanised to open countryside and by level of area deprivation. While the incidence of Perthes’ disease was found to be associated with indicators of the level of deprivation for areas, there was no evidence to suggest that there was an increased risk in urban areas; the highest rate was found in the most deprived rural category.

Seven children who had partial arrest of the growth plate after neonatal arterial cannulation, developed obvious skeletal changes in adolescence. Cannulation of the femoral artery produced ischaemia which led to four cases of ipsilateral shortening of the lower limb and one of partial arrest of the proximal femoral physis with subsequent coxa valga. The two arrests in the upper limb affected the humerus, ulna and radius, and the radius alone, after cannulation of the brachial and radial arteries, respectively. These late effects of cannulation are not widely appreciated, and may occur as a result of thrombosis rather than extravasation.

Although the technique of autogenous acetabular bone grafting has been widely used to augment containment of the acetabulum in total hip arthroplasty (THA) for developmental dysplasia, the role of this technique in improving long-term results remains controversial.

We present the long-term results of cemented THA with acetabular bone grafting in 112 patients (133 hips) in order to clarify the factors which affect the outcome. The mean follow-up was for 12.3 years (8 to 24). Kaplan-Meier survivorship analysis predicted a rate of survival of the acetabular component at 15 years of 96% (95% confidence interval (CI) 92 to 99) with revision for aseptic loosening as the endpoint, and of 75% (95% CI 65 to 85) when radiological loosening was used. Parametric survivorship analysis using the Cox proportional-hazards model indicated that trochanteric nonunion, lateral placement of the socket, and delayed trabecular reorientation of the bone graft were risk factors for loosening of the acetabular component.

Our findings have shown that autologous acetabular bone grafting is of value for long-term success provided that the risk factors are reduced.

We reviewed 120 consecutive primary total hip replacements in 109 patients in whom a Freeman uncemented metal-backed component had been used. Acetabular components were used with a Freeman neck-retaining stem in one of three configurations: cemented smooth stem, uncemented smooth stem or uncemented ridged stem. After a mean follow up of 72 months (62 to 113) there were nine cases of aseptic loosening; a survivorship at eight years of 83%. In addition, more than one-third of the remaining surviving cups showed loosening radiologically. The Freeman acetabular component demonstrated a characteristic pattern of loosening on radiographs. The high incidence of aseptic loosening can be explained by poor design and material failure.

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.

Glass ionomer cement (Ionocem) was developed for use in bone surgery and is reported to be notably biocompatible. Between 1991 and 1994 we performed revision operations for aseptic loosening of arthroplasties of the hip on 45 patients using this material in its granulate form (Ionogran) mixed with homologous bone as a bone substitute. Of these 45 patients, 42 were followed up for a mean of 42 months. Early reloosening of the acetabular component has occurred in ten after a mean of 30 months. Histological examination showed large deposits of aluminium in the adjacent connective tissue and bone. Osteoblastic function and bone mineralisation were clearly inhibited. The serum levels of aluminium were also increased. The toxic damage at the bone interface caused by high local levels of aluminium must be seen as an important factor in the high rate of early reloosening. Our findings cast doubt on the biocompatibility of this material and we do not recommend continuation of its further use in orthopaedic surgery.

We have analysed associated factors in 164 patients with acute compartment syndrome whom we treated over an eight-year period. In 69% there was an associated fracture, about half of which were of the tibial shaft. Most patients were men, usually under 35 years of age. Acute compartment syndrome of the forearm, with associated fracture of the distal end of the radius, was again seen most commonly in young men. Injury to soft tissues, without fracture, was the second most common cause of the syndrome and one-tenth of the patients had a bleeding disorder or were taking anticoagulant drugs. We found that young patients, especially men, were at risk of acute compartment syndrome after injury. When treating such injured patients, the diagnosis should be made early, utilising measurements of tissue pressure.

We performed an audit of 71 children with consecutive displaced, extension-type supracondylar fractures of the humerus over a period of 30 months. The fractures were classified according to the Wilkins modification of the Gartland system. There were 29 type IIA, 22 type IIB and 20 type III. We assessed the effectiveness of guidelines proposed after a previous four-year review of 83 supracondylar fractures. These recommended that: 1) an experienced surgeon should be responsible for the initial management; 2) closed or open reduction of type-IIB and type-III fractures must be supplemented by stabilisation with Kirschner (K-) wires; and 3) K-wires of adequate thickness (1.6 mm) must be used in a crossed configuration.

The guidelines were followed in 52 of the 71 cases. When they were observed there were no reoperations and no malunion. In 19 children in whom they had not been observed more than one-third required further operation and six had a varus deformity. Failure to institute treatment according to the guidelines led to an unsatisfactory result in 11 patients. When they were followed the result of treatment was much better. We have devised a protocol for the management of these difficult injuries.

The late results of treatment of calcaneal fractures in 17 children (19 fractures) were reviewed at a mean of 16.8 years after injury. With the exception of one patient, all fractures had been treated conservatively. At follow-up there were few complaints. All but two patients had full or slightly reduced mobility of the subtalar joint and unrestricted foot function, including the ability to walk comfortably on uneven surfaces. Minor radiological abnormalities of the hindfoot were common; there were two cases of post-traumatic osteoarthritis. Clinical scoring of the ankle and hindfoot using the American Orthopaedic Foot and Ankle Society rating system averaged 96.2 points.

The results suggest that up to 16.8 years after injury almost all children achieve excellent long-term functional results with conservative treatment of fractures of the os calcis. Open management may only be appropriate for adolescents with severe displacements.

We have carried out a prospective, longitudinal 30-year review of 95 adolescents who underwent total meniscectomy in one knee, and have compared the results with those observed 13 years earlier. All the medical records were scrutinised. Of the 63 patients reviewed clinically, 47 reported decreased sporting activity, although subjective satisfaction rose by 3% to 71%. The scores on the WOMAC osteoarthritis index differed significantly between patients grouped by subjective global assessment. Satisfactory function scores increased from 48% to 60%. In the 53 patients consenting to bilateral radiography of the knee, the incidence of narrowing of the articular cartilage in the operated knee increased significantly between the reviews (19% to 36%). Progression of degenerative change paralleled reduction in activity. Outcome measures were best after medial, intermediate after lateral and worst after double meniscectomy.

Accurate quantitative measurements of micromovement immediately after operation would be a reliable indicator of the stability of an individual component. We have therefore developed a system for measuring micromovement of the tibial component using three non-contact displacement transducers attached to the tibial cortex during total knee arthroplasty (TKA). Using this system we measured the initial stability in 31 uncemented TKAs. All the tibial components were fixed by a stem and four screws. The initial stability was defined as the amount of displacement when a load of 20 kg was applied. The mean subsidence was 60.7 μm and the mean lift-off was 103.3 μm.

We also studied the migration of the tibial component using roentgen stereophotogrammetric analysis (RSA) for up to two years after operation. Most migration occurred during the first six months, after which all prostheses remained stable. We defined migration as the maximum total point motion (MTPM) at two years after operation. The mean migration was 1.29 mm at two years.

Our results show that there was a significant correlation between the initial stability and migration (p < 0.05) and emphasise the importance of the initial stability of the tibial component.

We assessed the unloading effect of the patellar tendon-bearing (PTB) cast in five healthy volunteers using a new system for analysis of dynamic plantar pressure. We devised a method to improve the unloading effect of the PTB cast, and tested this using the same system.

Our findings showed that the conventional PTB cast only achieved unloading of 30% of the body-weight and that the part of the cast on the leg had a more important role in the unloading than that which was in contact with the patellar tendon. When the depth of the free space under the foot inside the PTB cast was 1, 2 and 3 cm, the unloading effect was 60%, 80% and 98%, respectively.

The unloading effect of the conventional PTB cast was disappointing at only 30% of body-weight. It was improved by producing a space between the sole of the foot and the cast, and was adjustable by altering the depth of this space.

Between 1986 and 1994, 13 patients with mobile painful arthritic elbows were treated by distraction interposition arthroplasty using fascia lata. The mean period of follow-up was 63 months. An elbow distractor/fixator was applied for three to four weeks to separate the articular surfaces and to protect the fascial graft.

Nine of the 13 patients (69%) had satisfactory relief from pain; eight (62%) had an excellent or good result by the objective criteria of the Mayo Elbow Performance score. Four have required revision to total elbow arthroplasty at a mean of 30 months with good results to date.

Instability of the elbow, both before and after surgery, was found to be associated with unsatisfactory results. The rate of success when the procedure was performed for inflammatory arthritis was similar to that for post-traumatic arthritis, about 67%. Eight complications occurred in six patients, all in the group with post-traumatic arthritis. Two of these required further surgical procedures such as transposition of the ulnar nerve or repair of hernia of the fascia lata.

Although less reliable than prosthetic replacement, distraction interposition arthroplasty is a useful option in the treatment of young, high-demand patients with arthritis of the elbow. It is rarely indicated in the presence of generalised inflammatory arthritis, but may be of value in those patients in whom the disease is limited primarily to the elbow.

Patients with short congenital amputations below the elbow often function as if they have had a disarticulation of the elbow. We have reviewed the results in six patients who had lengthening of such stumps by the Ilizarov technique to improve the fitting of prostheses.

The mean lengthening was 5.6 cm (3.4 to 8.4), and in two patients flexion contractures of the elbows were corrected simultaneously. Additional lateral distraction was used in one patient to provide a better surface on the stump. There were no major complications. All six patients were able to use their prosthesis at the latest follow-up after 39 to 78 months.

We enrolled 34 normal volunteers to test the hypothesis that there were two types of movement of the wrist. On lateral radiographs two distinct patterns of movement emerged. Some volunteers showed extensive rotation of the lunate with a mean range of dorsiflexion of 65°, while others had a mean range of 50°. The extensive rotators were associated with a greater excursion of the centre of articulation of the wrist. It is suggested that dynamic external fixation of a fracture of the distal radius carries with it the risk of stretching the ligaments or causing volar displacement at the site of the fracture.

We randomised prospectively 60 consecutive patients who were undergoing internal fixation of similar fractures of the ankle into two groups, one of which was treated by immobilisation in a below-knee cast and the other by a functional brace with early movement. All were instructed to avoid weight-bearing on the affected side. They were seen at 6, 12, 26 and 52 weeks. The functional rating scale of Mazur et al was used to evaluate the patients at each follow-up and we recorded the time of return to work. After one year the patients completed the SF-36 questionnaire.

By then 55 patients remained in the study, 28 (mean age 45.5 years) in group 1 and 27 (mean age 39.5 years) in group 2. Those in group 2 had higher functional scores at each follow-up but only at six weeks was this difference significant (p = 0.02). They also had higher mean SF-36 scores, but this difference was significant only for two of the eight aspects investigated. For patients gainfully employed, not on workers’ compensation, the mean time from surgery to return to work was 53.3 days for group 2 and 106.5 days for group 1; this difference was significant (p = 0.01). No patient developed a problem with the wound or had loss of fixation.

Our findings support the use of a functional brace and early movement after surgery for fractures of the ankle.

We carried out a cross-sectional study in 51 patients (81 feet) with a clawed hallux in association with a cavus foot after a modified Robert Jones tendon transfer. The mean follow-up was 42 months (9 to 88). In all feet, concomitant procedures had been undertaken, such as extension osteotomy of the first metatarsal and transfer of the tendon of the peroneus longus to peroneus brevis, to correct the underlying foot deformity. All patients were evaluated clinically and radiologically.

The overall rate of patient satisfaction was 86%. The deformity of the hallux was corrected in 80 feet. Catching of the big toe when walking barefoot, transfer lesions and metatarsalgia, hallux flexus, hallux limitus and asymptomatic nonunion of the interphalangeal joint were the most frequent complications. Hallux limitus was more likely when elevation of the first ray occurred (p = 0.012). Additional transfer of the tendon of peroneus longus to peroneus brevis was a significant risk factor for elevation of the first metatarsal (p < 0.0001).

The deforming force of extensor hallucis longus is effectively eliminated by the Jones transfer, but the mechanics of the first metatarsophalangeal joint are altered. The muscle balance and stability of the entire first ray should be taken into consideration in the management of clawed hallux.

Limb salvage is now customary in the treatment of primary bone tumours. The proximal tibia is a frequent site for these neoplasms but reconstruction, especially in children, is a formidable challenge. We reviewed 20 children with extendible replacements of the proximal tibia, all with a minimum follow-up of five years. Five died from their disease and, of the remaining 15, four had above-knee amputations for complications. Infection occurred in seven patients; in five it was related to the lengthening procedure. Aseptic loosening is inevitable in the younger children and only two have avoided a revision, amputation or other major complication; both were aged 12 years at the time of the initial surgery. Despite this, 11 children are alive with a functioning leg and a mean Musculoskeletal Tumour Society functional score of 83%. The lengthening mechanisms used in our series required extensive open operations. We are now using a simpler, minimally invasive, technique which we hope will decrease the incidence of complications. At present, the use of extendible prostheses of the proximal tibia remains an experimental procedure.

Multiple hereditary exostoses is an autosomal dominant skeletal disorder in which there are numerous cartilage-capped excrescences in areas of actively growing bone. The condition is genetically heterogeneous, and at least three genes, ext1, ext2 and ext3 are involved. The reported risk for malignant transformation to chondrosarcoma has been from 0.6% to 2.8%. We have reviewed six generations of a family with 114 living adult members, 46 of them with multiple exostoses. Four have had operations for chondrosarcoma, giving the risk for malignant transformation as 8.3% in this family. Clinical and radiological examination revealed two additional patients with a suspicion of malignancy, but in whom the histological findings were benign. Reported elsewhere in detail, genetic linkage analysis mapped the causative gene to chromosome 11 and molecular studies revealed a guanine-to-thymine transversion in the ext2 gene. Patients with multiple hereditary exostoses carry a relatively high risk of malignant transformation. They should be informed of this possibility and regularly reviewed.

An 18-year-old man who presented with weakness in his lower limbs, had an upper motor neurone lesion at the D12-L1 level. At laminectomy two stone-like objects were found which proved to be bundles of tiny pieces of wood. They are thought to have entered the cord through an abdominal penetrating injury sustained six years previously.

We treated 31 patients aged 65 years or more with cervical spondylotic myelopathy by microsurgical decompression and fusion at a single most appropriate level, in spite of MRI evidence of compression at several levels. Spinal cord potentials evoked at operation localised the level responsible for the principal lesion at C3-4 in 18 patients, C4-5 in 11 and at C5-6 in two. Despite the frequent coexistence of other age-related conditions, impairing ability to walk, the average Nurick grade improved from 3.5 before operation to 2.2 at a mean follow-up of 48 months. There was also good recovery of finger dexterity and sensitivity.

Operation at a single optimal level, as opposed to several, has the advantage of minimising complications, of particular importance in this age group.

We studied the effects of irradiation on the reintegration of autologous osteoarticular grafts over a period of 24 weeks in a canine model. In 16 foxhounds the medial femoral condyle was resected, irradiated and immediately replanted. In the control group resection and replantation were performed without irradiation. Reintegration was assessed by macroscopic analysis, histology, radiography and gait analysis.

Reintegration was equal at 12 weeks, but significantly inferior in the irradiated group after 24 weeks with delayed bone remodelling. The articular cartilage showed modest degeneration. Conventional radiography and histology showed corresponding changes. Limb function was adequate but the gait was inferior in the treated group.

Periprosthetic osteolysis is a major cause of aseptic loosening in artificial joint replacement. It is assumed to occur in conjunction with the activation of macrophages. We have shown in vitro that human osteoblast-like cells, isolated from bone specimens obtained from patients undergoing hip replacement, phagocytose fine particles of titanium alloy (TiAlV).

The human osteoblast-like cells were identified immunocytochemically by the presence of bone-specific alkaline phosphatase (BAP). With increasing duration of culture, a variable number of the osteoblastic cells became positive for the macrophage marker CD68, independent of the phagocytosis of particles, with a fine granular cytoplasmic staining which was coexpressed with BAP as revealed by immunodoublestaining. The metal particles were not toxic to the osteoblastic cells since even in culture for up to four weeks massively laden cells were vital and had a characteristic morphology. Cells of the human osteosarcoma cell line (HOS 58) were also able to phagocytose metal particles but had only a low expression of the CD68 antigen. Fluorescence-activated cell scanning confirmed our immunocytochemical results. Additionally, the cells were found to be negative for the major histocompatibility complex-II (MHC-II) which is a marker for macrophages and other antigen-presenting cells. Negative results of histochemical tests for tartrate-resistant acid phosphatase excluded the contamination by osteoclasts or macrophages in culture.

Our observations suggest that the osteoblast can either change to a phagocytosing cell or that the phagocytosis is an underestimated property of the osteoblast. The detection of the CD68 antigen is insufficient to prove the monocytic lineage. In order to discriminate between macrophages and osteoblasts additional markers should be used. To our knowledge, this is the first demonstration of cells of an osteoblastic origin which have acquired a mixed phenotype of both osteoblasts and macrophages.

Coating titanium alloy implants with titanium nitride (TiN) by the method of Powder Immersion Reaction Assisted Coating (PIRAC) produces a stable layer on their surface. We have examined the ability of the new TiN coating to undergo osseointegration. We implanted TiN-coated and uncoated Ti6Al4V alloy pins into the femora of six-month-old female Wistar rats.

SEM after two months showed a bone collar around both TiN-coated and uncoated implants. Morphometrical analysis revealed no significant differences between the percentage of the implant-bone contact and the area and volume of the bone around TiN-coated compared with uncoated implants. Electron-probe microanalysis indicated the presence of calcium and phosphorus at the implant-bone interface. Mineralisation around the implants was also confirmed by labelling with oxytetracycline. Strong activity of alkaline phosphatase and weak activity of tartrate-resistant acid phosphatase were shown histochemically. Very few macrophages were detected by the non-specific esterase reaction at the site of implantation.

Our findings indicate good biocompatibility and bone-bonding properties of the new PIRAC TiN coatings which are comparable to those of uncoated Ti6Al4V alloy implants.

After cemented total hip arthroplasty (THA) there may be failure at either the cement-stem or the cement-bone interface. This results from the occurrence of abnormally high shear and compressive stresses within the cement and excessive relative micromovement.

We therefore evaluated micromovement and stress at the cement-bone and cement-stem interfaces for a titanium and a chromium-cobalt stem. The behaviour of both implants was similar and no substantial differences were found in the size and distribution of micromovement on either interface with respect to the stiffness of the stem.

Micromovement was minimal with a cement mantle 3 to 4 mm thick but then increased with greater thickness of the cement. Abnormally high micromovement occurred when the cement was thinner than 2 mm and the stem was made of titanium.

The relative decrease in surface roughness augmented slipping but decreased debonding at the cement-bone interface. Shear stress at this site did not vary significantly for the different coefficients of cement-bone friction while compressive and hoop stresses within the cement increased slightly.