We reviewed the medical records of 115 patients with 130 hips with developmental dysplasia with complete dislocation in the absence of a neuromuscular disorder, spontaneous reduction with a Pavlik harness, and a minimum of 14 years’ follow-up. The mean age at the time of harness application was 4.8 months (1 to 12) and the mean time spent in the harness was 6.1 months (3 to 12). A total of 108 hips (83.1%) were treated with the harness alone and supplementary surgery for residual acetabular dysplasia, as defined by an acetabular index > 30°, was performed in 22 hips (16.9%).

An overall satisfactory outcome (Severin grade I or II) was achieved in 119 hips (91.5%) at a mean follow-up of 16 years (14 to 32) with a follow-up rate of 75%. Avascular necrosis of the femoral head was noted in 16 hips (12.3%), seven of which (44%) underwent supplementary surgery and nine (56%) of which were classified as satisfactory. The acetabular index was the most reliable predictor of residual acetabular dysplasia.

We studied 52 patients, each with a lumbosacral transitional vertebra. Using MRI we found that the lumbar discs immediately above the transitional vertebra were significantly more degenerative and those between the transitional vertebrae and the sacrum were significantly less degenerative compared with discs at other levels. We also performed an anatomical study using 70 cadavers. We found that the iliolumbar ligament at the level immediately above the transitional vertebra was thinner and weaker than it was in cadavers without a lumbosacral transitional vertebra.

Instability of the vertebral segment above the transitional vertebra because of a weak iliolumbar ligament could lead to subsequent disc degeneration which may occur earlier than at other disc levels. Some stability between the transitional vertebra and the sacrum could be preserved by the formation of either an articulation or by bony union between the vertebra and the sacrum through its transverse process. This may protect the disc from further degeneration in the long term.

We performed a paired study of mature patients with Perthes’ disease to compare the radiological results after treatment between conservatively- and surgically-treated groups. One patient was selected from each group to create the pairs for this study. Each pair was strictly matched for gender, body mass index, age at onset, stage at the first visit, necrotic area and radiological at-risk signs and each was assessed by comparing the values of six radiological measurements.

Eighteen pairs (36 hips) fitted the criteria. The radiological measurements which showed a statistically better result in the surgical groups were Mose’s method, the acetabular-head index and leg-length discrepancy. There were no statistical differences in the slope of the acetabular roof and the articulotrochanteric distance. Four hips in the conservative group were in Stulberg class II, five in class III and nine in class IV. In the surgical group, 13 were in Stulberg class II, four in class III and one in class IV.

We conclude that surgical treatment improved the sphericity of the femoral head and provided greater acetabular cover, but did not reform the acetabular roof.

It was noteworthy that a greater leg-length discrepancy and a smaller articulotrochanteric distance were not seen in the surgical group. Our study suggests that surgical treatment is preferable in patients with severe Perthes’ disease.

We undertook a radiographic analysis with pre-operative computed tomographic myelography in 78 patients with idiopathic scoliosis in order to analyse rotation of the spinal cord and to investigate its clinical significance. The angle of rotation of the cord had a statistically significant relationship to both that of the apical vertebra and the size of the primary curve. The relationship between the rotation of the cord and that of the apical vertebra was divided into three types. The cord rotated in the same direction as the apical vertebra in 55 patients and rotated in the opposite direction in the remaining 23 patients. In the first group, the angle of rotation of the cord was more than that of the vertebra in six patients, but less than it in 49 patients. These results suggest that the neuraxis in idiopathic scoliosis may be under tension in the axial dimension.

Dorsal root ganglion neurones with dichotomising axons are present in several species and are considered to play a role in referred pain. Clinically, patients with lesions in the lower lumbar discs occasionally complain of pain in the groin. We investigated the existence of dichotomising afferent neurones projecting axons both to the lumbar disc and to the groin skin, using the double fluorescent-labelling technique in rats.

We observed neurones labelled with a tracer applied at the ventral portion of the L5-L6 disc and another tracer placed on the groin skin in L1 and L2 dorsal root ganglia. Our results showed that the double-labelled neurones had peripheral axons which dichotomised into both the L5-L6 disc and the groin skin, indicating the convergence of afferent sensory information from the disc and groin skin. Our findings provide a possible neuroanatomical mechanism for referred groin pain in patients with disc lesions.

Degenerative changes of the knee often cause loss of extension. This may affect aspects of posture such as lumbar lordosis. A total of 366 patients underwent radiological examination of the lumbar spine in a standing position. The knee and body angles were measured by physical examination using a goniometer. Limitation of extension of the knee was significantly greater in patients whose lumbar lordosis was 30° or less. Lumbar lordosis was significantly reduced in patients whose limitation of extension of the knee was more than 5°. It decreased over the age of 70 years, and the limitation of extension of the knee increased over the age of 60 years. Our study indicates that symptoms from the lumbar spine may be caused by degenerative changes in the knee. This may be called the ‘knee-spine syndrome’.

In patients who underwent autogenous iliac bone grafting we studied prospectively injury to the lateral femoral cutaneous nerve (LFCN) in relation to the size (length, depth, width) of the graft. We also examined the neurological deficit, by questioning them about numbness and/or pain in the lateral thigh.

The risk of injury was significantly higher in those in whom the depth of the graft was more than 30 mm. With regard to the length of the graft the incidence of nerve injury was 20% when the graft was 45 mm long or more, 16% when it was between 30 mm and 45 mm long, and 8% when it was less than 30 mm long. We should inform patients of the possibility of such injury, and take size into consideration when harvesting grafts from the ilium.

We examined an39 patients (45 knees) who had undergone Elmslie-Trillat procedure for recurrent or habitual dislocation of the patella with a follow-up of more than ten years. The mean age at the time of surgery was 18.4 years; the mean follow-up was 161 months (120 to 238). Using Fulkerson’s functional knee score, 41 knees (91%) had an excellent or good result at a mean follow-up of 45 months, and this was maintained in 29 (64%) at the final review.

The main cause of deterioration in the clinical results was the onset or worsening of patellofemoral joint pain, not patellar instability.

We have observed the natural history without treatment of 46 limbs in 29 patients with infantile tibia vara and a metaphyseal-diaphyseal angle (MDA) of more than 11°. The femorotibial angle (FTA) and MDA were measured, and Langenskiöld’s classification of radiological changes in the proximal medial metaphysis of the tibia was used. In 22 limbs which were not in Langenskiöld stages II to III the condition resolved spontaneously without treatment. Of the remaining 24 which were in stages II to III, in 18 it resolved spontaneously by the age of six years, but six showed little or no improvement at the latest follow-up. It was impossible to differentiate by measuring the FTA or MDA whether spontaneous resolution could be expected before the age of four years. There was no difference in the rate of resolution of the deformity between those patients who had been treated by a brace and those who had received no treatment. We advise no initial treatment but review at six-monthly intervals until the age of four years, even in patients with Langenskiöld stage-II to stage-III deformity. When a deformity persisted or progressed we carried out a corrective osteotomy after the age of four years.

Based on a study using a retrograde neurotracer, we have previously found that the dorsal portion of the L5/6 disc in the rat is multisegmentally innervated by dorsal root ganglia (DRG) from the level of T13 to L6, and that sensory nerve fibres from DRG of T13, L1 and L2 pass through the paravertebral sympathetic trunks. In this study in newborn rats, we injected crystals of 1,1′-dioctadecyl-3,3,3′,3′-tetramethylinedocarbocyanine perchlorate (DiI) into the DRG of T13, L1 and L2 and showed DiI-labelled sensory nerve fibres in the dorsal portion of the discs from the level of T13/L1 to L5/6. Our results show that the dorsal portion of the lumbar discs is innervated by the DRG from levels T13 to L2.

We studied in vivo the talonavicular alignment of club foot in infants using MRI. We examined 26 patients (36 feet) with congenital club foot. The mean age at examination was 9.0 months (4 to 12). All analyses used MRI of the earliest cartilaginous development of the tarsal bones in the transverse plane, rather than the ossific nucleus. The difference in the mean talar neck angle (44.0 ± 8.1°) in club foot was statistically significant (p < 0.001) when compared with that of the normal foot (30.8 ± 5.5°). The difference between the mean angles in the group treated by operation (47.9 ± 6.7°) and those treated conservatively (40.1 ± 7.5°) was also statistically significant. The anatomical relationship between the head of the talus and the navicular was divided into two patterns, based on the position of the mid-point of the navicular related to the long axis of the head. In the operative group, 18 feet were classified as having a medial shift of the navicular and none had a lateral shift. In the conservative group, 12 showed a medial shift of the navicular and six a lateral shift. All nine unaffected normal feet in which satisfactory MRI measurements were made showed a lateral shift of the navicular.

Club feet had a larger talar neck angle and a more medially deviated navicular when compared with normal feet. This was more marked in the surgical group than in the conservative group.

We have examined the process of fusion of the intertransverse processes and bone graft in the rabbit by in situ hybridisation and evaluated the spatial and temporal expression of genes encoding pro-α1 (I) collagen (COL1A1), pro-α1 (II) collagen (COL2A1) and pro-α1 (X) collagen (COL10A1).

Beginning at two weeks after operation, osteogenesis and chondrogenesis occurred around the transverse process and the grafted bone at the central portion of the area of the fusion mass. Osteoblasts and osteocytes at the newly-formed woven bone expressed COL1A1. At the cartilage, most chondrocytes expressed COL2A1 and some hypertrophic chondrocytes COL10A1. In some regions, co-expression of COL1A1 and COL2A1 was observed. At four weeks, such expressions for COL1A1, COL2A1 and COL10A1 became prominent at the area of the fusion mass. From four to six weeks, bone remodelling progressed from the area of the transverse processes towards the central zone. Osteoblasts lining the trabeculae expressed a strong signal for COL1A1. At the central portion of the area of the fusion mass, endochondral ossification progressed and chondrocytes expressed COL2A1 and COL10A1.

Our findings show that the fusion process begins with the synthesis of collagens around the transverse processes and around the grafted bone independently. Various spatial and temporal osteogenic and chondrogenic responses, including intramembranous, endochondral and transchondroid bone formation, progress after bone grafting at the intertransverse processes. Bone formation through cartilage may play an important role in posterolateral spinal fusion.

Sterilisation by gamma irradiation in the presence of air causes free radicals generated in polyethylene (PE) to react with oxygen, which could lead to loss of physical properties and reduction in fatigue strength. Tissue retrieved from failed total hip replacements often has large quantities of particulate PE and most particles associated with peri-implant osteolysis are oxidised. Consequently, an understanding of the cellular responses of oxidised PE particles may lead to clarification of the pathogenesis of osteolysis and aseptic loosening.

We have used the agarose system to demonstrate the differential effects of oxidised and non-oxidised PE particles on the release of proinflammatory products such as interleukin-1β (IL-1β), IL-6, and tumour necrosis factor-α (TNF-α) from monocytes/ macrophages (M/M). Oxidised PE particles were shown to stimulate human M/M to phagocytose and to release cytokines. Oxidation may alter the surface chemistry of the particles and enhance the response to specific membrane receptors on macrophages, such as scavenger-type receptors.

We performed a retrospective review of 27 scoliotic patients with syringomyelia using MRI. Their mean age at the first MRI examination was 10.9 years, and at the final review 15.8 years. The mean ratio of the diameter of the syrinx to the cord on the midsagittal MRI (S/C ratio) decreased from 0.49 to 0.24; 14 patients showed a decrease of 50% or more (reduction group). In this reduction group, the cerebellar tonsillar herniation decreased from a mean of 11.3 mm to 6.0 mm, and some improvement in dissociated sensory disturbance was seen in nine of 13 patients. The scoliosis improved by 5° or more in six patients in the reduction group. Our results indicate that spontaneous shrinkage of syringomyelia in children is not unusual and is associated with improvement in the tonsillar herniation, the scoliosis and the neurological deficit.

We describe a 47-year-old woman with sciatic neuropathy caused by compression of the sacral plexus by posterior shift of the uterus.

We studied 23 patients with spondylolysis of the fifth lumbar vertebra (L5) and 20 with spondylolytic spondylolisthesis at this level. All were more than 40 years of age. The transverse processes at L5 were significantly wider in the former group than in the latter. We also dissected 56 cadavers to study the morphological relationship between the transverse process of L5 and the iliolumbar ligament, and found that the wider transverse process is associated with increased width of the posterior band of the iliolumbar ligament.

If a patient with pars defects has wide transverse processes at L5, the lumbosacral junction may be stabilised by wide posterior bands of the iliolumbar ligament and the fifth lumbar vertebra by the ligament, preventing anterior displacement.

The multifunctional adhesion molecule CD44 is a major cell-surface receptor for hyaluronic acid (HUA). Recent data suggest that it may also bind the ubiquitous bone-matrix protein, osteopontin (OPN). Because OPN has been shown to be a potentially important protein in bone remodelling, we investigated the hypothesis that OPN interactions with the CD44 receptor on bone cells participate in the regulation of the healing of fractures. We examined the spatial and temporal patterns of expression of OPN and CD44 in healing fractures of rat femora by in situ hybridisation and immunohistochemistry. We also localised HUA in the fracture callus using biotinylated HUA-binding protein.

OPN was expressed in remodelling areas of the hard callus and was found in osteocytes, osteoclasts and osteoprogenitor cells, but not in cuboidal osteoblasts which were otherwise shown to express osteocalcin. The OPN signal in osteocytes was not uniformly distributed, but was restricted to specific regions near sites where OPN mRNA-positive osteoclasts were attached to bone surfaces. In the remodelling callus, intense immunostaining for CD44 was detected in osteocyte lacunae, along canaliculi, and on the basolateral plasma membrane of osteoclasts, but not in the cuboidal osteoblasts. HUA staining was detected in fibrous tissues but little was observed in areas of hard callus where bone remodelling was progressing.

Our findings suggest that OPN, rather than HUA, is the major ligand for CD44 on bone cells in the remodelling phase of healing of fractures. They also raise the possibility that such interactions may be involved in the communication of osteocytes with each other and with osteoclasts on bone surfaces. The interactions between CD44 and OPN may have important clinical implications in the repair of skeletal tissues.

We have studied fracture-dislocation of the fifth lumbar vertebra in seven patients and reviewed 50 previously reported cases. Based on this information, we have classified the injury into five types: type 1, unilateral lumbosacral facet-dislocation with or without facet fracture; type 2, bilateral lumbosacral facet-dislocation with or without facet fracture; type 3, unilateral lumbosacral facet-dislocation and contralateral lumbosacral facet fracture; type 4, dislocation of the body of L5 with bilateral fracture of the pars interarticularis; and type 5, dislocation of the body of L5 with fracture of the body and/or pedicle, with or without injury of the lamina and/or facet.

Conservative treatment of fracture-dislocation of L5 is generally not effective because the lesion is fundamentally unstable. Planning of the operation should be made on the basis of the various types of injury.

It has been thought that lumbar intervertebral discs were innervated segmentally. We have previously shown that the L5-L6 intervertebral disc in the rat is innervated bilaterally from the L1 and L2 dorsal root ganglia through the paravertebral sympathetic trunks, but the pathways between the disc and the paravertebral sympathetic trunks were unknown.

We have now studied the spines of 17 rats to elucidate the exact pathways. We examined serial sections of the lumbar spine using immunohistochemistry for calcitonin gene-related peptide, a sensory nerve marker. We showed that these nerve fibres from the intervertebral disc ran through the sinuvertebral nerve into the rami communicantes, not into the corresponding segmental spinal nerve.

In the rat, sensory information from the lumbar intervertebral discs is conducted through rami communicantes. If this innervation pattern applies to man, simple decompression of the corresponding nerve root will not relieve discogenic pain. Anterior interbody fusion, with the denervation of rami communicantes, may be effective for such low back pain.

We investigated 21 pairs of twins for zygosity and idiopathic scoliosis. DNA fingerprinting confirmed that 13 pairs were monozygotic and eight were dizygotic. There was concordance for idiopathic scoliosis in 92.3% of monozygotic and 62.5% of dizygotic twins.

Of the 12 pairs of monozygotic twins concordant for idiopathic scoliosis, six showed discordant curve patterns but eight had differences in Cobb angle of less than 10°. Seven of the ten pairs of monozygotic twins had similar back shapes.

Our findings suggest that there is a genetic factor in the aetiology of idiopathic scoliosis; they also indicate that there is a genetic factor in both the severity of the curve and the general shape of the back.

We performed a prospective study using MRI to evaluate early necrosis of the femoral head in 48 patients receiving high-dose corticosteroids for the treatment of various autoimmune-related disorders. The mean interval from the initiation of corticosteroid therapy to the first MRI examination was 2 months (0.5 to 6). MRI was repeated, and the mean period of follow-up was 31 months (24 to 69).

Abnormalities were found on MRI in 31 hips (32%). The initial changes showed well-demarcated, band-like zones which were seen at a mean of 3.6 months after initiation of treatment with steroids. In 14 of these hips (45%) there was a spontaneous reduction in the size of the lesions about one year after treatment had started, but there was no further change in size with a longer follow-up.

We reviewed 33 patients (35 hips) after open reduction of congenital dislocation of the hip using Ludloff’s medial approach. The mean age at the time of operation was 14 months (5 to 29) and at the time of final follow-up 20.1 years (15 to 24) giving a mean duration of follow-up of 19.4 years (14 to 23). We evaluated the radiological results by the Severin classification and the extent of avascular necrosis using the criteria of Kalamchi.

At the latest review 16 hips (45.7%) were in Severin groups I or II (acceptable results) and 19 (54.3%) in Severin groups III, IV or V (unacceptable results). There was evidence of avascular necrosis in 15 hips (42.9%). The radiological results began to become worse at about the age of ten years when type-II avascular changes presented.

We conclude that Ludloff’s medial approach for open reduction is unsatisfactory for the treatment of congenital dislocation since about half of our patients required additional operations.

The afferent pathways of discogenic low-back pain have not been fully investigated. We hypothesised that this pain was transmitted mainly by sympathetic afferent fibres in the L2 nerve root, and in 33 patients we used selective local anaesthesia of this nerve.

Low-back pain disappeared or significantly decreased in all patients after the injection. Needle insertion provoked pain which radiated to the low back in 23 patients and the area of skin hypoalgesia produced included the area of pre-existing pain in all but one. None of the nine patients with related sciatica had relief of that component of their symptoms.

Our findings show that the main afferent pathways of pain from the lower intervertebral discs are through the L2 spinal nerve root, presumably via sympathetic afferents from the sinuvertebral nerves. Discogenic low-back pain should be regarded as a visceral pain in respect of its neural pathways. Infiltration of the L2 nerve is a useful diagnostic test and also has some therapeutic value.

We carried out MRI studies of 74 patients with end-plate and vertebral bone-marrow changes associated with degenerative lumbar disc disease. Abnormalities were classified into type A, with decreased signal intensities, and type B, with increased signal intensities on T1-weighted spin-echo images. Twenty-seven (73%) of the 37 patients with type-A changes had low back pain, in contrast to only four (11%) of the 37 patients with type-B changes. Lateral flexion-extension radiographs showed hypermobility in 26 patients (70%) with type-A changes, and in only six (16%) with type-B changes. Type-A changes correlated with segmental hypermobility and low back pain, while type-B changes were more common in patients with stable degenerative disc disease.

We used hydroxyapatite blocks to keep open Salter innominate osteotomies in 19 children (21 hips), avoiding the need for an autograft taken from the anterior superior iliac spine. Our results were satisfactory, with less deformation of the iliac bone, and a reduction of both blood loss and operating time.

We studied the use of gadolinium diethylenetriaminepentaacetic acid-enhanced MRI in the detection of pathological changes in the nerve roots of 25 patients with unilateral sciatica due to lumbar disc herniation. Enhancement was observed in the affected nerve roots within the root sleeve at the caudal edge of the herniation and was classified into three categories: grade 0, none; grade 1, enhancement restricted to a focal region within the sleeve; and grade 2, diffuse and homogeneous. The grade of enhancement correlated well with the severity of the sciatica, and was considered to be due to a disruption of the blood-nerve barrier, leading to oedema.

The exact measurement of femoral head cover is essential for an assessment of reduction of congenital dislocation of the hip. We have compared standard anteroposterior radiographs with computerised tomograms and thereby classified the shape of the acetabular roof into four types. We found that the CE angle of Wiberg is a more reliable measure of head cover when the lateral point of bony condensation of the roof is chosen as the reference point rather than the edge of the bone, where these two points do not overlap. We conclude that head cover can be more accurately determined in younger children with dysplastic hips by our 'refined' CE angle, than by the original method of Wiberg.

Hip arthrography was performed in 19 patients in the initial stage of Perthes' disease. Sphericity and subluxation were measured and it was found that subluxation was independent of the femoral head deformity. We therefore tried to identify the cause of early subluxation: in seven patients a swollen ligamentum teres was thought to be responsible, and was associated with medial pooling of the contrast medium. A swollen ligamentum teres was seen in another seven cases; the other five arthrograms were normal. These findings were further clarified by enhanced CT scans, which confirmed that ligament swelling may be an important cause of early subluxation.