Introduction: Degeneration of the intervertebral disc is characterised by loss of normal cell activity, disc matrix and loss of disc height. There is currently much interest in using cells to effect a biological repair in connective tissues, eg autologous chondrocyte implantation for cartilage repair. Intervertebral discs have a low cell density, with those cells present often being unhealthy and necrotic. Hence, identification of an alternative source of cells for autologous disc repair could be beneficial. Thus we have investigated other types of connective tissue cells to determine if they may be encouraged to undertake a disc cell phenotype.

Materials and Methods: Cells were enzymatically/mechanically extracted from bovine coccygeal discs (annulus and nucleus), skin, bone marrow, periosteum and tendon and the efficiency and proliferation rates assessed. Dermal fibroblasts and bone marrow cells were also grown in a 3D alginate system and compared to disc nucleus pulposus cells for phenotypic expression from 0–28 days. Cell phenotype was assessed via morphology, immunohistochemistry, Western blotting and RT-PCR for mRNA expression.

Results: All cell types could be extracted and proliferated in monolayer, with a flattened and fibroblast-like morphology. Proliferation was slowest for bone marrow cells (4 times slower than nucleus pulposus cells). Cells cultured in alginate became rounded with chondrocyte-like morphology. They remained viable for 4 weeks, but with little replication. Expression or production of proteoglycans, both aggrecan and the small proteoglycans (especially fibromodulin) and collagen types I, II and X was demonstrated for all cell types. There was, however, a difference in the timescale of production between some cell types.

Conclusions: Plasticity of different cell types is well known and the connective tissue cells investigated in this study are capable of responding to the environment in which they are cultured. They can synthesise matrix molecules typically produced by disc cells in vivo and hence warrant further investigation as a potential source of cells for biological repair of the intervertebral disc.

Background: Increased nerve growth into degenerated intervertebral discs is associated with discogenic low back pain [1]. Many of these growing nerves are in neo-vascularised areas of the tissue [1,2] and endothelial cells that penetrate the disc express neurotrophic factors [3]. Thus, disc neovascularisation and disc innervation may be closely linked. Whilst disc aggrecan has been found to inhibit sensory nerve growth in vitro [4], the effects of disc aggrecan on endothelial cells are unknown.

Methods/Results: Adapting in vitro assays used previously [4], with HMEC-1 and EAhy-926 cell lines as models of endothelial cell growth, we found that disc aggrecan inhibited endothelial cell migration in a dose-dependent manner. Endothelial cells traversed over collagen substrates until they encountered disc aggrecan substrates (1mg/ml human aggrecan), where they either stopped migrating or, more commonly, changed their direction of movement and aligned to the collagen:aggrecan border (Figure 1). After reaching the aggrecan border, some endothelial cells also migrated away from the disc aggrecan. At lower concentrations of disc aggrecan (0.01mg/ml), no such inhibition of endothelial cell growth was seen.

Conclusions: Loss of aggrecan, increased innervation and neovascularisation are all marked features of disc degeneration [1,2,5]. This study provides evidence that disc aggrecan inhibits endothelial migration and therefore supports a hypothesis that a loss of aggrecan from degenerated discs predisposes the tissue to vascular invasion.

Autologous chondrocyte implantation (ACI) has been used most commonly as a treatment for cartilage defects in the knee and there are few studies of its use in other joints. We describe ten patients with an osteochondral lesion of the talus who underwent ACI using cartilage taken from the knee and were prospectively reviewed with a mean follow-up of 23 months. In nine patients the satisfaction score was ‘pleased’ or ‘extremely pleased’, which was sustained at four years. The mean Mazur ankle score increased by 23 points at a mean follow-up of 23 months. The Lysholm knee score returned to the pre-operative level at one year in three patients, with the remaining seven showing a reduction of 15% at 12 months, suggesting donor-site morbidity. Nine patients underwent arthroscopic examination at one year and all were shown to have filled defects and stable cartilage. Biopsies taken from graft sites showed mostly fibrocartilage with some hyaline cartilage. The short-term results of ACI for osteochondral lesions of the talus are good despite some morbidity at the donor site.

Aims: To examine vibration levels produced by orthopaedic air tools and the prevalence of upper limb symptoms in orthopaedic surgeons. Methods: A preliminary measurement of vibration levels produced by six air powered orthopaedic saws was followed by a national survey of orthopaedic surgeons and controls. A health surveillance questionnaire of symptoms associated with Hand Arm Vibration Syndrome (HAVS) was sent to 1200 orthopaedic surgeons (test group) and 1200 gynaecological surgeons (controls). Results: Measured accelerations of the saws were 3.42 to 10.7 m/sec² using BSI standards and 90.5 to 182 m/sec⁻² using NIOSH standards. These vibration levels are compatible with those reported to cause significant upper limb symptoms

Survey responses were received from 741(61.7%) of the test group and 748 (62.3%) of the control group. A statistically significant increase in the prevalence of the neurological symptoms was seen among orthopaedic surgeons (p< 0.001). A significant increase in musculoskeletal problems (p< 0.008) and muscle pain (p< 0.004) was also found. No significant difference was seen in the prevalence of vascular symptoms. The neurological symptoms were not related to other potential medical causes. Conclusions: Orthopaedic surgeons report an excess of upper limb symptoms and these may be linked to vibration exposure at work.

Objective: To determine if (a) inflammatory mediators are present in herniated intervertebral discs and (b) if their presence correlates with inflammation of nerve roots or symptoms.

Design: Inflammation was assessed with gadolinium enhancement of MRI. Neurological compromise was measured. Disc tissue was examined for inflammatory mediators IL-1α and β, IL-6, MCP-1, TSG-6, iNOS, TNFα and thromboxane.

Patients: Sixty-five discs were removed from 64 patients undergoing surgery for disc prolapse.

Outcome measures: We developed (i) an MRI score to assess inflammation radiologically prior to surgery (n=28, mean 4.9±6.8 days), (ii) a Surgical Score to assess inflammation of the nerve roots at surgery (n=44), (iii) a Clinical Score to determine pain, disability and neurological compromise (n=17) and (iv) a Mediator Score to reflect the number and amount of inflammatory mediators present (n=20).

Results: Thirty percent of the prolapses in this study were extrusions, 19% sequestrations and 51% protrusions. Sixteen of the 28 patients with gadolinium had nerve root enhancement (86% of the extrusions, 57% of sequestrations, and 43% of protrusions), whilst 19 had enhancement of or around the disc herniation itself (71% of the extrusions, 86% of sequestrations and 57% of protrusions). The Mediator Scores were highest for the sequestrations (as was the Surgical Score) and lowest for the protrusions, but extruded discs had most IL-1α and β, IL-6, TNFα and thromboxane. Extruded discs had the highest Clinical Score and sequestrated the lowest.

Conclusions: Mediators produced in prolapsed disc appear to play an important role in inflammation of adjacent tissue and nerve roots. The type of mediator present and proximity of the prolapse to the nerve root may be the important factors in determining which pro-lapses are the most painful.

Objective: The shape of articular chondrocytes regulates their function, changes in response to mechanical load and is altered in osteoarthritis. We aimed to identify the shape of intervertebral disc cells in pathological and normal tissue.

Design: Immunohistology of human intervertebral discs using cytoskeletal markers to examine disc cell shape.

Subjects: Intervertebral discs from patients with degeneration (n=3), scoliosis (n=3), spondylolisthesis (n=3) and from non-pathological cadaveric spines (n=3).

Outcome measures: (i). Cell shape and (ii). Organisation/ content of cytoskeleton.

Results: In degenerate and normal discs, cells of the anulus fibrosus were generally elongated and bipolar, whilst those of the nucleus pulposus were rounded/oval. However, in localised areas, cells were observed with multiple cytoplasmic processes that extended into the discal matrix. In central regions of scoliotic and, most markedly, spondylolisthetic discs, such cells were more frequent. Their processes were vimentin positive (but F-actin negative) and reached up to 80μm in length. F-actin was clearly present in endothelial cells of blood vessels but absent in disc cells. In contrast, vimentin was expressed by disc cells within the discs’ inner regions, but not towards the outer anulus fibrosus.

Conclusions: The altered shape of disc cells in pathological tissue may reflect areas of abnormal loading. These changes are also likely to affect/reflect altered cell function and therefore have a role to play in the pathological process.

Background: The healthy, adult human disc is innervated but the nerves are restricted to the outer few millimetres of the annulus fibrosus. In degenerate discs with associated back pain, however, the nerves are more numerous and penetrate further in.

We have used a sheep model of intervertebral disc degeneration to monitor the presence and organisation of nerves in the disc as degeneration progresses. This model has been used to study morphological and bio-chemical changes of the disc as it degenerates, in addition to associated alterations in end-plate vascularity and vertebral bone remodelling. One aspect of this model which has not been studied to date is how the innervation of the disc may change with the onset of degeneration. This is the object of the present study.

Materials and Methods: Four-year old, skeletally mature Merino wether sheep (n=64) were divided randomly into lesion and control groups. A surgical incision was created in the anterolateral annulus in the L1–L2 and L3–L4 discs of the lesion group. The control group received the same retroperitoneal surgical approach but the annulus was not incised. Intact lumbar discs encompassed by adjacent vertebral bodies were removed at 3,6,12 and 26 months post operation. Specimens were fixed, decalcified and paraffin embedded before sectioning (7μ thick, vertical sagittal sections) and stained immunohistochemically with the neuronal marker, PGP9.5, together with standard histological stains.

Results: The incised region of the outer annulus underwent collagenous re-organisation, consistent with an active repair process as early as three months post-operatively. However, the inner annular lesion had a poor repair response and propagated with time, sometimes through to the nucleus. In contrast, remodelling of the outer annular lamellae occurred across the cut region. For example, in one sample at two years post injury there were up to six lamellae “bridging the gap”. Nerves were present in all samples but in the sham animals they were very few and confined to the very outer annulus or longitudinal ligament. In the operated animals, nerves were more extensive, occurring in the matrix adjacent to the fissure where there was often blood vessel ingrowth. The maximum number of nerves was seen at 12 months post-operatively, before diminishing in number at 24 months post-op. This paralleled the presence and extent of blood vessel penetration in this experimental model.

Conclusions: We have used an animal model to follow longitudinally the penetration of nerves into the ovine intervertebral disc in association with disc degeneration. Whilst we obviously cannot assess back pain in these animals, and not all nerves are nociceptive, nerves nevertheless are a pre-requisite for the perception of pain. Hence the greater numbers, size and penetration of nerves into degenerate discs demonstrated here has important implications not only for the aetiopathogenesis of degenerative disc disease but also for the treatment of its associated symptoms. Further characterisation of this innervation, i.e. whether autonomic or sensory, may provide an indication as to its nociceptive potential.

Introduction: Intervertebral disc degeneration occurs with ageing and is often associated with back pain. During such degeneration, gross morphological differences between the central nucleus pulposus (NP) and outer annulus fibrosus (AF) are lost and the disc loses hydration and height due to decreased proteoglycan content. The cartilage endplate may also become calcified and this blocks the passage of nutrients into the disc, causing cell death and further degeneration. A potential therapy of degeneration is “re-inflation” of the disc with the use of hydrogels seeded with autologous disc cells. In this study, we have assessed the ability of a variety of hydrogels to support intervertebral disc cell growth.

Method: Intervertebral disc cells were isolated enzymatically from bovine tails and cultured as a monolayer in 10% foetal calf serum in DMEM containing antibiotics and ascorbic acid. This stimulates the cells to proliferate and thereby produces increased cell numbers. The cells were then seeded onto various hydrogels including hyaluronic acid (HA), 2-hydroxyethyl methacrylate (HEMA), N’N’ dimethyl methacrylate (NNDMA) and polyacryloyl morpholine (AMO) before harvesting at set time points of 1, 3, 6 and 9 days for hyaluronic acid and 1, 7, 14, 21, and 28 days for the other hydrogels. Cell number, morphology, viability and adherence to or migration into the hydrogels were assessed. Cell proliferation was also determined by immunostaining for the Ki67 antigen.

Results: Disc cells became incorporated in the HA gel, adopted a spherical morphology and remained viable for up to nine days. However, after a few days, a large proportion of the cells began to migrate through the gel to form a monolayer on the bottom of the tissue culture well. These monolayered cells became fibroblastic and proliferated. NP cells appeared to proliferate to a greater extent than AF cells both in monolayer and in suspension. Ki67 antigen immunostaining confirmed cell proliferation. On the non-porous HEMA, NNDMA and AMO, both cell types adhered and adopted a fibroblast-like morphology. Cell adhesion was greatest to the HEMA. NNDMA and AMO had lower levels of cell adherence. Both cell types became incorporated into the porous materials and adopted a rounded morphology. Cell incorporation appeared to be greatest into porous HEMA.

Conclusion: These initial studies show that intervertebral disc cells will adhere to or migrate into a variety of hydrogels and remain viable. The morphology and proliferative capacity of cells derived from both the AF and NP were responsive to the structure of the hydrogel with which they were cultured. Thus, cells were able to become fibroblastic or chondrocytic. Further analyses will reveal whether matrix synthesis by disc cells is similarly responsive to the hydrogel format. The results of these experiments suggest that the hydrogels tested have potential as support matrices in intervertebral disc repair to provide relief from discogenic low-back pain.

Mature human intervertebral disc cells have generally been described as being either fibroblast-like or chondrocyte-like; i.e. appearing either elongated and bipolar or rounded/oval. Fibroblast-like cells are observed within the outer regions of the anulus fibrosus whilst chondrocyte-like cells are found in the more central regions of the disc. However, a few reports have noted that in some circumstances disc cells appear to extend more elaborate cytoplasmic processes into their surrounding extracellular matrix. In this study, we have examined healthy and pathological human intervertebral discs for the presence of the cytoskeletal elements, F-actin and vimentin.

Tissues examined included discs of no known pathology, discs with spondylolithesis, scoliosis specimens taken from the convex and concave sides, and degenerated discs. F-actin was not readily observed within discs cells but was a marked feature of vascular tissue within the disc and occasionally seen in infiltrating cells. Vimentin was more readily seen within cells of the inner anulus fibrosus and nucleus pulposus. In general, disc cell morphology was fibrocyte or chondrocyte-like; however, in spondylolisthetic discs, cells with numerous cytoplasmic projections were frequently observed.

The differential morphologies and cytoskeletal composition observed in disc cells may be indicative of variations in mechanical strains and/or pathologies, or indeed of cell function.

The aim of this study was to identify potential inflammatory mediators in herniated and non-herniated intervertebral disc. It has been suggested that inflammation of the nerve root is a pre-requisite for disc herniations to be symptomatic. What leads to this inflammation is a matter of conjecture; one possible cause may be inflammatory mediators released from the herniated disc tissue itself. In this study we have examined discs from individuals with and without disc herniations to determine if there is a different degree of occurrence.

Twenty two discs from 21 patients with disc herniation were examined together with four discs from patients with other disc disorders and five age-matched discs from individuals obtained at autopsy. Samples were studied for the presence of blood vessels and inflammatory cytokines: IL-1α and β, IL-6, INOS, MCP1, TNFα, TSG-6 and thromboxane.

Of the herniated discs 10 were protrusions, six extrusions and six sequestrations. There was less of all the cytokines in the non-herniated discs than found in the herniated, with very little immunostaining for iNOS or IL-1α in any samples. Staining was seen in all herniated samples for IL-1β, but in fewer for IL-six and MCP1 (86%), thromboxane (68%), TNFα (64%) and TSG-6 (59%). The presence of cytokines was strongly associated with the presence of blood vessels. Protruded discs had less TNFα and thromboxane than sequestrated or extruded discs.

Cytokines appear to play an active role in the aetiopathogenesis of disc herniations. Some may be involved in the stimulation of degradative enzymes and hence resorption of, for example, sequestrations, whereas others may be responsible for an inflammatory response in the surrounding tissues such as nerve roots.

Although an increased and deeper innervation of painful and degenerate intervertebral discs (IVDs) has been reported, the mechanisms that regulate nerve growth into the IVD are largely unknown. In other tissues, proteoglycans have been found to act as nerve guidance molecules that, generally speaking, inhibit nerve growth. As disc degeneration is characterised by a loss of proteoglycans, we assessed the effects of IVD proteoglycans on nerve growth and guidance.

Using in vitro assays of nerve growth, we found that human disc proteoglycans inhibited nerve attachment, neurite extension and induced sensory growth cone turning in a dose-dependent manner. Digestions with chondroitinase ABC or keratinase abrogated these inhibitory effects. Proteoglycans of the anulus fibrosus were more inhibitory than those from the nucleus pulposus.

Disc proteoglycans inhibit nerve growth and this inhibitory activity may dependent on proteoglycan glycosylation and/or sulfation. A loss of proteoglycans from degenerative discs may therefore predispose the discs to nerve invasion.

Tissue engineering is an increasingly popular method of addressing pathological disorders of cartilage. Recent studies have demonstrated its clinical efficacy, but there is little information on the structural organisation and biochemical composition of the repair tissue and its relation to the adjacent normal tissue. We therefore analysed by polarised light microscopy and immunohistochemistry biopsies of repair tissue which had been taken 12 months after implantation of autologous chondrocytes in two patients with defects of articular cartilage.

Our findings showed zonal heterogeneity throughout the repair tissue. The deeper zone resembled hyaline-like articular cartilage whereas the upper zone was more fibrocartilaginous. The results indicate that within 12 months autologous chondrocyte implantation successfully produces replacement cartilage tissue, a major part of which resembles normal hyaline cartilage.

The articular surface of the humeral head is usually described as facing posteromedially, making an angle of between 16 degrees and 35 degrees with the transepicondylar plane. At hemiarthroplasty the articular surface also appears to be offset posteriorly with respect to the humeral shaft. Coracoid impingement may occur if this offset is not accommodated. An analysis was made of 29 cadaveric humeri using an industrial co-ordinate measuring machine. The position of the centre of the head was defined with respect to the humeral shaft and transepicondylar plane. The humeral articular surface was found to be retroverted by 21.4 degrees and its centre offset posteriorly by 4.7 mm. Previous interpretation of retroversion did not take into account the posterior displacement, and this may be of importance in improving future prosthetic design.

Articular cartilage from the femoral heads of 27 patients having an arthroplasty for subcapital fracture was studied, and its mechanical and chemical properties compared to those of a group of 33 age-matched macroscopically normal autopsy specimens. Water and proteoglycan contents were measured, as were swelling ability, compressive and tensile strength of the cartilage, and the density of the underlying bone. Cartilage from the fracture specimens had a significantly reduced proteoglycan content, as measured by fixed charge density, and increased swelling ability. These results indicate that this group differs from the "normal" population and care should be taken before they are accepted as control material for studies on osteoarthritic cartilage. Another finding was that bone density was much the same in the fracture and the normal group. This casts some doubt upon the concept that patients who sustain subcapital fractures are more osteoporotic than the average for the same age range.

We studied the mechanical and biochemical properties of articular cartilage from 22 osteoarthritic femoral heads obtained at operation and 97 femoral heads obtained at autopsy. Cartilage from the zenith and from the antero-inferior aspect of each head was tested both in tension and in compression. Water content, swelling ability and proteoglycan content were measured, the cartilage was examined histologically and the density of the underlying bone was assessed. Fifty-five of the autopsy specimens were defined as macroscopically normal because they exhibited no progressive fibrillation patterns on staining with Indian ink; but significant changes in water content, bone density and tensile strength related to age were seen in this group. In 20 pairs of femoral heads which were both macroscopically normal, we found, surprisingly, that cartilage from the left and right sides of the same patient was sometimes very different. Compared with the normal autopsy specimens the osteoarthritic specimens had a significantly increased swelling ability, a lower proteoglycan content and impaired mechanical properties, being both weaker in tension and softer in compression. Abnormal autopsy specimens had values intermediate between those of osteoarthritic and normal groups. Results from this abnormal group suggest that there is no primary loss of proteoglycan in early osteoarthritis.

Specific antisera to collagen Types I, II and III and proteoglycan were used to investigate the distributions of these molecules in normal human intervertebral discs. Immunofluorescent staining indicated the presence of small amounts of Type III collagen located pericellularly in normal adult intervertebral discs. This finding had not been demonstrated previously by other methods. Similar specimens of intervertebral discs from 17 patients with scoliosis of varying aetiologies were examined, but no evidence was obtained for primary connective tissue defects. Secondary changes, especially marked vascularisation of the inner annulus, were apparent in a number of scoliotic discs, and some of these showed enhanced staining for collagen Type I and proteoglycan, and intercellular matrix staining for Type III collagen.