To date there is no medical treatment alternative to surgery for osteolysis after THA. In this proof-of-concept clinical trial we examined the effect of a human monoclonal antibody against osteoclasts versus placebo on osteolytic lesion activity in patients undergoing revision surgery.

Patients scheduled for revision for symptomatic osteolysis were randomised (1:1) to receive either denosumab 60mg or placebo subcutaneously eight weeks prior to operation. At surgery, biopsies from the osteolytic membrane-bone interface were taken for histomorphometric analysis of osteoclast number. Secondary outcome measures included systemic bone turnover markers.

22 subjects completed the study (10 denosumab). The denosumab group had 83% (−63 to −97), P=0.011 fewer osteoclasts at osteolytic lesion sites, 87% lower osteoclast surface (−65 to −95, P=0.009), and 72% lower eroded surface (−35 to −93, P=0.020) versus the placebo group. At surgery, serum CTX-I, TRAP5b and PINP were 80% (−65 to −95, p<0.001), 57% (−40 to −90, p<0.001), and 44% (−41 to −65, p<0.001) lower in the denosumab versus placebo groups, respectively. The rate of adverse events (denosumab 6, placebo 7) were similar between groups (P>0.05).

These data provide a biological basis for a definitive clinical trial using pain, function and prosthesis survival as the study endpoints. As osteolysis/ aseptic loosening is the leading cause of prosthesis failure world-wide, the establishment of a non-surgical solution would reduce patient suffering and dramatically reducing the cost to healthcare economies.

A recent French report suggested that cobalt metal ions released from total hip replacements (THRs) were associated with an increased risk of dilated cardiomyopathy and heart failure. If the association is causal the consequences would be significant given the millions of Orthopaedic procedures in which cobalt-chrome is used annually. We examined whether cobalt-chrome containing THRs were associated with an increased risk of all-cause mortality, heart failure, cancer, and neurodegenerative disorders.

Data from the National Joint Registry was linked to NHS English hospital inpatient episodes for 375,067 primary THRs with up to 14·5 years follow-up. Implants were grouped as either containing cobalt-chrome or not containing cobalt-chrome. The association between implant construct and the risk of all-cause mortality and incident heart failure, cancer, and neurodegenerative disorders was examined.

There were 132,119 individuals (35·2%) with an implant containing cobalt-chrome. There were 48,106 deaths, 27,406 heart outcomes, 35,823 cancers, and 22,097 neurodegenerative disorders. There was no evidence of an association that patients with cobalt-chrome implants had higher rates of any of the outcomes. For all-cause mortality there was a very small survival advantage for patients having a cobalt-chrome implant (restricted mean survival time 13·8=days, 95% CI=6·8-20·9).

Cobalt-chrome containing THRs did not have an increased risk of all-cause mortality, heart failure, cancer, and neurodegenerative disorders into the second decade post-implantation. Our findings will reassure clinicians and patients that primary THR is not associated with systemic implant effects.

Introduction

The development of an algorithm that provides accurate individualised estimates of revision risk could help patients make informed surgical treatment choices. This requires building a survival model based on fixed and modifiable risk factors that predict outcome at the individual level. Here we compare different survival models for predicting prosthesis survivorship after hip replacement for osteoarthritis using data from the National Joint Registry for England, Wales, Northern Ireland and the Isle of Man (NJR).

Introduction

Periprosthetic osteolysis resulting in aseptic loosening is a leading cause for total hip arthroplasty (THA) failure. Individuals vary in their susceptibility to osteolysis, and it is thought that heritable factors contribute to this variation. We conducted two genome-wide association studies to identify genetic risk loci associated with osteolysis and genetic risk loci associated with time to prosthesis failure due to osteolysis.

Background

Heterotopic ossification (HO) is lamellar bone formation in the soft tissues following trauma or joint replacement for osteoarthritis (OA). A genome wide association study of HO patients after total hip arthroplasty for OA has identified Kinesin Family Member 26B (KIF26B) as a gene associated with HO severity. KIF26B has previously been associated with HO in mice.

Hypothesis and aims: We hypothesised that Kif26b regulates the osteogenic trans-differentiation of myoblasts; a possible mechanism of HO. Using an in vitro model, we wished to establish whether Kif26b is involved in HO formation and to explore the molecular mechanism.

With the increasing demand for hip and knee replacement and move towards seven-day services, we examined whether this planned, elective surgery performed at the weekend is associated with a different 30-day mortality versus that performed between Monday and Friday.

The study dataset comprised 118,096 joint replacement episodes performed at the weekend and 1,233,882 episodes done on a weekday. We used a Kaplan-Meier framework to examine the 30-day cumulative mortality rate for all elective hip and knee replacements performed in England and Wales and recorded in the NJR between 1st April 2003 and 31st December 2014, with Cox proportional-hazards regression models to assess for time-dependent variation and adjust for identified risk factors for mortality.

For hip replacement the cumulative 30-day mortality was 0.15% (95%CI: 0.12–0.19) for patients operated on at the weekend versus 0.20% (0.19–0.21) for patients undergoing surgery during the normal working week. For knee replacement the cumulative 30-day mortality was 0.14% (0.11–0.17) for patients operated on at the weekend versus 0.18% (0.17–0.19) for patients undergoing surgery during the normal working week. The lower mortality associated with weekend operating was most apparent in the later years of the audit (2009 to 2014) and remained after adjustment for any differences in patient age, gender, American Society of Anaesthesiologist grade, surgeon seniority, surgical and anaesthetic practices, and thrombo-prophylaxis choice in weekend versus weekday operated patients.

Routine hip and knee replacements performed at the weekend in England and Wales and are not associated with an increased risk of post-operative mortality.

With the increasing demand for hip and knee replacement, and the increasing pressure to move towards routine seven-day services within the National Health Service, the trend towards weekend operating is set to increase. We aimed to determine whether planned, elective total hip and total knee replacement performed at the weekend is associated with a different 30-day mortality versus those performed between Monday and Friday.

We used National Joint Registry of England, Wales, Northern Ireland and Isle of Man (NJR) linked to Office for National Statistics (ONS) data. The study dataset comprised 118,096 joint replacement episodes performed at the weekend and 1,233,882 episodes done on a weekday. The main outcome measure was 30-day all-causes mortality. We applied a survivorship analysis using a Kaplan-Meier framework to examine the 30-day cumulative mortality rate for all elective hip and knee replacements performed in England and Wales between 1st April 2003 and 31st December 2014, with Cox proportional-hazards regression models to assess for time-dependent variation and adjust for identified risk factors for mortality.

For hip replacement the cumulative 30-day mortality was 0.15% (95%CI: 0.12–0.19) for patients operated on at the weekend versus 0.20% (0.19–0.21) for patients undergoing surgery during the normal working week. For knee replacement the cumulative 30-day mortality was 0.14% (0.11–0.17) for patients operated on at the weekend versus 0.18% (0.17–0.19) for patients undergoing surgery during the normal working week. The lower mortality associated with weekend operating was most apparent in the later years of the audit (2009 to 2014) and remained after adjustment for any differences in patient age, gender, American Society of Anaesthesiologist grade, surgeon seniority, surgical and anaesthetic practices, and thrombo-prophylaxis choice in weekend versus weekday operated patients.

Hip and knee replacements are routinely performed on Saturdays, and to a lesser extent on Sundays, in England and Wales and are not associated with an increased risk of post-operative mortality.

We have previously observed an increase in total bone mineral density and reduced bone turnover (TRAP5b and osteocalcin) in patients with well-functioning metal-on-metal hip resurfacing (MOMHR). Here, we provide data to support the hypothesis that osteoclast differentiation and function is altered in this patient population, and that this effect is transferrable through their serum.

Patients with well-functioning MOMHR (cases, n=18) at a median follow-up of 8 years were individually matched for gender, age and time-since-surgery to a low-exposure group consisting of patients with THA (controls, n=18). The monocyte fraction of patient peripheral blood was isolated and differentiated into osteoclasts on dentine wafers using RANKL and M-CSF supplemented media (osteoclastogenic media, OM). Cultures were monitored for the onset of resorption, at which point the cells were treated with OM, autologous serum or serum from matched MOMHR/THA donors, all supplemented with RANKL and M-CSF. At the end of the culture, cells were TRAP-stained and quantified using CellD Software Package, Olympus.

When cells were differentiated in standard osteoclastogenic media, the resorbing ability of osteoclasts derived from MOMHR patients was reduced 22%(p<0.0079) compared to THA. The resorbing ability of osteoclasts generated from MOMHR patients and differentiated in autologous serum was reduced 33%(p<0.0001), whilst matched THA serum caused a smaller reduction of 14%(p<0.01). When cells derived from THA patients were differentiated in autologous serum, the resorbing ability of osteoclasts was similarly reduced by 35%(p<0.0001), whilst the matched MOMHR serum also caused a reduction of 21%(p<0.0001).

This data suggests that prior exposure to higher circulating Co and Cr in patients with MOMHR reduces osteoclastogenesis, and that the detrimental effect on the functionality of mature osteoclasts is transferable through the serum. This has implications for systemic bone health of patients with MOMHR or modular taper junctions.

Systemic concentrations of metal ions (cobalt and chromium) are persistently elevated in patients with metal-on-metal hip resurfacing (MOMHR) compared to conventional total hip arthroplasty (THA). Several studies by us and others have described the detrimental effects of metal exposure on survival and function of various cell types in-vitro, but the mechanisms for these effects remain unclear. Epigenetic modifications following chronic metal exposure is a possible mechanism that could mediate these effects. Here we test the methylation status in genomic DNA from MOMHR (“cases”) and THA (“controls”) patient-groups, and its correlation with circulating metal levels.

The cohort consisted of 34 patients with a well-functioning MOMHR at a median follow-up of 9.75 years. These were individually matched for gender, age and time-since-surgery to a non-exposure group consisting of patients with THA. Genomic DNA was isolated from blood samples and cell composition estimated using the ‘estimateCellCounts’ function in ‘minfi R-package’. Methylation was assessed using the Illumina 450k BeadChip array analysing 426,225 probes. Logit model was fitted at each probe with case/control status as independent variable and covariates of gender, age, time-since-surgery, smoking, non-arthroplasty metal exposure, and cell composition. DNA methylation age was assessed using an online calculator (https://dnamage.genetics.ucla.edu/) and comparisons made between cases and controls, and correlated with circulating metal levels.

Cell distributions did not differ between the cases and controls (Wilcoxon test p<0.17) with no probe having an association at 5% FDR. Circulating metal levels and LVEDD also had no association with any probe at 5% FDR. There was no preferential age acceleration between cases and controls (Wilcox p<0.7), and it had no correlation with plasma-chromium or blood-cobalt levels (p<0.9).

In summary, large methylation changes following MOMHR seem to be absent, compared to THA. Future research with larger samples will be needed to clarify the presence and extent of small methylation changes.

Background

An extended trochanteric osteotomy (ETO) is a widely used approach for revision hip arthroplasty. Following an ETO it is common practice to use a long stemmed femoral prosthesis at the second stage to bypass the osteotomy. We propose that at the second stage, if the osteotomy has united, it is appropriate to use a standard length prosthesis, which preserves bone stock for any future revisions.

Introduction

Although DDH is one of the most common skeletal dysplasias (incidence 1.5 cases per 1000 births), it remains slow and costly to recruit large-scale patient cohorts for powerful genetic association studies. In this work we have successfully used the NJR as a platform to generate a DDH biobank of 907 individuals, upon which we have conducted the first ever genome-wide association study (GWAS) for DDH.

Introduction

In rheumatoid (RA) patients undergoing total joint arthroplasty, evidence suggests methotrexate should be continued without increase in post-operative infection. Prednisolone increases post-operative infection risk, but cannot be stopped without risk of Addisonian events. Insufficient evidence exists to clarify perioperative biologic agent management. We currently stop biologics 2 weeks prior to operation and reintroduced following wound healing.

Introduction

Aseptic loosening, the clinical endpoint of osteolysis, remains the leading cause of total hip arthroplasty (THA) failure, and is caused by a host response to wear debris that varies between individuals. Although several candidate gene studies have identified loci associated with osteolysis susceptibility, there have been no systematic studies at genome-wide level. We aimed to identify risk loci associated with osteolysis by conducting a genome-wide association study.

Introduction

Subtle variations in hip morphology associate with risk of hip osteoarthritis (OA). However, validated accurate methods to quantitate hip morphology using plain radiography are lacking. We have developed a Matlab-based software-tool (SHIPs) that measures 19 OA-associated morphological-parameters of the hip using a PACS pelvic radiograph. In this study we evaluated the accuracy and repeatability of the method.

Background and objectives

Local bone-related adverse events occur more frequently following metal-on metal hip resurfacing (MOMHR) versus convention total hip arthroplasty (THA). High local tissue levels of cobalt and chromium may contribute to impaired bone health, however the systemic effects on bone of exposure to elevated metal levels after MOMHR are unknown.

Introduction

Lumbar spinal pain can be a consequence of discogenic pain. After failure of conservative management, lumbar fusions are popular and considered the gold standard. However, these techniques are associated with significant morbidity.

A percutaneous trans-sacral technique may address these for L4/L5 and/or L5/S1 alone spinal fusions. Our unit has adopted this technique and presents our initial experience of 51 patients and preliminary results.

One possible mechanism by which metal-on-metal hip resurfacing (MOMHR) may be associated with prosthesis loosening, periprosthetic fracture, and femoral neck narrowing is through an increase in bone resorption by osteoclast cells. Whilst it is known that metal ions such as cobalt (Co) and chromium (Cr) ions (that are elevated locally and systemically after MOMHR), may affect osteoblast and macrophage activity in-vitro, little is known about the effect of these ions on osteoclasts. We examined whether these ions have an adverse effect on human peripheral blood derived osteoclasts at levels that are clinically relevant after MOMHR. Peripheral blood mononuclear cells from healthy donors were seeded onto dentine wafers, and treated to transform them into osteoclasts using standard techniques in the presence of various clinically relevant concentrations of Co2+, Cr3+, and Cr6+. After 3 weeks of culture osteoclast number and resorption pit formation was quantified using histological techniques. All 3 metal ions had a dose-dependent effect on both osteoclast formation and resorption activity. At ion levels found in serum after MOMHR, an increase in osteoclast formation and bone resorption was found, but at higher levels found in synovial fluid, osteoclast cell proliferation and resorption activity was decreased, likely due to a direct toxic effect of the ions on the cells (Figure 1). Cr6+ was more toxic than the other ions at higher concentrations. Our data suggest that metal ion release following MOMHR may increase osteoclast activity systemically that might have a deleterious effect on general and local bone health, and may contribute to the observed bone related complications of MOMHR.

Measurements of biochemical markers of bone turnover have been explored as a diagnostic tool for the detection of osteolysis after THA, but their predictive value in individual subjects has been poor. One explanation for this low diagnostic utility is that the mechanism of bone resorption in osteolysis may be different to that occurring in other high bone turnover states, such as osteoporosis, where these markers were principally developed. The aim of this study was to examine the role of the biomarkers urinary ααCTX-I and serum CTX-MMP, that are released in pathological rather than physiological bone turnover states, for detecting periprosthetic osteolysis in a case control study of 23 subjects with osteolysis and 26 controls. All samples were collected between the hours of 0800 and 1000 following an overnight fast, and were assayed using standard techniques. The demographic characteristics of the subjects in both groups were similar. Serum CTX-MMP was greater in the osteolysis versus the control group (P=0.001). Urinary ααCTX-I was similar between osteolysis and control groups (P> 0.05). A cut-off value of 5.50ng/mL CTX-MMP had a sensitivity of 91% (95% CI: 72 to 99) and specificity of 69% (48 to 96) detecting osteolysis (P=0.001). The same cut-off had a sensitivity of 100% (100 to 100) and specificity of 63% (44 to 79) for detecting femoral osteolysis (P=0.0004), and a sensitivity of 89% (65 to 98) and specificity of 58% (39 to 75) for identifying pelvic osteolysis (P=0.014). Serum CTX-MMP shows promise for further investigation as a sensitive bio-marker for detecting periprosthetic osteolysis.

In this 2-year randomised clinical trial we examined whether cemented femoral prosthesis geometry affects the pattern of strain-adaptive bone remodelling in the proximal femur after THA. 128 patients undergoing primary THA were randomised to receive a Charnley (shape-closed, no taper), Exeter (force-closed, double-tapered) or C-stem (forced-closed, triple-tapered) prosthesis. All received a cemented Charnley cup. Proximal femoral BMD change over 2 years was measured by DXA. Urine and serum samples were collected at pre-operative baseline and over 1 year post-operatively. N-telopeptides of type-I-collagen (NTX) was measured in urine as a marker of osteoclast activity and Osteocalcin (OC) in serum as a maker of osteoblast activity. Clinical outcome using the Harris and Oxford hip scores, and prosthesis migration measured using digitised radiographs (EBRA-Digital) were measured over 2 years. The baseline characteristics of the subjects in each group were similar (P> 0.05). Decreases in femoral BMD were observed over the first year for all prosthesis designs. Bone loss was greatest (14%) in the proximal medial femur (region 7). The pattern and amount of bone loss observed was similar between all prosthesis designs (P> 0.05). Transient rises in both osteoclast (NTX) and osteoblast (OC) activity also occurred over year 1, and were similar in pattern in the 3 prosthesis groups (p> 0.05). All prostheses showed migration patterns that were true to their design type and similar improvements in clinical hip scores were observed over the 2 year study. Differences in the proposed mechanism of load transfer between prosthesis and host bone in force-closed versus shape-closed femoral prosthesis designs in THA are not major determinants of prosthesis-related remodelling.

Introduction: Polymorphisms within genes encoding bone regulatory cytokines influence individuals’ susceptibility to osteolysis after THA. We aimed to determine whether single nucleotide polymorphisms (SNPs) within these genes influence the severity of these osteolytic lesions in 272 patients with established aseptic loosening.

Methods: Assessment of osteolytic lesions was made from pre-revision radiographs in conjunction with direct visualisation in those subjects undergoing surgery. Osteolytic lesions were defined as linear (AAOS pelvic and femoral osteolysis classification grade 0) or expansile, in the presence of segmental or cavitary defects (AAOS grade 1 or greater). We analysed 11 SNPs in the pro-inflammatory cytokines IL-1A, IL-1B, IL-1RA, IL-6 and TNF; 2 SNPs within the FRZB gene, which modulates osteoblast function; and 6 SNPS in the RANK/RANKL/OPG pathway, that modulates osteoclast function.

Results: Femoral Osteolysis: Carriage of the IL-6 −174C allele was 60% in the expansile osteolysis group versus 80% in the linear osteolysis group (χ2 test p=0.007). Carriage of the OPG −163G allele was 34% in the expansile osteolysis group versus 18% in the linear group (χ2 test p=0.03). The odds ratios for expansile osteolysis associated with carriage of IL-6-174G and OPG −163G were 2.7 (1.3 to 5.7, p=0.008) and 2.3 (1.1 to 5.0, p=0.03) respectively.

Acetabular Osteolysis: No differences in SNP genotype were found between osteolysis groups.

Discussion: The IL-6-174G allele and the OPG-163G allele are over-represented in subjects with expansile femoral versus linear osteolysis, but do not relate to severity of pelvic osteolysis. These differences in association may reflect differences in the mechanism of osteolysis between the bone sites, however, replication of the results are required to confirm this differential association.

Total hip arthroplasty (THA) wear debris induced macrophage expression of pro-inflammatory cytokines has been associated with osteolysis both in vitro and in animal and human subjects. Interleukin-1 receptor antagonist (IL-1RA) is an anti-inflammatory cytokine which may limit bone destruction. Polymorphisms (SNPs) within the IL-1RN gene are associated with differences in susceptibility to infectious and inflammatory conditions and disorders of bone remodelling. This study investigated the association between the IL-1RA+2018T/C SNP (rs419598) and osteolysis after THA, and with mRNA and protein expression in an in-vitro wear debris-macrophage stimulation assay.

611 North European Caucasians who had received a cemented THA for primary osteoarthritis were genotyped for the IL-1RN+2018 SNP using Taqman methods. 62 subjects with a Charnley THA were selected from the genotyping population. Control subjects had no radiographic osteolysis and the osteolysis group had previously undergone revision surgery for aseptic loosening. Peripheral blood mononuclear cells were extracted and stimulated with endotoxin-stripped titanium particles (TiCL, endotoxin level 0 Eu/ml) and endotoxin-stripped particles with adherent LPS added back (TiAB, endotoxin level 140 Eu/ml); non-stimulated and LPS-stimulated cells were used as negative and positive controls. Cell lysate IL-1RA mRNA levels were assessed by rqRT-PCT following a 3-hour stimulation. Cell supernatant IL-1RA protein levels were assayed after 24 hours stimulation using a multiplex method.

The IL-1RN+2018C allele was underrepresented in patients with osteolysis after THA versus control THA subjects (chi-squared test 5.96, P=0.015). After correction for other risk factors for osteolysis, the adjusted odds ratio for osteolysis associated with carriage of the IL-1RN+2018C SNP was 0.69 (0.48 to 0.99, p=0.048). IL-1RA mRNA expression increased linearly with IL-1RN+2018C allele copy number in gene-dose dependent manner (ANOVA p=0.013). The IL-1RA+2018C allele did not significantly affect IL-1RA protein expression (ANOVA p> 0.05), however a similar trend towards increased levels with increased C allele copy number was observed.

Carriage of the IL-1RA+2018C allele is associated with both a decreased risk of osteolysis after THA and increased IL-1RA mRNA expression in-vitro. The mechanism for this functional effect remains unclear, however these findings support the importance of the IL-1RA in osteolysis and aseptic loosening.

Introduction: Immune responses in patients susceptible to aseptic loosening may differ to those without this susceptibility. We compared stimulated cytokine mRNA and protein expression in peripheral blood mononuclear cells (PBMC) in 34 subjects (M:F 16:18; mean age 75 years) with previous revision surgery for aseptic loosening versus 28 subjects (14:14; 75 years) with well-fixed implants after Charnley THA for osteoarthritis.

Methods: Extracted PBMCs were stimulated with endotoxin (LPS 200ng/mL), endotoxin-free titanium particles (Ti, endotoxin level =0 Eu/mL), or particles with adherent LPS (TiLPS, 140 Eu/mL). Cell lysate IL-1α, IL-1β, IL-1RA, IL-6, IL-10, IL-18, and TNF mRNA were assayed after 3 hours stimulation using standard rqRT-PCR techniques. Cell supernatant IL-1β, IL-1RA, IL-6 and TNF protein were assayed after 24 hours stimulation using a multiplex method.

Results: mRNA and protein levels in non-stimulated cells were lower in revision versus control subjects for all cytokines (p< 0.05 all analyses). mRNA expression relative to baseline was greater in revision subjects versus controls for all cytokines and all modes of stimulation (LPS, Ti, and TiLPS, p< 0.05 all analyses). LPS induced the greatest inflammatory cytokine response at both the mRNA and protein level in both groups, TiLPS particles induced a more attenuated response, and responses to Ti particles were weakest. In the control group endotoxin free particles showed a negative cytokine mRNA response for IL-1α, IL-1β, and IL-6 (p< 0.05), and reduced protein levels for IL-1β, IL-1RA, IL-6, and TNF versus non-stimulated cells (p< 0.05).

Discussion: Patients with a susceptibility to aseptic loosening have lower baseline but greater stimulated immune responses versus patients without loosening that may contribute to the pathogenesis of aseptic loosening.

Introduction: The concept that aseptic loosening is a function of polyethylene wear has led to the use of cross-linked polyethylene in total hip arthroplasty (THA). We studied the relationship between polyethylene wear rate and aseptic loosening in order to model the potential effects of wear-reducing strategies on the failure rate for each prosthetic component.

Methods: 350 subjects who had previously undergone Charnley THA were divided into 3 groups: Controls (n=273); those with loosening of only the femoral stem (n=43); and those with only cup loosening (n=34). Poly-ethylene wear was measured using a validated method (EBRA). The relationship between wear rate and loosening was examined using logistic regression analysis, and estimates of the effect of wear rate modulation made using odds-ratios.

Results: The median annual wear rate in the controls (0.07mm) was lower than both stem looseners (0.09mm, p=0.002) and cup looseners (0.18mm, p< 0.001). The odds of cup loosening increased 4.7 times per standard deviation (SD) increase in wear rate above the reference (control) population (p< 0.001). The odds of stem loosening increased 1.7 times per SD, but was not independent of other risk factors (p> 0.05). The potential reduction in risk of loosening was calculated using the following formula: (OR^SD2)/(OR^SD1), where 1 and 2 are the new and old mean z-score wear rates. Thus, for a 25% or 50% reduction in wear rate, the incidence of cup loosening may reduce by 71% and 293%, respectively. The rate of stem loosening may, at best, reduce by 7% and 17%, respectively.

Discussion: Wear reduction strategies, such as cross-linked polyethylene, have the potential for a major impact on the incidence of cemented cup, but not stem, loosening.

Introduction: Activated peri-prosthetic macrophages release pro-inflammatory cytokines, including interleukin-6 (IL-6), that stimulate osteoclast activation and aseptic loosening. Natural sequence variations (polymorphisms) within the IL-6 gene promoter region are associated with diseases characterised by increased osteoclast activity, including osteoporosis, and affect IL-6 production in-vitro. We tested whether polymorphisms in the IL-6 gene promoter influence the risk of aseptic loosening after total hip arthroplasty (THA).

Methods: 614 Caucasians, 292 men and 322 women, mean age 75.8 years who had undergone primary cemented THA for idiopathic osteoarthritis a mean of 13.4 years previously were recruited. Peripheral blood was taken and DNA extracted using standard techniques. Subjects were genotyped for the IL-6 -174, -572, and -597 promoter single nucleotide polymorphisms using the Taqman 5′ nuclease method.

Results: The allele frequencies and carriage rates for both alleles at promoter positions −174, −572, and −597 were similar between controls and aseptic loosening subjects (Table, χ² P> 0.05 all comparisons).

Discussion: Although Il-6 has been implicated in the pathogenesis of aseptic loosening and the −174, −572, and −597 polymorphisms are associated with bone loosing pathologies, they do not appear to play a major role in aseptic loosening after THA.

Aseptic loosening arises when periprosthetic bone loss results in mechanical failure at the host-implant interface, and is the main factor limiting implant survival after total hip arthroplasty (THA). The aims of this study were to determine whether genetic variation is a risk factor for loosening, explore the metabolic mechanisms of periprosthetic bone loss, and determine whether bisphosphonates may prevent bone loss and enhance implant mechanical stability after THA. In a genetic association study (J Bone Mineral Res2003; 18:1995–2001) we found that carriage of the −238A allele within the promoter region of the TNF gene was an independent risk factor for aseptic loosening. A subsequent reporter gene assay showed differential TNF gene responsiveness between the –238A and –238G alleles to polyethylene particule stimulation (Calcified Tissue Int 2003; 72: 251-273). In a cross-sectional study (J Orthop Res 2003; 214: 691–696) we found that subjects with aseptic loosening had lower bone mineral density (BMD) in the region of the femoral calcar and higher urinary excretion of cross-linked collagen breakdown products than their counterparts with fixed femoral implants. In a randomised controlled trial we found that a single dose infusion of a bisphosphonate (pamidronate) reduced femoral bone loss over 2 years after THA, but did not affect pelvic bone loss or implant migration (J Bone Miner Res2002; 17: 1328). Transient increases in bone turnover markers occurred after surgery and were highly predictive of later femoral BMD change. The main predictor of early implant migration was patient age, but not periprosthetic BMD change. In summary, genetic as well as environmental factors affect implant survival after THA. Aseptic loosening is associated with regional changes in bone mass and turnover as well as focal osteolytic lesions. Bisphosphonate therapy is well tolerated after THA and has a clear biological effect. However, the impact of preventing early bone loss on late aseptic loosening remains unclear and awaits long term study.

[Winner, Robert Jones Gold Medal and Association Prize, 2003]

Dual energy X-ray absorptiometry (DXA) is a precise tool for measuring bone mineral density (BMD) around total joint prostheses. The Hologic ‘metal-removal hip’ analysis package (Hologic Inc, Waltham, Massachusetts) is a DOS-based analysis platform that has been previously validated for measurement of pelvic and proximal BMD after total hip arthroplasty (THA). This software has undergone a change in the operating platform to a Windows-based system that has also incorporated changes to DXA image manipulation on-screen. These changes may affect the magnitude of random error (precision) and systematic error (bias) when compared with measurements made using the previously validated DOS-based system. These factors could influence interpretation of longitudinal studies commenced using the DOS system and later completed using the Windows system. The aims of this study were to compare the precision and bias of pelvic and femoral periprosthetic BMD measurements made using the Windows versus the DOS analysis platform of the Hologic ‘metal-removal hip’ software. A total of 29 subjects (17 men and 12 women) with a mean age of 51years (SD±10), who had undergone hybrid THA using a cemented stem and uncemented cup. Subjects underwent duplicate DXA scans of the hemipelvis and proximal femur taken on the same day after a period for repositioning.. Scans were obtained with the patient lying supine in the scanner with the legs in extension and the foot in a neutral position. Scans were carried out using the same Hologic QDR 4500-A fan-beam densitometer in ‘metal-removal hip’ scanning mode. The DXA scan acquisitions were analysed using both the DOS and the Windows versions of the analysis software. The same observer made all analyses (NRS). Pelvic scans were analysed using a four region of interest model and femoral scans were analysed using a seven region of interest model. Precision was expressed as coefficient of variation (CV%) and compared between methods using the F-test. Systematic bias was examined using the Bland and Altman method and paired t-test. The CV% for the pelvic regions of interest (n=4) varied from 3.92 to 8.54 and from 2.36 to 5.96 for the Windows and DOS systems, respectively. The CV% for the net pelvic region was 3.04 and 2.36 for Windows versus DOS, respectively (F- test, p> 0.05). The CV% for the femoral regions of interest (n=7) varied from 1.58 to 4.14 and from 1.84 to 4.65 for the Windows and DOS systems, respectively. The CV% for the net femoral region was 1.75 and 1.51 for Windows versus DOS, respectively (F- test, p> 0.05). Absolute BMD values for the net pelvic region were similar (Bland-Altman, Windows minus DOS value mean = -1.0%, 95% CI −7.5 to 5.6; t-test p.0.05). Absolute BMD values for the net femoral region were also similar (Bland-Altman, Windows minus DOS value mean = 1.3%, 95% CI −8.3 to 10.8; t-test p.0.05). In summary precision of the measurements using the 2 operating systems was similar and there was no systematic bias between methods. These data suggest that scans analysed using each platform may be used interchangeably within the same study subjects, without the need of a calibration correction.

Peri-prosthetic bone loss may contribute to aseptic loosening after THA. The aims of this randomised controlled trial extension study were to study the effect of pamidronate therapy on Peri-prosthetic bone mineral density (BMD) and Peri-prosthetic osteolysis over 5 years after primary THA.

50 patients were enrolled in the study in 1998. All received a hybrid THA (Ultima-TPS stem, Plasmacup) for osteoarthritis. Subjects were randomised to receive either 90mg of pamidronate or placebo by intravenous infusion on the 5^th post-operative day. At 5 years 36 patients (41 Hips: placebo n=21, pamidronate n=20) returned for measurement of BMD and clinical and plain radiographic assessment. Five patients had died and nine had withdrawn from the study.

The effect of pamidronate in maintaining femoral bone mass in the region of the calcar previously reported at 2 years was maintained at 5 years (Gruen zone 6 pamidronate versus placebo ANOVA P=0.038; Gruen zone 7 ANOVA P=0.048). No differences in pelvic BMD were found between treatment groups at 5 years. Harris hip scores used to evaluate clinical outcome did not show any significant difference between the 2 groups over the 5-year period. (Mann Whitney p> 0.05). Isolated expansile osteolytic lesions were identified on AP radiographs of the hip at 5 years in 4 patients (2 placebo, 2 pamidronate; P> 0.05). One patient had a 5x9mm lytic lesion in the region of the femoral calcar, and 3 patients had pelvic lytic lesions in the region of the acetabular dome (largest measuring 20x10mm).

Single-dose peri-operative pamidronate therapy preserves femoral calcar bone mass over a 5 year period after THA. However, although the number of subjects with osteolysis is small, we have seen no difference in the rate of osteolytic lesions between treatment groups. Long term study of this patient group is required to examine the rate of aseptic loosening between the treatment groups.

The pattern and magnitude of pelvic periprosthetic bone loss around cementless metal-backed acetabular implants have previously been described. The pattern of periprosthetic BMD change around cemented all-polyethylene acetabular implants is unreported. The aims of this study were to determine the precision of pelvic BMD measurements around the Charnley cup and to examine the longitudinal pattern of BMD change over the first 2 years after surgery.

19 subjects who had previously received a Charnley cup for osteoarthritis underwent duplicate measurements of pelvic BMD after repositioning using an Hologic QDR 4500A densitometer. Scan analysis was carried out using a 4-region of interest model according to a protocol previously described. In-vivo precision was expressed as coefficient of variation (CV%) for each region of interest. The precision of pelvic periprosthetic BMD measurements were 7.7%, 9.8%, 10.8%, and 9.9% for regions 1 to 4, respectively.

Longitudinal BMD changes were measured over a 2 year period in 32 patients (mean age 74 years; 22 women) undergoing cemented THA for unilateral osteoarthritis (17 right-sided). Transient decreases in BMD were observed in regions 2 and 3 (behind the dome of the implant) at 3 months (−9.0% and −13.2%, respectively; P< 0.05) and at 1 year (−8.1% and −9.3%; P< 0.05). By 2 years there had been some recovery in bone mass (BMD−6.9% and −2.6% respectively). No significant changes in BMD for regions 1 and 4 (located at the rim of the implant) were found.

The precision of pelvic periprosthetic BMD measurements for the cemented Charnley cup are poorer than those we have previously reported for cementless cups and may be due, in part, to cement artifact. The pattern of BMD change observed for the Charnley implant suggests that load transfer between the implant and the pelvis occurs principally at the implant rim. The magnitude of bone loss is similar to that we have previously reported for cementless metal-backed acetabular implants.

The Exeter (Howmedica Ltd) and Ultima-TPS (Depuy Ltd) implants are both collarless, polished, double-tapered, cemented femoral implants. The Exeter is manufactured in stainless steel and has an excellent long-term survivorship. The Ultima-TPS is manufactured in cobalt-chrome and has been recently introduced. The aim of this study was to compare the early performance of these implants in a 2-year randomised clinical trial.

65 patients with unilateral hip osteoarthritis were randomised to receive either the Exeter or TPS stem. All received a Charnley Cup. Outcome measures included the Oxford Hip Questionnaire, proximal femoral bone mineral density (BMD) measured by dual energy x-ray absorptiometry, and implant subsidence measured using EBRA. At 2 years 43 patients (66%) were reviewed. 22 patients (mean age 70 years, 16 female, BMI 27.9Kg/m²) received the TPS implant, and 21 patients (mean age 70 years, 15 female, BMI 28.9Kg/m²) received the Exeter implant. 19 patients withdrew for reasons unrelated to the study, 2 died, and 1 was withdrawn after deep wound infection.

Complete Oxford hip scores were available pre-operatively and at 2 years in 37 patients (n=20 TPS). Median (IQR) pre-operative hip scores were 51 (43 to 54) and 48 (36 to 53) for the TPS and Exeter implants, respectively. At 2 years the hip scores improved to 24 (18 to 31) and 22 (16 to 31), respectively. There were no differences in scores between groups at each time-point. There were no differences in BMD between groups at pre-operative baseline, 3 months, 1 and 2 years (Gruen zones 1–7, all time-points; n=19 TPS, n=13 Exeter implants: P> 0.05). Maximum bone loss was seen in Gruen zone 7 at 2 years for bone implants (TPS-11%, Exeter -14%, P> 0.05). Measurement of subsidence over 2 years using EBRA was possible in 20 patients (n=7 TPS, n=13 Exeter). Mean subsidence at 2 years was 1.62mm for the TPS implant and 1.60mm for the Exeter implant (P> 0.05). There was no plain radiographic evidence of osteolysis in either group.

These data suggest that the early performance of the two implants studied is similar. However, long-term survivorship data is required to confirm their equivalency.

Phagocytosis of wear particles by perimplant macrophages results in cytokine release and osteoclast activation and osteolysis. Some investigators have proposed that this response may be mediated by adherent endotoxin. The aim of this study was to determine the role of endotoxin in modulating pro-inflammatory cytokine mRNA expression of macrophages when stimulated with titanium particles using relative quantitative real-time polymerase chain reaction (rqRT-PCR)

Human peripheral blood mononuclear cells were isolated from healthy subjects and plated in chamber slides. Three types of titanium particles were prepared; commercially pure titanium particles (cpTi), endotoxin stripped particles and endotoxin stripped particles with endotoxin (LPS) added back. Endotoxin levels of 450, 0 and 140 Eu/ml respectively were confirmed by high sensitivity Limulus Amebocyte Lysate assay. Macrophages were stimulated with particle concentrations of 0, 8.3, 83 and 830 particles per cell at time points 0 and 3 hours. LPS (200ng/ml) was used as a positive control. rqRT-PCR was performed using standard techniques.

Stimulation of human macrophages with cpTi demonstrated a significant dose dependent increase in TNFα, IL-1A, IL-1B and, IL-6. (Kruskal-Wallis p=0.01, p=0.017, p=0.001 and p=0.013 respectively). IL-18 mRNA levels were not increased (P> 0.05). The expression of mRNA following stimulation with the highest dose of titanium particles was similar to that following LPS stimulation. Endotoxin-free cpTi particles did not elicit any increase in mRNA expression above base line levels (P > 0.05, all cytokines). This lack of response was rescued in endotoxin-stripped particles with LPS added back. Particle dose dependent increases in cytokine mRNA levels were observed for TNFα, IL-1A, IL-1B and, IL-6 mRNA but not IL-18 (p=0.01, p=0.01, p=0.01, p=0.05 and p=0.> 0.05 respectively).

Our results show that adherent endotoxin plays a role in modulating particle induced pro-inflammatory cytokine mRNA expression in-vitro. Further study is required in evaluating the role of adherent endotoxin in vivo

Cytokine mediated activation of osteoclasts can lead to peri-implant osteolysis and aseptic loosening. The aim of this study was to determine the IL-1β and TNFα mRNA cytokine expression profile of human macrophages when stimulated with polyethylene particles using relative quantitative real-time polymerase chain reaction (rqRT-PCR).

Human peripheral blood monocytes or human monocytes from the cell line THP-1 were used in this study. rqRT-PCR conditions were optimized by stimulating human macrophages with 200ng/ml lipopolysaccharide (LPS). The median CV% value for duplicate measures was 12.6 (range 4.5–54). Stimulation assays were performed using unfractionated endotoxin-free commercial polyethylene particles (median size 7μm); or fractionated particles (size range 0.1–1.2μm). Human macrophages were stimulated with high dose unfractionated polyethylene particles at 0, 3500 or 10500 mm³/cell or with fractionated polyethylene particles at 0 and 100mm³/cell at time points 0 and 3 hours. Low dose unfractionated polyethylene stimulation was performed on THP-1 cells at 0, 50, 100, 1000 and 10000 mm³/cell. In all experiments LPS stimulation was used as a positive control. RNA was extracted and rqRT-PCR was performed using standard techniques

High dose unfractionated polyethylene stimulation did not result in a significant difference in cytokine mRNA levels between groups. Using fractionated polyethylene, a small increase in IL-1β mRNA was identified (21% versus maximal stimulation using LPS). Low dose unfractionated polyethylene stimulation of THP-1 cells demonstrated dose dependent decreases in TNFα and IL-1β mRNA expression that was not due to inhibition of RNA extraction or a decrease of cell viability.

Endotoxin-free polyethylene particles do not appear to be a major stimulus for IL-1β and TNFα mRNA production as measured by rqRT-PCR. We did observe a small positive effect on IL-1β mRNA expression using a fractionated polyethylene stimulus. However it remains unclear whether this effect is due to fractionation of particles into the submicron range or is due to introduction of endotoxin during the filtration process.

We compared the long-term clinical outcome scores of the Stanmore total hip arthroplasty (THA) in patients with rheumatoid arthritis (RA, n=26 subjects) versus osteoarthritis (OA, n=35 subjects) at a mean of 12 years after THA. Patients with RA were a mean of 11 years younger at review (66 years, P< 0.001) than those with OA. A greater proportion of RA patients had bilateral THA (19/26 versus 12/35, p=0.03), and were of Charnley grade C (23/26 versus 2/35, p< 0.001). The proportion of male versus female subjects and body mass index were similar between groups (p> 0.05 all comparisons). The overall SF-12 score and SF-12 physical component score were 8% and 15% poorer, respectively, in subjects with RA versus those with OA (P< 0.05). The hip-specific Oxford and Harris hip scores, however, were similar between groups (p> 0.05). Within the individual domains of the Harris hip score, patients with RA had poorer scores for walking distance, stair climbing, putting on of socks/shoes, and ability to enter public transport (p< 0.05 all comparisons). The other domains of pain, limp, use of walking aids, sitting, deformity and range of movement were similar between groups (p> 0.05). The observed differences in outcome scores between RA and OA groups were independent of age and whether the patient had bilateral THA (ANOVA, p> 0.05). Clinical outcome scores in the long term after THA are poorer in RA subjects versus OA. The principal differences occur in the ability to walk long distances, and the use of stairs and public transport.

The aim of this study was to determine whether there is evidence to support the [often quoted] concept of a threshold effect of implant wear rate on osteolysis risk after total hip arthroplasty (THA). The study design was a case control study of 115 subjects with osteolysis after Charnley THA for idiopathic osteoarthritis (mean age at primary surgery 61.1 years; M:F =49:66; osteolysis-free survival 10.9 years) compared with 115 individually case-matched subjects following Charnley THA for idiopathic osteoarthritis with no current radiographic evidence of osteolysis (mean age 61.3 years; M:F = 49:66; osteolysis-free survival 11.0 years).

Calculated median (interquartile range) annual linear wear rate (measured using the EBRA method) was 0.12mm (0.08 to 0.18) and 0.07mm (0.05 to 0.10) in the osteolysis and control groups, respectively (Wilcoxon, P< 0.001). Subjects were divided into wear quintiles based on wear rate (n=46 subjects per quintile). The proportions of osteolysis subjects in each successive wear quintile groups were 0.22, 0.39, 0.48, 0.61, and 0.80 (χ² P< 0.001). The proportion of subjects with osteolysis thus increased in a uniform manner with no evidence of a disproportionate increase between groups. The odds-ratio for osteolysis for each incremental increase in annual linear wear above the median wear rate in the control subjects was 2.4 (logistic regression analysis, 95% CI 1.7 to 3.3, P< 0.001).

In summary, the proportion of subjects with osteolysis increases steadily by wear quintile. Our data suggest a continuous gradient of risk for osteolysis associated with increasing annual wear rate in the Charnley prosthesis. We found no evidence to support the concept of a defined threshold above which the risk of osteolysis is disproportionately increased. The implication of this finding is that the goal of advances in bearing surface technologies should be aimed at the elimination of wear, rather than simply it’s reduction to below an arbitrarily-defined level

Information on the complication rates of revision THA is well documented. However, there is little data on functional outcome of revision THA. We aimed to determine the functional outcome of revision THA (n=72 subjects) versus individually matched THA controls. All subjects underwent THA for idiopathic osteoarthritis, and the same investigator made all clinical assessments. The mean ages (±SD) at primary THA were 61.3±7.2 years (THA revisions) and 61.1±7.4 years (THA controls). The male: female ratio was 36:36 in both groups. The groups were also individually matched for primary THA year (median 1984), presence of bilateral THA (43 subjects per group), and total follow up time (mean 14±4 years). Revision-free survival in the THA revision group was 9.8±3.9 years, and post revision follow up was 4.5±3.0 years.

Sixteen subjects had revision of 1 implant component and 56 had both revised. Allograft was required in 25 and 17 of the cup and stem revisions, respectively. The median (Interquartile range) Oxford and Harris Hip Scores in the revision and control groups were 28 (21 to 39) and 72 (60 to 86) versus 21 (16 to 32) and 89 (79 to 97), respectively (Wilcoxon, P< 0.001 both comparisons). The largest difference in Harris Hip Score was found in the function domain; revision THA median score 24 (17 to 36) versus 38 (28 to 44) in the controls (P< 0.001). Male subjects had slightly better outcomes versus females in both groups (P< 0.05). Revision of both versus 1 component, bilateral THA, age at revision, and use of allograft did not affect outcome (P> 0.05 all comparisons).

The clinical outcome of revision hip arthroplasty for aseptic loosening is worse than that of primary arthroplasty, principally in terms of function. However, use of allograft, number of components revised, and age at revision are not strongly associated with clinical outcome of revision surgery.

Total hip arthroplasty (THA) patients often require peri-operative blood transfusion. Variables that predict transfusion requirement may allow us to target cross-matching of individual patients.

153 patients underwent primary unilateral THA for osteoarthritis or rheumatoid arthritis during 2002 in our institution. 75 casenotes from these subjects were reviewed. Age, sex, diagnosis, weight, height, pre-operative haemoglobin (Hb) and haematocrit (Hct), anticoagulation type and timing were recorded, along with post-operative Hb and timing and quantity of any blood transfusion. Potential predictors of transfusion were examined using logistic regression analysis. ROC analysis was used to compare the relative predictive value of significant variables.

Mean (±SD) age at surgery was 67±11 years (53% females). Mean pre-operative Hb was 13.8±1.4g/dl, mean post-operative Hb was 10.2±1.0g/dl. 27 patients (36%) needed a transfusion; the most frequently given volume was 2 units and the mean number of units given was 0.85. The most common reason for transfusion was an asymptomatic low Hb (< 8.0g/dl). Pre-operative Hb and Hct were predictive of post-operative transfusion (logistic regression analysis P< 0.01). Age, gender, diagnosis and anticoagulation were not predictive. Using ROC analysis the optimal ‘cut-off’ value of pre-op Hb as a predictor was 12.7 g/dl, giving a sensitivity of 41% and a specificity of 88% for blood transfusion requirement. The optimal ‘cut-off’ for Hct was 0.41, sensitivity 74% and specificity 61%. There was no significant difference in the overall predictive value between these variables (comparison of area under ROC curves, P> 0.05).

In summary, subjects with a pre-operative Hb< 12.7 or Hct < 0.41 are more likely to require a blood transfusion after unilateral primary THA than those with an Hb or Hct above these values. In treatment centres where cross-matched blood is not available at short notice on demand, pre-operative cross-match of patients with blood counts below these values may be appropriate.

An acute phase of periprosthetic bone loss occurs following total hip arthroplasty (THA). Periprosthetic bone loss undermines implant support, may contribute to its failure, and complicates revision surgery as allograft may be required to replace lost bone. We assessed the effect of a single 90mg dose of the bisphosphonate pamidronate on early periprosthetic bone mineral density (BMD), biochemical markers of bone turnover, and clinical outcome in 47 men and women undergoing hybrid THA in a randomised, double-blinded, placebo-controlled trial.

The mean (± 95% CI) differences in BMD (area under BMD change.time curve) between those receiving pamidronate and those receiving placebo was 0.91(± 0.51) g.weeks/cm² for the proximal femur (P=0.002), and 0.80 (±0.60) g.weeks/cm² for the pelvis (P=0.009). Patients in the pamidronate group had suppression of all biochemical markers of bone turnover compared to placebo (P< 0.05), except for urinary free deoxypyridinoline. Both treatment groups experienced similar improvement in Harris hip and SF-36 UK outcome scores. The frequency of adverse events was similar in each treatment group (placebo 7/24, pamidronate 8/23, P> 0.05).

Acute periprosthetic bone loss following THA is due to a transient increase in bone turnover. A single dose infusion of pamidronate in the early post-operative period significantly reduces this bone loss, and is well tolerated.

We have used a modified technique of cervical osteotomy to treat a consecutive series of 23 patients with chronic slip of the upper femoral epiphysis. It has been successful in correcting both moderate and severe deformities with a low incidence of avascular necrosis, comparable to that seen after subtrochanteric osteotomies. We describe the operative details and discuss the features which make cervical osteotomy technically superior to intertrochanteric and subtrochanteric procedures.

Of a consecutive series of 117 one-year-old infants with 130 established dislocations of the hip, 11% failed to respond to primary surgical treatment. Genetic and iatrogenic factors accounted for half the failures. There were no obvious causes in the remainder, though a few had the superficial stigmata of spinal dysraphism, and by two years of age, most of the group had developed a lateral rotation posture of the affected leg associated with a relatively smaller foot on that side. Radiologically, the femoral head had drifted and rotated laterally out of the surgically deepened acetabulum, causing persistent subluxation. Although there was no clinical evidence of sensory or motor denervation, sensory spinal evoked potential tests revealed the presence of neurological deficits in the majority of patients in this group.

Total hip replacement for adults with unreduced congenital dislocation presents a difficult problem because soft-tissue contractures usually prevent sitting at the normal anatomical level. Extensive soft-tissue division or a high-level acetabulum leads to reduced function and poor fixation of the components. We describe a new technique for hip replacement in such cases. The shortened abductors and flexors are released proximally and excision of the upper third of the ilium allows them to be repaired without tension, while providing bone graft to reconstruct the acetabular roof. We report 12 such replacements in 10 patients with good results and few early complications.

A series of 75 patients who had undergone anterior cervical fusion between 1965 and 1977 were reviewed. The patients were divided into two groups: those in Group A had had the level of fusion indicated by cine radiography, whereas in Group B the level had been determined by plain radiographs and clinical symptoms and signs. Results showed that cine radiography was the more accurate diagnostic technique. Accurate diagnosis of the level to be fused, the careful clinical selection of patients and sound bony union were found to be vital to the success of anterior cervical fusion. The incidence of pseudarthrosis was significant in single-level fusions and was even greater in double-level fusions and in patients with a history of trauma, especially whiplash injuries. It was rare to develop recurrence of symptoms in adjacent levels after fusion of a level localised by cine radiography.

A prospective study of allergic contact dermatitis after metal-on-plastic total hip replacement was undertaken in 69 patients, of whom 54 were available for review after operation. Before operation six patients were metal sensitive, but only one remained so afterwards; this patient had given a clear history of metal sensitivity and a titanium prosthesis had therefore been used. In all six patients the result of the operation was good and no case of loosening occurred. Sixty-three patients had negative patch tests before operation; in none of these was metal sensitivity detected after operation. Cutaneous sensitivity to various metals is well documented after the insertion of metal-on-metal prostheses and in failed prostheses. We have not found any such increased sensitivity after metal-on-plastic hip replacement. There is little evidence of a direct causal relationship between metal sensitivity and subsequent loosening. The cutaneous sensitivity may be the consequence of loosening rather than its cause. Our results suggest that, providing metal-on-plastic prostheses are used, routine patch testing before hip replacement is no longer required.

The features of congenital elevation of the scapula are described for a group of 16 children. Fibrous bands which tether the scapula and limit its movements were discovered in most children. Vertical displacement osteotomy of the medial border of the scapula and division of the fibrous attachments have provided a reliable and safe treatment for 12 children during the past 10 years.

Thirty-one patients have been reviewed four and a half to thirteen years after total excision of the patella for fracture. This operation did not give the uniformly excellent results previously reported by some authors. The type of incision used was unimportant in the long term. Immobilisation in plaster-of-Paris for any period between one and eight weeks after operation had no adverse effect on the long-term results. There was no correlation between the amount of calcification or ectopic bone formation found in the patellar tendon and the degree of function or discomfort in the joint. There was no evidence that osteoarthritis is an inevitable sequel to patellectomy in man. Maximal recovery of knee function may take up to three years after patellectomy. In this series 22% of patients had excellent results, 39% good results and 39% poor results, according to defined criteria. The implications of these findings are discussed.

1. General joint laxity affecting more than three joints was found in 7 per cent of normal schoolchildren. Similar laxity was found in fourteen of a random series of forty-eight girls, and in nineteen of twenty-six boys, with non-familial congenital dislocation of the hip. Such laxity was also found in four of seven girls and five of seven boys with familial (first degree relative affected) congenital dislocation of the hip.

2. It is concluded that persistent generalised joint laxity, which is often familial, is an important predisposing factor to congenital dislocation of the hip in boys. It is less important in girls, except perhaps in familial cases, as in girls there is an alternative temporary hormonal cause of joint laxity.

1. The histories of 149 patients, coming to the Hospital for Sick Children within the first three years of life with congenital dislocation of the hip (191 dislocated hips), and treated by conservative methods, have been reviewed.

2. The patients with unilateral dislocations (107) have been divided into three groups, according to the angle of slope of the opposite acetabulum. This angle was measured on the first radiograph and related to the mean value for age and sex.

3. The opposite hip was classed as "normal" if the acetabular angle was below or within one standard deviation above the mean for sex and age; as "moderately shallow" if it was between one and two standard deviations above the mean; and as "shallow" if it was over two standard deviations above the mean. This grouping was found to have a direct bearing on the results of conservative treatment in unilateral cases. a) Those with "normal" opposite acetabula–accounting for most of the unilateral cases–responded well. b) Those with "moderately shallow" opposite acetabula responded variably. c) The group with "shallow" opposite acetabula usually failed to respond.

4. Most bilateral dislocations behaved as unilateral dislocations with shallow opposite hips.

5. Additional factors influencing the response to conservative treatment–sex, age at first attendance, family history, fragmentation of the femoral epiphysis and eccentric reduction–are discussed.