Aims

As the peak of the COVID-19 pandemic passes, the challenge shifts to safe resumption of routine medical services, including elective orthopaedic surgery. Protocols including pre-operative self-isolation, COVID-19 testing, and surgery at a non-COVID-19 site have been developed to minimize risk of transmission. Despite this, it is likely that many patients will want to delay surgery for fear of contracting COVID-19. The aim of this study is to identify the number of patients who still want to proceed with planned elective orthopaedic surgery in this current environment.

Periprosthetic joint infection (PJI) is a major complication affecting >1% of all total knee arthroplasties, with compromise in patient function and high rates of morbidity and mortality. There are also major socioeconomic implications. Diagnosis is based on a combination of clinical features, laboratory tests (including serum and articular samples) and diagnostic imaging. Once confirmed, prompt management is required to prevent propagation of the infection and further local damage. Non-operative measures include patient resuscitation, systemic antibiotics, and wound management, but operative intervention is usually required. Definitive surgical management requires open irrigation and debridement of the operative site, with or without exchange arthroplasty in either a single or two-stage approach. In all options, the patient's fitness, comorbidities and willingness for further surgery should be considered, and full intended benefits and complications openly discussed. Late infection almost invariably leads to implant removal but early infections and acute haematogenous infections can be managed with implant retention – the challenge is to retain the original implant, having eradicated infection and restored full function.

Debridement with component retention: Open debridement is indicated for acute postoperative infections or acute haematogenous infections with previously well-functioning joints. To proceed with this management option the following criteria must be met: short duration of symptoms - ideally less than 2–3 weeks but up to 6; well-fixed and well-positioned prostheses; healthy surrounding soft tissues. Open debridement is therefore not an appropriate course of management if symptoms have been prolonged – greater than 6 weeks, if there is a poor soft tissue envelope and scarring, or if a revision arthroplasty would be more appropriate due to loosening or malposition of the implant. It is well documented in the literature that there is an inverse relationship between the duration of symptoms and the success of a debridement. It is thought that as the duration of symptoms increases, other factors such as patient comorbidities, soft tissue status and organism virulence play an increasingly important role in determining the outcome.

There is a caveat. Based on our learning in the hip, when we see an acute infection where periprosthetic implants are used, it is much easier to use this time-limited opportunity to remove the implants and the associated biofilm and do a single-stage revision instead of just doing a debridement and a change of insert. This will clearly be experience and prosthesis-dependent but if the cementless implant is easy to remove, then it should be explanted. One critical aspect of this procedure is to use one set of instruments and drapes for the debridement and to then implant the new mobile parts and close using fresh drapes and clean instruments. Units that have gained expertise in single-stage revision will find this easier to do. After a debridement, irrigation, and change of insert, patients continue on intravenous antibiotics until appropriate cultures are available. Our multidisciplinary team and infectious disease experts then take over and will dictate antibiotic therapy thereafter. This is typically continued for a minimum of three months. Patients are monitored clinically, serologically, and particularly in relation to nutritional markers and general wellbeing. Antibiotics are stopped when the patients reach a stable level and are well in themselves. All patients are advised to re-present if they have an increase in pain or they feel unwell.

There has been an evolution in revision hip arthroplasty towards cementless reconstruction. Whilst cemented arthroplasty works well in the primary setting, the difficulty with achieving cement fixation in femoral revisions has led to a move towards removal of cement, where it was present, and the use of ingrowth components. These have included proximally loading or, more commonly, distally fixed stems. We have been through various iterations of these, notably with extensively porous coated cobalt chrome stems and recently with taper-fluted titanium stems. As a result of this, cemented stems have become much less popular in the revision setting.

Allied to concerns about fixation and longevity of cemented fixation revision, there were also worries in relation to bone cement implantation syndrome when large cement loads were pressurised into the femoral canal at the time of stem cementation. This was particularly the case with longer stems. Technical measures are available to reduce that risk but the fear is nevertheless there.

In spite of this direction of travel and these concerns, there is, however, still a role for cemented stems in revision hip arthroplasty. This role is indeed expanding.

First and foremost, the use of cement allows for local antibiotic delivery using a variety of drugs both instilled in the cement at the time of manufacture or added by the surgeon when the cement is mixed. This has advantages when dealing with periprosthetic infection. Thus, cement can be used both as interval spacers but also for definitive fixation when dealing with periprosthetic hip infection. The reconstitution of bone stock is always attractive, particularly in younger patients or those with stove pipe canals. This is achieved well using impaction grafting with cement and is another extremely good use of cement.

In the very elderly or those in whom proximal femoral resection is needed at the time of revision surgery, distal fixation with cement provides a good solution for immediate weight bearing and does not have the high a risk of fracture seen with large cementless stems.

Cement is also useful in cases of proximal femoral deformity or where cement has been used in a primary arthroplasty previously. We have learnt that if the cement is well-fixed then the bond of cement-to-cement is excellent and therefore retention of the cement mantle and recementation into that previous mantle is a great advantage. This avoids the risks of cement removal and allows for much easier fixation. Stems have been designed specifically to allow this cement-in-cement technique. It can be used most readily with polished tapered stems - tap out a stem, gain access at the time of revision surgery and reinsert it. It is, however, now increasingly used when any cemented stems are removed provided that the cement mantle is well fixed. The existing mantle is either wide enough to accommodate the cement-in-cement revision or can be expanded using manual instruments or ultrasonic tools. The cement interface is then dried and a new stem cemented in place.

Whilst the direction of travel in revision hip arthroplasty has been towards cementless fixation, particularly with tapered distally fixed designs, the reality is that there is still a role for cement for its properties of immediate fixation, reduced fracture risk, local antibiotic delivery, impaction grafting and cement-in-cement revision.

Restoring native hip biomechanics is crucial to the success of THA. This is reflected both in terms of complications after surgery such as instability, leg length inequality, pain and limp; and in terms of patient satisfaction. The challenge that remains is that of achieving optimal implant sizing and positioning so as to restore, as closely as possible, the native hip biomechanics specific to the hip joint being replaced. This would optimise function and reduce complications, particularly, instability. (Mirza et al., 2010). Ideally, this skill should also be reproducible irrespective of the surgeon's experience, volume of surgery and learning curve.

The general consensus is that the most substantial limiting factor in a THA is the surgeon's performance and as a result, human errors and unintended complications are not completely avoidable (Tarwala and Dorr, 2011). The more challenging aspects include acetabular component version, sizing and femoral component sizing, offset and position in the femoral canal. This variability has led to interest in technologies for planning THA, and technologies that help in the execution of the procedure. Advances in surgical technology have led to the development of computer navigation and robotic systems, which assist in pre-operative planning and optimise intra-operative implant positioning.

The evolution of surgical technology in lower limb arthroplasty has led to the development of computer navigation and robotics, which are designed to minimise human error and improve implant positioning compared to pre-operative templating using plain radiographs. It is now possible to use pre-operative computerised tomography (image-based navigation) and/or anatomical landmarks (non-imaged-based navigation) to create three-dimensional images of each patient's unique anatomy. These reconstructions are then used to guide bone resection, implant positioning and lower limb alignment.

The second-generation RIO Robotic Arm Interactive Orthopaedic system (MAKO Surgical) uses pre-operative computerised tomography to build a computer-aided design (CAD) model of the patient's hip. The surgeon can then plan and execute optimal sizing and positioning of the prostheses to achieve the required bone coverage, minimise bone resection, restore joint anatomy and restore lower limb biomechanics. The MAKO robotic software processes this information to calculate the volume of bone requiring resection and creates a three-dimensional haptic window for the RIO-robotic arm to resect. The RIO-robotic arm has tactile and audio feedback to resect bone to a high degree of accuracy and preserve as much bone stock as possible.

We have used this technology in the hip to accurately reproduce the anteversion, closure and center of rotation that was planned for each hip. Whilst the precise safe target varies from patient to patient, the ability to reproduce native biomechanics, to gain fixation as planned and to get almost perfect length and offset are a great advantage. Complications such as instability and leg length inequality are thus dramatically reduced.

We live in an era where younger, fitter, more active patients are presenting with the symptoms and signs of degenerative joint disease and require total knee and total hip arthroplasty at a young age. At the same time, this population of patients is living longer and longer and is likely to create new and more complex failure modes for their implants. The ideal solution is a biological one, whereby we can either prevent joint degradation or catch it in its early stages and avoid further deterioration. There may also be advances along the way in terms of partial arthroplasty and focal resurfacing that will help us prevent the need for total joint arthroplasty.

There are several tensions that need to be considered. Should we resurface / replace early, particularly now that we have access to navigation and robotics and can effectively customise the implants to the patient's anatomy and their gait pattern? This would allow good function at a young age. Or should we wait as long as possible and risk losing some function for the sake of preserving the first arthroplasty for the lifetime of the patient?

There are some key issues that we still do not fully understand. The lack of true follow-up data beyond 20 or 30 years is worrying. The data available tends to be from expert centers, and always has a dramatic loss to follow-up rate. We worry about bearing surfaces and how those materials will behave over time but we really do not know the effect of chronic metal exposure over several decades, nor do we really understand what happens to bone as it becomes more and more osteopenic and fragile around implants. We have largely recorded but ignored stress shielding, whereas this may become a very significant issue as our patients get older, more fragile, more sarcopaenic and more neurologically challenged. All the fixation debates that we have grappled with, may yet come back to the fore. Can ingrowth lead to failure problems later on? Will more flexible surfaces and materials be required to fit in with the elasticity of bone?

We have failed dramatically at translating the in vitro to the in vivo model. It seems that the in vitro model tells us when failure is going to occur but success in vitro does not predict success in vivo. We, therefore, cannot assume that long-term wear data from simulators will necessarily translate to the extreme situations in vivo where the loading is not always idealised, and can create adverse conditions.

We must, therefore, consider further how to improve and enhance our interventions. There is no doubt that the avoidance of arthroplasty needs to be at the heart of our thinking but, ultimately, if arthroplasty is to be performed, it needs to be performed expertly and in such a way as to minimise later failure. It also, clearly, needs to be cost-effective. The next stage will no doubt involve close cooperation between surgeons, engineers and industry partners to identify individualised surgical targets, select an appropriate prosthesis to minimise soft-tissue strain and develop a reproducible method of achieving accurate implantation. An ideal outcome can only be achieved by an appropriately trained surgeon selecting the optimal prosthesis to implant in the correct position in the well-selected patient.

In the longer term, our choice of implants and the way that they are inserted and fixed must take into account the evolving physiology of our patients, the nature of our devices and how to limit harm from them, and the long-term impact of the materials used which we sometimes still do not understand.

The infected joint arthroplasty continues to be a very challenging problem. Its management remains expensive, and places an increasing burden on health care systems. It also leads to a long and difficult course for the patient, and frequently a suboptimal functional outcome. The choice of a particular treatment program will be influenced by a number of factors. These include the acuteness or chronicity of the infection; the infecting organism(s), its antibiotic sensitivity profile and its ability to manufacture glycocalyx; the health of the patient; the fixation of the prosthesis; the available bone stock; and the particular philosophy and training of the surgeon.

For most patients, antibiotics alone are not an acceptable method of treatment, and surgery is necessary. The standard of care for established infection is two stage revision with antibiotic loaded cement during the interval period and parental antibiotic therapy for six weeks. Single stage revision may have economic and functional advantages, however. We have devised a protocol that dictates the type of revision to be undertaken based on host, organism and local factors.

Our protocol has included single stage revision using antibiotic loaded cement in both THA and TKA. This was only undertaken when sensitive organisms were identified pre-operatively by aspiration and appropriate antibiotics were available to use in cement. Patients with immunocompromise, multiple infecting organisms or recurrent infection were excluded. Patients with extensive bone loss that required allograft reconstruction or where a cementless femoral component was necessary were also excluded.

Our algorithm was validated first in the hip and extended to infected TKA in 2004. This protocol has now been applied in over 100 TKA revisions for infection between 2004 and 2009. Our single stage revision rate is now over 25%. We continue to see a lower reinfection rate in these carefully selected patients, with high rates of infection control and satisfaction and better functional and quality of life scores than our two stage revision cases.

Whilst our indications are arbitrary and not based on specific biomarkers, we present excellent results for selective single stage exchange. A minimum three year follow-up suggests that these patients have shorter hospital stays, higher satisfaction rates and better knee scores. An ongoing evaluation is in place. One stage revision arthroplasty for infection offers potential clinical and economic advantages in selected patients.

The infected joint arthroplasty continues to be a very challenging problem. Its management remains expensive, and places an increasing burden on health care systems. It also leads to a long and difficult course for the patient, and frequently a suboptimal functional outcome. The choice of a particular treatment program will be influenced by a number of factors. These include the acuteness or chronicity of the infection; the infecting organism(s), its antibiotic sensitivity profile and its ability to manufacture glycocalyx; the health of the patient; the fixation of the prosthesis; the available bone stock; and the particular philosophy and training of the surgeon. Although there have been multiple developments to enhance our ability to effect two-stage techniques whilst limiting inpatient stay, cost and patient morbidity - these include functional spacers, the use of local as well as systemic antibiotics, and home intravenous therapy programmes – there is nevertheless still a considerable morbidity and mortality to the two-stage process, and a massive cost to the patient who has to have two operations with an unpredictable interval period in between and to the local tissues which have already been damaged and are violated on two occasions. The push for one-stage surgery has generally been from centers who are passionate about that technique and has involved a combination of knowing the organism in question prior to surgery, a very radical debridement, the use of hinge / tumor-type implants and prolonged antibiotic therapy post-surgery. The last decade has seen an evolution whereby we have recognised that treatment may be tailored to the patient. There is a big difference between a relatively healthy host and someone with multiple comorbidities, and a big difference between infection with a relatively benign organism and polymicrobial infection with multi-resistant bacteria or fungi. There has, therefore, been increased interest in the use of single-stage revision in order to decrease morbidity, potentially decrease mortality and to decrease cost to the health care system. Single stage revision may have economic and functional advantages, however. We have devised a protocol that dictates the type of revision to be undertaken based on host, organism and local factors.

Whilst we believe that there is a role for both single- and two-stage techniques in our armamentarium, we have gradually evolved to increasing use of single-stage surgery. We use antibiotic-loaded cement whenever possible but can reconstruct most cases using semiconstrained implants without resorting to a hinge.

We continue to see a lower reinfection rate in these carefully selected patients, with high rates of infection control and satisfaction and better functional and quality of life scores than our two-stage revision cases. We use hinge reconstruction in less than 20% of cases.

Since its inception, knee arthroplasty has struggled to balance the requirements of relieving pain and restoring function in a durable way. Although highly successful in improving symptoms as measured by traditional outcome measures and achieving longevity, numerous studies have shown the problems that exist, even with well-implanted components of modern design. Some patients complain of ongoing functional limitation, discomfort, and pain. There are still many challenges in knee arthroplasty. We have a young population that is increasingly active that requires these procedures and yet they are living to a ripe old age and remaining ambulant into their 80s and 90s. We have focussed for the last decade on improving function and satisfaction in knee arthroplasty but we should not forget the fact that the highest failure rate is seen in our young patients and that we really do need a durable solution that will last several decades. There are several tensions that need to be considered. Should we resurface the knee early, particularly now that we have access to navigation and robotics and can effectively customise the implants to the patient's anatomy and their gait pattern? This would allow good function at a young age. Or should we wait as long as possible and risk losing some function for the sake of preserving the first arthroplasty for the lifetime of the patient?

Should we for example accept alignment paradigms that we know give us longevity or should we go with alternative kinematic or anatomical alignment techniques that may well give us better function but could compromise long-term fixation? Both registries and the long-term studies available suggest that we can expect good survivorship into the second decade for older patients and for some into the third decade, but data beyond that is sparse and is not available with contemporaneous implants. Changing the polyethylene in the knee may prove to be successful but may yet be nowhere near as beneficial as it has been in the hip. There has also been all too little work to consider the changing physiology of the bone. Will the increasing trend for cementless implants lead to longer lasting osseointegration or will it lead to periprosthetic fractures through areas of stress shielding? We have been spared somewhat thus far in the knee the issue of local metal ion effects and systemic issues that we have seen in the hip. If our implants last longer and are treated more brutally by an active patient population, we may yet see more problems. At the same time, we have to continue evolving our technologies and yet be cost effective and affordable. Our focus on operative efficiency, early discharge, rapid recovery and a return to full function must not compromise our goals and plans for implant longevity. The next stage will no doubt involve close co-operation between surgeons, engineers and industry partners to identify individual surgical targets, select an appropriate prosthesis to minimise soft-tissue strain and develop a reproducible method of achieving accurate implantation. However, in seeking to solve the problems seen in a proportion of arthroplasty patients, the achievements of ‘traditional’ total knee arthroplasty should not be overlooked. The results achieved by such methods in all three domains: pain relief, functional restoration and longevity, should act as baseline measures for newer techniques and designs. Improvements in any one domain should not be at the expense of another. An ideal outcome can only be achieved by an appropriately trained surgeon selecting the optimal prosthesis to implant in the correct position in the well-selected patient.

The infected joint arthroplasty continues to be a very challenging problem. No test has 100% diagnostic accuracy for PPI and the treating surgeon must correlate the clinical and radiographic presentation with a combination of blood tests, synovial fluid analysis, microbiological and histopathological evaluation of periprosthetic tissue and intra-operative inspection to reach a definitive diagnosis. Diagnosis should begin with a high index of suspicion for new onset of pain or symptoms in well-functioning joints. Plain radiographs may identify osteolysis or early signs of implant failure and should be promptly investigated further for PPI.

Peripheral blood ESR and CRP remain the most widely used next step for the diagnosis of PPI. Both these tests are widely available, inexpensive, and have a rapid turnaround time in laboratories. The results should be interpreted with caution due to their relative lack of specificity. The sensitivity and specificity values for CRP are approximately 88 and 74%, respectively; while that of ESR is slightly lower at 75 and 70%, respectively. The combined ESR and CRP tests are 96% sensitive for ruling out PPI but the specificity of this combination is as low as 56%. Advanced imaging modalities may be used as a part of the diagnostic algorithm. However, they require expert interpretation and are limited by availability and high costs. When available they have high sensitivity and specificity but their routine use is not recommended and indications have to be individualised in the light of clinical presentation.

In the presence of high clinical suspicion, the clinician should plan synovial fluid analysis. This provides a synovial fluid white cell count with differential cell count, specimen for culture and possibility of analyzing other synovial fluid markers. It is important to note that failed metal-on-metal hip arthroplasties can give a falsely elevated synovial fluid cell count when using automated cell counters. This can be overcome by manually counting cell numbers. Synovial fluid should be directly into blood culture bottles, and antibiotics should be withheld at least 2 weeks prior to aspiration, whenever possible. Cultures also help establish the organism, virulence and sensitivities that help plan subsequent treatment algorithm.

Periprosthetic tissue biopsy provides valuable information in microbiological diagnosis and workup of PPI. Routine use of gram staining is not recommended due to poor sensitivity. However, frozen section may have some role especially when performed by a skilled pathologist. Tissue culture remains the gold standard for diagnosis despite false-positive and false-negative results. Whenever possible multiple samples should be obtained to aid interpretation. A threshold of 2 to 3 positive specimens yielding indistinguishable microorganisms has been recommended to improve sensitivity. Acute inflammation, evidenced by neutrophilic infiltrate on fixed or frozen tissue, is suggestive of PPI and is defined as the presence of at least 5 neutrophils per high-powered field, in at least 5 separate microscopic fields.

Sonication of removed prosthetic components is used to dislodge the biofilm and the associated bacteria from the surface of the implant. The fluid surrounding the implant can be used for culture or analysis.

PCR testing: Synovial fluid aspirate, periprosthetic tissue or sonicate fluid may be subject to molecular diagnosis to amplify genetic material and improve microbiological diagnosis of PPI. This technique has shown increased sensitivity in patients who had received antibiotics within 14 days before implant removal. Results have to carefully interpreted with due consideration for possibility of false positive results.

The infected joint arthroplasty continues to be a very challenging problem. No test has 100% diagnostic accuracy for PPI and the treating surgeon must correlate the clinical and radiographic presentation with a combination of blood tests, synovial fluid analysis, microbiological and histopathological evaluation of periprosthetic tissue and intra-operative inspection to reach a definitive diagnosis. Diagnosis should begin with a high index of suspicion for new onset of pain or symptoms in well-functioning joints. Plain radiographs may identify osteolysis or early signs of implant failure and should be promptly investigated further for PPI.

Peripheral blood ESR and CRP remain the most widely used next step for the diagnosis of PPI. Both these tests are widely available, inexpensive, and have a rapid turnaround time in laboratories. The results should be interpreted with caution due to their relative lack of specificity. The sensitivity and specificity values for CRP are approximately 88 and 74%, respectively; while that of ESR is slightly lower at 75 and 70%, respectively. The combined ESR and CRP tests are 96% sensitive for ruling out PPI but the specificity of this combination is as low as 56%. Advanced imaging modalities may be used as a part of the diagnostic algorithm. However, they require expert interpretation and are limited by availability and high costs. When available they have high sensitivity and specificity but their routine use is not recommended and indications have to be individualised in the light of clinical presentation.

In the presence of high clinical suspicion, the clinician should plan synovial fluid analysis. This provides a synovial fluid white cell count with differential cell count, specimen for culture and possibility of analyzing other synovial fluid markers. It is important to note that failed metal-on-metal hip arthroplasties can give a falsely elevated synovial fluid cell count when using automated cell counters. This can be overcome by manually counting cell numbers. Synovial fluid should be directly into blood culture bottles, and antibiotics should be withheld at least 2 weeks prior to aspiration, whenever possible. Cultures also help establish the organism, virulence and sensitivities that help plan subsequent treatment algorithm.

Periprosthetic tissue biopsy provides valuable information in microbiological diagnosis and workup of PPI. Routine use of gram staining is not recommended due to poor sensitivity. However, frozen section may have some role especially when performed by a skilled pathologist. Tissue culture remains the gold standard for diagnosis despite false-positive and false-negative results. Whenever possible multiple samples should be obtained to aid interpretation. A threshold of 2 to 3 positive specimens yielding indistinguishable microorganisms has been recommended to improve sensitivity. Acute inflammation, evidenced by neutrophilic infiltrate on fixed or frozen tissue, is suggestive of PPI and is defined as the presence of at least 5 neutrophils per high-powered field, in at least 5 separate microscopic fields.

Sonication of removed prosthetic components is used to dislodge the biofilm and the associated bacteria from the surface of the implant. The fluid surrounding the implant can be used for culture or analysis.

PCR testing: Synovial fluid aspirate, periprosthetic tissue or sonicate fluid may be subject to molecular diagnosis to amplify genetic material and improve microbiological diagnosis of PPI. This technique has shown increased sensitivity in patients who had received antibiotics within 14 days before implant removal. Results have to carefully interpreted with due consideration for possibility of false positive results.

A large body of the orthopaedic literature clearly indicates that the cement mantle surrounding the femoral component of a cemented total hip arthroplasty should be at least 2 mm thick. In the early 1970s, another concept was introduced and is still in use in France consisting of implanting a canal filling femoral component line-to-line associated with a thin cement mantle. This principle has been named the “French paradox”. An explanation to this phenomenon has been provided by in-vitro studies demonstrating that a thin cement mantle in conjunction with a canal filling stem was supported mainly by cortical bone and was subjected to low stresses. We carried out a study to evaluate the in-vivo migration patterns of 164 primary consecutive Charnley-Kerboull total hip replacements. All prosthesis in the current series combined an all-polyethylene socket and a 22.2 mm stainless steel femoral head. The monobloc double tapered (5.9 degrees) femoral component was made of 316L stainless steel with a highly polished surface (Ra = 0.04 μm), a quadrangular section, and a neck-stem angle of 130 degrees. The stem was available in six sizes with a stem length (shoulder to tip) ranging from 110 mm to 160 mm, and a neck length ranging from 24 mm to 56 mm. For each size, the femoral component was available in two to four different diameters to adapt the implant to the medullary canal. Hence the whole range comprised a total of 18 standard femoral components. The femoral preparation included removal of diaphyseal cancellous bone to obtain primary rotational and varus/valgus stability of the stem prior to the line-to-line cementation. We used the Ein Bild Roentgen Analyse Femoral Component (EBRA-FCA) method to assess the subsidence of the femoral component. At the minimum 15-year follow-up, 73 patients were still alive and had not been revised at a mean of 17.3 years, 8 patients had been revised, 66 patients were deceased, and 8 patients were lost to follow-up. The mean subsidence of the entire series was 0.63 ± 0.49 mm (0 – 1.94 mm). When using a 1.5 mm threshold, only four stems were considered to have subsided. With revision of either component for any reason as the endpoint, the cumulative survival rate at 17 years was 90.5 ± 3.2% (95% CI, 84.2% to 96.8%). With radiological loosening of the femoral component as the endpoint, the cumulative survival rate at 17 years was 96.8 ± 3.1% (95% CI, 93.2% to 100%). This study demonstrated that, in most cases, a highly polished double tapered stem cemented line-to-line does not subside up to 18-year follow-up.

Treatment of recurrent dislocation: approximately: 1/3 of failures (probably higher in the absence of a clear curable cause).

In the US: most popular treatment option: constrained liners with high redislocation and loosening rates in most reports. Several interfaces leading to various modes of failures.

In Europe: dual mobility cups (or tripolar unconstrained): first design Gilles Bousquet 1976 (Saint Etienne, France), consisting of a metal shell with a highly polished inner surface articulating with a mobile polyethylene insert (large articulation). The femoral head is captured into the polyethylene (small articulation) using a snap fit type mechanism leading to a large effective unconstrained head inside the metal cup. With dual mobility, most of the movements occur in the small articulation therefore limiting wear from the large polyethylene on metal articulation.

Contemporary designs include: CoCr metal cup for improved friction, outer shell coated with titanium and hydroxyapatite, possible use of screws to enhance primary stability (revision), cemented version in case of major bone defect requiring bone reconstruction.

Increased stability obtained through an ultra-large diameter effective femoral head increasing the jumping distance.

Dual mobility in revision for recurrent dislocation provided hip stability in more than 94% of the cases with less than 3% presenting redislocation up to 13-year follow-up. A series from the UK concerning 115 revisions including 29 revisions for recurrent dislocation reported 2% dislocation in the global series and 7% re-dislocation in patients revised for instability. A recent report of the Swedish hip arthroplasty register including 228 patients revised for recurrent dislocation showed 99% survival with revision for dislocation as the endpoint and 93% with revision for any reason as the endpoint.

One specific complication of dual mobility sockets: intra-prosthetic dislocation (ie: dislocation at the small articulation): often asymptomatic or slight discomfort, eccentration of the neck on AP radiograph, related to wear and fatigue of the polyethylene rim at the capturing are through aggressive stem neck to mobile polyethylene insert contact (3rd articulation). Risk factors include: large and aggressive femoral neck design implants, small head/neck ratio, skirted heads, major fibrosis and periprosthetic ossifications.

Current (over ?) use in France: 30% of primary THA, 60% in revision THA.

Proposed (reasonable) indications: primary THA at high risk for dislocation, revision THA for instability and/or in case of abductors deficiency, Undisputed indication: recurrent dislocation.

Total knee arthroplasty (TKA) is widely accepted as a successful treatment option for the pain and limitation of function associated with severe joint disease. The ideal knee arthroplasty implant should provide reliable pain relief and normal levels of functional strength and range of motion. However, there are still a number of implant-specific problems following knee arthroplasty, such as irregular kinematics, polyethylene wear and poor range of motion.

MRI and cadaveric studies have highlighted important kinematics during movement of the native knee. In particular, flexion of the joint results in a phenomenon referred to as “roll back and slide”. This essentially describes posterior translation of the femur on the tibia which in turn has a two-fold biomechanical function: to increase the lever arm of the quadriceps and allow clearance of the femur from the tibia in deep flexion. During extension of the joint, the femur rolls forward increasing the lever arm of the hamstrings to act as a brake on hyperextension.

Additional rotation of the joint arises in the axial plane. This is attributed to the concave tibial plateau and relatively fixed meniscus on the medial compartment of the joint in comparison to a lateral convex plateau with a mobile meniscus. This asymmetry allows axial rotation of the lateral compartment over the medial compartment by up to 30 degrees. Subsequently, from extension to full flexion the tibia rotates internally on the femur and vice versa. The external rotation of the tibia on the femur that occurs during the terminal degrees of knee extension is often referred to as the “screw home mechanism” and results in tightening of both the cruciate ligaments locking the knee such that the tibia is in a position of maximum stability on the femur.

Numerous studies over the past two decades have characterised the in-vivo motions of knee replacements. Major conclusions from these studies are that the motion after knee arthroplasty generally does not replicate normal knee motions. In particular, many kinematic studies of unconstrained devices have demonstrated the femur sliding forwards rather than backwards with flexion. This paradoxical movement is also seen in many posterior cruciate retaining knees. This in turn has a negative outcome in range of movement, particularly in light of fluoroscopic studies highlighting strong positive correlations in weight-bearing flexion with femoral roll back. In contrast knee arthroplasties that retain both cruciate ligaments come closest to replicating normal knee motion and furthermore, provide greater stair climbing stability. It may therefore be presumed that this excessive AP motion in a well-designed prosthesis is attributed to a loss in the natural intrinsic stabilizing structures.

A number of studies to date have also highlighted close correlation between knee kinematics and functional strength. Generally, patients with knee replacement exhibit a significant loss of strength compared to normal. The common experimental findings is that knees with the highest intrinsic stability, whether provided by retained ligaments, conforming articular surfaces or post-cam substitution, exhibit the greatest functional strength in high-demand activities in TKA patients.

On the basis of this knowledge, it would be intuitive to choose a TKA design that attempts to restore natural knee joint stability. The medially conforming ‘ball and socket’ articulation of the medial tibio-femoral compartment is a design concept thought to provide stability through the complete arc of knee flexion. Clinical and biomechanical data from a number of centers suggests that this has been a successful evolution in TKA that will continue to benefit patients.

Periprosthetic fractures after total hip arthroplasty lead to considerable morbidity in terms of loss of component fixation, bone loss and subsequent functional compromise. The prevention, early recognition and appropriate management of such fractures are therefore critical. The pathogenesis of periprosthetic factors is multi-factorial. There are a number of intrinsic patient influences such as poor bone stock, biomechanics and compliance. There are also a host of extrinsic factors over which the surgeon has more control. The key tenets for fracture avoidance include careful planning, identifying the risk, choosing the correct implant, understanding the anatomy, and using appropriate surgical technique.

There are a number of recognised risk factors for periprosthetic hip fractures The prevalence of intraoperative fractures during total hip arthroplasty is higher in the patient with osteopenia / osteoporosis. Other conditions causing increased bone fragility, such as osteomalacia, Paget's disease, osteopetrosis, and osteogenesis imperfecta are also at a higher risk of intraoperative fracture. The use of more and more press fit cementless components has also increased the number of periprosthetic femoral fractures because of the force required to obtain such a fit. Complex deformities of the proximal femur, particularly when associated with a narrow medullary canal, may also increase the risk of intraoperative fractures. Revision surgery is associated with a higher risk of intraoperative fracture than primary hip replacement surgery. These fractures typically occur during hip dislocation, cement extraction, or reaming through old cement. Other risk factors for postoperative femoral fractures following total hip replacement include loosening of the prosthesis with cortical bone loss, local osteolysis, stress risers within the cortex, such as old screw holes, the ends of plates, or impingement of a loose stem against the lateral femoral cortex.

The management of periprosthetic fractures requires appropriate preoperative imaging, planning and templating, the availability of the necessary expertise and equipment, and knowledge of the potential pitfalls so that these can be avoided both intraoperatively and in follow-up. There is a danger that these cases fall between the expertise of the trauma surgeon and that of the revision arthroplasty surgeon. The past two decades have afforded us clear treatment algorithms based on fracture location, component fixation and the available bone stock. We still nevertheless face the enduring challenge of an elderly population with a high level of comorbidity who struggle to rehabilitate after such injuries. Perioperative optimization is critical as we have seen prolonged hospital stays, high rates of systemic complications and a significant short term mortality in this cohort.

We have also been presented with new difficult fracture patterns around anatomic cementless stems and in relation to tapered cemented and cementless stems, as well as biologically challenging transverse or oblique fractures at the tip of a stem. In many cases, fixation techniques are biomechanically and biologically doomed to fail and intramedullary stability, achieved through complex revision is required.

The sequelae of periprosthetic fractures include the financial cost of fixation or revision surgery, the associated morbidity and mortality in an elderly frail population, the difficulty with mobilization if the patient cannot fully weight bear, and a poor functional outcome in a proportion of cases. The battle over which patients or fractures require fixation and which require revision surgery continues.

The approach to total hip arthroplasty (THA) should allow adequate visualization and access so as to implant in optimal position whilst minimizing muscle injury, maintaining or restoring normal soft tissue anatomy and biomechanics and encouraging a rapid recovery with minimal complications. Every surgeon who performs primary hip arthroplasties will expound the particular virtues of his or her particular routine surgical approach. Usually this approach will be the one to which the surgeon was most widely exposed to during residency training.

There is a strong drive from patients, industry, surgeon marketing campaigns, and the media to perform THA through smaller incisions with quicker recoveries. The perceived advantage of the anterior approach is the lack of disturbance of the soft tissues surrounding the hip joint, less pain, faster recovery with the potential for earlier return to work, shorter hospital stay and improved cosmetic results. The potential disadvantages include less visibility, longer operation time, nerve injuries, femoral fractures, malposition and a long learning curve for the surgeon (and his / her patients).

The anterior approach was first performed in Paris, by Robert Judet in 1947. The advantages of the anterior approach for THA are several. First, the hip is an anterior joint, closer to the skin anterior than posterior. Second, the approach follows the anatomic interval between the zones of innervation of the superior and inferior gluteal nerves lateral and the femoral nerve medial. Third, the approach exposes the hip without detachment of muscle from the bone.

The mini-incision variation of this exposure was developed by Joel Matta in 1996. He rethought his approach to THA and his goals were: lower risk of dislocation, enhanced recovery, and increased accuracy of hip prosthesis placement and leg length equality. This approach preserves posterior structures that are important for preventing dislocation while preserving important muscle attachments to the greater trochanter. The lack of disturbance of the gluteus minimus and gluteus medius insertions facilitates gait recovery and rehabilitation, while the posterior rotators and capsule provide active and passive stability and account for immediate stability of the hip and a low risk of dislocation.

A disadvantage of the approach is the fact that a special operating table with traction is required. Potential complications include intraoperative femoral and ankle fractures. These can be avoided through careful manipulation of the limb. If a femoral fracture occurs, the incision can be extended distally by lengthening the skin incision downward along the anterolateral aspect of the thigh, and splitting the interval between the rectus femoris and the vastus lateralis.

The choice of approach used to perform a primary THA remains controversial. The primary goals are pain relief, functional recovery and implant longevity performed with a safe and reproducible approach without complications. The anterior approach is promising in terms of hospital stay and functional recovery. Although recent studies suggest that component placement in minimally invasive surgery is safe and reliable, no long-term results have been published. Further follow-up and development is necessary to compare the results with the posterior approach as most of the positive data is based on comparisons with the anterolateral approach.

The proposed benefits of the anterior approach are not supported by the current available literature. The issues regarding the difficult learning curve, rate of complications, operative time, requirement for trauma tables and image intensifier should be taken into account by surgeons starting with the anterior approach in THA.

Approximately 20% of patients undergoing primary unilateral total knee arthroplasty complain of severe pain in the contralateral knee, and 10% of patients who have had a primary total knee (TKA) undergo contralateral surgery within 1 year.

The number of patients suitable for primary TKA is rising, and so is the need for simultaneous bilateral TKA (BTKA) procedures. The advantages of single-stage BTKA include its low complication rates, high patient satisfaction, and cost-effectiveness. Others believe that there is increased morbidity and mortality.

The goal of reducing the exposure to repeated anaesthesia, total hospitalization and recovery time, and cost, while maintaining patient safety, is a laudable one. Our data suggest that bilateral TKA patients have a lower total operating time, use less pain medication, have a shorter hospital stay and lower overall treatment costs.

The cohort of patients selected for bilateral surgery in our unit is younger and has fewer comorbidities than unilateral controls. They have a high satisfaction rate and no increase in complication or readmission rates. We have seen a higher blood transfusion rate but no increase in cardiac, thromboembolic or septic complications.

The key to BTKA is patient selection and the implementation of efficient care and surgical pathways that includes a thorough pre-assessment, careful education and well-resourced aggressive post-operative physiotherapy. When appropriately applied, the benefits include a shorter overall recovery time and an accelerated return to everyday life and work.

The infected joint arthroplasty continues to be a very challenging problem. Its management remains expensive, and places an increasing burden on health care systems. It also leads to a long and difficult course for the patient, and frequently a suboptimal functional outcome. The choice of a particular treatment program will be influenced by a number of factors. These include the acuteness or chronicity of the infection; the infecting organism(s), its antibiotic sensitivity profile and its ability to manufacture glycocalyx; the health of the patient; the fixation of the prosthesis; the available bone stock; and the particular philosophy and training of the surgeon.

For most patients, antibiotics alone are not an acceptable method of treatment, and surgery is necessary. The standard of care for established infection is two-stage revision with antibiotic loaded cement during the interval period and parental antibiotic therapy for six weeks. Single stage revision may have economic and functional advantages however. We have devised a protocol that dictates the type of revision to be undertaken based on host, organism and local factors.

Our protocol has included single stage revision using antibiotic loaded cement in both THA and TKA. This was only undertaken when sensitive organisms were identified pre-operatively by aspiration and appropriate antibiotics were available to use in cement. Patients with immunocompromise, multiple infecting organisms or recurrent infection were excluded. Patients with extensive bone loss that required allograft reconstruction or where a cementless femoral component was necessary were also excluded.

Our algorithm was validated first in the hip and extended to infected TKA in 2004. This protocol has now been applied in over 100 TKA revisions for infection between 2004 and 2009. Our single stage revision rate is now over 25%. We continue to see a lower reinfection rate in these carefully selected patients, with high rates of infection control and satisfaction and better functional and quality of life scores than our two-stage revision cases.

Whilst our indications are arbitrary and not based on specific biomarkers, we present excellent results for selective single stage exchange. A minimum three-year follow-up suggests that these patients have shorter hospital stays, higher satisfaction rates and better knee scores. An ongoing evaluation is in place. One-stage revision arthroplasty for infection offers potential clinical and economic advantages in selected patients.

The approach to total hip arthroplasty (THA) should allow adequate visualization and access so as to implant in optimal position whilst minimizing muscle injury, maintaining or restoring normal soft tissue anatomy and biomechanics and encouraging a rapid recovery with minimal complications. The direct anterior approach (DAA) for THA was first performed in Paris, by Robert Judet in 1947. This procedure has since been performed consistently by a small group of surgeons and has recently gained great popularity. Access to the hip can be safely performed with one or two assistants. The advantages of the anterior approach for hip arthroplasty are several. First, the hip is an anterior joint, closer to the skin anterior than posterior. Second, the approach follows the anatomic interval between the zones of innervation of the superior and inferior gluteal nerves lateral and the femoral nerve medial. Third, the approach exposes the hip without detachment of muscle from the bone. Care must be taken to avoid cutting the lateral femoral cutaneous nerve which runs over the fascia of the sartorius. The mini-incision variation of this exposure was developed by Joel Matta in 1996. He rethought his approach to hip arthroplasty and by abandoning the posterior approach and adopting the anterior approach his goals were: lower risk of dislocation, enhanced recovery, and increased accuracy of hip prosthesis placement and leg length equality. This approach preserves posterior structures that are important for preventing dislocation while preserving important muscle attachments to the greater trochanter. The lack of disturbance of the gluteus minimus and gluteus medius insertions facilitates gait recovery and rehabilitation while the posterior rotators and capsule provides active and passive stability and accounts for immediate stability of the hip and a low risk of dislocation. Using the anterior approach, patients are allowed to mobilise their hip freely. The gluteus maximus and tensor fascia latae muscles insert on the iliotibial band which joins them and form a ´hip deltoid´. Lack of disturbance of these abductors and pelvic stabilisers is another benefit of the anterior approach and accelerates gait recovery. The lateral femoral cutaneous nerve is at risk when the fascia is incised between the tensor fascia latae and the sartorius muscle. Damaging it may lead to a diminished sensation on the lateral aspect of the thigh and formation of a neuroma.

A disadvantage of the approach is the fact that a special operating table with traction is required. Potential complications include intra-operative femoral and ankle fractures. These can be avoided through careful manipulation of the limb. If a femoral fracture occurs, the incision can be extended distally along the anterolateral aspect of the thigh, and splitting the interval between the rectus femoris and the vastus lateralis. In obese or muscular patients, where visibility is in doubt, an increase of the incision length will give the surgeon the required view.

The choice of approach used to perform a primary THA remains controversial. The primary goal of a hip replacement is pain relief, functional recovery and implant longevity performed with a safe and reproducible approach without complications. The anterior approach is promising in terms of hospital stay and functional recovery. Although recent studies suggest that component placement in minimally invasive surgery is safe and reliable, no long-term results have been published. Further follow-up and development is necessary to compare the results with the posterior approach. The proposed benefits of with the DAA are not supported by the current available literature. The issues regarding the difficult learning curve, rate of complications, operative time, requirement for trauma tables and image intensifier should be taken into account by surgeons starting with the DAA in THA.

The best surgical modality for treating chronic periprosthetic shoulder infections has not been established, with a lack of randomised comparative studies. This systematic review compares the infection eradication rate and functional outcomes after single- or two-stage shoulder exchange arthroplasty, to permanent spacer implant or resection arthroplasty.

Full-text papers and those with an abstract in English published from January 2000 to June 2014, identified through international databases, were reviewed. Those reporting the success rate of infection eradication after a single-stage exchange, two-stage exchange, resection arthroplasty or permanent spacer implant were included, with a minimum follow-up of 6 months and sample size of 5 patients.

Eight original articles reporting the results after resection arthroplasty (n = 83), 6 on single-stage exchange (n = 75), 13 on two-stage exchange (n = 142) and 8 papers on permanent spacer (n = 68) were included.

The average infection eradication rate was 86.7% at a mean follow-up of 39.8 months (SD 20.8) after resection arthroplasty, 94.7% at 46.8 months (SD 17.6) after a single-stage exchange, 90.8% at 37.9 months (SD 12.8) after two-stage exchange, and 95.6% at 31.0 months (SD 9.8) following a permanent spacer implant. The difference was not statistically significant.

Regarding functional outcome, patients treated with single-stage exchange had statistically significant better postoperative Constant scores (mean 51, SD 13) than patients undergoing a two-stage exchange (mean 44, SD 9), resection arthroplasty (mean 32, SD 7) or a permanent spacer implant (mean 31, SD 9) (p=0.029). However, when considering studies comparing pre- and post-operative Constant scores, the difference was not statistically significant.

This systematic review failed to demonstrate a clear difference in infection eradication and functional improvement between all four treatment modalities for established periprosthetic shoulder infection. The relatively low number of patients and the methodological limitations of the studies available point out the need for well designed multi-center trials to further assess the best treatment option of peri-prosthetic shoulder infection.

Total knee replacement (TKR) is considered the “gold standard” treatment for advanced osteoarthritis (OA) of the knee with good survivorship and functional outcomes. However up to 20% of patients undergoing TKR may have unicompartmental disease only. Treatment options for medial compartment arthritis can include both unicompartmental knee replacements (UKR) and TKR. While some surgeons favor TKR with a proven track record, others prefer UKR due to more normal joint kinematics, better proprioception and better motion. There is also a higher rate of return to sports amongst patients with UKR compared to TKR.

When considering all knee procedures, partial knee replacements account for 7–9%, primary TKRs for 83–88%, and revision knee replacements for 5–8%. Unicompartmental Knee Replacements comprise more than 90% of all partial knee replacement procedures. Proponents of UKR cite as advantages the preservation of normal knee kinematics, lower peri-operative morbidity, blood loss and infection risk compared with TKRs, as well as accelerated patient rehabilitation and recovery. However, partial knee replacements have a higher rate of revision than TKRs. This may be partly because they are inserted in patients with higher expectations, and partly because they are easier to revise. As a result, the volume of UKRs implanted has diminished over time and continues to decline.

We compared patient reported outcomes, satisfaction and perception of normality of the knee post-operatively between UKR and TKR. A single unit and single surgeon series of patients were recruited. Data was collated for 68 well-matched patients with more than 24 months follow-up. UKR was undertaken in patients with isolated medial compartment osteoarthritis; stable ACL and less than grade 3 lateral patellar changes of the Outerbridge classification. TKR was undertaken for the rest. The patients were assessed with validated knee scores including the Total Knee Function Questionnaire (TKFQ) which focuses on recreational and sporting outcomes as well as activities of daily living (ADL). Patient satisfaction and perception of knee normality was measured on a visual analogue scale.

Thirty-four patients with a TKR and 34 patients with a UKR were analyzed. The average ages in the TKR and UKR groups were 69.25 and 67.26 years, respectively. The patients were well-matched for demographics and had equivalent pre-operative morbidities and scores. The UKR group had better WOMAC (p=0.003), SF36 (physical: p<0.001 mental: p=0.25), Oxford knee (p<0.001) and Knee Society scores (p=0.002, function: p<0.001). The UKR group showed better outcomes in the TKFQ including exercise and sport (p= 0.02), movement and lifestyle (p=0.02) and the ADL (p=0.002). There was, however, no difference in patient satisfaction scores (p=0.41) and perception of how normal the knee felt between the two groups (p=0.99).

A UKR procedure confers better functional outcome in terms of recreation and sport compared to TKR procedures. While UKR is an appropriate choice in the elderly yet active patient with unicompartmental knee arthritis, satisfaction is similar to that of patients with a TKR reflecting higher pre-operative expectations. We believe that in appropriate centers, the UKR procedure is associated with excellent outcomes. UKR should have a secure place in the knee arthroplasty armamentarium provided current knowledge regarding patient selection and surgical technique is followed.