The first successful total hip replacement which is used in large numbers even today, was developed by John Charnley in the 1960s. Several iterations were necessary before the final design was arrived at, and even then, Charnley was continually striving to improve the design and the technique. Another innovator of the 1960s was George McKee from Norwich in the United Kingdom, whose metal-on-metal bearing in the McKee-Farrar design recently has been revisited because of the exceptionally low wear rate. 79 Muller produced his own metal-on-plastic total hip designs, innovated techniques, and popularized hip replacement throughout Europe. These hip designs, and many other approaches to treating the arthritic hip, have provided the foundation for the hundreds of designs that have been produced.
Predicting and evaluating the long-term performance of new or modified designs of total hip replacements is a challenging subject that is at the forefront of orthopaedic practice today. Fortunately, there is a strong incentive from designers, researchers, surgeons, manufacturers, and regulators, to minimize clinical problems. However, problems still occur, and although many can be attributed to design deficiencies or inadequate preclinical testing, other problems would have been difficult to predict in advance. It is suggested by the author that with an effective use of the knowledge and experience available today, the likelihood of future problems can be minimized, but it will be recognized this is such a complex area that there are no simple answers to ensuring that new hip designs of total hip replacement will be trouble-free.
In the current study, the legacy of successful designs from the past will be discussed first. The limitations of those designs will be examined, which will help to identify the opportunities for improvement. The process of the introduction of new designs that were intended to have one or more advantages will be studied. From this study, useful lessons will be deduced, which will be applied to make proposals for introducing designs with only modest improvements, to those with major differences to anything used previously. If the knowledge and experience available today were used more effectively, the likelihood of future problems could be minimized significantly. In an area where research is seen as the primary responsibility of the industry, immense benefits to patient outcomes could result from an influx of appropriately directed federal funding to the research community. Finally, numerous suggestions are made that point to a more holistic approach to innovative design, with a particular emphasis on a more effective use of the sophisticated design and evaluation tools that are available.
EVALUATION OF INNOVATIONS
The Charnley total hip replacement can be regarded as the reference design because of the experience and numerous published studies. 14,15,20,42,43,50,67,73,80,94,96 However, it must be emphasized that the results vary with the hospital, the surgeon group, the patient population, and which version of the Charnley total hip replacement was used. In a 5-year study of more than 1000 Charnley hip replacements performed across a United Kingdom health region, Fender et al 21 found 7.5% either revised or grossly loose, and reported a 5% dislocation rate. The authors suggested that in general orthopaedic practice, the failure rate is higher than in the specialist centers and may be more representative of the norm. Although the surgeon’s experience is likely to be an important factor, an additional point is that the surgical technique may not be sufficiently reproducible to obtain the ideal component placement and quality of cement-bone interface. Therefore, there are numerous ways in which innovation can be applied to improve the results of total hip replacement.
In writing about innovation, Huiskes 32 stated that for a new design, an explicit statement of purpose is required. Which problem is it trying to solve? The following examples will be grouped according to specific problems. Huiskes also showed that in attempting to solve one particular problem, another problem can be introduced inadvertently. This is apparent in many of the examples.
In the early years of hip replacement, fracture of the stem was a problem, which was solved by the use of improved metal processing, forged CoCr alloys, and Ti alloy. However, in the 1990s, fatigue failures occurred of a well-designed hip stem made from forged CoCr alloy. 93 The fracture occurred through the laser-etched lettering that produced a local stress concentration. Exacerbating factors were age of the patients (average, 61 years), weight of the patients (average, 96 kg), a poor cement mantle, and undersized stems. Even if the final product had been subjected to the International Standards Organization fatigue test as reported by Paul, 58,59 if right hips had been tested, the hips would have passed because the laser-etching only intruded on the anterolateral corner of left stems where the failures initiated. Whether the testing was thorough is unclear but there certainly was an element of unpredictability.
In 1996, Willert et al 91 reported on 28 cemented Ti alloy Muller stems revised for the effects of crevice corrosion. The conclusion was that “titanium alloy can no longer be recommended for cementation, but is as safe as ever for anchorage without cement.” Yet, Ti alloy stems continued to be sold, and in recent years, many stems have been removed because the patients had pain and effusion where severe corrosion of the stem was observed. That is an example of a manufacturer not using or not being aware of data that are readily available in the literature. Although it may be possible to delay or prevent such corrosion by surface treatments, because alternate metals are readily available, the use of Ti alloy with cement would have to be rigorously justified today.
Subsequent to the curing of acrylic cement, heat necrosis of bone has been proposed to result in bone resorption and loosening. 92 To address this problem, in the early 1990s, a new cold-curing cement called Boneloc was introduced in Europe. The new cement was compared in 30 patients with the widely used Palacos cement using the roentgenstereophotogrammetry migration method. 85 Much greater migrations were measured at all interfaces with the Boneloc compared with Palacos. Unfortunately, by that time, hundreds of cases had been done in many hospitals using Boneloc, with a very high rate of loosening after only a few years. A new cement with different mechanical and chemical properties than before has a high level of unpredictability and therefore needs more extensive evaluation. If the roentgenstereophotogrammetric method had been used as part of the design and evaluation process before full product release, widespread failure could have been avoided.
However, a simple design change to a stem should be more predictable. In the next example for improving fixation, an uncemented hip stem was designed in the mid1980s with a high neck cut and a large proximal area (Fig 1). Biomechanical testing showed a high resistance to torsional loading and less interface micromotion. 66 The distal stem was designed to avoid digging in such that thigh pain would be avoided. However, the clinical results were the opposite of what was predicted, with high incidences of migration, loosening and thigh pain. 57 The failure was not caused by the new design features themselves, but were attributable to the stem surface being completely smooth, allowing too much micromotion. This is an example of basing a design on a onedimensional theory, one factor, without sufficient consideration of the other design factors. Since this early version, the smooth stem now is used only with cement, whereas for uncemented use, the proximal stem surface has been roughened and coated with hydroxyapatite, producing greatly improved results.
In cemented hip stems, it would seem to be an advantage if the cement is bonded to the stem, especially because debonding and cement mantle cracking have been associated with failure in retrieved specimens. 36–38 However, in a study of two versions of the Iowa total hip replacement, a semismooth surface showed 6% revision and a rough surface showed 18% revision. 81 The explanation was that if debonding occurs with a rough surface, cement abrasion with particle release occurs. It was said that in this situation, a smooth surface is preferable. The issue of smooth versus rough is multifactorial, and in some clinical series, rough stems have performed well. From a materials science perspective, the debonding process with rough and precoated stems is complex. 53 However, the studies show that any system must work well under all conditions of cement technique, femoral anatomy, and patient function. This is the principle of ‘Robust Design’, which will be mentioned later.
For durable uncemented fixation, several authors 33,75 have taken the approach of achieving long-lasting fixation and making the procedure easier and more reproducible, by using uncemented grit-blasted stems that fit tightly distally and proximally (Fig 2). Retrievals have shown solid osseointegration against the corners of the implant. 49,97 The 5-to 9-year survivorship has been reported to be 98% for the stem. 17 A similar design in terms of overall shape, the proximally sinter-coated Taperloc has a similar survivorship, even in younger patients. 56 The high level of survivorship against loosening in the porous coated AML design, although at the expense of some proximal bone loss, is documented. 19
Another successful approach to uncemented fixation has been hydroxyapatite coating. The Omnifit design has a remarkable 10-year survivorship of the stem with no radiographic evidence of migration or interface bone resorption. 25 Another notable design is the Corail, which has grooves and hydroxyapatite coating over the entire stem. In several series, a head and socket combination was used that produced an accelerated polyethylene wear rate. 76,77 Because of the apparent design deficiencies, the socket is no longer used. However, despite the wear particles, there was 98.9% survivorship of the stem at 8 years and a complete absence of femoral osteolysis. It was proposed that extensive hydroxyapatite coating seals the entire interface and blocks particles.
From the above data, namely a high survivorship of several different uncemented designs, each with many studies conducted in different centers, it seems that uncemented hip stems may be more durable than cemented stems. This possibility should be given serious study because if true, more widespread use of uncemented designs could reduce the number of failures in the future.
Proximal Bone Resorption
Preventing proximal medial bone resorption has been seen by many to be an important problem because it could lead to overall stem loosening and fracture of the stem; it also compromises a revision procedure. The evidence is that the resorption is caused by the distal half of the stem carrying a high proportion of the frontal plane bending moment, stress-protecting the proximal bone. The Butel is an ingenious concept, which reduces stem stiffness by a split stem (Fig 3), and would be expected to reduce the stress protection in the bone. The short-term results were promising. 13 However, at longer followup, the failure rate was no less than 43%. 35 Again, this represents a design based on a unidimensional concept, where other features of the design were unsatisfactory (in this case the smooth surface) allowing excessive interface micromotion caused by the flexibility of the stem. Even in the early 1990s, the unsatisfactory performance of relatively smooth-surfaced uncemented stems in various forms was not generally recognized.
The Isoelastic hip replacement was another design with increased flexibility that suffered a similar fate (Fig 4). At 28 months followup, Andrew et al 3 reported that “these early results are encouraging, but the effectiveness must be determined by longer term follow-up.” At 8 years followup, there was 10% rate of revision for loosening. An additional 30% were radiographically loose. 65 The authors stated “The problem seems to be insufficient primary fixation of the proximal stem. Stress-shielding, micromotion, and polyethylene fragments, prevent the anticipated bony ingrowth.” This was an example of widely introducing a total hip replacement much too early when the design was dramatically different from anything that was manufactured previously, even including an unbonded smooth polymer surface interfaced with bone.
With the same goal of reducing bone remodeling by a reduced stem stiffness, the EPOCH hip was designed in the early 1990s as reported by Bobyn et al. 9 The surface was coated with fiber metal to minimize interface micromotion (Fig 5). To date, bone remodeling has not been evident, and in addition, pain has been almost entirely absent. This type of stem would have particular advantages in patients with large canal diameters. A well-controlled trial is ongoing in several centers to obtain data at longer followup before a decision is made on its expanded use.
Increased Loosening in Young Patients
An important question today is whether the same results seen in an older population are seen in younger more active patients. There are conflicting results, with cemented failure rates in younger patients reported at between 0% and 49% and uncemented failure rates reported at between 0% and 13% although the different followup times should be noted 85 (Table 1). In recent reports 14,18,45,56,64,80,82 long-term results in cemented stems are not encouraging, which suggests that uncemented results may be better. The data from the Swedish Joint Registry 26,55 support the view that the younger patients have worse results: “Aseptic loosening still constitutes the major problem. Specific patient cohorts have increased failure risks, especially younger patients.”
To design the ultimate total hip replacement that potentially could optimize results in all patients, a remarkable system called the Identifit was developed. 60 At surgery, a mold is taken of the femoral cavity. The mold is taken to an on-site facility where it is laser scanned. A custom total hip replacement then is made by computer numerical control machining in only 45 minutes, during which time the surgeon installs the socket. The latest development is that hydroxyapatite coating can be applied and it is expected that it will greatly enhance the long-term results, compared with the disappointing early results using an uncoated stem. 74
The experience of custom hip replacements in the Biomedical Engineering Department at the Royal National Orthopaedic Hospital, in Stanmore, England, has been positive so far. The service was set up 10 years ago, when the stems were designed from radiographs using proprietary software. 30,89 The total hip replacements have been used mainly in three centers and the numbers of surgical procedures have increased steadily as confidence in the system has increased. The study plan included preclinical design synthesis and analysis, consisting of stem-bone micromotion measurement, 28 photoelastic studies of bone strains, 29 and finite element analysis to predict the effect of various stem design features. 86,87 The finite element analysis was used to compare a lateral flare design with a straight stem design, which was unbonded for this comparison 86 (Fig 6). The lateral flare design transmitted the forces predominantly in the proximal region. In the straight stem, most of the load was transmitted more distally and the stem-bone interface pressures were much higher than for the lateral flare design. The lateral flare, described previously by Fetto et al 22 was preferable for more favorable bone stresses and axial stem stability.
The three basic stems for primary hip replacement and for revision hip replacement use a combination of a lateral flare, a collar, proximal grooves, hydroxyapatite coating, and distal cutting flutes (Fig 7). The particular stem features used in each case were based on the status of the bone. The ideal situation is that the lateral and medial flares interface with the cortical bone and a shorter stem can be used with this type of design in primary hip replacement. Radiographs show that trabecular attachment occurs onto the lateral flare and even can increase in density with time. In many revision situations, 87 not only is there loss of proximal bone, but osteolytic lesions around the distal tip can lead to bone fracture. An extended long-stem with cutting flutes has been effective in treating such cases. Such flutes add torsional stability to the composite structure. 90
The study plan included postclinical studies to test the basic hypotheses of the stem design. Axial migration was measured when cemented Charnley total hip replacements and Stanmore total hip replacements were used for comparison. 88 In 2 years, the computer assisted design-computer assisted manufacturing (CAD-CAM) primary hip replacements migrated less, whereas the migration for the CAD-CAM manufacturing revision hip replacements was similar to that of the primary cemented hip replacements. Data at 4 years showed minimal additional migration. To test for proximal bone density changes, dual-energy xray absorptiometry scanning was used, which showed complete preservation of proximal-medial bone up to 4 years followup. All of these data showed that the clinical results are so far consistent with that predicted by the preclinical design synthesis and analysis.
Using the CAD-CAM system, more than 1500 total hip replacements have been done in a 9-year period, in the ratio of approximately three primary surgeries to two revision surgeries. Only 1.3% of the stems have been revised. There were seven fractured stems (0.5%) caused by underestimating the minimum allowable distal stem diameter in patients undergoing revision surgery in whom proximal bone support was insufficient. Histologic examination of sections of some of these removals revealed intimate attachment of new bone to the hydroxyapatite surface, including in and around the grooves and flutes. There were only six failures caused by excessive subsidence and loosening, mostly in the early cases when the designs and techniques still were being developed, The other failures were caused by infection, a fractured femur, and dislocation. It was concluded that these results support continued use of the system clinically, because the results are likely to be superior to what could be expected with an off-the-shelf system, particularly in patients with abnormal geometry or with severe bone loss and deformity who are undergoing revision surgery.
Excessive Length of Conventional Stems
In the very earliest attempts at hip replacement, stems that engaged only the proximal part of the femur were used. This approach has come to be known as conservative hip replacement. The concept would involve less destruction of bone should failure occur, which is particularly valuable in younger patients. One recent example is the Thrust Plate design developed by Huggler and Jacob in 1980. 31 The original design was based on finite element analysis and mechanical testing. Two design iterations were introduced during the first 9 years of experience, followed by a 10-year followup of the most recent design. 4 Although the survivorship was 80% at 10 years, it was considered that the advantages merited its use in selected patients. This seems to be a valid method of developing a new design, such that an improved version of the device is developed in early trials, and the final design is subjected to adequate followup.
The Mayo hip replacement is another example of a conservative total hip replacement, which was developed by Morrey 59 (Fig 8). Clinical trials were initiated at one institution. 1 In a group of patients who were an average of 51 years of age, the survivorship for loosening was 98.2% at 10 years. Pain relief was achieved in more than 90% of patients. Because the canal was not required to be reamed, there was less blood loss compared with the amount of blood loss when conventional total hip replacement is done. Based on these data, the study now is being expanded to determine whether the results of the originator can be reproduced in other centers.
Higher Loosening Rate of Sockets than With Stems
Sockets have received less attention than stems in terms of design and analysis even though the loosening rates of cemented sockets usually have been higher than those for the stems in medium to long-term followup. 11,50,94 The Mecron screw-ring was introduced as a new cementless approach to socket fixation. Clinical followup at 5 to 9 years showed 33% of the prostheses were revised or awaiting revision. Eighty percent of all total hip replacements showed migration on radiographs. 16 However, the component was introduced to the market well before that clinical report was published. The authors stated: “The use of this prosthesis represents an example of a new design being used in a widespread uncontrolled fashion before the publication of adequate follow-up studies.” Other sockets with metal-backing have been designed without adequate attention given to the locking mechanism for the plastic, and the clearances between the plastic and the metal, resulting in failure modes such as splitting of the plastic insert. Finite element analysis subsequently was used to predict such overall mechanical effects especially with rimmed sockets, also showing the potential effect on the wear of the plastic. 46,47 These are examples of introductions of products without adequate preclinical or postclinical testing and analysis.
A success story in socket design in the past 15 years has been the Harris-Galante uncemented socket with a fiber-metal coated metal backing. 24 Even in young patients who averaged 52 years of age, the survivorship for fixation at 11 years was 97.7%. However, polyethylene wear averaged 0.15 mm per year, causing an incidence of osteolysis around stems and sockets. In the latter, osteolysis occurred around screws, possibly attributable to particles from back-side wear. The wear rate and the incidence of osteolysis was higher in the younger patients. Although with the use of newer plastics the wear rate can be reduced substantially, the holed versions of cups and screws only should be used when absolutely necessary.
It may be possible to increase the percentage of bone ingrowth into a porous surface by a change in the materials and structure. For example, a porous tantalum structure, in which the individual struts are roughened, has shown evidence of this 10 (Fig 9). Based on a canine model, the authors concluded: “No adverse response of any type was noted in the histological analyses, and bone of normal appearance contacted the tantalum struts. From the perspectives of bone ingrowth and biological fixation, the data from this study support the further research and development of porous tantalum for clinical research.”9 The construction of an acetabular component using this material is interesting in that the polyethylene is molded into the porous structure, eliminating the possibility of back-side wear, and preventing particles migrating into the metal-bone interface. Additional mechanical testing and controlled clinical evaluations are planned as the next stage in the process.
Polyethylene wear undoubtedly has been the major long-term problem of total hip replacement. In Europe, this has been addressed with ceramic-on-polyethylene or ceramic-on-ceramic. Unfortunately, occasional poor results such as using a type of Zr that was evidently sensitive to autoclaving, 2 and fracture of alumina femoral heads in one series, 27 has limited the benefits of using ceramic bearings in patients in the United States. A survey by one ceramic manufacturer making Biolox alumina femoral heads showed minimal fracture problems. 27 It was stated that “this wait and see attitude is absolutely unacceptable in today’s world.”27 It may be that in the United States, insufficient use has been made of ceramics technology in specific indications. 39 In recent years, there has been a revival of metal-on-metal in Europe and the United States, not without justification.
An intriguing approach to avoiding wear has been investigated in the United Kingdom. This is a cushion-form bearing using a deformable polyurethane liner, which generates fluid film lubrication. 8 “The composite cushion hips produced negligible wear compared to . . . conventional UHMWPE cups. This study has demonstrated for the first time the beneficial effects of fluid film lubrication.”8 But even after extensive research, the problem of the interface between the polyurethane and the metal shell still remains.
There is, however, an overriding question today: whether the new cross-linked polyethylenes recently developed 48,54,55,61 will reduce clinical wear to such a low level that metal or ceramic femoral heads in such polyethylene will solve the wear problem altogether. The data using hip simulating machines, 36 and clinical data of early types of cross-linked ultrahigh molecular weight polyethylene 68,96 would support this view. For each particular version, staged introduction and specific studies using roentgenstereophotogrammetry and standard evaluation methods are expected in the coming years.
Nonreproducibility of Technique and Surgical Error
The study of the Swedish Joint Registry 51 showed major variations in results and techniques between different centers, which led to technique improvements. “The fact that we have information about surgical technique from every Department for almost two decades has had a profound effect on technique improvement in Sweden.”51 This relates back to another study, 21 in which results of patients who were treated in general orthopaedic centers could be much worse than those treated in specialist centers. Robots guided by computers are widely used in industry for achieving precision and reproducibility. Is it not logical to propose that such technology has a major place in orthopaedics? In Europe, instrument navigation systems are widely used in spinal surgery and new applications are imminent. Approximately 10 years ago in the United States, Paul et al 70 initiated the development of the Robodoc system for inserting hip stems. This system, and a European lookalike, now are used in many centers in Europe (Fig 10). To date, showing major benefits over conventional methods has been controversial, other than achieving accurate placement in the position of choice, which might explain the relatively low level of support for robotic surgery in the United States. However, this may be too cautious because the use of robotic or computer-assisted surgery, such as by the use of navigation systems, is likely to provide important advantages in the long term.
Having now considered these representative examples of innovation, answers can be proposed to two practical questions.
The first key question is how to determine whether a particular implant will produce the best possible clinical results for routine use. In this context, certain traditional designs such as the Charnley, the Exeter, and the Lubinus already are considered acceptable based on their long-term followup. However, long-term followup is not necessarily definitive because many designs have been modified since or during the studies, and the variables between followup studies usually are different. 51 There must be a mechanism for accepting other designs that are logical extensions of existing systems, or variants with likely benefits. The above examples of innovation, and many others, show that even simple changes do not necessarily lead to a trouble-free device. This is often because the input to the design was too limited, without all factors being taken into consideration. In contrast, the vast experience and knowledge of the orthopaedic community today could be effectively applied to the design and evaluation of these basic designs for routine use. The proposal is that an authoritative group of individuals formulate a specification of the design features of total hip replacement that produces successful results, for routine use particularly in older patients. Cemented and uncemented types of total hip replacements must be addressed. Some specific designs can be mentioned as examples, but not with the intention of a restriction to those designs only. As reported by Murray 62 this approach already has a precedence in the Concensus meetings sponsored by the National Institutes of Health and the American Academy of Orthopaedic Surgeons on various subjects.
The second key question is how to be reasonably certain that a device with significantly new features, materials, or techniques, is better and whether and when it should be used. This question applies to devices for routine use and for specific applications such as in younger patients and revision hip replacement. A key requirement is a staged introduction appropriate to the magnitude of the unpredictability of the new device, and the potential benefit that can be gained by their use. In terms of the design, has it been subjected to a sufficiently extensive preclinical and postclinical testing regime? To address this question, the principles of Taguchi can be applied. 84 Taguchi proposed that designs must be functionally robust, such that the effect of input variation on performance should be minimized. Hips should function well regardless of the surgical technique that was used or use by the patient. This means that the evaluation should use a matrix of tests and test conditions, and that this evaluation should have been used to optimize the design. Too frequently, a limited number of tests are done on a hip that already has been designed. It is emphasized that total hip replacements (or other implants) should be designed using the design process, and not based solely on a unidimensional concept, whether tested or not, or on a limited number of scientific experiments. The design process is a systematic process for the formulation and evaluation of a design, and includes all aspects and working conditions of the intended design. The proposal is that federal funding should be provided for the specification and development of extensively researched and validated test methods by the research community.
Evolution of the System for Introducing New Devices
It is important that there is an efficient mechanism for introducing new devices with sound principles that are similar to previous devices, but which may have some simple but important improvements: such devices are covered by the Food and Drug Administration under the 510K route. 23 Even then, there must be a cautionary introduction to ensure that nothing unexpected occurs. But when is a change a change? According to Food and Drug Administration guidelines, 23 an important criterion is that “the change does not alter the fundamental scientific technology of the device.” If the innovation is clearly different such that even after a range of preclinical evaluations, the clinical performance still is unpredictable, a more staged introduction is required. The Food and Drug Administration has guidance documents for such a case, and a compendium of individual test methods that can be appropriately applied to each new device. These methods include International Standards Organization 34 and the American Society for Testing and Materials 4 standards, and literature reports, thus using data over a wide spectrum of experience. The involvement of manufacturing companies in this process of developing test methods is essential.
In Europe, the regulatory system is more general, having three tiers of performance standards, notably general requirements for all implants, particular requirements for families (total joint replacements) and specific requirements for types (total hip replacements). Emphasis is placed on performance standards with pass and fail criteria, which clearly is an ideal method for specifying a test. In the case of total hip replacement, the only specific requirements stated are surface finish and sphericity, fatigue testing, and wear testing, all based on established standards of the International Standards Organization. The drawback is that the standards of the International Standards Organization take many years to develop under the present system, 71 and an extension to other sources of test methods seems warranted.
There are numerous test methods available for a wide range of design aspects, and advances already have been made in developing more sophisticated and reliable test methods. 12 For example, Stolk et al 82 have shown that in terms of predicting the strains in a hip stem cemented into a plastic test bone, finite element analysis can produce results that are very close to the results of physical tests. This would point toward an finite element analysis model as an initial design and evaluation method. Bergmann et al 6,7 have produced an invaluable compendium of data of the forces in the hip in a range of activities based on data from patients with telemetrized total hip replacements. Paul 72 produced a broad survey for all of the joints of the lower extremity. Such data can be used for comprehensive laboratory analyses and tests. Paravic et al 69 reported on many different tests for evaluating fit, strength, and stresses in the bone-stem system. Crosbie from the Department of Health in the United Kingdom is advocating roentgenstereophotogrammetry to measure implant migration as an early predictor of implant performance, particularly relevant to assessing the interface behavior (unpublished data: Crosbie GA: Work performed at the Medical Devices Agency, United Kingdom Department of Health, London, United Kingdom, 1998). This proposal is well supported by numerous studies conducted by Scandinavian researchers. 40,41,44,52,58,67,78
Total joint replacements provide major benefits to large numbers of patients at relatively little cost. Additional substantial improvements still can be made in outcome and performance by modest improvements, or by more innovative approaches including the combination of artificial and tissue engineered materials. To achieve these advances, an appropriate allocation of federal funding is needed for studies in this area, on subjects including design methods, biomechanical testing, biomaterials, and surgical technique. In the past, such research often has been regarded as the responsibility of industry. However, the benefits are more general and can involve complex scientific methodologies implying that a broader spectrum of funding sources should be available. There is an ever-increasing volume of lectures and articles, on the subject of hip replacement, such that a clear and informed view often is difficult to formulate. Conferences often are not sufficiently informative because of the increasing pressure for more presentations, which leads to less discussion. However, meetings that bring together a group of experts from various disciplines, to discuss a particular subject in depth, are very effective. A example is the National Institutes of Health consensus meeting on total hip replacement, which was held in 1994. 62 As noted earlier, there are no simple answers, but there are signs that improvements by innovation are steadily being made and that such innovations are being introduced more carefully than in the past.
The author thanks the Hip Society for the invitation to present this paper. The author thanks the following people for providing information, slides, and discussion: Alan Ashby, Georg Bergmann, Roy Crowninshield, Andy Crosbie, John Fisher, Jorge Galante, Rudolf Geesink, Paul Gregg, William Harris, Rik Huiskes, Hilaire Jacob, Johann Kärrholm, Michael Mulier, John Paul, Robert Poss, Ted Wendt, and Karl Zweymuller.
1. Adams RA, Kessler M, Morrey BF: A conservative femoral replacement for total hip arthroplasty: A prospective study. J Bone Joint Surg In Press 2000.
2. Allain J, Le Mouel S, Goutallier D, et al: Poor eight-year survival of cemented zirconia-polyethylene total hip replacement. J Bone Joint Surg 81B:835–842, 1999.
3. Andrew TA, Flanagan JP, Gerundini M, et al: The Isoelastic, noncemented total hip arthroplasty. Preliminary experience with 400 cases. Clin Orthop 206:127–138, 1986.
4. American Society for Testing and Materials: Standards of the American Society for Testing and Materials. West Conshohocken, PA, http://www.astm.org.
5. Bereiter H, Huggler AH, Jacob HAC: Development, experience and follow-up with the thrust plate. Proceedings of the European Federation of National Asociation of Orthopaedics and Traumatology, Brussels, 75, 1999.
6. Bergmann G, Graichen F, Rohlmann A: Hip joint loading during walking and running, measured in two patients. J Biomech 26:969–990, 1993.
7. Bergmann G, Graichen F, Rohlmann A: Is staircase walking a risk for the fixation of hip implants? J Biomech 28:535–553, 1995.
8. Bigsby RJA, Anger DD, Jin ZM, et al: A comparative tribological study of the wear of composite cushion cups in a physiological hip joint simulator. J Biomech 31:363–369, 1998.
9. Bobyn JD, Mortimer ES, Glassman AH, et al: Producing and avoiding stress shielding. Clin Orthop 274:79–96, 1992.
10. Bobyn JD, Toh K-K, Hacking A, et al: Tissue response to porous tantalum acetabular cups. J Arthroplasty 14:347–354, 1999.
11. Bourne RB, Rorabeck CH, Skutek, et al: The Harris Design-2 total hip replacement fixed with so-called second-generation cementing techniques. J Bone Joint Surg 80A:1775–1780, 1998.
12. Browne M, Langley RS, Gregson PJ: Reliability theory for load bearing biomedical implants. Biomaterials 20:1285–1292, 1999.
13. Butel J, Robb JE: The isoelastic hip prosthesis followed for 5 years. Acta Orthop Scand 59:258–262, 1988.
14. Callaghan JJ, Forest EE, Olejniczak JP, et al: Charnley total hip arthroplasty in patients less than fifty years old. J Bone Joint Surg 80A:704–714, 1998.
15. Charnley J: The long-term results of low-friction arthroplasty of the hip performed as a primary intervention. J Bone Joint Surg 54B:61–71, 1973.
16. Chell J, Howard PW: Migration and failure of the Mecron screw-in acetabular prosthesis. J Arthroplasty 13:638–641, 1998.
17. Delaunay CP, Kapandji AI: Primary total hip arthroplasty with the Karl Zweymuller first-generation cementless prosthesis. A 5–9 year retrospective study. J Arthroplasty 11:643–652, 1996.
18. Emery DFG, Clarke HJ, Grover ML: Stanmore total hip replacement in younger patients. J Bone Joint Surg 79B:240–246, 1997.
19. Engh Jr CA, Culpepper WJ, Engh WJ: Long-term results of use of the AML in total hip arthroplasty. J Bone Joint Surg 79A:177–184, 1997.
20. Feller JA, Kay PR, Hodgkinson JP, et al: Activity and socket wear in the charnley low-friction arthroplasty. J Arthroplasty 9:341–345, 1994.
21. Fender D, Harper WM, Gregg PJ: Outcome of Charnley total hip replacement across a single health region in England. J Bone Joint Surg 81B:577–581, 1999.
22. Fetto JF, Bettinger P, Austin K: Re-examination of hip biomechanics during unilateral stance. Am J Orthop 605–612, 1995.
23. Food and Drug Administration: The New 510(k) Paradigm. Alternative Approaches to Demonstrating Substantial Equivalence in Premarket Notifications. Rockville, MD, United States Department of Health and Human Services, Food and Drug Administration. Center for Devices and Radiological Health, March, 1998.
24. Galante JO: Cementless acetabular fixation. Proceedings of the International Society for Technology in Arthroplasty, Chicago 25, 1999.
25. Geesink RGT, Hoefnagels NHM: Six-year results of hydroxyapatite-coated total hip replacement. J Bone Joint Surg 77B:534–547, 1995.
26. Herberts P, Malchau H: How outcome studies have changed total hip arthroplasty practices in Sweden. Clin Orthop 344:44–60, 1997.
27. Heros RJ, Willmann G: Ceramics in total hip arthroplasty: History, mechanical properties, clinical results, and current manufacturing state of the art. Semin Arthroplasty 9:114–122, 1998.
28. Hua J, Walker PS: Relative motion of hip stems under load: An in vitro study of symmetrical, asymmetrical and custom asymmetrical designs. J Bone Joint Surg 76A:95–103, 1994.
29. Hua J, Walker PS: Closeness of fit of uncemented stems improves the strain distribution in the femur. J Orthop Res 13:339–346, 1995.
30. Hua J, Walker PS, Muirhead-Allwood SK, et al: The rationale for CAD-CAM uncemented custom hips: An interim assessment. Hip Int 5:52–62, 1995.
31. Huggler A, Jacob H: A new approach towards hip prosthesis design. Arch Orthop Trauma Surg 97:141–144, 1980.
32. Huiskes R: Failed innovation in total hip replacement. Acta Orthop Scand 64:699–716, 1993.
33. Huo MH, Martin RP, Zatorski LE, et al: Total hip arthroplasty using the Zweymuller stem implanted without cement. J Arthroplasty 10:793–799, 1995.
34. International Standards Organization for Standardization: Standards of the International Standards Organization for Standardization. Geneva, Switzerland. http://www.iso.ch.
35. Jacobsson S-A, Djerf K, Gillquist J, et al: A prospective comparison of Butel and PCA hip arthroplasty. J Bone Joint Surg 75B:624–629, 1993.
36. Jasty M, Bragdon CR, O’Connor DO, et al: Marked improvement in the wear resistance of a new form of UHMWPE in a physiologic simulator. Trans Orthop Res Soc 22:785, 1997.
37. Jasty M, Goetz DD, Bragdon CR, et al: Wear of polyethylene acetabular components in total hip arthroplasty. J Bone Joint Surg 79A:349–358, 1997.
38. Jasty M, Maloney WJ, Bragdon CR, et al: The initiation of failure in cemented femoral components of hip arthroplasties. J Bone Joint Surg 73B:551–558, 1991.
39. Jazrawi LM, Bogner E, Valle CJD, et al: Wear rates of ceramic-on-ceramic bearing surfaces in total hip implants. J Arthroplasty 14:781–787, 1999.
40. Karrholm J, Borssen B, Lowenhielm G, et al: Does early micromotion of femoral stem prostheses matter? J Bone Joint Surg 76B:912–917, 1994.
41. Karrholm J, Malchau H, Snorrason F, et al: Micromotion of femoral stems in total hip arthroplasty. J Bone Joint Surg 76A:1692–1705, 1994.
42. Kavanagh BF, Dewitz MA, Ilstrup DM, et al: Charnley total hip arthroplasty with cement. J Bone Joint Surg 71A:1496–1503, 1989.
43. Kavanagh BF, Wallrichs S, Dewitz M, et al: Charnley low-friction arthroplasty of the hip. J Arthroplasty 9:229–234, 1994.
44. Kiss J, Murray DW, Turner-Smith AR, et al: Migration of cemented femoral components after THR. J Bone Joint Surg 78B:796–801, 1996.
45. Kronick JL, Barba ML, Paprosky WG: Extensively coated femoral components in young patients. Clin Orthop 344:263–274, 1997.
46. Kurtz SM, Bartel DL, Edidin AA: The effect of gaps on contact stress and relative motion in a metal-backed acetabular component. Trans Orthop Res Soc 19:243, 1994.
47. Kurtz SM, Gabriel SM, Bartel DL: The effect of non-conformity between metal-backing and polyethylene inserts in acetabular components for total hip arthroplasty. Trans Orthop Res Soc 18:434, 1993.
48. Kurtz SM, Muratoglu OK, Evans M, et al: Advances in the processing, sterilization, and crosslinking of ultra-high molecular weight polyethylene for total joint arthroplasty. Biomaterials 20:1659–1688, 1999.
49. Lintner F, Zweymuller K, Brand G: Tissue reactions to titanium endoprostheses: Autopsy studies in four cases. J Arthroplasty 1:183–195, 1986.
50. Madey SM, Callaghan JJ, Olejniczak JP, et al: Charnley total hip arthroplasty with use of improved techniques of cementing. J Bone Joint Surg 79A:53–64, 1997.
51. Malchau H, Herberts P: Prognosis of total hip replacement (The Swedish Joint Registry). http://www.jru.orthop
52. Malchau H, Karrholm J, Wang YX, et al: Accuracy of migration analysis in hip arthroplasty. Acta Orthop Scand 66:418–424, 1995.
53. Mann KA, Bhashyam S: Mixed-mode fracture toughness of the cobalt-chromium alloy/polymethylmethacrylate cement interface. J Orthop Res 17:321–328, 1999.
54. McKellop HA, DiMaio WG, Shen F-W, et al: Wear of gamma radiation crosslinked PE acetabular cups after aging and against roughened removed heads. Trans Orthop Res Soc 24:76, 1999.
55. McKellop HA, Shen F-W, Lu B, Campbell P, et al: Development of an extremely wear-resistant ultra high molecular weight polyethylene for total hip replacements. J Orthop Res 17:157–167, 1999.
56. McLaughlin JR, Lee KR: Plasma spray porous-coated total hip arthroplasty: A 13-year survivorship analysis in patients age 50 and under. Proceedings of the American Academy of Orthopaedic Surgeons 143, 1997.
57. Menon DK, McCreath SW: 5-to 8-year results of the Freeman press-fit hip arthroplasty without HA coating. J Arthroplasty 14:581–588, 1999.
58. Mjoberg B, Selvik G, Hawsson LI, et al: Mechanical loosening of total hip prostheses: A radiographic and roentgen stereophotogrammetric study. J Bone Joint Surg 68B:770–773, 1986.
59. Morrey BF: Short-stemmed uncemented femoral component for primary hip arthroplasty. Clin Orthop 249:169–175, 1989.
60. Mulier JC, Mulier M, Brady LP, et al: A new system to produce intraoperatively custom femoral prostheses from measurements taken during the surgical procedure. Clin Orthop 249:97–112, 1989.
61. Muratoglu OK, Bragdon CR, O’Connor DO, et al: A comparison of 5 different types of highly crosslinked UHMWPEs: Physical properties and wear behavior. Trans Orthop Res Soc 24:77, 1999.
62. Murray DG, (ed): Total Hip Replacement. National Institutes of Health Consensus Statement 12. Bethesda, MD, National Institutes of Health 1–31, 1994.
63. Murray DW, Carr AJ, Bulstrode CJ: Which primary total hip replacement? J Bone Joint Surg 77B:520–527, 1995.
64. Neumann L, Freund KG, Sorensen KH: Total hip arthroplasty with the Charnley prosthesis in patients fifty-five years old and less. Fifteen to twenty one year results. J Bone Joint Surg 78A:73–79, 1996.
65. Niinimaki T, Puranen J, Jalovaara P: Total hip arthroplasty using isoelastic femoral stems. J Bone Joint Surg 76B:413–418, 1994.
66. Nunn D, Freeman MAR, Tanner KE, et al: Torsional stability of the femoral component of hip arthroplasty. J Bone Joint Surg 71B:452–455, 1989.
67. Onsten I, Akesson K, Besjakov J, et al: Migration of the Charnley stem in rheumatoid arthritis and osteoarthritis. J Bone Joint Surg 77B:18–22, 1995.
68. Oonishi H, Saito M, Kadoya Y: Wear of high-dose gamma irradiated polyethylene in total joint replacement—Long term radiological evaluation. Trans Orthop Res Soc 22:97, 1998.
69. Paravic V, Noble PC, McCarthy JC: The fit of cementless stems in the femur: Robot versus surgeon. Trans Orthop Res Soc 23:418, 1998.
70. Paul HA, Bargar WL, Mittelstadt B, et al: Development of a surgical robot for cementless total hip arthroplasty. Clin Orthop 285:57–66, 1992.
71. Paul JP: Development of standards for orthopaedic implants. Proc Inst Mech Eng [H] 211:119–126, 1997.
72. Paul JP: Strength requirements for internal and external prostheses. J Biomech 32:381–393, 1999.
73. Raut VV, Siney PD, Wroblewski BM: Cemented revision Charnley low-friction arthroplasty in patients with rheumatoid arthritis. J Bone Joint Surg 76B:909–917, 1994.
74. Robinson RP, Clark JE: Uncemented press-fit total hip arthroplasty using the Identifit custom-moulding technique. J Arthroplasty 11:247–254, 1996.
75. Robinson RP, Deysine GR, Green TM: Uncemented total hip arthroplasty using the CLS Stem. J Arthroplasty 11:286–292, 1996.
76. Rokkum M, Brandt M, Bye K, et al: Polyethylene wear, osteolysis and acetabular loosening with an HA-coated hip prosthesis. J Bone Joint Surg 81B:582–589, 1999.
77. Rokkum M, Reigstad A: Total hip replacement with an entirely hydroxyapatite-coated prosthesis: 5 years follow-up of 94 consecutive hips. J Arthroplasty 14:689–700, 1999.
78. Ryd L, Linder L: On the correlation between micromotion and histology of the bone-cement interface. J Arthroplasty 4:303–309, 1989.
79. Schmalzreid TP, Peters PC, Maurer BT, et al: Long-duration metal-on-metal total hip arthroplastics with low wear of the articulating surfaces. J Arthroplasty 11:322–331, 1996.
80. Sochart DH, Porter ML: Long-term results of cemented Charnley low-friction arthroplasty in patients aged less than 30 years. J Arthroplasty 13:123–131, 1998.
81. Sporer SM, Callaghan JJ, Olejniczak JP, et al: The effects of surface roughness and polymethylmethacrylate precoating on the radiographic and clinical results of the Iowa hip prosthesis. J Bone Joint Surg 81A:481–492, 1999.
82. Stolk J, Verdonschott NB, Huiskes R, et al: Strains in a composite hip joint reconstruction obtained through FEA and experiments correspond closely. Proceedings of the 12th Annual International Symposium for Technology in Arthroplasty, Chicago 1999.
83. Sullivan PM, MacKenzie JR, Callaghan JJ, et al: Total hip arthroplasty with cement in patients who are less than fifty years old. J Bone Joint Surg 76A:863–869, 1994.
84. Taguchi G: On Robust Technology Development. American Society of Mechanical Engineers, New York, 1993.
85. Thanner J, Freij-Larsson C, Karrholm J, et al: Evaluation of Boneloc. Acta Orthop Scand 66:207–214, 1995.
86. Walker PS, Culligan SG, Hua J, et al: The effect of a lateral flare feature on uncemented hip stems. Hip Int 9:71–80, 1999.
87. Walker PS, Culligan SG, Hua J, et al: Stability and bone preservation in custom designed revision hip stems. Clin Orthop 372:1–10, 2000.
88. Walker PS, Mai SF, Cobb AG, et al: Prediction of clinical outcome of THR from migration measurements on standard radiographs. J Bone Joint Surg 77B:705–714, 1995.
89. Walker PS, Robertson DD: Design and fabrication of cementless hip stems. Clin Orthop 235:25–34, 1988.
90. Whiteside LA, Arima J, White SE, et al: Fixation of the modular total hip femoral component in cementless total hip arthroplasty. Clin Orthop 298:184–190, 1994.
91. Willert H-G, Broback L-G, Buchhorn GH, et al: Crevice corrosion of cemented titanium alloy stems in total hip replacements. Clin Orthop 333:51–75, 1996.
92. Willert HG, Ludwig J, Semlitsch M: Reaction of bone to methylmethacrylate after hip arthroplasty. J Bone Joint Surg 56A:1368–1382, 1974.
93. Woolson ST, Millbauer JP, Bobyn JD, et al: Fatigue fracture of a forged cobalt-chromium-molybdenum femoral component inserted with cement. J Bone Joint Surg 79A:1842–1848, 1998.
94. Wroblewski BM, Fleming PA, Siney PD: Charnley low-frictional torque arthroplasty of the hip. J Bone Joint Surg 81B:427–430, 1999.
95. Wroblewski BM, Siney PD: Charnley low-friction arthroplasty of the hip: Long-term results. Clin Orthop 292:191–201, 1993.
96. Wroblewski BM, Siney PD, Dowson D, et al: Prospective clinical and joint simulator studies of a new total hip arthroplasty using alumina ceramic heads and cross-linked polyethylene cups. J Bone Joint Surg 78B:280–285, 1996.
97. Zweymuller KA, Lintner FK, Semlitsch MF: Biologic fixation of a press-fit titanium hip joint endoprosthesis. Clin Orthop 235:195–206, 1988.
Richard F. Santore, MD; and Richard A. Brand, MD, Guest Editors