Philosophies of Stem Designs in Cemented Total Hip Replacement

Abstract

Stem designs, which have different design features, may produce similar clinical survival curves. Alternatively, some designs that are considered to be similar in design produce different survival rates. In this paper, design aspects of cemented femoral total hip replacement stems, how they can be grouped to design philosophies, and how they may affect the failure process are discussed. In addition, explanations of unsuccessful designs are posed to learn from previous mistakes and improve understanding of design aspects that affect the longevity of cemented femoral stem designs.

A variety of cemented stem designs are available on the orthopedic market. Most of the current designs perform well with survival rates more than 90% after 10 years. However, younger patients have lower survival rates of 72%-85% after 10 years.^1-3

Scandinavian Register reports show that researchers and clinicians have succeeded in continuous improvement total hip replacements (THR). Nevertheless, we are still far from our final goal which is to create a generation of THR designs that last a lifetime, particularly for younger patients.

Researchers must understand the failure mechanisms involved in aseptic loosening of hip prostheses and be able to separate good and bad prosthetic design features before they are able to improve current THR designs. Huiskes⁴ provided a basis for discriminating between different failure scenarios of THR reconstructions. However, these failure scenarios do not specifically identify which prosthetic design parameters are responsible for success or early failure.

The purpose of this article is to provide general guidelines on prosthetic design features such as prosthetic shape, surface roughness, and stem material. More specifically, this article will discuss how these features affect the mechanical failure process of cemented stems in terms of stem-cement debonding, cement abrasion, stem burnishing, and failure of the cement mantle. Design philosophies of cemented stems that have been proven to work or fail are also discussed. This article may serve as a basis for new THR designs or function as a discussion document for individuals who work in THR design development.

Design Features

Surface Roughness
Migration studies suggest that in all THR prostheses, the stem migrates within the cement mantle.⁵ Thus, one can hypothesize that all current stem designs debond from the cement within a limited period. In the past, it has been postulated that debonding can be prevented by increasing the surface roughness or by using stems that had been precoated with polymethylmethacrylate.^6-8 Little evidence exists that these surface treatments create a long-lasting bond between the stem and the cement. The interfacial strength is not adequate with rough surfaces because rough surfaces usually have an incomplete attachment to the cement mantle, resulting in substantial pores at the stem-cement interface and weakening of the interfacial bond.^9,10 If one adopts the hypothesis that all stems debond from the cement mantle, then one must consider the consequences after stem-cement debonding, including the abrasive behavior of the debonded stem. On the one hand, rougher surfaces seem to have greater abrasive potential. This finding is illustrated by Crowninshield et al.⁷ They applied a cyclic displacement on stems while compressing the stems against bone cement. The stems had various roughness values, and the study confirmed that cement abrasion increases with surface roughness.

However, one can argue that the application of a cyclic displacement on a rough metal is an inaccurate experiment because it does not simulate what occurs clinically. Clinically, a cyclic force is applied, and a rougher surface may increase the friction coefficient at the stem-cement surface, reducing the cyclic motions and possibly the abrasive potential. This procedure was analyzed with finite element micro-models, and a relationship of surface roughness and cyclic micro-motions was established.¹¹ Subsequently, these micro-motions were applied to a micro-model, simulating the asperities of the surface of the stem. The peak stresses in the cement around the asperities of the roughness profile were calculated. The cyclic micro-motions were maximal for a surface roughness of 0 µm. However, due to the absence of asperities on the metal surface, the local cement stresses remained low (Figure 1A). At a roughness value of Ra=15 µm, the local cement stresses were high, indicating a high abrasive mechanism. When the surface roughness was increased again, local cement stresses were reduced again due to reduced cyclic motions caused by the improved “grip” of the metal surface on the cement (Figure 1B).

Figure 1: Results from a finite element micro-analysis of the surface roughness of a straight, tapered unbonded stem. Von Mises stress distribution in the cement around the asperities of the roughness profile of the stem surface (A). Surface roughness values are: 0 (a), 5(b), 15(c), and 30(d) µm. Local stresses around the asperities of the stem surface showed a maximum at 15 µm; beyond that value, the local stresses reduced with surface roughness indicated a lower abrasive potential (B).

This example shows the complexity of stem-cement interface mechanics and cement abrasion. The surface roughness beyond which the abrasive potential diminishes depends on factors such as the prosthetic design, loading conditions, and location. Likewise, it is also difficult to state which surface finish generates negligible cement abrasion.

After reviewing the literature, one can conclude that, in some cases, roughened stems have been shown to fail earlier than polished stems of the same implant. Examples of such stems are the Exeter stem (Stryker, Kalamazoo, Mich) and the Iowa stem.^1,12 In addition, Muller et al¹³ analyzed the in-vivo failure behavior of a stem with a surface roughness of Ra=2 µm. They concluded that the stem was not stable and generated excessive osteolysis, particularly around defects in the cement mantle. On the other hand, Von Knoch et al¹⁴ analyzed 11 retrieved femoral cobalt-chromium stems that had a surface roughness of Ra=1 µm (comparable to the Exeter matte surface finish). They found no marks of burnishing on the stem surfaces, which suggested that the stems had been stable and had not produced significant cement abrasion. Similarly, the Spectron prosthesis (Smith & Nephew Orthopaedics, Inc., Memphis, Tenn) has a roughness of Ra=2.8 µm and has an excellent survival record.^1,5 These clinical findings illustrate the complex interactions of design features of femoral stem designs that cannot be judged on an individual parametric basis. Nevertheless, the question “What is rough and what is polished?” is often posed, and I believe that it is possible to define a classification of abrasive potential in relation to the surface roughness of cemented femoral stems:

Polished: Ra<1 µm. Usually causes little to no cement abrasion.

Matte: Ra<2 µm. Causes no excessive abrasion, but if the stems are designed to create large micro-motions, cement abrasion and surface burnishing may occur.

Rough: Ra>2.0 µm. Surface finish can be applied only at locations where micro-motions are expected to be minimal, usually at the proximal level.

Shape
To facilitate the discussion about the shape of cemented femoral stems, the loading modes applied to the reconstruction should be considered. The loading modes include axial forces, bending forces, and rotational forces. The mechanism of load-transfer of the axial and bending forces is primarily affected by the mid-frontal plane shape of the reconstruction, whereas the cross-sectional shape primarily determines the rotational stability of the design.

Mid-frontal shape. Axial forces can be transferred from stem to cement by different mechanisms. As discussed by Huiskes et al,¹⁵ designs can be classified as “force-closed” or “shape-closed.” The force-closed design relies on a taper that transfers the load to the cement at the stem-cement interface. The external load is in mechanical equilibrium with the (frictional) forces at the stem-cement interface. As the cement creeps or micro-cracks accumulate in the cement, the frictional forces (and circumferential forces) are reduced. Therefore, the stem will migrate to increase the fictional forces and to balance the external forces. Examples of this design include the Exeter stem (Stryker, Kalamazoo, Mich), the collarless polished taper (Zimmer, Inc., Warsaw, Ind), and the C-stem (DePuy Orthopaedics, Warsaw, Ind).

The shape-closed design has features that transfer a large portion of the axial load directly to the cement. These features can be collars, ridges, or profiles. An anatomic design is also considered a shape-closed design feature. These features contribute to the mechanical stability of the implant, even after debonding of the stem-cement interface.

The way bending forces are transferred to the cement depends on the bending stiffness of the stem relative to the cement-bone construct. Stress calculations indicate that almost all designs have stress concentrations at the proximal and distal region. By changing the stiffness, the areas with high stresses can be adapted. A relatively stiff stem will transfer more load distally, whereas a flexible stem will generate higher stresses in the proximal region. When using a flexible material such as titanium, surgeons should avoid including a proximal section with a small proximal mediolateral (ML) dimension because the stem will be too flexible, resulting in early debonding and high micro-motions at the stem-cement interface.

Cross-sectional shape. The first designs of the cemented hip prostheses were designed to withstand loads in the frontal plane (ie, they had to withstand the axial and bending forces). In the past decade, it has become increasingly apparent that the design also must withstand the rotational anteroposterior (AP) forces that are exerted on the reconstruction. Rotational forces can be substantial when patients rise from a chair or climb stairs, and not all cemented stems are equally suited to withstand these forces.¹⁶ Migration studies have also demonstrated the rotational movement of some implants.^5,17 Thus, rotational resistance has become an important parameter of the stem. Stems with a circular cross-sectional shape have a smaller rotational stability. Therefore, the cross-sectional shape should be rectangular or irregular instead of circular to improve the rotational stability.

Figure 2: Cross section of an example of a retrieved specimen showing stem-cement debonding and cement cracking at the corner of the stem.

Stems with proximal-distal profiles along the surface also have an improved rotational stability. However, a disadvantage of irregular cross-sectional shapes is that they may create stress intensities at the stem-cement interface and in the cement, causing stem-cement debonding and cement cracks (Figure 2).

Rotational stability also depends on the cross-sectional size of the implant. The high failure rate of the rough surfaced pre-coated stems showed a trend between failure and small size.¹⁸ This trend suggests that the rotational stability of the smaller stems was inadequate, resulting in rotational migration and subsequent failure of the reconstruction. This finding indicates that surgeons should be critical of the rotational stability of the stem, especially in heavy patients with a large offset. For this reason, surgeons should be reluctant to implant small stem sizes in heavy patients.

In terms of rotational stability, the cross-sectional shape should not be circular but should not include sharp corners which could act as stress risers. Although definite limitations to the fillet radius of the corners of the stem are difficult to provide, they should be at least 2 mm (preferably >3 mm). Most current designs that perform well clinically fulfill these requirements.

Stem Material
The materials commonly used in cemented femoral stems are cobalt-chromium, stainless steel, and titanium. The use of titanium is attractive because its stiffness is closer to the stiffness of bone and bone cement as compared to the stiffness of the other two materials. Titanium implants have also been selected for use because of their modularity and subsequent intraoperative flexibility. However, cemented titanium designs have received a reputation for failing earlier than the cobalt-chromium or stainless steel designs.^19-22 On the other hand, there are reports, which show satisfying results with titanium stems. It seems that titanium implants have two disadvantages.

Titanium has a stiffness of approximately 50% compared to cobalt-chromium or stainless steel. As a result, titanium implants behave more flexibly in the cement mantle than the cobalt-chromium or stainless steel designs, increasing the potential of stem-cement debonding in case it is combined with a slim design. This result is comparable to that of the isoelastic cementless stems that were implanted in the 1980s. These implants failed largely because of the “failed-ingrowth” failure scenario.^4,23 The higher flexibility of the stem increases proximal cement stresses. In designs with a bulky proximal shape, these stresses may be within an acceptable range, but in designs with small proximal dimensions, particularly in the ML direction, cement stresses may be too high, resulting in cement failure.

The second disadvantage of a titanium design is its susceptibility to crevice corrosion. This type of corrosion is driven by the generation of a gap (the crevice) between the stem and the cement. Willert et al²⁴ were the first to report crevice corrosion and occasional reports have appeared since then.^25,26 In contrast, no reports about corrosion in the stainless steel or cobalt-chromium designs have surfaced. Although the impact of crevice corrosion on clinical failure rates is unclear, it is evident that crevice corrosion is a disadvantage of cemented titanium implants.

The disadvantages of titanium cemented stems have caused experts to be concerned about using titanium for cemented applications. The flexibility of slim implant designs may become a complication in heavy patients with small femurs.

Design Philosophies

The existence of a clinical database representing more than three decades of survival data would lead one to believe that distilling the optimal design parameters of prosthesis stems would not be a challenge. However, survival analysis of THRs is a complex matter. Even clinically, the failure criteria are often unclear (eg, revision or radiographic loosening).

In large studies such as the national registers, survival of the cup and stem are sometimes not separated, and the same prosthetic system sometimes performs well in one study but insufficiently in another study. This discrepancy illustrates the numerous factors affecting THR survival.

In addition, it has become clear that a design feature can have a negative effect for one particular prosthetic design, but no effect or a beneficial effect for another prosthetic design. As an example, surface finish can be considered as a prosthetic design parameter. According to the Swedish Register, the Exeter matte stem with a surface finish of approximately Ra=1 µm produced significantly worse results than the polished Exeter stem, which had a surface roughness of about Ra=0.02 µm. This finding suggests that a rough surface finish would lead to inferior survival.

However, in the same Register, stems with rough stem surfaces performed clinically very well. Hence, a combination of inferior design features can lead to ineffective implants. Thus, one must plan an implant design with a design philosophy in mind instead of depending upon individual design features. When a design philosophy is adopted, all design features can be chosen to match the philosophy, optimizing clinical performance.

Mixing Design Philosophies

To demonstrate the importance of the paradigm of design philosophies, one could consider two design philosophies of prosthetic stems that perform clinically well, mix the design features, and discuss the potential consequences. As an example, we will discuss the Exeter stem and the Lubinus SPII (Waldemar-Link GmbH, Hamburg, Germany) stem designs (Figure 3).

Figure 3: Exeter stem with a tapered shape and a polished surface (A). Lubinus SPII stem with an anatomical shape, a collar, stem profiles, and a matte surface finish (B).

The Exeter stem is double tapered, has no collar, has a highly polished surface, is symmetric, and is made of stainless steel. The philosophy behind the design is that the stem anticipates stem-cement debonding, distributes the stresses evenly in the cement mantle (no collar or ridges are present), and accommodates creep and stress relaxation in the cement mantle.

The Lubinus SPII stem has an anatomical shape, longitudinal profiles, a matte surface finish, a collar, and is made of cobalt-chromium. The matte surface finish, profiles, anatomical shape, and collar of the Lubinus SPII stem are design parameters that promote the philosophy of maximal mechanical stability of the stem in the cement mantle, even if the stem debonds from the cement mantle. This philosophy correlates well with the small migration rates of this stem.⁵

For the Exeter stem, higher migration values are reported, designating it as a “migrating” stem with a force-closed load transfer mechanism.¹⁵ In this type of stem, a polished surface finish, tapered shape, and no collar are beneficial. If the Exeter stem had a collar, then the migration would decrease. A major part of the load would be transferred at the proximal level and would probably lead to more cement damage in this region. If the surface roughness of the Exeter stem were increased, then the stem may debond from the cement at a later stage. But once the stem was debonded, the prosthesis would produce more cement (and stem) abrasion, leading to an imperfect fit of the stem in the cement mantle, pathways for wear debris transportation, and higher local cement stresses.

If the Exeter stem was changed from a straight tapered design to an anatomic design, such as that of the Lubinus SPII shape, then migration would be reduced. However, the stem would continue to migrate more than the stem of the Lubinus SPII design and create gaps between the stem and cement due to the polished surface finish and the absence of a collar. Migration and gap formation lead to worn particle pathways and an uneven stress distribution in the cement mantle, resulting in earlier failure of the reconstruction.

In a clinical study performed by Kärrholm et al,⁵ the design philosophies are mixed intentionally. The study includes three types of Lubinus stems, a polished version without a collar, a pre-coated stem with a collar, and a standard version. Logically, the polished stem without a collar subsides more than the other two components, which subside a comparable amount. However, the polished Lubinus stem migrates 0.3 mm after 1 year, whereas the Exeter stem migrates 1 mm after 1 year.⁵ It is understandable that the polished Lubinus stem migrates less than the Exeter stem because the polished Lubinus stem has an anatomical shape and surface profiles. These two features limit the migration of the stem and localize cement stresses. The collarless polished Lubinus SPII stem, which is used only in this clinical experiment, has a clear mixture of features from the force-closed and shape-closed design philosophies. Therefore, the Lubinus SPII stem should perhaps not be implanted because it may cause more cement damage and earlier loosening than the Exeter stem in the long-term.

Figure 4: Scientific Hip Prosthesis had a small ML dimension at the resection level and in the tip region.

In the paper by Huiskes et al,¹⁵ the Scientific Hip Prosthesis (SHP) (Biomet, Warsaw, Ind) stem was used as an example of the shape-closed design. The SHP stem was designed in the mid 1980s with the assumption that the stem would remain bonded to the cement. Subsequently, a finite element computer simulation calculated the optimal shape to minimize cement stresses.²⁷ The stem had a small AP dimension at the proximal and distal region (Figure 4).

Experimental work has confirmed that this shape works in terms of more global cement stresses.²⁸ However, current migration data suggest that most stems debond from the cement and that the assumption of a bonded interface is incorrect. For the SHP prosthesis, this has become evident in migration data reported by Nivbrant and colleagues.²⁹ A debonded SHP stem has only a limited number of features that represent a shape-closed design.

In retrospect, the SHP design was created using a mixed design philosophy. If the stem had been fitted with a collar, then the shape-closed theory would be more applicable, less migration would occur, and survival would most likely have been better.

The Capital Hip: Absence of a Design Philosophy

Choosing a good design philosophy is more difficult than it seems. The best way to design a stem that works well is to learn from earlier mistakes. An example of a mistake is the Capital hip design, which was marketed in the 1990s by 3M Health Care Ltd., particularly in the United Kingdom. In 1997, The Capital hip design was taken off the market when reports of high failure rates appeared in the media and literature.^30,31

The Capital design is often described as a copy of the Charnley design. However, an examination of the designs shows that the surface roughness of the Capital design is higher than the surface roughness of the Charnley design. Also, the Capital design represents elements of .four designs. 3M Health Care Ltd. made two prosthetic geometries (the round back and the flanged design) (Figure 5), both of which could be made of stainless steel or titanium. No scientific consideration seems to have influenced these design variations, and no scientific design philosophy seems to have been used. Only the argument that the stems of the Capital design resemble the stems of the Charnley design influenced these variations. These round back, flanged, stainless steel, and titanium design variations affected the clinical survival rates of the stem prostheses (Table).

Figure 5: Charnley hip system (A) and the Capital hip system (B). Note that the flanged Capital hip differs from the flanged Charnley stem.

The surface of the Capital hip stem was treated with a shot-blasted procedure to achieve roughness. Titanium is softer than stainless steel, resulting in differences in the surface roughness of the titanium stems (Ra=1.1 µm) and the stainless steel (Ra<1.1 µm) stems.³² In addition, the flanged design in the Capital design did not replicate the Charnley flanged design due to patent restrictions. Therefore, the flange on the Capital design was less pronounced than the flange on the Charnley design.

McGrath et al³³ described the failure mechanism of the titanium Capital design (primarily of the flanged type). The failure began with lateral-proximal debonding, cement abrasion, migration, and osteolysis. From a mechanical point of view, these observations are logical. The titanium stems, in combination with the relatively slim design, are more flexible than the stainless steel versions. This flexibility promotes early debonding of the stem.

After debonding of the Capital titanium stem, cement (and stem) abrasion occurred as a result of the interaction between high surface roughness and high micro-motions, causing the high flexibility of the implant. The high flexibility of the titanium stems caused them to create higher cement stresses than the stainless steel stems. Consequently, more cement damage accumulated, resulting in the increased likelihood of cement crack formation in titanium stems.

Janssen et al³⁴ have demonstrated this mechanism. They simulated the mechanical failure mechanism around the Capital hip stems using finite element techniques. They found more cement cracks around the titanium stems than they found around the stainless steel stems (Figure 6A).

The round back stems may have performed better clinically than the flanged stems because the round back stems had better rotation stability than the flanged stems. This finding has been reported by Ramamohan et al³¹ and became apparent in the finite element simulations performed by Janssen et al,³⁴ in which the roundback stem was more susceptible for out-of-plane loading (stair climbing) than the roundback design (Figure 6B).

Figure 6: Results of computer finite element simulation of the 4 Capital hip stems. Accumulation of damage in the complete cement mantle. Note that the flanged design produced more damage than the round back, and the titanium implants produced more damage than the stainless steel implants (A). Migration of the femoral head relative to the bone in the model after the damage and creep simulation. Note that the migration in posterior direction of the head of the flanged design was greater than the migration in the roundback design, indicating the inferior rotational stability of the flanged design (B).

As stated earlier, not all Capital hip designs failed in large numbers. The stem most similar to the Charnley stem was the stainless steel roundback Capital stem, which has a low failure rate (Table). For the patients treated with Capital hip designs and for 3M Health Care Ltd., it was unfortunate that approximately 50% of the Capital hip stems were of the titanium, flanged design type (Table). If the majority of the Capital hip stems had been of the round back stainless steel type, then the high failure rate of the flanged, titanium design may not have been as apparent.

Table

Details of the Capital Hip Design Stem Design Stem Material Loosening Rate [%]* Distribution [%] Flanged Titanium 9.1 48.2 Roundback Titanium 6.1 20.5 Flanged Stainless steel 4.4 22.0 Roundback Stainless steel 1.5 9.4 Source: Royal College of Surgeons of England, 2001.a; 2001.b. *Loosening rate until issue Hazard Notice in 1997.

Conclusion

The clinical survival of cemented femoral components depends on many design factors, but is also affected by patient and surgeon factors. It is this complex interaction of factors that makes it difficult to identify the design parameters that result in a successful implant. The examples of failed implants included in this article highlight this complex interaction, illustrating the urgent need for reliable, pre-clinical test methods, careful introduction of new implants to the orthopedic community, and rigid post-marketing surveillance of these new devices.⁴

The principal attitude toward novel designs should be to establish a design philosophy and check if all design features contribute to optimize the design philosophy, and to decrease the sensitivity of the design to patient- and surgeon-related factors. In this way, researchers and surgeons can learn from previous mistakes and improve the performance of cemented THR components, particularly for younger patients.

References

1. Herberts P, Malchau H, Garellick G. Annual report 2003, The Swedish National Hip Arthroplasty Register. 2004.
2. Puolakka TJ, Pajamaki KJ, Halonen PJ, Pulkkinen PO, Paavolainen P, Nevalainen JK. The Finnish Arthroplasty Register: Report of the hip register. Acta Orthop Scand. 2001; 72:433-441.
3. Havelin LI, Engesaeter LB, Espehaug B, Furnes O, Lie SA, Vollset SE. The Norwegian Arthroplasty Register: 11 years and 73,000 arthroplasties. Acta Orthop Scand. 2000; 71:337-353.
4. Huiskes R. Failed innovation in total hip replacement. Diagnosis and proposals for a cure. Acta Orthop Scand. 1993; 64:699-716.
5. Kärrholm J, Nivbrant B, Thanner J, Anderberg C, Börlin N, Herberts P, Malchau H. Radiosstereometric evaluation of hip implant design and surface finish. Scientific Exhibition, AAOS, Orlando, USA, 2000.
6. Chen CQ, Scott W, Barker TM. Effect of metal surface topography on mechanical bonding at simulated total hip stem-cement interfaces. J Biomed Mater Res. 1999; 48:440-446.
7. Crowninshield RD, Jennings JD, Laurent ML, Maloney WJ. Cemented femoral component surface finish mechanics. Clin Orthop. 1998; 355:90-102.
8. Davies JP, Harris WH. Strength of cement-metal interfaces in fatigue: Comparison of smooth, porous and precoated specimens. Clin Mater. 1993; 12:121-126.
9. Verdonschot N, Huiskes R. Surface roughness of debonded straight-tapered stems in cemented THA reduces subsidence but not cement damage. Biomaterials. 1998; 19:1773-1779.
10. Race A, Miller MA, Ayers DC, Cleary RJ, Mann KA. The influence of surface roughness on stem-cement gaps. J Bone Joint Surg Br. 2002; 84:1199-1204.
11. Verdonschot N, Tanck E, Huiskes R. Effects of prosthesis surface roughness on the failure process of cemented hip implants after stem-cement debonding. J Biomed Mater Res. 1998; 42:554-559.
12. Collis DK, Mohler CG. Comparison of clinical outcomes in total hip arthroplasty using rough and polished cemented stems with essentially the same geometry. J Bone Joint Surg Am. 2002; 84:586-592.
13. Muller RT, Heger I, Oldenburg M. The mechanism of loosening in cemented hip prostheses determined from long-term results. Arch Orthop Trauma Surg. 1997; 116:41-45.
14. Von Knoch M, Bluhm A, Morlock M, Von Foerster G. Absence of surface roughness changes after insertion of one type of matte cemented femoral component during 2 to 15 years. J Arthroplasty. 2003;18:471-477.
15. Huiskes R, Verdonschot N, Nivbrant B. Migration, stem shape, and surface finish in cemented total hip arthroplasty. Clin Orthop. 1998; 355:103-112.
16. Bergmann G, Deuretzbacher G, Heller M, et al. Hip contact forces and gait patterns from routine activities. J Biomech. 2001; 34:859-871.
17. Gill HS, Alfaro-Adrian J, Alfaro-Adrian C, McLardy-Smith P, Murray DW. The effect of anteversion on femoral component stability assessed by radiostereometric analysis. J Arthroplasty. 2002; 17:997-1005.
18. Sylvain GM, Kassab S, Coutts R, Santore R. Early failure of a roughened surface, precoated femoral component in total hip arthroplasty. J Arthroplasty. 2001; 16:141-148.
19. Bowditch M, Villar R. Is titanium so bad? Medium-term outcome of cemented titanium stems. J Bone Joint Surg Br. 2001; 83:680-685.
20. Jergesen HE, Karlen JW. Clinical outcome in total hip arthroplasty using a cemented titanium femoral prosthesis. J Arthroplasty. 2002; 17:592-599.
21. Lichtinger TK, Schurmann N, Muller RT. Early loosening of a cemented hip endoprosthesis stem of titanium. Unfallchirurg. 2000; 103:956-960.
22. Ebramzadeh E, Normand PL, Sangiorgio SN, et al. Long-term radiographic changes in cemented total hip arthroplasty with six designs of femoral components. Biomaterials. 2003; 24:3351-3363.
23. Niinimaki T, Puranen J, Jalovaara P. Total hip arthroplasty using isoelastic femoral stems. A seven- to nine-year follow-up in 108 patients. J Bone Joint Surg Br. 1994; 76:413-418.
24. Willert HG, Broback LG, Buchhorn GH, et al. Crevice corrosion of cemented titanium alloy stems in total hip replacements. Clin Orthop. 1996; 333:51-75.
25. Thomas SR, Shukla D, Latham PD. Corrosion of cemented titanium femoral stems. J Bone Joint Surg Br. 2004; 86: 974-978.
26. Hallam P, Haddad F, Cobb J. Pain in the well-fixed, aseptic titanium hip replacement. The role of corrosion. J Bone Joint Surg Br. 2004; 86:27-30.
27. Huiskes R, Boeklagen R. Mathematical shape optimization of hip prosthesis design. J Biomech. 1989; 22:793-804.
28. Peters CL, Bachus KN, Craig MA, Higginbotham TO. The effect of femoral prosthesis design on cement strain in cemented total hip arthroplasty. J Arthroplasty. 2001; 16:216-224.
29. Nivbrant B, Karrholm J, Soderlund P. Increased migration of the SHP prosthesis: Radiostereometric comparison with the Lubinus SP2 design in 40 cases. Acta Orthop Scand. 1999; 70:569-577.
30. Massoud SN, Hunter JB, Holdsworth BJ, Wallace WA, Juliusson R. Early femoral loosening in one design of cemented hip replacement. J Bone Joint Surg Br. 1997; 79:603-608.
31. Ramamohan N, Grigoris P, Schmolz W, Chappell AM, Hamblen DL. Early failure of stainless steel 3M Captial femoral stem. J Bone Joint Surg Br. 2000; 82:71.
32. Royal College of Surgeons of England. An Investigation of the Performance of the 3M Capital Hip System, 2001.
33. McGrath LR, Shardlow DL, Ingham E, et al. A retrieval study of capital hip prostheses with titanium alloy femoral stems. J Bone Joint Surg Br. 2001; 83:1195-1201.
34. Janssen DW, Aquarius R, Stolk J, Verdonschot N. Inferior design characteristics of the Capital Hip can be detected by numerical analysis. ORS, 2005.

Author

From the Orthopaedic Research Laboratory, Radboud University Nijmegen Medical Center Nijmegen, The Netherlands.