Adhesion and reliability of interfaces in cemented total joint arthroplasties

Debonding of the prosihctielpolymelhylmcthacrylate interface has been implicated in the initial failure process of cemented total hip arthroplasties. However, little quantitative understanding of the debonding process, as well as of the optimum interface morphology for enhanced resistance to debonding, exists. Accordingly, a fracture-mechanics approach has been used in which adhesion at the interface is characterized in terms of the interface fracture energy. G (J/m²), and shown to be a strong function of the morphology, debonding length, and loading mode of the interface. Double-cantilever-beam and four-point-flexure fracture-mechanics samples containing four clinically relevant prosthetic surface preparations were prepared to survey a range of interface roughness and loading modes. Adhesion at the interface could not be characterized with a single-valueu material property but was found to exhibit resistance-curve behavior in which resistance to debonding increased with both the initialdebond extension and the roughness of the interface. Values of debonding. initiation. G_o, were relatively insensitive to the roughness of the surface and the loading mode, whereas steady-state fracture resistance of the interface. G_ss, increased significantly with the roughness and shear loading of the interface. These quantitative results suggest that debonding of the prosthetic/polymethyl methacrylate interface may be primarily attributed to surface interactions such as interlocking and the pullout of rough asperities thai occur behind the debond tip. A simple mechanics analysis of such interactions was performed and revealed increases in the fracture resistance of the interface that were consistent with experimentally measured values.     

The Frank Stinchfield Award. Influence of cement technique on the interface strength of femoral components

The optimal surface finish for polymethylmethacrylate cemented femoral components remains controversial. Concerns about early debonding of the prosthesis-cement interface have led surgeons to use roughened surfaces to enhance the cement-prosthesis bond. However, loosening of roughened stems is associated with the generation of excessive wear debris. The purpose of the current study was to determine whether the time to cementation influenced the cement-prosthesis bond of four roughened cobalt chrome surfaces (60 grit-blasted, 10 grit-blasted, 10 grit-blasted with polymethylmethacrylate precoating, glass bead-blasted) and one polished cobalt chrome surface. Fixation strength was assessed using mechanical pushout and tensile testing. Roughened and polymethylmethacrylate precoated surfaces had significantly greater tensile and shear strengths at early cementation times compared with polished surfaces. However, roughened components had significant decreases in tensile and shear strengths as cementation time increased from 2 to 4 minutes and 2 to 6 minutes. In contrast, tensile and shear strengths for the polished surface were significantly lower than for the roughened surfaces and did not change with longer cementation times. When using a roughened or precoated cemented femoral component, the surgeon should consider cementing earlier with wetter cement to maximize the cement-prosthesis bond. When implanting a polished femoral component, it is preferable that the cement is doughy, because the cement-prosthesis bond is not influenced by the wetness of the cement and it is easier to maintain the orientation of the femoral component.     

The influence of support conditions in the loading fixture on failure mechanisms in the push-out test: a finite element study

The usefulness of the push-out test as an indicator of interface strength was evaluated using finite element models of intact and partially failed cylindrical push-out specimens loaded against a rigid annular support. The irregular stress distributions that were found in intact specimens depended more on interface conditions at the loading fixture than on a 35% increase in interface area. The maximum stress at the interface was a tensile stress. Critical energy release rates for interface failure were calculated for flawed specimens in which flaw size was either 10 or 100 microns, and for boundary conditions at the loading fixture that were either fixed or slipping in the radial direction. The critical energy release rates depended heavily on the support boundary conditions. Thus, the results of parametric push-out tests can be reasonably compared only for specimens that are very similar in geometry and that are loaded in very carefully controlled fixtures.     

The significance of stem-cement loosening of grit-blasted femoral components

This study analyzed 15 patients who underwent revision for loosening at the stem-cement interface. The femoral components were from the same manufacturer and had grit-blast roughened surfaces. An apparent radiographic deficiency in the cement mantle was present in at least one zone in 1 3 patients. In 9 of 12 patients with localized osteolysis, the osteolysis developed in a zone with an apparent radiographic cement mantle defect. Loosening occurred due to tension failure of the stem-cement interface followed by axial subsidence and movement into relative retroversion. Motion between the stem and the cement mantle fueled an abrasive wear mechanism between the roughened metal surface and the cement mantle, generating excessive metal and cement particles that gained access to endosteal bone via defects in the cement mantle and resulting in localized osteolysis. Although the roughened surface played a central role in these failures, it is unlikely the layer of polymethylmethacrylate (precoat) played a role in the mechanism of failure. In some cases, debonding occurred as a result of tension failure of the metal-precoat interface. In others, tension failure occurred within the cement mantle, leaving the precoat and some cement from the mantle on the stems. There was no difference in the mechanism of failure of stems with precoat proximally compared to stems with precoat proximally and distally. One stem had no precoat; findings in this patient were indistinguishable from the others. The significance of debonding depends on the surface roughness of the stem. Debonding carries a poorer prognosis with a rougher stem surface because of abrasive wear with the generation of numerous metal and cement particulates, which can lead to rapid osteolysis if there are cement mantle defects. Stems with a higher metal-cement bond strength may require a higher quality cement mantle for long-term success.     

Primary hybrid total hip arthroplasty with a roughened femoral stem: integrity of the stem-cement interface

One hundred and two consecutive cemented femoral stems were evaluated in 92 patients at an average 9-year follow-up and a minimum 5-year follow-up (range, 5-14 years). The stem used was cobalt chromium with a collar, normalization steps, and a roughened surface (Ra 40); the stem was inserted using contemporary cementing techniques. This series demonstrated a femoral component aseptic loosening rate of 2.0% and a femoral component survivorship of 97.2 +/- 2.0% at 10 years. One of 2 failed stems was revised at 95 months for failure at the cement-bone interface. The second failed stem showed failure at the cement-bone interface with incomplete debonding radiographically at 65 months. The remaining femoral components did not demonstrate any evidence of debonding at the stem-cement interface. These results compare favorably with other series of cemented femoral stems, as well as with those with a polished surface.     

A finite element study of the initiation of failure of fixation in cemented femoral total hip components.

In order to study initial mechanisms of failure in cemented femoral total hip components, an anatomically accurate three-dimensional linear finite element model was constructed and verified against experimental strain measurements in the cement mantle. Good agreement was found between predicted and measured strains. The likelihood of failure initiation due to cement-prosthesis debonding and crack initiation at voids was studied for loading conditions simulating both one-legged stance and stair climbing. The "out of plane" forces involved in stair climbing appear to be the greatest threat to the fixation of total hip replacements. In stair climbing, cement-prosthesis debonding and pore crack initiation were probable in the proximal anteromedial region of the cement mantle, and near the distal tip of the implant. The proximal stresses in stair climbing were higher than the distal stresses in either stair climbing or one-legged stance.     

Shear and tensile strength of hydroxyapatite coating under loading conditions

The shear and tensile strength of a hydroxyapatite (HA) coating on a femoral component was studied after physiological loading conditions in 8 German Shepherds. A proximal macrostructure on the stem was used to protect this region from shear stresses. Another four implantations with uncoated components were used as controls. In vitro testing of the HA layer demonstrated excellent tensile strength and stability to surface deformation. The loaded implants were tested at 6, 12, and 24 weeks. At 6 weeks the HA-coated components could easily be removed by axial loading, whereas the HA layer remained undamaged on the metal. However, pull out tests of implants older than 12 weeks showed complete debonding of the HA layer from the non-macrostructured surface due to shear forces in all cases. Debonding of the HA layer was also observed with microradiography. The macrostructured surface prevented dislodging of the component from this area at pull out test by distributing shear forces. Unlike in uncoated implants, considerable amounts of bone remained attached onto the HA macrostructure when the surrounding femur was pulled out. Shear forces cause debonding of the HA layer, while tensile stress affects failure within the bone. Physiological loading partially produces gaps at the interface so direct transmission of tensile forces onto the bone is lost, and the coating-metal interface becomes the weak point in the system.     

Monitoring the integrity of the cement—metal interface of total joint components in vitro using acoustic emission and ultrasound

Debonding of the cement—metal interface of cemented femoral components of total hip arthroplasty has been shown from clinical and autopsy material to be a common occurrence. Experimentally, debonding has been shown to increase markedly the strains in the adjacent cement mantle. Studies of autopsy-retrieved specimens demonstrate that debonding of the cement—metal interface is a key initiating event in loosening of cemented femoral components of total hip arthroplasty. However, both the radiographic and autopsy evidence of cement—metal interfacial debonding exist after the fact, that is, after debonding has occurred. The lack of prospective data showing that debonding does indeed occur under physiologic loading and occurs prior to other forms of failure of fixation leaves uncertain the issue of debonding and its role in initiating loosening of cemented femoral components. Knowing when, where, and to what extent the cement—metal interface debonds is critical information in understanding the process of loosening of cemented femoral components. Such information would contribute to improving the durability of stems and improving cementing techniques. In this study, the two nondestructive techniques of acoustic emission and ultrasonic evaluation of the cement—metal interface of cemented femoral stems of total hip arthroplasty were combined to investigate when, where, and to what extent cement—metal debonding occurred in vitro in simulated femurs loaded physiologically in fatigue in simulated single-leg stance. Debonding of the cement—metal interface of a cemented femoral component in this model was both an initiating event and a major mechanism of compromise of the cement—metal interface. Additional acoustic emission signals arose from cracks that developed in the cement.     

Biomechanical study of pins in cementing of contained proximal tibia defect

Defects from curettage for giant cell tumors of bone frequently have been reconstructed with bone cement with or without reinforcement pins. The biomechanical basis for the addition of reinforcement pins was examined using a model of a contained defect in the proximal tibia. Fifty-four cadaveric proximal tibia in matched pairs were divided into five test groups: intact tibia, medial metaphyseal contained defect, defect reconstructed with cement alone, defect reconstructed with cement and pins inserted within the medullary canal, and defect reconstructed with cement and pins inserted through the cortex. Specimens were tested to failure during one cycle of compressive loading. Defect specimens were significantly weaker and less stiff than intact specimens, establishing the validity of the model-contained defects. For the reconstructions, there was no statistically significant difference in load to failure, stiffness, energy to failure, or displacement for the polymethylmethacrylate treatment alone when compared with matched specimen receiving polymethylmethacrylate and pins treatment. Similarly, there was no statistical difference in biomechanical properties in comparing matched specimens treated with polymethylmethacrylate alone or polymethylmethacrylate/pins (cortex). For contained defects of the proximal tibia that are typical after curettage for giant cell tumor, there appears to be no biomechanical advantage to use of reinforcement pins in the cement.     

Osteon pullout in the equine third metacarpal bone: effects of ex vivo fatigue.

An important concept in bone mechanics is that osteons influence mechanical properties in several ways, including contributing to toughness and fatigue strength by debonding from the interstitial matrix so as to "bridge" developing cracks. Observations of "pulled out" osteons on fracture surfaces are thought to be indicative of such behavior. We tested the hypothesis that osteon pullout varies with mode of loading (fatigue vs. monotonic), cortical region, elastic modulus, and fatigue life. Mid-diaphseal beams from the dorsal, medial, and lateral regions of the equine third metacarpal bone were fractured in four point bending by monotonic loading to failure under deflection control, with or without 10(5) cycles of previous fatigue loading producing 5000 microstrain (15-20% of the expected failure strain) on the first cycle; or sinusoidal fatigue loading to failure, under load or deflection control, with the initial cycle producing 10,000 microstrain (30-40% of the expected failure strain). Using scanning electron microscopy, percent fracture surface area exhibiting osteon pullout (%OP.Ar) was measured. Monotonically loaded specimens and the compression side of fatigue fracture surfaces exhibited no osteon pullout. In load-controlled fatigue, pullout was present on the tension side of fracture surfaces, was regionally dependent (occurring to a greater amount dorsally), and was correlated negatively with elastic modulus and positively with fatigue life. Regional variation in %OP.Ar was also significant for the pooled (load and deflection controlled) fatigue specimens. %OP.Ar was nearly significantly greater in deflection controlled fatigue specimens than in load-controlled specimens (p=0.059). The data suggest that tensile fatigue loading of cortical bone eventually introduces damage that results in osteonal debonding and pullout, which is also associated with increased fatigue life via mechanisms that are not yet clear.     

Mechanical failure of hydroxyapatite- and polysulfone-coated titanium rods in a weight-bearing canine model.

Three types of material that have shown potential as coatings for orthopaedic implants were studied. Using a weight-bearing canine model, Ti-6A1-4V femoral intramedullary rods coated with (1) sintered titanium beads, (2) plasma-sprayed hydroxyapatite, and (3) silyl coupled polysulfone beads were evaluated for mechanical strength and bone ingrowth. The model was designed to secure optimal prosthetic stability by obtaining maximal bony ingrowth during an initial non-weight-bearing phase, then stressing the implant during a full-weight-bearing phase. None of the rods coated with titanium beads failed. All 17 polysulfone-coated rods failed, 13 of them at the interface between the polysulfone coating and the titanium core. Of 18 rods coated with hydroxyapatite, 15 suffered implant breakdown at the interface between the hydroxyapatite coating and the titanium core. This may be due to dissolution of the plasma-sprayed hydroxyapatite in vivo. Testing of retrieved specimens from both hydroxyapatite- and polysulfone-coated implants showed that the shear strength at the coating-rod interface had decreased to less than 40% of the shear strength at manufacture. Despite mechanical failure, histologic study showed extensive bone ingrowth or apposition onto both the polysulfone and hydroxyapatite coatings.     

Surface coating to improve the metal-cement bonding in cemented femur stems

INTRODUCTION: Hydrolytic debonding of the metal-cement interface is one of the main reasons for aseptic loosening in cemented hip arthroplasty. MATERIALS AND METHODS: BiContact femur stems (CoCrMo-/TiAl(6)V(4)-alloy) were coated by a silica/silane interlayer coating system. The stems were cemented into artificial femurs. The cyclical loading (DIN ISO 7206-4) was performed within a hip-simulator. Uncoated stems (CoCrMo-/TiAl(6)V(4)-alloy) were prepared and loaded the same way. After loading, the metal-cement and the bone-cement interfaces were analysed. Unloaded uncoated and unloaded coated BiContact stems served as a control. RESULTS: The coated loaded stems showed a significant reduction in debonding and cement failure (P /= 0.9). There was no significant difference between CoCrMo- and TiAl(6)V(4)-stems (P >/= 0.05). CONCLUSION: The silica/silane coating significantly decreased hydrolytic debonding at the metal-bone cement interface with consecutively less cement failure.     

Early failure of precoated femoral components in primary total hip arthroplasty

BACKGROUND: In an effort to decrease the rate of aseptic loosening, certain cemented femoral components were designed to have a roughened or textured surface with a methylmethacrylate precoating. Reports differ as to whether this step has increased or decreased the rate of failure. This study was designed to evaluate this issue. METHODS: Five hundred and fourteen hips treated with a cemented Harris Precoat stem (Zimmer, Warsaw, Indiana) were evaluated clinically and radiographically and compared with 254 hips treated with an uncoated Harris Design-2 stem (Howmedica, East Rutherford, New Jersey). Prostheses that had been removed at revision were examined. The cementing and surgical techniques were identical and the population demographics were similar for these two groups. RESULTS: The mean durations of follow-up were 8.4 and 13.5 years for the Precoat and uncoated Design-2 stems, respectively. At those times, at least forty-nine (9.5%) of the 514 Precoat components and at least ten (3.9%) of the 254 uncoated Design-2 stems had failed (p = 0.006). Five Precoat stems fractured, and no uncoated Design-2 stems fractured. Component failure was associated with use in young, active, heavy men with a diagnosis of avascular necrosis and generally with the use of smaller components. The cementing technique was satisfactory in the majority of the patients, and there were no qualitative differences in cementing technique between the hips that failed and those that did not. The mechanisms of failure of the Precoat prostheses included bone-cement loosening, focal osteolysis, stem fracture, and prosthesis-cement debonding. Fractures of smaller components occurred as a result of fatigue failure and were associated with good distal fixation but proximal stem loosening. CONCLUSIONS: The rate of failure of roughened, precoated, cemented femoral components was considerably higher and occurred earlier than that of femoral components that were neither textured nor precoated with methylmethacrylate. Younger patients with avascular necrosis had a higher risk of failure; however, this factor alone did not completely explain the differences in outcome between these two components. The causes of aseptic loosening are multifactorial and may be related to component design and size as well as to precoating and surface finish.     

In vitro influence of stem surface finish and mantle conformity on pressure generation in cemented hip arthroplasty

Background and purpose Under physiological loads, debonded cemented femoral stems have been shown to move within their cement mantle and generate a fluid pump that may facilitate peri-prosthetic osteolysis by pressurizing fluid and circulating wear debris. The long-term physiological loading of rough and polished tapered stems in vitro has shown differences in performance, with greater interface pressures generated by the rough stems. In this study we investigated the individual effects of stem surface finish, degree of mantle wear, and mode of loading on the stem pump mechanism.Method Rough and polished stems were loaded under different regimes in artificially worn cement mantles that permitted either 2 or 5 degrees of rotational stem movement, and the interface pressures were compared.Results The pressures generated by the rough and polished stems were similar in either type of mantle. The pattern of pressure generation in the 2-degree mantles was similar to the pressures generated by rough stems after long-term loading, but the high posterior wall pressures fell and the tip pressures increased in the 5-degree mantles. The torsional loads were principal drivers of pressure generation in all areas of the interface other than the implant tip, where axial loading predominated.Interpretation Femoral stems with rotational instability under cyclic torsional loads generate elevated interface fluid pressures and flows independently of stem surface finish. The rough surface finish is only important in creating this instability in tapered stems.     

Metal/cement interface strength in cemented stem fixation

To characterize the strength of the interface between stem-type metal implants and bone cements, a fracture mechanics parameter was used. This parameter, the critical strain energy release rate (G_c), was determined from “push-out” tests of cylindrical specimens. The specimens, formed using molds of bone, were maintained and tested at body temperature. The strength of interfaces formed with cancellous bone surrounding the cement mantle was significantly less than the strength of those formed in apposition to cortical bone. A marked degradation of strength was found with saline immersion for SS316LVM/cement interfaces formed with Zimmer regular, Simplex-P, and Zimmer LVC cements. After 60 days of immersion the interface G_c was only 10–20% of the value for bulk cement. Interfaces formed with thin-film polymethylmetharcrylate-precoated metals (SS316LVM, Co-Cr-Mo, and Ti-6A1-4V) yielded “dry” G_c values one order of magnitude greater than those measured with interfaces formed with uncoated metals. Moreover, the strength of precoated SS316LVM/cement interfaces formed with all three brands of cement did not change after saline immersion for 60 days.     

The fixation of the cemented femoral component. Effects of stem stiffness, cement thickness and roughness of the cement-bone surface

After cemented total hip arthroplasty (THA) there may be failure at either the cement-stem or the cement-bone interface. This results from the occurrence of abnormally high shear and compressive stresses within the cement and excessive relative micromovement. We therefore evaluated micromovement and stress at the cement-bone and cement-stem interfaces for a titanium and a chromium-cobalt stem. The behaviour of both implants was similar and no substantial differences were found in the size and distribution of micromovement on either interface with respect to the stiffness of the stem. Micromovement was minimal with a cement mantle 3 to 4 mm thick but then increased with greater thickness of the cement. Abnormally high micromovement occurred when the cement was thinner than 2 mm and the stem was made of titanium. The relative decrease in surface roughness augmented slipping but decreased debonding at the cement-bone interface. Shear stress at this site did not vary significantly for the different coefficients of cement-bone friction while compressive and hoop stresses within the cement increased slightly.     

Interface gap after implantation of a cemented femoral stem in pigs

We studied the interface gap around cemented femoral stems. Fresh pig femora were used. Bone cement mixed under vacuum or at atmospheric pressure was injected into the femoral canal and a cobalt chrome stem was then implanted. The femora were sectioned transversely from the minor trochanter and distally by using a high-pressure water cutter. Most of the interfaces had intimate contact. However, in all specimens, small gaps were found at the bone-cement and cement-stem interfaces. The gaps at the interfaces between the bone and cement and the cement and stem were measured, using a computerized video digital system. They occupied about 10% of the circumference at the bone-cement interface and about 15% of the circumference at the cement-stem interface, irrespective of the mixing procedures. Most gaps were less than 100 u at the interfaces. In conclusion, cemented implants in the animal model showed that small gaps are found at the interfaces directly after implantation. These gaps may be weak points and initiate debonding when loading the prostheses.     

Mechanical characteristics of the bone-graft-cement interface after impaction allografting.

Impaction allografting is an attractive procedure for the treatment of failed total hip replacements. The graft-cement-host bone interface after impaction allografting has not been characterized, although it is a potential site of subsidence for this type of revision total hip reconstruction. In six human cadaveric femurs, the cancellous bone was removed proximally and local diaphyseal lytic defects were simulated. After the impaction grafting procedure, the specimens were sectioned in 6 mm transverse sections and push-out tests were performed. From the adjacent sections the percentage cement contact of the PMMA cement with the endosteal bone surface was determined. The host bone interface mechanical properties varied significantly along the femur largely due to different interface morphologies. The apparent host bone interface shear strength was highest around the lesser trochanter and lowest around the tip of the stem. A significant positive correlation was found between the percentage cement contact and the apparent host bone interface shear strength (r2 = 0.52). The sections failed in 69% of the cases through a pure host bone interface failure without cement or allograft failure, 19% failed with local cement failure, and 12% with a local allograft failure. The apparent host bone interface strength was on average 89% lower than values reported for primary total hip replacements and were similar to cemented revisions proximally and lower distally. This study showed that cement penetration to the endosteal surface enhanced the host bone-graft interface.     

Analysis of 16 retrieved proximally cemented femoral stems

Sixteen proximally cemented, collared, and distally splined, Bridge Hip femoral stems with a matte proximal surface and smooth distal surface were retrieved because of loosening. Electron microscopy, with correlated elemental analysis, identified titanium particulate embedded in the internal surface of the cement mantle. Data supported the observations that loosening of the femoral stems was related to proximal debonding at the cement-implant interface, loosening at the proximal cement-bone interface, and inherent rotational instability. Cement-implant interface debonding resulted in the proximally matte femoral stem surface abrading with the opposing cement mantle, resulting in particulate and osteolysis in some cases. Careful consideration of implant design and clinically relevant biomechanical testing protocols should be considered before the clinical introduction of future proximally cemented femoral stems.     

Interface gap after implantation of a cemented femoral stem in pigs.

We studied the interface gap around cemented femoral stems. Fresh pig femora were used. Bone cement mixed under vacuum or at atmospheric pressure was injected into the femoral canal and a cobalt chrome stem was then implanted. The femora were sectioned transversely from the minor trochanter and distally by using a high-pressure water cutter. Most of the interfaces had intimate contact. However, in all specimens, small gaps were found at the bone-cement and cement-stem interfaces. The gaps at the interfaces between the bone and cement and the cement and stem were measured, using a computerized video digital system. They occupied about 10% of the circumference at the bone-cement interface and about 15% of the circumference at the cement-stem interface, irrespective of the mixing procedures. Most gaps were less than 100 mu at the interfaces. In conclusion, cemented implants in the animal model showed that small gaps are found at the interfaces directly after implantation. These gaps may be weak points and initiate debonding when loading the prostheses.