Abstract
Cylindrical porous-coated implants were placed in the distal femoral metaphyses of twenty dogs and were subjected to zero, twenty, forty, or 150 micrometers of oscillatory motion for eight hours each day for six weeks with use of a specially designed loading apparatus. The in vivo skeletal responses to the different magnitudes of relative motion were evaluated.Histological analysis demonstrated growth of bone into the porous coatings of all of the implants, including those that had been subjected to 150 micrometers of motion. However, the ingrown bone was in continuity with the surrounding bone only in the groups of implants that had not been subjected to motion or that had been subjected to twenty micrometers of motion; in contrast, the implants that had been subjected to forty micrometers of motion were surrounded in part by trabecular bone but also in part by fibrocartilage and fibrous tissue, and those that had been subjected to 150 micrometers of motion were surrounded by dense fibrous tissue. Trabecular microfractures were identified around three of the five implants that had been subjected to forty micrometers of motion and around four of the five that had been subjected to 150 micrometers of motion, suggesting that the ingrown bone had failed at the interface because of the large movements.The architecture of the surrounding trabecular bone also was altered by the micromotion of the implant. The implants that had stable ingrowth of bone were surrounded by a zone of trabecular atrophy, whereas those that had unstable ingrowth of bone were surrounded by a zone of trabecular hypertrophy. The trabeculae surrounding the fibrocartilage or fibrous tissue that had formed around the implants that had been subjected to forty or 150 micrometers of motion had been organized into a shell of dense bone tangential to the implant (that is, a neocortex outside the non-osseous tissue).CLINICAL RELEVANCE: The findings of the present study quantitate the in vivo patterns of bone ingrowth and remodeling that occur in association with different magnitudes of micromovement of porous-coated implants. Small movements (zero and twenty micrometers) are compatible with stable ingrowth of bone and atrophy of the surrounding trabecular bone, whereas larger movements (forty and 150 micrometers) result in less stable or unstable ingrowth of bone, the formation of fibrocartilage or fibrous tissue around the implant, and hypertrophy of the surrounding trabecular bone. This study not only quantified the magnitudes of relative micromotion that cause these different skeletal responses but also may help in the interpretation of radiographs of patients who have a porous-coated prosthesis.
Biological fixation of porous-surfaced implants plays a major role in contemporary total hip-replacement operations2-5,8,12. Growth of bone into the porous surface can provide the implant with long-term stability. However, achieving and maintaining such ingrowth remains difficult. Relative movement between the prosthesis and the bone is an important factor that influences bone ingrowth and remodeling. There is, however, a limited amount of data on the direct effects of known magnitudes of such movement on the responses of the surrounding bone.
Radiographic, histological, and biomechanical studies of implant-bone composites retrieved post mortem from experimental animals or from patients who had a total hip replacement without cement have provided some information regarding the skeletal responses to physiologically loaded implants3,10,14,15. Those studies demonstrated that ingrowth of bone provides porous-coated implants with excellent stability, with only a few micrometers of relative micromovement occurring at the bone-implant interface. In contrast, ingrowth of fibrous tissue provides less stability and may result in hundreds of micrometers of relative movement at the interface.
However, the effects of defined amounts of relative micromovement on ingrowth and remodeling of tissue apparently have not been quantitated previously in a controlled manner. Radiographic follow-up studies of the skeletal architecture around load-bearing implants in patients who had a prosthesis inserted without cement have shown altered patterns of remodeling, such as atrophy or hypertrophy of bone or the formation of radiodense lines or a pedestal, but the reason for these changes remains unknown2,8.
In the present study, cylindrical porous-coated implants were placed in the distal femoral metaphyses of dogs and were subjected to zero, twenty, forty, or 150 micrometers of oscillatory motion for eight hours a day for six weeks. By histological evaluation of sections of the implant-bone composite, it was possible to determine the skeletal responses to the different magnitudes of relative micromovement.
*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. Funds were received in total or partial support of the research or clinical study presented in this article. The funding source was the William H. Harris Foundation, Boston, Massachusetts.
†Orthopaedics Biomechanics Laboratory and Hip and Implant Surgery Unit, Department of Orthopaedic Surgery, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts 02114.
An in vivo loading device was specially designed to be rigidly attached to the canine femur with skeletal fixation pins and was used to apply a fixed amount of oscillatory rotational motion to cylindrical porous-coated implants that had been placed in the lateral femoral condyles of twenty skeletally mature foxhounds.
The porous-coated implants were made of a titanium-alloy (Ti-6Al-4V) core and commercially pure titanium fiber-mesh that was diffusion-bonded onto the core. The implants were eight centimeters long and had an outer diameter of 7.2 millimeters. The fiber mesh was one millimeter thick and covered a ten-millimeter-long section of one end of the implant. It was necessary to machine two flat sides on the implant in this region in order to press-fit the mesh before sintering.
The mechanical and electrical components of the in vivo loading device were housed in a rectangular box, measuring 127 by fifty-one by thirteen millimeters, on a platform that was fixed to the canine femur with external fixation pins (Fig. 1). A twelve-volt direct-current motor with an adjustable offset cam and a special lever arm was used to oscillate the implant at one hertz. The resulting displacement at the bone-implant interface could be set between zero and 250 micrometers.
The torques applied to the implant were measured indirectly by determining the strains in the lever arm that oscillated the implant. The displacement of the implant was measured at a fixed radius from the center of rotation of the implant by placing a one-centimeter-long projection on the assembly that held the implant. The amount of oscillation of the cylindrical implant was measured with use of a miniature, spring-loaded, Hall-effect displacement transducer (Micro Strain, Burlington, Vermont) that was accurate to within one micrometer; the transducer was held in the housing against the projection and was calibrated in angular degrees (Fig. 1). The linear motion at the outer surface of the fiber mesh was calculated with the equation: S = r x theta, where r = 3.6 millimeters (the radius of the implant) and theta is the angular displacement in radians.
The implantation procedure was performed with the dogs under general anesthesia. After the right hindlimb had been prepared and draped, the right lateral femoral condyle was exposed through a small incision over the epicondyle. An alignment jig was used to establish the position and orientation of the loading platform relative to the bone, and a threaded 3.2-millimeter-diameter pin was driven through the guide-hole in the platform and into the femoral condyle at a right angle to the long axis of the femur. The loading platform then was attached rigidly to the femoral diaphysis in the desired position with use of three 4.5-millimeter-diameter Schanz screws (Synthes USA, Paoli, Pennsylvania) that were drilled through the platform and into the femur. Set screws were used to secure the Schanz screws to the platform.
With the platform rigidly secured to the femur, a 6.4-millimeter-diameter hole was drilled into the lateral femoral condyle to a depth of twenty-five millimeters with use of location guides on the platform. The hole then was enlarged with a seven-millimeter-diameter side-cutting reamer with use of an additional guide to center the reamer within the hole. This provided for a diametrical interference of 0.2 millimeter for the 7.2-millimeter-diameter cylindrical implants.
A separate porous-coated implant that was ten millimeters long but was otherwise identical to the experimental implant was impacted into the deepest portion of the hole to serve as a motionless internal control. Then, the experimental implant was impacted into the hole so that the entire porous surface was surrounded by cancellous bone. This implant was attached to the bearing assembly. The lever arm and the motor then were assembled, and the housing was covered and sealed. The wounds were left open. Radiographs were made immediately postoperatively to verify the positions of the device and the fixation pins (Fig. 2).
The motor was powered eight hours a day for six weeks, starting twenty-four hours after the operation. A twelve-volt battery pack with custom-built timer circuitry powered the motor. The battery was housed in a custom-designed canine body jacket that was worn by the dogs. The dogs were allowed unrestricted activity, and the pin sites were cleaned daily. Data on torque and displacement were obtained with use of a custom-built data-acquisition system, and the apparent stiffness at the bone-implant interface was calculated in newton-meters per degree.
The dogs were divided into four groups of five dogs each, and the implants in each group were subjected to zero, twenty, forty, or 150 micrometers of oscillatory motion. In the first group, the platform was attached to the femur but the motor was not powered; the implants therefore were not subjected to motion. In the second group, an angular displacement of 0.3 degree produced twenty micrometers of motion at the bone-implant interface. In the third group, an angular displacement of 0.6 degree produced forty micrometers of motion at the interface. In the fourth group, an angular displacement of 2.0 degrees produced 150 micrometers of motion at the interface.
The dogs were killed at six weeks. The entire femur and the jig were retrieved. The femur then was sectioned in the sagittal plane just medial to the most medial border of the implant and perpendicular to the longitudinal axis of the implant with use of a high-speed, water-cooled diamond saw to expose the interface between the bone and the porous coating. The device then was powered, and the motion at the implant-bone interface was viewed under a high-resolution dissecting microscope and was recorded with use of a video camera that yielded an image with a magnification of 200 times. A calibrated ruler-grid was imposed on the image, and the motions that occurred within the implant, at the interface between the bone and the porous coating, and in the surrounding bone were measured.
The distal aspect of the femur, with the implant still in place, was fixed in 4 per cent buffered formalin, dehydrated in graded alcohols, and embedded in polymethylmethacrylate. It then was sectioned perpendicular to the longitudinal axis of the implant with use of a high-speed water-cooled diamond saw to yield ten serial sections, each one-half millimeter thick. Radiographs of the sections were made with a high-resolution technique. Alternate sections from both the experimental and the control implants then were fixed to flat acrylic slabs and were ground to a thickness of twenty micrometers with use of a ceramic disk grinder. The ground sections were stained with hematoxylin and eosin and were examined under a light microscope. The remaining sections were coated with gold and were examined under a scanning electron microscope that was equipped with a backscatter detector. Bone growth into the porous mesh was quantified with use of a computer-assisted image-analysis system7. The area density of bone next to and ten millimeters away from the porous coating also was measured with use of this image-analysis system. After quantification of bone ingrowth, these sections were prepared for histological evaluation by grinding, as has been described, and were stained with trichrome or toluidine blue.
All animals recovered uneventfully from the operation. There were no fractures or infections. All of the loading devices remained intact and functional throughout the experimental period.
With the calibrated grid, it was confirmed that the actual displacements of the implants were zero, twenty, forty, and 150 micrometers (in the absolute coordinate system) and that no slip had occurred at the external fixation frame or at the junctions between the Schanz screws and the bone. All of the deformation occurred either at the bone-implant interface or within the ten to twenty millimeters of bone surrounding the implant.
Evaluation of the motion at the bone-implant interface with use of the dissecting microscope demonstrated that trabecular bone was in intimate contact with the porous coatings of the implants that had been subjected to zero and twenty micrometers of oscillatory motion. When the implants were moved, the bone surrounding the porous coating was deformed and there was no relative motion between the implant and the adjacent bone. In contrast, when the implants in the third group were moved, there was forty micrometers of relative motion at the bone-implant interface in some areas and deformation of the surrounding bone in others. In marked contrast, when the implants in the fourth group were moved, there was 150 micrometers of relative motion along the entire implant-bone interface and no deformation of the surrounding trabecular bone.
Histological analysis demonstrated bone within the porous coatings of all of the implants, including those that had been subjected to 150 micrometers of movement (Figs. 3, 4 and 5). The bone within the porous coating was predominantly woven bone, but mature trabecular bone could be found in some areas. Quantitative analysis showed that the amount of bone within the porous coating, expressed as a percentage of the porous layer, was similar among the various groups; with the numbers available, no significant difference was found between any of the groups with respect to the amount of bone ingrowth. Furthermore, a paired t test showed no significant differences between the experimental and control implants with regard to the percentage of bone ingrowth (Table I).
There were differences among the groups, however, with regard to the histological appearance of the tissue within the porous coating. Trabecular bone and bone-marrow cells were found within the porous coatings of the implants that had been subjected to zero and twenty micrometers of motion. In many areas, the ingrown bone was in intimate contact with the titanium wires without an intervening layer of fibrous tissue. Some of the ingrown bone was mature and contained osteoid seams. Well developed marrow spaces also were evident. Fibrous tissue had formed only in the deepest layer of the porous coating, adjacent to the solid substrate (Fig. 3); the substrate was surrounded by a layer of fibrous tissue that was approximately fifty micrometers thick, with the collagen bundles arranged parallel to the substrate. In contrast, the porous coatings of the implants that had been subjected to forty micrometers of motion contained a mixture of trabecular bone, fibrocartilage, and fibrous tissue (Fig. 4). The porous coatings of the implants that had been subjected to 150 micrometers of motion contained trabecular bone and densely organized fibrous tissue (Fig. 5), but no marrow elements or fibrocartilage were found.
There also were marked differences among the groups with regard to the histological appearance of the interface between the porous coating and the surrounding trabecular bone. In the first two groups of implants (those that had been subjected to zero and twenty micrometers of motion), the bone within the porous coating was intimately connected to the surrounding bone. However, the implants that had been subjected to forty micrometers of motion were surrounded by a mixture of trabecular bone, fibrocartilage, and fibrous tissue; in some areas the ingrown bone was in continuity with the surrounding bone, whereas in other areas it was separated from the surrounding bone by the fibrocartilage or fibrous tissue. The implants that had been subjected to 150 micrometers of motion were entirely surrounded by a layer of dense fibrous tissue that was one to two millimeters thick. The collagen bundles in the fibrous tissue generally were tangential to the surface of the implant. It is important to note that, in most areas, the ingrown bone was not in continuity with the surrounding bone.
Trabecular microfractures were noted around three of the five implants that had been subjected to forty micrometers of motion and four of the five implants that had been subjected to 150 micrometers of motion. In the former group, the fractured trabeculae were bridged by fibrocartilage in some areas, which indicated an attempt at repair (Fig. 4). In the latter group, the fractured trabeculae had irregular edges and were bridged by dense fibrous connective tissue (Fig. 6). Some of the trabeculae were devitalized and were seen to be undergoing resorption and reinforcement with newly formed bone.
The patterns of remodeling in the region of trabecular bone surrounding the implants also had been altered dramatically. The implants that had been subjected to zero or twenty micrometers of motion were surrounded by an area of relative trabecular atrophy that extended five millimeters radial to the implants (Fig. 7). Some of the trabeculae were organized as radial spokes extending from the implant surface, but the general architecture of the trabecular bone was preserved in this region. Histomorphometric analysis showed that the area density of bone surrounding the porous coating was a mean (and standard deviation) of 36 ± 34 per cent less than that ten millimeters away from the porous coating.
In marked contrast, the implants that had been subjected to forty or 150 micrometers of motion were surrounded by an area of trabecular hypertrophy that extended five millimeters radial to the implants (Fig. 7). The area density of bone surrounding the porous coating was a mean (and standard deviation) of 78 ± 72 per cent greater than that ten millimeters away from the porous coating in these groups. Adjacent to the porous coating, the trabeculae were reoriented in a parallel direction with respect to the surface of the implants. The hypertrophy and reorientation of the trabeculae resulted in the formation of a dense shell of bone around the implant. This shell was separated from the porous coating by the fibrocartilage and fibrous tissue that surrounded the implants that had been subjected to forty and 150 micrometers of motion, respectively. The trabecular hypertrophy and reorientation, which were more pronounced in the latter group, represented an attempt to form a so-called neocortex outside the porous coating.
The trabecular bone surrounding the internal control implants (those that had not been attached to the loading apparatus) had the same histological features as that surrounding the implants that had been attached to the device but had not been subjected to motion.
The mean apparent torsional stiffness at the bone-implant interface immediately after the operation and six weeks postoperatively was 0.88 ± 0.25 and 1.25 ± 0.45 newton-meters per degree, respectively, for the implants that had been subjected to twenty micrometers of motion; 0.77 ± 0.43 and 0.54 ± 0.13 newton-meter per degree for the implants that had been subjected to forty micrometers of motion; and 0.25 ± 0.10 and 0.16 ± 0.10 newton-meter per degree for the implants that had been subjected to 150 micrometers of motion. The initial stiffness at the bone-implant interface of the implants that had been subjected to zero micrometers of motion was not tested as the motors were not powered during the study period. Six weeks postoperatively, the mean apparent torsional stiffness in this group was 0.99 ± 0.34 newton-meter per degree, which with the numbers available was not significantly different than that of the implants that had been subjected to twenty micrometers of motion. The implants that had been subjected to 150 micrometers of motion were significantly less stable than those that had been subjected to forty micrometers of motion (p < 0.0005), which in turn were less stable than those that had been subjected to twenty micrometers of motion (p < 0.005).
Although it generally is believed that relative micromovements at the bone-implant interface influence the growth of bone into and remodeling of bone around porous-surfaced implants, the exact relationship between such motion and the changes in the skeletal architecture has remained elusive because of the difficulty of simulating these effects in vivo. The special loading device used in the present study made it possible to apply controlled oscillatory micromovements to porous-coated implants in vivo and to evaluate the skeletal responses to these movements.
We found that twenty micrometers of oscillatory micromovement was compatible with stable ingrowth of bone and that such movement did not decrease the amount of bone that formed around the implant compared with the amount that formed around a stationary control. Movements of this magnitude were absorbed by the elastic deformation of the surrounding cancellous bone. Micromovements of forty micrometers were less compatible with stable ingrowth of bone; the implants that had been subjected to movements of this magnitude were surrounded by fibrocartilage or fibrous tissue in some areas and by bone in others. As there was almost no osseous continuity between the ingrown bone and the surrounding bone when the implants had 150 micrometers of movement, the threshold of micromovement that prevented stable ingrowth of bone was thus established for the conditions of this study.
The surprising finding, however, was that bone had formed within the porous coatings of all of the implants, including those that had been subjected to forty or 150 micrometers of motion. This bone was not always in continuity with the surrounding bone. This finding suggests that the bone within the porous coating had formed de novo rather than extending from the surrounding bone. Osteogenic cells may have infiltrated the porous coating during the initial healing phases immediately after the injury caused by the reaming, and progression along osteoblastic lines then may have occurred within the porous coating. Such cells also may have infiltrated the porous coating during the daily sixteen-hour period when the motor was not powered. These high levels of micromovement, however, were not compatible with the development of stable interfaces. Histological evaluation showed evidence of trabecular microfractures around some of the implants that had been subjected to forty or 150 micrometers of motion and bridging of these fractures by fibrocartilage or fibrous tissue. Although there was bone within the porous coatings of these implants, the apparent loss of connection between the ingrown bone and the surrounding bone suggested that the high levels of micromovement actually had disrupted the healing of the bone at the interface. There have been anecdotal reports of the failure of such bone in both animals and humans. Russotti et al., in a canine model, observed bone within the hydroxyapatite coating of some implants that had subsided and become loose by the time the animals were killed. In a previous study6, we observed bone within the porous coating of five femoral components that had been retrieved from patients who had a revision procedure because of radiographic and clinical evidence of loosening. Thus, it appears that, if a certain threshold of motion is exceeded, the bone-implant interface can fail and the implant can be enveloped by fibrous tissue even if bone already has formed within the porous coating. Such a failure may occur after an initially successful result if a patient participates in excessive physical activity or gains an unusual amount of weight.
The current study provided several insights into the early phases of bone-remodeling around porous-coated implants and the effects of micromovement on such remodeling. A zone of relative atrophy, extending several millimeters from the implant, developed around the implants that had not been subjected to micromovement, whereas a zone of relative hypertrophy, also extending several millimeters from the implant, developed around the implants that had been subjected to a high level of micromovement (forty or 150 micrometers). Engh et al.3 evaluated serial radiographs of patients who had had a total hip arthroplasty without cement and classified the implants, on the basis of certain radiographic features, as having ingrowth of bone or ingrowth of fibrous tissue. They suggested that ingrowth of bone leads to stress-related bone loss, whereas ingrowth of fibrous tissue does not. Our findings support their interpretation. We also noted that the adjacent trabeculae were oriented perpendicular to the implants that had stable ingrowth of bone, which is consistent with the so-called trabecular streaming observed on clinical radiographs of patients who have a porous-coated implant.
The region of trabecular hypertrophy that was observed around the implants that had been subjected to 150 micrometers of motion was of particular interest. The trabecular architecture also had been altered considerably in this group, with the trabeculae adjacent to the fibrous layer having been reorganized into a dense shell of new bone parallel to the surface of the implant. Radiodense lines that are separated from an implant by a one or two-millimeter-wide radiolucent area, as sometimes are observed on follow-up radiographs, raise concerns about prosthetic stability. Such radiodense lines most likely are related to reorientation and condensation of the trabeculae, as was observed in the current study. If that is the case, it is likely that the implant is surrounded by fibrous tissue. The reason for the formation of the dense shell of bone is speculative, but it may be related to the local stress states. A dense shell of new bone, without an intervening layer of fibrous tissue, also has been found around stable implants inserted with cement9. The reason for the trabecular hypertrophy also is speculative, but it may be related to local fractures of the trabeculae and reinforcement with new bone. Local trabecular hypertrophy occasionally is observed around the non-porous-coated distal end of partially porous-coated components; high stresses and trabecular damage may be involved in such instances as well.
Extrapolation of the data from the canine model used in the present study to humans must be done with caution. We made some assumptions in this study. For example, we subjected the implants to torsional loading, which resulted in shear forces at the interface. Loading of femoral components inserted without cement may be more complex. However, femoral components usually have greater instability under conditions such as stair-climbing1,3,9, during which they are subjected to torsional loading. Also, we powered the motors for only eight hours each day in order to simulate the average loading of a joint prosthesis, even though activity levels may vary widely among different patients. We measured the apparent stiffness at the interface by dividing the torque by the displacement. This measurement does not truly represent the stiffness of the interface, which may depend on the compliance of the adjacent bone or soft tissues. The bone in the distal femoral metaphysis may have different mechanical properties than the bone in the endosteal canal of the femur, which also could influence the apparent stiffness. However, Ramamurti et al. showed that, within a certain range, the elastic modulus of trabecular bone had no effect on the slip characteristics of porous-coated implants.
Despite these limitations, the micromovements that we observed are consistent with those observed in postmortem retrieval studies. In a study of fourteen femora that had been retrieved post mortem from patients who had had a total hip arthroplasty without cement twelve to ninety-three months earlier, Engh et al.3 investigated the movement of implants by simulating single-limb-stance and stair-climbing loads in vitro with use of a special loading fixture. They found that the relative motion between the implant and the porous coating was less than forty micrometers in the thirteen femora that had ingrowth of bone. The relative motion of the one implant that did not have ingrowth of bone was 150 micrometers. Their findings are consistent with those of the current study, in which different magnitudes of motion were induced in a controlled fashion. Thus, it appears that micromovements of forty micrometers or less are compatible with complete or partial ingrowth of bone, whereas micromovements of 150 micrometers prevent osseous stability.
Burke, D. W.; O'Connor, D. O.; Zalenski, E. B.; Jasty, M.; and Harris, W. H.: Micromotion of cemented and uncemented femoral components. J. Bone and Joint Surg.,73-B(1): 33-37, 1991.73-B(1)33
1991
Engh, C. A.; Bobyn, J. D.; and Glassman, A. H.: Porous-coated hip replacement. The factors governing bone ingrowth, stress shielding, and clinical results. J. Bone and Joint Surg.,69-B(1): 45-55, 1987.69-B(1)45
1987
Engh, C. A.; O'Connor, D.; Jasty, M.; McGovern, T. F.; Bobyn, J. D.; and Harris, W. H.: Quantification of implant micromotion, strain shielding, and bone resorption with porous-coated anatomic medullary locking femoral prostheses. Clin. Orthop.,285: 13-29, 1992.28513
1992
[PubMed]
Galante, J.; Rostoker, W.; Lueck, R.; and Ray, R. D.: Sintered fiber metal composites as a basis for attachment of implants to bone. J. Bone and Joint Surg.,53-A: 101-114, Jan. 1971.53-A101
1971
Jasty, M., and Harris, W. H.: Observations on factors controlling bony ingrowth into weight-bearing, porous, canine total hip replacements. In Non-Cemented Total Hip Arthroplasty, Bristol-Myers/Zimmer Orthopaedic Symposium, pp. 175-189. Edited by R. H. Fitzgerald, Jr. New York, Raven Press, 1988.
Jasty, M.; Bragdon, C. R.; Maloney, W. J.; Haire, T.; and Harris, W. H.: Ingrowth of bone in failed fixation of porous-coated femoral components. J. Bone and Joint Surg.,73-A: 1331-1337, Oct. 1991.73-A1331
1991
Jasty, M.; Bragdon, C. R.; Schutzer, S.; Rubash, H.; Haire, T.; and Harris, W. H.: Bone ingrowth into porous coated canine total hip replacements. Quantification by backscattered scanning electron microscopy and image analysis. Scan. Microsc.,3: 1051-1057, 1989.31051
1989
Lord, G., and Bancel, P.: The Madreporic cementless total hip arthroplasty. New experimental data and a seven-year clinical follow-up study. Clin. Orthop.,176: 67-76, 1983.17667
1983
[PubMed]
Maloney, W. J.; Jasty, M.; Burke, D. W.; O'Connor, D. O.; Zalenski, E. B.; Bragdon, C.; and Harris, W. H.: Biomechanical and histologic investigation of cemented total hip arthroplasties. A study of autopsy-retrieved femurs after in vivo cycling. Clin. Orthop.,249: 129-140, 1989.249129
1989
[PubMed]
Pilliar, R. M.; Lee, J. M.; and Maniatopoulos, C.: Observations on the effect of movement on bone ingrowth into porous-surfaced implants. Clin. Orthop.,208: 108-113, 1986.208108
1986
[PubMed]
Ramamurti, B. S.; Orr, T. E.; DiGioia A. M. III; Bragdon, M.; Jasty, M.; and Harris, W. H.: An investigation of the implant/bone interface of an in-vivo micromotion canine experiment using finite element analysis. Trans. Orthop. Res. Soc.,19: 244, 1994.19244
1994
Rosenberg, A., and Galante, J.: Cementless total hip replacement. In The Hip and Its Disorders, pp. 971-1006. Edited by M. E. Steinberg. Philadelphia, W. B. Saunders, 1991.
Russotti, G. M.; Okada, Y.; Fitzgerald, R. H., Jr.; Chao, E. Y. S.; and Gorski, J. P.: Efficacy of using a bone graft substitute to enhance biological fixation of a porous metal femoral component. In The Hip. Proceedings of the Fourteenth Open Scientific Meeting of The Hip Society, pp. 120-154. St. Louis, C. V. Mosby, 1987.
Turner, T. M.; Sumner, D. R.; Urban, R. M.; Rivero, D. P.; and Galante, J. O.: A comparative study of porous coatings in a weight-bearing total hip-arthroplasty model. J. Bone and Joint Surg.,68-A: 1396-1409, Dec. 1986.68-A1396
1986
Walker, P. S.; Schneeweis, D.; Murphy, S.; and Nelson, P.: Strains and micromotions of press-fit femoral stem prostheses. J. Biomech.,20: 693-702, 1987.20693
1987
[PubMed]