.
`United States Patent
`
`[19]
`
`IIIIIIIllllllllllllllllllllIllllllllllllllllllllllllIllllllllllllllllllllll
`U8005282861A
`[11] Patent Number:
`
`5,282,861
`
`Kaplan
`
`[45] Date of Patent:
`
`Feb. 1, 1994
`
`75
`
`[56]
`
`[54] OPEN CELL TANTALUM STRUCI‘URES
`FOR CANCELLOUS BONE IMPLANTS AND
`'
`CELL AND TISSUE RECEPTORS
`luv to : Richard B. K 1
`, B
`I
`en r
`Calif.
`up u ever y
`1
`I
`[73] Assignee: mmet’ Pacoima, Calif
`21
`l.
`.:
`,
`8
`I
`] App No
`850 11
`[22] Filed:
`MEL 11, 1992
`,
`_
`[51]
`Int. Cl.5 .......................... A61F 2/28, A61F 2/02,
`A61F 2/54; AOIN 1/02
`[52] US. Cl. ........................................ 62632/31/6gé6222/71/12;
`[58] Field of Search ................. 623/16 11 667427/2-
`433 173’ 201.1, 2022, 206, 207
`/
`References Cited
`U.S. PATENT DOCUMENTS
`3,677,795
`7/1972 Bokros et al. ........................... 623/2
`3992 725 11/1976 Homsy
`623/11
`
`4,392,828
`7/1933 Ehrnford .....
`433/201.1x
`4,457,984 7/1984 Otani et a1.
`....................... 433/201.1
`4,737,411
`4/1988 Graves, Jr. et a1.
`.............. 623/16 X
`
`fill ,
`1
`5’
`
`4,775,598 10/1988 Jaeckel ................................ 428/550
`7/1989 von Recum et a1.
`.
`.. 427/2 X
`4,846,834
`
`7/1991 Ducheyne ............. 623/16
`5,030,233
`5,152,794 10/1992 Davidson .............................. 623/16
`FOREIGN PATENT DOCUMENTS
`2093701
`9/1982 United Kingdom .................. 623/16
`Primary Examiner—David Isabella
`.
`Assistant Examiner—Dinh X. Nguyen
`Attorney, Agent, or Firm—Charles H. Schwartz;
`Ellsworth R. Roston
`
`[57]
`ABSTRACT
`A tantalum open cell structure is formed by chemical
`“PM “13°51‘10“ °nt° *1 ”mula‘ed carbon ‘0“ “’3'
`strate. Tantalum has a long history of use as an implant
`material, in both bone and soft tissue. This lightweight,
`strong, porous structure, mimicking the microstructure
`of natural cancellous bone, acts as a matrix for the in-
`corporam" 0f b°“e’ prov‘dmg Opt‘mal Pemeab‘hty
`and a high surface area to encourage new bone in-
`STOWth‘
`
`18 Claims, 4 Drawing Sheets
`
`100
`
`{00
`
`I02
`
`102
`
`Page 1 of 11
`
`ZIMMER EXHIBIT 1020
`
`ZIMMER EXHIBIT 1020
`
`Page 1 of 11
`
`

`

`US. Patent
`
`Feb. 1, 1994
`
`» Sheetlof4
`
`V
`
`5,282,861
`
`[00
`
`[02
`
`F/G.
`
`/
`
`Page 2 ofll
`
`Page 2 of 11
`
`

`

`US. Patent
`
`Feb. 1,1994
`
`Sheet 2 of 4
`
`5,282,861
`
`[00
`
`F/G. 2
`
`Page 3 ofll
`
`Page 3 of 11
`
`

`

`US. Patent
`
`Feb. 1, 1994
`
`Sheet 3 of 4
`
`5,282,861
`
`
`
`[04
`
`104
`
`F/G. 3
`
`Page 4 ofll
`
`Page 4 of 11
`
`

`

`US. Patent
`
`Feb. 1, 1994
`
`Sheet 4 of 4
`
`5,282,861
`
`FIG. 4
`
`GASES IN
`
`(H2) (CI2)
`
`2/4
`
`
`
`REACT/0N
`CHAMBER
`200
`
`RES/537g?
`205
`
`TANTALUM
`”5 7’“
`2’0
`
`CHLOR/NATION
`CHAMBER
`202
`
`FOAM SUBSTRATE
`2/2
`
`GRAPH/ TE
`I
`_ — SUSCEPTOR
`
`(HOT WALL FURNACE)
`
`
`
`
`
`[II/A
`I/I/I/IA
`
`O 204
`O
`INDUCTION
`II/I
`HEATING COIL —>O
`
`
`208
`/[
`_//1
`//////
`
`W
`
`
`00000
`
`7/
`
`
`
`O O O O
`
`
`
`222
`
`EXHAUST
`
`Page 5 ofll
`
`Page 5 of 11
`
`

`

`1
`
`5,282,861
`
`OPEN CELL TANTALUM STRUCTURES FOR
`CANCELLOUS BONE IMPLANTS AND CELL AND
`TISSUE RECEPTORS
`
`BACKGROUND OF THE INVENTION
`
`The need for a cancellous bone substitute and/or cell
`and tissue receptive material is significant. For example,
`cancellous autografts provide a porous framework
`within which revascularization occurs and against
`which new bone is layered, and also provide a popula-
`tion of osteoprogenitor cells and a complement of bone
`growth-inducing factors. Grafting, however, requires
`surgery to obtain the material, and a viable substitute is
`desirable. It is here that the concept of artificial biocom-
`patible implants becomes of interest. Extensive studies
`over the last two decades have shown that to duplicate
`the success of cancellous grafts, an implant should serve
`as a porous framework. Indeed, early research demon-
`strated that an interconnected porous material is toler-
`ated by the body, and encourages new bone growth,
`better than the same material in solid form.
`The replacement of diseased, destroyed, or degenera-
`tive bone and tissues consumes time and financial re-
`sources from a large segment of the surgical commu-
`nity, in both medicine and dentistry. Clinical and scien-
`tific work is directed at facilitating regeneration of tis-
`sues in affected patients so that normal biomechanical
`and physiologic functions can resume. In some patients,
`full restoration of function with normal tissue is achiev-
`able, while in others, prostheses are biologically at-
`tached to restore function. The specialty science de-
`voted to the study of substances utilized for implants in
`medicine and dentistry, biomaterials,
`is a young field
`that has taken tremendous strides in the last 20 years.
`Over the same period, dental implantology has evolved
`from early attempts by a few enthusiasts to a fully rec-
`ognized branch of dentistry.
`Although indispensable for survival, the body’s natu-
`ral defense mechanisms, by which materials identified
`as nonself are rejected, have been the nemesis of sur-
`geons using prostheses or implantable devices. It
`is
`necessary to minimize the rejection mechanism as much
`as possible. Certain biomaterials have been identified
`having apparently limited reactions to the body’s de-
`fense mechanisms. These materials can be placed on a
`continuum that extends from relatively chemically reac-
`tive to completely nonreactive or passive. Generally,
`the more nonreactive the material is in vivo, the better
`the performance that can be expected.
`Matching the requisite biomechanical requirements
`for an implant with the environment of surrounding
`tissues has been a formidable challenge. Significant
`progress was made in resolving this problem in the early
`19705, when the importance of porosity was first recog-
`nized. Later work showed that certain physical parame-
`ters of the porosity affect the type of tissue and the rate
`of ingrowth. The degree of interconnectivity and the
`nominal pore size were found to be critical factors in
`determining the success of an implant. Maximum inter-
`connectivity, or the absence of “dead ends”, was found
`to facilitate ingrowth. These studies showed that pore
`sizes less than 10 um prevent ingrowth of cells; pore
`sizes of 15-50 pm encourage fibrovascular ingrowth;
`pore sizes to 50—150 um result in osteoid formation; and
`pore sizes of greater than 150 um facilitate the ingrowth
`of mineralized bone.
`
`10
`
`15
`
`20
`
`25
`
`30
`
`35
`
`45
`
`50
`
`55
`
`65
`
`Page 6 ofll
`
`2
`Bone ingrowth into the voids of a porous material
`provides ideal skeletal fixation for the permanent im~
`plants used for the replacement of bone segments lost
`due to any number of reasons, or in total joint prosthe-
`ses. Biological compatibility, intimate contact with the
`surrounding bone, and adequate stability during the
`early period of bone ingrowth have been identified as
`important requirements, along with proper porosity.
`The optimal porous material should have good crack
`resistance, particularly under impact, and a compliance
`comparable to that of bone. The material should also
`make the manufacture of implants of precise dimensions
`easy, and permit the fabrication of either thick or thin
`coatings on load-bearing cores.
`One prerequisite for successful ingrowth is that the
`implant be placed next to viable bone. In fact, the pres-
`ence of bone within the implant has become presump-
`tive evidence of osteoconductive properties: that is, the
`ability of bone to grow into a porous structure when the
`structure is placed next to bone. Initially, the cells that
`interface the implant convert to bone, then the front of
`regenerated bone progresses into the implant. This pro—
`cess is known as osseointegration, meaning the achieve-
`ment of direct contact between living bone and implant.
`The research, development, and manufacture of syn-
`thetic porous implants having the physical properties
`required to promote bone ingrowth have proved to be a
`major endeavor. Implants with porous surfaces of me-
`tallic, ceramic, polymeric, or composite materials have
`been studied extensively over the last two decades. A
`significant early advance in this area was made with the
`development of “replamineform” materials, so termed
`because they replicate actual life forms. These materials
`are based on the three-dimensional microstructure of
`certain marine invertebrates (best represented by corals
`and echinoids), which is uniform and completely per-
`meable. The replamineform process utilizes the inverte-
`brate microstructure as a template to make porous
`structures of other materials.
`The most commonly used substance for porous bio-
`materials is calcium hydroxyapatite (HA), which is the
`largest chemical constituent of bone. Other nonmetallic
`materials frequently used in porous form for implants
`include the ceramics tricalcium phosphate (TCP), cal-
`cium aluminate, and alumina, carbon; various polymers,
`including polypropylene, polyethylene, and polyoxy-
`methylene (delrin); and ceramic-reinforced or -coated
`polymers. Unfortunately, ceramics, while strong, are
`very brittle and often fracture readily under loading;
`and polymers, while possessing good ductility, are ex-
`tremely weak. The very nature of these materials can
`restrict their clinical dental and orthopedic applications.
`Metals, on the other hand, combine high strength and
`good ductility, making them attractive candidate mate
`rials for implants (and effectively the most suitable for
`load-bearing applications). Many dental and orthopedic
`implants contain metal, most often titanium or various
`alloys such as stainless steel or vitallium (cobalt-chromi-
`um-molybdenum). Ceramic-coated metals are also used,
`typically HA or TCP on titanium. Additionally, a large
`variety of metals are used internally in biomedical com-
`ponents such as wire, tubing, and radiopaque markers.
`Many existing metallic biomaterials, however, do not
`easily lend themselves to fabrication into the porous
`structures that are most desirable for bone implants.
`These materials (e.g. stainless steel, cobalt-based alloys)
`exhibit the necessary properties and biocompatibility as
`long as only a smo‘oth, bulk shape in a metallurgically
`
`Page 6 of 11
`
`

`

`5,282,861
`
`3
`perfect state is needed. The machining or other treat-
`ment needed to obtain a porous or surface-textured
`shape for interlocking with skeletal tissue can have a
`detrimental effect on the properties and biocompatibil-
`ity, and can even result in material failure. For example,
`the hexagonal crystal structure of titanium makes it
`susceptible to cracks and fractures, as has been seen in
`the case of dental implants. Some porous metallic mate-
`rials (e.g. flame- or plasma-sprayed titanium, porous
`sintered powder metallurgy materials) do not match the 10
`structure of cancellous bone sufficiently well to ensure
`successful ingrowth and integration. Also, most metals
`and alloys currently in use are subject to some degree of
`corrosion in a biological environment. Finally, the high
`densities of metals can make them undesirable from a 15
`weight standpoint.
`'
`SUMMARY OF THE INVENTION
`
`5
`
`4
`lum (atomic number 73, atomic weight 180.95, density
`16.68 g/cm3) is a transition element (periodic group
`VB), a highly refractory (melting point 2996° C.),
`strong, ductile metal with excellent oxidation and cor-
`’ rosion resistance. These properties led to its early inves-
`tigation, in both animal and human experiments, as a
`potential human implant material. Early evidence of
`excellent tissue acceptance, combined with low corro-
`sion, has led to the use of tantalum as a surgical implant
`material and its use in a variety of applications, includ-
`ing pacemaker electrodes, wire, foil and mesh for nerve
`repair, cranioplasty plates, contrast media for airwave
`radiographic studies, radiopaque markers for following
`bone growth, ligation clips, and more recently on an
`experimental basis in femoral endoprostheses.
`The crystal structure of tantalum is body-centered
`cubic, giving it excellent ductility due to the six possible
`slip planes. It is so corrosion-resistant that it resists the
`attack of most chemical agents; tantalum pacemaker
`electrodes have exhibited excellent corrosion resistance
`both in vitro and in vivo. This inertness likely accounts
`for the good tissue compatibility of the base metal as
`well, whereas a noble metal such as gold, though con-
`sidered corrosion-resistant, is not sufficiently biocom-
`patible due to its catalytic surface.
`Comparative studies have demonstrated that tanta-
`lum does not inhibit cell growth and indeed becomes
`tightly enveloped by new osseous tissue soon after im-
`plantation, whereas dental gold and cobalt-based alloys
`can inhibit cell growth and cause bone resorption. With
`tantalum, osseous ingrowth has been demonstrated
`right up to and into implants. Complete, strong, long-
`term osseointegration has been demonstrated with tan-
`talum implants in both dental and orthopedic applica-
`tions, under both unloaded and heavily loaded condi-
`tions, for implantation periods as long as eight to twelve
`years.
`In addition, tantalum has an elastic modulus close to
`that of bone, much closer than any of the other high-
`strength metals and alloys commonly used for implants;
`this too may well contribute to the favorable reaction
`with bone. With its greater ductility, excellent corro-
`sion resistance, good workability, and demonstrated
`biocompatibility, tantalum clearly can be regarded as an
`excellent alternative to the metals and alloys presently
`in use and under development for bone implants.
`BRIEF DESCRIPTION OF THE DRAWINGS
`
`FIG. 1 is a perspective view of an open cell tantalum
`structure constructed in accordance with the present
`invention;
`FIG. 2 is an enlarged view of the surface of the tanta-
`lum structure of FIG. 1;
`FIG. 3 is a detailed view of small sections of the
`material of FIGS. 1 and 2; and
`FIG. 4 is illustrative of one method of making the
`tantalum structure of the present invention.
`DESCRIPTION OF THE PREFERRED
`EMBODIMENT
`
`Cancellous, or spongy, bone is composed of a porous
`space-frame structure formed of open spaces defined by
`interconnected trabeculae, oriented along lines of prin-
`cipal stresses. At the microstructural level, the trabecu-
`Iae are composed of layers of lamellar bone. Cancellous
`bone has anisotropic mechanical properties, i.e. differ-
`ent structural behavior along different orientations.
`Along the axis of the major channels, cancellous bone
`
`New materials are enabling the design of innovative,
`and increasingly biocompatible, replacements for dam- 20
`aged human tissues. In the present invention, reticulated
`open cell carbon foam is infiltrated with tantalumby the
`chemical vapor deposition (CVD) process. It should be
`noted that niobium, which has similar chemical and
`mechanical properties to tantalum, may also be used as 25
`well as appropriate alloys of tantalum and niobium. For
`example, other metals such as niobium, hafnium and/or
`tungsten could be alloyed with the tantalum or hafnium
`and/or tungsten with niobium to change modulus and-
`/or strength. Therefore, any reference to tantalum 'is 30
`not meant to be an exclusion of other metals.
`
`The carbon foam is infiltrated by chemical vapor
`deposition (CVD). The resulting lightweight, strong,
`porous structure, mimicking the microstructure of natu-
`ral cancellous bone, acts as a matrix for the incorpora— 35
`tion of bone or reception of cells and tissue. The pores
`of the matrix are connected to one another to form
`continuous, uniform channels with no dead ends. This
`intricate network of interconnected pores provides op-
`timal permeability and a high surface area to encourage 40
`cell and tissue ingrowth, vascularization, and deposition
`of new bone.
`
`The result is a new biomaterial that, when placed next
`to bone or tissue, initially serves as a prosthesis and then
`functions as a scaffold for regeneration of normal tis- 45
`sues. The new biomaterial fulfills the need for an im-
`plant modality that has a precisely controllable shape
`and at the same time provides an optimal matrix for cell
`and bone ingrowth. Additionally, the physical and me-
`chanical properties of the porous metal structure can be 50
`specifically tailored to the particular application at
`hand. This new implant offers the potential for use in
`alveolar ridge augmentation, periodontics, and orthog-
`nathic reconstruction. As an effective substitute for
`autografts, it will reduce the need for surgery to obtain 55
`those grafts. It is useful in orthopedic applications as
`well.
`The present invention may also be used for tooth
`replacement because of the ability to induce tissue and
`bone growth even in the face of mildly infectious condi- 60
`tions. For example, an artificial tooth can be joined to
`an open cell tantalum stem and positioned in an appro-
`priately sized hole in the jaw. The gum is allowed to
`rest against the artificial tooth and some of the stem to
`form a seal.
`Tantalum was selected as the material of choice based
`on its good mechanical properties, excellent corrosion
`resistance, and demonstrated biocompatibility. Tanta-
`
`65
`
`Page 7 ofll
`
`Page 7 of 11
`
`

`

`5
`exhibits elastic behavior with sudden brittle failure at
`
`5,282,861
`
`5
`
`ultimate load in tension. When loaded with a tensile
`force whose line of action is skewed with respect to the
`channel axis of the bone, the stress-strain curve is para-
`_ bolic with plastic deformation and greater energy ab-
`sorption. It is therefore stiffer (has higher tensile and
`compressive moduli) but fails at a lower strain when
`loaded parallel to the predominant spicular direction
`than when loaded in other directions. These properties
`are important because they serve to absorb shock and 10
`distribute load in the vicinity of the articular surfaces of
`joints.
`Any material to be used as a substitute for cancellous
`bone must therefore allow elastic deformation and load
`distribution. In addition, the material must not produce 15
`load concentrations, particularly if placed close to the
`underlying surface of articular cartilage, which might
`increase the local stresses on the articular surface and
`lead to wear and damage of the surface.
`Cancellous bone demonstrates remodeling behavior 20
`according to Wolff‘s Law: that is, with the form being
`given, bone adapts to the loads applied to it. The con-
`verse is also true, and equally important: where loads
`are not applied, bone tends to resorb. An implant, then,
`must distribute stresses throughout its structure,
`the 25
`ingrowing bone, and the surrounding bone in order to
`avoid bone resorption and weakening caused by stress
`shielding.
`The density of cancellous bone is 0.7 g/cm3; its tensile
`modulus 0.2—0.5 GPa; tensile strength 10—12 MPa; and 30
`strain to failure 5-7%. Compared to cortical bone, can-
`cellous bone is é-i as dense (indicating its porous na-
`ture); l/lO—l/ZO as stiff; and five times as ductile. The
`mechanical properties of the two types, though, actu-
`ally represent a continuum, reflecting the behavior of a 35
`relatively uniform material (bone) modified by differ-
`ences in density and structure.
`Based on experiments with hydroxyapatite implants,
`ingrowth and maturation of new bone are more rapid
`from a cancellous bone region than from cortical bone, 40
`with the tissue-implant interface reaching peak shear
`strength in dogs in 8 weeks. The process may take
`longer in humans, with remodeling still possible up to 2
`years postoperation. Inadequate device designs may
`produce continued stress shielding remodeling as long 45
`as 9—10 years postoperation.
`Materials for osseous, or bone, implants must be rigid
`and stress-resistant, while avoiding self-concentration
`of stresses that result in stress shielding. Also, osseous
`implants should ideally reside in the bone without inter- 50
`fering with bone remineralization, the natural process
`by which the body replenishes bone. The implant
`should be able to be precisely shaped and placed for
`optimal interface and performance. Finally, non-resorp-
`tion would be a beneficial quality for implants used in 55
`load-bearing applications, and/or those in which com-
`plete bone ingrowth is not possible.
`Critical to the performance of a porous implant is the
`completeness of its interconnectivity. This is essential
`because constrictions between pores and isolated, dea- 60
`dend pockets can limit vascular support to ingrowing
`tissues; ischemia of the ingrowing bone cells results in
`failure of the implant. Incomplete vascularization or a
`reduction in the neovascularity also makes an implant
`vulnerable to bacterial colonization. Implants lacking 65
`completely interconnected porosity-can also result in
`aberrant mineralization, stress shielding,
`low fatigue
`strength, and/or bulk displacement.
`
`Page 8 ofll
`
`6
`The open cell metal structure of the present invention
`offers highly interconnected, three-dimensional poros-
`ity that is uniform and consistent, a structure exception-
`ally similar to that of natural cancellous bone. In this
`way it is superior to other porous metallic implant mate-
`rials, whose “porosity” is artificially produced via some
`form of surface treatment that does not result in a truly
`complete, open porosity. Examples of these methods
`include macroscopic porous coatings (e.g. metal micro-
`spheres or wires sintered or otherwise attached to a
`bulk surface); microscopic surface porosity (e.g. metal
`powder particles flame- or plasma-sprayed onto a bulk
`surface); and controlled surface undulations machined
`into a bulk surface.
`Although certain porous ceramic materials do offer
`full porosity (e.g. the replamineform process for hy-
`droxyapatite), they have properties inferior to metals as
`discussed previously. The open cell metal structure is
`osteoconductive, like other porous implants. Also, it is
`entirely biocompatible, based on the demonstrated bi-
`ocompatibility of tantalum.
`Allowing full mineralization is another extremely
`important property required of bone substitute materi-
`als. The highly organized process of bone formation is a
`complex process and is not fully understood. There are,
`however, certain prerequisites for mineralization such
`as adequate pore size, presumably larger than 150 pm
`with interconnect size in the range of 75 pm. A pore
`diameter of 200 um corresponds to the average diame-
`ter of an osteon in human bone, while a pore diameter of
`500 um corresponds to remodeled cancellous bone. The
`open cell metal structures of the present invention can
`be fabricated to virtually any desired porosity and pore
`size, and can thus be matched perfectly with the sur-
`rounding natural bone in order to provide an optimal
`matrix for ingrowth and mineralization. Such close
`matching and flexibility are generally not available with
`other porous implant materials.
`One concern with an implant must be the potential
`for stress shielding. According to Wolff‘s law, bone
`grows where it
`is needed (that is, where there is a
`stress). Stress on a bone normally stimulates that bone to
`grow. With an implant, it is primarily the stress/strain
`field created in the tissue around an implant that con-
`trols the interface remodeling. Stress shielding occurs
`when an overly stiff implant carries stresses that were
`previously applied to the bone in that area; it can result
`in inhibition of mineralization and maturation of the
`ingrowing bone, and/or the resorption of existing natu-
`ral bone.
`
`An implant, then, must distribute stresses throughout
`its structure, the ingrowing bone, and the surrounding
`bone in order to avoid bone resorption and weakening
`caused by stress shielding. Because metals are stronger
`than natural bone, this would seem to be a concern with
`a metallic implant in that the implant would itself focus
`and bear direCtly the majority of local loads and stresses
`that would ordinarily be placed on the bone, thus de-
`priving both the existing and new bone of those forces
`which, in effect, help keep it at optimal strength.
`The unique structure and properties of the open cell
`metal structures of the present
`invention, however,
`avoid this drawback altogether. The deposited thin
`films operate as an array within the porous metal body,
`contributing their exceptional mechanical properties to
`the structure at large. One result of this effect is that
`imposed loads are distributed throughout the body. In
`the case of a open cell metal bone implant, stresses are
`
`Page 8 of 11
`
`

`

`_
`
`5
`
`7
`distributed into both the ingrowing new bone and the
`surrounding existing bone as well, thereby providing
`both the old and new bone with the normal, healthy
`forces they require.
`In fact, with the ability to finely tailor the open cell
`metal structure’s properties during the fabrication pro-
`' cess, an implant can be designed to distribute stresses in
`a given direction(s), depending on the needs of the
`specific application at hand. The bonding of regener-
`ated bone to the implant also helps to transfer stresses 10
`directly to the bone in and around the implant; this
`sharing of biofunction is a consequence of the compos-
`ite nature of the implant/bone structure. The advantage
`of these metal structures over other porous implant
`materials is especially strong in this area. Ceramics lack 15
`sufficient mechanical properties to begin with, and no
`current implant material, either ceramic or metallic,
`possesses the unique properties of the metal structure as
`described here.
`In the present invention, useful lightweight refrac- 20
`tory structures are made by the chemical vapor deposi-
`tion (CVD) of a small amount of metallic material such
`as tantalum or niobium (or combination of these materi-
`als with other materials to form alloys) into a reticulated
`(porous) vitreous carbon foam. The density of the resul- 25
`tant body is purposely maintained at substantially below
`full density, resulting in a structure with extremely
`favorable properties. The basic approach involves the
`use of a low-density carbon foam, which is infiltrated
`with the desired material by CVD to provide uniform 3O
`thin films on all ligaments. These thin films provide
`exceptional strength and stiffness to the ligaments, with
`the expenditure of very little weight. Thin CVD films
`can provide much higher mechanical properties than
`can bulk materials. Such quasi-honeycomb materials 35
`have remarkably high specific strength and stiffness.
`This process does not endeavor to densify the body
`fully, although it is possible to do so, and useful parts
`can be so fabricated. In the present invention, only thin
`films are deposited on the interior surfaces of the vitre- 40
`ous carbon foam,
`taking advantage of the apparent
`unusual mechanical properties of the thin films which,
`when operating as an array in the body as a whole,
`produce unusual properties for the entire body. Using a
`porous carbon with extremely high porosity and small 45
`pore size takes advantage not only of the properties of
`thin films, but of short beams as well.
`It is important to note that the structural integrity of
`the fabricated structure is provided by the deposited
`thin films themselves, rather than by the carbon foam 50
`substrate. These films have much higher moduli of elas-
`ticity than do the thin sections of vitreous carbon in the
`foam substrate. Because the deposited films are so thin
`and short, they show great strength, not unlike the high
`strength experienced in very fine fibers or filaments. 55
`Their support of the mechanical load ensures that fail-
`ure does not occur in the carbon.
`The open cell metal structures of the present inven-
`tion are fabricated using the tantalum metal film and
`carbon substrate combination, with the film deposited 60
`by CVD, to form the structure shown in FIG. 1 which
`mimics bone closely in having open spaces 100 intercon-
`nected by ligaments 102. With the variables available in
`both the materials and the fabrication process, it is possi-
`ble to obtain the simultaneous optimization of multiple 65
`properties (e.g. strength, stiffness, density, weight) for
`the given application of substitution for bone. FIGS. 2
`and 3 are scanning electron photomicrographs showing
`
`5,282,861
`
`8
`the ligamental structure of the metal-infiltrated reticu-
`lated carbon foam and an individual coated ligament in
`cross-section, respectively. In FIG. 3 it can be seen that
`each ligament is formed by a carbon core 104 covered
`by a thin film 106 of metal such as tantalum, niobium or
`alloys of each.
`_
`Another major advantage of the open cell metal
`structure of the present invention is that it is readily
`shapeable to nearly any configuration, simple or com-
`plex, simply by shaping the raw carbon substrate prior
`to metal infiltration. This facilitates exact contouring of
`the implant for the specific application and location;
`precise placement is enhanced and bulk displacement is
`prevented. Additionally, it appears that any final sha—
`ping/trimming needed at surgery can be accomplished
`on the final device using conventional dental or ortho-
`pedic equipment available at the time of surgery.
`The optimal conditions for fracture healing and long-
`term stability can be met if an implant can be designed
`allowing for motionlessness along all
`the interfaces
`necessary for a stable anchorage, thereby excluding (to
`the greatest extent possible) all outside influences on the
`remodeling process and allowing the local stress/strain
`field to control.
`
`Following implantation and initial tissue ingrowth,
`the metal foam device stays where it is placed without
`retention aids, a reflection of precise contouring and the
`rapid ingrowth of fibrovascular tissue to prevent dis-
`lodgement. The binding between bone and implant
`stabilizes the implant and prevents loosening. These
`implants thus will not need to be held in place by other
`means (e. g. sutures or cement); rather, the growth of a
`natural bone-to-bone seal is encouraged by the nature of
`the implant itself. Tissue ingrowth would not be a con-
`tributing factor to device retention for a period follow-
`ing implantation, however, until a substantial amount of
`ingrowth had occurred.
`The ability to precisely contour the device, along
`with its “Velcro-like” surface texture that provides
`multipoint contact with the surrounding tissue,
`is of
`some aid in retention, although mechanical aids may
`still be necessary at first. If needed, sutures would seem
`to lend themselves well to use with the open cell metal
`structure, while compatibility studies with cement and
`other bonding aids have been identified as an area of
`future investigation.
`Broad-scale clinical adoption of bone grafting onto
`the alveolar ridge and for certain orthognathic recon-
`struction has been hindered by the well-established
`problem of resorption. Hydroxyapatite implants un-
`dergo some degree of chemical dissolution, often limit-
`ing their effectiveness as porous bone implants. Studies
`have shown that too-rapid degradation can inhibit the
`ongoing regeneration of bone throughout the implant.
`A permanent, nonresorbing implant can afford long-
`term maintenance of the augmentation and thereby
`overcome the resorption problem. However, perma-
`nent implants can be vulnerable to infection, loosening,
`or extrusion due to a lack of chemical or biomechanical
`compatibility and/or incomplete cellular ingrowth.
`An open cell metal implant, being metallic, will un-
`dergo no resorption, and its anticipated complete bi-
`ocompatibility and osteoconductivity render such con-
`cerns moot. Non-resorption is also beneficialin load-
`carrying applications where complete bone ingrowth
`cannot be achieved; the continued presence of the tanta-
`lum structures, with their superior mechanical proper-
`ties, is beneficial in such circumstances.
`
`Page 9 ofll
`
`Page 9 of 11
`
`

`

`5,282,861
`
`9
`The advantages of the open cell metal structure for
`bone implants are summarized as follows:
`a. lightweight, low-density
`b. very strong
`0. biocompatible
`three-dimensional
`d. high interconnected, uniform,
`porosity with high void fraction; structure similar
`to natural cancellous bone, with resultant os-
`teoconductivity
`e. fabricable to virtually any desired porosity/pore 10
`sxze
`
`5
`
`f. excellent mechanical properties
`g. imposed loads distributed throughout the structure
`and into both the ingrowing new bone and the
`surrounding existing bone as well, avoiding stress 15
`shielding
`h. readily shapeable to most desired configurations
`i. non-resorbing
`j. nearly all physical and mechanical properties can
`be tailored for a specific application, due to the 20
`number of fabrication variables available to be
`
`manipulated and the versatility of the CVD pro-
`cess.
`
`FIG. 4 illustrates an apparatus for depositing the
`metal, such as tantalum, on the carbon foam substrate. 25
`A reaction chamber 200 encloses a chlorination cham-
`ber 202 and a hot wall furnace 204. A resistance heater
`206 surrounds the chlorination chamber 202 and an
`induction heating coil 208 surrounds the reaction cham-
`ber 200 to heat the hot wall furnace 204.
`Tantalum metal 210 is located within the chlorination
`
`30
`
`chamber 202 and a carbon foam substrate 212 is posi-
`tioned within the hot wall furnace. Chlorine