Acta Biomaterialia 14 (2015) 217–225

Contents lists available at ScienceDirect

Acta Biomaterialia

journal homepage: www.elsevier .com/locate /actabiomat

Additively manufactured porous tantalum implants

http://dx.doi.org/10.1016/j.actbio.2014.12.0031742-7061/� 2014 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.

⇑ Corresponding author at: 3D Systems – LayerWise NV, Grauwmeer 14, 3001Leuven, Belgium. Tel.: +32 16 946414; fax: +32 16 946401.

E-mail address: ruben.wauthle@3dsystems.com (R. Wauthle).

Ruben Wauthle a,b,⇑, Johan van der Stok c, Saber Amin Yavari d, Jan Van Humbeeck e, Jean-Pierre Kruth a,Amir Abbas Zadpoor d, Harrie Weinans d,f, Michiel Mulier g, Jan Schrooten e,h

a KU Leuven, Department of Mechanical Engineering, Section Production Engineering, Machine Design and Automation (PMA), Celestijnenlaan 300B, 3001 Leuven, Belgiumb 3D Systems – LayerWise NV, Grauwmeer 14, 3001 Leuven, Belgiumc Orthopaedic Research Laboratory, Department of Orthopaedics, Erasmus University Rotterdam Medical Centre, Rotterdam, The Netherlandsd Faculty of Mechanical, Maritime, and Materials Engineering, Delft University of Technology (TU Delft), Mekelweg 2, 2628 CD Delft, The Netherlandse KU Leuven, Department of Materials Engineering, Kasteelpark Arenberg 44, PB2450, 3001 Leuven, Belgiumf Department of Orthopedics & Department of Rheumatology, UMC Utrecht, Heidelberglaan 100, 3584 CX Utrecht, The Netherlandsg KU Leuven, Department of Orthopaedics, Weligerveld 1, 3212 Pellenberg, Belgiumh KU Leuven, Prometheus, Division of Skeletal Tissue Engineering, Bus 813, O&N1, Herestraat 49, 3000 Leuven, Belgium

a r t i c l e i n f o a b s t r a c t

Article history:Received 24 July 2014Received in revised form 25 November 2014Accepted 4 December 2014Available online 11 December 2014

Keywords:TantalumSelective laser meltingPorous biomaterialsBone regeneration

The medical device industry’s interest in open porous, metallic biomaterials has increased in response toadditive manufacturing techniques enabling the production of complex shapes that cannot be producedwith conventional techniques. Tantalum is an important metal for medical devices because of its goodbiocompatibility. In this study selective laser melting technology was used for the first time to manufac-ture highly porous pure tantalum implants with fully interconnected open pores. The architecture of theporous structure in combination with the material properties of tantalum result in mechanical propertiesclose to those of human bone and allow for bone ingrowth. The bone regeneration performance of theporous tantalum was evaluated in vivo using an orthotopic load-bearing bone defect model in the ratfemur. After 12 weeks, substantial bone ingrowth, good quality of the regenerated bone and a strong,functional implant–bone interface connection were observed. Compared to identical porous Ti–6Al–4Vstructures, laser-melted tantalum shows excellent osteoconductive properties, has a higher normalizedfatigue strength and allows for more plastic deformation due to its high ductility. It is therefore con-cluded that this is a first step towards a new generation of open porous tantalum implants manufacturedusing selective laser melting.

� 2014 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.

1. Introduction

Today, porous metal orthopedic implants are breaking newground in skeletal reconstructive surgery and more specifically intotal hip replacement, recently referred to as ‘‘the operation ofthe century’’ [1]. While total hip arthroplasty has become a routinetreatment for hip osteoarthritis and as the number of surgicalinterventions increased from the 1970s on, the number of neces-sary revision operations has also increased. The most commoncauses for revisions are mechanical loosening, infection and insta-bility/dislocation [2]. A large portion of these failures are due topolyethylene wear and periprosthetic osteolysis and, morerecently, metallosis [1,3]. When this causes large bone defectsand cavities in the bony structure and when the dimensions of

these defects become too large, there is a chance of aseptic loosen-ing of the implant. These bone defects need to be reconstructedand filled during revision operations with new structures on whichnew prosthesis elements are attached. These new structures can beautografts, polymethylmethacrylate (bone cement) or allografts. Inthe case of large bone defects, the structural bone and compactedgrafts are currently the best solution to create a mechanical stablereconstruction, which can withstand postoperative mechanicalloads [4,5]. Given the increased demand for allograft materialand its limited availability, more and more surgeons have startedto use artificial bone substitute materials. Preferably, thesesubstitute materials should provide initial fixation and long-termstability for surrounding prostheses. Porous metals have the abilityto allow for bone ingrowth and avoid stress shielding by reducingstiffness without too high a loss in strength and thus are suitable tobe used as bone substitute materials in load-bearing applications.The first porous implants only had porous coatings of cobalt–chrome, but were soon replaced by surgical grade 5 Ti–6Al–4V,

218 R. Wauthle et al. / Acta Biomaterialia 14 (2015) 217–225

which is still the most widely used material for porous biomateri-als [1]. An alternative with much potential is tantalum (Ta).

Ta is a hard, ductile, highly chemically resistant material withgood apposition to human bone [6]. It has been successfully usedin clinical applications as a biomaterial since the 1940s [6,7], butbecause it is both expensive and difficult to machine [8], the useof Ta as a biomaterial has been limited. The biocompatibility andnon-toxic behavior of Ta in general have been reported previously[9,10]. The attachment, proliferation and differentiation of humanosteoblasts [11] on Ta are excellent in comparison to the morecommonly used surgical grade 5 Ti–6Al–4V [12,13]. More recently,surface modifications [14–20] or protein coatings [21] to enhancethe biological performance of Ta have been investigated. Despiteits promising biological properties, Ta is not considered as anappropriate material for large implants due to its high densityand high price. On the other hand, the high density of Ta is anadvantage for high-contrast applications, such as bone markersfor radiostereometry [22–26]. To avoid stress shielding, Ta ismostly limited to open porous structures in biomedical applica-tions. Knowing that bulk Ta is difficult to process, and thereforecommonly produced as a powder, it can also be applied as a coat-ing on both solid [13,27] and open porous implant surfaces [28,29].

In the early 2000s, a new porous Ta biomaterial for acetabularcups was introduced (Trabecular Metal™ (TM), Zimmer, Warsaw,IN, USA). This highly innovative biomaterial is created by deposit-ing a Ta coating upon open porous carbon matrices [29]. Ever sinceits introduction, the use and range of applications using TM haveexpanded. The increased interest in Ta, mainly thanks to TM, canclearly be illustrated by the increased number of publications onTa in the last 10 years. Today, TM is the most commonly used bio-material containing Ta in orthopedics [30–36]. Over 250 publica-tions and 800,000 surgeries worldwide [37] illustrate its non-toxic and osteoconductive behavior, but without actual proof ofoutperforming other commonly used materials such as surgicalgrade 5 Ti–6Al–4V [38–42]. It has proven bone ingrowth both inanimal studies [29,43–45] and retrievals from clinical cases[46,47], although some publications mention issues like implantfailure and the brittle deformation behavior of TM [48–50]. Fur-thermore, the mechanical properties of TM are close to those ofhuman bone [51–54] and most uses of TM can be found in hip[55–62], knee [63–67], spinal [40,68–70] and other orthopedicapplications [71]. Despite the clinical success of TM, no otherimplant manufacturers use Ta as raw material for orthopedicimplants. Surgical grade 5 titanium and other titanium alloys arestill the current standard for porous biomaterials used in orthope-dics [36,72]. Several manufacturing techniques, like furnace sinter-ing, plasma spraying, laser/electron beam melting, lost wax castingand vapor deposition techniques, are used to manufacture poroustitanium biomaterials [73–75].

Recently, new attempts have been made to produce both solidand porous Ta parts using novel manufacturing techniques suchas laser-engineered net shaping (LENS™) [76,77], spark plasmasintering [78] and selective laser melting (SLM) [79,80]. SLM isan additive manufacturing technology in which a focused laserbeam melts thin layers of metal powder together in order to createfully dense functional parts [81]. Some of the present authors wereinvolved in previous research in which the first successfulmanufacturing of functional Ta parts by SLM was reported [80].It is now possible to make nearly 100% dense parts that fulfill boththe chemical and mechanical requirements for ISO 13782‘Unalloyed Tantalum for Surgical Applications’ [82]. Therefore,SLM manufacturing of porous Ta implants could lead to unexploredopportunities in orthopedics by tailoring mechanical propertiesand innovative implant designs with predictable mechanical prop-erties. The research presented in this work examines SLM as a newmethod to manufacture highly porous pure Ta bone replacement

structures with controlled mechanical properties. Following mor-phological and mechanical characterization, an in vivo assay in aload-bearing orthotopic animal model was conducted.

2. Materials and methods

2.1. Materials and manufacturing

Porous Ta structures were manufactured from Ta powder usingthe selective laser melting technology (3D Systems – LayerwiseNV, Leuven, Belgium). The unit cell used as the microarchitectureof these porous structures was a dodecahedron (Fig. 1B), with anaverage strut size of 150 lm and an average pore size of 500 lm,which resulted in an overall open porosity of ±80%. This specificunit cell was chosen in order to compare the obtained results withthose of previous studies that used identical dodecahedron struc-tures made by SLM out of Ti–6Al–4V ELI powder [83–87]. In thiswork, the same spherical pure Ta powder (chemical compositionaccording to ISO 13782 [88], measured purity of 99.99% by ICP-MS), with particle size ranging from 10 to 25 lm, as in Ref. [80],was used (powder particle size distribution measured by laser lightdiffraction: D10 = 12.8 lm, D50 = 18.4 lm, D90 = 26.4 lm; seeFig. 1A). The production was performed in an inert atmosphereand the samples were built on top of a solid Ti substrate. After pro-duction, the samples were removed from the substrate using wireelectro discharge machining. Porous structures in the shape of a ratfemur defect, with a maximum diameter of 4 mm and a height of6 mm, including a central channel that corresponds to the femurcanal, were manufactured for filling the segmental defect createdin the animal model (Fig. 1C) [83,84,87]. Cylindrical porous speci-mens with a diameter of 10 mm and a height of 15 mm were man-ufactured for morphological analysis, static and dynamicmechanical testing and an in vitro cytotoxicity test (Fig. 1D).

2.2. Morphological analysis

The overall porosity was measured using dry weighing andArchimedes measurements on five different cylindrical samplesprior to their being used for mechanical evaluation. Dry weighingoccurred under normal atmosphere conditions, and the overallporosity was calculated by dividing the actual weight by the theo-retical weight of the macro volume using a theoretical density of16.6 g cm�3 for pure Ta. Archimedes measurements are based ona combination of dry weighing and weighing in pure ethanol.The overall porosity is then calculated by dividing the actualvolume by the macro volume. All weighing measurements wereperformed on an OHAUS Pioneer balance. The geometry andsurface of the porous structures were viewed using secondary elec-trons in a Philips SEM XL30 FEG equipped with a Schottky type ofgun. A Leica DMILM 12 V/100 W light optical microscope was usedto evaluate metallographic cross-sections.

2.3. Mechanical testing

2.3.1. Static mechanical testingStatic mechanical testing of cylindrical porous samples was car-

ried out in accordance with the standard ISO 13314 [89]. All testswere carried out using an Instron 5985 mechanical testingmachine (30 kN load cell) by applying a constant deformation rateof 1.8 mm min�1. Flat hard metal endplates were used and onlyvertical movement was allowed. The strain was measured by theactual displacement of the crossheads and all tests were doneunder normal atmospheric conditions. Each static compression testresulted in a stress–strain curve (Fig. 3), for which the followingvalues were calculated: plateau stress (rpl) as the arithmetic mean

D

15 m

m

figure 2A - B

figure 2C - D

6 m

m

50 µm

BA

C

Fig. 1. SEM picture of the used spherical Ta powder for SLM manufacturing (A), 3-Dvisual representation of the dodecahedron unit cell (B), top and side view of thefemur shaped porous Ta implant (C), 3-D visual representation of the cylindricaltest specimen (D) with indication of the SEM and cross-sectional views of Fig. 2.

R. Wauthle et al. / Acta Biomaterialia 14 (2015) 217–225 219

of the stresses between 20 and 30% compressive strain; plateauend stress (r130) and strain (eple) as the point in the stress–straincurve at which the stress is 1.3 times the plateau stress; quasi-elastic gradient (E) as the gradient of the straight line determinedwithin the linear deformation region at the beginning of the com-pressive stress–strain curve; and yield strength (ry) as the com-pressive 0.2% offset stress. In this context, the quasi-elasticgradient is closest to the concept of elastic modulus, which is usedfor solid materials. In order to facilitate understanding and com-parison between the results of this study and those of similar stud-ies on solid and porous materials, the quasi-elastic gradient will bereferred to as the elastic modulus. Nevertheless, the exact defini-tions presented above should be kept in mind when interpretingthe data.

2.3.2. Dynamic mechanical testingCompression–compression fatigue tests were carried out using

an identical set-up as reported before [86], using a hydraulic testframe (MTS, Minneapolis, US) with a 25 kN load cell. The loadingfrequency was fixed at 15 Hz (sinusoidal wave shape) and a con-stant load ratio of R = 0.1 was used. Hard metal endplates wereused with circular walls with heights of 1 mm and with inner radiislightly (1 mm) larger than the diameter of the specimens. Thir-teen different values of maximum force were chosen, resulting inapplied stress levels between 0.23 and 0.9 rpl. The samples wereconsidered to have failed once they lost +90% of their stiffness.The S–N curve of porous Ta was established by plotting the abso-lute values of stress vs. the number of cycles to failure for all testedsamples.

2.4. Biological evaluation

The biological and bone regeneration performance of porous Tawas evaluated through an in vitro cytotoxicity test, an in vivo seg-mental bone defect model and the quantification of bone ingrowthby histological analysis.

2.4.1. Cytotoxicity testBiocompatibility testing was performed by means of an in vitro

cytotoxicity test according to ISO 10993-5 [90] (Toxikon Europe

NV, Leuven, Belgium). The observed viability for the L929 mamma-lian cells exposed to the test sample at 41 h observation is used asthe test criterion and the tested biomaterial is considered non-cytotoxic if the percentage of viable cells is equal to or greater than70% of the untreated control. Serum-supplemented minimumessential medium was used as the cell culture medium.

2.4.2. Bone defect modelFor the functional in vivo evaluation of the open porous Ta

implants, load-bearing segmental bone defects were used asdescribed earlier [83,84,87]. In brief, a critical-sized femoral bonedefect was grafted with a porous Ta implant in eight male Wistarrats. The Animal Ethics Committee of the Erasmus Universityapproved the study and Dutch guidelines for care and use of labo-ratory animals were followed. Prior to surgery, rats were adminis-tered one dose of antibiotics (enrofloxacin, 5 mg kg�1 body weight)through subcutaneous injection. Surgery was performed asepti-cally under general anaesthesia (1–3.5% isoflurane) as follows: firstthe right femur was exposed through a lateral skin incision andblunt division of the underlying fascia. Then a 23 mm long poly-ether ether ketone (PEEK) plate was fixed to the anterolateral planeusing six bicortical titanium screws (Ø 0.8 � 6.5 mm). The perios-teum was removed from approximately 8 mm of the mid-diaphy-seal region before a 6 mm cortical bone segment was removedwith a wire saw and a tailor-made saw guide. The 6 mm femur-shaped implants were press-fitted into the defect. Finally, the fas-cia and skin were sutured using Vicryl 5-0 and pain medication(buprenorphine, 0.05 mg kg�1 body weight) was administeredthrough subcutaneous injection twice a day for 3 days. The ratswere sacrificed after 12 weeks with an overdose of pentobarbital(200 mg kg–1 body weight). Afterwards implant fixation was deter-mined on explants using X-ray images acquired using a SkyScan1076 (Bruker micro-CT NV, Kontich, Belgium).

2.4.3. Histological evaluationHistology was performed on two specimens of the total group

(n = 8: 2� histological evaluation, 1� spare histological evaluation,5� ex vivo testing) to qualitatively study the amount of boneingrowth and to examine the bone–tantalum interface. The speci-mens were selected by two medically trained co-authors as beingrepresentative of the whole group (the specimens with the leastand most visible amount of bone formation were selected). Speci-mens were first preserved and dehydrated, then embedded inmethylmethacrylate. Serial sections of about 100 lm were madeusing a microtome saw (longitudinal cuts, anterior–posterior),which were then polished to 50 lm, stained using Stevenel’s blueand finally counterstained using von Gieson’s picrofuchsin. As aresult, bone stains red, fibrous tissue stains blue and cartilagestains purple. Stained sections were examined using a LeicaM165 FC fluorescent stereomicroscope. One additional specimenwas kept as a spare in case further histological examination wasnecessary, but was not ultimately used.

2.5. Ex vivo testing

The strength of the implant–bone connection was evaluatedafter explantation on the five remaining specimens by means ofa torsion test, as described earlier [91]. In brief, both ends of eachfemur were embedded in a cold-cured epoxy resin (Technovit4071, Heraeus Kulzer, Germany) after removal of the PEEK fixationplate. On the upper clamping side, a Cardan joint was used toensure that the specimens were subjected to pure rotation withoutbending. The lower sides of the specimens were simply fixed. Thetests were performed until failure with a rotation rate of 0.5� s�1

using a static mechanical testing machine (Zwick GmbH, Ulm,Germany). The torsional strength (maximum torque to failure,

220 R. Wauthle et al. / Acta Biomaterialia 14 (2015) 217–225

N mm) and maximum rotation (degree) were determined andreported.

3. Results

3.1. Morphological properties

Dry weighing resulted in a mean overall open porosity of79.9 ± 0.2%. According to the Archimedes measurement, the strutdensity was 99.0 ± 0.2% and the overall porosity was 79.7 ± 0.2%.Fig. 2A and B shows scanning electron microscopy (SEM) picturesof the top view at different magnifications and Fig. 2C and D showslight optical microscopy (LOM) pictures at a certain cross-sectionof a prepared sample of the regular porous structure.

3.2. Mechanical properties

The results of the static compression tests are summarized inTable 1. Due to the ductile behavior of the porous Ta material, nomaximum compressive stress (rmax) or strain at maximum com-pressive stress (emax) could be registered. Fig. 3A shows a represen-tative stress–strain curve and the ductile behavior of the porous Tastructure during static compression testing. The repeatability ofthe mechanical properties of the porous Ta is illustrated byFig. 3B, in which the stress–strain curves for 0–20% strain of alltested samples are shown. The dynamic compression test resultsare shown in Fig. 4 through an S–N curve consisting of 13 datapoints obtained by compression–compression fatigue testing. Thehorizontal line at the level of the yield strength differentiatesbetween low cycle fatigue strength (above ry and mainly plasticdeformation) and high cycle fatigue strength (below ry and mainlyelastic deformation). Since three samples did not fail at 106 cycles,

Fig. 2. Morphological properties of open porous SLM-processed Ta: SEM pictures ofthe top view at different magnifications (A and B) and LOM pictures of a cross-section at different magnifications (C and D). The red dotted lines in A and C are ofcorresponding size.

Table 1Results of the static compression tests of open porous SLM-processed Ta according toISO 13314.

Units Mean St. Dev.

Yield stress, ry MPa 12.7 0.6Plateau stress, rpl MPa 21.8 0.9Plateau end stress, r130 MPa 28.3 1.2Plateau end strain, eple % 36.1 0.4Elastic modulus, E (=quasi-elastic gradient) GPa 1.22 0.07

the highest of these three values (7.35 MPa) is considered as anindication of the fatigue limit, Sf, of the porous Ta biomaterial.

3.3. Biological properties

The cytocompatibility of SLM-produced Ta implants was deter-mined in vitro using a cytotoxicity test. For this test, 82% viabilitywas observed, and the cellular response obtained from the posi-tive control extract (4% viability) and the negative control extract(108% viability) confirmed the suitability of the test system. Thetest sample is considered to be non-cytotoxic and meets therequirements of ISO 10993-5, thus showing that SLM-producedTa is cytocompatible.

Fig. 5 shows X-ray images and histology images of the twoimplants that were processed for histology after 12 weeksin vivo. In the X-ray images (Fig. 5A and G), the porous Ta implantcan clearly be recognized thanks to the high contrast caused by thehigh atomic number of the Ta material. However, this X-ray-absorbing property makes it difficult or impossible to draw anyconclusions on bone formation and bone ingrowth into the porousTa implant based on radiological images. The only observationsthat could be made are that in Fig. 5A there is a radiolucent line vis-ible, indicating a weak implant–bone connection at the bottom,probably resulting in no bone ingrowth (or very little), whilst atthe connection at the top no radiolucent line is visible. Also, inFig. 5G no radiolucent lines are present at the implant–bone inter-face and in this case the bone has clearly grown around theimplant.

Two explant specimens were selected for histological analysisas being representative for the whole group. Three images ofcross-sections at different locations and two detailed images ofthe explants are shown in Fig. 5B–F and H–L, respectively. Allcross-sections show some amount of bone ingrowth inside the por-ous Ta implant. In Fig. 5B, C and F the X-ray observation of a badimplant–bone connection at the bottom is confirmed by no visiblebone ingrowth, but the purple staining indicates the presence ofcartilage-like tissue at these locations (Fig. 5B, C and E). Further-more, thorough bone ingrowth from the top with restoration ofthe femoral canal can be seen in Fig. 5C and soft tissue is observedin Fig. 5B and F. For the second specimen (Fig. 5H, I and L), the bonehas grown not only around the implant, but also deep inside thepores of the implant, resulting in an almost full bridging of thedefect at all cross-sections. At some locations only small gaps inthe new bone growth are seen, and the presence of cartilage canbe noticed. As shown by the detailed images in Fig. 5D, J and K, agood implant–bone interface is established since the bone hasgrown closely to the Ta surface.

3.4. Mechanical properties of ex vivo specimens

Five explant specimens were biomechanically evaluated usingtorsion testing for which the results of four specimens can be foundin Table 2 and Fig. 6, which shows the rotation–torque curves. Twospecimens did not fail at the maximum torque (450 N mm) of thetest setup, whereas three samples did fail at an average maximumtorque of 331.3 N mm and an average rotation of 59.1�. Two of thespecimens that failed showed a fracture at the implant–bone inter-face and one specimen had a fracture in the bone region. One of thespecimens that did not fail was excluded from the test results sincethe rotation–torque curve showed some irregularities.

4. Discussion

It has been shown previously that it is possible to produce solidTa parts by SLM that meet the requirements of ISO 13782 in terms

0 5 10 15 200

5

10

15

20

Compressive strain [%]

Com

pres

sive

str

ess

[MPa

]

sample 1sample 2sample 3sample 4sample 5average

0 5 10 15 20 25 30 35

0

5

10

15

20

25

30

Compressive strain [%]

Com

pres

sive

str

ess

[MPa

]

stress−strain curveplateau stressquasi−elastic gradient0.2 % offset

eple

σ130

σpl

σy

BA

Fig. 3. Static mechanical properties of open porous SLM-processed Ta: representative compressive stress–strain curve and graphical representation of the calculated valuesry, rpl, r130 and E (A) and overview of the individual and average stress–strain curves of all five tested specimens (B).

103 104 105 1060

5

10

15

20

25

Cycles to failure, N

Max

imum

str

ess,

S [M

Pa]

σy = 12.7 MPa

σpl = 21.8 MPa

Sf = 7.35 MPa

Fig. 4. Dynamic mechanical properties of open porous SLM-processed Ta: S–N curve obtained by compression–compression fatigue testing with indication of the plateaustress (rpl), the yield stress (ry) and the fatigue limit (Sf).

R. Wauthle et al. / Acta Biomaterialia 14 (2015) 217–225 221

of both chemical composition and mechanical properties[80,82,88]. In this study, the SLM technology was used to producefor the first time porous structures consisting of a dodecahedronunit cell with an overall porosity of 80%, a strut size of 150 lmand a pore size of 500 lm. Morphological analysis revealed theregular dodecahedral architecture of the porous structure with ahighly repeatable overall porosity. The small difference betweenthe dry weighing and Archimedes method can be explained bythe small amount of enclosed pores inside the struts (Fig. 2E).Other strut defects are caused by the strut surface roughness andthe imperfect alignment of the cross-sectional plane. Therefore itcan be concluded that the SLM technology is able to produce veryfine porous Ta structures with high repeatability. This quality isalso noticeable in the mechanical properties, and is important ifthis technology is to be considered for use in the serial manufactur-ing of implants.

The static mechanical properties, like yield strength(12.7 ± 0.6 MPa) and elastic modulus (1.22 ± 0.07 GPa), are in therange of human cancellous bone and are hence favorable for low-ering stress shielding effects, although it should be noted thatthe elastic modulus of the porous Ta implants evaluated here isbelow the stiffness of most human cortical bone [6]. Comparedto similar porous structures in Ti–6Al–4V ELI made by SLM[85,86,92], the stress–strain curve of porous Ta does not reach afirst local maximum due to the intrinsic ductile behavior of theTa material (comparative data not shown) [80,82,93,94]. Insteadof failure at the local maximum, continuous (plastic) deformationnow occurs. This different deformation behavior is expected sinceconventionally processed pure tantalum has a higher ductilitycompared to Ti–6Al–4V ELI [6]. As already postulated by themorphological properties, the repeatability of the process isconfirmed by the static mechanical properties since all calculated

A B C D,E F

G H I J,K L

Fig. 5. X-ray and histological images of open porous SLM-processed Ta: explant specimen 1 (A–F) and explant specimen 2 (G–L), including detailed interface views forspecimen 1 (D and E) and for specimen 2 (J and K). The scale bar indicates 1 mm.

Table 2Torsion test results of four open porous SLM-processed Ta explants after 12 weeks.

Specimen Max. torque(N mm)

Rotation at max.torque (�)

Fracture location

1 324.6 54.6 Implant–boneinterface

2 324.4 65.7 Bone3 345.0 56.9 Implant–bone

interface4 >450.0 / Not brokenAverage 331.3* 59.1*

* Specimen 4 not included.

0

100

200

300

400

0 25 50 75 100 125 150

Torq

ue [N

mm

]

Rotation [°]

Specimen 1Specimen 2Specimen 3Specimen 4

Fig. 6. Rotation–torque curves of four explant specimens tested by torsion testing.The failed samples are shown in black, while specimen 4 (solid gray line) did not failat the maximum torque (450 N mm) of the test set-up.

222 R. Wauthle et al. / Acta Biomaterialia 14 (2015) 217–225

values – except for the elastic modulus – have standard deviationslower than 5% of the nominal values.

The fatigue behavior has been evaluated at 13 different stresslevels, of which six are above and seven are below the yieldstrength. Despite the fact that tests at these stress levels were onlyperformed once per level, a shift in the trend can be noticed whenpassing the yield strength. However, this observation and the indi-cation of the fatigue limit Sf should be interpreted with care, sincefurther tests are required in order to confirm the results. Althoughlarge and plastic deformation might be considered as implant fail-ure, the authors believe this partially explains the good implantin vivo performance observed in this study. Since the plastic defor-mability of the implant can also be used for customizing theimplant shape to the bone defect, it is important to characterizethe fatigue properties above the yield strength. Normalizing thefatigue limit by the yield strength results in an endurance limitof 0.58 ry, which is much higher when compared to identical por-ous structures in Ti–6Al–4V ELI, which have an estimated endur-ance limit of 0.12 ry for an open porosity range from 66 to 84%[86]. Even for absolute values, porous Ta (7.35 MPa) apparentlyhas higher fatigue strength than porous Ti–6Al–4V ELI (4.18 MPa)[86]. The reason for the good fatigue behavior of porous Ta canbe explained by its high ductility, which lowers crack initiationand propagation by softening the material when loaded [95].Again, since no statistics can be done, further tests have to be con-ducted in order to confirm the observed trends.

The cytotoxicity test also confirmed the non-cytotoxicity of theporous Ta biomaterial after SLM processing; however, to evaluate

R. Wauthle et al. / Acta Biomaterialia 14 (2015) 217–225 223

the in vivo functionality of the new biomaterial, an animal exper-iment was done. In this experiment, eight rats were implantedwith a 6 mm porous Ta implant to reconstruct a critical-sized fem-oral bone defect. Since Ta has a high atomic weight, it readilyabsorbs X-rays, and it is therefore difficult to use standard evalua-tion techniques like radiographic or 3-D computed tomographyfollow-up, as was performed in previous studies [83,84,87]. Never-theless, implant fixation can still be observed from the presence ofradiolucent lines, though no conclusions about the amount of boneingrowth can be drawn based on the radiographic images. Histo-logical analysis of two explant specimens confirmed the observa-tion of fixation based on X-rays by the presence of bone insidethe porous Ta implant at the sides with no radiolucent lines (topof Fig. 5A vs. B, C and F and top and bottom of Fig. 5G vs. H, Iand L). The depth of bone ingrowth varied between the cross-sections within a single specimen, but one cross-section of eachspecimen was at least more than 50% of the total length of thedefect. This resulted in an almost 100% bridging of the defect forthe second specimen. The regenerated bone grows closely to theTa surface and demonstrates the apposition of Ta to bone, whichis likely to enable a continuous load transfer. The lack of boneingrowth in the first specimen (at the bottom of Fig. 5B, C and F)could be explained by less initial fixation during surgery. Mechan-ical instability is a fairly reasonable cause, but other factors mayhave influenced the bone regeneration. Nevertheless, the inter-implant differences are not considered to have an influence sincethe differences in morphological and mechanical properties arevery small. Despite the mechanical instability, a stimulating bio-logical effect is evidenced by cartilage formation (Fig. 5B, C andE). In a biologically preferential environment, micromotion maylead to bone generation through a phase of cartilage formation[96,97]. The PEEK plate in this animal model was used to stabilizethe defect, and there is bone visible along the PEEK plate in Fig. 5L.Since there was also a thin interface of fibrous tissue present, bonegrowth was through the porous Ta implant rather than along theosteoconductive PEEK plate.

A good implant–bone interface connection is also evidenced bythe biomechanical torsion test results. Two out of five explantspecimens did not fail. One specimen failed due to a fracture inthe host bone and two specimens failed at the implant–bone inter-face. The maximum torque for the three failed specimens did notdiffer significantly, indicating that the implant–bone interfacewas at least as strong as the host bone. This assumption is in agree-ment with the previously reported maximum torque value of146.7 ± 19.1 N mm of intact femurs in a comparable study in whichthe same animal model was used [98].

Based on the radiographical and histological analysis and bio-mechanical evaluation of these porous Ta implants after 12 weeksin vivo, it can be concluded that this new biomaterial functionswell in a biomechanically loaded environment. Bone was seen toregenerate and grow into this porous biomaterial except for oneside of a histologically examined specimen, resulting in a stableand strong reconstruction. The favorable biological properties ofthe Ta material that facilitate cell attachment and proliferation[9–13], and the preferential mechanical properties of the porousstructure that likely avoid stress shielding effects by transferringloads in a stimulating way [99], are considered to be the main rea-sons for the excellent implant performance. To progress from thefirst promising results of this new porous Ta biomaterial, further,more extensive tests should be carried out on large animal modelsto confirm the findings of this study. Also, the potential risk forinfection deep inside the pores of large porous implants and theeffects and optimization of the microporosity are important totake into consideration when translating this porous tantalumbiomaterial to human implants.

This work has illustrated for the first time that SLM technologycan become a robust method for manufacturing porous Taimplants, allowing for almost full design freedom to create anyinterconnected porous structure with controllable mechanicalproperties and personalized outer geometries. Also, no additionalsurface modification treatment (e.g. etching, anodizing andhydroxyapatite plasma-sprayed coating) was applied to improveimplant–host interaction, whereas this is commonly done withTi–6Al–4V implant surfaces. It should be noted that the high costof the material and the difficulty in making radiological interpreta-tions are currently major disadvantages to be considered beforethis material can be used clinically in medical implants. Given boththe advantages and disadvantages, SLM-processed, porous Taoffers new opportunities for innovative implant designs, both forstandard and patient-specific orthopedic implants. Small implanttypes (e.g. dental implants, spinal implants, small joints andextremities) benefit from less material consumption, whereashighly porous structures can be applied as a thin layer on top ofa solid substrate for larger, load-bearing applications for which afast and solid anchoring of the implant is required. Compared toidentical Ti porous structures, SLM-produced Ta shows excellentosteoconductive properties even without any surface treatments,has a higher normalized fatigue strength and allows for a higherformability due to its excellent ductile properties. The latter couldlead to hitherto unexplored applications easing surgical handlingand intraoperative manipulation of the implant to obtain an opti-mal implant–bone fit. Further research on this topic could investi-gate the optimization of geometric and mechanical properties foroptimal load transfer and bone ingrowth for different applications,and the enhancement of cell attachment and proliferation by mod-ifying the surface characteristics.

5. Conclusions

In this study an additive manufacturing technology, selectivelaser melting, was used to manufacture a highly open porous(80%) pure Ta implant. The morphological and mechanical evalua-tion of this biomaterial demonstrated the high repeatability of theSLM process. With a yield strength of 12.7 MPa, an elastic modulusof 1.22 GPa and a ductile deformation mechanism, porous Ta exhib-its mechanical properties that are in the range of cancellous boneand appear to allow for bone ingrowth. Moreover, with a fatiguelimit of 7.35 MPa, the investigated material has relatively highresistance to cyclic loading. A cytotoxicity test as part of the biolog-ical evaluation raised no concerns over the biocompatibility of theSLM-processed material and an in vivo rat segmental bone defectmodel was used to investigate the osteoconductive and biomechan-ical performance of the porous material. Substantial bone ingrowthafter 12 weeks was shown by histological analysis, with almost fullbridging of the created defect in isolated cases. Torsion testing ofthe explants indicated a strong implant–bone interface connectionand the high strength of the repaired bone defect. Altogether, it canbe concluded that, based on this initial study, selective laser melt-ing can be used to manufacture highly porous, pure Ta orthopedicimplants with interesting mechanical properties and promisingin vivo performance for the animal model used in this study.

Acknowledgements

This research was funded by the agency for Innovation byScience and Technology (IWT) of the Flemish government throughBaekeland mandate ‘‘IWT 100228’’ as well as Project P2.04BONE-IP of the research program of the BioMedical MaterialsInstitute, co-funded by the Dutch Ministry of Economic Affairs.

224 R. Wauthle et al. / Acta Biomaterialia 14 (2015) 217–225

Appendix A. Figures with essential colour discrimination

Certain figures in this article, particularly Figs. 1–5 are difficultto interpret in black and white. The full colour images can be foundin the on-line version, at http://dx.doi.org/10.1016/j.actbio.2014.12.003.

References

[1] Learmonth ID, Young C, Rorabeck C. The operation of the century: total hipreplacement. Lancet 2007;370:1508–19.

[2] Bozic KJ, Kurtz SM, Lau E, Ong K, Vail TP, Berry DJ. The epidemiology of revisiontotal hip arthroplasty in the United States. J Bone Joint Surg Am2009;91:128–33.

[3] Sundfeldt M, Carlsson LV, Johansson CB, Thomsen P, Gretzer C. Asepticloosening, not only a question of wear: a review of different theories. ActaOrthop 2006;77:177–97.

[4] Villanueva M, Rios-Luna A, Pereiro De Lamo J, Fahandez-Saddi H, Bostrom MP.A review of the treatment of pelvic discontinuity. HSS J 2008;4:128–37.

[5] Brubaker SM, Brown TE, Manaswi A, Mihalko WM, Cui Q, Saleh KJ. Treatmentoptions and allograft use in revision total hip arthroplasty the acetabulum. JArthrop 2007;22:52–6.

[6] Helsen JA, Missirlis Y. Biomaterials – a tantalus experience. Heidelberg: Springer;2010.

[7] Black J. Biological performance of tantalum. Clin Mater 1994;16:167–73.[8] Balla VK, Bose S, Davies NM, Bandyopadhyay A. Tantalum – a bioactive metal

for implants. JOM 2010;62:61–4.[9] Matsuno H, Yokoyama A, Watari F, Uo M, Kawasaki T. Biocompatibility and

osteogenesis of refractory metal implants, titanium, hafnium, niobium,tantalum and rhenium. Biomaterials 2001;22:1253–62.

[10] Johansson CB, Hansson HA, Albrektsson T. Qualitative interfacial studybetween bone and tantalum, niobium or commercially pure titanium.Biomaterials 1990;11:277–80.

[11] Findlay DM, Welldon K, Atkins GJ, Howie DW, Zannettino ACW, Bobyn D. Theproliferation and phenotypic expression of human osteoblasts on tantalummetal. Biomaterials 2004;25:2215–27.

[12] Stiehler M, Lind M, Mygind T, Baatrup A, Dolatshahi-Pirouz A, Li H, et al.Morphology, proliferation, and osteogenic differentiation of mesenchymalstem cells cultured on titanium, tantalum, and chromium surfaces. J BiomedMater Res, Part A 2008;86:448–58.

[13] Tang Z, Xie Y, Yang F, Huang Y, Wang C, Dai K, et al. Porous tantalum coatingsprepared by vacuum plasma spraying enhance BMSCs osteogenicdifferentiation and bone regeneration in vitro and in vivo. PLoS One2013;8:e66263.

[14] Miyazaki T, Kim H-M, Kokubo T, Ohtsuki C, Kato H, Nakamura T. Mechanism ofbonelike apatite formation on bioactive tantalum metal in a simulated bodyfluid. Biomaterials 2002;23:827–32.

[15] Garbuz DS, Hu Y, Kim WY, Duan K, Masri BA, Oxland TR, et al. Enhanced gapfilling and osteoconduction associated with alendronate-calcium phosphate-coated porous tantalum. J Bone Joint Surg Am 2008;90:1090–100.

[16] Justesen J, Lorentzen M, Andersen LK, Hansen O, Chevallier J, Modin C, et al.Spatial and temporal changes in the morphology of preosteoblastic cellsseeded on microstructured tantalum surfaces. J Biomed Mater Res, Part A2009;89:885–94.

[17] Wang N, Li H, Wang J, Chen S, Ma Y, Zhang Z. Study on the anticorrosion,biocompatibility, and osteoinductivity of tantalum decorated with tantalumoxide nanotube array films. ACS Appl Mater Interfaces 2012;4:4516–23.

[18] Paganias CG, Tsakotos GA, Koutsostathis SD, Macheras GA. Osseous integrationin porous tantalum implants. Indian J Orthop 2012;46:505–13.

[19] Miyazaki T, Kim HM, Kokubo T, Miyaji F, Kato H, Nakamura T. Effect of thermaltreatment on apatite-forming ability of NaOH-treated tantalum metal. J MaterSci - Mater Med 2001;12:683–7.

[20] Miyazaki T, Kim HM, Miyaji F, Kokubo T, Kato H, Nakamura T. Bioactivetantalum metal prepared by NaOH treatment. J Biomed Mater Res2000;50:35–42.

[21] Lord MS, Modin C, Foss M, Duch M, Simmons A, Pedersen FS, et al. Monitoringcell adhesion on tantalum and oxidised polystyrene using a quartz crystalmicrobalance with dissipation. Biomaterials 2006;27:4529–37.

[22] Karrholm J, Gill RH, Valstar ER. The history and future of radiostereometricanalysis. Clin Orthop Relat Res 2006;448:10–21.

[23] Karrholm J, Herberts P, Hultmark P, Malchau H, Nivbrant B, Thanner J.Radiostereometry of hip prostheses. Review of methodology and clinicalresults. Clin Orthop Relat Res 1997:94–110.

[24] Valstar ER, de Jong FW, Vrooman HA, Rozing PM, Reiber JH. Model-basedRoentgen stereophotogrammetry of orthopaedic implants. J Biomech2001;34:715–22.

[25] Valstar ER, Gill R, Ryd L, Flivik G, Borlin N, Karrholm J. Guidelines forstandardization of radiostereometry (RSA) of implants. Acta Orthop2005;76:563–72.

[26] Valstar ER, Nelissen RGHH, Reiber JHC, Rozing PM. The use of roentgenstereophotogrammetry to study micromotion of orthopaedic implants. JPhotogramm Remote Sens 2002;56:376–89.

[27] Duan Y, Liu L, Wang L, Guo F, Li H, Shi L, et al. Preliminary study of thebiomechanical behavior and physical characteristics of tantalum (Ta)-coatedprostheses. J Orthop Sci 2012;17:173–85.

[28] Li X, Wang L, Yu X, Feng Y, Wang C, Yang K, et al. Tantalum coating on porousTi6Al4V scaffold using chemical vapor deposition and preliminary biologicalevaluation. Mater Sci Eng C Mater Biol Appl 2013;33:2987–94.

[29] Bobyn JD, Stackpool GJ, Hacking SA, Tanzer M, Krygier JJ. Characteristics ofbone ingrowth and interface mechanics of a new porous tantalum biomaterial.J Bone Joint Surg Br 1999;81:907–14.

[30] Gulotta LV, Wiznia D, Cunningham M, Fortier L, Maher S, Rodeo SA. What’snew in orthopaedic research. J Bone Joint Surg Am 2011;93:2136–41.

[31] Mantripragada VP, Lecka-Czernik B, Ebraheim NA, Jayasuriya AC. An overviewof recent advances in designing orthopedic and craniofacial implants. J BiomedMater Res Part A 2013;101:3349–64.

[32] Russell RD, Estrera KA, Pivec R, Mont MA, Huo MH. What’s new in total hiparthroplasty. J Bone Joint Surg Am 2013;95:1719–25.

[33] Levine B, Della Valle CJ, Jacobs JJ. Applications of porous tantalum in total hiparthroplasty. J Am Acad Orthop Surg 2006;14:646–55.

[34] Levine BR, Sporer S, Poggie RA, Della Valle CJ, Jacobs JJ. Experimental andclinical performance of porous tantalum in orthopedic surgery. Biomaterials2006;27:4671–81.

[35] Bobyn JD, Poggie RA, Krygier JJ, Lewallen DG, Hanssen AD, Lewis RJ, et al.Clinical validation of a structural porous tantalum biomaterial for adultreconstruction. J Bone Joint Surg Am 2004;86-A(Suppl. 2):123–9.

[36] Levine B. A new era in porous metals: applications in orthopaedics. Adv EngMater 2008;10:788–92.

[37] Zimmer Holdings, Inc., Annual Report 2012.[38] Welldon KJ, Atkins GJ, Howie DW, Findlay DM. Primary human osteoblasts

grow into porous tantalum and maintain an osteoblastic phenotype. J BiomedMater Res, Part A 2008;84:691–701.

[39] Sagomonyants KB, Hakim-Zargar M, Jhaveri A, Aronow MS, Gronowicz G.Porous tantalum stimulates the proliferation and osteogenesis of osteoblastsfrom elderly female patients. J Orthop Res 2011;29:609–16.

[40] Blanco JF, Sanchez-Guijo FM, Carrancio S, Muntion S, Garcia-Brinon J, delCanizo MC. Titanium and tantalum as mesenchymal stem cell scaffolds forspinal fusion: an in vitro comparative study. Eur Spine J 2011;20(Suppl.3):353–60.

[41] Jonitz A, Lochner K, Lindner T, Hansmann D, Marrot A, Bader R. Oxygenconsumption, acidification and migration capacity of human primaryosteoblasts within a three-dimensional tantalum scaffold. J Mater Sci –Mater Med 2011;22:2089–95.

[42] Schildhauer TA, Peter E, Muhr G, Koller M. Activation of human leukocytes ontantalum trabecular metal in comparison to commonly used orthopedic metalimplant materials. J Biomed Mater Res, Part A 2009;88:332–41.

[43] Hacking SA, Bobyn JD, Toh K, Tanzer M, Krygier JJ. Fibrous tissue ingrowth andattachment to porous tantalum. J Biomed Mater Res 2000;52:631–8.

[44] Reach Jr JS, Dickey ID, Zobitz ME, Adams JE, Scully SP, Lewallen DG. Directtendon attachment and healing to porous tantalum: an experimental animalstudy. J Bone Joint Surg Am 2007;89:1000–9.

[45] Deglurkar M, Davy DT, Stewart M, Goldberg VM, Welter JF. Evaluation ofmachining methods for trabecular metal implants in a rabbit intramedullaryosseointegration model. J Biomed Mater Res B Appl Biomater 2007;80:528–40.

[46] Sambaziotis C, Lovy AJ, Koller KE, Bloebaum RD, Hirsh DM, Kim SJ. Histologicretrieval analysis of a porous tantalum metal implant in an infected primarytotal knee arthroplasty. J Arthrop 2012;27(1413):e5–9.

[47] Hanzlik JA, Day JS, Ingrowth Retrieval Study Group. Bone ingrowth in well-fixed retrieved porous tantalum implants. J Arthrop 2013;28:922–7.

[48] Kasliwal MK, Baskin DS, Traynelis VC. Failure of porous tantalum cervicalinterbody fusion devices: two-year results from a prospective, randomized,multicenter clinical study. J Spinal Disorders Tech 2013;26:239–45.

[49] Kwong Y, Desai VV. The use of a tantalum-based augmentation patella inpatients with a previous patellectomy. Knee 2008;15:91–4.

[50] Nebosky PS, Schmid SR, Pasang T. Formability of porous tantalum sheet-metal.IOP Conf Ser Mater Sci Eng 2009;4:012018.

[51] Zardiackas LD, Parsell DE, Dillon LD, Mitchell DW, Nunnery LA, Poggie R.Structure, metallurgy, and mechanical properties of a porous tantalum foam. JBiomed Mater Res 2001;58:180–7.

[52] Shimko DA, Shimko VF, Sander EA, Dickson KF, Nauman EA. Effect of porosityon the fluid flow characteristics and mechanical properties of tantalumscaffolds. J Biomed Mater Res B Appl Biomater 2005;73:315–24.

[53] Vivanco J, Fang Z, Levine D, Ploeg HL. Evaluation of the mechanical behavior ofa direct compression molded porous tantalum–UHMWPE construct: amicrostructural model. J Appl Biomater Biomech 2009;7:34–42.

[54] Niinomi M, Nakai M, Hieda J. Development of new metallic alloys forbiomedical applications. Acta Biomater 2012;8:3888–903.

[55] Kosashvili Y, Backstein D, Safir O, Lakstein D, Gross AE. Acetabular revisionusing an anti-protrusion (ilio-ischial) cage and trabecular metal acetabularcomponent for severe acetabular bone loss associated with pelvicdiscontinuity. J Bone Joint Surg Br 2009;91-B:870–6.

[56] Deirmengian GK, Zmistowski B, O’Neil JT, Hozack WJ. Management ofacetabular bone loss in revision total hip arthroplasty. J Bone Joint Surg Am2011;93:1842–52.

[57] Issack PS. Use of porous tantalum for acetabular reconstruction in revision hiparthroplasty. J Bone Joint Surg Am 2013;95:1981–7.

R. Wauthle et al. / Acta Biomaterialia 14 (2015) 217–225 225

[58] Veillette CJ, Mehdian H, Schemitsch EH, McKee MD. Survivorship analysis andradiographic outcome following tantalum rod insertion for osteonecrosis ofthe femoral head. J Bone Joint Surg Am 2006;88(Suppl. 3):48–55.

[59] Macheras GA, Kateros K, Koutsostathis SD, Tsakotos G, Galanakos S, PapadakisSA. The Trabecular Metal Monoblock acetabular component in patients withhigh congenital hip dislocation: a prospective study. J Bone Joint Surg Br2010;92:624–8.

[60] Macheras GA, Papagelopoulos PJ, Kateros K, Kostakos AT, Baltas D, KarachaliosTS. Radiological evaluation of the metal-bone interface of a porous tantalummonoblock acetabular component. J Bone Joint Surg Br 2006;88:304–9.

[61] Mulier M, Rys B, Moke L. Hedrocel trabecular metal monoblock acetabularcups: mid-term results. Acta Orthop Belg 2006;72:326–31.

[62] Simon JP, Bellemans J. Clinical and radiological evaluation of modulartrabecular metal acetabular cups. Short-term results in 64 hips. Acta OrthopBelg 2009;75:623–30.

[63] Dunbar MJ, Wilson DA, Hennigar AW, Amirault JD, Gross M, Reardon GP.Fixation of a trabecular metal knee arthroplasty component. A prospectiverandomized study. J Bone Joint Surg Am 2009;91:1578–86.

[64] Meneghini RM, Lewallen DG, Hanssen AD. Use of porous tantalummetaphyseal cones for severe tibial bone loss during revision total kneereplacement. J Bone Joint Surg Am 2008;90:78–84.

[65] Howard JL, Kudera J, Lewallen DG, Hanssen AD. Early results of the use oftantalum femoral cones for revision total knee arthroplasty. J Bone Joint SurgAm 2011;93:478–84.

[66] Minoda Y, Kobayashi A, Iwaki H, Ikebuchi M, Inori F, Takaoka K. Comparison ofbone mineral density between porous tantalum and cemented tibial total kneearthroplasty components. J Bone Joint Surg Am 2010;92:700–6.

[67] Hayakawa K, Date H, Tsujimura S, Nojiri S, Yamada H, Nakagawa K. Mid-termresults of total knee arthroplasty with a porous tantalum monoblock tibialcomponent. Knee 2014;21:199–203.

[68] Fernandez-Fairen M, Sala P, Dufoo Jr M, Ballester J, Murcia A, Merzthal L.Anterior cervical fusion with tantalum implant: a prospective randomizedcontrolled study. Spine 2008;33:465–72.

[69] Lofgren H, Engquist M, Hoffmann P, Sigstedt B, Vavruch L. Clinical andradiological evaluation of Trabecular Metal and the Smith–Robinson techniquein anterior cervical fusion for degenerative disease: a prospective, randomized,controlled study with 2-year follow-up. Eur Spine J 2010;19:464–73.

[70] Sinclair SK, Konz GJ, Dawson JM, Epperson RT, Bloebaum RD. Host boneresponse to polyetheretherketone versus porous tantalum implants forcervical spinal fusion in a goat model. Spine 2012;37:E571–80.

[71] Frigg A, Dougall H, Boyd S, Nigg B. Can porous tantalum be used to achieveankle and subtalar arthrodesis?: a pilot study. Clin Orthop Relat Res2010;468:209–16.

[72] Levine BR, Fabi DW. Poröse Metalle in orthopädischen Anwendungen – EineÜbersicht [Porous metals in orthopedic applications – a review]. MaterialwissWerkstofftech 2010;41:1001–10.

[73] Marin E, Fedrizzi L, Zagra L. Porous metallic structures for orthopaedicapplications: a short review of materials and technologies. Eur OrthopTraumatol 2010;1:103–9.

[74] Vandenbroucke B, Kruth J-P. Selective laser melting of biocompatible metalsfor rapid manufacturing of medical parts. Rapid Prototyping J2007;13:196–203.

[75] Van Bael S, Chai YC, Truscello S, Moesen M, Kerckhofs G, Van Oosterwyck H,et al. The effect of pore geometry on the in vitro biological behavior of humanperiosteum-derived cells seeded on selective laser-melted Ti6Al4V bonescaffolds. Acta Biomater 2012;8:2824–34.

[76] Balla VK, Banerjee S, Bose S, Bandyopadhyay A. Direct laser processing of atantalum coating on titanium for bone replacement structures. Acta Biomater2010;6:2329–34.

[77] Balla VK, Bodhak S, Bose S, Bandyopadhyay A. Porous tantalum structures forbone implants: fabrication, mechanical and in vitro biological properties. ActaBiomater 2010;6:3349–59.

[78] Zhang YS, Zhang XM, Wang G, Bai XF, Tan P, Li ZK, et al. High strength bulktantalum with novel gradient structure within a particle fabricated by sparkplasma sintering. Mater Sci Eng, A 2011;528:8332–6.

[79] Fox P, Pogson S, Sutcliffe CJ, Jones E. Interface interactions between poroustitanium/tantalum coatings, produced by selective laser melting (SLM), on acobalt–chromium alloy. Surf Coat Technol 2008;202:5001–7.

[80] Thijs L, Montero Sistiaga ML, Wauthle R, Xie Q, Kruth J-P, Van Humbeeck J.Strong morphological and crystallographic texture and resulting yield strengthanisotropy in selective laser melted tantalum. Acta Mater 2013;61:4657–68.

[81] Kruth JP, Levy G, Klocke F, Childs THC. Consolidation phenomena in laser andpowder-bed based layered manufacturing. CIRP Ann Manuf Technol2007;56:730–59.

[82] Wauthle R, Kruth J-P, Montero M, Thijs L, Van Humbeeck J. New opportunitiesfor using tantalum for implants with additive manufacturing. Eur Cells Mater2013;26:15.

[83] Van der Stok J, Van der Jagt OP, Amin Yavari S, DeHaas MF, Waarsing JH, Jahr H,et al. Selective laser melting-produced porous titanium scaffolds regeneratebone in critical size cortical bone defects. J Orthop Res 2013;31:792–9.

[84] van der stok J, Wang H, Amin Yavari S, Siebelt M, Sandker M, Waarsing JH, et al.Enhanced bone regeneration of cortical segmental bone defects using poroustitanium scaffolds incorporated with colloidal gelatin gels for time- and dose-controlled delivery of dual growth factors. Tissue Eng A 2013;19:2605–14.

[85] Campoli G, Borleffs MS, Amin Yavari S, Wauthle R, Weinans H, Zadpoor AA.Mechanical properties of open-cell metallic biomaterials manufactured usingadditive manufacturing. Mater Des 2013;49:957–65.

[86] Amin Yavari S, Wauthle R, van der stok J, Riemslag AC, Janssen M, Mulier M,et al. Fatigue behavior of porous biomaterials manufactured using selectivelaser melting. Mater Sci Eng C Mater Biol Appl 2013;33:4849–58.

[87] Amin Yavari S, van der Stok J, Chai YC, Wauthle R, Tahmasebi Birgani Z,Habibovic P, et al. Bone regeneration performance of surface-treated poroustitanium. Biomaterials 2014;35:6172–81.

[88] ISO, 13782: Implants for surgery – metallic materials – unalloyed tantalum forsurgical implant applications. ISO; 1996.

[89] ISO. 13314: Mechanical testing of metals – ductility testing – compression testfor porous and cellular metals. ISO; 2011.

[90] ISO. 10993–5: Biological evaluation of medical devices – Part 5: Tests forin vitro cytotoxicity. ISO; 2009.

[91] Amin Yavari S, van der Stok J, Ahmadi SM, Wauthle R, Schrooten J, Weinans H,et al. Mechanical analysis of a rodent segmental bone defect model: the effectsof internal fixation and implant stiffness on load transfer. J Biomech2014;47(11):2700–8. http://dx.doi.org/10.1016/j.jbiomech.2014.05.006.

[92] Ahmadi SM, Campoli G, Amin Yavari S, Sajadi B, Wauthle R, Schrooten J, et al.Mechanical behavior of regular open-cell porous biomaterials made ofdiamond lattice unit cells. J Mech Behav Biomed Mater 2014;34:106–15.

[93] Köck W, Paschen P. Tantalum – processing, properties and applications. JOM1989;41:33–9.

[94] Schussler M, Droegkamp RE. ASM handbook. Materials Park, OH: ASMInternational; 1990.

[95] Ritchie RO. Mechanisms of fatigue-crack propagation in ductile and brittlesolids. Int J Fract 1999;100:55–83.

[96] Goodship AE, Kenwright J. The influence of induced micromovement upon thehealing of experimental tibial fractures. J Bone Joint Surg Br 1985;67:650–5.

[97] Morgan EF, Gleason RE, Hayward LN, Leong PL, Palomares KT.Mechanotransduction and fracture repair. J Bone Joint Surg Am2008;90(Suppl. 1):25–30.

[98] Van der Stok J, et al. Osteostatin-coated porous titanium implants improveearly bone regeneration in cortical bone defects in rats. 2014 (submitted forpublication).

[99] Huiskes R, Ruimerman R, van Lenthe GH, Janssen JD. Effects of mechanicalforces on maintenance and adaptation of form in trabecular bone. Nature2000;405:704–6.

本文献由“学霸图书馆-文献云下载”收集自网络，仅供学习交流使用。

学霸图书馆（www.xuebalib.com）是一个“整合众多图书馆数据库资源，

提供一站式文献检索和下载服务”的24 小时在线不限IP

图书馆。

图书馆致力于便利、促进学习与科研，提供最强文献下载服务。

图书馆导航：

图书馆首页 文献云下载 图书馆入口 外文数据库大全 疑难文献辅助工具