Journal of Orthopaedic Translation (2014) 2, 177e184

Available online at www.sciencedirect.com

ScienceDirect

journal homepage: http: / /ees.elsevier .com/jot

ORIGINAL ARTICLE

Degradation and in vitro cellematerialinteraction studies on hydroxyapatite-coated biodegradable porous iron for hardtissue scaffolds

Nurizzati Mohd Daud a, Ng Boon Sing a, Abdul Hakim Yusop a,Fadzilah Adibah Abdul Majid b, Hendra Hermawan a,c,*

a Faculty of Biosciences and Medical Engineering, Universiti Teknologi Malaysia, Johor Bahru, Malaysiab Faculty of Chemical Engineering, Universiti Teknologi Malaysia, Johor Bahru, Malaysiac Department of Mining, Metallurgical and Materials Engineering & CHU de Quebec Research Center,Laval University, Canada

Received 26 May 2014; received in revised form 10 July 2014; accepted 10 July 2014Available online 29 July 2014

KEYWORDSBiodegradable metal;Bone scaffold;Fibroblast;Mesenchymal;Porous iron

* Corresponding author. Dept. of Minþ1 418 6562160.

E-mail address: hendra.hermawan

http://dx.doi.org/10.1016/j.jot.2014.2214-031X/Copyright ª 2014, Chinese

Summary This paper describes degradation and cellematerial interaction studies on hydroxy-apatite (HA)-coated biodegradable porous iron proposed for hard tissue scaffolds. Porous ironscaffolds are expected to serve as an ideal platform for bone regeneration. To couple theirinherentmechanical strength, pureHA andHA/poly(ε-caprolactone) (HA/PCL)were coated ontoporous iron using dip coating technique. TheHA/PCLmixturewas prepared to provide amore sta-ble and flexible coating than HA alone. Degradation of the samples was evaluated by weight lossand potentiodynamic polarisation. Human skin fibroblast (HSF) and human mesenchymal stemcells (hMSC) were put in contact with the samples and their interaction was observed. Resultsshowed that coated samples degraded w10 times slower (0.002 mm/year for HA/PCL-Fe,0.003 mm/year for HA-Fe) than the uncoated ones (0.031 mm/year), indicating an inhibition ef-fect of the coating on degradation. Both HSF and hMSCmaintained high viability when in contactwith the coated samples (100e110% control for hMSC during 2e5 days of incubation), indicatingthe effect of HA in enhancing cytocompatibility of the surface. This study provided early evi-dence of the potential translation of biodegradable porous iron scaffolds for clinical use in ortho-pedic surgery. However, further studies including in vitro and in vivo tests are necessary.Copyright ª 2014, Chinese Speaking Orthopaedic Society. Published by Elsevier (Singapore) PteLtd. All rights reserved.

ing, Metallurgical and Materials Engineering, Laval University, Quebec City, G1V 0A6, Canada. Tel.:

.1@ulaval.ca (H.Hermawan).

07.001Speaking Orthopaedic Society. Published by Elsevier (Singapore) Pte Ltd. All rights reserved.

178 N. Mohd Daud et al.

Introduction

Bone scaffolds have been used to facilitate the regenera-tion of new bone tissues and maintain a balance betweentemporary mechanical support, degradation, and cellgrowth. Separate from the widely chosen polymers, porousbiodegradable metals have recently been viewed as po-tential materials for hard tissue scaffolds [1]. The inherentstrength and ductility owned by metals are the key featuresthat make them more appealing than polymers for hardtissue applications. Biodegradable metal scaffolds haveshown interesting mechanical properties, especiallyYoung’s modulus and toughness, which are close to that ofhuman bone with tailored degradation behaviour. Forinstance, the Young’s modulus of magnesium (Mg;41e45 GPa) is closer to that of cortical bone [2]. Despitehaving higher mechanical properties in fully dense state,(i.e., Young’s modulus Z 211 GPa), pure iron (Fe) strengthscan be rendered having mechanical properties closer tothose of bone by altering its porosity. Porous iron-ephosphorus (FeeP) alloys fabricated through powdermetallurgy have resulted in an elastic modulus of 2.3 GPa,which is comparable to that of typical bone [3,4] Morerecently, porous pure Fe with a pore size of 450 um and 88%porosity exhibited a compressive strength of 0.33 MPa thatfalls within the range of cancellous bone strength [5e8]. Inessence, the porosity and pore sizes of the porous biode-gradable metals can be altered to obtain desired mechan-ical properties, degradation behaviour, and cellematerialinteraction. In addition, the porosity must be inter-connected to provide spaces for osteogenesis while main-taining optimum mechanical properties of the scaffolds [9].

Mg and its alloys are the most studied biodegradablemetals and the recent coating technology has enable Mg tohave a more controllable degradation rate suitable for bonescaffold applications [10e14]. The porous structure of Mghas been proven to play a significant role in cell growth andproliferation [15,16]. Meanwhile, porous Fe was introducedmuch more recently [8,17,18]. The in vitro cytotoxicity teston three types of porous Fe manufactured by replicationmethod, FeeMg, Fe, and Fe-carbon nanotubes, has shownproliferation of osteoblastic cells [18] and the degradationproduct of Fe was proven to be nontoxic to endothelial cells[19]. Despite its appealing in vitro cytocompatibility, theinherent high strength of Fe will easily provide the requiredinitial strength of the scaffolds to stabilize the affectedbone. Compared to Mg or polymeric scaffolds, higherstrength of Fe will allow more flexible control on the porousstructure to meet specific bone strength requirements[1,18].

Cell attachment, migration, differentiation, and prolif-eration in the porous structure are among the importantcellematerial interaction parameters that determine thesuitability of a material for scaffolds. Starting with cellattachment, an attractive-to-cell surface should be pre-pared. Hydroxyapatite (HA) is a well-known bioactiveceramic material having similar chemical composition tohuman bone and has excellent bone bonding ability [20].The use of HA coatings on metallic implants have beenreported to stimulate faster bone cell attachment, result-ing in an improvement healing rate and bone strengthduring the early stage of implantation [21,22]. However, HA

alone is brittle and to overcome this issue, a coating of HAand poly(ε-caprolactone) (PCL) composite was proposed.This resulted in a more stable and flexible coating withoutcracking or delamination compared with the single HAcoating [23]. This study will be the first to investigatecellematerial interaction as valid evidence to propose HA-coated porous biodegradable iron as novel bone scaffolds.Samples of pure porous Fe, HA-coated porous Fe, and HA/PCL-coated porous Fe were tested for degradation andbioactivity. The interaction of human skin fibroblast(HSF1184) and human mesenchymal stem cells (hMSC) onthe coated and uncoated samples were observed. Staticimmersion (weight loss) and potentiodynamic polarisation(PDP) tests were performed to investigate the degradationbehaviour.

Materials and methods

Sample preparation

Interconnected-pores porous pure-Fe sheets (purity 99.9%,pore size Z 450 mm, porosity Z 88%) were kindly providedby Alantum Corporation, Korea. According to the manu-facturer, the porous pure-Fe sheet was made via the poly-mer space holder method. Samples of the porous pure-Fe(10 mm � 10 mm � 1.6 mm) were coated with hydroxy-apatite (HA, SigmaeAldrich, USA) and composite of hy-droxyapatite/poly(ε-caprolactone) (HA/PCL) using a dipcoating method (Dip Coater PTL-MMB, MTI Corp, USA) toproduce HA-coated porous Fe (HA-Fe) and HA/PCL-coatedporous Fe (HA/PCL-Fe), respectively. The HA suspensionwere prepared by dissolving crystalline micro-HA powdersin methanol at 10% weight/volume ratio at room temper-ature and homogeneously stirred for 72 hours. PCL pellets(Mw Z 65,000, SigmaeAldrich) and HA powders (1:1 ratio)were dissolved in chloroform at 15% weight/volume ratioand then homogeneously stirred for 72 hours. The dipcoating process was performed at a constant withdraw anddown speeds of 200 mm/minute five times, and the coatedsamples were let to cure at room temperature for 24 hours[24]. Microstructure and surface morphology of the sampleswere analysed by a scanning electron microscope (SEM,Hitachi TM3000, Japan) coupled with energy-dispersive X-ray spectroscopy (EDX, SwiftED 3000, Oxford Instruments,UK).

Degradation tests

Static immersion test were carried out in minimum essen-tial medium solution (MEM, Gibco, Australia) similarly usedfor cell culture medium. Three samples of pure-Fe, HA-Fe,and HA/PCL-Fe were immersed in 100 mL MEM at 37�C for 7days, 14 days, and 21 days in a CO2 incubator. Weight of thesamples was measured prior to and after the immersiontest. The samples were rinsed in deionized water andethanol and brushed gently as specified by the AmericanSociety for Testing and Materials (ASTM) G1-03 standard[25] followed by air and vacuum drying for 48 hours tocompletely remove all degradation products prior toweighing. Some samples were incubated for 25 days and5 mL of eluates were then taken out and tested for Fe ion

Hydroxyapatite-coated biodegradable porous iron 179

concentration using an inductively-coupled plasma massspectrometry (ICP-MS, PerkinElmer, USA).

PDP testing was carried out using a potentiostat (Ver-sastat 3, Princeton Applied Research, USA) on the samesamples as for immersion but using simulated body fluid(SBF), as suggested by Kokubo and Takadama [26], as theelectrolyte solution at room temperature. The three elec-trodes system was used where graphite, Ag/AgCl (KCl3.5M), and the samples served as the counter, reference,and working electrodes, respectively. Samples with 1 cm2

exposed area were stabilized for 5 minutes at their opencircuit potential (OCP) prior to the test. Tafel curve wasscanned at a range of �0.25 V versus the Ecorr with a rate of1.666 mV/s.

Cellematerial interaction evaluation

Three cellematerial interaction parameters (cell viability,attachment, and morphology) were evaluated. HSF 1184(European Collection of Cell Culture, UK) and hMSC cells(ScienCell Research Labs, USA) were used. Three groups ofsamples (pure-Fe, HA-Fe, and HA/PCL-Fe) were cleaned insterile phosphate buffered saline (PBS) containing 10% an-tibiotics (100 U/100 mg/mL penicillin/streptomycin), rinsedwith a flow of sterile water, and autoclaved at 120�C for 20minutes as final sterilisation steps [27]. The HSF cells werecultured in a MEM supplemented with 10% (v/v) fetal bovineserum (FBS) (Gibco), together with 1% antibiotics. The hMSCcells were cultured in a mesenchymal stem cell medium(MSCm, ScienCell Research Labs) supplemented with 25 mLFBS, 5 mL Pen/strep, and 10 mL of mesenchymal stem cellgrowth supplement (MCGS, ScienCell Research Labs) pro-ducing 500 mL complete medium. Cultures were incubatedin a humidified atmosphere containing 5% CO2 at 37�C.

As much as 2 � 105 cells/mL of HSF and 5 � 104 cells/mLof hMSC were seeded on the samples in a 12-well plateusing direct method assay according to the ASTM F813standard [28]. The viability of the cell toward sample wasmeasured after 24 hours of incubation and was assessed bytrypan blue exclusion test whereas metabolic activity weredetermined based on MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium] assay after 2 days, 3 days, and 5days. Cell attachment on the sample surfaces was evalu-ated under a fluorescence microscope (Fluoview FV1000,Olympus, USA) using HSF cells prepared the same way as forcell viability. The adhered HSF cell was detached by tryp-sinization and further staining by acridine orange/

Figure 1 Surface morphology of fabricated samples: (A) purHA Z hydroxyapatite; PCL Z poly (ε-caprolactone).

propidium iodide (AO/PI, SigmaeAldrich). The PI identifiesdead cells by giving off a red colour whereas the AO, whichis permeable to live cells, identifies live cells by its greencolour [29]. For cell morphology evaluation, the sampleswere seeded with 1 � 105 cells/mL concentrated number ofhMSC, incubated for 2 hours in a 5% CO2 humidified atmo-sphere at 37�C. Then, 2 mL medium was added to each welland incubated for 48 hours. The samples were withdrawnfrom the culture plate and washed with PBS, fixed with2.5% glutaraldehyde for 1 hour, dehydrated sequentially in50%, 70%, 90% and 100% ethanol for 10 minutes each andobserved under the SEM.

All results were expressed as mean � standard deviation(SD) of three replicates. Statistical analysis was done usingSigmaPlot (Systat Software, San Jose, USA) and Student t-test for multiple comparisons. The level of significance wasdetermined when p < 0.05.

Results

Coating characterisation

Fig. 1 shows surface morphology of the pure-Fe, HA-Fe, andHA/PCL-Fe samples. HA and HA/PCL deposits are bothevidently seen to cover the Fe struts and some pores aswell. As measured by the EDX, the deposits were composedof 51.6 wt% O, 5.7 wt% P, 4.5 wt% Ca, 38.3 wt% Fe for HA-Fe, and 36.1 wt% O, 2.9 wt% P, 5.1 wt% Ca, and 8.8 wt%Fe for HA/PCL-Fe. Both coatings are characterized by arough and embossed surface, whereas HA were sparselyembedded in the PCL phase.

Degradation test

Fig. 2A shows weight loss of the samples after immersion inMEM at different time period. The weight loss of pure-Feand HA-Fe increased until 14 days and decreased at 21days. Fig. 2B shows weight loss and Fe ion release of thesamples after 25 days of immersion. The HA-Fe sampleexhibited the highest weight loss with 32.79% followed bythe pure-Fe and HA/PCL-Fe samples which were 8.07% and2.53% respectively. Fe ion concentration was found higherin the media of pure-Fe sample than those of HA/PCL-Feand HA-Fe samples.

Fig. 3 shows typical Tafel curves of the samples obtainedfrom the potentiodynamic polarisation tests in SBF. The

e Fe, (B) HA-Fe, and (C) HA/PCL-Fe samples. Fe Z iron;

Figure 2 Weight loss of the samples are shown. (A) In minimum essential medium solution for different immersion times. (B)After 25 days’ immersion and its corresponding iron ion release.

Figure 3 Potentiodynamic polarisation curves of thesamples.

180 N. Mohd Daud et al.

pure-Fe sample shows active, passive, and transpassivecorrosion behaviours that were not observed in both coatedsamples. The coating could have acted as a barrier foranodic dissolution, hence slowing down the formation of apassive layer. Higher overpotentials were required to breakdown the HA/PCL and HA coating. An interesting phenom-enon was observed for the HA-Fe sample that rendered anOCP to be approximately 200 mV more cathodic comparedto the HA/PCL-Fe and pure-Fe. This may indicate that theHA/PCL coating created a less noble surface towardcorrosion in SBF at the early stage of degradation. Overall,as depicted in the Tafel curves, pure-Fe sample possessesthe highest current density (whighest degradation rate),whereas current densities of HA-Fe and HA/PCL-Fe arealmost comparable. Table 1 details degradation/corrosion

Table 1 Degradation rate of the pure-Fe, HA-Fe, and HA/PCL-Fe samples.

Parameter HA/PCL-Fe HA-Fe Pure-Fe

Ecorr (mV) �643 �835 �650Icorr (A/cm

2) 0.20E-6 0.27E-6 0.26E-5Degradation rate (mm/y) 0.002 0.003 0.031

Fe Z iron; HA Z hydroxyapatite; PCL Z poly(ε-caprolactone).

parameters calculated by Tafel extrapolation. It can beseen that the coated samples degraded about 10 timesslower (0.002 mm/year for HA/PCL-coated, 0.003 mm/yearfor HA-coated) than the uncoated ones (0.031 mm/year).

Cellematerial interaction observation

Fig. 4 shows viability of HSF 1184 and hMSC cultured withthe samples at Day 2, Day 3, and Day 5. A significantdecrease (p < 0.05) of HSF cell viability is observed forpure-Fe and HA-Fe groups (Fig. 4A). Meanwhile, HA/PCL-Fe

Figure 4 Cell viability. (A) Human skin fibroblast 1184 cells.(B) Human mesenchymal stem cells cultured with the samplesat Day 2, Day 3, and Day 5 (n Z 3). *p < 0.05. **p > 0.05.Fe Z iron; HA Z hydroxyapatite; PCL Z poly(ε-caprolactone).

Hydroxyapatite-coated biodegradable porous iron 181

group maintained higher cell viability at all incubation pe-riods than the other groups. The high cell viability of allgroups after 2 days of incubation indicates that Fe ionconcentration had no influence on the cell’s metabolicactivity.

Fig. 5 shows fluorescent images of HSF cells detachedfrom the surface of samples. Cells were observed on thesurface of pure-Fe sample as indicated by white arrows(Fig. 5A). The blue arrows indicate cells that were not wellstained and some cells were lysed as exposed to thecorrosion product on the HA-Fe sample (Fig. 5B). Theaggregated and clump cells were also observed on thesurface of HA/PCL-Fe sample as indicated by yellow arrows(Fig. 5C).

Fig. 6 shows SEM images of the hMSC seeded and incu-bated on the samples for 72 hours. There is no obviouschange on the morphology and distribution of cellsobserved on pure-Fe (Fig. 6A and B) and HA/PCL-Fe samples(Fig. 6E and F). The hMSCs grew well on the surfaces of HA-Fe (Fig. 6C and D). The cells developed an extended filo-podia and long pseudopods as observed on the cell mem-brane. The cell size enlarged, indicating spread to the HA-Fe surface.

Discussion

Coating characteristic and degradation behaviour

Porous iron was proposed for hard tissue scaffolds and itspotentiality was evaluated in the current study. Porouspure Fe, and HA-coated and HA/PCL-coated porous pure Fesamples were compared for their degradation andcellematerial interaction behaviours. From morphologicalobservation (Fig. 1), the overall porosity of both sampleswas not significantly affected because the coating homo-geneously coats the metal struts and no blockage in thepores was observed. The interconnected pores are impor-tant for cell migration and proliferation because they havebeen studied for osteoblasts and mesenchymal stem cells aswell as for vascular formation [30]. However, detail coatingcharacterisation (thickness, adherence, etc.) was not donein this work because the work was focused more oncellematerial interactions.

The coated samples experienced the lowest degradationrate than the uncoated ones as determined by ion releasemeasurement and PDP (Figs. 2 and 3). The decrease in the

Figure 5 Human skin fibroblast cells detached from the followFe Z iron; HA Z hydroxyapatite; PCL Z poly(ε-caprolactone).

degradation rate by coating can be related to the accu-mulation of phosphate products from the HA and MEM, asdetected by the EDX. This is consistent with a phosphatingtreatment study, which observed corrosion inhibition of Mgsamples by phosphate deposition [31], and this effect couldalso be prominent for Fe samples [32]. Furthermore, thesurface of pure-Fe and HA-Fe samples were covered by abrownish hydroxide layer that prevented direct contact ofFe struts with the solution or impeded the diffusion of Feions [33]. The EDX result on the layer composed of 51.6 wt%O and 38.3 wt% Fe. The dissolution of iron in any simulatedbody fluid could form ferrous hydroxide, Fe(OH)3. The lowweight loss of the HA/PCL-Fe sample could be attributed tothe slowly degrading PCL because of its hydrophobic char-acter and high crystallinity [34]. The addition of HA parti-cles into PCL further decreased the degradation rate of PCLby partial neutralisation of the acidic degradation acidicby-product [35]. An in vivo study has demonstrated that thedegradation time of the HA/PCL composite extended to >2years [36]. However, the degradation rate determination byweight loss may cause bias because it also counted theweight loss of the coating. Therefore, Fe ion release wasmeasured (Fig. 2B). As the immersion test was prolonged,the media became more alkaline (pH > 8) and in this con-dition Fe2þ transforms to Fe3þ and forms solid Fe(OH)3. Thisreaction should be more pronounced in the media of theHA-Fe sample, resulting in lesser Fe ion concentrationdetected by the ICP-MS.

Cellematerial interaction

Viability of HSF cells decreased when in contact with thesamples as incubation time was prolonged, and hMSC cellsmaintained its viability over the incubation periods (Fig. 4).With the prolonged incubation, Fe ion concentrationincreased thereby lowered HSF cells’ survival. The staticcondition of incubation retained the toxic degradationproduct in contact with the HSF cells. A dynamic cell cul-ture could increase cell proliferation because of a constantsupply of fresh medium and removal of toxic degradationproduct [37]. Meanwhile, the viability of hMSC in responseto the samples shows more positive results than those ofHSF cells results. The hMSC maintained their viability at100% versus control up to 5 days incubation (Fig. 4B). Sta-tistical analysis indicates that the cell proliferation on theHA-Fe group was significantly higher than other groups

ing surfaces: (A) pure-Fe, (B) HA/PCL-Fe, (C) HA-Fe samples.

Figure 6 Morphologies of hMSC cells before and after 3 days’ incubation. (A, B) Pure Fe sample. (C, D) HA-Fe sample. (E, F) HA/PCL-Fe sample. Fe Z iron; HA Z hydroxyapatite; PCL Z poly(ε-caprolactone).

182 N. Mohd Daud et al.

(p < 0.05) after 1 day of incubation with hMSC and 5 days ofincubation with HSF cells. The release of calcium andphosphate from the dissolved HA affects the proliferationand metabolic activities of the cells. This is consistent witha study done by Li et al [38] that showed better cyto-compatibility of osteoblast on iron (III) doped HA using thewet chemical method.

Both cells were observed to preferably attach and growactively on the coated surfaces, especially on the HA-coated surface (Fig. 5). This observation showed that anumber of cells were well adhered to bioactive poroussurfaces. Bioactive materials are capable to form bondsthrough bone-like HA layers and biological collagen withmaterial surface in physiological fluid. Within 2 days ofculture, the cell density on the surfaces was significant,suggesting that the coated porous surfaces had favourablebiological properties [39]. Furthermore, the MSC preferredto align on the HA-coated surface indicated by its growing

filopodia (Fig. 6). This could be related to the surfacewettability of HA-Fe sample because HA is known for itsstrong hydrophilicity [40]. The hydrophilicity of the HA-Fesurface may have fallen within the optimal wettabilityrange that is good for cell adhesion [41]. Overall, thebioactivity of iron was attained by HA coating. This workprovides early evidence of the potentiality of HA-coatedporous Fe to be used as bone scaffolds. Some potentialapplications include bone defect repair and cancerous bonetreatment, especially when the coating is loaded withanticancer drugs. However, further investigations should beperformed to give more information on biocompatibilitysuch as more in vitro and in vivo cytotoxicity of this Fe-based porous scaffold. Iron is known as essential for thefundamental biological process of cells and body and isbeneficial for many enzymes and the main component ofhemoglobin. Its degradability and high ductility makes Fepreferable for biodegradable stent and has shown positive

Hydroxyapatite-coated biodegradable porous iron 183

results in a study using rabbits [42]. However, excess Fe canbe deleterious to the human body because it can induce theformation of reactive oxygen intermediate by a series ofprocesses that eventually may lead to various pathologicalconditions such as neurodegenerative disease [43]. More-over, not much information is available on bone-relatediron toxicity.

Conclusion

HA-coated porous Fe was proposed for hard tissue scaffoldsand its potentiality was evaluated in the current study. HA-coated and HA/PCA samples were found to degrade slowerthan the uncoated ones, indicating the inhibiting effect ofthe coating on degradation process. HSF and hMSC werefound to maintain high viability when in contact with thecoated samples indicating the effect of hydroxyapatite inenhancing cytocompatibility of the surface. Moreover, thehMSC preferred to attach on the HA-coated samples indi-cated by the filopodia formation. Finally, this study pro-vided evidence of the early cytocompatibility of porous Fe,and further studies will focus on its characterisation forhard tissue replacement. The authors confirm that thereare no known conflicts of interest associated with thispublication and there has been no significant financialsupport for this work that could have influenced itsoutcome.

Acknowledgements

This study was supported by Malaysian Minister of Educationand Universiti Teknologi Malaysia through the FundamentalResearch Grant Scheme (FRGS) number R.J130000.7836.4F123. The authors would like to thank Mrs. Nor Samsiah(TCERG group) for the guidance and laboratory facilitiesthroughout the experiments, and Alantum Corporation, Koreafor providing porous iron sheets.

References

[1] Yusop AH, Bakir AA, Shaharom NA, Abdul Kadir MR,Hermawan H. Porous biodegradable metals for hard tissuescaffolds: a review. Int J Biomater 2012;2012:10.

[2] Staiger MP, Pietak AM, Huadmai J, Dias G. Magnesium and itsalloys as orthopedic biomaterials: a review. Biomaterials2006;27:1728e34.

[3] Quadbeck P, Hauser R, Kummel K, Standke G, Stephani G,Nies B, et al. Iron based cellular metals for degradable syn-thetic bone replacement. In: PM2010 World Congress-PM Bio-materials, Florenz, Italy; 2010.

[4] Black J, Hastings G. Handbook of biomaterial properties.Springer; 1998.

[5] Teo J, Wang SC, Teoh SH. Preliminary study on biomechanicsof vertebroplasty: a computational fluid dynamics and solidmechanics combined approach. Spine 2007;32:1320e8.

[6] Rahman CV, Kuhn G, White LJ, Kirby GT, Varghese OP,McLaren JS, et al. PLGA/PEG-hydrogel composite scaffoldswith controllable mechanical properties. J Biomed Mater ResB 2013;101:648e55.

[7] Misch CE, Qu Z, Bidez MW. Mechanical properties of trabecularbone in the human mandible: implications for dental implant

treatment planning and surgical placement. J Oral MaxillofacSurg 1999;57:700e6.

[8] Yusop AH, Hermawan H. Synthesis and development ofpolymers-infiltrated porous iron for temporary medical im-plants: a preliminary result. Adv Mater Res 2013;686:331e5.

[9] Karageorgiou V, Kaplan D. Porosity of 3D biomaterial scaffoldsand osteogenesis. Biomaterials 2005;26:5474e91.

[10] Wen CE, Yamada Y, Shimojima K, Chino Y, Hosokawa H,Mabuchi M. Compressibility of porous magnesium foam: de-pendency on porosity and pore size. Mater Lett 2004;58:357e60.

[11] Zhuang H, Han Y, Feng A. Preparation, mechanical propertiesand in vitro biodegradation of porous magnesium scaffolds.Mater Sci Eng C 2008;28:1462e6.

[12] Wen CE, Mabuchi M, Yamada Y, Shimojima K, Chino Y,Asahina T. Processing of biocompatible porous Ti and Mg.Scripta Mater 2001;45:1147e53.

[13] Tang J, Wang JL, Xie XH, Zhang P, Lai YX, Li YD, et al. Surfacecoating reduces degradation rate ofmagnesiumalloy developedfor orthopaedic applications. J Orthop Trans 2013;1:41e8.

[14] Nguyen TL, Staiger MP, Dias GJ, Woodfield TBF. A novelmanufacturing route for fabrication of topologically-orderedporous magnesium scaffolds. Adv Eng Mater 2011;13:872e81.

[15] Simon JL, Roy TD, Parsons JR, Rekow ED, Thompson VP,Kemnitzer J, et al. Engineered cellular response to scaffoldarchitecture in a rabbit trephine defect. J Biomed Mater Res A2003;66:275e82.

[16] Tan L, Gong M, Zheng F, Zhang B, Yang K. Study on compres-sion behavior of porous magnesium used as bone tissue engi-neering scaffolds. Biomed Mater 2009;4:015016.

[17] Wegener B, Sievers B, Utzschneider S, Muller P, Jansson V,Roßler S, et al. Microstructure, cytotoxicity and corrosion ofpowder-metallurgical iron alloys for biodegradable bonereplacement materials. Mater Sci Eng B 2011;176:1789e96.

[18] Ori�nakova R, Ori�nak A, Bu�ckova LM, Giretova M, Medvecky �L,Labbanczova E, et al. Iron based degradable foam structuresfor potential orthopedic applications. Int J Electrochem Sci2013;8:12451e65.

[19] Zhu S, Huang N, Xu L, Zhang Y, Liu H, Sun H, et al. Biocom-patibility of pure iron: in vitro assessment of degradation ki-netics and cytotoxicity on endothelial cells. Mater Sci Eng C2009;29:1589e92.

[20] Ong JL, Chan DCN. Hydroxyapatite and their use as coatings indental implants: a review. Crit Rev Biomed Eng 2000;28:667e707.

[21] Wang Y, Liu L, Guo S. Characterization of biodegradable andcytocompatible nano-hydroxyapatite/polycaprolactoneporous scaffolds in degradation in vitro. Polym Degrad Stabil2010;95:207e13.

[22] Li T, Lee J, Kobayashi T, Aoki H. Hydroxyapatite coating bydipping method, and bone bonding strength. J Mater Sci MaterMed 1996;7:355e7.

[23] Jo JH, Li Y, Kim SM, Kim HE, Koh YH. Hydrox-yapatite/Poly(epsilon-Caprolactone) double coating on mag-nesium for enhanced corrosion resistance and coatingflexibility. J Biomater Appl 2013;28:617e25.

[24] Mohd Yusoff MF, Kadir MRA, Iqbal N, Hassan MA, Hussain R.Dipcoating of poly (ε-caprolactone)/hydroxyapatite compos-ite coating on Ti6Al4V for enhanced corrosion protection. SurfCoat Technol 2014;245:102e7.

[25] ASTM G1e03. Standard practice for preparing, cleaning, andevaluating corrosion test specimens. West Conshohocken:ASTM; 2003.

[26] Kokubo T, Takadama H. How useful is SBF in predicting in vivobone bioactivity? Biomaterials 2006;27:2907e15.

[27] Sukmana I, Vermette P. Polymer fibers as contact guidance toorient microvascularization in a 3D environment. J BiomedMater Res A 2010;92:1587e97.

184 N. Mohd Daud et al.

[28] ASTM F813. Standard practice for direct contact cell cultureevaluation of materials for medical devices. West Con-shohocken: ASTM; 2007.

[29] Crawford JM, Milford E, Case K, Braunwald NS. Vital fluores-cent staining of human endothelial cells, fibroblasts, andmonocytes: assessment of surface morphology. Ann ThoracSurg 1989;48:S100e1.

[30] Kuboki Y, Takita H, Kobayashi D, Tsuruga E, Inoue M, Murata M,et al. BMP-induced osteogenesis on the surface of hydroxy-apatite with geometrically feasible and nonfeasible struc-tures: topology of osteogenesis. J Biomed Mater Res 1998;39:190e9.

[31] Xu L, Zhang E, Yang K. Phosphating treatment and corrosionproperties of MgeMneZn alloy for biomedical application. JMater Sci Mater Med 2009;20:859e67.

[32] Zhang E, Chen H, Shen F. Biocorrosion properties and bloodand cell compatibility of pure iron as a biodegradablebiomaterial. J Mater Sci Mater Med 2010;21:2151e63.

[33] Moravej M, Purnama A, Fiset M, Couet J, Mantovani D. Elec-troformed pure iron as a new biomaterial for degradablestents: in vitro degradation and preliminary cell viabilitystudies. Acta Biomaterialia 2010;6:1843e51.

[34] Causa F, Netti P, Ambrosio L, Ciapetti G, Baldini N, Pagani S,et al. Poly-ε-caprolactone/hydroxyapatite composites forbone regeneration: in vitro characterization and humanosteoblast response. J Biomed Mater Res A 2006;76:151e62.

[35] Li H, Chen Y, Xie Y. Photo-crosslinking polymerization toprepare polyanhydride/needle-like hydroxyapatite biode-gradable nanocomposite for orthopedic application. MaterLett 2003;57:2848e54.

[36] Coombes AGA, Rizzi SC, Williamson M, Barralet JE, Downes S,Wallace WA. Precipitation casting of polycaprolactone forapplications in tissue engineering and drug delivery. Bio-materials 2004;25:315e25.

[37] Farack J, Wolf-Brandstetter C, Glorius S, Nies B, Standke G,Quadbeck P, et al. The effect of perfusion culture on prolif-eration and differentiation of human mesenchymal stem cellson biocorrodible bone replacement material. Mater Sci Eng B2011;176:1767e72.

[38] Li Y, Widodo J, Lim S, Ooi C. Synthesis and cytocompatibilityof manganese (II) and iron (III) substituted hydroxyapatitenanoparticles. J Mater Sci 2012;47:754e63.

[39] Bogdanski D, Epple M, Esenwein SA, Muhr G, Petzoldt V,Prymak O, et al. Biocompatibility of calcium phosphate-coated and of geometrically structured nickeletitanium(NiTi) by in vitro testing methods. Mater Sci Eng A 2004;378:527e31.

[40] Tihan TG, Ionita MD, Popescu RG, Iordachescu D. Effect ofhydrophilicehydrophobic balance on biocompatibility of poly(methyl methacrylate) (PMMA)ehydroxyapatite (HA) compos-ites. Mater Chem Phys 2009;118:265e9.

[41] Cheng J, Liu B, Wu YH, Zheng YF. Comparative in vitro studyon pure metals (Fe, Mn, Mg, Zn and W) as biodegradablemetals. J Mater Sci Technol 2013;29:619e27.

[42] Peuster M, Hesse C, Schloo T, Fink C, Beerbaum P, vonSchnakenburg C. Long-term biocompatibility of a corrodibleperipheral iron stent in the porcine descending aorta. Bio-materials 2006;27:4955e62.

[43] Papanikolaou G, Pantopoulos K. Iron metabolism and toxicity.Toxicol Appl Pharmacol 2005;202:199e211.