H O S T E D B Y

Progress in Natural Science

Materials International

Available online at www.sciencedirect.com

http://dx.doi.org1002-0071/& 20

nCorrespondinE-mail addrePeer review u

Progress in Natural Science: Materials International 24 (2014) 466–471

Original Researchwww.sciencedirect.com

www.elsevier.com/locate/pnsmi

In vitro and in vivo studies on biodegradable magnesium alloy

Lida Houa, Zhen Lia, Yu Pana, Li Dua, Xinlin Lia, Yufeng Zhenga,b, Li Lia,n

aCenter for Biomedical Materials and Engineering, Harbin Engineering University, Harbin 150001, ChinabDepartment of Materials Science and Engineering, College of Engineering, Peking University, Beijing 100871, China

Received 2 July 2014; accepted 1 September 2014Available online 4 November 2014

Abstract

The microstructure, mechanical property, electrochemical behavior and biocompatibility of magnesium alloy (BioDe MSM™) were studied inthe present work. The experimental results demonstrated that grain refining induced by extrusion improves the alloy strength significantly from162 MPa for the as-cast alloy to 241 MPa for the as-extruded one. The anticorrosion properties of the as-extruded alloy also increased.Furthermore, the hemolysis ratio was decreased from 4.7% for the as-cast alloy to 2.9% for the as-extruded one, both below 5%. BioDe MSM™alloy shows good biocompatibility after being implanted into the dorsal muscle and the femoral shaft of the New Zealand rabbit, respectively, andthere are no abnormalities after short-term implantation. In vivo observation indicated that the corrosion rate of this alloy varies with differentimplantation positions, with higher degradation rate in the femur than in the muscle.& 2014 Chinese Materials Research Society. Production and hosting by Elsevier B.V. All rights reserved.

Keywords: Magnesium alloy; Extrusion; Performance; Biocompatibility

1. Introduction

Recently, the materials extensively used in the clinical applica-tions mainly include stainless steel, cobalt-based alloy and titaniumalloy. However, these biomedical metallic materials have their owndisadvantages. For example, wear debris and metal ions (Cr, Ni, V,Al etc.) [1–3] resulting from abrasion of the material in vivo mighthave cytotoxicity, cause local hypersensitivity or inflammation [4],decrease biocompatibility and lead to tissue necrosis. Additionally,the unmatched mechanical performance especially the elasticmodule of these conventional materials shall induce stress shieldingeffect [5,6], and increase the liability of secondary bone fracture[7]. Moreover, these materials could not be biodegraded in vivo, sothey need to be taken out through operation after the human bodyfunction restoration [8]. The biodegradable materials under devel-opment nowadays have been focused on the polymers and someceramics. However, for the former, the mechanical strength isusually poor, and brittleness is a big problem for the latter.

/10.1016/j.pnsc.2014.09.00214 Chinese Materials Research Society. Production and hosting by

g author. Tel./fax: þ86 45 8251 8644.ss: lili_heu@hrbeu.edu.cn (L. Li).nder responsibility of Chinese Materials Research Society.

As a new revolutionary biomedical metallic material, Mg alloyhas been put special focus [9]. It could be used as biodegradablematerial in medical field since Mg alloy has active chemicalproperty [10,11] and is liable to be corroded in the physiologicalenvironment after implantation without causing any toxic and sideeffects [12]. Mg ions play a very important role in metabolism[13,14], of which the content is only inferior to potassium ions inhuman body. As a main constituent of living body bone, Mgcould boost the formation of teeth and bone and play a regulatingrole in the metabolism of the mineral substance of the bone [15].Further, mechanical performances such as plasticity, stiffness andprocessability of Mg alloys are superior to those of the bioresorb-able polymers, e.g. polylactic acid, in extensive clinical use [16].Mg and its alloys are suitable to be used in the blood vesselintervention and orthopedics due to the close elastic modulus tothat of bone [17–20]. The most delighted thing to the researchersis that to date there is no adverse case report globally on the Mgalloys implanted in blood vessel or bone in the animal studies [21].The elastic deformation capability of Mg alloy is acknowl-

edged to be poor. However, Mg alloy with high elasticity couldbe obtained by extrusion, because extrusion could induce moreintensive three-dimensional compressive stress in comparison

Elsevier B.V. All rights reserved.

L. Hou et al. / Progress in Natural Science: Materials International 24 (2014) 466–471 467

with rolling [22]. Extrusion could also increase the strength andrefine the grain, and accordingly the as-extruded Mg alloy ispromising [23,24].

In this study, a new medical biodegradable Mg alloy (BioDeMSM™) was developed, and the feasibility of BioDe MSM™alloy as biodegradable implant was investigated through micro-structure observation, mechanical property testing, corrosion test-ing, in vitro hemocompatibility analysis and in vivo animal studies.

2. Materials and methods

2.1. Alloy melting

The BioDe MSM™ alloy composition used in the present workwas Mg–3 wt%Sn–0.5 wt%Mn. The pure Mg ingots, Mg–Mnmaster alloy and pure Sn ingots, calculated according to the aboveformula were added to the electric-resistance crucible and heated toapproximate 740 1C with protection of RJ-2 covering agent. Afterthe alloys were fullly melted, the melt was stirred for 10 min forhomogenization and then remained for 10 min. After that, the alloymelt was eventually casted with the protection of mixture gas ofCO2/SF6 at 720 1C.

2.2. Extrusion process

Four-column hydraulic press (YA32-200A) manufacturedby Tianjin Metal Forming Machine Tool Works was utilized inthe present work. The raw material size was ∅40 mm in diameterand 250 mm in length, Forward extrusion was adopted at thetemperature of 370 1C, with a extrusion rate of 1.5 m/min, anextrusion ratio of 20 and lubrication of colloidal graphite.

2.3. Microstructure and phase analysis

Axiovert 200 MAT metallurgical microscope was used formicrostructure observation. The extruded samples were degreasedwith acetone firstly, and then grinded with 1000#, 1500# and2000# abrasive paper followed by polishing. The polished samplesurface was first eroded with 3% nitric acid for 10 s, and then with5% picric acid for 10–20 s.

Philips X0Pert PRO X-ray diffraction system was used toanalyze the phase constitution of the as-extruded Mg alloys withCuKα as X ray source, Kα1 with λ of 1.541, Kα2 with λ of 1.544,Kα1/Kα2=0.5 and the scanning angle range as 20–801. Jade5 softwas used for result analysis.

2.4. Mechanical property testing

Tensile properties were tested using universal (Instron 3365,USA) with a crossing head rate of 0.5 mm/min. At least 3effective data were obtained for each sample group to ensurethe statistical analysis.

2.5. Electrochemical corrosion test

Galvanostat/potentiostat (EG&G PAR 273) was used forcorrosion test. A three-electrode system was employed which

was composed of a counter electrode of platinum foil, a referenceelectrode of saturated calomel electrode and a work electrodeof tested sample. The electrolyte volume was 200 ml and theexposure area was 0.283 cm2 for the test samples. The scanningpotential range was �2.0–�0.8 V with a scanning rate of0.01 mV/s.

2.6. Hemolysis

The cleaned samples were in the silanized glass tube. 10 ml0.9%NaCl solution was added in to the tube, and then kept in a37 1C water bath for 30 min. 0.2 ml fresh diluted blood withanticoagulant was added to the tube, shaken slightly for bettermix and kept in the water bath for 60 min, and then becentrifuged for 5 min. In the same condition, 10 ml 0.9%NaClsolution with 0.2 ml fresh diluted blood with anticoagulant wastaken as negative control, and 10 ml distilled water with 0.2 mlfresh diluted blood with anticoagulant was taken as positivecontrol. The supernatants after centrifugation were transferredto cuvettes for absorbency testing using spectrophotometer at awavelength of 545 nm. Hemolysis was calculated according tothe following formula:

Hemolysis ¼ ½AbsorbancyðtestÞ�Absorbancyðnegative controlÞ�=½Absorbancyðpositive controlÞ�Absorbancyðnegative controlÞ� � 100%

2.7. Animal testing

Ten adult rabbits (provided by the Animal ExperimentCenter of the 2nd affiliated Hospital of Harbin MedicalUniversity) with 2.0–2.5 kg weight were randomly dividedinto 3 groups with two rabbits in each group (marked groupA and group B for the test samples and a control group). Mgalloy samples were implanted in the dorsal muscle in group Aand in the femoral bone in group B.Blood samples for biochemical test (UniCelDxC 800, BECK-

MAN COULTERCo. Ltd.) were taken from marginal ear vein ofthe rabbits before experiment and at different implantation time-points. X-ray examination (MIKASA HF100H, MIKASA X-RAYCo. Ltd.) was carried out on the implanted area at different follow-up times for observation of the implant degradation. Rabbitswere sacrificed at the planned time points using 0.3% sodiumpentobarbital solution for overdose an aesthesia, and the organtissue samples of the heart, liver, kidney, spleen etc. were obtained.Pathological examination with HE staining was conducted forobservation of any pathological changes around the implant.

3. Results and discussion

3.1. Microstructures

Fig. 1 shows the metallographic structure of BioDe MSM™alloy at different conditions. It can be seen that the micro-structure of as-cast alloy is inhomogeneous dendrite with thegrain size differs greatly. Secondary phase mainly precipitatesin the grain boundary and in the form of small particles of

Fig. 1. Microstructure patterns of BioDe MSM™ alloy: (a)as-cast and (b)as-extruded.

Fig. 2. XRD patterns of BioDe MSM™ alloy.

0

50

100

150

200

250

300

350

400

Elo

ngat

ion%

ElongationUTS

Stre

ngth

/MPa

as-cast as-extruded

0

5

10

15

20

25

30

Fig. 3. Mechanical properties of BioDe MSM™ alloy in different conditions.

-1.8

-1.6

-1.4

-1.2

-1.0

-0.8

Pote

ntia

l (V

vs S

CE

)

As-castAs-extruded

L. Hou et al. / Progress in Natural Science: Materials International 24 (2014) 466–471468

black after hot extrusion. In addition, the microstructure isobviously refined and the grain size is smaller than the as-caststructure. This is attributed to magnesium alloy's dynamicrecrystallization in hot extrusion process [25] accompanied bythe large deformation and high deformation temperature. Fig. 2displays the XRD spectra of BioDe MSM™ alloy at differentconditions, and the secondary phase can be indexed as Mg2Sn.

1E-6 1E-5 1E-4 1E-3 0.01

-2.0

Curren Density (A/cm2))

Fig. 4. Potentiodynamic polarization curves of BioDe MSM™ alloys indifferent conditions.

Table 1Average electrochemical parameters of BioDe MSM™ alloys in differentcondition.

BioDe MSM™ alloy Ecorr (V) Icorr (μA/cm2)

3.2. Mechanical property

Fig. 3 shows tensile test results of as-cast and as-extrudedBioDe MSM™ alloy. It can be seen that the tensile strength ofas-extruded BioDe MSM™ alloy is higher than that of the as-cast one with an increment of 80 MPa. As shown in Fig. 1, finegrain was obtained in the extruded alloy due to dynamicrecrystallization during hot extrusion, resulting in grain refin-ing strengthening. After extrusion, Mg alloys exhibits highductility and the elongation rises to 22%.

As-cast �1.80 26.9As-extruded �1.61 2.31

3.3. Electrochemical measurement

Fig. 4 shows the potentiodynamic polarization curves ofBioDe MSM™ alloy in as-cast and as-extruded states. Asshown in Fig. 4, as-cast and as-extruded Mg alloys show quitesimilar polarization behavior. The corrosion potential of as-

extruded alloy is higher than that of as-cast alloy, but there isno obvious difference between them. Table 1 presents theaverage electrochemical parameters calculated for the BioDeMSM™ alloy according to AS™ G102. It is obvious that the

Fig. 5. The X-ray imaging of BioDe MSM™ samples inter-femur bone implanted in rabbit.

L. Hou et al. / Progress in Natural Science: Materials International 24 (2014) 466–471 469

corrosion potential (Ecorr) of as-extruded BioDe MSM™ alloy(�1.61 V) is superior to that of as-cast BioDe MSM™ alloy(�1.80 V), and the result of corrosion current (Icorr) indicatesthat as-extruded BioDe MSM™ alloy (2.31 μA/cm2) ismore resistant to corrosion than as-cast alloy (26.9 μA/cm2).According to the electrochemical results listed in the Table 1,as-extruded BioDe MSM™ alloy exhibits excellent antic-orrosion properties over as-cast magnesium alloy.

3.4. Blood compatibility

The hemolysis rate of the as-cast BioDe MSM™ alloy is4.7%, which is in the acceptable range of human body (5%).The hemolysis rate of the as-extruded BioDe MSM™ alloy is2.9%. For both the as-cast and as-extruded BioDe MSM™alloy, the hemolysis is below the recommended value of 5%according to ISO 10993-4:2002.

3.5. Animal implant experiments

Fig. 5 shows the X-ray manifestations at different time pointsafter the implantation of the extruded BioDe MSM™ alloy.A large amount of gas aggregated around magnesium alloy rod,as shown in Fig. 5(b) and (h).This is because of generatedhydrogen gas, which derived from reaction with body fluid forthe implanted the magnesium alloy. Gas accumulated in the fibrouscapsule to form a visible air mass because of the hysteresis andlimitedness of metabolism, thus the amount of gas formed in perunit time was too large, beyond the ability of the body to normallymetabolize the gas, to be effectively metabolized. However, as timeprolonged, the gas mass appearing in the early stage of implanta-tion disappeared (as shown in Fig. 5(e) and (i)), illustrating that thehydrogen produced by magnesium alloy degradation could benormally metabolized and excreted, which was considered to be

realized through some ways of the body, such as gas was absorbedby the body into the blood and exhausted through the lung, and thecomparison of X ray examination suggested that gas around thefemur disappeared earlier than that around back muscles, and itmay be caused by the activity of experimental animals. The bloodcirculation in the implantation sites (the outer end of femur) wasexuberant and the metabolism of the gas was faster. Witte et al.[26] and Zhang et al. [27] also found the gas mass in the fibrouscapsule around the magnesium alloy in animal experiment, whichcan be rapidly absorbed, and the gas release in the degradationprocess of magnesium did no harm to the animal body. More-over, Gao et al. [28] indicated that an appropriate release rateof hydrogen would not cause damage to the body, and onthe contrary, it played a role in eliminating free radical andinflammation.As can be seen in Fig. 5, magnesium alloy rod implanted

into different positions showed different degradation rates. Thedegradation rate of the rod implanted into subcutaneous tissuewas relatively slow. The alloy rod almost shows no signs ofdegradation in the second week. Obvious degradation featurewas observed on the surface of alloy rod at the fourth weekand the sixth week. The implant is still remained in the tenthweek, but the volume was reduced in a large scale, yet theremainder was still intact. The alloy implanted into femurexhibited a more rapid corrosion rate, as obvious feature ofdegradation could be observed in the fourth week. The alloyrod deepening into the bone marrow cavity displayed fuzzyand rough surface, and the alloy embedded deeply into thebone cavity degraded seriously in the sixth week, with smallerdiameter and serious fuzziness and roughness. Until the tenthweek, the alloy rod could not be fully developed and the mainpart of the rod was degraded, leaving only a small amount ofresidual alloy in irregular shape. It is considered that thedifferent degradation behavior was derived from the different

L. Hou et al. / Progress in Natural Science: Materials International 24 (2014) 466–471470

implantation position in the body. The slower degradationspeed of the alloy rod was caused by the wrapping by thetissue when the material was implanted into the back. Whenimplanted into femur, the degradation speed was faster becausemost of the alloy bar was located in the bone marrow cavityand it suffered from scouring by body fluid. Yang et al. [29]carried out an AZ31 alloy experiments in vivo and proved thatthe different environment of the implantation played a impor-tant role in the degradation of magnesium alloy in vivo.

In previous work [30], magnesium ion concentration in serumfluctuate each time point within 1.15–1.75 mmol/L range, wherethe normal range of serum magnesium concentration in experi-mental animals is 0.82–2.22 mmol/L. Serum magnesium concen-trations in the present experimental animals during the differentperiods after surgery is shown in Fig. 6. Seen from Fig. 6, Mg2þ

concentration at 1 day post implantation reaches a peak concentra-tion between 1.6 and 1.8 mmol/L, then decreases in the followingthe other days. The degradation of BioDe MSM™ alloy implants

0 7 14 21 28 35 42 49 560.8

1.0

1.2

1.4

1.6

. 1.8

2.0

Mg2+

Conc

entr

atio

n/m

mol

L-1

Implanting Time/d

subcutaneousorthopedicscontrol

Fig. 6. Serum magnesium concentrations in the experimental animals duringthe different periods after surgery.

Heart Liver

Subc

utan

eous

O

rtho

pedi

cs

a b

e f g

c

Fig. 7. Histological images of abdominals rabbits organs after 4 week

has little effect on the blood concentration of magnesium ion,which indicates liver and kidney function in experimental animalsare in good condition and the excess Mg2þ was excreted inexperimental animals. Hartwig [25] believed that intake of a highconcentration of magnesium ions would not cause adversereactions, Powerful excretory system of kidney and storage bufferfunction of bones can make the body to maintain balance serummagnesium concentration.Fig. 7 is the histological images of rabbit abdominal organs after

4 weeks of inter-muscle implantation of BioDe MSM™ alloysamples, including heart (Fig. 7(a) and (e)), liver (Fig. 7(b) and (f)),kidneys (Fig.7(c) and (g)), and spleen (Fig.7(d) and (h)). All organsshow no obvious abnormalities. The cell structure does not changein the morphology, and no inflammatory cell infiltration. Theseresults indicate that BioDe MSM™ alloy showed good biocom-patibility with animal organs (heart, liver, kidney, spleen). It can beinferred that the alloy implanted in animals at the early stage isbiosafe, experimental animals is tolerated with degradation pro-ducts of BioDe MSM™ alloy.

4. Conclusions

(1)

s of

The microstructure of extruded BioDe MSM™ alloy becomesmore homogeneous compared to that of as-cast alloy and thegrains are significantly refined. Being different from theuneven distribution (granular) in as-cast BioDe MSM™ alloy,the precipitated Mg2Sn phases are distributed along the grainboundary in the form of small particles in extruded alloy.

(2)

The comprehensive mechanical properties of BioDeMSM™ alloy have been notably improved by the extru-sion. The tensile strength and elongation of the extrudedalloy are 241 MPa and 22%, respectively. Besides, theanticorrosion properties of extruded BioDe MSM™ alloyare superior to that of as-cast alloy.

(3)

In vivo test indicates that magnesium concentration in theserum fluctuates in the normal range, and there are no notably

Kidney Spleen

h

d

inter-muscle implantation of magnesium alloy samples.

L. Hou et al. / Progress in Natural Science: Materials International 24 (2014) 466–471 471

histological changes for the various organs and tissues. Thisresult suggests that the implantation of BioDe MSM™ alloydo not produce adverse impacts on body fluid circulation,immune and urinary system of the studied animals, confirminghigh biosafety of BioDe MSM™ alloy. Due to the goodmechanical properties, blood biocompatibility and biosafety,the extruded BioDe MSM™ shows great potential for novelmedical implant material.

Acknowledgments

This work was supported by the National Natural Science Fundfor Young Scientists of China (Grant no. 51301049), the NationalHigh Technology Research and Development Program of China(863 Program, Grant no. 2009AA03Z423), the FundamentalResearch Funds for the Central Universities.

Reference

[1] K.S. Katti, Colloids Surf. B 39 (2004) 133–142.[2] S.B. Goodman, Biomaterials 28 (2007) 5044–5048.[3] M. Malliti, P. Page, C. Gury, E. Chomette, F. Nataf, F.X. Roux, Neurosurgery

54 (2004) 599–604.[4] T.B. Kinraide, U. Yermiyahu, J. Inorg. Bio-Chem. 101 (2007) 1201–1213.[5] Y.B. Ren, K. Yang, Y. Liang, Mater. Rev. 16 (2002) 12–15.[6] O. Böstman, H. Pihlajamäki, Biomaterials 21 (2000) 2615–2621.[7] J. Nagels, M. Stokdijk, P.M. Rozing, J. Shoulder Elb. Surg. (2003) 35–39.[8] H.R. Tao, C. Jiang, Int. J. Orthop. 29 (2008) 293–294.[9] H. Hermawan, D. Dubé, D. Mantovani, J. Biomed., Mater. Res. A 93

(2011) 1–11.

[10] C.M. Serre, M. Papillard, P. Chavassieux, J.C. Voegel, G.J. Boivin,J. Biomed. Mater. Res. 42 (1998) 626–633.

[11] H. Kuwahara, Y. Al-Abdullat, N. Mazaki, S. Tsutsumi, T. Aizawa,Mater. Trans. JIM 42 (2001) 1317–1321.

[12] F. Witte, N. Hort, C. Vogt, S. Cohen, K.U. Kainer, R. Willumeit,F. Feyerabend, Curr. Opin. Solid State Mater. Sci. 12 (2008) 63–72.

[13] H.C. Lukaski, Am. J. Clin. Nutr. 72 (2000) 585–593.[14] Z.C. Zhang, Z.Y. Xu, S.J. Zhang, Stud. Trace Elem. Health 23 (2006) 67–69.[15] J. Vormann, Mol. Asp. Med. 24 (2003) 27–37.[16] M.P. Staiger, A.M. Pietak, J. Huadmai, G. Dias, Biomaterials (2006)

1728–1734.[17] E.D. Mcbride, J. Am. Med. Assoc. 111 (1938) 2464–2467.[18] F. Witte, V. Kaese, H. Haferkamp, H.E. Switzer, A. Meyer-Lindenberg,

C.J. Wirth, H. Windhagen, Biomaterials 26 (2005) 3557–3563.[19] Y. Wan, G. Xiong, H. Luo, H.F. He, Y. Huang, X. Zhou, Mater. Des. 29

(2008) 2034–2037.[20] Z. Li, X. Gu, S. Lou, Y. Zheng, Biomaterials 29 (2008) 1329–1344.[21] Y.F. Zheng, G.X. Nan, N. Li, W.R. Zhou, Mater. China 30 (2011) 30–39.[22] C.Q. Xia, P. Liu, F.Z. Ren, B.H. Tian, Acta Met Sin. 20 (2006) 89–92.[23] B.Y. Hang, Non-Ferrous Metals Manual, first ed., CIP, Peking, 2009.[24] K. Se, W.X. Li, Heat Treat. Met. 27 (2002) 8–11.[25] A. Hartwig, Mutat. Res. 475 (2001) 113–121.[26] F. Witte, J. Fischer, J. Nellesen, H.A. Crostack, V. Kaese, A. Pisch,

H. Windhagen, Biomaterials 27 (2006) 1013–1018.[27] S.X. Zhang, X.N. Zhang, C.L. Zhao, J.N. Li, Y. Song, C.Y. Xie,

H.R. Tao, Y. Zhang, Y.H. He, Y. Jiang, Y.J. Bian, Acta Biomater. 6(2010) 626–640.

[28] J.C. Gao, D. Hou, C.J. Song, J. Funct. Mater. 43 (2012) 2577–2583.[29] K. Yang, L.L. Tan, Y.B. Ren, B.C. Zhang, G.D. Zhang, H.J. Ai, Mater.

China 28 (2009) 25–30.[30] G.D. Zhang, J.J. Huang, K. Yang, H.J. Ai, Acta Met. Sin. 43 (2007)

1186–1190.