See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/321725032

Magnesium alloy based interference screw developed for ACL reconstruction

attenuates peri-tunnel bone loss in rabbits

Article in Biomaterials · December 2017

DOI: 10.1016/j.biomaterials.2017.12.007

CITATIONS READS

99 893

8 authors, including:

Jiali Wang Yuanhao Wu

Chinese University of Hong Kong Peking Union Medical College Hospital
33 PUBLICATIONS 3,500 CITATIONS 27 PUBLICATIONS 1,644 CITATIONS

SEE PROFILE SEE PROFILE

Huafang Li Yang Liu

University of Science and Technology Beijing Peking University School of Stomatology
66 PUBLICATIONS 4,659 CITATIONS 34 PUBLICATIONS 2,135 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Huafang Li on 19 December 2017.

The user has requested enhancement of the downloaded file.

 Biomaterials 157 (2018) 86e97

Contents lists available at ScienceDirect

Biomaterials
journal homepage: www.elsevier.com/locate/biomaterials

Magnesium alloy based interference screw developed for ACL

reconstruction attenuates peri-tunnel bone loss in rabbits
Jiali Wang a, 1, Yuanhao Wu b, 1, Huafang Li a, 1, Yang Liu c, Xueling Bai d, Wingho Chau a,
Yufeng Zheng b, c, **, Ling Qin a, d, *
a
Musculoskeletal Research Laboratory of Department of Orthopaedics & Traumatology and Innovative Orthopaedic Biomaterial and Drug Translational
Research Laboratory of Li Ka Shing Institute of Health Sciences, The Chinese University of Hong Kong, Hong Kong, China
b
Center for Biomedical Materials and Tissue Engineering, Academy for Advanced Interdisciplinary Studies, Peking University, Beijing 100871, China
c
State Key Laboratory for Turbulence and Complex System and Department of Materials Science and Engineering, College of Engineering, Peking University,
Beijing 100871, PR China
d
Center for Translational Medicine Research and Development, Institute of Biomedical and Health Engineering, Chinese Academy of Sciences, Shenzhen
518055, China

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

Article history: Peri-tunnel bone loss after anterior cruciate ligament (ACL) reconstruction is often observed clinically,
Received 4 August 2017 which may detrimentally affect tendon graft integration with surrounding bone tissue. Biodegradable
Received in revised form magnesium (Mg) based fixators in terms of interference screws may be suitable for fixation of the tendon
5 December 2017
graft due to their favorable effects on promotion of new bone formation. However, the poor mechanical
Accepted 10 December 2017
Available online 11 December 2017
strength of Mg is still one of the major challenges for its clinical applications. The addition of alloying
elements into Mg is one of the strategies to improve their mechanical properties. Here, we prepared
magnesium (Mg)-(4 and 6 wt%) zinc (Zn)-(0.2, 0.5, 1 and 2 wt%) strontium(Sr) alloys and tested their
Keywords:
MgZnSr alloy
potential for attenuating peri-tunnel bone loss in ACL reconstruction. The optimal (6 wt%) Zn and (0.5 wt
ACL reconstruction %) Sr contents were screened with respect to the microstructures, mechanical properties and corrosion
Peri-tunnel bone loss behavior of these alloys. As compared to pure Mg, Mg-6Zn-0.5Sr rods and screws showed significantly
Torsion test higher torque and torsional stiffness in both numerical and experimental analysis. The in vitro cyto-
compatibility of Mg-6Zn-0.5Sr alloy was assessed with MTT test and fluorescence assay. The Mg-6Zn-
0.5Sr interference screw was designed for fixation of the tendon graft to the femoral tunnel in a rab-
bit model of ACL reconstruction, with a commercially available poly-lactide (PLA) screw for comparison.
In vivo high resolution peripheral quantitative computed tomography (HR-pQCT) scanning was per-
formed to measure the degradation behavior of Mg-6Zn-0.5Sr interference screws and peri-tunnel bone
quality at 0, 6, 12 and 16 weeks post-surgically. Mg-6Zn-0.5Sr interference screw was completely
degraded within 12 weeks after surgery. The peri-tunnel bone loss was significantly attenuated in the
Mg-6Zn-0.5Sr group when compared to the PLA group. Importantly, the bony ingrowth rapidly filled the
cavity left by the complete degradation of Mg-6Zn-0.5Sr screws at 16 weeks. In histological analysis,
more bone formation was observed in peri-tunnel region in the Mg-6Zn-0.5Sr group in comparison to
the PLA group at 6 and 16 weeks after surgery. The femur-tendon graft-tibia complex was harvested at
the end of week 6 and 16 post-operation for tensile testing. The maximum load to failure was signifi-
cantly improved in the Mg-6Zn-0.5Sr group at week 16 post-operation. Therefore, our results indicate
the potential clinical application of MgZnSr based interference screws in ACL reconstruction.
© 2017 Published by Elsevier Ltd.

* Corresponding author. Musculoskeletal Research Laboratory of Department of

Orthopaedics & Traumatology and Innovative Orthopaedic Biomaterial and Drug 1. Introduction
Translational Research Laboratory of Li Ka Shing Institute of Health Sciences, The
Chinese University of Hong Kong, Hong Kong, China. Anterior cruciate ligament (ACL) tear is one of the most common
** Corresponding author. Center for Biomedical Materials and Tissue Engineering, types of knee injuries and may subsequently lead to a higher risk of
Academy for Advanced Interdisciplinary Studies, Peking University, Beijing 100871,
China.
developing knee osteoarthritis (OA) in patients due to the loss of
E-mail addresses: yfzheng@pku.edu.cn (Y. Zheng), qin@ort.cuhk.edu.hk (L. Qin). knee stability if no clinical treatment is performed [1,2]. In cases of
1
These authors contributed equally to this work. ACL rupture, the patients may have to require ACL reconstruction

https://doi.org/10.1016/j.biomaterials.2017.12.007
0142-9612/© 2017 Published by Elsevier Ltd.
 J. Wang et al. / Biomaterials 157 (2018) 86e97 87

(ACLR) surgery to restore their knee function and prevent OA and In this study, we prepared a series of Mg-Zn-Sr ternary alloys for
its progression [3,4]. Although most of patients can return to sports screening one of available candidates with superior mechanical
after surgery, there is still about 25% failure rate in clinical out- strength and osteopromotive properties for in vivo testing using
comes [5], which may be ascribed to traumatic, technical, infection, experimental ACLR model, especially with focus on prevention of
and biologic factors [6]. In addition to these causes, peri-tunnel postoperative peri-tunnel bone loss.
bone loss, which is often observed in clinical studies, may result
in less bone around for tendon graft integration and subsequently 2. Materials and methods
impair the tendon graft-bone junction healing [7]. Hence, the
reduction of bone loss around tunnels is one of strategies to pro- 2.1. Alloys preparation
mote the graft healing.
More and more researchers and surgeons have shown increasing The alloys developed for the present study were as-extruded
interests in the Research & Development (R&D) of Mg or its alloys Mg-(4, 6 wt%)Zn-(0.2, 0.5, 1, 2 wt%)Sr ternary. Pure Mg (99.9 wt
based implants due to their excellent biocompatibility, degradation %), Zn (99.9 wt%) and Sr (99.9 wt%) ingots were selected as raw
and osteopromotive properties [8e13]. Till now, a series of novel materials. The Mg-Zn-Sr ternary alloys ingots were prepared by
Mg based alloys, which possess improved corrosion resistance or permanent mold casting at a pouring temperature of 720
C and air
higher mechanical strength, have been developed as potential or- cooling. The as-cast billets of the alloys were solution treated at
thopaedic devices in different animal models to mimic relevant 340
C for about 4 h and then quenched in water. Finally, the so-
clinical indications for both biosafety and bio-efficacy assessment lution treated alloy samples were hot extruded directly into bars at
[14e18]. In addition, extensive work has also been performed on 320
C with an extrusion speed and extrusion ratio of 20 mm/min
how to propose a reliable in vitro model with relevant factors and 12, respectively. Disk alloy samples with a diameter of 12 mm
influencing corrosion performance of Mg metals, for more precise and a height of 2 mm were directly cut from the as-extruded bars.
pre-screening information of in vivo degradation behavior of po- All the disk sample surfaces were then mechanically ground on SiC
tential Mg based devices, which has greatly accelerated the R&D of abrasive papers up to 2000 grit. Subsequently, the ground disks
novel Mg based implants [19,20]. The released Mg ions, which are were washed in acetone, absolute ethyl alcohol, deionized water
accompanied by the degradation of Mg implants, have been re- and finally dried in air. The actual chemical composition of the
ported to enhance osteogenic activity of mesenchymal stem cells ternary alloys, which was determined by using inductively coupled
(MSCs) [21], so the high-purity Mg interference screws have been plasma atomic emission spectrometry (ICP-AES) (Leeman Labs,
recently developed for fixation of the tendon graft in animals un- USA), is provided in Table 1.
derwent ACLR with encouraging results [22e25]. Indeed, high-
purity Mg has close elastic modulus to the cortical bone [26], 2.2. Microstructure characterization
which can mitigate the stress shielding effects to facilitate new
bone formation. However, the ultimate mechanical strength of the Prior to the microstructure characterization, the disk samples
high-purity Mg fixators is still not sufficient to bear the load at were etched with 4 vol% nitric acid distilled water solution. The
weight-bearing skeletal sites and even during the surgery, for microstructure of the Mg-Zn-Sr alloys was investigated by utilizing
example during screw tightening (Supplementary Fig. 1), which is an optical microscope (Olympus BX51M, Japan). The phases of the
one of the most challenging issues that prevents clinical applica- as-extruded alloys were identified by an X-ray diffractometer (XRD)
tions of pure Mg implants [27]. The addition of alloying elements (Rigaku DMAX 2400, Japan) using a Cu Ka radiation with a scanning
which are soluble in Mg has been widely considered as one of angle from 10
to 90
and a scanning rate of 4
/min.
effective strategies to improve the mechanical strength of Mg based
materials [28]. Although, AZ31 (aluminum 3 wt% and zinc 1 wt%) 2.3. Tensile testing
interference screws have been previously reported to restore the
knee stability and graft function of goats without observation of any The tensile property of the as-extruded alloys was measured at
adverse effects [29], the addition of aluminum into the Mg based room temperature on a universal testing machine (INSTRON 5969,
implants is still a serious clinical concern on inducing potential USA) with a crosshead speed about 1 mm/min. The specimens were
neuronal injury. Among the candidates of alloying elements with cut into slices with an electrical discharge wire according to the
low toxicity, Zinc (Zn) is an important element with a relatively high ASTM-E8-04 standard from the extruded bars parallel to the extru-
solubility in Mg and also is able to enhance mechanical properties of sion direction [33]. Three parallel samples were used for each alloy.
the alloy because of its solid solution strengthening and aging
strengthening effects [28]. The zinc distribution in the a-Mg phase 2.4. Immersion test for morphological analysis of alloys
is not uniformly distributed in Mg-Zn alloys with Zn concentrations
higher than 6% (the maximal solid solubility), otherwise the pre- The immersion tests were carried out in Hank's solution with
cipitation of Mg-Zn phase will be formed at the dendrite edges, surface area to solution volume ratio of 1 cm2: 20 ml in accordance
which will ultimately impair the corrosion resistance of Mg-Zn al-
loys [28]. Taking both mechanical properties and corrosion resis- Table 1
tance as main concerns, the preparation of ternary alloys may open Nominal and actual composition of Mg-Zn-Sr ternary alloys.
up a new path for the R&D of orthopaedic devices attributed to their Nominal Composition Actual Composition (wt.%)
combined effects of alloying elements on the microstructure of the
Zn Sr Si Ni Cu Fe Mg
alloys. Strontium (Sr) and calcium (Ca), belongs to the same group
of Mg in the periodic table and shares similar chemical and Mg-4Zn-0.2Sr 3.82 0.24 0.053 <0.001 0.0048 0.0032 Bal.
Mg-4Zn-0.5Sr 4.12 0.56 0.055 <0.001 0.0034 0.0022 Bal.
metallurgical properties [30]. Low contents of Ca and Sr were found
Mg-4Zn-1Sr 4.14 1.20 0.061 <0.001 0.0037 0.0033 Bal.
to slow down the degradation rate [28]. From the viewpoint of Mg-4Zn-2Sr 3.96 1.92 0.053 <0.001 0.0038 0.0020 Bal.
biological effects, Ca ions compete with Mg ions for binding sites, Mg-6Zn-0.2Sr 5.88 0.19 0.044 <0.001 0.0032 0.0025 Bal.
leading to lower influx of Mg ions into cells for biological reactions Mg-6Zn-0.5Sr 5.93 0.62 0.049 <0.001 0.0038 0.0038 Bal.
[31]. Consequently, Sr, which is an osteopromotive mineral element Mg-6Zn-1Sr 5.89 1.08 0.053 <0.001 0.0028 0.0024 Bal.
Mg-6Zn-2Sr 6.21 2.15 0.042 <0.001 0.0052 0.0036 Bal.
[32], is more suitable as a third alloying element in Mg-Zn alloys.
88 J. Wang et al. / Biomaterials 157 (2018) 86e97

with ASTM-G31-72 [34]. In order to avoid evaporation of the Hank's using plane elements to define the rigid surface of the screwdriver.
solution, the tubes were all sealed during the immersion period and The FE model consisted of two parts, i.e. the interference screw and
the temperature of the immersion medium was kept at 37
C with the screw driver [35]. Quadratic tetrahedral elements of types were
water bath. The alloy samples were removed after 3 days' immer- used in the model, in which 61982 elements were generated for the
sion, gently rinsed with distilled water and dried at room temper- screw while 60 hexahedron elements were created for the screw
ature. The surface morphology of the alloys after immersed was driver. The screw driver was modelled as rigid and the interaction
characterized by environmental scanning electronic microscopy between the screw and the driver was set as frictionless contact.
(SEM) (FEI Quanta 200 FEG, USA). Fig. 1A shows the finite elements mesh for the screw. The tip of the
screw (highlighted area in Fig. 1B) was clamped while a controlled
2.5. Measurement of in vitro degradation of alloys rotation at 0.02 radian/s (Fig. 1C) was prescribed for the screw
driver. To model the mechanical response of Mg and Mg-6Zn-0.5Sr
The hydrogen evolution volume and degradation rate of MgZnSr screws, an elastoplastic constitutive model was used. In this model,
ternary alloys were detected by volumetric measurements. Three yield strength sy ¼ 117 MPa, Young's modulus E ¼ 45.3 GPa and
samples prepared in cylindrical disk were immersed in Hank's so- Poisson ration ¼ 0.30 was used for Mg-6Zn-0.5Sr screw while
lution with surface area to Hank's solution volume ratio of 1 cm2: sy ¼ 24 MPa, E ¼ 44.8 GPa and n ¼ 0.35 was applied for high purity
40 ml. Hydrogen evolution volume was detected every 4 h in the Mg screw according to our obtained data. The Equivalent plastic
first two days, and then detected every 12 h. strain and Maximum principal stress in the screw were examined.
In addition, electrochemical test was applied using Metrohm
Auto workstation (Metrohm Autolab, Netherland) in a standard 2.7. Torsion test of rods
three-electrode cell in a beaker containing 250 ml Hank's solution.
The alloy disks were used as working electrode with 0.20 cm2 of The torsion test was carried out on Mg-6Zn-0.5Sr rods with a
exposed area. A platinum electrode was worked as counter elec- dimension of 410 mm  125 mm at a constant rate of 0.25 rpm in a
trode and a saturated calomel electrode (SCE) was worked as torsion test machine (INSTRON E1000, USA) according to ISO 7800-
reference electrode. The polarization curves were used to estimate 2012 standards [36]. High purity Mg (99.99 wt%) rods with the
the corrosion current density (icorr) and corrosion potential (Ecorr) same size was prepared as the control. The torque (Nm) versus
by Tafel extrapolation methods. twist angle (deg) was recorded for the calculation of maximal tor-
que (Nm). Six samples were used for each group.
2.6. Numerical stimulation of torsion test in screw
2.8. Cyto-compatibility test
To gain the understanding on the distribution of the strains and
stresses occurring in the screw's complex geometry during the Mg-6Zn-0.5Sr specimens were immersed into Dulbecco's
torsion test, the experimental setup was reproduced virtually and modified eagle medium (DMEM, Invitrogen, USA) supplemented
analyzed by using the finite element analysis (FEA) method with with 10% fetal bovine serum (Gibco, Thermo Scientific, USA) for
ABAQUS software version 6.12 (Hibbit Inc., Rhode Island, USA). For 72 h under cell culture conditions (5% CO2, 95% humidity, 37
C)
achieving this aim, we created a mesh of tetrahedral elements with a fixed mass ratio to medium volume (0.2 g/ml) for preparing
within the 3D geometry of Mg-6Zn-0.5Sr and high-purity screw, extracts according to ISO 10993-12 [37]. The ion concentrations of

Fig. 1. The proposed finite elements model for the screw tightening. (A) Images of finite elements mesh for the screw and driver. (B) The boundary condition set for the analysis. The
screw tip (highlighted area in red) was fixed when a rotation was applied. (C) A controlled rotation rate of 0.02 radian/s in clockwise direction was prescribed for the screw driver.
(For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)
 J. Wang et al. / Biomaterials 157 (2018) 86e97 89

Mg, Zn and Sr in the extracts were measured by ICP-AES (Leeman pass filter to minimize the noise (Sigma ¼ 1.2, support ¼ 2.0,
Labs, USA). The original extract is called 1  . The extract was then threshold ¼ 145). Therefore, bone volume (BV), trabecular number
diluted with fresh cell culture medium for 3 and 6 times (called 3 (Tb.N), trabecular thickness (Tb.Th) and trabecular separation
and 6 , respectively). Bone marrow stem cells (BMSCs) isolated (Th.Sp) of peri-tunnel tissue were calculated.
from rats at passage 3 were cultured in these extracts for 24 h and
48 h, respectively, according to our previous protocol [38]. Cell 2.11. Histological and histomorphological analysis
viability was evaluated using both 3-(4,5-Dimethylthiazol-2-yl)-
2,5-Diphenyltetrazolium Bromide (MTT) and live/dead fluores- These animals were sacrificed at 6 and 16 weeks after surgery.
cence assays following manufacturer's protocol. To label newly formed bone, the fluorochrome calcein green
(10 mg/kg) was intramuscularly injected into rabbits assigned for
2.9. Animal surgery the two time points at 3 and 13 weeks post-surgically prior to
xylenol orange (90 mg/kg) administration at 5 and 15 weeks post-
The Mg-6Zn-0.5Sr rods with 5 mm in diameter were used for surgically, respectively. Both decalcified and undecalcified samples
the fabrication of specially designed hollow interference screw that at 6 and 16 weeks were used for histological examination. Briefly,
was similar to commercially available screw for human applications all these harvested femoral samples were fixed in 4% neutral buffer
[39]. Prior to animal surgery, the Mg-6Zn-0.5Sr screws were man- formalin for 48 h and then embedded in paraffin or methyl-
ufactured on a computerized numerically controlled lathe (SZ-20C, methacrylate (MMA) according to our previously published pro-
Sowin, Shenzhen, China). All screws were ultrasonically cleaned in tocols prior to tissue sectioning at mid-tunnel region for the
absolute acetone and ethanol successively, followed by washing in following analysis [40]. Stevenel blue-Van Gieson-Alizarin Red S
distilled water prior to sterilization with 25 kGy of 60Co radiation. (SVA) staining was performed in MMA sections at a thickness of
The sterilized screws were then packaged in vaccum. The outer approximate 100 mm [41]. Paraffin sections with 5 mm in thickness
diameter and length of these screws were fixed as 3 and 8 mm, were stained with hematoxylin and eosin (H&E). Six samples were
respectively, which was suitable for fixation of the tendon graft in used in each group at each postoperative time point.
rabbits. Details of Mg-6Zn-0.5Sr screws were listed in
Supplementary Fig. 2. PLA screws with similar dimension param- 2.12. Tensile testing of femur-tendon graft-tibia complexes
eters were purchased (Inion, Tampere, Finland). As shown in
Supplementary Fig. 3, ACLR surgeries in skeletally matured male The femur-tendon graft-tibia complexes (FTGTC) at 6 and 16
New Zealand White rabbits, which was approved by the Animal weeks post-surgically were harvested and then stored in 80
C
Experimentation Ethics Committee (AEEC) of the Chinese Univer- prior to biomechanical testing after thawing at room temperature.
sity of Hong Kong (AEEC No. 13-014/MIS-5), were performed ac- Except for the ACL graft, the suture on the tibial side and all the soft
cording to our previously published protocol [40]. Briefly, under tissue were carefully removed while interference screws were kept
anesthesia, the long digital extensor tendon graft was harvested. in femoral tunnels prior to the fixation of femur and tibia with
Then, a medial parapatella arthrotomy was performed to expose custom-designed jigs in a uniaxial mechanical testing machine
the knee joint prior to patellar dislocation. The infrapatellar fat (H25K-S, Hounsfield Test Equipment LTD, UK). The tensile tests
tissue was removed to expose the joint cavity for ACL transection. were performed in FTGTC with the knee flexed to 90
with a pre-
Afterwards, bone tunnels of 2.5 mm diameter were created in the load of 1 N and a rate of 50 mm/min to record the maximal load to
femur and tibia through the footprint of the original ACL by using a failure. Six samples were used for each group at each time point.
bone drill via the transtibial technique to allow the placement of
the harvested tendon graft. As the synovial fluid in the cavity of 2.13. Statistical analysis
knee joint may result in fast degradation of Mg interference screw if
the inside-out insertion direction is applied, which may ultimately We used Student's t-test for comparisons between two groups
cause the rapid loss of mechanical integrity in Mg screws, the and one-way ANOVA with Tukey's post hoc test for multiple com-
outside-in screw insertion direction was applied in this study. Due parisons between two groups at various postoperative time points.
to insufficient cancellous bone in the proximal tibia, the extra- All data were expressed as mean ± standard deviation (SD) with the
articular graft fixation at the exit of the tibia was performed by significant level set as p < .05 using SPSS 17.0 software (SPSS Inc.
using non-absorbable suture. Finally, the wound was sutured layer Chicago, IL).
by layer. Both Mg-6Zn-0.5Sr and PLA interference screws were used
in this study. 3. Results

2.10. High resolution peripheral quantitative computed tomography 3.1. Alloy microstructure and mechanical properties
(HR-pQCT) analysis
Fig. 2A shows the X-ray diffraction results of the as-extruded
HR-pQCT (Scanco Medical AG, Switzerland) was used to monitor Mg-Zn-Sr ternary alloys. Fig. 2A-a: Mg7Zn3 phases were only
and quantify the degradation rates of Mg-6Zn-0.5Sr screws in peri- detected in Mg-4Zn-0.5Sr alloy. The diffraction intensity of the
tunnel mass after surgery. 6 rabbits in each group (Mg-6Zn-0.5Sr MgZn second phases in Mg-4Zn-1Sr alloy were higher than that of
and PLA) were used for in vivo CT scanning at defined postoperative Mg-4Zn-2Sr alloy, indicated that higher amounts MgZn phases in
time points during the entire experimental period. The rabbits the Mg-4Zn-2Sr alloy. The intensity of Mg17Sr2 second phase was
were anesthetized for taking X-rays at week 0, 6, 12 and 16 after much higher than Mg-4Zn-2Sr alloy. Fig. 2A-b: Mg17Sr2 second
operation with a X ray-tube set at 60 kV, 1 mA and 82 mm in phases were only detected in Mg-6Zn-0.2Sr alloy. Both the intensity
isotropic voxel size. The volume of the Mg-6Zn-0.5Sr screws were of MgZn and Mg7Zn3 second phases were increased with Sr
measured via 3D reconstruction of a series of 2D tomography im- contents.
ages. In addition, the peri-tunnel bone tissue with an area of The microstructure of the Mg-Zn-Sr ternary alloys were shown
0.0585 cm2 was selected as region of interest (ROI) for 3D recon- in Fig. 2B. In the Mg-4Zn-(0.2, 0.5) Sr alloys, the second phase
struction based on a series of 2D tomography images. The resulting particles were mainly distributed in the grains. However, for the
gray-images were segmented using a fixed threshold and a low- Mg-4Zn-(1, 2) Sr alloys, the second phases were mainly distributed
90 J. Wang et al. / Biomaterials 157 (2018) 86e97

Fig. 2. Microstructure and mechanical properties of the as-extruded Mg-Zn-Sr ternary alloys. (A) X-ray diffraction results of Mg-4Zn-(0.2, 0.5, 1, 2)Sr alloys (a) and Mg-6Zn-(0.2, 0.5,
1, 2)Sr alloys (b). (B) Optical micrographs showing the microstructures of Mg-4Zn-0.2Sr (a1), Mg-4Zn-0.5Sr (b1), Mg-4Zn-1Sr (c1), Mg-4Zn-6Zn (d1), Mg-6Zn-0.2Sr (a2), Mg-6Zn-
0.5Sr (b2), Mg-6Zn-1Sr (c2) and Mg-6Zn-2Sr (d2). (C) Yield strength, ultimate tensile strength and elongation of Mg-4Zn-(0.2, 0.5, 1, 2)Sr alloys (a) and Mg-6Zn-(0.2, 0.5, 1, 2) alloys
(b) in tensile tests. Mg-6Zn-0.5Sr, which was indicated by the triangle, showed best mechanical properties among the ternary alloys.

in the grain boundaries. Courser grain boundaries can be seen in fluctuation with different Sr contents and Mg-4Zn-0.2Sr alloy
Mg-4Zn-2Sr alloy. For the alloys containing 6% Zn, the second possess the highest UTS. The elongation rate of Mg-4Zn-xSr
phases were mainly distributed along the grain boundaries. Mg- decreased when the addition of Sr was over 0.5%. In terms of Mg-
6Zn-0.5Sr alloy exhibited much lower second phase contents 6Zn-ySr (y ¼ 0.2, 0.5, 1 and 2) alloys shown in Fig. 2C-b, the YS
than other 3 alloys. decreased with higher Sr content when y was lower than 2. How-
Fig. 2C showed the mechanical properties of the as-extruded ever, Mg-6Zn-0.5Sr exhibited highest UTS and elongation rate.
Mg-Zn-Sr ternary alloys. As we can see from Fig. 2C-a, the yield
strength (YS) of Mg-4Zn-xSr (x ¼ 0.2, 0.5, 1 and 2) alloys was 3.2. Corrosion behavior of alloys
proportional to the Sr content with x lower than 2. Interestingly,
when Sr content further increased up to 2%, the YS of Mg-4Zn-xSr The surface morphology of the as-extruded Mg-Zn-Sr alloys was
alloys decreased. The ultimate tensile strength (UTS) showed little depicted after immersed in Hank's solution for 3 days (Fig. 3A).
 J. Wang et al. / Biomaterials 157 (2018) 86e97 91

Fig. 3. In vitro corrosion behavior of the alloys. (A) Corrosion morphology of the as-extruded Mg-Zn-Sr ternary alloys after immersed in Hank's solution for 3 days: (a1) Mg-4Zn-
0.2Sr; (b1) Mg-4Zn-0.5Sr; (c1) Mg-4Zn-1Sr; (d1) Mg-4Zn-6Sr; (a2) Mg-6Zn-0.2Sr; (b2) Mg-6Zn-0.5Sr; (c2) Mg-6Zn-1Sr; (d2) Mg-6Zn-2Sr. (B) Hydrogen evolution measurement. (a)
and (b): hydrogen evolution volume; (c) and (d): hydrogen evolution rate. (C) Potentiodynamic polarization curves for the as-extruded Mg-Zn-Sr ternary measured in Hank's
solution: (a) Mg-4Zn-(0.2, 0.5, 1, 2) Sr alloys; (b) Mg-6Zn-(0.2, 0.5, 1, 2) Sr alloys.
92 J. Wang et al. / Biomaterials 157 (2018) 86e97

Many corrosion holes or cracks were observed in Mg-4Zn-2Sr and higher maximum torque angle than Mg screw. More importantly,
Mg-6Zn-2Sr alloys, indicating poor corrosion resistance for Mg-Zn- the maximum torque in Mg-6Zn-0.5Sr screw is about 5-fold greater
Sr ternary alloys with high Sr content. Hydrogen evolution and than that in Mg screw (Fig. 4C).
electrochemical measurements were performed for quantitative As shown in Fig. 5, Mg-6Zn-0.5Sr rods presented much higher
assessment of corrosion resistance of Mg-Zn-Sr alloys. As shown in maximum torque than high-purity Mg rods with the same
Fig. 3B, Mg-4Zn-1Sr and Mg-6Zn-0.5Sr showed lowest hydrogen dimension parameters at the same testing conditions (38.54 ± 0.35
evolution rate during the entire immersion period in Mg-4Zn-xSr vs. 25.25 ± 0.06 Nm, p < .01). Although the experimental data of
and Mg-6Zn-ySr ternary alloys (x, y ¼ 0.2, 0.5, 1 and 2), respec- both Mg-6Zn-0.5Sr and high-purity Mg interference screws was
tively. The electrochemical results from the potentiodynamic po- not tested, the experimental torque of Mg-6Zn-0.5Sr and high-
larization curves were supportive to above findings (Fig. 3C and purity Mg rods was principally consistent with the numerical
Table 2). findings, demonstrating a better material candidate for Mg-6Zn-
0.5Sr metal as potential Mg based fixators in orthopaedic field.
3.3. Torsion test
3.4. Cyto-compatibility of Mg-6Zn-0.5Sr
Fig. 4A shows the distribution of equivalent accumulated plastic
strains. The maximal values were observed near the head of both
The Mg, Zn and Sr concentrations of the extract of the Mg-6Zn-
Mg and Mg-6Zn-0.5Sr screws. The distribution of the maximal
0.5Sr rods were shown in Fig. 6A. Cells with severe cytotoxicity
principal stress was shown in Fig. 4B. Similarly, the highest stress
concentrations were found near the screw head. Besides, it was
found that much higher peak stress was obtained in Mg-6Zn-0.5Sr
screw when compared to Mg screw. Mg-6Zn-0.5Sr screw showed

Table 2
Electrochemical corrosion data of as-extruded alloys in Hank's solution.

Sample Corrosion Current Density Corrosion Rate

(mA/cm2) (mm/year)

Mgr-4Zn-0.2S 9.617 ± 1.909 0.220 ± 0.044

Mg-4Zn-0.5Sr 5.858 ± 0.569 0.127 ± 0.013
Mg-4Zn-1Sr 2.352 ± 1.013 0.054 ± 0.023
Mg-4Zn-2Sr 7.036 ± 1.231 0.161 ± 0.028
Mg-6Zn-0.2Sr 14.178 ± 0.852 0.325 ± 0.018
Mg-6Zn-0.5Sr 2.980 ± 0.308 0.068 ± 0.007
Mg-6Zn-1Sr 5.268 ± 0.192 0.119 ± 0.007
Fig. 5. Maximum torque of the high-purity Mg and Mg-6Zn-0.5Sr rods (10 mm in
Mg-6Zn-2Sr 14.461 ± 1.559 0.331 ± 0.036
diameter) (n ¼ 6). Student's t-test, **p < .01.

Fig. 4. Numerical stimulation of torsion test in Mg and Mg-6Zn-0.5Sr screws. (A) Equivalent accumulated plastic strains of Mg and Mg-6Zn-0.5Sr screws. (B) Maximum principal
stress of Mg and Mg-6Zn-0.5Sr screws. (C) Curves on numerical torque versus angle for Mg and Mg-6Zn-0.5Sr screws.
 J. Wang et al. / Biomaterials 157 (2018) 86e97 93

Fig. 6. Cyto-compatibility evaluation of Mg-6Zn-0.5Sr alloy. (A) Mg, Zn and Sr ion levels in the extract. (b) BMSCs viability test in the original extract and a series of diluted extracts.
(C) representative fluorescent images for live (green)/dead (red) cell assay (100X). (For interpretation of the references to colour in this figure legend, the reader is referred to the
Web version of this article.)

were detected in the original extract, whereas the extract diluted by the peri-screw tissue was observed in the Mg-6Zn-0.5Sr group. At
6 times led to non-cytotoxic results (Fig. 6B). Similarly, a large week 16 post-operation, mature bone eventually replaced the
number of apoptotic cells and only a few live cells were observed in fibrous tissue in the Mg-6Zn-0.5Sr group. The degradation of Mg-
the original extract. However, the extract can remarkably improve 6Zn-0.5Sr screw was accompanied with the bony ingrowth, so
the cell survival at a dilution of 6X (Fig. 6C). the cavity left by the complete degradation of screws at week 16
post-operation was fully filled by the newly formed bone.
3.5. HR-pQCT analysis of peri-tunnel bone mass and Mg-6Zn-0.5Sr
screw degradation 3.7. Tensile testing of femur-tendon graft-tibia complexes

The degradation rate of the Mg-6Zn-0.5Sr interference screws There was no significant difference in the maximal load to
and the changes of the peri-tunnel bone mass in both Mg-6Zn- failure between Mg-6Zn-0.5Sr group and PLA group at week 6 post-
0.5Sr and PLA groups were shown in Fig. 7. The Mg-6Zn-0.5Sr operation. However, the load to failure in Mg-6Zn-0.5Sr group was
screws showed a complete degradation within 16 weeks after significantly higher than that in PLA group at week 16 post-
surgery (Fig. 7AeC). The gas formation, which was accompanied by operation (Fig. 9).
the degradation of the screws, was observed in the peri-implant
tissue (Fig. 7B). The cavity left by the degradation of screws was 4. Discussion
then filled by the newly formed bone tissue at week 16 post-
operation. More importantly, the degradation of Mg-6Zn-0.5Sr In the present study, we developed a novel Mg alloy based
screws did not deteriorate peri-tunnel bone mass via measure- interference screw with suitable dimension parameters for fixation
ment of BV, Tb.N, Tb.Th and Tb.Sp after surgical implantation, while of the tendon graft in ACL reconstruction without concerns about
the implantation of PLA screws significantly decreased BV, Tb.N and mechanical failure and peri-tunnel bone loss after surgery
Tb.Th and increased Tb.Sp in peri-tunnel bone tissue within 16 (Supplementary Fig. 4). The addition of Zn and Sr elements into Mg
weeks after surgery (Fig. 7D). alloys can optimize their mechanical and corrosion properties [28].
In addition, Zn and Sr are known as osteopromotive elements for
3.6. Histological analysis fabrication into potential medical devices or drugs [32,42], so we
explored a series of Mg-xZn-ySr alloys (x ¼ 4, 6 while y ¼ 0.2, 0.5, 1,
As shown in Fig. 8, more bone was observed in the peri-tunnel 2) in this study. Mg-6Zn-0.5Sr alloy showed best corrosion resis-
region in the Mg-6Zn-0.5Sr group when compared to the PLA tance and mechanical properties among these ternary alloys.
group at both 6 and 16 weeks post-operation according to histo- Compared to previously used high purity Mg implant or screw [23],
logical sections stained with H&E and SVA, which were also sup- Mg-6Zn-0.5Sr sample had much higher maximum torque in both
ported by those with fluorescent labelling. The degradation experimental and numerical torsion tests. Cytotoxicity test sug-
products of the Mg-6Zn-0.5Sr screws induced the formation of the gested that Mg-6Zn-0.5Sr alloy could be considered as a safe
fibrous tissue around the screws at week 6 post-operation, which biomaterial for developing potential Class III medical devices ac-
was not observable in the PLA group. Meanwhile, the gas voids in cording to our previously proposed guidelines [38]. Importantly,
94 J. Wang et al. / Biomaterials 157 (2018) 86e97

Fig. 7. HR-pQCT analysis of Mg-6Zn-0.5Sr interference screws and peri-tunnel bone mass. (A) The representative 3D reconstructed models of the femora in rabbits with ACL
reconstruction. The screws and the targeted region of the peri-tunnel bone were indicated by the black and the pink arrows, respectively. (B) The representative 2D tomography
images showing the gas (indicated by the white arrows) around the screws. (C) The volume changes of Mg-6Zn-0.5Sr screws in rabbits within 16 weeks after surgery (n ¼ 4). (D) The
peri-tunnel bone mass (BV, Tb.N, Tb.Th and Tb.Sp) in Mg-6Zn-0.5Sr and PLA groups within 16 weeks after operation (n ¼ 6). One-way ANOVA with Tukey's post hoc test, *p < .05,
Scale bar: 1 mm. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

the peri-tunnel bone loss was significantly attenuated in the Mg- deposition of the hard brittle second phases around the grain
6Zn-0.5Sr group when compared to the PLA group, resulting in a boundaries can easily lead to intercrystalline brittle fracture in al-
significantly higher load to failure in the Mg-6Zn-0.5Sr group at loys in load-bearing conditions [28], the control of the second
week 16 post-operation. phases is critical for the development of Mg alloy based ortho-
paedic devices. Taking both tensile strength and elongation rate as
4.1. The optimal chemical composition of Mg-xZn-ySr ternary alloys the concerns, Mg-6Zn-0.5Sr alloys exhibited superior mechanical
properties over other Mg-Zn-Sr ternary alloys.
The incremental addition of Zn and Sr in the Mg-Zn-Sr ternary The amounts of the deposited second phases also played a vital
alloys lead to distinctive second phase composition and distribu- role in the corrosion behavior the Mg-Zn-Sr ternary alloys. It had
tion. The Zn-containing second phases were Mg7Zn3 and MgZn been widely regarded that the second phase exhibited higher
phases while the Sr-containing second phase was Mg17Sr2, which corrosion potential than a-Mg matrix, so small amounts of second
are in agreement with the previous study for as-cast Mg-5Zn-Sr phases may improve the corrosion resistance of the alloys. How-
alloys [43]. The mechanical strength can benefit from the deposited ever, large volume fraction of the second phases would lead to the
second phases in the alloys by dispersion strengthening mecha- increased formation of galvanic corrosion couples between a-Mg
nism [44,45]. For Mg-4Zn-ySr (y ¼ 0.2, 0.5, 1 and 2) alloys, the matrix and the second phases and subsequently resulted in accel-
second phase particles with round shape were mainly distributed erated corrosion rate of the alloys. This may be the reason why Mg-
in the grains for the Mg-4Zn-0.2Sr and Mg-4Zn-0.5Sr alloys. When 4Zn-(0.5, 1)Sr and Mg-6Zn-(0.5, 1)Sr alloys showed relatively lower
Sr contents increased to 1 and 2%, more second phases were corrosion rate and hydrogen evolution volume. Taken together, Mg-
deposited along the grain boundaries. In terms of Mg-6Zn-ySr 6Zn-0.5Sr alloy has excellent mechanical properties and relatively
(y ¼ 0.2, 0.5, 1 and 2) alloys, the volume fraction of second phase lower corrosion rate, demonstrating a higher potential as medical
in Mg-6Zn-0.5Sr alloy was much lower than other 3 alloys. As the devices over other Mg-Zn-Sr ternary alloys.
 J. Wang et al. / Biomaterials 157 (2018) 86e97 95

Fig. 8. Representative histology of the peri-tunnel tissue in the PLA and Mg-6Zn-0.5Sr groups at 6 and 16 weeks after surgery. F: fibrous tissue; T: tendon graft. Scar bar: 200 mm.
The gas voids in surrounding tissue and the Mg-6Zn-0.5Sr were indicated by the black and the white arrows, respectively. The original bone tunnel was labelled by the white dotted
circle.

forces [47], provided a pre-screening information regarding

maximum torque of Mg and Mg-6Zn-0.5Sr screws as potential fix-
ators in ACL reconstruction. Much higher maximum principal stress
was found distributed in the region of Mg-6Zn-0.5Sr screw head in
comparison to that of high-purity Mg screw head, contributing to 5-
fold increase in maximum torque of Mg-6Zn-0.5Sr screw. Actually,
the findings from the torsion test on rods were consistent with FEA
results. Mg-6Zn-0.5Sr showed significantly higher maximum torque
than high purity Mg rods, indicating a better material candidate for
Mg-6Zn-0.5Sr fabricated into interference screws. Of note that we
successfully inserted the whole body of Mg-6Zn-0.5Sr interference
screw into bone tunnels without causing any breakage in screw
Fig. 9. Tensile property of femur-tendon graft-tibia complexes tested at week 6 and 16 head while the top portion of high purity screw was remained
post-operation after ACL reconstruction surgery in rabbits (n ¼ 6). Student's t-test, outside of bone tunnels due to the failure in screw head
*p < .05.
(Supplementary Fig. 1). As longer interference screws can provide
better fixation ability in tendon graft [48], the use of Mg-6Zn-0.5Sr
may fix tendon graft in bone tunnels more tightly in the initial
4.2. The superior mechanical properties of Mg-6Zn-0.5Sr over high
healing stage, which is crucial for the graft healing prior to stable
purity Mg as potential interference screws in ACL reconstruction
interzone structure formed at the tendon-bone interface [49].
Although the use of high purity Mg interference screws has been
reported to promote the osteointegration of the tendon graft into 4.3. The effects of Mg-6Zn-0.5Sr screws on peri-tunnel bone tissue
bone tunnels and inhibit graft degradation [22e24], it is still a in animals after ACL reconstruction
serious clinical concern for the application of high purity Mg fixators
in clinical trials due to their insufficient maximum torque The peri-tunnel bone loss usually occurs in the early time post
(Supplementary Fig. 1). As the maximum torque of screw is pro- ACL reconstruction operation, which may be ascribed to high
portional to the yield strength [46], the Mg-6Zn-0.5Sr based screw expression of matrix metallopeptidase 1 (MMP1), MMP13 and
have higher maximum torque than high purity Mg screw, which CD68þ cells at the tendon-bone surface and peri-tunnel tissue [50],
then reduce the concern regarding the torque levels beyond the may pose potential micro fracture of surrounding trabecular bone
yield point during the insertion of screw into the bone. It is evident due to reduced bone strength and subsequently affect the graft
that the failure site after tensile test is near the screw head at osteointegration [51]. Although Mg-based implants have been
rotation in both experimental and numerical torsion test, so the increasingly reported to enhance bone formation and decrease
insertion of Mg screw into bone tunnels may cause the breakage in MMP13 expression levels as potential orthopaedic devices [12,22],
the screw head if the screw has low maximum torque. FEA, which the use of high purity Mg interference screw did not effectively
has been widely applied in orthopaedic field for estimating the attenuate peri-tunnel bone mass according to our previous
mechanical responses of medical devices to the applied external studies [23]. We found that Mg ions could not only positively
96 J. Wang et al. / Biomaterials 157 (2018) 86e97

promote osteoblastgenesis, but also increase osteoclastogensis group, Mg-6Zn-0.5Sr interference screw significantly increased
(Supplementary Fig. 5), which was consistent with Wu's findings peri-tunnel bone mass in rabbits after ACL reconstruction within 16
on enhanced local bone turnover [52]. As Sr can enhance new bone weeks both radiographically and histologically, which ultimately
formation by activating Wnt/Catenin signaling while decrease bone contributed to a significant increase in the maximal load to failure
resorption [32,53], the addition of Sr into Mg alloys for fabrication for FTGTC in the Mg-6Zn-0.5Sr group with comparison to the PLA
into medical devices may be a potential strategy to increase bone group. In conclusion, the use of Mg-6Zn-0.5Sr interference screws
mass. In addition, our previous studies showed that Zn based im- could provide sufficient mechanical strength to fix the tendon graft
plants have shown beneficial effects on new bone formation [54]. It during the entire graft healing period and effectively attenuate the
has been reported that the addition of Zn into Mg matrix may peri-tunnel bone loss as potential fixators in ACL reconstruction.
behave synergic effects on osteogenic differentiation of MSCs
attributed to the anabolic role of Zn ions in ZIP1 (a ubiquitous zinc Acknowledgements
transporter for zinc influx) and inhibitive role of Mg ions in ZnT1
(zinc exporters) [55]. Therefore, the degradation of Mg-Zn-Sr alloy This work was supported by the NSFC/RGC Joint Research
based interference screw may potentially reduce peri-tunnel bone Scheme with the Grant No. 51361165101 in Mainland China and
loss. Compared to the PLA group, Mg-6Zn-0.5Sr screw effectively N_CUHK449/13 in Hong Kong and partially supported by a grant
attenuated peri-tunnel bone loss as demonstrated by HR-pQCT from the Research Grants Council of the Hong Kong Special
based bone histomorphometry (BV, Tb.N, Tb.Th and Tb.Sp values). Administrative Region, China (Ref No. T13-402/17-N).
In histology, more trabecular bone around bone tunnels was
observed in the Mg-6Zn-0.5Sr group with comparison to the PLA Appendix A. Supplementary data
group. The released ions from Mg-6Zn-0.5Sr screw resulted in
larger fluorescent positive area in surrounding bone tissue around Supplementary data related to this article can be found at
tunnels, indicating promoted new bone formation. More trabecular https://doi.org/10.1016/j.biomaterials.2017.12.007.
bone around bone tunnels can minimize bone fracture in tunnel
surface and maintain the fixation strength of the tendon graft into
References
tunnels for better healing. Therefore, we observed a significant
increase in the maximal load to failure of FTGTC in the Mg-6Zn- [1] N.A. Mall, P.N. Chalmers, M. Moric, M.J. Tanaka, B.J. Cole, B.R. Bach Jr.,
0.5Sr group when compared to the PLA group at week 16 post- G.A. Paletta Jr., Incidence and trends of anterior cruciate ligament recon-
struction in the United States, Am. J. Sports Med. 42 (10) (2014) 2363e2370.
operation, indicating the beneficial effects of Mg-6Zn-0.5Sr
[2] A. Ajuied, F. Wong, C. Smith, M. Norris, P. Earnshaw, D. Back, A. Davies,
metals on the graft healing. Although the released metal ions Anterior cruciate ligament injury and radiologic progression of knee osteo-
from the Mg-6Zn-0.5Sr screws can contribute to the graft healing, arthritis: a systematic review and meta-analysis, Am. J. Sports Med. 42 (9)
the production of the gas accompanied with the degradation of the (2014) 2242e2252.
[3] C.L. Ardern, N.F. Taylor, J.A. Feller, K.E. Webster, Fifty-five per cent return to
Mg-6Zn-0.5Sr screws may deteriorate the surrounding tissue via competitive sport following anterior cruciate ligament reconstruction sur-
the formation and accumulation of gas voids. Interestingly, gery: an updated systematic review and meta-analysis including aspects of
although the Mg-6Zn-0.5Sr screws showed a relatively fast degra- physical functioning and contextual factors, Br. J. Sports Med. 48 (21) (2014)
1543e1552.
dation rate within the graft healing period, we did not observe the [4] Y.F. Chen, S. Lin, Y.X. Sun, X.H. Pan, L.B. Xiao, L.Y. Zou, K.W. Ho, G. Li, Trans-
accumulation of a large amount of gas in the peri-implant tissue lational potential of ginsenoside Rb1 in managing progression of osteoar-
both radiographically and histologically, which might be ascribed thritis, J. Orthop. Translat. 6 (2016) 27e33.
[5] G.V. Kamath, J.C. Redfern, P.E. Greis, R.T. Burks, Revision anterior cruciate
to the buffering role of the knee joint space for the diffusion and ligament reconstruction, Am. J. Sports Med. 39 (1) (2011) 199e217.
excretion of the gas in the local tissue [56]. [6] G. Samitier, A.I. Marcano, E. Alentorn-Geli, R. Cugat, K.W. Farmer, M.W. Moser,
There are some limitations in this study. Although rabbit is the Failure of anterior cruciate ligament reconstruction, Arch. Bone Joint Surg. 3
(4) (2015) 220e240.
most frequently used animal model to mimic patients underwent
[7] P.P.Y. Lui, Y.W. Lee, T.Y. Mok, Y.C. Cheuk, K.M. Chan, Alendronate reduced peri-
ACL reconstruction (Supplementary Fig. 6), we still need to perform tunnel bone loss and enhanced tendon graft to bone tunnel healing in anterior
more studies in larger animals such as sheep and goats for bio- cruciate ligament reconstruction, Eur. Cell. Mater. 25 (2013) 78e96.
[8] D. Zhao, S. Huang, F. Lu, B. Wang, L. Yang, L. Qin, K. Yang, Y. Li, W. Li, W. Wang,
efficacy assessment. The use of the larger animals indicates
S. Tian, X. Zhang, W. Gao, Z. Wang, Y. Zhang, X. Xie, J. Wang, J. Li, Vascularized
higher demands on animal holding and testing with significant bone grafting fixed by biodegradable magnesium screw for treating osteo-
high costs, but they move via walking rather than hopping, which necrosis of the femoral head, Biomaterials 81 (2016) 84e92.
may provide more precise pre-screening information for the clin- [9] J.W. Lee, H.S. Han, K.J. Han, J. Park, H. Jeon, M.R. Ok, H.K. Seok, J.P. Ahn, K.E. Lee,
D.H. Lee, S.J. Yang, S.Y. Cho, P.R. Cha, H. Kwon, T.H. Nam, J.H. Han, H.J. Rho,
ical trials. As the goat has the most similar anatomy to the human K.S. Lee, Y.C. Kim, D. Mantovani, Long-term clinical study and multiscale
knee, we will select goat as the animal model to conduct ACL analysis of in vivo biodegradation mechanism of Mg alloy, Proc. Natl. Acad. Sci.
reconstruction for the next stage prior to clinical trial application USA 113 (3) (2016) 716e721.
[10] H. Windhagen, K. Radtke, A. Weizbauer, J. Diekmann, Y. Noll, U. Kreimeyer,
[57]. In addition, in terms of the R&D of available Mg based inter- R. Schavan, C. Stukenborg-Colsman, H. Waizy, Biodegradable magnesium-
ference screws for clinical use, it is still necessary to improve the based screw clinically equivalent to titanium screw in hallux valgus sur-
corrosion resistance of currently used Mg-6Zn-0.5Sr. Micro-arc gery: short term results of the first prospective, randomized, controlled
clinical pilot study, Biomed. Eng. Online 12 (2013) 62.
oxidation (MAO) treatment will be adopted to control the degra- [11] D.W. Zhao, F. Witte, F.Q. Lu, J.L. Wang, J.L. Li, L. Qin, Current status on clinical
dation rate of Mg-6Zn-0.5Sr screws in our future studies [58]. applications of magnesium-based orthopaedic implants: a review from clin-
ical translational perspective, Biomaterials 112 (2017) 287e302.
[12] Y. Zhang, J. Xu, Y.C. Ruan, M.K. Yu, M. O'Laughlin, H. Wise, D. Chen, L. Tian,
5. Conclusion D. Shi, J. Wang, S. Chen, J.Q. Feng, D.H. Chow, X. Xie, L. Zheng, L. Huang,
S. Huang, K. Leung, N. Lu, L. Zhao, H. Li, D. Zhao, X. Guo, K. Chan, F. Witte,
This study developed and tested a series of Mg-(4 and 6 wt%) H.C. Chan, Y. Zheng, L. Qin, Implant-derived magnesium induces local
neuronal production of CGRP to improve bone-fracture healing in rats, Nat.
Zn-(0.2, 0.5, 1 and 2 wt%) Sr alloys as potential biodegradable
Med. 22 (10) (2016) 1160e1169.
fixators for ACL reconstruction. Mg-6Zn-0.5Sr showed superior [13] D. Noviana, D. Paramitha, M.F. Ulum, H. Hermawan, The effect of hydrogen gas
mechanical properties and corrosion resistance over other formu- evolution of magnesium implant on the postimplantation mortality of rats,
lations or alloys. Mg-6Zn-0.5Sr has significant superior maximum J. Orthop. Translat. 5 (2016) 9e15.
[14] F. Witte, J. Fischer, J. Nellesen, H.A. Crostack, V. Kaese, A. Pisch, F. Beckmann,
torque than that of high purity Mg in both experimental and nu- H. Windhagen, In vitro and in vivo corrosion measurements of magnesium
merical torsion tests. Compared to traditional degradable PLA alloys, Biomaterials 27 (7) (2006) 1013e1018.
 J. Wang et al. / Biomaterials 157 (2018) 86e97 97

[15] F. Witte, V. Kaese, H. Haferkamp, E. Switzer, A. Meyer-Lindenberg, C.J. Wirth, [37] ISO, 10993-12: Biological Evaluation of Medical Devices e Part 12: Sample
H. Windhagen, In vivo corrosion of four magnesium alloys and the associated Preparation and Reference Materials, International Organization of Standards,
bone response, Biomaterials 26 (17) (2005) 3557e3563. 2012.
[16] J. Kuhlmann, I. Bartsch, E. Willbold, S. Schuchardt, O. Holz, N. Hort, D. Hoche, [38] J. Wang, F. Witte, T. Xi, Y. Zheng, K. Yang, Y. Yang, D. Zhao, J. Meng, Y. Li, W. Li,
W.R. Heineman, F. Witte, Fast escape of hydrogen from gas cavities around K. Chan, L. Qin, Recommendation for modifying current cytotoxicity testing
corroding magnesium implants, Acta Biomater. 9 (10) (2013) 8714e8721. standards for biodegradable magnesium-based materials, Acta Biomater. 21
[17] E. Willbold, X. Gu, D. Albert, K. Kalla, K. Bobe, M. Brauneis, C. Janning, (2015) 237e249.
J. Nellesen, W. Czayka, W. Tillmann, Y. Zheng, F. Witte, Effect of the addition of [39] A. Weiler, H.J. Windhagen, M.J. Raschke, A. Laumeyer, R.F.G. Hoffmann,
low rare earth elements (lanthanum, neodymium, cerium) on the biodegra- Biodegradable interference screw fixation exhibits pull-out force and stiffness
dation and biocompatibility of magnesium, Acta Biomater. 11 (2015) similar to titanium screws, Am. J. Sports Med. 26 (1) (1998) 119e128.
554e562. [40] C.Y. Wen, L. Qin, K.M. Lee, K.M. Chan, Peri-graft bone mass and connectivity as
[18] K. Bobe, E. Willbold, I. Morgenthal, O. Andersen, T. Studnitzky, J. Nellesen, predictors for the strength of tendon-to-bone attachment after anterior cru-
W. Tillmann, C. Vogt, K. Vano, F. Witte, In vitro and in vivo evaluation of ciate ligament reconstruction, Bone 45 (3) (2009) 545e552.
biodegradable, open-porous scaffolds made of sintered magnesium W4 short [41] C. Maniatopoulos, A. Rodriguez, D.A. Deporter, A.H. Melcher, An improved
fibres, Acta Biomater. 9 (10) (2013) 8611e8623. method for preparing histological sections of metallic implants, Int. J. Oral
[19] Y. Koo, H.B. Lee, Z. Dong, R. Kotoka, J. Sankar, N. Huang, Y. Yun, The effects of Maxillofac. Implants 1 (1) (1986) 31e37.
static and dynamic loading on biodegradable magnesium pins in vitro and [42] H.F. Li, X.H. Xie, Y.F. Zheng, Y. Cong, F.Y. Zhou, K.J. Qiu, X. Wang, S.H. Chen,
in vivo, Sci. Rep. 7 (1) (2017) 14710. L. Huang, L. Tian, L. Qin, Development of biodegradable Zn-1X binary
[20] J. Wang, Y. Jang, G. Wan, V. Giridharan, G.L. Song, Z. Xu, Y. Koo, P. Qi, J. Sankar, alloys with nutrient alloying elements Mg, Ca and Sr, Sci. Rep. 5 (2015)
N. Huang, Y. Yun, Flow-induced corrosion of absorbable magnesium alloy: in- 10719.
situ and real-time electrochemical study, Corros. Sci. 104 (2016) 277e289. [43] M.X. Cheng, J.H. Chen, H.G. Yan, B. Su, Z.H. Yu, W.J. Xia, X.L. Gong, Effects of
[21] S. Yoshizawa, A. Brown, A. Barchowsky, C. Sfeir, Magnesium ion stimulation minor Sr addition on microstructure, mechanical and bio-corrosion prop-
of bone marrow stromal cells enhances osteogenic activity, simulating the erties of the Mg-5Zn based alloy system, J. Alloy. Comp. 691 (2017)
effect of magnesium alloy degradation, Acta Biomater. 10 (6) (2014) 95e102.
2834e2842. [44] J.F. Nie, Effects of precipitate shape and orientation on dispersion strength-
[22] P.F. Cheng, P. Han, C.L. Zhao, S.X. Zhang, X.N. Zhang, Y.M. Chai, Magnesium ening in magnesium alloys, Scripta Mater. 48 (8) (2003) 1009e1015.
inference screw supports early graft incorporation with inhibition of graft [45] J.D. Robson, C. Paa-Rai, The interaction of grain refinement and ageing in
degradation in anterior cruciate ligament reconstruction, Sci. Rep. UK 6 magnesium-zinc-zirconium (ZK) alloys, Acta Mater. 95 (2015) 10e19.
(2016). [46] J. Schatzker, F. Meutstege, W.D. Prieur, Manual of Internal Fixation in Small
[23] J. Wang, J. Xu, W. Fu, W. Cheng, K. Chan, P.S. Yung, L. Qin, Biodegradable Animals, Springer- Verlag, Berlin, 1998.
magnesium screws accelerate fibrous tissue mineralization at the tendon- [47] C.C. Hsu, C.K. Chao, J.L. Wang, S.M. Hou, Y.T. Tsai, J. Lin, Increase of pullout
bone insertion in anterior cruciate ligament reconstruction model of rabbit, strength of spinal pedicle screws with conical core: biomechanical tests and
Sci. Rep. 7 (2017) 40369. finite element analyses, J. Orthop. Res. 23 (4) (2005) 788e794.
[24] P.F. Cheng, P. Han, C.L. Zhao, S.X. Zhang, H.L. Wu, High-purity magnesium [48] A. Herrera, F. Martinez, D. Iglesias, J. Cegonino, E. Ibarz, L. Gracia, Fixation
interference screws promote fibrocartilaginous entheses regeneration in the strength of biocomposite wedge interference screw in ACL reconstruction:
anterior cruciate ligament reconstruction rabbit model via accumulation of effect of screw length and tunnel/screw ratio. a controlled laboratory study,
BMP-2 and VEGF, Biomaterials 81 (2016) 14e26. BMC Muscoskel. Disord. 11 (2010) 139.
[25] J.L. Wang, J.K. Xu, B. Song, D.H. Chow, P.S.H. Yung, L. Qin, Magnesium (Mg) [49] J. Nyland, R. Krupp, J. Greene, R. Bowles, R. Burden, D.N. Caborn, In situ
based interference screws developed for promoting tendon graft incorpora- comparison of varying composite tibial tunnel interference screws used for
tion in bone tunnel in rabbits, Acta Biomater. 63 (2017) 393e410. ACL soft tissue graft fixation, Knee 22 (6) (2015) 554e558.
[26] J.L. Wang, J. Tang, P. Zhang, Y.D. Li, J. Wang, Y.X. Lai, L. Qin, Surface modifi- [50] P.P. Lui, Y.W. Lee, T.Y. Mok, Y.C. Cheuk, Peri-tunnel bone loss: does it affect
cation of magnesium alloys developed for bioabsorbable orthopedic implants: early tendon graft to bone tunnel healing after ACL reconstruction? Knee Surg.
a general review, J. Biomed. Mater. Res. B Appl. Biomater. 100B (6) (2012) Sports Traumatol. Arthrosc. 23 (3) (2015) 740e751.
1691e1701. [51] P.P. Lui, Y.W. Lee, T.Y. Mok, Y.C. Cheuk, Local administration of alendronate
[27] T.T. Tang, L. Qin, Translational study of orthopaedic biomaterials and devices, reduced peri-tunnel bone loss and promoted graft-bone tunnel healing with
J. Orthop. Translat. 5 (2016) 69e71. minimal systemic effect on bone in contralateral knee, J. Orthop. Res. 31 (12)
[28] Y.F. Zheng, X.N. Gu, F. Witte, Biodegradable metals, Mat. Sci. Eng. R 77 (2014) (2013) 1897e1906.
1e34. [52] L.L. Wu, B.J.C. Luthringer, F. Feyerabend, A.F. Schilling, R. Willumeit, Effects of
[29] K.F. Farraro, K.E. Kim, S.L. Woo, J.R. Flowers, M.B. McCullough, Revolutionizing extracellular magnesium on the differentiation and function of human oste-
orthopaedic biomaterials: the potential of biodegradable and bioresorbable oclasts, Acta Biomater. 10 (6) (2014) 2843e2854.
magnesium-based materials for functional tissue engineering, J. Biomech. 47 [53] D.P. Wornham, M.O. Hajjawi, I.R. Orriss, T.R. Arnett, Strontium potently in-
(9) (2014) 1979e1986. hibits mineralisation in bone-forming primary rat osteoblast cultures and
[30] X.N. Gu, X.H. Xie, N. Li, Y.F. Zheng, L. Qin, In vitro and in vivo studies on a Mg- reduces numbers of osteoclasts in mouse marrow cultures, Osteoporos. Int. 25
Sr binary alloy system developed as a new kind of biodegradable metal, Acta (10) (2014) 2477e2484.
Biomater. 8 (6) (2012) 2360e2374. [54] H.F. Li, X.H. Xie, Y.F. Zheng, Y. Cong, F.Y. Zhou, K.J. Qiu, X. Wang, S.H. Chen,
[31] T. Dudev, C. Lim, Competition among Ca2þ, Mg2þ, and Naþ for model ion L. Huang, L. Tian, L. Qin, Development of biodegradable Zn-1X binary alloys
channel selectivity filters: determinants of ion selectivity, J. Phys. Chem. B 116 with nutrient alloying elements Mg, Ca and Sr, Sci. Rep. UK 5 (2015).
(35) (2012) 10703e10714. [55] Y. Yu, G. Jin, Y. Xue, D. Wang, X. Liu, J. Sun, Multifunctions of dual Zn/Mg ion
[32] F. Yang, D.Z. Yang, J. Tu, Q.X. Zheng, L.T. Cai, L.P. Wang, Strontium enhances co-implanted titanium on osteogenesis, angiogenesis and bacteria inhibition
osteogenic differentiation of mesenchymal stem cells and in vivo bone for- for dental implants, Acta Biomater. 49 (2017) 590e603.
mation by activating Wnt/Catenin signaling, Stem Cell. 29 (6) (2011) [56] D. Zhao, T. Wang, K. Nahan, X. Guo, Z. Zhang, Z. Dong, S. Chen, D.T. Chou,
981e991. D. Hong, P.N. Kumta, W.R. Heineman, In vivo characterization of magnesium
[33] ASTM, Standard Test Methods for Tension Testing of Metallic Materials, 2004. alloy biodegradation using electrochemical H2 monitoring, ICP-MS, and XPS,
[34] ASTM, Standard Practice for Laboratory Immersion Corrosion Testing of Acta Biomater. 50 (2017) 556e565.
Metals, 1972. [57] B.L. Proffen, M. McElfresh, B.C. Fleming, M.M. Murray, A comparative
[35] C.R.M. Roesler, G.V. Salmoria, A.D.O. More, J.M. Vassoler, E.A. Fancello, Torsion anatomical study of the human knee and six animal species, Knee 19 (4)
test method for mechanical characterization of PLDLA 70/30 ACL interference (2012) 493e499.
screws, Polym. Test. 34 (2014) 34e41. [58] J. Tang, J. Wang, X. Xie, P. Zhang, Y. Lai, Y. Li, L. Qin, Surface coating reduces
[36] ISO, Metallic Materials-Wire-Simple Torsion Test, International Organization degradation rate of magnesium alloy developed for orthopaedic applications,
for Standardization, 2012. J. Orthop. Translat. 1 (2013) 41e48.

View publication stats