lable at ScienceDirect

Biomaterials xxx (2013) 1e9

Contents lists avai

Biomaterials

journal homepage: www.elsevier .com/locate/biomater ia ls

Early osseointegration of a strontium containing glass ceramicin a rabbit model

Arumugan Sabareeswaran a, Bikramjit Basu b,*, Sachin J. Shenoy c, Zahira Jaffer d,Naresh Saha b, Artemis Stamboulis d,*

aHistopathology Laboratory, Biomedical Technology Wing, Sree Chitra Tirunal Institute for Medical Sciences and Technology, Trivandrum 695012, Indiab Laboratory for Biomaterials, Materials Research Centre, Indian Institute of Science, Bangalore 560012, IndiacDivision of In Vivo Models and Testing, Biomedical Technology Wing, Sree Chitra Tirunal Institute for Medical Sciences and Technology,Trivandrum 695012, IndiadBiomaterials Group, School of Metallurgy and Materials, University of Birmingham, Edgbaston, Birmingham B15 2TT, UK

a r t i c l e i n f o

Article history:Received 3 August 2013Accepted 21 August 2013Available online xxx

Keywords:OrthopaedicStrontiumBioglassOsteoporosisIn vivo BiocompatibilityMicro-CT

* Corresponding authors.E-mail addresses: bikram@mrc.iisc.ernet.in (B. Bas

(A. Stamboulis).

0142-9612/$ e see front matter � 2013 Elsevier Ltd.http://dx.doi.org/10.1016/j.biomaterials.2013.08.070

Please cite this article in press as: SabareesBiomaterials (2013), http://dx.doi.org/10.101

a b s t r a c t

The most important property of a bone cement or a bone substitute in load bearing orthopaedic implantsis good integration with host bone with reduced bone resorption and increased bone regeneration at theimplant interface. Long term implantation of metal-based joint replacements often results in corrosionand particle release, initiating chronic inflammation leading onto osteoporosis of host bone. An alter-native solution is the coating of metal implants with hydroxyapatite (HA) or bioglass or the use of bulkbioglass or HA-based composites. In the above perspective, the present study reports the in vivobiocompatibility and bone healing of the strontium (Sr)-stabilized bulk glass ceramics with the nominalcomposition of 4.5SiO2e3Al2O3e1.5P2O5e3SrOe2SrF2 during short term implantation of up to 12 weeksin rabbit animal model. The progression of healing and bone regeneration was qualitatively and quan-titatively assessed using fluorescence microscopy, histological analysis and micro-computed tomography.The overall assessment of the present study establishes that the investigated glass ceramic is biocom-patible in vivo with regards to local effects after short term implantation in rabbit animal model.Excellent healing was observed, which is comparable to that seen in response to a commercially availableimplant of HA-based bioglass alone.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

In last few decades, various materials are being researched foruse as bone implants [1]. Metals are mostly used as load bearingprosthesis. However, the need for good integration of implant withadjacent host bone through bone remodelling at the interfacenecessitated the use of coating of Ti or stainless steel or CoCrMoalloy with calcium phosphate or glass ceramic materials. However,aseptic loosening of joint prosthesis over long term implantationhas resulted in the release of ultrafine particles, which initiates aseries of events beginning with the influx/activation of a largenumber of inflammatory cells, release of cytokines and stimulationof host bone remodelling cells, the osteoclasts, which togetherleads to bone resorption. Osteoporosis, a commonly reported

u), a.stamboulis@bham.ac.uk

All rights reserved.

waran A, et al., Early osseoin6/j.biomaterials.2013.08.070

disease among post-menopausal women and the ageing popula-tion, is characterized by an imbalance in the functionality of oste-oblasts (bone forming) and osteoclast (bone resorption) cells,leading to the disruption of bone remodelling [5]. This conse-quently makes the bone porous and prone to fracture. The pre-vention of such bone resorption is a primary requirement formaintenance of the implant in the elderly patients, who otherwisehave to undergo revision surgery.

Motivated from the recent use of strontium ranelate (SrR) underthe trade name, Protelos for the treatment of osteoporosis, anumber of research groups have investigated the biocompatibilityof Sr-containing glass-ceramics and bioactive glasses [2e5].Strontium incorporated into bone cements has been found toimprove bone formation as well as reduce bone resorption in vivo.Therefore, there has been a considerable research interest todevelop Sr-containing bone substitute materials. Although, thein vivo biocompatibility of Sr-containing glass ceramics has notbeen comprehensively studied, several studies reported the in vitrobiocompatibility of multi-component glass-ceramics. For example,

tegration of a strontium containing glass ceramic in a rabbit model,

Fig. 1. Gross images of a rabbit femur immediately after the implantation of (a) HA-bioglass composite (control) and (b) strontium containing glass ceramic (test).

A. Sabareeswaran et al. / Biomaterials xxx (2013) 1e92

SiO2eP2O5eNa2OeCaO based glass ceramics with 0e100% Sr-substitution for Ca, supported increased cell growth of osteoblasts(human osteosarcoma cell line, Saos-2) as well as significantlyhigher alkaline phosphatase expression (differentiation marker)with the highest being measured for both 50 and 100% Sr-substitution [6]. Most importantly, a typical resorption activity ofosteoclasts was also observed in these glass ceramics. Based on thehistomorphometrical analysis to compare the efficacy of Sr-dopedhydroxyapatite (HA) on fast resorption of allograft and bonegrowth as bone graft extender at Ti alloy/host bone interface,Vestermark et al. reported that Sr-HA can induce 1.2 fold increase inbone growth along with 1.4 fold increase in neobone formation incomparison to Sr-free HA-based bone graft extender after 4 weeksof osseointegration in a dog animal model [4]. It was suggested thatCa9.51Sr0.49 (PO4)6(OH)2 enhanced the bone defect healing withincreased neobone formation as well as delaying the allograftresorption after 4 weeks.

A systematic and detailed study to assess the neobone forma-tion of Sr-containing biomaterials over various time points was notcarried out in any of the above mentioned studies. Considering thatSiO2eAl2O3eP2O5eSrOeSrF2 glass ceramics are in vitro cytocom-patible, this specific glass ceramic composition is being investi-gated for in vivo biocompatibility in the present work in a rabbitmodel. A detailed analysis using micro-computed tomography,fluorescence microscopy, of the in vivo tested implant/tissue sam-ples enabled us to evaluate the bone regeneration process bothquantitatively and qualitatively.

2. Materials and methods

2.1. Materials

The glasses with the composition 4.5SiO2e3Al2O3e1.5P2O5e3SrOe2SrF2 wereprepared by a melting-quench route, as described earlier [7]. Appropriate amountsof analytical grade oxide powders supplied by SigmaeAldrich of silica (SiO2),alumina (Al2O3), phosphorus pentoxide (P2O5), strontium carbonate (SrCO3) andstrontium fluoride (SrF2), all supplied by SigmaeAldrich were ball milled togetherand melted in a platinum crucible at 1475 �C for 2 h. The melt was then quenched indeionised water in order to obtain amorphous frit glass. The frit was then ball milledin a planetary ball mill using agate jar-agate ball and acetone as a grinding mediumfor 12 h at a speed of 250 rpm. The ball milled powders were dried overnight at100 �C and the dried powders were sieved in order to obtain a particle size of lessthan 45 mm. The pressureless sintering of the powder compact was carried out at1200 �C for 2 h in air in a conventional sintering furnace. All the samples were foundto be around 95% dense. After sintering, the glass was crystallised to a strontiumfluorapatite phase and a Sr-celsian phase (feldspar) as identified by XRD analysis(not shown).

The above processed glass-ceramics are used as test implants and designated asLG26Sr throughout the paper. Commercially available Biograft� HABG (IFGL Bio-ceramics, India) was used as the control implant. This new generation bone graftsubstitute material, processed by a solegel based technique, contains a patentedcomposition of synthetic hydroxyapatite (bone mineral) and calcium phosphatesilicate. This is a unique bioactive composite, which is one of the osteoinductive andresorbable bioceramics, currently being manufactured by IFGL Bioceramics, India.Recent studies reported a faster healing of bone defects using HABG [8,9]. Biograft�

(designated throughout this document as HABG), is specially designed for repairingintra-bony defects including bone cavities and defects arising due to cyst, tumour ortrauma. The surface of samples, Sr-containing glass ceramics and HABG appeared tobe irregularly porous.

2.2. In vivo implantation

Cylindrical shaped (6 mm � 2 mm) test and control implants of 4.5SiO2e

3Al2O3e1.5P2O5e3SrOe2SrF2 and Biograft� HABG respectively, were implanted inrabbits to study bone healing. The implants were subjected to ultrasound cleaning indistilled water for 10 min, dried overnight at 70 �C in a hot air oven (MemmertGmbH) and sterilized by autoclaving at a pressure of 15 psi at 121 �C for 15 min. Allthe animal experiments in the study were carried out with prior approval of theInstitute Animal Ethics Committee (IAEC).

Both test and control materials were implanted in the femur of 10 New ZealandWhite Rabbits (Sctb: Nncl NZW) of body weight more than 2 kg and of either sexesas per ISO 10993-6. The experimental animals were anaesthetized with xylazine(5 mg/kg body weight), ketamine (35 mg/kg body weight), and midazolam (0.3 mg/

Please cite this article in press as: Sabareeswaran A, et al., Early osseoinBiomaterials (2013), http://dx.doi.org/10.1016/j.biomaterials.2013.08.070

kg body weight) and maintained by continuous propofol infusion (0.4 mg/kg bodyweight/min). The animals were controlled on lateral recumbency and a linearincision was made on the cranio-lateral aspect of the thigh under standard asepticprecautions. The shaft of the femur was exposed after a blunt dissection through thejunctions of the tensor Fascia Lata and Vastus Lateralis muscle. Three cortical defectsof approximately 2.5e3 mm diameter were created approximately 10 mm apart inthemid shaft of each femur with a 1.5 mm drill using surgical micromotor (SUNI TM,Expert system SATELEC, France) with continuous saline irrigation. Each experi-mental animal received 3 test implants (Sr-containing glass ceramic designated asLG26Sr) in the left femur and 3 control implants (designated as HABG) in the rightfemur. A total of 30 numbers of LG26Sr and 30 numbers of HABG implants wereimplanted. Each implant was pre-soaked with saline and press fitted into the defect(Fig. 1). Implant position was confirmed by radiography. The surgical wound wasclosed in layers. Both ceftriaxone (20 mg/kg body weight) and meloxicam (0.25 mg/kg body weight) were administered intramuscularly for 5 days post-implantation.

At 4 and 12weeks post-implantation, 5 animals were euthanized by an overdoseof thiopentone sodium at each time period (total of 15 LG26Sr and 15 HABG implantsat 4 and 12 weeks). Also, the implantation sites were grossly examined for healing ofbone defects. The femurs with the implant materials were removed and fixed in 10%neutral buffered formalin.

2.3. Bone labelling

Bone labelling to identify site of new bone deposition post-implantation over aperiod of 4 and 12 weeks was pursued in one rabbit of each time period (3 LG26Srand 3 HABG implants at 4 and 12 weeks respectively) by sequential intramuscularadministration of two different flurochromes, namely xylenol orange and alizarinred at specified time intervals. Table 1 provides the dose and schedule of dyeadministration.

2.4. Micro-CT analysis

Mineralization of newly formed bone, overall morphology of the neobonearound the implant and quantification of neobone fractionwith respect to host bonewere carried out using high-resolution X-ray micro-computed tomography (mCT 40,

tegration of a strontium containing glass ceramic in a rabbit model,

Fig. 2. Digital radiographic image of LG26Sr (black dots) and HABG (radio-opaque)implants in the rabbit femur.

Fig. 3. Histological images of the host bone-implant interface in the case of controlimplant (HA-bioglass) after implantation of (a) 4 weeks and (b) 12 weeks.

Table 1Flurochrome dosage schedule used in vital staining during the implantation in thepresent work.

Experimentalanimal groups

Time point and fluorochrome dosage

2 weeks 4 weeks 8 weeks

4 weeks group Xylenol orange (25 mg/kg) e

12 weeks group Xylenol orange (25 mg/kg) Calcein blue(7 mg/kg)

Alizarin red(10 mg/kg)

A. Sabareeswaran et al. / Biomaterials xxx (2013) 1e9 3

Scanco Medical, Switzerland). The scanning was carried out with a slice thickness/slice increment¼ 16 mm, X-ray beam energy¼ 70 kV and X-ray intensity¼ 114 mA. Inparticular, the specific area of the scanned region containing one implant wasselected and the reconstruction of the selected region was made using the cone-beam convergence/back projection algorithm-based software. Based on the den-sity variation of the host bone and the implant, the 3D image of the implant wasextracted on the basis of the distinction between ROI (region of interest) and VOI(volume of interest). The overall scan time was 3,000,000 ms and the entire scandistance was 12,288 mm. From the experimentally measured CT database, a cylin-drical region of interest (ROI) was selected for analysis corresponding to the corticalbone during implantation. In order to evaluate bone regeneration within the defect,ROI was further sectioned transversely as top, medium and bottom and radially asouter, inner and core shells. Also, thresholds were suitably applied to images of eachsample in order to segment the newly formed bone from the residual implant. Afterthresholding, the bone volume (BV) was determined by counting the total number ofbone voxels and multiplying by their known volume, while the total volume (TV)was determined by counting the bone and non-bone voxels.

2.5. Histopathological analysis and fluorescence microscopy

The cross-sectional blocks of bone with implant were cut using a low speed saw(ISOMET 2000, Beuhler) and further fixed in 10% neutral buffered formalin. Blockswere further dehydrated in ascending grades of alcohol, cleared in alcoholic acetone(1:1 v/v) mixture and immersed in two changes of 100% alcohol. The bone blockswere washed out twice using methyl-methacrylate and were embedded in Poly-methylmethacrylate resin. Multiple serial sections (100e150 mm) of implant withadjacent bone were cut from resin blocks using a high speed precision diamond saw(ISOMET 5000, Beuhler), ground and polished in a variable speed grinder-polisher(ECOMET 3000, Beuhler). Subsequently, the thin sections were stained with Ste-venel’s blue, followed by counter staining with Van Gieson’s Picro-Fuchsin. Thestained sections were analysed using a trinoccular light microscope (NIKON Eclipse600, Japan). Several images from bone material interface were captured using adigital camera (DXM1200, Nikon), attached to the light microscope.

The identification of flurochrome labelled new bone deposition was carried outby observing selected unstained sections under a fluorescence microscope (Nikon E600, Tokyo, Japan) and the fluorescence images were captured using a DXM 1200camera attached to the microscope. The G-2A filter with specific filter set up for red(excitation: 530e560 nm, emission: 580 nm) and orange (excitation: 546 nm,emission: 580 nm) was used in order to excite the used fluorochromes at theiroptimal wavelength.

3. Results

The recovery of all the experimental animals post-implantationwas uneventful. There was no mortality or morbidity. The implantsites were healed. There was no evidence of infection or necrosis.All implants were identified by digital radiography and the LG26Srimplants were found to be notably more radio opaque (Fig. 2).

3.1. Histological observations

All implants were found to be press fit in the cavities in corticalbone. In all the histology sections, cortical and trabecular bonewereoriented in a vertical direction from the host bone to the implantsurface. All implants were rectangular and irregularly porous andpresent in the cortical bone and extended into the marrow space.The HABG implants appeared yellowish brown (in web version)(Fig. 3) and LG26Sr implants were greyish white (Fig. 4) underbright field microscopy. Both in case of HABG and LG26Sr implantsite, foci of bone remodelling with rosettes of osteoblasts and os-teoclasts were noted at the interface. Foci of ossified bone matrix,

Please cite this article in press as: Sabareeswaran A, et al., Early osseointegration of a strontium containing glass ceramic in a rabbit model,Biomaterials (2013), http://dx.doi.org/10.1016/j.biomaterials.2013.08.070

Fig. 4. Histological images of host bone/implant interface in the case of test implant(LG26Sr) after implantation of (a) 4 weeks and (b) 12 weeks.

A. Sabareeswaran et al. / Biomaterials xxx (2013) 1e94

osteoid and cells were observed in the pores and grooves of boththe implant groups (Figs. 3 and 4). Mature newwoven and lamellarbone with Haversian systems were observed both at the periostealand endosteal aspects of the cavity with new bone deposition withosteoid and osteoblasts extending onto the implant in marrowspace in both the implant groups. Thin layers of woven bone werealso seen on the implant surface in marrow space in both theimplant groups. There was no evidence of inflammation or bonenecrosis at the implant site in both the implant groups. Multipleareas of indistinguishable apposition were noticed between thebone and the implant. It may be of interest tomention here that in arecent study carried out by Tripathi et al., osseointegration ofpolymer-ceramic hybrid biocomposites with similar histologicalfeatures was observed in a rabbit model at the bone-implantinterface [10].

Fig. 5. Fluorescence images of host bone-implant interface, at 12 weeks post-implantation to illustrate the deposition of new bone at the interface with HABG (a)and LG26Sr glass ceramic (b), as evident from the fluorescent labeling/staining withxylenol orange.

3.2. Bone labelling

In addition to histological investigation, the widely used tech-nique of polychromatic fluorochrome labelling was adopted in thepresent study, which provides direct evidence for bone formationin vivo at different time intervals. The dosage treatment is sum-marized in Table 1. It can be reiterated here that the polychromatic

Please cite this article in press as: Sabareeswaran A, et al., Early osseoinBiomaterials (2013), http://dx.doi.org/10.1016/j.biomaterials.2013.08.070

fluorescent stains using calcium-binding fluorochromes weredeposited at various active sites of biomineralisation in vivo withdifferent fluorochrome colours providing sequential informationregarding the accretion and progression of bone formation [11]. InFig. 5, all the intravitally administrated fluorochrome labels wereclearly visible with alizarin red providing red signatures (in webversion). The accumulation sequence of these fluorochrome labelsindicated that initial bone formation started at the periphery of thebone defects and the active ingress of new bone. A clear enhance-ment of bone formation on the surface of the LG26Sr ceramics wasclearly observed and this is comparable with the control implant(Fig. 5). The first signature of ossification was observed on theimplant surface after 4 weeks of implantation, indicating that thebone formation started as early as 2 weeks post-implantation (seeTable 1 and Fig. 5).

3.3. Micro-CT analysis

Micro-CT is a useful technique to characterise the natural femurin rabbits [12e16] as well as to quantify the bone regenerationaround synthetic implants [15e20]. Micro-CT systems use

tegration of a strontium containing glass ceramic in a rabbit model,

A. Sabareeswaran et al. / Biomaterials xxx (2013) 1e9 5

effectively micro-focal X-ray projections rotated through multipleviewing directions in order to obtain 3-D reconstructed images,which provide spatial distribution maps of linear attenuation co-efficients determined by the energy of X-ray source and sample/bone composition. As part of the histomorphometric study, director indirect methods can be used to analyse the bone density,thickness, spacing as well as connectivity and anisotropy onthreshold and reconstructed images. In efforts to improve the ef-ficacy of micro-CT analysis, Buie et al. developed a fully automatedsegmentation algorithm to efficiently extract periosteal andendosteal surfaces of cortex and such a model has been tested fortibia or radius of mouse, rat and human with nominal isotropicresolution of 10e82 mm [14]. In another study, Voor et al. studiedthe influence of five different scanning conditions with increased/reduced X-ray power against base settings (86kv, 110 mA) on thehistomorphic measurements of rabbit femurs [16]. The extensiveanalysis revealed the repeatability and reproducibility inmeasuringbone volume or trabecular bone separation or thickness, irre-spective of scanning conditions. Nair et al. reported the neoboneformation of (97.5 � 1.9)% around the porous biodegradable tri-phasic bioceramic (CaSiO3 þ TCP þ HA)-coated HA [HASi ceramic]after 12 months in goat animal model [20]. This study establishesthe osteoconductive nature of HASi to regenerate/remodel bone ina segmental defect (2 cm) in a goat’s femur.

Fig. 6. Microcomputed tomographs showing three-dimensional morphology of newly form(c) 12 weeks as well as HA-bioglass based control implant after implantation of (b) 4 week

Please cite this article in press as: Sabareeswaran A, et al., Early osseoinBiomaterials (2013), http://dx.doi.org/10.1016/j.biomaterials.2013.08.070

In the present work, the analysis of the micro-CT results hasbeen made in reference to the following aspects: a) morphology ofneobone and implant material in three dimensions after carefulreconstruction of a number of slices and extracting the host bonefrom the reconstructed image, b) grey scale image revealing howthe implant was placed into the cylindrical defect at the intra-medullary region, c) bone mineralization of implant as measuredby the bone mineral density on 2D micro-CT images and d) 3Dcolour images to evaluate the thickness variation of the tissues onand around the implant surface. As far as the first aspect is con-cerned, the 3D morphology of the implant after 4 and 12 weeks ofimplantation revealed a rough surface morphology with a highratio of bone volume (BV) to total volume (TV). It was clear fromextensive micro-CT analysis that both the HABG and LG26Sr im-plants were in physical contact with the neighbouring host bone.Surface morphology of both implants at both time periods revealeda rough surface. As shown in Fig. 6, the BV/TV ratio significantlyincreased from 0.62 to 0.81 for LG26Sr implant with increasingimplantation time from 4 to 12 weeks. On the other hand, in thecase of HABG control implant, the BV/TV ratio increased only from0.77 to 0.80 with increasing implantation time from 4 to 12 weeks,respectively. Such observations clearly reveal excellent boneregeneration at 4 weeks for the HABG control implant, while boneregeneration is comparable after 12 weeks for both LG26Sr and

ed bone over residual LG26Sr based test implant after implantation of (a) 4 weeks ands and (d) 12 weeks.

tegration of a strontium containing glass ceramic in a rabbit model,

A. Sabareeswaran et al. / Biomaterials xxx (2013) 1e96

HABG (Fig. 6c and d). As shown in Fig. 7a and c, both the control andtest implants were in physical contact with the neighbouring hostbone. Tissue growth around the implant was suggested by somefiner features present in the CT-images (Fig. 7). Overall, the CT-images suggested that the LG26Sr implant showed a good inte-gration with the bone within the cylindrical defects, irrespective ofthe implantation time.

The quantification of the bone mineralization density (BMD) ofthe newly formed bone and the BMD distribution graph across the2D slices showed extremely high values of around 2800 mgHA/ccm, irrespective of the implantation time of up to 12 weeks.However, the baseline BMD value remained almost similar for 4and 12 weeks of implantation in the case of the control implant(Fig. 8b and d). Additionally, in the case of the control implant, aperfect continuity of the newly formed bone with the surroundinghost bone was observed at all time points. In contrast, a sharp in-crease in BMD over a few micrometre of width as one crosses fromthe host tissue to the test implant (LG26Sr) was recorded at all thetime points (Fig. 7b and d). This increase as well as much higherBMD in the case of the test implants must be due to a highermineral density of the test implants (prior to implantation), used inthe present study. Nevertheless, the density (BMD) of the newlyformed bone on both test and control implants was uniformthroughout the tissue construct.

In order to illustratemore details, 3D colour tomography imagesof the test implants after 4 and 12 weeks are shown in Fig. 9. After 4

Fig. 7. 3D micro-CT tomographs in grey scale images revealing the physical integration of2 mm diameter as well as selected 2D slices along with the bone mineralisation density dtimetre] acquired from newly formed bone at post-implantation of (a and b) 4 weeks and

Please cite this article in press as: Sabareeswaran A, et al., Early osseoinBiomaterials (2013), http://dx.doi.org/10.1016/j.biomaterials.2013.08.070

weeks of post-implantation, the test implant was found to becovered with tissues of around 1 mm of thickness with focalthickness of 1.5 mm or above (Fig. 9a), indicating good integrationin the osseous system. Similarly, the entire LG26Sr implant in theintermedullary cavity of the cylindrical defect was covered uni-formly by relatively thicker tissue of 2 mm or above after 12 weekspost-implantation (Fig. 9b).

4. Discussion

The longevity of orthopaedic prostheses depends upon theimplant fixation and interfacial stability. Aseptic loosening andfailure of an implant are often reported due to instability andincomplete anchorage at the bone/implant interface [21]. There aredifferent approaches to improve the stability and anchorage andthese approaches include bioactive fixation using compositionallytailored biomaterials, cemented fixation, mechanical bone implantinterlocking, etc. This study investigates the bioactive fixation of aSr-containing glass ceramic, based on the glass composition4.5SiO2e3Al2O3e1.5P2O5e3SrOe2SrF2. The osteointegration of theLG26Sr glass ceramic implant with the host bonewas assessed overtwo different time points of 4 and 12 weeks of implantation.

The studies related to the in vivo biocompatibility of calciumphosphate-based biomaterials [22e25] are much more in com-parison to glass-ceramics [26e28]. Gauthier et al. [27] studied theefficacy of biphasic calcium phosphates (BCP) with two different

LG26Sr based test implants in the intermedullary cylindrical shaped cortical cavity ofistribution plot [mgHA/ccm indicating milligram of hydroxyapatite, HA per cubic cen-(c and d) 12 weeks.

tegration of a strontium containing glass ceramic in a rabbit model,

Fig. 8. Representative 3D micro-CT tomographs in grey scale images revealing the physical integration of HA-bioglass based control implants in the intermedullary cavity of 2 mmdiameter as well as selected 2D slices along with the bone mineralisaiton density distribution plot [mgHA/ccm indicating milligram of hydroxyapatite, HA per cubic centimetre]acquired from newly formed bone, respectively, post-implantation for two different time points: (a, b) 4 weeks and (c, d) 12 weeks.

A. Sabareeswaran et al. / Biomaterials xxx (2013) 1e9 7

pore sizes of 80e200 mm or 200e500 mm using implantation ex-periments for 6 weeks in critical sized bone defects in a rabbitanimal model. An extensive use of micro-CT analysis revealed ahigh interconnectivity of the new bone structure with morequantified bone volume being measured with BCP of 80e200 mmsize (BV/TV¼ 38.5� 5.5) in comparison to BCP of 200e500 mm (BV/TV ¼ 24.8). Vogel et al. reported the bone regeneration ability ofbioglass particles of three different compositions (45S5:45SiO2e

24.5Na2Oe24.5CaOe6P2O5;52S:52SiO2e21Na2Oe21CaOe6P2O5and 55S; 55SiO2e19.5Na2Oe19.5CaOe6P2O5, all compositions inwt%), when implanted in the distal femoral epiphysis in rabbits for7, 28 and 84 days [29]. A major observation of this study is thein vivo degeneration of the investigated bioglass particles either toSr-rich remnants or to CaP-rich shells as well as the finding ofhigher number of multinuclear giant cells (MNGC) on the degradedbioglass remnants. In another study, the same research group re-ported an increased number of MNGCwith increasing implantationtime from 28 days to 84 days with the highest number of MNGCsbeing measured in the case of the widely researched 45S5 bioglass[28]. In a relatively recent study, Gorustovich et al. reported theeffectiveness of Sr-containing 45S5 bioglass for bone tissue repairin a rat model [26]. The addition of 6 wt% SrO to substitute for CaOin 45S5 bioglass did not cause any statistically significant differencein bone bonding behaviour in terms of affinity index, beingmeasured as the neobone length in contact with the implant sur-face (expressed as % of total length of implant) after 30 days ofimplantation in rat tibia. The overall affinity index (w89%) of 45S5-

Please cite this article in press as: Sabareeswaran A, et al., Early osseoinBiomaterials (2013), http://dx.doi.org/10.1016/j.biomaterials.2013.08.070

6% SrO bioglass is quite notable after implantation in rat tibia for 4weeks. The histomorphometric analysis revealed an increase in theaffinity index from 73.6� 3.5% after 3 months to 85.2� 2.7% after 6months of implantation. Importantly, 45S5 bioglass had muchlarger affinity index in the rat model than that measured (30e60%)with an identical glass composition in a rabbit model after similarimplantation period, as reported by Vogel et al. [28]. In anotherstudy, Li et al. reported the progression of bone bonding of Sr-HAbased bioactive bone cement into the cancellous bone of the iliaccrest of rabbits [30]. In a different study to assess the influence ofdicalcium phosphate coatings on the stability of Ti- implants inrabbit model, micro-CT analysis revealed BV/TV ratio of 44.6% in thecase of coated Ti after 6 weeks of implantation [22].

In the present work, the BV/TV ratio in the case of LG26Sr glassceramic was measured to be 80% after 12 weeks of implantation.The BV/TV ratio in the case of LG26Sr increased significantly withimplantation time, implying greater bone regeneration capabilitywith time. The much more new bone deposition at the LG26Srinterface in comparison to that deposited to the HABG interface asearly as 2 weeks post-implantation is also evident by flurochromelabelled images (Fig. 5). Furthermore, the radio opacity of LG26Srimplants is an added advantage for clinical identification.

In our earlier research, the rabbit animal model is used toestablish the osseointegration of HA-mullite [25] and HA-CaTiO3[31] composites. In both cases, the short term implantation over 12weeks confirmed good in vivo biocompatibility and in particular,confirmed the lack of in vivo toxicity of mullite reinforcement [25]

tegration of a strontium containing glass ceramic in a rabbit model,

Fig. 9. 3D micro CT tomographs in colour contrast of the neobone around LG26Srbased test implant at two different time points: (a) 4 weeks and (b) 12 weeks. Thecolour scheme shows the variable thickness of around 1 mm higher in the neobonearea. (For interpretation of the references to colour in this figure legend, the reader isreferred to the web version of this article.)

A. Sabareeswaran et al. / Biomaterials xxx (2013) 1e98

or the better osteogenesis property of electroconductive HA-CaTiO3

than sintered HA [31]. In both these studies, only histologicalanalysis was carried out. In this perspective, the combinationof histological, micro-CT and polychrome fluorescent analysistogether establish the osteoconductive, non-biodegradable andosseointegration property with excellent close appositionwith hostbone of 100% Sr substituted glass ceramics. Such glass ceramics canalso be used as an alternative treatment option for osteoporosis.

Systemic administration of strontium ranelate (SR) has beenfound to improve fixation of hydroxyapatite (HA) - coated titanium

Please cite this article in press as: Sabareeswaran A, et al., Early osseoinBiomaterials (2013), http://dx.doi.org/10.1016/j.biomaterials.2013.08.070

screws in ovariectomized (OVX) rats [32]. The present incorpora-tion of Strontium in glass ceramic assures a regular delivery ofstrontium to host and new bone, thus facilitating good bonedeposition. This would definitely improve fixation of joint pros-thesis and concurrently reduce activation of osteoclasts and pre-vent long term host bone resorption. Further in vitro studies withosteoclasts and long term in vivo studies in animal models ofosteoporosis are required to find out the beneficial effect of stron-tium in older patients with joint replacement.

5. Conclusions

The in vivo implantation study in a rabbit model revealed thatthe investigated glass ceramics with the nominal composition of4.5SiO2e3Al2O3e1.5P2O5e3SrOe2SrF2 are non-biodegradable andosteoconductive. Overall, the investigated implant materialsenabled dynamic healing in a rabbit animal model. The boneregeneration ability of the investigated glass ceramic was quanti-tatively similar to that of hydroxyapatite-based bioglass (controlimplant) after 12 weeks of implantation, as determined usingmicro-computed tomography analysis. The polychrome sequencelabelling of fluorochromes confirmed both the bony reconstructionand new bone formation around the Sr-containing glass ceramic.These results were similar and comparable to that observed withthe bioactive HA-bioglass composite, used as control implant in thepresent study. In summary, the present study establishes theosseointegration of Sr-containing glass ceramics during short termimplantation in a rabbit model.

Acknowledgements

The authors would like to thank the UK-India Education andResearch Initiative (UKIERI), administered by Department of Sci-ence and Technology (DST), Government of India and the Britishcouncil, UK, for the research funding. The authors also thank DrKalliyana Krishnan, Sree Chitra Tirunal Institute for Medical Sci-ences and Technology (SCTIMST), Trivandrum, India for assistanceprovided during the micro-CT evaluation. The authors also thankDr. H. K. Varma and Dr. Mira Mohanty, SCTIMST, Trivandrum, Indiafor providing us the control implant samples and for critical com-ments during the revision of the manuscript, respectively. Theauthors also acknowledge the fruitful discussion with Mr. ParthaMukherjee, JV Scientific Instruments Pvt. Ltd., India, whileanalyzing the micro-CT results.

References

[1] Basu B, Katti D, Kumar A. Advanced biomaterials fundamentals, processingand applications. Hoboken, NJ: John Wiley & Sons Ltd; 2009.

[2] Ravi ND, Balu R, Sampath Kumar TS. Strontium-substituted calcium deficienthydroxyapatite nanoparticles: synthesis, characterization, and antibacterialproperties. J Am Ceram Soc 2012;95(9):2700e8.

[3] Goel A, Rajagopal RR, Ferreira JM. Influence of strontium on structure, sin-tering and biodegradation behaviour of CaOeMgOeSrOeSiO2eP2O5eCaF2glasses. Acta Biomater 2011;7(11):4071e80.

[4] Vestermark MT, Hauge EM, Soballe K, Bechtold JE, Jakobsen T, Baas J. Stron-tium doping of bone graft extender. Acta Orthop 2011;82(5):614e21.

[5] Hamdy NA. Strontium ranelate improves bone microarchitecture in osteo-porosis. Rheumatology 2009;48(iv):9e13.

[6] Gentleman E, Fredholm YC, Jell G, Lotfibakhshaiesh N, O’Donnell MD, Hill RG,et al. The effects of strontium-substituted bioactive glasses on osteoblasts andosteoclasts in vitro. Biomaterials 2010;31(14):3949e56.

[7] Hill RG, Stamboulis A, Law RV, Clifford A, Towler MR, Crowley C. The influenceof strontium substitution in fluorapatite glasses and glass-ceramics. J Non-Cryst Solids 2004;336(3):223e9.

[8] Acharya NK, Mahajan CV, Kumar RJ, Varma HK, Menon VK. Can iliac crestreconstruction reduce donor site morbidity?: a study using degradablehydroxyapatite-bioactive glass ceramic composite. J Spinal Disord Tech2010;23(4):266e71.

tegration of a strontium containing glass ceramic in a rabbit model,

A. Sabareeswaran et al. / Biomaterials xxx (2013) 1e9 9

[9] Nair MB, Suresh Babu S, Varma HK, John A. A triphasic ceramic-coated poroushydroxyapatite for tissue engineering application. Acta Biomater 2008;4(1):173e81.

[10] Tripathi G, Gough JE, Dinda A, Basu B. In vitro cytotoxicity and in vivoosseointegration property of compression moulded HDPE-HA-Al2O3 hybridbiocomposites. J Biomed Mater Res A 2013;101(6):1539e49.

[11] Oest ME, Dupont KM, Kong HJ, Mooney DJ, Guldberg RE. Quantitativeassessment of scaffold and growth factor-mediated repair of critically sizedbone defects. J Orthop Res 2007;25(7):941e50.

[12] Numata Y, Sakae T, Nakada H, Suwa T, LeGeros RZ, Okazaki Y, et al. Micro-CTanalysis of rabbit cancellous bone around implants. J Hard Tissue Biol2007;16(2):91e3.

[13] Guldberg RE, Ballock RT, Boyan BD, Duvall CL, Lin AS, Nagaraja S, et al.Analyzing bone, blood vessels, and biomaterials with microcomputed to-mography. IEEE Eng Med Biol Mag 200;22(5):77e83.

[14] Buie HR, Campbell GM, Klinck RJ, MacNeil JA, Boyd SK. Automatic segmen-tation of cortical and trabecular compartments based on a dual thresholdtechnique for in vivo micro-CT bone analysis. Bone 2007;41(4):505e15.

[15] Pazzaglia UE, Zarattini G, Giacomini D, Rodella L, Menti AM, Feltrin G.Morphometric analysis of the canal system of cortical bone: an experimentalstudy in the rabbit femur carried out with standard histology and micro-CT.Anat Histol Embryol 2010;39(1):17e26.

[16] Voor MJ, Yang S, Burden RL, Waddell SW. In vivo micro-CT scanning of arabbit distal femur: repeatability and reproducibility. J Biomech 2008;41(1):186e93.

[17] Meinel L, Betz O, Fajardo R, Hofmann S, Nazarian A, Cory E, et al. Silk basedbiomaterials to heal critical sized femur defects. Bone 2006;39(4):922e31.

[18] Takemoto M, Fujibayashi S, Neo M, Suzuki J, Kokubo T, Nakamura T. Me-chanical properties and osteoconductivity of porous bioactive titanium. Bio-materials 2005;26(30):6014e23.

[19] Guldberg RE, Duvall CL, Peister A, Oest ME, Lin AS, Palmer AW, et al. 3D im-aging of tissue integration with porous biomaterials. Biomaterials2008;29(28):3757e61.

[20] Nair MB, Varma H, Shenoy SJ, John A. Treatment of goat femur segmentaldefects with silica coated hydroxyapatite-one year follow-up. Tissue Eng PartA 2010;16(2):385e91.

Please cite this article in press as: Sabareeswaran A, et al., Early osseoinBiomaterials (2013), http://dx.doi.org/10.1016/j.biomaterials.2013.08.070

[21] Sundfeldt M, Carlsson LV, Johansson CB, Thomsen P, Gretzer C. Aseptic loos-ening, not only a question of wear: a review of different theories. Acta Orthop2006;77(2):177e97.

[22] Simank HG, Stuber M, Frahm R, Helbig L, van Lenthe H, Muller R. The influenceof surface coatings of dicalcium phosphate (DCPD) and growth and differen-tiation factor-5 (GDF-5) on the stability of titanium implants in vivo. Bio-materials 2006;27(21):3988e94.

[23] Rai B, Oest ME, Dupont KM, Ho KH, Teoh SH, Guldberg RE. Combination ofplatelet-rich plasma with polycaprolactone-tricalcium phosphate scaffolds forsegmental bone defect repair. J Biomed Mater Res A 2007;81(4):888e99.

[24] Shao XX, Hutmacher DW, Ho ST, Goh JC, Lee EH. Evaluation of a hybridscaffold/cell construct in repair of high-load-bearing osteochondral defects inrabbits. Biomaterials 2006;27(7):1071e80.

[25] Nath S, Basu B, Mohanty M, Mohanan PV. In vivo response of novel calciumphosphate-mullite composites: results up to 12 weeks of implantation.J Biomed Mater Res B Appl Biomater 2009;90(2):547e57.

[26] Gorustovich AA, Steimetz T, Cabrini RL, Porto Lopez JM. Osteoconductivity ofstrontium-doped bioactive glass particles: a histomorphometric study in rats.J Biomed Mater Res A 2010;92(1):232e7.

[27] Gauthier O, Muller R, von Stechow D, Lamy B, Weiss P, Bouler JM, et al. In vivobone regeneration with injectable calcium phosphate biomaterial: a three-dimensional micro-computed tomographic, biomechanical and SEM study.Biomaterials 2005;26(27):5444e53.

[28] Vogel M, Voigt C, Knabe C, Radlanski RJ, Gross UM, Muller-Mai CM. Devel-opment of multinuclear giant cells during the degradation of bioglass particlesin rabbits. J BiomedMater Res A 2004;70(3):370e9.

[29] Vogel M, Voigt C, Gross UM, Muller-Mai CM. In vivo comparison of bioactiveglass particles in rabbits. Biomaterials 2001;22(4):357e62.

[30] Li YW, Leong JC, Lu WW, Luk KD, Cheung KM, Chiu KY, et al. A novel injectablebioactive bone cement for spinal surgery: a developmental and preclinicalstudy. J Biomed Mater Res 2000;52(1):164e70.

[31] Mallik PK, Basu B. Better early stage osteogenesis of electroconductiveHA-CaTiO3composites in a rabbit animal model. J BiomedMater Res A 2013 in press.

[32] Bain SD, Jerome C, Shen V, Dupin-Roger I, Ammann P. Strontium ranelateimproves bone strength in ovariectomized rat by positively influencing boneresistance determinants. Osteoporos Int 2009;20(8):1417e28.

tegration of a strontium containing glass ceramic in a rabbit model,