Multi-layer porous fiber-reinforced composites for implants: In vitrocalcium phosphate formation in the presence of bioactive glass

Abstract

Objectives

Glass-fiber-reinforced composites (FRCs), based on bifunctional methacrylate resin, have recently shown their potential for use as durable cranioplasty, orthopedic and oral implants. In this study we suggest a multi-component sandwich implant structure with (i) outer layers out of porous FRC, which interface the cortical bone, and (ii) inner layers encompassing bioactive glass granules, which interface with the cancellous bone.

Methods

The capability of Bioglass ® 45S5 granules (100–250 μm) to induce calcium phosphate formation on the surface of the FRC was explored by immersing the porous FRC-Bioglass laminates in simulated body fluid (SBF) for up to 28 d.

Results

In both static (agitated) and dynamic conditions, bioactive glass granules induced precipitation of calcium phosphate at the laminate surfaces as confirmed by scanning electron microscopy.

Significance

The proposed dynamic flow system is useful for the in vitro simulation of bone-like apatite formation on various new porous implant designs containing bioactive glass and implant material degradation.

Introduction

Durable, biostable fiber-reinforced composites (FRC), which were originally started to be developed for applications in dentistry in early 1960s took until mid-1990s until they made their way into the routine clinical use in various types of dental restorations . Among the applications of prosthetic and restorative dentistry, FRC has been tested as implant material in craniofacial, orthopedic and oral implantology. The implant material is based on co-polymer matrices out of polymethyl methacrylate (PMMA) or bisphenol-A-dimethacrylate (BisGMA) and triethyleneglycoldimethacrylate (TEGDMA) reinforced with E-glass fibers. The potential of FRCs as implant material for craniofacial bone reconstructions has also been demonstrated . In cell culture and in vivo studies, the photopolymerized BisGMA-TEGDMA resin system has shown good biocompatibility and bone formation capability .

Further enhancement of the FRC biocompatibility can be attained by mimicking fibrous structure of bone at the implant–bone interface . To achieve this goal, we suggest a sandwich implant structure based on multiple components. The outer implant layers, which interface the cortical bone, are made of porous FRC, while the inner layers, which interface the cancellous bone, comprise a biologically active component. This implant structure would allow attachment, proliferation, and migration of cells leading to vascularization and rapid bone in-growth for the stabilization of the implant in medium and long terms . 45S5 Bioglass ® (BG), a biocompatible and osteoconductive bone graft substitute material , is approved by the U.S. Food and Drug Administration for certain clinical applications in orthopedic and cranio-maxillofacial surgery . For this reason this bioactive glass was chosen to be used as biologically active component in the FRC implant structures.

The formation of a bone-like apatite layer on the implant surface is essential for the achievement of bone-bonding and osteoconductivity . In contact with body fluids BG is capable of forming such an apatite layer . The layer formation can also be simulated in vitro . In addition, it has been demonstrated that the presence of BG can lead to apatite formation on the surface of adjacent bioinert polymer materials . The induction of this so called “halo” effect to the sandwich FRC implant structures is expected to enhance the implant attachment and also to improve the bone ingrowth into the porous FRC layers .

The goal of this work was to study the calcium phosphate (CaP) formation induced on the surfaces of the FRC implant by BG granules. Therefore, experimental in vitro set-ups were developed to describe the reactions taking place upon immersion of composite structures in simulated body fluid (SBF) in both static and dynamic conditions with various arrangements of the bioactive component. Detailed knowledge of the in vitro and in vivo reactions is essential for the development of different implant structures for craniofacial applications.

Materials and methods

Experimental arrangements

The capability of bioactive glass granules to induce CaP formation on the composite surfaces during immersion in SBF was studied by using two different experimental conditions, static and dynamic. In the static tests, FRC specimens were incubated in SBF in the presence and in the absence of separate BG granules. Incubated specimens included (1) FRC1 : control polished resin specimens without reinforcement, (2) FRC2 : porous FRC, and (3) FRC3 : laminate FRC structures. Specimen types FRC1 and FRC2 were included to study the influence of the distance between the composite surfaces and BG granules on the in vitro behavior. FRC3 was a surrogate of a cranial implant. In the dynamic test, a flow of SBF was passed through the FRC3 specimens. Details of experimental set-ups are shown in Fig. 1 .

Fig. 1
Scheme of static (a) and dynamic (b) immersion test set-up: left: pump system, right: Cross section of FRC3 specimen in tube reactor.

FRC2 was made of a randomly oriented chopped silanized E-glass veil impregnated within the BisGMA/TEGDMA polymer matrix. This composite was used to create multi-component laminates ( FRC3 ) which had two layers of BG granules entrapped between three layers of porous FRC. The BG granules were retained by the polymer, but not embedded into it.

Preparation of the specimens

The FRC1 specimens were prepared by pouring resin mixture consisting of BisGMA (50 wt%), TEGDMA (50 wt%), camphorquinone (0.7 wt%) and DMAEMA (0.7 wt%) into putty molds and then polymerized by light curing. Subsequently, the specimens were ground (LaboPol-21 Grinding Machine, 300 rpm, Struers A/S, Rødovre, Denmark) to the final dimension of 9 mm × 59 mm × 2.9 mm (each ±0.2 mm) and the surfaces were polished with grit 4000 SiC paper. The photo-initiated polymerization of the specimens was carried out in three stages to ensure minimal content of residual monomers. First, the specimens were pre-cured by a dental hand cure device (Elipar ® S10 LED curing light, 3M/ESPE, Seefeld, Germany) with an exposure time of 120 s (intensity: 1200 mW/cm 2 , wavelength: 430–480 nm). Second, the specimens were post-cured in a vacuum light oven (Visio Beta vario, 3M/ESPE, Seefeld, Germany) for 15 min in ambient temperature and finally in a light oven (Targis Power, Ivoclar Vivadent AG, Schaan, Lichtenstein) for 25 min at a temperature of 95 °C.

The FRC2 specimens were prepared from 1 mm thick E-glass fiber sheets (Ahlstrom Glassfiber Oy, Kotka, Finland). The randomly oriented fibers in the sheets had a diameter of 10 μm and a length of 8 mm, and were bound together by an acrylic binder with a methacrylate silane coupling agent. Pieces of 9 mm × 60 mm in size were cut from these sheets and scaled. The cut pieces were then impregnated with 1–1.5 ml of the BisGMA/TEGDMA resin mixture, and were left for 48 h in the dark. After the impregnation, the excessive resin was removed by compressing and the specimens were then light cured as described above. These procedures ensured that 75–81 wt% resin remained in the FRC2 specimen structure, which was expected to result in a total porosity of 39–49 vol% with more than 90% of functional (open) porosity . The average pore size was larger than 100 μm. The reinforcing E-glass fibers in the specimens were covered by polymer resin.

The FRC3 specimens were prepared by assembling two pieces of FRC2 with fiber bundles (everStick ® Ortho, Ø 0.75 mm, StickTech Ltd., Turku, Finland) serving as spacers ( Fig. 2 ) between the layers to allow the incorporation of the BG granules. Each of the two BG layers contained 45S5 Bioglass ® granules of the 100–250 μm fraction (composition in wt%: SiO 2 : 45, Na 2 O: 24.5, CaO: 24.5, P 2 O 5 : 6). The two sides of the FRC2 pieces along the spacers were sealed with additional resin prior to final light curing ( Fig. 2 ). Smaller specimens (10 mm × 10 mm × 4 mm), containing 60 ± 5 mg of BG per layer were used in the static test. Larger specimens (9.3 mm × 30 mm × 4 mm), comprising 200 ± 10 mg of BG per layer were applied in the dynamic test. The BG content of both smaller and larger FRC3 was around 38% of the total specimen weight. In previous tests, the packed arrangement of the granules in between the FRC2 layers enabled wetting throughout the composite structure .

Fig. 2
(a) Scheme of cranial implant design: A – intact bone, B – brain tissue, C – dura mater, D – screws for implant fixation, E – dense and F – porous FRC layers with Everstick ® space holders (gray) and BG (not marked) in between layers. (b) Scheme of FRC3 specimen.

After preparation the specimens were washed in ultrapure water (milliQ quality), then washed with ethanol, dried at ambient temperature and stored in a desiccator. Specimens for the static immersion test were additional cleaned by sonication for 5 min.

In vitro SBF immersion

The in vitro bioactivity of the specimens was studied using SBF prepared according to Kokubo’s formulation (c-SBF) . The SBF was prepared by dissolving reagent chemicals of NaCl, NaHCO 3 , KCl, K 2 HPO 4 ·3H 2 O, MgCl 2 ·6H 2 O, CaCl 2 ·2H 2 O, and Na 2 SO 4 into deionized water. The fluid was buffered at physiological pH 7.40 at 37 °C with (CH 2 OH) 3 CNH 2 and 2 M-HCl.

In the static test condition, the specimens (two per group) were placed vertically in 10 ml polypropylene flasks as shown in Fig. 1 a. BG granules (Bioglass ® 45S5, 100 mg/ml, 300–500 μm fraction) were added to the bottom of the flask where appropriate, see Table 1 for details. Contact between the BG granules and the FRC1 or FRC2 specimen was ensured. Thereafter 7 ml of SBF was added. However, the small FRC3 specimens were incubated with 5 ml of SBF in horizontal orientation. The specimens were kept at 37 °C for either 3 and/or 7 d in a shaking incubator (Stuart S1500) using the speed 120 rpm. The solution flow was within the laminar range but induced some movement of glass particulates in the solution. The pH of the mixed SBF solutions was determined at 37 °C (pH meter: MP225, Mettler Toledo) at several time points during the immersion.

Table 1
Specimen groups used in each static and dynamic immersion test. Immersion times were 3 and 7 d in the static test and 7, 14 and 28 d in the dynamic test. N : Number of tests of each immersion time. In the dynamic test, two specimens were immersed in parallel.
Test Group name N Specimen SBF
Type Size (mm 3 ) Volume (ml) Immersion with BG granules Flow rate (ml/min)
Static SBF control 2/2 None n/a 7 No n/a
Static FRC1 control 0/2 FRC1 9 × 59 × 2.9 7 No n/a
Static FRC2 control 0/2 FRC2 9 × 60 × 1 7 No n/a
Static BG 1/1 None n/a 7 Yes n/a
Static FRC1+BG 2/2 FRC1 9 × 59 × 2.9 7 Yes n/a
Static FRC2+BG 2/5 FRC2 9 × 60 × 1 7 Yes n/a
Static FRC3 1/1 FRC3 10 × 10 × 4 5 Yes a n/a
Dynamic FRC3 2/2/1 FRC3 9.3 × 30 × 4 22 (circulated) Yes a 13

a BG granules were part of the FRC3 specimens.

In the dynamic set-up, 22 ml SBF was circulated through larger FRC3 within a reactor. This closed system, shown in Fig. 1 b, consisted of a peristaltic pump (type Isomatec Reglo Quick, IDEX Health & Science GmbH, Wertheim, Germany), a water bath to keep the circulating solution at 37 °C and two parallel specimen chambers made of reinforced PE with an inner diameter ( d I ) of 9.4 mm. The flow rate was determined as 13 ml/min at the main connector tubing (Tygon ST ® 3603, Saint-Gobain Performance Plastics, US) with d I = 3.2 mm and was estimated as 3 ml/min in each of the specimen chambers. Both ends of the FRC3 were covered with filter paper fixed by PE holders to homogenize the flow conditions. Prior to the test, the FRC3 were pre-immersed in SBF for 10 min in vacuum to fill the porous structure with liquid. Thereafter, the specimen chambers were connected to the pump and the reactor (including the specimens) was cleaned by pumping in ethanol, H 2 O (twice) and finally SBF. The total pre-immersion time was 20 min. The FRC3 specimens were kept in the flow of SBF for 7 d, 14 d or 28 d with the exchange of the solution after every 7 d. The pH of the SBF was detected before and after the immersion.

After the removal of incubated specimens, in static and dynamic tests, 0.5 ml samples of SBF were taken for the analysis of ionic concentrations. The ionic concentrations of Ca, P, K, Mg, Na, P, and Si were determined with an inductively coupled plasma-optical emission spectrometer (ICP-OES, Optima 5300 DV, Perkin Elmer). The removed FRC specimens were washed in H 2 O (twice), followed by ethanol (once) and then dried overnight at 60 °C in an oven. The weight of the specimens in dry state was measured before and after the immersion. The dried specimens were stored in a desiccator prior to characterization by scanning electron microscopy (SEM).

Characterization of precipitation layers

Characterization of the external and cross-sectional surfaces of the specimens were performed by SEM (Gemini 1530, LEO Oberkochen/Carl Zeiss, Germany) coupled with energy-dispersive X-ray spectroscopy (EDX, UltraDry Silicon Drift Detector, Thermo Scientific, Madison, WI, US). For the examination in cross-section, the specimens were embedded in epoxy resin (EpoFix ® , Struers A/S) filled with carbon active p.a. (Merck, Darmstadt, Germany) to enhance contrast against the BisGMA/TEGDMA resin. After polymerization and setting of the embedding resin, the specimens were sectioned along the main axis and then polished with SiC paper up to 4000 grit. The SEM analysis of the specimens from the static set-up was performed at various positions along the specimens (at 0, 3, 8, 11, 22, 33, 44, 58 mm distance from the bottom end). Additionally, the reaction layer formation on the BG granules was examined on their surfaces and in cross-section. Cross-sections of the FRC3 from the dynamic test were analyzed at 0, 15 and 30 mm distance from the solution inlet surface of the specimen.

Statistical analysis

Statistical analysis was performed using the SPSS (Statistical Package of Social Science, SPSS Inc., Chicago, USA) software. The equality of variances was analyzed by Levene test and where appropriate the equality of means test with Independent-samples T -test. Data which was not following normal distribution was analyzed via Kruskal–Wallis Test and Mann–Whitney U -test. The level of statistical significance was considered to be 0.05.

Materials and methods

Experimental arrangements

The capability of bioactive glass granules to induce CaP formation on the composite surfaces during immersion in SBF was studied by using two different experimental conditions, static and dynamic. In the static tests, FRC specimens were incubated in SBF in the presence and in the absence of separate BG granules. Incubated specimens included (1) FRC1 : control polished resin specimens without reinforcement, (2) FRC2 : porous FRC, and (3) FRC3 : laminate FRC structures. Specimen types FRC1 and FRC2 were included to study the influence of the distance between the composite surfaces and BG granules on the in vitro behavior. FRC3 was a surrogate of a cranial implant. In the dynamic test, a flow of SBF was passed through the FRC3 specimens. Details of experimental set-ups are shown in Fig. 1 .

Fig. 1
Scheme of static (a) and dynamic (b) immersion test set-up: left: pump system, right: Cross section of FRC3 specimen in tube reactor.

FRC2 was made of a randomly oriented chopped silanized E-glass veil impregnated within the BisGMA/TEGDMA polymer matrix. This composite was used to create multi-component laminates ( FRC3 ) which had two layers of BG granules entrapped between three layers of porous FRC. The BG granules were retained by the polymer, but not embedded into it.

Preparation of the specimens

The FRC1 specimens were prepared by pouring resin mixture consisting of BisGMA (50 wt%), TEGDMA (50 wt%), camphorquinone (0.7 wt%) and DMAEMA (0.7 wt%) into putty molds and then polymerized by light curing. Subsequently, the specimens were ground (LaboPol-21 Grinding Machine, 300 rpm, Struers A/S, Rødovre, Denmark) to the final dimension of 9 mm × 59 mm × 2.9 mm (each ±0.2 mm) and the surfaces were polished with grit 4000 SiC paper. The photo-initiated polymerization of the specimens was carried out in three stages to ensure minimal content of residual monomers. First, the specimens were pre-cured by a dental hand cure device (Elipar ® S10 LED curing light, 3M/ESPE, Seefeld, Germany) with an exposure time of 120 s (intensity: 1200 mW/cm 2 , wavelength: 430–480 nm). Second, the specimens were post-cured in a vacuum light oven (Visio Beta vario, 3M/ESPE, Seefeld, Germany) for 15 min in ambient temperature and finally in a light oven (Targis Power, Ivoclar Vivadent AG, Schaan, Lichtenstein) for 25 min at a temperature of 95 °C.

The FRC2 specimens were prepared from 1 mm thick E-glass fiber sheets (Ahlstrom Glassfiber Oy, Kotka, Finland). The randomly oriented fibers in the sheets had a diameter of 10 μm and a length of 8 mm, and were bound together by an acrylic binder with a methacrylate silane coupling agent. Pieces of 9 mm × 60 mm in size were cut from these sheets and scaled. The cut pieces were then impregnated with 1–1.5 ml of the BisGMA/TEGDMA resin mixture, and were left for 48 h in the dark. After the impregnation, the excessive resin was removed by compressing and the specimens were then light cured as described above. These procedures ensured that 75–81 wt% resin remained in the FRC2 specimen structure, which was expected to result in a total porosity of 39–49 vol% with more than 90% of functional (open) porosity . The average pore size was larger than 100 μm. The reinforcing E-glass fibers in the specimens were covered by polymer resin.

The FRC3 specimens were prepared by assembling two pieces of FRC2 with fiber bundles (everStick ® Ortho, Ø 0.75 mm, StickTech Ltd., Turku, Finland) serving as spacers ( Fig. 2 ) between the layers to allow the incorporation of the BG granules. Each of the two BG layers contained 45S5 Bioglass ® granules of the 100–250 μm fraction (composition in wt%: SiO 2 : 45, Na 2 O: 24.5, CaO: 24.5, P 2 O 5 : 6). The two sides of the FRC2 pieces along the spacers were sealed with additional resin prior to final light curing ( Fig. 2 ). Smaller specimens (10 mm × 10 mm × 4 mm), containing 60 ± 5 mg of BG per layer were used in the static test. Larger specimens (9.3 mm × 30 mm × 4 mm), comprising 200 ± 10 mg of BG per layer were applied in the dynamic test. The BG content of both smaller and larger FRC3 was around 38% of the total specimen weight. In previous tests, the packed arrangement of the granules in between the FRC2 layers enabled wetting throughout the composite structure .

Fig. 2
(a) Scheme of cranial implant design: A – intact bone, B – brain tissue, C – dura mater, D – screws for implant fixation, E – dense and F – porous FRC layers with Everstick ® space holders (gray) and BG (not marked) in between layers. (b) Scheme of FRC3 specimen.

After preparation the specimens were washed in ultrapure water (milliQ quality), then washed with ethanol, dried at ambient temperature and stored in a desiccator. Specimens for the static immersion test were additional cleaned by sonication for 5 min.

In vitro SBF immersion

The in vitro bioactivity of the specimens was studied using SBF prepared according to Kokubo’s formulation (c-SBF) . The SBF was prepared by dissolving reagent chemicals of NaCl, NaHCO 3 , KCl, K 2 HPO 4 ·3H 2 O, MgCl 2 ·6H 2 O, CaCl 2 ·2H 2 O, and Na 2 SO 4 into deionized water. The fluid was buffered at physiological pH 7.40 at 37 °C with (CH 2 OH) 3 CNH 2 and 2 M-HCl.

In the static test condition, the specimens (two per group) were placed vertically in 10 ml polypropylene flasks as shown in Fig. 1 a. BG granules (Bioglass ® 45S5, 100 mg/ml, 300–500 μm fraction) were added to the bottom of the flask where appropriate, see Table 1 for details. Contact between the BG granules and the FRC1 or FRC2 specimen was ensured. Thereafter 7 ml of SBF was added. However, the small FRC3 specimens were incubated with 5 ml of SBF in horizontal orientation. The specimens were kept at 37 °C for either 3 and/or 7 d in a shaking incubator (Stuart S1500) using the speed 120 rpm. The solution flow was within the laminar range but induced some movement of glass particulates in the solution. The pH of the mixed SBF solutions was determined at 37 °C (pH meter: MP225, Mettler Toledo) at several time points during the immersion.

Table 1
Specimen groups used in each static and dynamic immersion test. Immersion times were 3 and 7 d in the static test and 7, 14 and 28 d in the dynamic test. N : Number of tests of each immersion time. In the dynamic test, two specimens were immersed in parallel.
Test Group name N Specimen SBF
Type Size (mm 3 ) Volume (ml) Immersion with BG granules Flow rate (ml/min)
Static SBF control 2/2 None n/a 7 No n/a
Static FRC1 control 0/2 FRC1 9 × 59 × 2.9 7 No n/a
Static FRC2 control 0/2 FRC2 9 × 60 × 1 7 No n/a
Static BG 1/1 None n/a 7 Yes n/a
Static FRC1+BG 2/2 FRC1 9 × 59 × 2.9 7 Yes n/a
Static FRC2+BG 2/5 FRC2 9 × 60 × 1 7 Yes n/a
Static FRC3 1/1 FRC3 10 × 10 × 4 5 Yes a n/a
Dynamic FRC3 2/2/1 FRC3 9.3 × 30 × 4 22 (circulated) Yes a 13
Only gold members can continue reading. Log In or Register to continue

Nov 28, 2017 | Posted by in Dental Materials | Comments Off on Multi-layer porous fiber-reinforced composites for implants: In vitrocalcium phosphate formation in the presence of bioactive glass
Premium Wordpress Themes by UFO Themes