EVALUATION OF AN IBAD THIN-FILM PROCESS AS AN ALTENATIVE METHOD FOR SURFACE INCORPORATION OF BIOCERAMICS ON DENTAL IMPLANTS. A STUDY IN DOGS

Thin-film bioceramic coatings are potential alternatives to overcome the limitations provided by other commercially available coating techniques like PSHA, where variable bioceramic dissolution added to a metalloceramic weak link are process- inherent. The purpose of this investigation was to determine the overall and site specific (to 0.5 mm from implant surface) levels of osseoactivity around a thin-film (IBAD processed) coated titanium alloy implant versus a non surface modified (sand-blasted/acid etched) titanium alloy implant in a canine model. The surgical model comprised the proximal tibiae epiphyses with four implants placed in each limb remaining for 2 and 4 weeks in-vivo. 10 mg/Kg oxytetracycline was administered 48 hours prior to euthanization. The limbs were retrieved by sharp dissection, reduced to blocks, and subsequently nondecalcified processed for fluorescent microscopy. Micrographs (20x mag) were acquired around the implant perimeter and merged for overall biological response evaluations, and four micrographs (40x mag. subdivided in rectangles) were acquired along one of the implant sides for tetracycline labeled area fraction quantification. The results showed biocompatible and osseoconductive properties for the thin-film coated and uncoated titanium alloy implants. Tetracycline labeled area fraction analyses showed that the thin-film coated implants presented significantly higher overall and site specific osseoactivity levels at 2 and 4 weeks. The site specific osseoactivity values were significantly higher compared to overall values for control and thin-film coated implants at both times in-vivo. According to the results obtained in this study, thin-film coated implants enhanced biological response at the early implantation times evaluated.


INTRODUCTION AND BACKGROUND
The success of dental implants requires their anchorage in bone in order to withstand functional loading.This idea is accepted since both archaeology and histology records provide evidence of dental implants endosseous integration 5 .
Significant evolution on both surgical and restorative aspects of dental implantology have occurred since the term osseointegration 2 (direct bone apposition at the surface of either titanium or titanium alloys) was defined by a Swedish research group in the late seventies 1 .This term has been constantly redefined and by no means represents the complexity of the phenomena occurring at the bone-biomaterial interface.
While predictable outcomes have been reported since endosseous' implants early days following the classical twostage technique protocol, a desire for treatment time decrease while maintaining success rates reported above 90% [2][3] has been demonstrated by practitioners and patients.
Many attempts have been made on the manufacturing processes of dental implants in order to improve biological response of the host to materials.For this purpose, surface engineering methods have been under constant investigation, once it was known that some surface modifications notably changed the in-vivo performance of biomaterials 7,21,28 .
Several engineering processes have been used to modify the surface of dental implants in an attempt to increase bone wound healing kinetics and decrease treatment time frames 20,22 .Among popular surface modifications are the incorporation of calcium-and phosphate-based bioceramics to the surface of commercially pure titanium and titanium alloys in the form of apatites 29 or phases of other stoichiometry (calcium to phosphate ratios) 20 .The elemental components of these phases are found in the composition of natural bone, leading to a rationale for employment of biomaterials synthetically manufactured to resemble these compositions as implant materials 20,22,29 .
Plasma Spraying of Hydroxyapatite (PSHA) is by far the most commonly used coating technique for bioceramic incorporation on dental implants to the present day due to its processing versatility and simplicity, where virtually all implant bulk designs may receive a continuous coating layer on its surface [19][20]22 . Itsmanufacturing process has been thoroughly described 19 , and PSHA coatings have been shown to elicit earlier biological responses around implants [8][9]11,[17][18] .Limitations concerning PSHA processed bioceramic coatings are the variable dissolution rates due to the inherent multiphase microstructure obtained through this process 15,19- 20,25,30 , added to the presence of a metalloceramic weak link between the bulk metallic substrate and bioceramic coating, which relies on mechanical interlocking for its integrity and maintenance 20,22 .
In an attempt to overcome the limitations of the PSHA process while still benefiting from the increased osseoconductive properties provided by bioceramic coatings, thin-films of highly controlled microstructures and thicknesses have been engineered on the surface of dental implants.These thin-films may be deposited by a variety of techniques including sol-gel, Pulsed Laser Deposition (PLD), Ion Beam Assisted Deposition (IBAD), and others 20,22 .A potential advantage of thin-film processes is the tailored bioceramic dissolution as a function of time in-vivo, providing implant surface exposure as implantation time elapses, enabling direct bone contact to the implant surface.This direct bone contact to the metallic substrate favors the bone-biomaterial interface mechanical competence by avoiding metalloceramic weak links between the metallic substrate and coating, as found on PSHA coated implants 20,22 .
Due to the dynamic modeling/remodeling nature of bone during wound healing and homeostasis, there is a need for specific labeling tools for hard tissue kinetics' histomorphometric assessment.A bone tissue marker is defined as any identifiable feature, naturally occurring or artificially induced, which permits the location of a given bone surface in anatomical space at a known moment in time 26 .The use of tetracycline (TC) as a tissue marker was introduced 23- 24 , reviewed 4,13 and has been recently applied 6.16 due to its fluorescent properties, which allow for determination of osteoblastic activity (osseoactivity) in nondecalcified specimens.TC use has been extensive as a bone research tool regarding location and kinetics of bone formation and growth.This methodology has been utilized to evaluate overall and site specific bone activity levels around dental implants in various in-vivo models 10,14,27 , including dogs and humans.These studies 10,14,27 indicated that the modeling/remodeling kinetics at regions adjacent to the implant surface (to approximately 1 mm from surface) may have significant differences compared to regions away from the surface at various times in-vivo, and it has been hypothesized 15 that short and long term stability of dental implants are related to this region of increased bone activity.
The purpose of this investigation was to determine the overall and site specific (to 0.5 mm from implant surface) levels of osseoactivity around a thin-film Ca-and P-based bioceramic coated titanium alloy (Ti-6Al-4V) implant manufactured by the IBAD process versus a non surface modified (sand-blasted/acid etched) titanium alloy implant by means of stereological techniques 12 (quantitative microscopy) in a laboratory dog model.

Materials
The as-processed, sterilized, and packaged sand-blasted/ acid-etched titanium alloy and thin-film coated titanium alloy implant rods were provided by the manufacturer (Bicon, Inc.Boston, MA-USA).These were 10 mm in length by 4 mm in diameter.The number of devices was 32 and included an experimental (thin-film bioceramic coated, n=16) and a control group (sand-blasted/acid-etched, n=16).No detail regarding surface topography and chemistry was provided by the manufacturer.

Surgical Model and Clinical Aspects
The surgical model comprised 4 mid-size class A adult (closed bone growth plates) mongrel dogs in good health.The dogs followed a 2-week housing period before the first surgical procedure and 4 weeks post-operatively.The project was conducted after IRB approval in an AALAC approved facility.
The surgical site was the proximal tibiae epiphyses, with four implants placed in each limb.Each dog provided a 2-and 4-week comparison between experimental and control surfaces per four-implant location through sequenced surgical procedures.The left limb was used for the 4-week evaluation and the right limb for the 2-week evaluation.The surgeries were conducted under full anesthesia following sterile methodologies.

Surgical Site Preparation and Implantation
The proximal tibiae were exposed subperiostally and 4 equi-spaced holes were drilled through sequential burs (under external saline irrigation).The implants were then inserted into the trabecular mid-region with its top in contact with the tibiae proximal cortical plate.A polymeric cover screw was threaded into the implant top and standard layered procedures were employed for soft tissue closure.Forty-eight hours prior to euthanization, 10 mg/kg oxytetracycline was administered subcutaneously to provide fluorescent labeling for histomorphometric analyses (single label).

Specimen Preparation
At necropsy, the proximal tibia was exposed by sharp dissection.The upper one half of the bone was removed and contact radiographed to confirm implant location and orientation.The limbs were reduced to blocks with the implant in its center, which were subsequently processed to thin sections approximately 20 µm in thickness with the metallic implant in place through standard procedures for optical light microscopy.

General Biocompatibility Evaluation
The nondecalcified specimens were placed under an optical microscope equipped with an ultra violet source at 20x magnification, and 12 micrographs were acquired around the implant perimeter.The micrographs were merged by a computer software (Adobe Photoshop, San Jose, CA-USA), and the implant perimeter and surrounding bone structure was obtained for analysis.Qualitative evaluation regarding biocompatibility was performed for the different groups at both evaluation times in-vivo.

Tetracycline Labeling Quantification
Quantification of the tetracycline labeled bone area fraction was performed by acquiring 4 micrographs (40x magnification) along one side of the implant (total implant length covered at this magnification).Each of the four micrographs was subdivided into rectangles (0.5 mm base and 2.5 mm height) comprising 0.5 mm steps from the implant surface (Figure 1), and a 9-point grid was randomly placed 6 times for each micrograph subdivision for point-count 12 stereologic inferences.This procedure implied 24 tetracycline labeled bone area fraction measurements for each micrograph and a total of 96 measurements per implant.
The overall quantification of tetracycline labeled bone area fraction was assessed by considering all measured quantities for the 4 micrographs and their respective subdivisions for statistical analysis.
Investigation of the tetracycline labeled bone area fraction at the site in close proximity to the implant surface (to 0.5 mm from implant surface) was performed by only considering measurements obtained from the subdivision adjacent to the implant surface for the 4 micrographs acquired for statistical analysis.

Statistical Analyses
The confidence interval (CI) for each parameter evaluated through the quantitative microscopy technique described above was calculated at the 95% level of significance through the following equations: CI= [mean value ± t (standard error)], standard error = [standard deviation/(n 1/ 2 )], where t= t value associated with the number of degrees of freedom and level of significance, and n= number of observations for the parameter under evaluation 12 .
FIGURE 1-Representation of a micrograph (at 40x original magnification) subdivision into 0.5 mm base by 2.5 mm height rectangles for quantification of tetracycline labeled bone area fraction biocompatible and have osseoconductive properties, as per direct bone-to-implant contact at regions of cortical (proximal and distal plates) and trabecular bone [1][2][3] .This phenomenon regarded as osseointegration 2 is a desirable feature when considering endosseous dental implants for functional load bearing applications [1][2][3]5 .
The absence of a thin-film bioceramic coating on the experimental groups' implants indicated that partial or total coating dissolution occurred as implantation time elapsed in-vivo.This feature supports opportunities for direct bone contact to the underlying metallic substrate of thin-film coated implants, avoiding the occurrence of weak links between the bioceramic coating and metallic substrate invivo, where mechanical failure is likely to occur [19][20]22 . Thepartial or total coating dissolution that occurred in-vivo was possibly related to the low thicknesses thin-film coated implants and/or its microstructural phase composition 15,30 .
The presence of fluorescent labels at regions in proximity and away from the implant surface in all specimens demonstrated the effectiveness of oxytetracycline as a tissue marker 4,6,13,16,[23][24]26 , enabling quantification of relative degrees of bone activity around the implants 10,14,27 at both times invivo.
The overall tetracycline labeled area fraction quantification revealed significantly higher values for the thin-film coated implants compared to control implants at 2 and 4 weeks in-vivo, demonstrating a significant effect 7,21,27 of the surface treatment on bone kinetics, as previously reported by different methodologies [8][9]11,[17][18] considering bioceramic coatings obtained through different processes.
The site specific (to 0.5 mm from implant surface) tetracycline labeled bone area fraction quantification showed the same qualitative trends within groups when compared to the overall tetracycline labeled area fraction quantification, but presented significant higher values for all groups at both times in-vivo (no CI overlaps between identical groups in Tables 1 and 2).These significantly higher values at the region adjacent to the implant surface indicate that the modeling/remodeling rates have higher values compared to regions away from the implant surface, and these values potentially decrease to physiologic levels as a function of distance from the implant surface 10,14,27 .
Alteration in bone kinetics found in both overall and site specific tetracycline labeled area fraction quantifications of bone around thin-film coated implants may be beneficial regarding both potential decreases in osseointegration time 14,20,22,29 (short term stability) and long term implant stability maintenance 14 .It is important to note that no relationship between increased bone activity and increased bone-to-implant contact, higher bone-biomaterial interface mechanical properties, or short-and long-term implant treatment success ratios have been presented in a concise manner to date, and both in-vitro, in-vivo, biomechanical, and controlled clinical research protocols are desirable for addressing these issues.Protocols involving the full physical and chemical characterization of the thin-film coating used throughout this study would also provide valuable insight on relating bone kinetics to coating characteristics, and may be subject of future research.

CONCLUSIONS
Bilateral proximal regions of dog tibia were utilized to study the effect of a thin-film coated (IBAD processed) on bone activity at times 2 and 4 weeks after implantation, and according to the stereological (quantitative microscopy) results obtained, it can be concluded that the Ca-and Pbased bioceramic thin-film (IBAD processed) coated implants were biocompatible, osseoconductive, and presented significantly higher overall and site specific (to 0.5 mm distance from implant surface) tetracycline labeled area fraction values on dog's bone.
Direct bone contact to the metallic substrate was achieved for thin-film coated implants while still benefiting from the surface modification's effect on bone kinetics at both times in-vivo, supporting the rationale for thin-film coatings as candidates to overcome limitations inherent to commercially available bioceramic coatings processes like PSHA.

AKNOWLEDGEMENTS
This project was partially funded by Bicon Dental Implants, Inc. Boston, MA.This study was the first of a series of histomorphometric, biomechanical, and physico/ chemical characterization investigations concerning IBAD

TABLE 2 -
Summary statistics for tetracycline labeled bone area fraction quantification at the region adjacent to the implant surface (to 0.5 mm) quantification for thin-film coated (experimental) and titanium alloy (control) at 2 and 4 weeks in-vivo a,b, and c-statistical group CI overlap at 95% level of significance.