Grafting in peri-implant bone defects by in-situ polymer deposition using a 3D pen – in vitro/ ex vivo study

Guided Bone Regeneration (GBR) aims to gain or maintain bone volume due to the use of barrier membranes that act for this purpose. This research aims at grafting polymeric filaments into preformed peri-implant bone defects in porcine condyles in vitro/ex vivo, stabilized and grafted with poly(lactic acid) (PLA) and poly(vinyl alcohol) (PVA) polymeric filaments, printed in-situ with a 3D printing pen. Nine porcine condyles received bone defects of 8 mm diameter and 7 mm depth, where occurred the installation of conical implants of 3.5x10 mm. After forming the bone gap region, above the apical bone anchorage, we divided the Poof Bodies (PB) according to the polymeric fill used: G.Control – without filling in the bone gap; G.PLA – with PLA scaffolds and G.PVA – with PVA scaffolds. In another step, the PVA and PLA 3D membranes were compared with the dense polytetrafluoroethylene membrane (PTFE-d). Subsequently, the SkyScan 1172 microtomograph (Bruker- μCT, Konti ch, Belgium) analyzed the PB. The analysis corresponding to the total porosity revealed no statistical difference between G.Control (70.44%), G.PLA (59.99%), and G.PVA (57.66%). The closed porosity showed a statistical difference between G.Control (75.509%) and G.PVA (189.19%) and between G.PVA and G.PLA (79,093%). This study demonstrated the possibility of the polymeric filaments of PVA and PLA to fill the bone defects created, revealing an intimate contact on the surface of the implants used. The data suggested a higher porosity of the PVA filament when applied to bone defects or membrane shape. polimérico utilizado: G. Control sin relleno en el gap óseo; G. PLA – con estructura de PLA y G.PVA – con estructura de PVA. En otra etapa, fueron comparadas las membranas de PVA y PLA 3D con la membrana de politetrafluoretileno denso (PTFE-d). analizados microtomógrafo (Bruker- análisis significativa significativa una


Introduction
Among the different matrix manufacturing techniques, bioprinting applied directly to an injured site during surgery stands out, called intraoperative bioprinting or in-situ or in-vivo bioprinting, allowing the reconstruction of different craniomaxillofacial features such as bones, skin, and composite tissues, in an associated or isolated way (Moncal et al., 2021).
This developing process allows bone remodeling through the association of osteoprogenitor cells during cell proliferation, differentiation, and matrix formation (Wang, et al., 2020).
The use of manual printers and 3D pens has been popularized in recent years due to their low cost and practicality.
Porous polyethylene does not reabsorb nor degrades and has the advantage of allowing internal vascular and soft tissue growth after one week and internal bone growth after three weeks. The porous polyethylene is characterized as nonantigenic, anti-allergic, non-absorbable, highly stable, easily fixated, and available in various formats for reconstruction (Maia, et al., 2010).
PLA is a linear aliphatic synthetic polymer that exhibits biodegradability, biological absorption, biocompatibility, mechanical tensile strength, modulus of elasticity, thermal stability, and low environmental impact. In addition, its melting point is 180 °C (Santana et al., 2018).
PVA is a biodegradable, water-soluble, non-toxic, hydrophilic synthetic polymer produced through the hydrolysis of vinyl acetate (Prasadh, et al., 2018;Basa et al., 2021), widely used in topical pharmaceutical, ophthalmic formulations and drug-loaded pharmaceutical microsponges. In solid pharmaceutical formulations, PVA can be used as tablet coating. This polymer is also the most widespread support in 3D printing (Basa et al., 2021).
Bone defects undoubtedly tend to be a challenge for several areas of dentistry, particularly in oral and maxillofacial surgery, periodontics, and implantology. Repair of these bone defects is challenging even when the correct Guided Bone Regeneration (GBR) technique is applied (Prado et al., 2006;Araujo, et al., 2022).
In parallel, osseointegration is the direct anchoring of a dental implant through the formation of bone tissue around the implant without the growth or development of fibrous tissue at the bone-implant interface, which ensures its stability (Mantovani Junior, 2006;Araujo et al., 2022). In this context, the different bone defects are challenging in the fixation of implants and may face partial or total fenestrations of the vestibular wall, leading to exposure of the threads of this implant after its installation (Consolaro, et al., 2010).
As biomaterials are developed, certain defects can be corrected, leading to bone volume gain. The properties of these biomaterials range from osteogenic to osteoinductive, releasing bone morphogenic proteins and allowing undifferentiated mesenchymal cells to differentiate into osteoblasts and chondroblasts, initiating the bone remodeling process (Okamoto, et al., 1973).
Thus, this study aims to analyze the grafting of polymeric filaments in peri-implant bone defects performed in porcine condyles in vitro/ex vivo, stabilized and grafted with PLA and PVA polymeric filaments, printed in-situ with a 3D printing pen.

Methodology
This study was submitted to the ethics committee on animal use of the Grande Rio University under the number 045/2021. This research was based on the quali-quanti method. Qualitative analyzes were based on the information and details provided through the 3D scanning images. Quantitative analyzes were based on reconstruction software and analysis of 3D tomographic images. Thus, the analyzes were completed facilitating the understanding of the results obtained (Pereira, et al., 2018).
In this qualitative-quantitative in vitro and ex-vivo research we used five dissected porcine mandibles had the condyle region sectioned with carborundum discs. Then, nine condyles received bone defects using a trephine drill with 8 mm diameter and 7 mm depth, at an angle of 90° in relation to the bone surface and at 500 RPM with a Driller motor (BLM350 -Driller).
We store the specimens at -2º C, remaining at room temperature until reaching 23º C during the tests.
Subsequently, we installed one conical implant of 3.5x10 mm (Singular, Parnamirim, RN) in each condyle, totaling nine PB, following the milling sequence with lance drills and 2.0 and 3.5 twist drills. Located in the central region of the defects, the implants had an apical intraosseous anchorage of 4 mm and a final installation torque from 10 to 20 N at 500 RPM.
After forming the bone gap region above the apical bone anchorage, the proof bodies (PB) were divided according to the polymeric filling used: group G.Control with no filling in the bone gap; Group G.PLA with PLA scaffolds, and Group G.PVA with PVA scaffolds. A 3D pen deposited the filaments in situ. We studied the groups in triplicate for each condition (Table 1 and Figure 1).

3D Pen
We used commercial filaments of PVA and PLA of 0.75 mm. A 3D pen extrudes the filaments at 200°C and at a speed of 30 mm/s (Saywe SMA-1 Plus). Filament deposition occurred directly in the bone gaps until the filling reached the cortical bone crest of the PB.

Proof Body Final installation torque
Group and scaffold used

Microtomographic Analysis
A SkyScan 1172 microtomography (µCt) (Bruker-μCT, Kontich, Belgium) analyzed all PB and membranes. The μCT had the following common parameters for image acquisition in all stages: voltage of 50 Kvp, source current of 800 μA, flatfield correction, Al 0.5 filter, pixel size of 18.99 μm, exposure time of 4000 ms, rotation step of 0.5 and Frame Averaging 3.
The NRecon program (SkyScan, Kontich, Belgium) reconstructed and processed the images obtained from μCT adjusting the parameters: ring artifacts reduction in 7, Beam-hardening correction in 46%, Gaussian kernel smoothing and defect pixel masking in 50%, and misalignment compensation.
Then, the DataViewer program (SkyScan, Kontich, Belgium) analyzed the reconstructed images for visualization and 2D evaluation of the coronal, transverse, and sagittal axes, using the Hounsfield Unit to evaluate the pixel intensity of the artifacts found in the PB.
The CTan program (SkyScan, Kontich, Belgium) also analyzed the images and delimited the regions of interest, proceeding to the thresholding and binarization of the images, adjusting the histogram to evidence the suggestive artifacts of voids and bubbles. Ultimately, the CtVox software (SkyScan, Kontich, Belgium) presented the 3D images for visual evaluation.
GraphPad Prisma 5.01 software (Graph Pad Software Inc) performed the statistical analysis of the data, with analysis of variance (One-Way ANOVA) and complementary Tukey post-test, with a significance level of 5% (p < 0.05).

PVA and PLA Membranes
In parallel, we extruded the PVA and PLA filaments on a glass surface of 3x3 cm, in a single layer of deposition, using the same extrusion parameters applied above. These PVA and PLA membranes were developed in triplicate and analyzed on µCt.
This step involves a membrane based on dense polytetrafluoroethylene (PTFE-d), Ti-250 cytoplast membrane (Implacil De Bortoli, Osteogenics Biomedical, São Paulo, Brazil) with titanium reinforcement as a control sample. This membrane is non-resorbable and used in GBR surgeries, with porosity less than 0.3 µm, characterized by having greater resistance to bacterial penetration, protecting the bone and the implant below this membrane (Maridati, et al., 2016).
For the microtomographic analysis of the PTFE membrane, we selected a quadrant composed exclusively of PTFE, without titanium reinforcement, considering the generation of artifacts created during the scanning of the metal structure. Here, the adequated parameters due to the low radiolucids of the PTFE are 50 Kvp voltage, source current of 800 μA, flat-field correction, no filter, pixel size of 18.99 μm, exposure time of 1700, rotation step of 0.3, and Frame Averaging 3.

Results
From the analysis performed at µCt, information corresponding to total porosity (Po ( (Fig. A.1), G.PLA (Fig. B.1), and G.PVA (Fig. C.1), with an indication of the axial section area (red line) binarized, in the CTan program, in the G.Control (Fig. A.2), G.PLA (Fig. B.2) and G.PVA (Fig. C.2) groups, demonstrating the grafting relationship performed by the polymers in the bone gaps. The contact relationship of the polymers with the surface of the implants analyzed in the 3D images through the CtVox program demonstrates the integrated surface of the implant in the G.Control group (Fig. A.3 and A.4) and the filling of the polymers with a surface coating of the implants in G.PLA (Fig. B.3 and B.4)  The analysis detected the density difference between the polymers. In the reconstruction of G.PVA, a less dense polymer, we notice intercalated areas of porosity. In contrast, the images of G.PLA, a polymer denser than PVA, revealed areas with horizontal extensions and a better distribution of the radiodense regions ( Figure 2). These data corroborated the statistical difference found for closed porosity in G.PVA and G.PLA groups. The average object surface (Obj.S mm 2 ) is lower in the PVA membrane (2,693 Obj.S mm 2 ) in relation to the PLA membrane (3,256 Obj.S mm 2 ), demonstrating a higher filling ratio in the PLA membrane due to the higher average, but this Research, Society andDevelopment, v. 11, n. 14, e301111436234, 2022 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v11i14.36234 8 data also suggests a higher porosity in the PVA membrane, justified by the lower filling of the volume in the area of interest.

PVA and PLA membranes
The PTFE membrane presented 6,468 Obj.S mm 2 , explaining the higher density of this membrane compared to PLA and PVA. After analyzing this set of data corroborated by the closed porosity, we can suggest a greater porosity of the PVA membrane.
The analysis of the 3D reconstructed membranes in CtVox and Ctan programs suggested the presence of numerous pores of different sizes, indicating greater intensity in the PVA membrane. As it is a manual 3D deposition, the spaces between the deposited filaments are noted. The binarized reconstructions in the Ctan also suggested a higher pore intensity in the PVA membranes than the PLA membranes (Figure 3). In contrast, PTFE membrane reconstruction did not show visible pores, suggesting a highly compact material (Figure 3).

Discussion
This study is based on using different regenerative techniques associated with biomaterials. The main objective of this technique is to obtain the vertical and horizontal bone volume in bone edges, aiming to minimize bone remodeling after tooth extraction for the installation of integrated bone implants (Costa et al., 2021;Sanz et al., 2019).
Different biodegradable and non-biodegradable materials are investigated in the application of GBR and should present biocompatibility, semipermeability, integration into host tissues, clinical maneuverability, and space maintenance capacity (Warrer, et al., 1993).
If bone defects are not reduced, they can interfere with or even make prosthetic rehabilitation impossible. During the installation of osseointegrated implants in regions with atrophy or severe bone defects, stability can be compromised, causing premature losses, besides aesthetic and functional defects. Therefore, inadequate bone volume for implant placement is a critical clinical problem, determining the need to insert bone grafts to ensure adequate bone volume and provide greater stability, leading to a better prognosis (Herford & Dean, 2011).
The present study adopted in-situ PVA and PLA filaments in the proof of concept. Still, the data suggested the complete filling of the bone defect using a 3D pen, which may represent a future development in the GBR bone grafting technique.
Based on the microarchitecture characteristic of the scaffold, the feasibility and level of precision of the manufacturing process can be evaluated (Ho & Hutmacher, 2006). The microarchitectures of the scaffolds also influence the mechanical strength and biological functionality, and pore size, porosity, and surface area/volume ratio can be evaluated through analysis (Ho & Hutmacher, 2006). The analysis of the membranes indicated a lower porosity in the PTFE membrane, followed by the PLA and PVA, as the one with greater porosity. The data are justified once the PTFE membrane is nonabsorbable, and this high porosity is unnecessary. Structurally, the membranes may present perforations to improve the conditions for bone neoformation. Rigid membranes, such as PTFE membranes, act as a substrate so isolated cells can fix and grow to form tissue (Costa et al., 2016).
Contrastingly, the PVA membrane with greater porosity indicates a possible absorption characteristic in the body. The absorbable membranes are advantageous as they do not require a second surgical time for removal, being degraded or incorporated by the body throughout the regeneration process (Rakhmatia, et al., 2013). In this context, absorbable membranes propose greater bone formation in relation to non-absorbable membranes (Costa et al., 2021).
The biological membrane materials must be physically capable of allowing adequate modeling of the graft, the implant, and the bone structure. A hardness is necessary so it does not deform after being adjusted in place nor move easily, thus preventing inflammation (Costa et al., 2016). The versatility of developing PVA or PLA membranes or scaffolds, directly at the surgical site, may encompass these different requirements in this type of biomaterial.
High porosity may represent a low tensile strength since stress causes the existing cracks to spread (De Oliveira, et al., 2007), however, porosity is desirable in scaffolds for biomaterial use in grafting procedures. The microtomographic analyses obtained information on total porosity, open porosity, closed porosity, and surface objects, identifying in the last two items desirable porosity specifications associated with the uniform distribution of polymers in PVA or PLA scaffolds, presenting better closed porosity and surface objects in the first filament.
The high porosity and surface area for volume are fundamental characteristics of uniform cellularity, enabling tissue fixation and neoformation. An ideal porosity would be 90%, allowing the diffusive transport of cells within the scaffolds (Ho & Hutmacher, 2006). Nevertheless, such a high porosity could compromise the mechanical properties of scaffolds (Ho & Hutmacher, 2006). In the presented data, the G.PVA group reached a closed porosity of approximately 90%, while the total porosity in G.PLA and G.PVA groups was close to 60%, suggesting the possibility of being biomaterials susceptible to cell neoformation.
Once the total porosity is based on the volume of all open and closed pores, of the VOI selected in each sample, the closed porosity data is divergent. However, the Po(cl)% analysis considers the total solids plus the volume of closed pores within the VOI, suggesting the filling of the greater peri-implant bone defect in the G.PVA group, followed by the G.PLA Previously, Ho and Hutmacher (2006) reported using µCT in scaffolding research, as they characterized the morphology in poly (L-lactide-co-DL-lactide) and poly-e-caprolactone scaffolds manufactured through extrusion deposition.
Thermoplastic semi-crystalline polymers are excellent candidates for developing tissue-engineered scaffolds as they are easy to process and present adjustable properties. In parallel, synthetic polymers can be an option for scaffolds applicability in hard tissues, presenting physical characteristics such as material stiffness, roughness, and topography that are desirable for cell neoformation (Calore et al., 2021). Calore et al. (2021) indicated that in an osteogenic environment, human mesenchymal stromal cells responded more to surface roughness than to surface stiffness of different scaffolds.
The applicability of in-situ printed PLA and PVA filaments on bone defects generating the stability of osseointegrated implants has not yet been explored. Therefore, the data exposed in this study bring an innovative possibility of polymeric biomaterials in implant dentistry to minimize early loss of osseointegrated implants and its usage in the GBR technique.

Conclusion
This study demonstrated the possibility of using a 3D pen in situ as an instrument in GBR surgery, completely filling peri-implant bone defects with the implant already installed. In addition, the polymeric filaments of PVA and PLA demonstrated the ability to fill the bone defects created, revealing an intimate contact on the surface of the implants.
The data suggested a higher porosity of the PVA filament when applied to bone defects or membrane shape.
Once this is a proof-of-concept study, additional data are still necessary until the feasibility of applying the materials presented can be concluded.