Recently there has been a burgeoning increase in

Transcription

1 Build Angle: Does It Influence the Accuracy of 3D-Printed Dental Restorations Using Digital Light-Processing Technology? Reham B. Osman, BDS, MSc, PhD 1 /Nawal Alharbi, BDS, MSc 2 /Daniel Wismeijer, DDS, PhD 3 Purpose: The aim of this study was to evaluate the effect of the build orientation/build angle on the dimensional accuracy of full-coverage dental restorations manufactured using digital light-processing technology (DLP-AM). Materials and Methods: A full dental crown was digitally designed and 3D-printed using DLP-AM. Nine build angles were used: 90, 120, 135, 150, 180, 210, 225, 240, and 270 degrees. The specimens were digitally scanned using a highresolution optical surface scanner (IScan D104i, Imetric). Dimensional accuracy was evaluated using the digital subtraction technique. The 3D digital files of the scanned printed crowns (test ) were exported in standard tessellation language (STL) format and superimposed on the STL file of the designed crown [reference ] using Geomagic Studio 2014 (3D Systems). The root mean square estimate (RMSE) values were evaluated, and the deviation patterns on the color maps were further assessed. Results: The build angle influenced the dimensional accuracy of 3D-printed restorations. The lowest RMSE was recorded for the 135-degree and 210-degree build angles. However, the overall deviation pattern on the color map was more favorable with the 135-degree build angle in contrast with the 210-degree build angle where the deviation was observed around the critical marginal area. Conclusions: Within the limitations of this study, the recommended build angle using the current DLP system was 135 degrees. Among the selected build angles, it offers the highest dimensional accuracy and the most favorable deviation pattern. It also offers a self-supporting crown geometry throughout the building process. Int J Prosthodont 2017;30: doi: /ijp.5117 Recently there has been a burgeoning increase in the implementation of digital tools in the field of restorative dentistry. Digital fabrication technologies involve computer-aided design/computer-assisted manufacture (CAD/CAM) techniques through either subtractive (milling) or additive (3D printing) manufacturing (AM). 1 Lecturer, Removable Prosthodontics Department, Faculty of Dentistry, Cairo University, Giza, Egypt; Research Associate, Department of Oral Implantology and Prosthetic Dentistry, Academic Centre for Dentistry Amsterdam (ACTA), Universiteit van Amsterdam and Vrije Universiteit, Amsterdam, The Netherlands. 2 PhD Candidate, Oral Implantology and Prosthetic Dentistry, Academic Centre for Dentistry Amsterdam (ACTA), Universiteit van Amsterdam and Vrije Universiteit, Amsterdam, The Netherlands; Lecturer, Prosthetic Dental Science Department, College of Dentistry, King Saud University. Riyadh, Saudi Arabia. 3 Professor of Oral Implantology and Prosthetic Dentistry and Head of the Department of Oral Implantology and Prosthetic Dentistry, Academic Centre for Dentistry Amsterdam (ACTA), Universiteit van Amsterdam and Vrije Universiteit, Amsterdam, The Netherlands. Correspondence to: Dr Nawal Murshed Alharbi, Department of Oral Implantology and Prosthetic Dentistry, Academic Centre for Dentistry Amsterdam (ACTA), Gustav Mahlerlaan 3004, 1081 LA Amsterdam, The Netherlands. nawalmurshed@gmail.com 2017 by Quintessence Publishing Co Inc. AM has a unique advantage over conventional milling production methods: it produces practically no waste material, there is no restriction in geometric shape of the products, and tolerance of milled parts is no longer an issue. 1,2 This allows AM technologies to be a key component in the mass production of parts with special geometric requirements. 2 The fabrication of fixed dental crown and bridge restorations with their unique buccal, lingual, mesial, and distal contours and sophisticated occlusal outlines is an example. Among the various AM techniques, digital light processing (DLP) is gaining increased popularity in the production of dental parts. 3 5 In a DLP build process, a highly complex structure is fabricated on a layer-bylayer basis directly from 3D data, whereby consecutive liquid photoactivated monomer layers are exposed to UV light and cured based on the final shape of the required product. The DLP process involves a digital micromirror device (DMD) that is used to dynamically define a mask image that is projected on the resin surface. 6 9 DMDs consist of hundreds of thousands of individually moving micromirrors that control the reflection path of light. Each pixel of the image corresponds to an individual micromirror, the orientation of which can be switched among several degrees based on the geometry of the part to be printed. 9, The International Journal of Prosthodontics

2 Osman et al Fig 1 (right) STL of the designed restoration (the occlusal and fitting surface). Antirotation feature (a); flat occlusal surface (b); tapered axial wall with 6-degree angle (c). Fig 2 (below) 3D-printed specimens using different build angles. a b c Buccal Lingual Build angle (degrees) Direct capture of the geometric features of the object during the build process is limited by the specific machine design and the fact that photoactivated monomers often do not change optical characteristics in the polymerization process. 9 Customization of build angle/orientation during the build process is one of the factors that can improve the geometrical accuracy as well as the structural properties of the final 3D-printed part by using the full capabilities of the light source. 11 In the same context, a study by Alharbi et al 12 evaluated influence of build angle and support configuration (thick vs thin) on the dimensional accuracy of full-coverage dental restorations printed using stereolithography (SLA) technology. The findings of the study revealed that both factors influence the dimensional accuracy of the printed parts. The optimal build angle should provide self-supported geometry and thus require minimal support structure during the build process. 12 However, similar studies in the dental literature investigating the influence of various technical factors involved in DLP technique on the accuracy of printed dental restorations are lacking. Therefore, the aim of the current study was to evaluate the effect of the build orientation/build angle on the dimensional accuracy of full-coverage dental restorations manufactured using DLP-AM technology. Printing of Specimens Materials and Methods A virtual full-coverage crown restoration was designed on a virtual die simulating an actual tooth preparation using the 3-Shape Dental System TM CAD solution. The crown design is shown in Fig 1. Crown design and preparation have been previously described in detail. 12 The digital data of the crown restoration were exported in standard tessellation language (STL) format, which was used for the 3D printing of dental crowns using DLP technology. The crowns were fabricated using nine different build angles; 90, 120, 135, 150, 180, 210, 225, 240, and 270 degrees (Fig 2). Initially, the crown was rotated 90 degrees where the support structure was located on the lingual surface, and this was considered the start build angle. Subsequently, the object was rotated at nine different build angles until the support was located on the buccal surface of the crown. The support structure was generated semiautomatically; the software automatically configured any surface that needed support within each build angle. Any support structure automatically generated at the fitting surface of the crown was manually removed. The location of support was generated based on the Volume 30, Number 2,

3 Influence of Build Angle on Accuracy of 3D-Printed Dental Restorations Build angle (degrees) Fig 3 Red areas represent crown surfaces where support structure was generated in different build angles. Z-axis Build platform Printed object Resin vat Y-axis X-axis 30 LED light source Motor Lens Build platform DMD Fig 4 Red arrows represent surfaces with < 30 degrees that must be supported. Fig 5 DLP printing technique. Table 1 Material Properties of NextDent C&B Property Model Value Brookfield viscosity at 23 degrees Pa ASTM D2162 Flexural strength MPa ISO 10477:2003 Flexural modulus MPa ISO 10477:2003 Charpy impact strength unnotched kj/mm 2 ISO 179: 2010 Water sorption < 30 μg/mm 2 ISO 10477:2004 Water solubility < 5 μg/mm 2 ISO 10477:2004 Hardness shore (D) ISO 868:2003 angle between the building platform and the long axis of the object; any horizontal surface forming an angle of < 30 degrees with the long axis was supported as shown in Figs 3 and 4. The crowns were printed using RapidShape D30 DLP printer (Rapid Shape) with each crown fabricated at the center of the build platform. The DLP printer uses an LED light source, a DMD device/chip, a lens, a resin vat, and a build platform moving along the z-axis (Fig 5). The DMD device is composed of several micromirrors that dynamically reflect the light either toward (on) or away from the vat (off) to create light or dark pixels, respectively. The LED light used had a wavelength of 405 nm (narrow spectrum wavelength of nm) and 10.0 W/m 2 energy output. The build platform size was mm and the resolution was 1,080 1,920 pixels. The pixel size was mm, and the layer thickness was 50 µm. The x-y accuracy of this DLP system as reported by the manufacturer is ± 29 µm. 13 Fig 5 presents a schematic drawing describing the mechanism of the DLP process. All crowns were printed using NextDent C&B material (Shade A2, NextDent) with similar printer settings. The material properties as reported by the manufacturer are presented in Table 1. Specimen Preparation for Analysis After printing, all specimens were cleaned with 96% alcohol for 5 minutes and postcured for 30 minutes using ultraviolet curing unit LC 3DPrint Box (NextDent) following the manufacturer s instructions. 14 The specimens were stored in a lightproof box and tested within 10 days of fabrication. 184 The International Journal of Prosthodontics

4 Osman et al RMSE (mm) Build angle (degrees) Fig 6 RMSE values in all build angles. Table 2 Deviations of all Build Angles Build angle (degrees) RMSE (mm) Maximum deviation (+) (mm) Maximum deviation ( ) (mm) Average deviation (+) (mm) Average deviation ( ) (mm) SD (mm) Analysis All specimens were digitally scanned using a high-resolution optical surface scanner (IScan D104i, Imetric). Prior to scanning, the scanner was calibrated according to the manufacturer s instructions. The specimens were examined for any manufacturing defects and sprayed with a thin layer of antireflective powder (3D Scan Spray, Helling). The accuracy of the printed crowns was then evaluated using a digital subtraction technique. The STL files of the scanned printed crowns (test ) and that of the designed crown (reference ) were exported in Geomagic Studio (3D Systems). The exported files were aligned to have the same coordinate systems. The alignment was further refined using automatic best-fit alignment based on the closest point algorithm (ICP) system. Prior to the alignment process, the support structure was virtually removed to eliminate any potential error during the procedure. The accuracy was then evaluated through root mean square estimate values (RMSE) and deviation patterns on color maps. All specimens were scanned and analyzed by one trained operator. Statistical Analysis Intraobserver reliability was assessed using intraclass correlation coefficient. To assess the observer bias, measurements of all build angles were taken twice by one observer with an interval of 15 days between the two measurements. Furthermore, possible errors from optical scanning were excluded by assessing the repeatability of the measurements six times in one randomly selected build angle. Results The intraclass correlation coefficient was 0.99, which reflects high reliability of the measurements and eliminates any possible observer bias. The repeatability of the measurements was high as is evident from the RMSE values of the six repeated measurements of the randomly selected build angle (120 degrees). RMSE values of all the build angles are shown in Table 2 and Fig 6. Maximum deviation was observed with the 90-degree build angle followed by 180- and 270-degree angles where the RMSE values were mm, mm, and mm, respectively. The Volume 30, Number 2,

5 Influence of Build Angle on Accuracy of 3D-Printed Dental Restorations Buccal Lingual Fig 7 Deviation pattern on the color-map. Deviation within mm marked in green, mm marked in red, mm marked in dark blue. Positive deviation of fitting surface at close vicinity to support structure and on opposing wall to support Attachment of support structure Buccal Buccal Buccal Lingual Lingual Lingual 90 degrees 180 degrees Attachment of support structure Support structure located on the occlusal surface 270 degrees Fig 8 Fitting surface view of build angles 90, 180, and 270 degrees. Test Reference Test Reference An average deviation of 100 µm decrease of the crown margins relative to the reference crown was observed on the color maps for all the build angles (Fig 7). Furthermore, an increase in the positive deviation was observed on the fitting surfaces of printed crowns in areas in close vicinity to the support structure and on the opposing wall as denoted by the red areas on the color maps as observed with build angles 90, 180, and 270 degrees (Fig 8). In other words, the wall of the printed crowns in the previously described areas will be thicker in comparison to the reference, thus the expected tighter fit (Fig 9). Discussion Fig 9 Greater wall thickness of printed restoration relative to the reference. minimum RMSE values were recorded for 135- and 210-degree build angles (0.049 mm). However, scrutiny of the deviation patterns of both angles on the color map reveals that the 135-degree build angle exhibits a more favorable deviation pattern that is located further away from the critical marginal area. The aim of this study was to evaluate the effect of the support/build angle on the dimensional accuracy of full-coverage dental restorations fabricated using the DLP-AM technique. Nine full-coverage dental crowns were 3D printed using nine different build angles. To minimize and/or eliminate all potential handling and processing errors, several measures were taken. Each crown was printed individually in the center of the build platform. One trained practitioner performed the fabrication and postcuring process. The restoration was designed on a customized virtual die to eliminate any error 186 The International Journal of Prosthodontics

6 Osman et al encountered from digital impression of a real tooth preparation. 15,16 Further, the virtual die allowed for a simple flat design of fitting surfaces of the printed crowns and thus enhanced the digital superimposition process. The antirotational feature in the fitting surface of the crown prevented any rotational movement during the scanning and analysis procedures. 12 The results were interpreted taking into consideration the RMSE values and the deviation pattern on the color maps. Maximum, minimum, and average deviation values were not considered when the accuracy was evaluated. The deviation values can be either positive or negative, which when averaged to provide the arithmetic mean can preclude any existing actual difference. 17 In this study, the minimum RMSE value was mm for both 135- and 210-degree build angles, whereas the maximum value was mm for the 90-degree build angle. This can be explained by the fact that angulation of the crown with 135- and 210-degree build angles offered the most self-supporting geometry in comparison to other build angles. This assumption can be confirmed by observing the increase and decrease pattern of the RMSE values related to the difference in supported and unsupported surfaces of the printed crowns. The highest RMSE values were observed with build angles 90, 180, and 270 degrees, where most of the fitting surfaces were unsupported, whereas lower values were observed for the angles in between with the increase in the selfsupported geometry of the printed crowns (Fig 6). Though the RMSE value for both 135- and 210-degree build angles was the same, the pattern of deviation on the color map was different. The deviation pattern in the 135-degree build angle was more favorable and located farther from the critical margin area. Further analysis of the color map revealed positive deviation on the fitting surfaces close to the support structure and on the opposing wall. This finding can be attributed to the upward movement of the building platform during the fabrication process and sagging of the material under its own weight. On the other hand, the positive deviation on the wall opposing the support structure can be explained by the curing pattern of the DLP technique. In the DLP system, the whole layer is illuminated at once, and those walls may be exposed to overcuring during processing of the subsequent layers. This precuring is expected considering the geometry of the fitting surface of the crown featuring a hollow cylinder. The recorded RMSE values for all the build angles other than 90, 180, and 270 degrees were smaller than the pixel size of the build platform. Therefore, the cause of deviation may be inherited from the nature of the DLP system. 9 A previous study that evaluated the geometric accuracy of the printed crown fabricated using SLA technique reported a maximum RMSE value of mm, which is still less than the minimal deviation reported in this study (0.049 mm). This can be attributed to the difference between the two manufacturing techniques. The precision of DLP printing is further influenced by the optical specifications integrated in the system; DMD device, lens quality, pixel size, and platform resolution. 7,9 Furthermore, these results are in line with various reports in the literature stating SLA technique to be one of the most accurate techniques amongst additive manufacturing technologies. 18,19 However, it would be wise to conclude that no single 3D-printing technique is superior to another, but that a properly selected technique fits the intended use. 5,20 The results of this study revealed that a build angle of 135 degrees offers the lowest deviation and the most favorable deviation pattern. However, the results of this study should be carefully interpreted when using different DLP systems, other printing materials, and different designs of the printed parts. Nevertheless, the selected build angle for any system must ensure high accuracy, self-supported geometry, minimal support structure, and minimal fabrication time. In this study, one aspect of DLP technology was explored. Factors to be evaluated in the future could include the depth of cure/light intensity and the position of the part within the build platform. Future developments in materials applied and in the different technical aspects of DLP systems, such as the quality of the lens and DMD device and the light source, are ongoing and mandatory. This will guarantee a leading place for DLP technology among other 3D-printing techniques in the dental field. Limitations of this study are acknowledged and include limited generalizability of the result due to different printers specifications and materials, but the general concept explaining the findings remains unchanged. Further drawbacks involve inherited errors associated with any digital subtraction procedure and potential scanning and slicing errors. However, the knowledge added to this field outweighs any potential limitation and presents a stepping stone for further studies in this field. Conclusions Within the limitations of this study, the recommended build angle using the current DLP system was 135 degrees. Among the selected build angles, it offers the highest dimensional accuracy and the most favorable deviation pattern. It also offers a self-supporting crown geometry throughout the building process. Volume 30, Number 2,

7 Influence of Build Angle on Accuracy of 3D-Printed Dental Restorations Acknowledgments The authors would like to thank Imetric 3D SA for providing the digital scanner used for the study. In addition, they would like to thank Koen Brongers from TTL Zutphen laboratory for his help in designing the restoration. This project is supported by a scholarship grant number 2/ from King Saud University, Riyadh, Kingdom of Saudi Arabia. The authors reported no conflicts of interest related to this study. References 1. Berman B. 3-D printing: The new industrial revolution. Business Horizons 2012;55: Azari A, Nikzad S. The evolution of rapid prototyping in dentistry: A review. Rapid Prototyp J 2009;15: Hoang LN, Thompson GA, Cho SH, Berzins DW, Ahn KW. Die spacer thickness reproduction for central incisor crown fabrication with combined computer-aided design and 3D printing technology: An in vitro study. J Prosthet Dent 2015;113: Sancho-Puchades M, Fehmer V, Hämmerle C, Sailer I. Advanced smile diagnostics using CAD/CAM mock-ups. Int J Esthet Dent 2015;10: van Noort R. The future of dental devices is digital. Dent Mater 2012;28: Wu GH, Hsu SH. Review: Polymeric-based 3D printing for tissue engineering. J Med Biol Eng 2015;35: Mitteramskogler G, Gmeiner R, Felzmann R, et al. Light curing strategies for lithography-based additive manufacturing of customized ceramics. Additive Manuf 2014;1 4: Zhou C, Chen Y, Yang Z, Khoshnevis B. Digital material fabrication using mask-image-projection-based stereolithography. Rapid Prototyp J 2013;19: Andersen UV, Pedersen DB, Hansen HN, Nielsen JS. In-process 3D geometry reconstruction of objects produced by direct light projection. Int J Adv Manuf Technol 2013;68: Lee MP, Cooper GJ, Hinkley T, Gibson GM, Padgett MJ, Cronin L. Development of a 3D printer using scanning projection stereo lithography. Sci Rep 2015;5: Oropallo W, Piegl LA. Ten challenges in 3D printing. Eng Comput 2016;32: Alharbi N, Osman RB, Wismeijer D. Factors influencing the dimensional accuracy of 3D-printed full-coverage dental restorations using stereolithography technology. Int J Prosthodont 2016:29: Welcome to the future of 3D dental manufacturing. Germany: Rapidshape, Accessed 19 December Instruction for use: NextDent C&B (crown and bridge). The Netherlands: NextDent, 2014: com/wp-content/uploads/2016/05/ifu-nextdent-cb- INCBIIa201601UK.pdf. Accessed 19 December Persson AS, Andersson M, Odén A, Sandborgh-Englund G. Computer aided analysis of digitized dental stone replicas by dental CAD/CAM technology. Dent Mater 2008;24: Jeon JH, Choi BY, Kim CM, Kim JH, Kim HY, Kim WC. Threedimensional evaluation of the repeatability of scanned conventional impressions of prepared teeth generated with white- and blue-light scanners. J Prosthet Dent 2015;114: Congalton RG, Green K. Assessing the Accuracy of Remotely Sensed Data: Principles and Practices, ed 2. Boca Raton, FL: CRC, Melchels FP, Feijen J, Grijpma DW. A review on stereolithography and its applications in biomedical engineering. Biomaterials 2010;31: Liu Q, Leu MC, Schmitt SM. Rapid prototyping in dentistry: Technology and application. Int J Adv Manuf Tech 2006; 29: Abduo J, Lyons K, Bennamoun M. Trends in computer-aided manufacturing in prosthodontics: A review of the available streams. Int J Dent 2014;2014: Literature Abstract Association of Endodontic Lesions with Coronary Artery Disease Much attention has been given to the association between periodontitis and coronary artery disease (CAD), and this study examines the link with endodontic lesions (ELs). A sample of 508 adults who underwent coronary artery angiography were dentally, clinically, and radiographically examined. They were assigned to one of four groups: no significant CAD, stable CAD, acute coronary syndrome (ACS), and ACS-like but no CAD. The presence of periapical widened spaces and apical rarefactions were determined from panoramic radiographs. Subgingival bacterial samples were taken, and blood samples were drawn. The incidence of 1 widened periapical spaces was 50.4% and 1 apical rarefaction was 22.8%. In total, 51.2% of all teeth with apical rarefactions had been root treated. The results indicated an association between ELs and cardiologic outcomes, especially ACS, which was more evident in untreated ELs. ELs were associated with subgingival P endodontalis and serum IgG levels, indicating a systemic immune response, which may be the potential mechanism of induction of systemic inflammation and endotoxemia. The high prevalence of both diseases warrants further study. Liljestrand J M, Mäntylä P, Paju S, et al. J Dent Res 2016;95: References: 40. Reprints: sagepub.com/journalspermissions.nav doi: / jdr.sagepub.com. johnliljestrand@helsinki.fi Steven Soo, Singapore 188 The International Journal of Prosthodontics