Now Reading
Comparison of 3-dimensional printing technologies on the precision, trueness, and accuracy of printed retainers

Comparison of 3-dimensional printing technologies on the precision, trueness, and accuracy of printed retainers



American Journal of Orthodontics and Dentofacial Orthopedics, 2022-04-01, Volume 161, Issue 4, Pages 582-591, Copyright © 2021 American Association of Orthodontists


Introduction

This study aimed to evaluate the differences in the precision, trueness, and accuracy of 3-dimensional (3D) printed clear orthodontic retainers fabricated using printer systems with different printing technologies.

Methods

Retainers (n = 15) were 3D printed using 4 different printers: stereolithography (SLA), digital light processing (DLP), continuous DLP, and polyjet photopolymer (PPP) printers. Printed retainers were transformed into a digital image through a cone-beam computed tomography scan and compared with the original image using 3D superimposition analysis software. At previously chosen landmarks (R6, L6, R3, L3, R1, and L1), intaglio surfaces of the retainers were compared to that of the reference model. The intercanine and the intermolar width measurements were also assessed. A discrepancy of up to 0.25 mm between the printed retainer and the reference retainer intaglio surfaces indicated accuracy and clinical acceptability. Precision and trueness were also determined. Root mean square and percent of points within the tolerance level were calculated for precision and trueness for each retainer. Statistical significance was set at P <0.05.

Results

Interrater correlation coefficient indicated good agreement. Statistically significant differences were found between printer types among the 6 landmarks and the arch width measurements. When evaluating tolerance level and root mean square, statistically significant differences in median precision and trueness among each printer type were found.

Conclusion

Retainers fabricated by SLA, DLP, continuous DLP, and PPP technologies were shown to be clinically acceptable and accurate compared to the standard reference file. Based on both high precision and trueness, SLA and PPP printers yielded the most accurate retainers.

Highlights

  • Three-dimensional printed clear retainers are accurate as compared to a reference file.

  • In this research, various 3D printing technologies were studied.

  • SLA and PPP printers show greater accuracy.

  • Based on high trueness and precision values, PPP and SLA were deemed the most accurate overall.

Recent advances in dental technology have allowed dentists to use intraoral scanners to create digital models for diagnosis and treatment planning. Furthermore, one particular advancement has been the 3-dimensional (3D) printing of appliances and surgical guides directly from a digital file.

Thermoformed clear retainers are widely used for retention after the completion of orthodontic treatment. Thermoplastic appliances such as aligners and clear retainers are currently fabricated on plaster casts or 3D printed models. However, if clear retainers can be accurately 3D printed directly from a digital image of teeth, the extra step of creating a physical model may be eliminated. Furthermore, 3D printing may help to decrease the cost while increasing efficiency and improving patient comfort.

In the literature, the first attempt to fabricate a 3D printed retainer was made by Nasef et al in 2014. The appliance was built directly from a cone-beam computed tomography (CBCT) scan. In a later study, the same authors compared the accuracy of 3D printed and vacuum-formed traditional retainers and concluded that 3D printing was a reliable method for fabricating these appliances.

In 2019, Cole et al also evaluated the accuracy and precision of digitally printed retainers by comparing them to vacuum-formed traditional retainers. The authors concluded that the differences up to 0.5 mm between the intaglio surfaces of the 2 groups were insignificant. However, the threshold value is different from a study by Johal et al, in which the fit of thermoformed retainers was tested under laboratory conditions, and a discrepancy of up to 0.25 mm was used as a threshold for a clinically acceptable fit.

Finally, the findings of a recent study by Jindal et al reported that 3D printed aligners were geometrically more accurate and durable than thermoformed appliances.

Today there are several 3D printing systems available in digital dentistry. One of the most established technologies, stereolithography (SLA), consists of a photosensitive resin, a platform for building the object, and an ultraviolet (UV) light to cure the resin on the platform ( Fig 1 , A ). In this additive manufacturing process, the UV light cures each layer, and as the layer is cured, the tray descends to add more liquid resin until a shape is formed. The laser continues to trace and form each layer of the previous layer, building from the bottom until the final product is obtained. Digital light processing (DLP) is another 3D printing system used to create an object ( Fig 1 , B ). A DLP printer operates the same way as an SLA printer, but in contrast, a projector instead of a laser cures the entire layer each time. Another variation of the DLP printer is a continuous digital light processing (cDLP) printer. This is a new approach in which the build plate constantly moves vertically, allowing light to cure the polymers without interruption. Finally, a polyjet photopolymer (PPP) printer operates similarly to a standard inkjet printer, except it also has a vertical component in which the UV light cures the resin as the material is jetted out by nozzles ( Fig 1 , C ). The object is cured layer by layer at a time from the bottom up. Once the liquid resin is sprayed on the platform, the UV light attached to the printer head simultaneously cures the material as it is printed. This printer allows for the smallest microns of build layers, leading to better print quality. PPP printers are limited to a slice thickness of 16 μm and 32 μm. It is possible to mix multiple materials using a PPP, unlike an SLA that uses only one material for the model and the supports. Therefore, a gel-like soluble resin can be used for supports on a PPP printer. The 2 key differences with printing technologies are the smaller dimensions, soluble support material, and multi-material capabilities for the PPP. The SLA also has faster printing times and, therefore, may be better for larger-scale prints. In summary, the different types of printers vary in the resin material used, the print layer heights, and the curing modalities.

Schematic drawings showing the technology used for (A) SLA, (B) DLP, and (C) PPP printers.
Fig 1
Schematic drawings showing the technology used for (A) SLA, (B) DLP, and (C) PPP printers.

Previous research on the accuracy and precision of 3D printed models reported significant differences among 3D printing technologies. Specifically, the PPP and DLP printers were more precise than the SLA systems, with the PPP printers also showing the highest accuracy. However, there is limited information on the most accurate and precise printer system for fabricating clear retainers to date. The purpose of this study was to evaluate the precision, trueness, and accuracy of clear retainers 3D printed with different technologies.

Material and methods

This study used a previously created standard tessellation language (STL) file of a maxillary retainer to fabricate the samples. Ball markers, 1.5 mm in diameter, were digitally placed using Autodesk Meshmixer software (Autodesk, Inc, San Rafael, Calif) on the mesiobuccal cusps of the right and left first molars (R6, L6), the cusp tips of the canines (R3, L3), and the middle of the incisal edges of the central incisors (R1, L1). These reference points helped to eliminate errors during the superimposition process later in the study.

To fabricate clear retainers, the following 3D printers were used: SLA printer (Form 3; Formlabs Inc, Somerville, Mass), DLP printer (MoonRay, SprintRay Inc; Los Angeles, Calif), cDLP printer (Envision One cDLM Dental; EnvisionTEC, Dearborn, Mich), and PPP printer (Objet Eden260VS, Stratasys, Eden Prairie, Minn). These systems were chosen to represent the leading technologies currently available in digital dentistry.

A total of 60 retainers (n = 15, each) were printed using the resin specific to each system on the basis of the manufacturer’s recommendation via private communication. When the sample size in each of the 4 groups is 15, a one-way analysis of variance will have 80% power (at a 0.05 significance level) to detect a difference in means characterized by a variance of means of 0.003, assuming that the common standard deviation is 0.12, which was demonstrated by Cole et al.

Because the purpose of the study was to test the accuracy and precision of different printing technologies, the manufacturer’s recommendations specific to each printer system were strictly followed for the resin material, print angulation, and postprocessing procedures.

Retainers in the PPP group were printed horizontally at a 16 μm layer thickness. In contrast, the appliances in the SLA, DLP, and cDLP groups were fabricated at 100 μm, and angulation was recommended by the manufacturer ( Fig 2 ). With the supports still attached, the appliances were scanned using CBCT to create their corresponding digital images. CBCT technology has previously been an essential tool in creating STL files from dental (elastomeric and nonelastomeric) impressions and dental plaster casts. The settings were chosen according to the research by Ammoun et al using CBCT (iCAT FLX V10; Imaging Sciences International, Hatfield, Pa) with standard postoperative implant scan parameters (120 kVp; 5 mA; 0.3-mm voxel size; scan time, 2 seconds; field view of no more than 10 cm in height × 16 cm in depth; and 283, 582, or 291 mGy/cm 2 ). As clear retainers are vulnerable to distortion and breakage because of their thin structure, the removal of the supports was carried out digitally to reduce errors so that a valid comparison of the 3D printed retainers to the original file could be made.

Postprocessed printed retainers with (A) SLA, (B) DLP, (C) cDLP, and (D) PPP technology.
Fig 2
Postprocessed printed retainers with (A) SLA, (B) DLP, (C) cDLP, and (D) PPP technology.

The intaglio surfaces of the printed retainers (experimental) and the reference retainer (control) were superimposed using a best-fit method on a 3D analysis software (Geomagic Control; 3D Systems, Rock Hill, SC). A total of 6 previously chosen reference points (R6, L6, R3, L3, R1, and L1) were used to evaluate the accuracy of the pairings in the vertical dimension ( Fig 3 ). In addition, intercanine width (ICW) and intermolar width (IMW) calculations were performed to make comparisons of the 2 images in the transverse dimension ( Fig 3 ). Measurements were performed by 2 operators independently and in random order (O.A.N.).

Landmarks, ICW, and IMW calculations with the superimposition to reference STL.
Fig 3
Landmarks, ICW, and IMW calculations with the superimposition to reference STL.

Trueness, or the closeness of the printed retainer to its true model, was determined by comparing the digital data of the printed retainer to the reference STL file ( Fig 4 ). To determine the precision, 5 of the 15 printed retainers were randomly chosen for each printer, and comparisons were made. These combinations included retainers 1:2, 1:3, 1:4, 1:5, 2:3, 2:4, 2:5, 3:4, 3:5, 4:5 resulting in a total of 10 comparisons per printer group. Using the 3D analysis software, root means square error (RMSE) values and the percent of points within the tolerance level (inTOL) were calculated ( Fig 4 ).

The experimental retainer to reference STL superimposition analysis output with RMS (mm) and inTOL (%).
Fig 4
The experimental retainer to reference STL superimposition analysis output with RMS (mm) and inTOL (%).

Precision and trueness of the retainers were assessed when a low RMS value and a high inTOL percentage indicated a good fit. Finally, accuracy representing the combination of both precision and trueness was assessed.

Statistical analysis

The interrater correlation coefficient was used to determine the agreement between 2 raters for the ICW and IMW assessments and measurements at the 6 landmarks. Analysis of variance (ANOVA) was used to test for differences in accuracy at each of the 6 designated reference points on the basis of the type of printer, and post-hoc pairwise comparisons were adjusted using Tukey honestly significant difference tests. Two-way student t tests were used to evaluate if the mean distance of the canine and molar measurements from the printed retainers was different from the calculated distance of the original STL file. Additional ANOVA models were used to assess precision and trueness between printer types on the basis of RMSE values. Estimated means and 95% confidence intervals (CIs) were reported. Because of the nonnormality of the RMSE data, nonparametric Kruskal-Wallis tests were used to assess precision and trueness between printer types for percent within the defined tolerance (±0.25). Post-hoc pairwise comparisons were adjusted using Dunn multiple comparison tests. Median values and bootstrap 95% CIs using 10,000 samples were reported. All statistical analyses were performed in R (version 3.6.1, R Development Core Team; University of Auckland, New Zealand). All statistical tests were assessed at the α = 0.05 significance level.

Results

The interrater correlation coefficient indicated good agreement between 2 raters for all the landmarks ( Table I ). The differences in the measurement values between the 2 raters were so small that the agreement was also reported as the percent agreement within 0.10 mm ( Table I ). At all reference points, values were within 0.10 mm at least 95% of the time, indicating clinical insignificance. Therefore, measurements from only 1 rater were used for further analyses.

Table I
ICC between the 2 raters for each landmark
Landmarks ICC (95% CI) % with difference <0.10 mm P
R6 0.876 (0.801-0.924) 100.00 <0.0001
L6 0.701 (0.544-0.810) 98.30 <0.0001
R3 0.894 (0.814-0.938) 98.30 <0.0001
L3 0.933 (0.882-0.961) 100.00 <0.0001
R1 0.804 (0.693-0.878) 96.70 <0.0001
L1 0.749 (0.614-0.842) 95.00 <0.0001
ICW 0.990 (0.984-0.994) 98.30 <0.0001
IMW 1.000 (1.000-1.000) 100.00 <0.0001
ICC , interrater correlation coefficient.

Table II shows the mean discrepancy between the digital image of printed retainers and the reference model for each measurement and printer group. There were statistically significant differences between the printers among each of the 6 reference points. Overall, a low estimated mean discrepancy at each landmark indicated a close fit between the printed retainers and the reference retainer ( Fig 5 ). There was a significant variation within the printers for the least mean difference at the molar, canine, and incisor landmarks, with the PPP printer showing the lower mean difference in the incisors (R1, L1), the DLP printer in the canines (R3, L3), and the cDLP and SLA printers in the molars (R6, L6). Nevertheless, the estimated mean difference for each printer type at each landmark fell within the 0.25 mm tolerance level, as shown as a dashed line in Figure 5 .

Table II
Post-hoc Tukey honestly significant difference tests showing the estimated mean discrepancy between the digital image of the printed retainers and the original STL image for each measurement across each printer
Landmark Printer type Estimated mean discrepancy (mm) Standard error P
R6 <0.0001
SLA 0.075 a 0.013
DLP 0.149 b 0.013
cDLP 0.075 a 0.013
PPP 0.148 b 0.013
L6 <0.0001
SLA 0.089 a,b 0.011
DLP 0.102 b 0.011
cDLP 0.047 a 0.011
PPP 0.149 c 0.011
R3 0.0083
SLA 0.095 a,b 0.018
DLP 0.045 a 0.018
cDLP 0.095 a,b 0.018
PPP 0.138 b 0.018
L3 <0.0001
SLA 0.11 b 0.011
DLP 0.043 a 0.011
cDLP 0.129 b 0.011
PPP 0.093 b 0.011
R1 0.0056
SLA 0.098 a,b 0.019
DLP 0.165 b 0.019
cDLP 0.156 b 0.019
PPP 0.081 a 0.019
L1 0.0007
SLA 0.11 a,b 0.015
DLP 0.128 b 0.015
cDLP 0.153 b 0.015
PPP 0.06 a 0.015
ICW <0.0001
SLA 0.314 a 0.03
DLP 0.642 c 0.03
cDLP 0.469 b 0.03
PPP 0.262 a 0.03
IMW <0.0001
SLA 0.588 b 0.057
DLP 1.108 c 0.057
cDLP 0.997 c 0.057
PPP 0.145 a 0.057
Note. Values with the same superscripted letter (ie, a, b, or c) are not significantly different.

Bar plot illustrating the mean difference in print accuracy for each landmark across each print type.
Fig 5
Bar plot illustrating the mean difference in print accuracy for each landmark across each print type.

The ICW and IMW measurements showed differences in accuracy between the printers ( Fig 6 ). ICW and IMW values for the retainers in the PPP printer group with the mean values of 34.54 mm and 50.10 mm were closest to the original values of the reference retainer (34.80 mm and 51.15 mm, respectively). This finding indicated that the retainers in the PPP group exhibited almost an identical replication with only a negligible dimensional change. The retainers in the SLA group had 34.48 mm and 49.57 mm for the ICW and IMW values, respectively. Samples in the DLP and cDLP groups had the highest discrepancy for these measurements.

Bar plot illustrating how the mean distance (mm) across canine and molar points compares to the original STL file for each type of printer.
Fig 6
Bar plot illustrating how the mean distance (mm) across canine and molar points compares to the original STL file for each type of printer.

When evaluating the percent of points within the tolerance level (inTOL), statistically significant differences in both the median precision and trueness among each printer type were found ( Table III ). For precision, the cDLP samples were shown to have the highest precision with a median of 99.9% of points within the preset tolerance (bootstrap CI, 99.7- 100.0). However, post-hoc analyses revealed that cDLP retainers were not statistically different from the DLP retainers. For trueness, the SLA retainers were found to have the highest trueness with a median of 94.9% of points within the preset tolerance bounds (bootstrap CI, 89.4-99.1), yet they were not statistically different from PPP samples.

Table III
Evaluation of precision and trueness for inTOL and RMS across each printer type
Measure Printer (%) P
SLA DLP cDLP PPP
inTOL
Precision # 97.9 (85.9-99.8) b 99.4 (98.8-100.0) a,b 99.9 (99.7-100.0) a 97.6 (92.7-98.9) b 0.0015
Trueness # 94.9 (89.4-99.1) b 75.3 (74.0-76.6) a 82.6 (80.7-83.9) a 93.3 (88.4-94.4) b <0.0001
RMS ∗∗ (mm)
Precision ## 0.111 (0.0905-0.1312) c 0.067 (0.0467-0.0873) a,b 0.052 (0.0317- 0.0724) a 0.101 (0.0804-0.1210) b,c 0.0005
Trueness ## 0.121 (0.102-0.140) a 0.242 (0.223-0.261) c 0.203 (0.184-0.221) b 0.151 (0.132-0.170) a <0.0001
Note. Values with the same superscripted letter (ie, a, b, or c) are not significantly different.

Kruskal-Wallis test with post-hoc Dunn’s multiple comparison tests.

# median (95% bootstrap CI using 10,000 samples).

∗∗ ANOVA with post-hoc Tukey’s multiple comparison tests.

## mean (95% CI).

Similarly, the RMS data indicated statistically significant differences in the mean precision and trueness values among each printer type ( Table III ). Once again, the cDLP retainers exhibited the highest precision (0.052 mm; 95% CI, 0.0317-0.0724) but low trueness (0.184 mm; 95% CI, 0.184-0.221). Although the SLA samples had the highest trueness (0.121 mm; 95% CI, 0.102-0.140), they exhibited low precision (0.111 mm; 95% CI, 0.0905-0.1312). Post-hoc analyses indicated that the cDLP retainers were statistically different from the SLA and PPP but not DLP retainers. The SLA samples were statistically different from the cDLP and DLP but not the PPP retainers.

Discussion

Previous studies have shown that 3D printed models replicate the occlusion accurately to be reliable for diagnosis and treatment planning in orthodontics. However, fabricating retainers and clear aligners directly from a digital scan is still very much in its infancy, and there is limited information on the accuracy of these appliances. This study evaluated the accuracy of 3D printed retainers fabricated by 4 different printing technologies. Previously determined 6 reference points were used to superimpose the digital image of the printed retainer on the original STL image of the reference retainer. This method consisted of measuring the mean differences across these landmarks to evaluate the closeness of the fit of the samples to the reference retainer on intaglio surfaces, similar to Cole et al. However, in the current study, retainers were 3D printed with 1.5 mm ball markers that were previously placed virtually on the digital images to aid with choosing the landmarks. The addition of these reference points helped to eliminate operator error during the superimposition process. In addition, the RMS feature of the Geomagic software provided the overall fit error when precision and trueness of the printed retainers were assessed. This approach was similar to the technique used by Kim et al.

Discrepancies in the ICW and IMW measurements between the sample and the reference retainers indicate that the resin has experienced dimensional change during the printing process. The intraarch distortion and the polymerization shrinkage may be attributed to the print angulation when printing the retainers. It should be noted that the retainers in the PPP group were printed at a horizontal orientation as opposed to the samples in the other printer groups at 30°-45° of angulation. PPP printers can print using 2 simultaneous resins. Therefore, a soluble gel-like support resin can allow for horizontal printing without distortion of the main resin. In addition, it is well known that printing at a horizontal orientation will result in more accurate objects. , Therefore, a possible explanation for less distortion in the PPP group may be the horizontal print orientation.

Another factor contributing to the distortion and shrinkage of the resin may be the layer height used when printing the retainers. The z-axis is printed at a constant layer height on the basis of the micron settings of the printer and resin. The PPP group samples were printed at the smallest micron setting (16μm) compared with the other printers (100 μm) and thus had the smallest printing height and the smoothest finish. Although this may seem to indicate better accuracy, previous studies have argued that smaller build height does not necessarily lead to greater accuracy because of a greater potential for error.

Based on the study by Boyd and Waskalic, for an aligner to cause tooth movement, a minimum of 0.15 to 0.25 mm distance needs to exist between the cast and the appliance when the aligner is active. Johal et al used the same threshold in their study when evaluating the fit of thermoformed retainers. Similarly, Cole et al reported discrepancies between the traditional vacuum-formed retainer and reference model of 0.1-0.2 mm and 0.1-0.3 mm for commercially vacuum-formed retainers. Although a precise fit is essential in thermoformed appliances for successful retention, , there are no previous studies that evaluated the precision and trueness of vacuum-formed retainers in patients. Shrinkage observed in the ICW and IMW in these 3D printed retainers may affect the clinical suitability; therefore, future studies are warranted to evaluate the fit on casts.

The retainers in all 4 printer groups showed mean discrepancies less than 0.25 mm at the 6 reference points, indicating accuracy within the clinical acceptance tolerance. However, it must be noted that to evaluate whether a printed retainer is accurately reproduced from its digital file, the overall retainer fit is more important than its adaptation at selected landmarks. The cDLP and the SLA samples showed less error in the molar region, the DLP retainers in the canine area, and the PPP and SLA samples in the incisor region. A previous study, that examined the effect of print orientation on the dimensional accuracy reported that aligners printed at a 45° angle had the smallest deviation from the original file. In addition, the authors noted that depending on the orientation used, there were deviations at different parts of the aligners among different directions used, but consistently at the same locations within the group itself. Therefore, it was concluded that printing orientation affects the accuracy of the aligners. Similarly, in the current research, differences in deviations from the original file were found among the groups. Why a retainer had a better fit at 1 landmark vs another may be due to several reasons, including print angulation, location of the model on the baseplate, postprocessing procedures of the retainers, overexposure of some layers as the build develops because of a clear resin, and finally, errors from the CBCT scan.

In this study, the overall fit method provided a more accurate and reliable assessment of precision and trueness than the techniques used in previous studies. For precision, the total percentage of points <0.25 mm mean difference, expressed as a percentage within the inTOL, was 99.9% and 99.4% for the cDLP and DLP retainers, respectively. Although these values indicated excellent precision, the relatively low inTOL percentages (82.6% and 75.3%) showed low trueness for these printers. Because accuracy is defined as the combination of precision and trueness, the cDLP and DLP retainers were deemed less accurate than the SLA and PPP retainers that had high percentages in both precision and trueness. Samples in the cDLP and DLP groups were physically smaller than the reference retainer, as evidenced by the increased discrepancy in the ICW and IMW values.

The precision and trueness of the retainers were also evaluated with the interpretation of the RMS data. RMS allows the offset error to be represented more accurately because the effect of positive and negative values of the difference of measurements of 3D structures is eliminated. According to the RMS analyses, the SLA and PPP retainers were deemed to have precision, trueness, and accuracy.

In the literature, the information on the precision of 3D printed retainers is limited. In a study by Kim et al, the PPP printers were shown to have the highest precision, followed by the DLP and SLA printers for printing 3D models. The DLP showed the highest precision in our study, followed by PPP and SLA for printing 3D retainers.

In this study, 4 print failures were observed during the printing process. In the cDLP group, 3 samples exhibited cupping at their canine cusp tips. Cupping, a printing defect, occurs when a hollow area traps air and causes high suction during the print process. Cupping will occur unless drainage holes are added to prevent the uncured resin from getting trapped inside the print. If the cup is shallow, angling the pattern may be sufficient to minimize the cupping effect. Most printing software can recognize this issue and adjust the model by changing the orientation or adding supports with vent holes in these at-risk areas. In our study, it appears that the cDLP printer software was unable to either identify or correct this phenomenon, resulting in failures at the canine regions of the retainer.

The fourth failure occurred in the PPP printer during postprocessing. Each retainer was subjected to a waterjet cleaning method described by the manufacturer, in which the soluble support material used to maintain the retainer's dimensional stability was removed by a high-powered waterjet. The gel-like support material was able to keep the stability of the retainer and helped to avoid the need to angle the retainer. However, because of the thin width of the retainers, the high-powered water jet did overpower the first retainer causing some distortion. Although this retainer looked similar to the others, it did show a slight deformation in its x-y plane, which was enough to consider it a failure. Therefore, a more careful method of postprocessing with more fragile structures such as retainers is indicated in the removal of the soluble supports.

Although this study focused on the overall fit and the accuracy of specific landmarks, certain important reference points could have been overlooked. In addition, although the overall fit for the retainers was clinically acceptable on the basis of the average mean discrepancy in RMS values, there were retainers with differences in certain areas reaching almost 1.0 mm, or 4 times the accepted threshold ( Fig 5 , the minimum and maximum values.) Similarly, Cole et al found that the discrepancy in the middle of the incisor edge of central incisors (R1, L1) and the midfacial point of the central incisors (FR1, FL1) were greater than the 0.25 mm threshold. These findings suggest that the printing capabilities may be limited in the anterior region regardless of print angulation Once the errors in the facial surfaces of the anterior teeth are corrected, a future study evaluating the fit of 3D printed retainers in orthodontic patients may shed light on the clinical performance of these appliances.

The large deviations from the original intaglio surfaces were more commonly observed in the DLP and cDLP retainers than in the SLA and PPP appliances. In many instances, these disparities were located at smooth surface areas such as the buccal and lingual of the posterior teeth in red and blue colors on color-coded superimpositions, indicating a lack of good fit. In contrast, the superimposition of an ideal retainer would display a homogenous green color . Similarly, in a recent study that evaluated the accuracy of aligners fabricated on an SLA printer using different angulations, deviations from the original STL file were reported at specific locations. For example, at the 45° print angulation, overbuilding in the central grooves and the lingual of the maxillary anterior teeth were observed.

Furthermore, each print angulation used resulted in deviations at certain locations specific to that printer setting. Nevertheless, the study determined the little effect of the print orientation on the overall dimensional accuracy. The presence of location-specific deviations was thought to affect the clinical usefulness of the printed aligners. In the current study, print angulation was not intentionally standardized because of the manufacturer’s recommendations which may have introduced bias.

Another limitation of the current study was that the retainers were compared to a digital reference retainer. For a valid comparison of the 3D printed retainers to the original file, the supports were removed digitally. This adaptation may not represent the clinical implications as support removal would be necessary. Unfortunately, to date, there is no information on how well vacuum-formed retainers fit once delivered clinically. Therefore, future clinical studies comparing the 3D printed retainers to a thermoformed retainer along with the fit of these retainers on patients are warranted.

Although the resins used in this study are available for intraoral applications, they are not currently marketed for retainer fabrication. At this time, there is no approved commercially available biocompatible resin for fabricating clear orthodontic retainers directly from a digital image file; however, a new compatible resin material may soon become available for 3D printing of retainers and aligners. Future studies with an approved resin material will be useful for evaluating the mechanical and physical properties and the potential for clinical applications.

In summary, this study has compared different 3D technologies. However, the results of this study need to be interpreted with caution. Looking at various factors such as precision, trueness, and accuracy, one can argue that one printer is better in one aspect, another printer is better in another aspect. Furthermore, although there were statistically significant differences between printers with different technologies, there may not be a clinical significance. When the new technology is adopted in dentistry, precision, accuracy, cost, and ease of use should also be considered.

You're Reading a Preview

Become a DentistryKey membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here