The effect of post-ultraviolet light curing on the accuracy of direct-printed aligners: an in vitro study

Clear aligner therapy^1,2 has become a widely accepted treatment option due to the increasing demand for aesthetic orthodontic treatment. Its origin can be traced back to 1945, when Kesling¹ introduced the concept of a tooth positioner. With recent advances in computer-aided design and computer-aided manufacturing, direct-printed aligners have been identified as an improvement over printed study models upon which aligners are fabricated. The concept of three-dimensional (3D) printing was first developed by Hull in 1984.³ 3D-printed technology aims to make the aligner fabrication workflow easier and more accurate compared to the traditional thermoforming process.⁴

Special resins are used for the direct printing of aligners, which removes the need for 3D model fabrication via the conventional thermoforming process. Stereolithography, direct light processing, fused deposition modelling or polyjet printers are used in the direct printing process.

When printed, the aligners are in a ‘green state’ containing many reactive free radicals which need to undergo post-ultraviolet light (UV) curing. This produces a more cross-linked polymer network which in turn, increases the rigidity and improves the mechanical properties of the material, making it fit for intraoral use.^3,5–8 An optimal duration of post UV curing is applied for each resin to ensure optimal aligner strength without disruption of other mechanical properties or the accuracy of fit. The post-curing protocol must therefore be material specific for both time and temperature.

Since inaccuracies in aligner fabrication can lead to undesired tooth movement, the dimensional accuracy of the printed resins is an imperative. According to previous studies, the dimensional accuracy of Dental LT resin has been compared at different post-UV curing durations and different print orientations.^5,6 However, location-specific deviations have not been assessed. Moreover, no studies have been found which assess Dental LT V2 resin. Deviations over critical areas could be detrimental to the entire treatment plan, and so, if there is an alteration in curing duration, it is necessary to determine if certain areas on the aligner are more prone to deviation than others. Knowledge of these affected locations can lead to the incorporation of over-corrections in aligner design in order to avoid unwanted tooth movement.

The present study therefore aimed to investigate the effect of different post-curing durations on the location-specific accuracy of direct-printed aligners produced using Dental LT V2 resin.

The null hypothesis is that there is no effect of different post-curing durations on the location-specific accuracy of direct-printed aligners.

The present study was approved by the Institutional Ethical Committee of Maulana Azad Institute of Dental Sciences (MAIDS) (Reference number: 592). The research was conducted in the Department of Orthodontics and Dentofacial Orthopaedics, MAIDS, New Delhi. Sample size estimation was conducted according to the study by McCarty et al.⁶ to determine dimensional inaccuracy while printing the aligners at different orientations.

A master maxillary Standard Tesselation Language (.STL) file of thickness 0.5 mm was obtained by scanning an ideal maxillary cast using an intraoral scanner (iTero Element Flex; Align Technology, Inc, San Jose, Calif., USA) (Figure 1). The file was prepared for printing using the PreForm Software (version 2.19.3; Formlabs, Somer-ville, Mass, USA) and a computer processor whose characteristics were: Intel ^® Core (TM); 1.70GHz; 6 GB RAM. According to McCarty,⁶ the files were spaced more than 5 mm apart and oriented at 45° to the build platform, which avoided a large footprint, a cupping effect and the generation of a large number of supports. The support structures were then generated automatically by the software and edited to avoid the inclusion of supports on essential areas such as the intaglio surface (inner tissue facing surface) and cusp tips.

Figure 1.

Thirty copies of the master aligner file were printed on a Form 3B printer (Formlabs, Somer-ville, Mass, USA) using a clear class IIa biocompatible resin⁸ (Dental LT V2, Formlabs) (Figure 2). The aligners were printed at a print layer height of 100 μ m. Six print runs were carried out to print all of the aligners since the build platform was small. Post-processing was conducted according to the manufacturer’s instructions.^9,10 The printed aligners were removed from the build platform and rinsed in FormWash containing 99% isopropyl alcohol for 15 min. Subsequently, the aligners were removed and soaked in fresh isopropyl alcohol (99%) for 5 min⁹ before air drying for 30 min with the assistance of compressed air. The printed aligners were randomly and equally assigned into three groups and placed in Form Cure (Figure 3) for post-UV curing for either 0 min (No cure); 40 min or 60 min (n = 10 each) at 80°C and 405 nm.¹⁰

Figure 2.

Figure 3.

Flush cutters were used to remove the supports from each aligner and a rag wheel provided a final polish (Figure 4). After post-processing, the aligners in each group were numbered and individually sprayed with Spotcheck SKD S2 aerosol developer to facilitate scanning. The spray coating was thinned with compressed air to minimise the chances of deviations (apparent deviations that can arise due to the thickness of the spray coating).

Figure 4.

The aligners were scanned using an intraoral scanner (iTero Element Flex; Align Technology, Inc, San Jose, Calif., USA) to produce an .STL file.¹¹ The 30 digital files obtained were individually superimposed on the master aligner .STL file using an automated best-fit algorithm (Geomagic Control, version 2015.1.1; 3D Systems, Rock Hill, SC). A 0.25 mm tolerance was set in the software to analyse the differences between the surfaces of the master digital file and each superimposed aligner scan.^6,12

Ten comparison points were marked on the digital file and a point-to-point comparison was performed with the help of the software to determine location-specific deviations. The points marked were the mesiobuccal cusps of right and left 1st and 2nd molars (R6, R7; L6, L7); the cusp tips of the right and left canines (R3, L3) and the incisal edges of the right and left central incisors (R1, L1); and the right and left central incisor midfacial points (FR1, FL1). This was in accordance with a previous study conducted by Williams et al.¹³

Reports for each superimposition which included average positive and negative deviations, were generated. Intra-examiner reliability was tested by re-doing the superimposition by the same examiner after an interval of 2 weeks.

The Spotcheck spray was applied to facilitate the scanning of the 3D-printed aligners since past errors have been encountered in the scanning process.

As depicted in Table I and Figure 5, the values of deviation exceeded the 0.25 mm threshold in the ‘No cure’ group and were maximum at the posterior landmarks; however, they were not statistically significant.

Figure 5.

In the ‘40 and 60 min cure’ group, the average positive and negative deviations fell within reported limits of clinical acceptability (0.25 mm) and the values were statistically significant. A negative deviation signified contraction and a positive deviation signified expansion at the various landmarks in the three groups. The maximum deviation was observed in the anterior region in the ‘40 and 60 min cure’ group. The values were positive at the smooth surfaces of the anterior teeth, signifying expansion.

The mean discrepancy between the reference .STL file and the 3D-printed retainers across the 10 landmarks in the 3 groups is shown in Table II.

R6, FL1 and L7 landmarks presented with statistically significant mean differences as shown in Table II and Figure 6.

Figure 6.

The difference in deviation between the landmarks in the ‘40 and 60 minutes cure’ group was only 0.02 mm at R6 and 0.03 mm at FL1. However, it was 0.11 mm at L7. The deviation between the same landmarks was much greater in the ‘No cure’ group when compared to the ‘40 and 60 min cure’ groups.

Images generated by Geomagic Control software allowed for the visualisation of the 3D deviations associated with the landmarks and the samples from each of the 3 groups (Figure 7). Areas appearing in warm colours indicated a positive deviation and in cool colours indicated a negative deviation.

Figure 7.

The intraclass correlation coefficients for the same rater at different intervals for each landmark are presented in Table III.

According to the post hoc results (Table IV), the difference in composite deviation was statistically significant for R6 between the ‘40 min cure and 60 min cure’ groups with the ‘No cure’ group. Similarly, the mean difference in composite deviation for FL1 and L7 was also statistically significant for the ‘40 min cure’ and the ‘60 min cure’ groups with the ‘No cure’ group.

Aligners have become an accepted and integral part of orthodontic treatment.² Since their inception, this treatment option has undergone continuing advances, with a significant breakthrough being the incorporation of 3D printing technology for the fabrication of aligners. Initially used to print 3D models, this concept is now being applied to directly print aligners, and to therefore remove the extra printing model step, upon which aligners would be fabricated by thermoforming. This has proved to be a major development.⁴ Compared to the model resins, the resins used for the fabrication of clear aligners are different and a thorough knowledge of their material properties, post-processing procedures and other parameters is imperative to understand the implications on resulting treatment.

The need to assess the dimensional accuracy of aligners arises from a discrepancy of fit that could be a potential cause for the lack of movement in planned orthodontic movement.^14,15 Depending on the deviation’s location, undesired tooth movement or other clinical concerns can arise due to an imprecise aligner fit which could jeopardise the expected treatment outcome. The failure of teeth to follow a planned movement pathway is a common complication in clear aligner therapy.¹⁶

Several studies have been conducted using the Dental LT 1 resin to identify the factors which could affect the accuracy of 3D-printed aligners.^4,6,17 However, the studies without exception, assessed the overall accuracy of the aligners using a best-fit algorithm, whose validity has been questioned.¹⁸ A landmark-based alignment method used by Cole et al. was found to be more precise.¹⁹

In the present study, the landmarks chosen were at the anterior as well as the posterior cusp tips, similar to those used by Allison et al.,¹³ to provide uniform representation. Since the aligner encompasses the entire tooth surface, the facial surfaces of the central incisors were also assessed. To enhance the depiction of the results, coloured deviation maps were generated using warm and cool colours to signify positive and negative deviations, respectively (Figure 7).

According to Boyd and Vlaskalic, for an aligner to produce 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. If the discrepancy between the aligner and the tooth is greater than 0.25 mm at a site where 0.25 mm movement is prescribed, no clinically appreciable tooth movement will occur.¹² A similar threshold was used in the present study when assessing the deviation at different landmarks.

New resins like Tera Harz have been developed and Koenig et al. compared the material to thermoformed aligner resin and confirmed its greater accuracy.²⁰ However, there has been no study that has compared the superiority of this resin to Dental LT V2 resin in relation to accuracy and precision. All previous studies have compared this resin to a thermoformed aligner material and so Dental LT V2 resin was used to determine its use as an alternative resin, rather than restricted to the direct 3D-printing of aligners.

Tera Harz resin has been used to print aligners using LCD (liquid crystal display) and DLP printers (digital light processing that uses a projector to cure an entire layer at a time). However, the Dental LT resin is compatible with SLA printers (stereolithography printers that use a laser to cure and builds an object from down to up, only).²¹

Post-processing is an important step in the finishing of 3D-printed components. The product obtained initially by 3D printing is in a ‘green’ state, indicating an incomplete cure. The resin can be considered as a highly cross-linked network of polymer chains within which reactive groups can further cross-link and stabilise the polymer network. The objective of post-curing is to link as many unreacted groups as possible and therefore maximise the material properties.⁷

According to the manufacturer, the optimum curing time suggested for this resin is 60 min. It is important to follow instructions for material use to ensure that optimal properties are reached. ²²

Excessive post-curing of certain resins can cause brittleness or warping and inadequate curing may lead to an incompletely cured polymer which may leak and produce cytotoxic effects.²² The post-curing protocol must therefore be specific for both time and temperature to avoid excessive curing, and is therefore, unique to each resin.

There have been no reports which have assessed the effect of post-UV curing duration on the location-specific deviation of the aligners printed using the Dental LT V2 resin. Identifying the specific sites that are more prone to deviation is helpful as overcorrections can be built into the aligner design at the beginning to compensate for discrepancies. This will help predict the accuracy of tooth movement and safeguard the overall treatment plan.

The results of the present study showed that more deviations were identified in the landmarks of the ‘No cure’ group but the values were not statistically significant. The increase in deviations can be ascribed to the increased flexibility of the aligners in the ‘No cure’ group. Since it is obligatory to expose all of the 3D-printed components to UV curing, the inclusion of the ‘No cure’ group provided a comparative control group.

In the ‘40-min cure’ group, the maximum deviation was noted at the incisal edge of the anterior teeth which was similar to the findings of Cole et al.¹⁹ In the ‘60 min cure’ group, the deviation was maximum at the smooth surface of the incisors. This might be a result of the increased rigidity of the resin subjected to post-UV curing, which may have not allowed the material to flex over the contour heights of teeth or adapt to the undercuts.

In the ‘40 and 60 min cure’ group, a positive deviation signified expansion, which was similar to the results obtained by Williams et al.¹³ However, the deviations were within the clinically acceptable range of 0.25 mm in contrast to the studies by Alison et al. and Cole et al.^13,19 in which the deviations at the middle of the incisal edge and the midfacial surface of the central incisors exceeded the 0.25 mm threshold.

The null hypothesis was rejected as there was a variation in accuracy at different post UV curing durations. Post-UV curing of the aligners affected the anterior region more compared to the posterior region; however, the deviation values were comparable in the ‘40 and 60 min cure’ groups.

The limitations of the present study include the addition of a spray to enable scanning but may have also contributed to the deviations observed. However, the exact contribution was not quantified. Since the present study was conducted in vitro and on an ideal cast, future in vivo studies are warranted to substantiate the accuracy of the Dental LT V2 resin.

The Dental LT V2 resin allowed printing of aligners at a single resolution and the print layer height could also not be altered. The accuracy of a future resin designed for intraoral use may have the capability to print at lower resolutions, and at different print layer heights, thereby perhaps improving accuracy, precision, and ultimately, the fit of the aligner.