Surface roughness and marginal adaptation of stereolithography versus digital light processing threedimensional printed resins: An in-vitro study

Abstract

Aim:

The aim of this study was to assess surface roughness and marginal adaptation of Stereolithographic versus Digital Light Processed three-dimensional (3D) printed provisional resins.

Materials and Methods:

A 3-unit fixed partial denture (FPD) preparation was done on ideal model irrespective to 44–46. The Model was scanned and a 3-unit FPD was designed using 3-shape software. The STL file was transferred to two different 3D printers – Sprintray digital light processing (DLP) and Formlabs stereolithography (SLA). Eight samples were printed per group (total of 16 samples) using C and B temporary tooth-colored resin and cured according to the manufacturer's instructions. Marginal adaptation was checked for six surfaces per tooth for all the samples using a stereomicroscope. Surface roughness was also calculated for four samples from each group before and after polishing (pumice slurry + rouge and cotton buff) using a contact profilometer.

Results:

The mean maximum marginal gap overall, was seen for the DLP group on the mesiobuccal surface of the first premolar, i.e., 178.8 ± 8.35 μm, while the minimal marginal gap was seen for the SLA group on the mesiolingual surface of first molar − 32.5 ± 7.07 μm. Furthermore, all the DLP samples showed a statistically significant higher mean marginal gap as compared to SLA samples (P < 0.005). All the samples showed surface roughness within the acceptable range. There was a statistically significant difference noted in Rz (roughness depth) before and after polishing (P < 0.05).

Conclusion:

3D printed temporary resin FPD via SLA showed a much better marginal adaptation (49.6 μm mean marginal gap for 46 and 106.8 μm for 44) as compared to those printed via DLP (101.8 μm mean marginal gap for 46 and 157.5 μm for 44). All the samples showed an acceptable surface roughness.

Clinical Relevance:

3D printed temporaries have shown good marginal fit and adaptation and are a viable choice in patients where temporaries has to be given for long term before a final prosthesis can be fabricated (especially for full mouth rehabilitations).

INTRODUCTION

The introduction of computer-aided design and computer-aided manufacturing (CAD/CAM) has revolutionized the field of restorative dentistry.[1] Prosthesis fabricated using CAD/CAM has several advantages such as reduced patient appointments, excellent tissue adaptation, and mechanical properties.[2,3,4] Digital manufacturing includes subtractive methods using computer-aided milling and additive method using 3-dimensional (3D) printing. Milling technologies have the inherent disadvantage of being unable to sculpt complicated details such as undercuts and intaglio geometry, as well as the ability to create only one unit at a time. Additive manufacturing, on the other hand, can render complicated geometry and has the potential to be considerably more resourceful as it does not involve the wear of rotary tools or wastage of materials.[5]

A total of seven additive manufacturing technologies were determined by the ASTM committee F42-Stereolithography (SLA), material jetting, material extrusion or fused deposition modeling, binder jetting, powder bed fusion, sheet lamination, and direct energy deposition.[6,7] In the SLA process, a build platform is immersed in liquid resin, which is subsequently polymerized by an ultraviolet (UV) laser. Each layer's cross-section is traced by the laser.[8]

The layer thickness can be customized by the user and dictates how much distance the build platform will drop into the photopolymer vat to allow the uncured resin to cover the previously cured layer. This process is repeated till the printing is complete.[9] The ASTM classifies digital light processing (DLP) into an identical category as SLA because their technologies are similar. The cross-section of object to be printed is projected by a matrix of microscopic mirrors with a semiconductor chip which is also known as a digital micromirror device. Some printers also use an arc lamp to project the image on the vat of liquid photopolymer through a DLP projector with a UV clear window under safe lighting. The printed object is rather drawn upward from the liquid resin than being submerged in this technique.[10,11] SLA and DLP are the most common technologies used in dentistry worldwide.

Recently, CAD/CAM technology has been extensively used to manufacture provisional crowns for patients via an indirect method.[12] Good quality provisional crowns are required to safeguard the prepared teeth and periodontal tissues.[13] They are also utilized to preserve the function and esthetics of the oral cavity. Internal fit and marginal adaptation are critical for any restoration's long-term clinical success.[14,15] Microleakage and plaque accumulation can be exacerbated by poor marginal fit, leading to cement disintegration, recurrent decay, and periodontal inflammation. As a result, extra attention should be paid to the marginal adaptation of restorations.[16,17]

The characteristics of the material such as marginal fit, hardness, and roughness affect the stability of color, in turn, influencing esthetic appearance, preservation of occlusal relationship, and bacterial adhesion which will lead to a biofilm formation. If the material used and the technique for fabrication of temporary crowns is adequate, the final result will be of superior quality and, hence, maintain the integrity of the periodontium.[18]

The aim of this study was to assess surface roughness and marginal adaptation of stereolithographic versus. DLP 3D printed provisional crowns. The null hypothesis stated that there is no difference between the surface roughness and marginal adaptation of provisional crowns fabricated by DLP or SLA 3D printers.

MATERIALS AND METHODS

Sample preparation

An ideal tooth preparation was done on a typodont model (NISSIN Typodont Jaw Model) with respect to the first premolar and molar, while the second premolar was removed to simulate a partially edentulous condition for fabrication of a fixed dental prosthesis (FDP). The preparations were evaluated by two different faculty members for the presence of undercut or any defect. The preparations were scanned by a Medit^® Lab Scanner and files were exported in STL format and transferred to 3Shape^® CAD software. A 3-unit FDP was designed and was transferred to two different types of 3D printers-Sprint Ray^®(DLP) and Form Labs^®(SLA). Eight samples per group were 3D printed (total of 16 samples) using NextDent C and B temporary tooth-colored resin. The build angle and layer thickness were rendered identical for both types of printers. The residual surface monomer was cleaned using 99.9% ethyl alcohol and support structures were clipped flush with the printed structure before polymerization using specific light cure units for the two printers according to the manufacturer's guidelines [Figure 1a-c].

Figure 1.

The sequential methodology of the study. (a) Tooth preparation, (b1): FormLabs printer, (b2): SprintRay printer, (c1): DLP sample, (c2): SLA sample, (d): Stereomicroscope, (e1): Contact Goniometer, (e2): Goniometer tip on sample. DLP: Digital light processing, SLA: Stereolithography

Marginal adaptation

All the samples were seated on the ideal model without any internal surface modifications or adjustments. The typodont teeth were removed from the model for ease of recording. All the provisional FDP samples were evaluated under a stereomicroscope (LYNX, Lawrence Mayo) with ×2 by a blinded reviewer for six surfaces per tooth-buccal, lingual, mesiobuccal, mesiolingual, distobuccal, and distolingual surfaces. A digital scale incorporated in the stereomicroscope software was used [Figure 1d].

Surface roughness

Four samples each from the two groups were randomly selected and were subjected to a surface roughness test by a contact goniometer (Ossila^®) with a measurement accuracy of ±1° and a measurement range of 5°–180°. A droplet is placed on the substrate which is then gradually tilted. The advancing angle is measured at the front edge of the droplet just before the droplet becomes unpinned and starts to move. The receding contact angle is measured at the rear of the droplet before the trailing edge starts to move. Measurements were made both before and after surface polishing by pumice slurry, followed by rouge and a cotton buff [Figure 1e]. The values were evaluated by a blinded investigator and the same surface was chosen for all the samples (buccal surface) for roughness calculation.

Statistical analysis

All the values obtained were tabulated and coded in a spreadsheet and then transferred to IBM SPSS v23.0 software (IBM Corp. Armonk, NY). Due to the scaled nature of data, parametric tests were chosen. Unpaired/independent t-test was performed to find if significant differences are there between the two groups and the P value was calculated.

RESULTS

The group statistics of the marginal gaps seen on the six surfaces of both the abutments for all samples is summarized in Table 1. The mean maximum marginal gap overall was seen for the DLP group on the mesiobuccal surface of the first premolar, i.e., 178.8 ± 8.35 μm while the minimal marginal gap was seen for the SLA group on the mesiolingual surface of the first molar −32.5 ± 7.07 μm. The mean marginal gaps seen across the surfaces of the first molar were significantly less than that seen with the first premolar. Furthermore, all the DLP samples showed a statistically significant higher mean marginal gap as compared to SLA samples (P < 0.005).

Table 1.

Descriptive statistics of marginal gap seen on the six surfaces examined

Group statistics
	SLA or DLP	Mean	Std. deviation
Marginal gap 44 buccal surface	SLA	0.1538	0.01408
	DLP	0.1975	0.01581
Marginal gap 44 lingual surface	SLA	0.1075	0.01282
	DLP	0.1712	0.01246
Marginal gap 44 mesiobuccal surface	SLA	0.2000	0.01195
	DLP	0.2788	0.00835
Marginal gap 44 mesiolingual surface	SLA	0.2087	0.01808
	DLP	0.2638	0.02326
Marginal gap 44 distobuccal surface	SLA	0.1775	0.02252
	DLP	0.2600	0.01069
Marginal gap 44 distolingual surface	SLA	0.1463	0.01188
	DLP	0.1837	0.00744
Marginal gap 46 buccal surface	SLA	0.0813	0.01246
	DLP	0.1600	0.01069
Marginal gap 46 lingual surface	SLA	0.0588	0.01458
	DLP	0.1125	0.00886
Marginal gap 46 mesiobuccal surface	SLA	0.0425	0.01035
	DLP	0.0837	0.01302
Marginal gap 46 mesiolingual surface	SLA	0.0325	0.00707
	DLP	0.0875	0.01282
Marginal gap 46 distobuccal surface	SLA	0.0400	0.00926
	DLP	0.0863	0.01188
Marginal gap 46 distolingual surface	SLA	0.0425	0.01165
	DLP	0.0813	0.01126

The mean surface roughness (R_a) was 0.24 ± 0.07 μm for SLA samples and 0.28 ± 0.05 μm for DLP samples. There was no statistically significant difference between the two groups The difference in mean surface roughness values of the samples before and after polishing are displayed in Table 2 along with paired t-sample test.

Table 2.

Paired samples t-test for surface roughness of various samples pre- and postpolishing

Paired samples test
	Paired differences, mean (µ)±SD	t	df	Significant (two-tailed)
Pair 1: Roughness average SLA prepolishing - Roughness average SLA postpolishing	0.00020±0.00045	1.000	4	0.374
Pair 2: Roughness depth SLA prepolishing - Roughness depth SLA postpolishing	0.00780±0.00396	4.402	4	0.012
Pair 3: Roughness average DLP prepolishing - Roughness average DLP postpolishing	0.00020±0.00045	1.000	4	0.374
Pair 4: Roughness depth DLP prepolishing - Roughness depth DLP postpolishing	0.00640±0.00182	7.878	4	0.001

SLA: Stereolithography, DLP: Digital light processing, SD: Standard deviation

DISCUSSION

The null hypothesis states that there was no difference between the DLP and SLA samples in the marginal adaptation and surface roughness. The null hypothesis was partially rejected as SLA samples showed a statistically significant difference in terms of marginal adaptation when compared with DLP samples (P < 0.005), but no difference was found between the surface roughness values. However, a significant reduction in roughness depth (R_z) was observed after polishing the samples (P < 0.05).

The clinical outcome of dental restorations depends highly on marginal adaptation.[17] In general, the precision of marginal fit is determined by tooth preparation, impression technique, restorative materials and technology used for fabricating them, and even on the luting cement. In the previous publications, the average discrepancy in the marginal fit has been reported to be in the range from 177 to 400 μm for interim crowns.[18,19] McLean and Von[20] reported a marginal gap of 120 μm to be clinically acceptable while Boening et al.[21] claimed that a marginal gap between 100 and 200 μm lies within the clinically permissible range for a definitive prosthesis. In the current study, the DLP samples showed a mean marginal gap of 101.8 ± 11.42 μm for the first molar and 157.5 ± 13 μm for the first premolar. On the other hand, SLA samples showed much lower mean marginal gap values, i.e., 49.6 ± 10.9 μm for the first molar and 106.8 ± 15.22 μm for the first premolar. Although both groups showed values within the clinically acceptable range, SLA samples had statistically significant better marginal adaptation as compared to DLP samples (P < 0.005).

Previous studies[22,23] have already demonstrated that surface roughness significantly influences the extent of microbial adhesion to the denture base. The microbial attachment was increased on rougher surfaces, with roughness values ranging between 0.1 and 0.4 mm. Smooth interim restorations are essential to avoid biofilm accumulation and to maintain healthy periodontal tissues. The surface roughness values obtained in the current study lie in the threshold of clinical relevance, as described by Quirynen et al. and Bollen et al.[24,25]

Although there is no literature comparing SLA versus DLP technology in the construction of provisional crowns, studies comparing these technologies for the printing of 3D dental models are abundant. A study by Kim et al.[26] demonstrated that the models printed by SLA technique were more accurate in terms of measurement of teeth the arch as compared to the DLP technique; however, DLP was superior in precision. As the printers with SLA technology complete one layer by curing the resin point by point via laser projection, the slow space of the mirror reflecting the beam of laser is bound to generate the error. On the other hand, DLP technology is faster because it employs a projector to cure the material layer by layer, reducing the inaccuracy that comes with repetitive printing. When we examined the two processes, we found that the SLA technology, which uses a lower x-y resolution and thinner layer thickness, was more exact than the DLP technique, although it was less precise due to variations in the manufacturing technique. A recent systematic review[27] also demonstrated that models printed by SLA technology were more accurate but had a wider range of mean errors.

CONCLUSION

In this study, we did a comparative evaluation of marginal adaptation and surface roughness of SLA versus DLP 3D printed commercially available crown and bridge provisional restorative material. There was no difference in surface roughness between both techniques. Furthermore, the temporary restorations printed via SLA showed a significantly less marginal gap as compared to DLP ones. Future work utilizing the same methodology can be carried out clinically to provide a definitive protocol.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.