Dimensional stability of 3D-printed fixed prosthetic restorations

Aim The aim of these in vitro study was to compare the accuracy and stability of four 3D printed resin full arch restorations at 24 hours and 7 days from fabrication.

Materials and methods A reference full arch gypsum model was scanned with Aadva lab scanner. A complete maxillary arch restoration was designed using the CAD software “Exocad (Exocad GmbH, Germany, 2010)”. After the designing procedure, the STL file was exported to the DLP printer. Four biocompatible (II class) resins developed for temporary crown and bridges have been chosen to realize the prosthetic restorations. Eight models were printed for each resin  using the DLP printer Asiga MAX UV (wavelength= 385nm; pixel resolution = 62 µm). A total of 32 printed restorations were included in this study. The surface was lightly dusted with powder (Occlusal Spray,Larident) and then were scanned using the lab scanner “Aadva Lab Scanner 2” to generate STL files. The scans were performed in four different moments: 1. Time = 0: scans are done immediately after printing; 2. Time = 24h: scans are done 24 hours after the first scan; 3. Time = 48h: scans are done 48 hours after the first scan; 4. Time = 7days: scans are done 7 days after the first scan. Eight models for each resin were scanned in four different moments, for a total of 32 scans for each resin and 128 total scans. The STL files were exported to a surface matching software (Geomagic Control X; 3D systems, Rock Hill, SC). The scan at T=0 of each model was taken as the reference model and the 24h scan was superimposed to the refence model. After the superimposition, the “3D compare” function was used to create color surface maps. A maximum critical value of ± 100 mm (0,100mm) and a maximum nominal value of ± 25 mm (0,025mm) was set for color spectra. Each superimposition allows to obtain two percentage (%) values: correspondence and variation. Thanks to these two values it was possible to quantify the behavior of tested prosthetic restorations over the first day (T=24h), the second day (T=48h), and the seven following days (T=7days) since printing. In order to analyse each resin, the two % values (correspondence and variation) of each 3D printing were collected and their arithmetic mean was calculated to have one % for each resin in each of the four times. One-way ANOVA was used for comparing different measurement errors between groups.

Results For all 3D-printed resins a change of stability was recorded; these changes were affected by time and type of resin. There were statistically significant differences among the four resins after 24 hours and 7 days.

Conclusions Within the limitations of this study, the results of the present research rejected two the null hypothesis as the time factor and different resins had effect on the stability of the 3D printed resin restorations. The Tempt PRINT resin showed do be more stable than the other three tested resins after 1 and 7 days. The variance in stability of FreePrint, C&B MFH and Temp PRINT was not statistically significant between 1 and 7 days, whilst VarseoSmile Crown showed a statistical significance change at the two times controls.