Accuracy Evaluation of Implant Impressions: A Prospective Study

The aims of this clinical study are: 1) to compare the accuracy of conventional versus digital impression techniques for implant impressions; 2) to assess the passive fit of the prosthesis constructed from the most accurate impression.

Study design The ethical approval will be obtained from the research ethics board. Thirty edentulous patients who received 4 - 6 implant and indicated for implant-supported fixed prosthesis will be invited. All participants will require to read and sign the informed consent form before initiation of the treatment. The impression procedures will be performed by one experienced prosthodontist. The prostheses will be evaluated by an independent prosthodontist.

Sample size calculation The sample size is calculated based on the tests for paired means. With α= 0.05, β= 0.2, mean of distance deviation, IOS= 17 μm, and standard deviation= 34 μm, a sample size of 25 is necessary, which would have the power of 90% to detect the difference. Therefore, a total of 25 patients will be invited to participate in this clinical study. This sample size is more than previous studies evaluating IOS and SPG workflows (15 - 17 patients). However, it has been reported that the dropout rate in clinical studies is 20%. For this reason, a total of 30 patients will be invited to compensate for the anticipated loss of patients.

Treatment procedure Conventional impression Abutment-level impression copings will hand tightened onto the abutments and splinted with autopolymerizing acrylic resin (Pattern Resin; GC), and allowed to polymerize. A custom open-tray splinted impressions will be made with a polyether material (Impregum Penta Soft; 3M ESPE). After 5 minutes, the impression copings will be unscrewed from the implants, and the impression will be removed. Type IV dental stone (dentostone 220; dentona AG) will be poured into the impression to generate a definitive cast. The models will be digitalized with a high-resolution laboratory scanner (E4; 3Shape A/S) with an accuracy of 4 mm to obtain the 3D data of the models in standard tessellation language (STL) format.

Digital impressions After tightening the scan bodies (CARES® RC Mono Scan body, Straumann, Basel, Switzerland), the intraoral scanner will be used according to the manufacturer's recommendations. Three scanners will be used since they have different ways of obtaining 3D images and working principles. The technology of Trios scanner (TRIOS 4, 3Shape A/S) depends on confocal microscopy, ultrafast optical scanning and auto-correction; Cerec Primescan (Dentsply Sirona) depends on optical triangulation and confocal microscopy; Medit i700 (Medit, Seoul, Korea) depends on 3D in-motion video technology, 3D full-color streaming capture and advance algorithm to stitch images. The scanning path will begin with the occlusal surfaces, then turned to the palatal or lingual surfaces, and lastly to the buccal surfaces of the scan bodies. Then, removing the scan body with a trim tool from the occlusal view and exporting the STL file.

Accuracy measurement All STL files generated from the intraoral scan (STL digital) and the STL files generated from cast digitization (STL conventional) will be imported into 3D inspection software (Geomagic, 3D Systems Inc., Rock Hill, SC) for merging and evaluation of 3D deviations. Following software best-fit alignment, the 3D comparison will be used to evaluate the 3D deviations between the digital files (STL files from IOS and digitized conventional working casts) limited to the area of scan bodies. Assessment of the root man square (RMS) deviations will be the primary outcome. The effect of the type of edentulous jaw (maxilla vs mandible) on the 3D deviations will be the secondary outcome, assessing whether the jaw has an effect on the amount of 3D deviations.

Prostheses fabrication The final restorative phase included definitive with one-piece, screw-retained zirconia implant-supported prostheses. The prosthesis will be constructed from the most accurate workflow after confirmation of the passive fit at the try-in appointment. In addition, the fit of the zirconia implant-supported prostheses will be confirmed prior to delivery, clinically and radiographically.

Statistical analysis Statistical analysis will be performed by SPSS version 29 (IBM Corp, Armonk, NY). Data will be presented as mean, median, and standard deviation. The normality of data will be tested by the Shapiro-Wilk test. The generalized linear mixed model will be used to compare the influence of scan bodies and impression copings on the accuracy of digital and conventional impressions. The fixed factors will be the 'IOS', 'type of scan body', 'type of impression coping' and their interaction. The precision will be calculated using the gamma distribution with the log-link function. The p values will be adjusted by the Bonferroni method for pairwise comparison (>.05).

All clinical outcomes will be assessed by one examiner. Since some patients will received more than one implant, linear mixed effect models (LMM) will be fitted using restricted maximum likelihood estimations to assess the effect of the follow-up time on repeated measurements of marginal bone loss, plaque index, calculus index, gingival index, bleeding index, and probing depth while accounting for within-person variability. LMM calculates means with standard errors as it estimates variability across multiple samples of a population (i.e., the different patients, implants, and repeated measures). The multivariable models included the fixed effects of the baseline value and follow-up in months. The interaction between the baseline and follow-up value (baseline value*follow-up) will be tested for model improvement. The included random effect will be the patient (i.e., random intercepts). The random effect of the follow-up (i.e., random slopes) will be also tested. Statistical comparisons will be performed using the type III analysis of variance test with Satterthwaite's method to estimate the degrees of freedom. A significance level of p< 0.05 will be chosen for all the analyses.