In vitro evaluation of the effect of auxiliary geometric device on measurement trueness and scanning time in full-arch implant impressions

Abstract

PURPOSE

This in vitro study aimed to evaluate the effect of clinically practical auxiliary geometric devices (AGD) on measurement trueness and scan time in full-arch implant-supported prostheses, focusing on different intraoral scanners (IOS) and implant angulations.

MATERIALS AND METHODS

Four implants were planned in an edentulous maxillary arch and divided into two groups based on posterior implant angulation: Model A (Parallel) and Model B (30°). Each model was evaluated under three auxiliary geometric devices (AGD) application types (std, agd1, agd2), and scanned using three different intraoral scanners (IOSs) [3Shape Trios 3 (T), Medit i700 (M), and Cerec Primescan (PS)], resulting in nine groups per model (n = 10). Scanning times and faulty scans were recorded. A conventional impression (CON) was taken as a control and digitized with a desktop scanner. Trueness was analyzed using Geomagic Control X, with statistical significance set at P < .05.

RESULTS

Group B had higher RMS values than Group A (40.3 µm vs. 34.7 µm). T_std in Group A (51.4 µm) and T_std (53.1 µm) and Con (50.7 µm) in Group B exceeded the acceptable deviation limit. AGDs reduced deviations to acceptable levels in the Trios 3. The Primescan scanner had the shortest scanning times. AGD use, especially in T_agd2 and M_agd2, shortened scan times and eliminated erroneous scans.

CONCLUSION

The use of AGD has a significant impact on scanners’ scanning trueness and time.

INTRODUCTION

With the advancement of digital technologies in dentistry, the use of intraoral scanners (IOSs) has become increasingly common in clinical practice.¹ The advantages of these scanners include enhanced patient comfort, reduced clinical and laboratory steps, improved cross-infection control, increased time and material efficiency, and the elimination of impression and plaster distortions.²,³,⁴,⁵

According to the International Organization for Standardization (ISO), trueness is defined as the degree of closeness between the arithmetic mean of multiple test results and the true or accepted reference value.⁶,⁷ Trueness can be analyzed based on root mean square (RMS) values calculated through three-dimensional comparisons.⁷ The trueness of digital implant scans is of great importance, as it significantly affects the outcome of dental implant restorations.⁸,⁹ Failure to accurately transfer the three-dimensional position of the implant into the CAD software may lead to biological and mechanical complications in the final restorations.¹⁰

Studies have shown that the trueness of IOSs in short edentulous span or single-unit implant-supported prosthesis measurements is clinically acceptable.¹¹,¹² However, the use of IOSs in multi-unit or full arch implant-supported prostheses is controversial.⁵,⁹,¹³,¹⁴ This may be due to two main reasons. Firstly, due to the limited field of view of the scanner, multiple images must be captured and subsequently stitched together by the software algorithm to construct a complete virtual model, which may lead to increased cumulative error.¹⁵,¹⁶ Secondly, the presence of translucent, mobile soft tissue in long edentulous areas and the lack of anatomical landmarks to guide the scanner may reduce trueness.¹⁷,¹⁸ In addition, scanner type, scanning protocol, implant distribution, inter-implant distance, the presence of angulated implants, and operator experience are also among the key factors that influence the accuracy of full-arch implant scans.¹⁹

To minimize the effects of these errors, various methods and devices have been developed. Studies have investigated the impact of techniques such as scanning the palate, splinting scan bodies, fabricating custom scan bodies, designing different auxiliary geometric devices (AGD), and painting the palate on the trueness of IOS. However, the results remain controversial.⁶,¹³,²⁰,²¹,²²,²³,²⁴,²⁵,²⁶,²⁷,²⁸ The effectiveness of AGD, which is used to compensate for anatomical deficiencies in long edentulous spans, varies depending on factors such as the geometry, surface characteristics, optical reflectivity, its proximity to the crest, and the scanning technology employed.¹³,²³,²⁹ These proposed methods and materials require additional steps, including the separate fabrication of AGDs or labor-intensive and time-consuming clinical procedures for case-specific adjustments.⁶,²⁷,³⁰ Moreover, these devices and techniques have generally been evaluated only in terms of scanning trueness, while their potential to reduce scanning time or minimize scanning errors has largely been overlooked.

The aim of this study was to evaluate the effect of clinically applicable, cost-effective, and universally adaptable AGDs on the trueness and scanning time of full-arch implant impressions obtained using different IOSs and implant angulations. Additionally, the study compared the trueness of digital impressions with that of the conventional splinted open-tray technique.

The null hypotheses, H1; there is no difference in the trueness of full arch implant digital impressions using different intraoral scanners and AGD. H2; implant angulation has no significant effect on the trueness of full arch digital and conventional impressions. H3; there is no effect of different intraoral scanners and AGD usage on the scanning time of full arch implant digital impressions.

MATERIALS AND METHODS

In this in vitro study, two main models were designed using SolidWorks^® (SolidWorks^® Corp., 2013, Waltham, MA, USA), and STP files were generated (Model A/B). Digital scan bodies (Bilimplant, Istanbul, Turkey) were positioned parallel to the 13, 23, 16, and 26 tooth locations in Model A. In Model B, the scan bodies at the 13 and 23 tooth positions were placed parallel, while those at the 16 and 26 positions were angulated 30° distally (Fig. 1). Subsequently, STL files were obtained.

Fig. 1. The jaws modeled in SolidWorks are; (A) model A, (B) model B.

The STL files were imported into Exocad, where digital analog placement was completed. The models were then fabricated using a digital light processing (DLP) based 3D printer (Asiga MAX, ASIGA, Sydney, Australia). After printing, the analogs and scan bodies were manually secured into the models. To minimize potential impression errors, the surrounding structures were stabilized using pattern resin.

To obtain the primary data for comparing the trueness of full-arch implant-supported impressions, a desktop 3D scanner (3Shape D2000; 3Shape A/S, Copenhagen, Denmark) was used. The scanner was calibrated according to the manufacturer’s instructions. Each model was scanned five times, and the reference dataset was randomly selected from these scans.

During scanning, two different AGDs were used to compensate for the lack of anatomical reference in the edentulous spaces between the scan bodies. For the first AGD, artificial acrylic teeth (Eray Deluxe, Ankara, Turkey) were selected, as IOS are optimized for scanning tooth morphology, and these materials are easily accessible to dentists in clinical settings. For the second AGD, a custom material was designed using SolidWorks with dimensions of 60.0 × 46.5 × 1.5 mm (Fig. 2). Small semi-circular rings were incorporated into the design to fit over the scan bodies. Additionally, indentations and protrusions were added to the surface to enhance scanning trueness.²⁹ To determine the most suitable printing material, preliminary tests were conducted using polymethyl methacrylate (PMMA), polylactic acid (PLA), polyethylene terephthalate glycol (PETG), and thermoplastic polyurethane (TPU). Based on these tests, TPU filament (Porima TPU Flex, Porima Polymer Technologies) with a Shore A95 hardness was selected for 3D printing. The Artillery Sidewinder-X2 3D printer was used for fabrication.

Fig. 2. Specially designed auxiliary geometric device (agd2): (A) lateral view showing the thickness (x = 1.5 mm), (B) top-sectional view indicating the width (y = 60 mm) and height (z = 46.5 mm), and (C) semi-circular ring structure designed to fit over the scan bodies.

Three different application types were defined according to the use of AGD: scans performed without AGD (std), scans using an artificial acrylic tooth as a reference (agd1), and scans using a specially designed auxiliary geometric device (agd2) (Fig. 3).

Fig. 3. Representation of the three application types defined according to the use of AGD: (A) standard scanning group (std), (B) scanning with an artificial acrylic tooth reference (agd1), and (C) scanning with a specially designed device.

Initially, digital scans of both models were performed using three different IOS (3Shape Trios 3, Medit i700, and Primescan) and with three different AGD configurations (std, agd1, and agd2) (n = 10). For each combination of scanner, model (Model A and Model B), and AGD application type, 10 scans were taken, resulting in a total of 180 digital scans. Subsequently, conventional impressions of the same models were taken using the splinted open-tray impression technique (n = 10), resulting in a total of 20 stone models, which were then digitized. The workflow used in this study is illustrated in Figure 4.

Fig. 4. Study groups.

For the IOS group, 3 different scanners were used: Trios 3 (3Shape Trios A/S, Copenhagen, Denmark; T), Medit i700 (Medit, Seoul, Korea; M) and Primescan (Dentsply Sirona Dental, Bensheim, Germany; PS). The specifications of the scanners are given in Table 1.

Table 1. The IOSs used in the study.

Scanner	Manufacturer	Scanning technology	Images per second	Imaging system	Software version
Trios 3 (T)	3Shape, Copenhagen, Denmark	Confocal Microscopy	3000	Ultrafast and Multimodal Imaging, Photographic	23.1.3
Medit i700 (M)	Medit, Seoul, Korea	Active Triangulation	FPS: 70	Stereoscopic Vision, 3D Motion Video Technology, and 3D Full-Color Stream Capture	3.3.2
Primescan (PS)	Dentsply Sirona, Bensheim, Germany	Active Triangulation	50.000	Video Recording, Smart Pixel Sensor	5.2.7

All scans were performed under identical environmental conditions and by the same operator (S.İ), a prosthodontist with three years of experience in the use of intraoral scanners, following the protocols recommended by the respective IOS manufacturers. To standardize experience across devices, the operator conducted 10 trial scans for each scanner before starting the actual measurements. The scans started from region 16 and proceeded sequentially toward region 26. Before each scan, the device was recalibrated. Throughout the scanning process, the scan bodies were not removed.

After each scan, the scanning duration was recorded. The scanning time was measured from the start of the scan until its completion. Any post-processing of the scan data was not included in the calculation of scanning time.

Errors such as image drift or overlapping artifacts during scanning were documented, and these scans were discarded. Before finalizing the scans, it was ensured that all surfaces of the scan bodies were properly captured and that the intended scanning area was fully covered. Once verified, the scans were saved, and STL files were exported.

Following the completion of digital impressions, the scan bodies were removed from the models and replaced with open-tray impression copings. The copings were splinted using pattern resin (GC America, Chicago, IL, USA). After polymerization, the resin splint was sectioned between each coping using a separating disc to minimize distortion. Following verification of individual coping stability, the sections were reconnected with additional resin applied with a brush, ensuring a rigid and unified splint. A minimum setting time of 20 minutes was allowed prior to the impression procedure.³¹ Conventional impressions were made using the open-tray impression technique with Zhermack Elite HD+ Fast Set polyvinyl siloxane (Zhermack, Italy) (n = 10). Digital scan bodies were placed on the stone models, ensuring that their positions matched those in the intraoral scanning models. For comparative measurements, the stone models were scanned using a 3Shape D200 scanner, and the STL files were generated.

In this study, trueness was analyzed based on RMS values calculated through three-dimensional comparisons. All STL files were imported into Geomagic Control X 2020 (3D Systems, Rock Hill, SC, USA) for three-dimensional analyses and measurements. The registration process was performed with the ‘best fit algorithm’. Firstly, the buccal parts of each scan group were used for registration. In the agd2 group, the upper open parts of the scan bodies were selected due to the AGD. The selected parts were superimposed using the ‘Use Selected Data Only’ and ‘Use Only Reliable Measured Data’ options (Fig. 5). After overlapping, RMS values were obtained automatically. Based on previous studies in the literature, clinically acceptable threshold for RMS values has been defined as 30 − 50 µm.¹³,³²

Fig. 5. The ‘best fit alignment’ method used for aligning the obtained scans and the buccal surface taken as the reference during alignment (A). To ensure standardization, the visible scan body area was selected in all scans (B).

The data were analyzed using IBM SPSS V23 and the WRS2 package in R software. The Shapiro-Wilk test was used to assess the normality of the data distribution. For variables that followed a normal distribution based on model and group, a two-way analysis of variance (ANOVA) was applied. For variables that did not follow a normal distribution, a two-way robust ANOVA was performed. Multiple comparisons were conducted using the Bonferroni test. The results were presented as mean ± standard deviation and median (minimum – maximum). A significance level of P < .050 was considered statistically significant.

RESULTS

The RMS values for all groups are presented in Table 2.

Table 2. Comparison of RMS values according to models and groups.

Group	Model		Total		Q	P
	A	B
T_std	51.4 (34.3 - 69.6)	53.1 (33.2 - 66.9)	52.3 (33.2 - 69.6)a	Model	4.074	0.043
T_agd1	43.3 (30.8 - 57.1)	40.6 (30.7 - 66.6)	41 (30.7 - 66.6)ab	Group	3.462	< .001
T_agd2	38.6 (29.6 - 51.3)	40 (30.8 - 55.5)	39.3 (29.6 - 55.5)ab	Model*Group	7.005	0.637
M_std	24 (19.2 - 37.3)	49.8 (33.1 - 98.7)	35.2 (19.2 - 98.7)abc
M_agd1	32.9 (22.1 - 52.7)	35.6 (21 - 43.3)	33.8 (22.1 - 52.7)bc
M_agd2	24.8 (20.3 - 44.4)	23.8 (16.3 - 39.4)	24.3 (16.3 - 44.4)c
PS_std	31.1 (25.3 - 49.4)	40.2 (33.8 - 54)	38.5 (25.3 - 54)b
PS_agd1	30.1 (27 - 39.6)	34.1 (23.5 - 42.6)	32 (23.5 - 42.6)bc
PS_agd2	36.8 (29.4 - 42)	37.6 (30.7 - 47.6)	36.8 (29.4 - 47.6)b
CON	40 (26.6 - 81.7)	50.7 (28.3 - 58.8)	46.5 (26.6 - 81.7)ab
Total	34.7 (19.2 - 81.7)	40.3 (16.3 - 98.7)	938.4 (16.3 - 98.7)

Median (minimum – maximum); a – c: There is no difference between the main effects with the same letter.

The main effect of model was statistically significant on the RMS median values (P = .043). While the median value in Model A was 34.7 µm, the median value in Model B was 40.3 µm. Group main effect was statistically significant on RMS median values (P < .001). The highest RMS median value was obtained in the T_std group of Model B (53.1 µm), while the lowest median value was obtained in the M_agd2 group of Model B. Model and group interaction was not statistically significant on RMS median values (P = .637). Considering the RMS median values, the clinically unacceptable deviation values (> 50 µm) were shown by T_std (51.4 µm) in Model A, and T_std (53.1 µm) and Con (50.7 µ m) in Model B (Fig. 6). It was found that the used AGDs reduced the amount of deviation in the T scanner to an acceptable level.

Fig. 6. The box plot showing the RMS values of Model A and B. The clinically acceptable threshold (50 µm) is indicated by the red line.

The comparison of scanning times obtained according to the models and groups is provided in Table 3.

Table 3. Comparison of scanning time values according to models and groups.

Group	Model		Total		Q	P
	A	B
T_std	90.5 (83 - 100)AB	103 (95 - 118)İ	97.5 (83 - 118)a	Model	28.452	< .001
T_agd1	79 (63 - 86)ACDE	96 (80 - 107)ABİ	83 (63 - 107)b	Group	83.688	< .001
T_agd2	64.5 (54 - 78)CDF	74.5 (61 - 83)CDE	70.5 (54 - 83)c	Model*Group	32.284	< .001
M_std	82.5 (76 - 87)ACE	96.5 (87 - 109)Bİ	87 (76 - 109)ab
M_agd1	78 (65 - 80)CDE	86 (72 - 97)ABE	79.5 (65 - 97)b
M_agd2	69 (63 - 72)D	73.5 (69 - 77)CDE	70.5 (63 - 77)c
PS_std	47.5 (40 - 56)FGH	54.5 (48 - 60)FH	51 (40 - 60)d
PS_agd1	45 (39 - 48)GH	44.5 (40 - 55)GH	45 (39 - 55)e
PS_agd2	39 (32 - 47)G	45 (42 - 51)GH	43.5 (32 - 51)e
Total	69.5 (32 - 100)	76 (40 - 118)	72 (32 - 118)

Q: Robust ANOVA; Median (minimum – maximum); a – e: There is no difference between interactions with the same letter.; A – İ: There is no difference between interactions with the same letter.

When the scan time values were analyzed, the main effect of model was found to be significant (P < .001). The median value was 69.5 seconds in Model A and 76 seconds in Model B. While the use of AGD shortened the scan time in all 3 scanners, this effect was significant in T_agd2 and M_agd2. In both models, the 3 all scan obtained with of the PS scanner gave significantly lower scan times than the T and M scanners.

In both models, 3 faulty images were obtained in the T_std and M_std groups (Fig. 7). In Model B, one faulty image was obtained in the PS_std group. After using both AGDs, the number of faulty scans dropped to 0.

Fig. 7. An example image of scanning errors caused by image stitching during digital scanning.

DISCUSSION

Based on the findings of this study, the first null hypothesis, which stated that different intraoral scanners and various AGDs do not significantly affect the trueness of full-arch implant-supported digital impressions, was not rejected. Although Primescan demonstrated higher trueness compared to the other scanners, the effect of AGD usage was not statistically significant. However, it was considered clinically relevant, as it reduced deviation values to acceptable levels, particularly in cases involving angled implants. The second hypothesis was rejected due to the significantly higher RMS values observed in models with angulated implants. The third hypothesis was also rejected, as different intraoral scanners and the use of AGDs were found to have a significant impact on the scanning time of full-arch implant-supported digital impressions.

The jaw models were designed using SolidWorks and Exocad software to ensure accurate placement of implants at the desired angulations and parallelism prior to 3D printing. The symmetry of the jaw models, the consistent anterior implant positions, and the identical soft tissue emergence locations of the posterior implants with different angulations contributed to standardization in the comparative study.

In this study, two clinically practical methods were employed to compensate for the lack of anatomical reference in full-arch digital impressions taken with IOSs and to facilitate the stitching process. Agd1, which aims to eliminate the lack of anatomical reference between the scan bodies, was preferred because it is easily fixed to the edentulous crest with tissue adhesive and the clinician has easy access to the acrylic tooth. The agd2, which aims to create an optical bridge between the scan bodies and is specially designed and manufactured for this purpose, offers advantages such as easy attachment to the scan bodies due to its flexible material structure, adaptability to scan bodies with different diameters and geometries, and adaptability to differently positioned implants with its flexibility feature. Additionally, when compared to other scan body splinting methods reported in the literature,²⁴,²⁷,³³ this approach can be considered a practical alternative for clinical applications, as it significantly reduces the chairside clinical procedure time. Most studies in the literature recommend taking a second impression and using case specific AGDs.²⁰,²³,²⁷,²⁹,³⁰ In this study, the aim was to evaluate the effect of clinically practical AGDs, which are universally applicable rather than case-specific, on the trueness of full-arch implant impressions and scanning time. Both AGDs were designed to be adaptable to scan bodies of different positions and diameters, ensuring ease of clinical application.

It is known that the most accurate and precise impression technique for cases to be treated with full arch implants is the open tray impression method using type A silicone splinted with pattern resin.²⁶,³⁴ Additionally, many studies comparing digital impression trueness have included conventional impressions taken with the splinted open-tray technique as a reference for comparison.²⁶,³³,³⁵,³⁶ In this study, as recommended in the literature,³⁷ conventional impressions were taken using the open-tray technique with splinting.

In many studies,¹⁴,²² the “best-fit algorithm” has been used to superimpose reference and test scans. This methodology allows for the calculation of point-to-point distances between the test surface and reference scans. Once the best-fit algorithm is initiated, inconsistencies between the reference and test data can be automatically computed, and the RMS value can be generated. In general, a lower RMS value indicates a higher degree of alignment between the superimposed virtual models.²⁶

This study is in accordance with the studies which reported that implant angulation has a significant effect on impression trueness.³⁸,³⁹ The presence of implant angulation can reduce the scanner’s accessibility and scanning angle, thereby impairing its ability to accurately digitize the geometry of the scan bodies. In the conventional measurement technique, although not statistically significant, higher RMS values were found in model B compared to model A and these values exceeded the clinically acceptable limit values in model B. These results are consistent with studies reporting that implant angulation reduces the trueness of conventional impressions.⁴⁰,⁴¹,⁴² The obtained data suggest that digital impression techniques may offer a more advantageous option in cases involving angulated implants, particularly when appropriate scanner selection and the use of AGDs are applied.

In the comparison among std groups, Primescan demonstrated significantly higher trueness than Trios 3. The differences in scanning technologies, imaging systems, and algorithmic processing among scanners can lead to significant variations in trueness. The results of this study are consistent with previous research indicating that the Primescan scanner provides more accurate results than the Trios 3.⁵,³³,⁴³

An average RMS value between 30 and 50 µm has been recommended to prevent biological and mechanical complications.³² The reduction of previously exceeded RMS values to below 50 µm with the use of AGDs is considered to offer a clinical advantage (Fig. 6). Although no statistically significant difference was observed across all scanners, the use of AGD reduced deviations exceeding the clinical threshold in both models for the Trios 3 and in Model B for the Medit i700. Additionally, a decrease in RMS values was noted for all scanners in Model B. The distal angulation of the implants may have enabled the intraoral scanner to capture a broader view of the mesial surface of the scan body, potentially increasing its visible surface area. This increased visibility might have supported the simultaneous capture of the occlusal and axial surfaces of the scan body, along with the AGD, which could have facilitated more accurate digital alignment. Consequently, this may have helped reduce possible errors during the image stitching process within the digital workflow. In addition, implant angulation can increase the edentulous span between scan bodies, which may lead to greater error accumulation during the scanner’s data processing phase.³⁸,⁴⁴ The additional reference points provided by AGDs in these edentulous areas may support the stitching process and help reduce potential error accumulation.

In an in vitro study by Pozzi et al.,⁶ scan bodies were scanned with an intraoral scanner (Trios 3) in two subgroups as splinted and non-splinted using AGD in a model with 4 implants with distal 17° and 30° and it was found that the use of AGD positively affected the accuracy of the full arch digital impression by reducing linear and angular deviations at the most critical posterior implant positions. In another study, the effect of artificial landmark and scan body splinting on the impression accuracy of a mandibular model with 6 parallel implants was investigated on different scanners (TRIOS 4, Virtuo Vivo, Medit i700, iTero Element 5D and Primescan). The Primescan scanner gave better results for all techniques than the other scanners. Using artificial landmarks or splinting the scan body did not affect the accuracy of digital scans with the Virtuo Vivo, Medit i700 and Primescan, but splinting significantly improved accuracy with the Trios 4 scanner.⁴⁵

The differences observed in scanning times among scanners can be attributed to variations in scanning technologies and algorithms.⁴⁶ Trios 3 operates using a confocal microscopy-based imaging principle combined with ultra-fast optical sectioning, enabling the acquisition of up to 3,000 images per second. Medit i700 employs an active triangulation system with a maximum frame rate of 70 frames per second. Primescan utilizes an advanced active triangulation technology combined with confocal microscopy, integrated with a smart pixel sensor capable of capturing 50,000 images per second and performing deep scans with enhanced depth resolution.⁴⁷ Additionally, the image area captured per scan may also contribute to these differences. The scanning times with Primescan were found to be significantly faster compared to the other scanners. This suggests that Primescan may be the most suitable option for situations where fast scanning is required. With the Primescan, a larger area can be scanned at one time compared to Medit i700 and Trios 3. The larger field of view of the scan window can minimize the error from image stitching by scanning larger areas.⁴⁸ Dönmez et al.⁴⁹ reported that Primescan has higher time efficiency than Trios 3.

An increase in scanning time was observed in the T_std and M_std groups due to implant angulation. This finding may suggests that the undercut areas created by the inclined surfaces of angulated implants required additional scanning efforts. The use of AGDs was observed to eliminate the effect of angulation on scanning time. Specifically, the use of agd2 significantly reduced scanning times for both the Trios 3 and Medit i700 scanners across both models. This may indicate that the continuous surface morphology of agd2, which provides uninterrupted reference points between scan bodies, is more effective in the stitching process compared to complex tooth morphologies (agd1).

In multiple implant scans, the scanner may fail to recognize scan bodies due to their similar morphology, leading to incorrect image stitching (Fig. 7). The use of AGDs, particularly with the Medit i700 and Trios 3, reduced image stitching errors, thereby minimizing the need for rescanning by the clinician.

The reduced scanning time is particularly noteworthy, as it may minimize the influence of patient-related factors on trueness by reducing the risk of saliva contamination and patient movement.⁵⁰ These findings suggest that AGDs can enhance clinical workflow efficiency by shortening chairside scanning time and decreasing the need for rescanning in complex cases.

In this study, the Primescan scanner demonstrated the most successful results among all devices, standing out with low RMS values (< 50 µm), short scanning times, and the fewest scanning errors (only one faulty scan in Model B). This superior and consistent performance may be attributed to Primescan’s advanced technological features. Moreover, its performance was not affected by either implant angulation or the use of AGD, suggesting that Primescan can operate reliably even in complex clinical scenarios.

Consistent with Garcia-Martinez et al.,²⁸ AGD use did not significantly affect Primescan’s trueness; however, unlike their findings, no significant reduction in scanning time was observed in our study. This discrepancy may be due to methodological differences such as scan body geometry, implant distribution, AGD design, or scanner software versions. Overall, these findings suggest that Primescan is a reliable and efficient option for full-arch implant scanning, even without the use of auxiliary devices.

In full-arch restorations with immediate loading, the successful and timely delivery of the prosthesis is directly dependent on a highly accurate impression-taking process. Any error during this stage may disrupt the treatment workflow or even prevent the placement of the final restoration. The use of AGD in a fully digital workflow may offer certain advantages. In particular, agd2 is hypothesized to reduce light reflection by creating a matte surface between scan bodies, especially when it partially covers bleeding or translucent soft tissues, thereby potentially facilitating intraoral scanning.¹³ However, this possibility remains theoretical, and prospective clinical studies are needed to validate the effectiveness of AGD use under immediate loading conditions.

One of the limitations of this study is the use of only four implant systems and a single angulation variation. Evaluating different implant numbers, angulations, and orientations could help more comprehensively determine the trueness and limitations of both IOS and AGD. Additionally, studies conducted with a broader range of scanners and different scanning protocols would better elucidate their contribution to impression accuracy and precision. In this study, only the implant-level impression technique was utilized, and evaluating abutment-level impression techniques could enhance the generalizability of the results. Moreover, the evaluation of only trueness data in this study represents another limitation. To enable a more comprehensive interpretation of the results, the analysis should be supported by the inclusion of precision measurements as well. Although the RMS methodology used may overlook certain details, such as linear and angular displacements, it intuitively reflects overall inconsistencies. Future studies should incorporate additional analytical methods to determine in which plane and which region of the arch the deviation are more pronounced. Finally, due to the in vitro design of this study, intraoral factors (e.g., saliva, patient movement, soft tissue dynamics, and ambient light) were not considered. Therefore, the trueness and precision of scanners and AGDs in clinical settings need to be validated through in vivo studies.

CONCLUSION

This in vitro study demonstrates the potential of AGDs to improve the trueness and efficiency of full-arch digital implant impressions, particularly when using intraoral scanners such as Trios 3 and Medit i700. Although the Primescan exhibited consistently high performance without the use of AGDs, the benefits observed in scanners more sensitive to anatomical variables and implant angulation suggest that AGDs may enhance the trueness and reliability of digital impressions in such cases. In complex clinical scenarios involving tilted implants and extensive edentulous spans, AGD-assisted digital workflows may serve as a viable alternative to conventional impression techniques. These findings underscore the importance of appropriate scanner and device selection in full-arch prosthodontic rehabilitation and highlight the need for further in vivo studies to evaluate the clinical applicability of AGDs, supporting the integration of fully digital workflows in implant prosthodontics.