Keywords
degree of deviation - All-on-4 - trueness - intraoral scan - precision - accuracy
Introduction
Implant impression represents a critical step in the fabrication of an implant-supported
fixed dental prosthesis. This procedure necessitates the accurate transmission of
the implants' position and angle within the mouth, which is essential for achieving
optimal treatment outcomes.[1]
[2] Impression making can be executed through two primary methods: the conventional
method, which employs elastomeric impression materials to record implant positions
via a physical cast, and the digital approach, which utilizes optical scanning technologies,
including intraoral scanners (IOSs) and laboratory-based scanners (LBSs). Regardless
of the chosen method, the objective remains to accurately transfer the intraoral position
of dental implants to either a physical working cast or a digital virtual model.[2]
IOSs are sophisticated dental devices designed for capturing direct optical impressions.
They consist of a handheld camera (hardware), a computer, and specialized software.[3] These advanced devices utilize three-dimensional (3D) in-motion video imaging technology
alongside measurement systems to capture the shape and size of dental arches, thereby
reproducing 3D models of the teeth, soft tissues of the oral cavity, and implant scan
bodies.[4] The advent of such technologies has significantly transformed the field of dentistry,
particularly in the domains of prosthodontics and implantology, by markedly reducing
scanning time when compared with conventional impression methods, eliminating the
need for physical cast storage, and minimizing patient discomfort during the impression-making
process.[5]
While conventional impressions have long been regarded as the gold standard in clinical
dentistry, their usage has been accompanied by several challenges, including the necessity
for material preparation, susceptibility to impression distortion, high sensitivity
to technique, protracted clinical workflows, and increased patient discomfort.[6]
[7] Conversely, digital impressions present distinct advantages such as enhanced efficiency,
precision, and seamless integration with computer-aided design and manufacturing (CAD/CAM)
systems.[7] The IOS technology has found successful applications across various dental situations,
including fixed partial dentures, complete dentures, maxillofacial prostheses, and
implant dentistry.[5]
[8] In the context of implant-supported prosthetics, IOSs have improved processes spanning
from treatment planning to the fabrication of definitive prostheses.[9] This evolution has not only transformed the daily routines of both dental practitioners
and technicians, but has also facilitated a more streamlined and precise approach
to impression making.
Presently, a diverse array of scanners is available on the market, employing various
optical technologies, acquisition methods, and reconstruction algorithms.[10]
[11]
[12]
[13] IOSs predominantly utilize laser beam or structured light technology,[4]
[5] while LBSs incorporate three main technologies: laser beam, structured light, and
contact-based methods.[14] Over the last 10 years, there has been a significant rise in the accessibility of
optical IOSs, each based on different technologies that may influence the quality
of clinical outcomes.[15]
[16]
The exact positioning of dental implants is essential for the success and durability
of prostheses supported by implants.[17]
[18] The positioning of implants has a significant impact on biomechanical stability,
esthetic outcomes, and functional efficacy; thus, the accuracy of the impression process
is of utmost importance.[19] Within the domain of implant dentistry, the impression technique is essential for
effectively transferring the intraoral position and angular orientation of dental
implants onto the gypsum cast, which plays a vital role in the success of implant-borne
full-arch fixed prostheses, such as the All-on-4 therapeutic approach.[19] Achieving a passive fit of the prosthesis is crucial in full-arch cases[20] and is integral to preventing complications such as mechanical stress and eventual
implant failure.[10] While the literature reports that the acceptable range for passive fit varies between
10 and 150 µm, it is recommended that this range be confined to between 30 and 50 µm
to mitigate the risk of mechanical and biological complications.[21]
[22]
The process of obtaining optical impressions for All-on-4 edentulous arches poses
considerable challenges, particularly when dealing with tilted implants.[23] The lack of reference points (teeth) in completely toothless jaws makes the stitching
process challenging. Additionally, the presence of movable tissues and saliva, combined
with the pink hue of the soft tissue, further complicates full-arch imaging.[23] Additionally, factors such as the dimensions, colors, and design of the scanning
body, in conjunction with the positioning and accessibility of the implants, significantly
affect the accuracy of full-arch implant scanning.[24] Furthermore, the employment of various IOSs appears to have a substantial impact
on the fidelity of implant position documentation.[25]
Accuracy is defined as the “closeness of agreement between a measured quantity value
and a true quantity value of a measurand.”[25]
[26] This concept encompasses two primary components: trueness and precision, which,
while distinct, are complementary. Trueness pertains to the ability of a measurement
to closely align with the actual value of the quantity being assessed, whereas precision
reflects the consistency or repeatability of measurements taken from the same object.[10] For an IOS to function effectively, it must be capable of accurately capturing all
intricate details of the scan, producing a virtual 3D model that closely resembles
the reference model with minimal deviations. This is critical to ensure that the digital
impressions generated by the IOS are reliable and suitable for clinical applications.
In the present study, the authors prioritize trueness over precision, focusing on
evaluating the degree to which the IOSs can replicate the actual geometry of the scanned
object. The emphasis on trueness is warranted because it directly impacts clinical
outcomes, ensuring that digital impressions accurately reflect actual anatomical structures'
dimensions, which are vital for effective treatment planning, appropriate prosthesis
fit, and the minimization of errors that could result in complications.[27] While precision guarantees consistency in repeated measurements, it does not necessarily
ensure accuracy in reflecting true anatomical values; therefore, trueness is the more
pertinent metric for assessing the performance of IOS technology in this research.
Despite the critical importance of these aspects, limited studies have investigated
the accuracy of contemporary IOSs. Lots of existing research focuses on first-generation
devices and has not evaluated their performance in All-on-4 implant scenarios, highlighting
a significant gap in understanding the capabilities of newer IOSs. This study aims
to address this gap by conducting an in vitro investigation to assess the impact of various IOSs on the trueness of scanning in
All-on-4 implants situated in a completely edentulous mandible. The null hypotheses
(H0) formulated for this study postulate that: (1). there exists no statistically significant
difference in the degree of deviation of scan bodies' position for All-on-4 implants
between the various methods of impressions and (2) there is no statistically significant
difference in the dimensional discrepancy of the scan body zone for All-on-4 implants
between the various methods of impressions.
Materials and Methods
Model Preparation
A total of four SuperLine implants (4.5 × 10 mm; Dentium, South Korea) were placed
in an All-on-4 configuration by an experienced periodontist, complying with the manufacturer's
guidelines. The setup included two implants placed vertically in the canine region,
and two implants angulated distally in the second premolar area, all embedded in a
one-piece artificial jaw model (EasyinSmile, United States).
Grouping
The study involved two IOSs: the TRIOS 5 (3Shape, Copenhagen K, Denmark), designated
as T, and the Runyes 3DS 3.0 (Runyes, Zhejiang, China), referred to as R. Additionally,
a conventional silicone impression method was utilized, denoted as C. In total, 30
impressions were obtained, with n = 10 for each method.
Impression Taking
An initial reference scan was conducted on the artificial jaw using an LBS (KaVo ARCTICA
AutoScan, Kavo Dental, Germany), incorporating four scan bodies (4.0 × 10 mm, SuperLine,
Dentium, South Korea) securely attached to the implants ([Fig. 1]). All scan bodies were hand-tightened initially and then secured to the implants
using a calibrated torque wrench (Dentium) set to the manufacturer's recommended 15 Ncm.
[Table 1] summarizes the properties of the scanners used in this study. This scanning procedure
was executed by a certified technician in accordance with the manufacturer's protocols,
and the resultant data were exported in Standard Tessellation Language (STL) format.
After the reference scan, the scan bodies were left undisturbed to facilitate subsequent
impressions utilizing the two IOSs. Calibration of both scanning devices was performed
to ensure precision, followed by a standardized scanning protocol involving a one-way
sweep (buccal, occlusal, and lingual surfaces). A scan was deemed complete when it
thoroughly encompassed all significant features of the master model without notable
gaps and captured the entirety of the scan bodies. This process was meticulously replicated
10 times for each scanner type, strictly adhering to the manufacturer's specifications,
with output in STL format for further analysis.
Table 1
Properties of the scanners used in the study
|
Scanner
|
Runyes 3DS 3.0
|
TRIOS 5
|
KaVo ARCTICA AutoScan
|
|
Manufacturer
|
Runyes, Zhejiang, China
|
3Shape, Copenhagen K, Denmark
|
Kavo Dental, Biberach, Germany
|
|
Imaging principle
|
Optical continuous video collection
|
Blue laser and structured light technology
|
Structured light and laser triangulation
|
|
Light source
|
LED light source without radiation
|
LED as a light source
|
Striped light projection
|
|
Scan speed
|
90 s
|
Captures 2,400 images per second
|
1 min for a single stump as framework
3 min for a three-unit bridge
|
|
Accuracy
|
<20 μm
|
100 μm
|
20 μm
|
Abbreviation: LED, light-emitting diode.
Fig. 1 Reference STL scan.
In the conventional impression protocol, the scan bodies were extracted from the implants,
and open-tray impression copings (4.0 × 15 mm, SuperLine, Dentium) were installed
and hand-tightened at the implant level. The impression copings were then fastened
using a calibrated torque wrench (Dentium) adjusted to the manufacturer's suggested
setting of 15 Ncm. Despite existing literature recommending the open-tray splinting
technique and abutment level impression for angled implants to enhance full-arch accuracy,
this approach was deliberately not utilized. The intention was to replicate the digital
scanning conditions and to precisely identify the shortcomings of the silicone impression
technique without incorporating additional measures that could mitigate these imperfections.
A custom open-tray featuring self-curing acrylic resin (Vertex Trayplast, Vertex Dental,
The Netherlands) was fabricated; the resin was blended and shaped into a dough state
before curing. The final trays underwent refinement through grinding with a red acrylic
bur and polishing with a white stone bur. An addition silicone (Monophase, Take 1
Advanced, Kerr, United States) was then applied to the custom tray for the impression.
The tray was positioned over the artificial jaw, adhering to the manufacturer's specified
working time for complete polymerization prior to removal. After tray removal, implant
analogs (DANSE, SuperLine, Dentium) were secured to the impression copings. Light
body wash silicone (Take 1 Advanced, Kerr, United States) was poured around the impression
copings within the fitting surface until polymerization was complete as a soft tissue
replica, followed by the pouring of gypsum (Type V) to produce a master model. The
same custom tray was utilized repeatedly 10 times to yield a total of 10 master casts.
Then, the scan bodies were seated onto the implant analogs and scanned with the laboratory
scanner (KaVo ARCTICA AutoScan, Kavo Dental, Biberach, Germany), with each master
cast scanned individually and exported as an STL file. Ultimately, 30 STL scans were
produced, with n = 10 for each method.
Alignment and Trueness Analysis
Each scan was imported into Blue Sky Plan software (version 4.12, Blue Sky Bio, United
States) for alignment with the reference scan. This alignment was based on four predefined
reference points corresponding to the buccally oriented notch marks on each scan body
across all implants ([Fig. 2]). The aligned STL files were processed in Autodesk Meshmixer software (version 3.5.474,
California, United States) to assess the degree of deviation between the impression
scans and the reference scan. The dimensional analysis of the All-on-4 scan body zone
was conducted across the X, Y, and Z axes for all scans and the reference using Autodesk Meshmixer. Here, the X-axis represented the mesiodistal dimension, the Y-axis reflected the apico-occlusal dimension, and the Z-axis delineated the buccolingual dimension. This comprehensive evaluation included
a total of 90 readings, with 10 measurements per group across all dimensions for comparative
statistical significance against the reference scan.
Fig. 2 Alignment of the impression scan (right image) to the reference scan (left image)
using the four defined reference points corresponding to the notch marks on each scan
body.
Statistical Analysis
Normality of data distribution and variance equality were assessed through the Shapiro–Wilk's
and Levene's tests, respectively (p < 0.05). Statistical analysis was executed using Brown–Forsythe one-way analysis
of variance (ANOVA) along with Tamhane's post hoc tests to elucidate group differences
(p < 0.05). All statistical procedures were conducted using IBM SPSS version 22 (SPSS
Inc., Chicago, Illinois, United States).
Results
The dataset exhibited normal distribution; however, variances were found to be unequal
(p < 0.05). Consequently, Brown–Forsythe one-way ANOVA coupled with Tamhane's post hoc
analysis was employed for the evaluation of the data.
[Table 2] presents the descriptive statistics pertaining to the degree of deviation analysis.
Notably, the conventional impression yielded the lowest mean deviation (0.82 mm),
whereas the TRIOS 5 group exhibited the highest mean (1.11 mm). The readings from
the conventional impression group demonstrated the lowest precision, as indicated
by a standard deviation (SD) of ± 0.16. In contrast, the Runyes 3DS and TRIOS 5 groups
displayed comparable precision (SD ± 0.05 and 0.06, respectively). Significant differences
were observed among all groups, as detailed in [Table 3]. [Fig. 3] illustrates the box plots of the degree of deviation readings across the various
groups, highlighting the wide spread of values within the conventional impression
group, along with the identification of one outlier within the TRIOS 5 group.
Fig. 3 Box plot of the degree of deviation between the different impression types from the
reference. Note that one outlier was spotted in the TRIOS 5 group. DD = degree of
deviation.
Table 2
Descriptive statistics of the degree of deviation analysis mean in millimeter
|
Group
|
N
|
Mean
|
Standard deviation
|
|
C
|
10
|
0.82
|
0.16
|
|
R
|
10
|
1.02
|
0.05
|
|
T
|
10
|
1.11
|
0.06
|
Abbreviations: C, conventional impression; R, Runyes 3DS; T, TRIOS 5.
Table 3
Degree of deviation's mean analysis
|
(I) Sample
|
(J) Sample
|
Mean difference (I − J)
|
Standard error
|
Significance
|
|
R
|
C
|
0.20
|
0.05
|
0.01[a]
|
|
T
|
R
|
0.08
|
0.02
|
0.01[a]
|
|
T
|
C
|
0.29
|
0.05
|
0.00[a]
|
Abbreviations: C, conventional impression; R, Runyes 3DS; T, TRIOS 5.
a The mean difference is significant at the 0.05 level.
Descriptive statistics pertaining to the dimensional discrepancies in the scan body
zone are presented in [Table 4]. The reference file measurements for the scan body zone were recorded at 41.80 mm
in the X dimension, 11.25 mm in the Y dimension, and 16.05 mm in the Z dimension. Notably, discrepancies were most pronounced along the Z-axis, with all three groups exhibiting greater deviations from the reference measurements
in this axis compared with the others.
Table 4
Descriptive statistics of the abutment zone dimensional discrepancy
|
Axis
|
Impression type
|
N
|
Mean
|
Standard deviation
|
|
X (mesiodistally)
|
Control (LBS)
|
41.8000
|
|
|
C
|
10
|
41.8620
|
0.11679
|
|
R
|
10
|
41.8790
|
0.07475
|
|
T
|
10
|
41.9580
|
0.08854
|
|
Y (apico-occlusally)
|
Control (LBS)
|
11.2500
|
|
|
C
|
10
|
11.5110
|
0.18912
|
|
R
|
10
|
11.2150
|
0.10470
|
|
T
|
10
|
11.3240
|
0.11266
|
|
Z (buccolingually)
|
Control (LBS)
|
16.0500
|
|
|
C
|
10
|
15.9950
|
0.03719
|
|
R
|
10
|
15.9980
|
0.04492
|
|
T
|
10
|
16.1100
|
0.02625
|
Abbreviations: C, conventional impression; LBS, laboratory-based scanner; R, Runyes
3DS; T, TRIOS 5.
The conventional impression method demonstrated the least discrepancies in the X- and Z-axes while exhibiting the highest levels of discrepancy solely in the Y-axis. This method displayed the lowest precision in the X-axis (SD ± 0.11679) and Y-axis (SD ± 0.18912). Conversely, Runyes 3DS revealed the least discrepancies in the
Y-axis and exhibited superior precision in both the X-axis (SD ± 0.07475) and Y-axis (SD ± 0.10470). In contrast, the TRIOS 5 images were characterized by the most
significant discrepancies in the X- and Z-axes while maintaining the highest precision in the Z-axis (SD ± 0.02625) (as detailed in [Table 4]). The conventional impression method identified one outlier in the X-axis, whereas the Runyes 3DS and TRIOS 5 scans revealed two and one outliers in the
Y-axis, respectively. Additionally, the TRIOS 5 scans presented two outliers in the
Z-axis, as illustrated in [Figs. 4] to [6].
Fig. 4 Box plot of the dimensional discrepancy in the X-axis (mesiodistally) between different impression methods compared with the reference.
Note that one outlier was spotted in the conventional impression group.
Fig. 5 Box plot of the dimensional discrepancy in the Y-axis (apico-occlusally) between different impression methods compared with the reference.
Note that two and one outliers were spotted in Runyes 3DS and TRIOS 5 groups, respectively.
Fig. 6 Box plot of the dimensional discrepancy in the Z-axis (buccolingually) between different impression methods compared with the reference.
Note that two outliers were spotted in the TRIOS 5 group.
Statistical analysis indicated that the conventional impression method exhibited significant
discrepancies in both the Y- and Z-axes. The TRIOS 5 scans also displayed significant discrepancies in the X- and Z-axes. Meanwhile, the Runyes 3DS readings were significantly discrepant solely in
the Z-axis; however, the results approached significance in the X-axis with a borderline, p = 0.051 ([Table 5]).
Table 5
Dimensional discrepancy's mean analysis
|
Dependent variable
|
(I) sample
|
(J) sample
|
Mean difference (I − J)
|
Standard error
|
Significance
|
|
X (mesiodistally)
|
Control (LBS)
|
C
|
−0.06200
|
0.03693
|
0.559
|
|
R
|
−0.07900
|
0.02364
|
0.051
|
|
T
|
−0.15800a
|
0.02800
|
0.002a
|
|
Y (apico-occlusally)
|
Control (LBS)
|
C
|
−0.26100a
|
0.05980
|
0.011a
|
|
R
|
0.03500
|
0.03311
|
0.899
|
|
T
|
−0.07400
|
0.03563
|
0.343
|
|
Z (buccolingually)
|
Control (LBS)
|
C
|
0.05500a
|
0.01176
|
0.007a
|
|
R
|
0.05200a
|
0.01420
|
0.031a
|
|
T
|
−0.06000a
|
0.00830
|
0.000a
|
Abbreviations: C, conventional impression; LBS, laboratory-based scanner; R, Runyes
3DS; T, TRIOS 5.
aThe mean difference is significant at the 0.05 level.
Discussion
The present study highlights statistically significant differences in dimensional
discrepancies and deviations in scan body positions for All-on-4 implants across various
impression methods, which ultimately led to the rejection of both null hypotheses.
It appears that the performance of the conventional impression group is comparatively
aligned with that of the Runyes 3DS group, while the TRIOS 5 group yielded less favorable
results. While the conventional impression method demonstrated the best trueness (lowest
mean deviation), it also exhibited the highest variability (SD ± 0.16 mm). This lower
precision is likely attributable to the numerous manual steps involved, including
the pouring of gypsum and the potential for analog positioning error. In contrast,
the scan bodies utilized in the TRIOS 5 and Runyes 3DS groups were firmly secured
in place during the scanning process, which could account for the more consistent
results observed in these digital impression groups. It is noteworthy that the conventional
impression method recorded the lowest mean degree of deviation (0.82 mm), and significant
statistical differences were identified when comparing the digital methods to the
conventional group, thereby supporting the notion that the conventional method remains
a highly reliable option in this context. Interestingly, the greatest dimensional
discrepancies for the conventional impression group were noted along the Y-axis (apico-oclusally), which may be attributable to the frequent insertion and removal
of impression posts. Cumulatively, the dimensional discrepancy analysis revealed that
the Runyes 3DS group exhibited the least discrepancy from the reference, followed
by the conventional impression and TRIOS 5 groups, respectively. These findings present
a clinical trade-off: the conventional method may, on average, get closer to the true
position but with less predictability, while digital methods offer very consistent
results that may be consistently offset by a certain error.
The clinical relevance of the observed deviations must be considered. The mean deviations
reported (0.82–1.11 mm) exceed the widely cited threshold of 150 µm considered acceptable
for a passive prosthetic fit.[20]
[21] This suggests that for All-on-4 cases, a single, nonsplinted digital impression
without verification may carry a risk of clinically significant misfit. This underscores
the necessity of employing clinical strategies to improve accuracy, such as using
verification jigs or splinting techniques, regardless of the impression method chosen.
A critical consideration in this study is the selection of the LBS as the reference
model. While industrial-grade coordinate measuring machines offer superior nominal
accuracy, the use of a high-precision LBS is a well-established methodology in dental
literature for in vitro comparisons of IOSs.[11]
[23]
[27] It provides a clinically relevant benchmark, as it represents the real-world gold
standard to which IOS data are compared in dental laboratories. However, it is acknowledged
that the stated accuracy of the LBS (20 µm) is in the same order of magnitude as the
precision claimed by some IOSs. This means that the absolute trueness values reported
(e.g., deviations of 0.8–1.1 mm) should be interpreted with caution, as they may include
a component of error from the reference itself. Nevertheless, the relative comparison
between the test groups remains valid, as all were compared against the same reference
under identical conditions.
Recent studies examining the accuracy of various impression methods for complete-arch
implant rehabilitation offer valuable insights that enhance the understanding of the
effectiveness and limitations of these methodologies. The investigation by Zingari
et al[28] provides significant comparisons among photogrammetry, IOS, and conventional impressions.
Utilizing a standardized maxillary polymer resin model, their research indicates that
photogrammetry achieves the highest levels of trueness and precision. This finding
highlights the promising potential of photogrammetry for facilitating superior accuracy,
likely due to its advanced digitization process. Conversely, the IOS method demonstrated
lower accuracy, prompting important considerations regarding the reliability of digital
impressions, especially in complex implant cases requiring high precision. In a complementary
study, Sallorenzo and Gómez-Polo[29] compared the implant-specific PiC camera with the conventional TRIOS 3 scanner,
revealing that the PiC camera provided significantly greater trueness and precision.
Moreover, the research by Kaya and Bilmenoglu[30] focused on the performance of 14 IOSs in the context of the All-on-4 treatment approach.
By simulating a clinical environment, this study underscored the necessity of consistency
in digital impressions across diverse devices. The study highlighted the superiority
of some scanners to scan these complex scenarios. However, the study also highlighted
the need for caution regarding potential data loss due to artifacts or reflections,
which is crucial for selecting the most suitable scanners for complex impression-taking
procedures. Lyu et al[31] contributed further by comparing intraoral scanning results with conventional impressions
across various scan ranges. Their findings suggest that larger scanning areas can
negatively impact accuracy, particularly in cross-arch scans. The various evaluation
methods employed indicate that trueness is significantly affected not only by the
impression method but also by the specific clinical context, emphasizing the importance
of careful consideration in choosing scanning ranges. The study conducted by Vieira
et al[32] focused on the accuracy of digital intraoral versus conventional elastomeric impressions
for fabricating complete implant-supported bars on a four-implant master model. Five
zirconia bars were produced for each impression type using CAD/CAM technology, with
vertical misfit measured in micrometers through a scanning electron microscope. The
statistical analysis revealed no significant difference in vertical misfit between
the two methods, regardless of screw configuration. Nonetheless, the study's scope
was limited by its in vitro setting and focus on specific materials, which suggests the need for further exploration
of additional outcome measures.
Collectively, these findings reflect a positive trend in the advancement of digital
impression technology for full-arch implant rehabilitation. However, the accuracy
of these methods can vary significantly based on technique, equipment, and clinical
scenarios. Future research should aim to bridge the gap between in vitro studies and clinical applications to ensure the adoption of the most accurate and
reliable impression methods, ultimately enhancing patient outcomes.
The current study's in vitro design presents certain limitations that may impact its applicability to clinical
practice. First, the controlled environment does not fully capture the complexities
present in oral conditions, where factors such as mouth opening ability, mucosal mobility,
saliva, and gingival crevicular fluid significantly influence the accuracy of digital
scans. Second, the comparison was made using the 3D analysis software, which is an
acceptable method of analysis. However, it is more meaningful to evaluate the difference
between the actual prosthesis made from these impressions to test the seating and
passive fit directly. Moreover, the manual tightening of scan bodies can introduce
variability; any unintentional micromovement during the impression process may result
in deviations between reference and conventional impressions. In contrast, the IOSs
maintained a constant configuration of both scan bodies and jaw structures between
reference and digital scans. Additionally, the IOS scanned the same jaw multiple times,
whereas each conventional impression corresponds to a unique master model. These considerations
may complicate the interpretation of the results and highlight the importance of conducting
further investigations into these variables.
Conclusion
Within the limitations of this in vitro study, the following conclusions can be drawn:
-
The impression method had a statistically significant impact on the trueness and precision
of scan body position capture in an All-on-4 mandibular model.
-
The conventional open-tray impression technique demonstrated the highest trueness
(lowest mean deviation from the reference) but the lowest precision (highest variability).
-
The Runyes 3DS IOS demonstrated the highest precision (lowest variability), whereas
the TRIOS 5 IOS exhibited the lowest trueness (greatest mean deviation from the reference).
-
Dimensional discrepancies were most pronounced along the Z-axis (buccolingual) for all groups.
The observed deviations highlight the inherent challenges in capturing full-arch implant
positions. The superior precision of digital impressions suggests they are highly
reproducible, but the overall trueness results indicate that the direct application
of a single digital scan for complex All-on-4 prostheses may require validation. Future
research should prioritize in vivo studies that correlate these dimensional measurements with the actual clinical fit
of the final prosthesis to determine the clinical significance of these differences.
