Accuracy of Intraoral Scanners Versus Polyvinyl Siloxane Impression in Partially Edentulous Implant Rehabilitations: An In Vitro Comparison

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Prosthesis

Title: Evaluation of the accuracy of intraoral scanner versus analog impression in implant supported rehabilitations: an in vitro study

Manuscript ID: prosthesis-3927221

 

Reviewer Comments

- Title: The authors may keep the current title or, alternatively, make it slightly more focused on the superior accuracy of digital methods, which is the main finding: “Accuracy and Consistency of Intraoral Scanners Versus Polyvinyl Siloxane Impression in Implant Rehabilitations: An In Vitro Comparison.”

- Abstract: Authors may include a brief reference to clinical significance in the Conclusions section of the Abstract.

- Introduction: It is suggested that the authors strengthen the last paragraph of the Introduction by adding more key references to justify the specific need to compare the two intraoral scanners with PVS in this context, demonstrating more robustly what is unknown in the literature.

- Results: Authors may add an introductory/summary paragraph that briefly summarises the main findings.

- Discussion: It is suggested that authors include a sentence in this section acknowledging that, although the Best-Fit overlay method is operator-independent, the fact that the evaluator is not blinded may introduce a potential risk of bias and should be considered a limitation.

- Conclusion: The conclusion section lacks a direct and final mention of the clinical implication of the finding.

Author Response

Comment 1 - Title: The authors may keep the current title or, alternatively, make it slightly more focused on the superior accuracy of digital methods, which is the main finding: “Accuracy and Consistency of Intraoral Scanners Versus Polyvinyl Siloxane Impression in Implant Rehabilitations: An In Vitro Comparison.”

We thank the reviewer for this valuable suggestion. To better emphasize the main finding of our study, we have slightly modified the title as follows: "Accuracy of intraoral scanners versus Polyvinyl Siloxane Impression in Implant Rehabilitations: An In Vitro Comparison."

 

Comment 2 - Abstract: Authors may include a brief reference to clinical significance in the Conclusions section of the Abstract.

We appreciate this suggestion. We have added a sentence to highlight the clinical relevance of the findings in the Conclusions section of the Abstract: "Within the limitations of this in vitro study, the higher accuracy and consistency of digital impressions may translate into improved clinical efficiency and prosthetic fit in implant rehabilitations"

 

Comment 3 - Introduction: It is suggested that the authors strengthen the last paragraph of the Introduction by adding more key references to justify the specific need to compare the two intraoral scanners with PVS in this context, demonstrating more robustly what is unknown in the literature.

We thank the reviewer for this important suggestion. We have added a new paragraph with updated references that highlight the current gaps in the literature and the need for direct comparative assessments between intraoral scanners and conventional PVS techniques in multi-implant scenarios. The revised paragraph now reads as follows: "However, current evidence remains inconsistent regarding the accuracy (trueness and precision) of digital versus conventional impression techniques in implant-supported rehabilitations. Several studies and reviews report high or comparable accuracy for IOS relative to PVS in many clinical/in-vitro settings [9–11], whereas other investigations highlight limitations for full-arch or multi-implant spans (e.g., cumulative stitching and strategy-dependent errors), sometimes favoring conventional approaches for long spans [12–14]. Moreover, direct head-to-head comparisons between different IOS systems and conventional PVS under standardized conditions are still limited; specifically, evidence comparing Trios3Shape® and Carestream CS 3600™—two widely adopted scanners with distinct acquisition technologies—remains relatively scarce, although available in vitro data do exist [15–17]."

 

Comment 4 - Results: Authors may add an introductory/summary paragraph that briefly summarises the main findings.

We appreciate the reviewer’s constructive comment. We have added an introductory paragraph at the beginning of the Results section summarizing the key findings for greater clarity: 

"A total of 51 models were analyzed. Table 1 reports the mean deviation (trueness) and standard deviation (precision) values for each impression technique. Both intraoral scanners—Trios 3Shape® and Carestream CS 3600®—showed lower mean deviation and variability compared with the conventional PVS impressions. Specifically, mean ± SD values were 13 ± 79 μm for Trios 3Shape®, 12 ± 82 μm for CS 3600®, and 25 ± 213 μm for PVS.

The ANOVA test (Table 3) revealed significant differences among the three techniques (p < 0.001). Pairwise comparisons using the Z-test (Table 4) confirmed that both digital systems were significantly more accurate than PVS (p < 0.001), while no statistically significant difference was found between the two scanners (p = 0.60).

Figure 1 illustrates the overall distribution of deviations. All techniques achieved accuracy within the clinically acceptable threshold of 30 μm. However, the digital systems displayed narrower dispersion and higher consistency, as evidenced by their lower SD values (Tables 5 and 6).

When analyzing signed deviations (Table 2), Trios 3Shape® tended to produce slightly more negative deviations, whereas PVS showed a prevalence of positive deviations, indicating dimensional overestimation. Carestream CS 3600® demonstrated a balanced distribution around zero."

Comment 5 - Discussion: It is suggested that authors include a sentence in this section acknowledging that, although the Best-Fit overlay method is operator-independent, the fact that the evaluator is not blinded may introduce a potential risk of bias and should be considered a limitation.

We appreciate this suggestion and added a paragraph: "Although the Best-Fit overlay used for surface superimposition is algorithm-driven, our workflow included a manual pre-alignment step and the evaluator was not blinded to the impression technique. These aspects may introduce a potential risk of bias and should be considered a limitation of the present study."

 

Comment 6 - Conclusion: The conclusion section lacks a direct and final mention of the clinical implication of the finding.

We appreciate this helpful observation. The Conclusions section has been revised to explicitly address the clinical relevance of our findings: "Within the limitations of this in vitro study, all three impression techniques—TRIOS® 3Shape, CS 3600® Carestream Dental, and PVS analog impressions—demonstrated high intrinsic accuracy, with mean deviations below 30 μm from the reference model.

Both digital systems showed superior accuracy and consistency compared with the conventional PVS method. These findings suggest that intraoral scanning can provide more reliable and reproducible impressions, reducing potential errors related to the analog workflow.

From a clinical perspective, the higher trueness and precision of digital impressions may improve the passive fit of implant-supported restorations, enhance efficiency in laboratory communication, and increase patient comfort during impression procedures.

Further in vivo studies are needed to confirm these advantages under clinical conditions and to evaluate their long-term impact on prosthetic success and implant survival."

Reviewer 2 Report

Comments and Suggestions for Authors

The introduction needs to be enriched, as this has been a much discussed issue in recent years, it is necessary to point out what is new in this study, what is missing so far or how it differs from similar studies.

Photos of the studies can be added to demonstrate how the measurements were performed.

In the discussion, point out what is new and useful that is not known until now.

Author Response

Comment 1 - The introduction needs to be enriched, as this has been a much discussed issue in recent years, it is necessary to point out what is new in this study, what is missing so far or how it differs from similar studies.

We thank the reviewer for this observation. The Introduction section has been expanded in the revised version to better highlight the gaps in the current literature, the rationale for comparing Trios 3Shape® and CS 3600™ intraoral scanners with PVS, and the novelty of the present study. Relevant recent references have been added accordingly (Introduction, last paragraph): 

"However, current evidence remains inconsistent regarding the accuracy (trueness and precision) of digital versus conventional impression techniques in implant-supported rehabilitations. Several studies and reviews report high or comparable accuracy for IOS relative to PVS in many clinical/in-vitro settings [9–11], whereas other investigations highlight limitations for full-arch or multi-implant spans (e.g., cumulative stitching and strategy-dependent errors), sometimes favoring conventional approaches for long spans [12–14]. Moreover, direct head-to-head comparisons between different IOS systems and conventional PVS under standardized conditions are still limited; specifically, evidence comparing Trios3Shape® and Carestream CS 3600™—two widely adopted scanners with distinct acquisition technologies—remains relatively scarce, although available in vitro data do exist [15–17]."

These additions directly address the reviewer’s request to clarify what is new in the present work and how it differs from previous studies.

Comment 2 - Photos of the studies can be added to demonstrate how the measurements were performed.

We thank the reviewer for this suggestion. The aligment procedure has been described in detail in the Materials and Methods section and, to further clarify, a representative image illustrating the Best-Fit overlay process has been added (Figure 1 and figure 2).

Comment 3 - Research design, methods, results, and conclusions can be improved.

We appreciate the reviewer’s general comment. Following the first revision, substantial improvements have been made in all major sections of the manuscript, including a more detailed description of the study design and statistical methods, an expanded and clarified Results section, a more comprehensive Discussion addressing potential limitations, and a revised Conclusion explicitly outlining the clinical implications of the findings.

"This in-vitro study was based on 51 master models obtained from partially edentulous clinical cases previously treated at the Implantology and Prosthodontics Department of the University of Milan. Digital and conventional impressions were compared under standardized laboratory conditions. The master models were generated from digital patient data (CBCT and intraoral scans) but no patients were directly involved in the experimental phase; therefore, ethical approval and informed consent were not required for this in-vitro analysis."

" A total of 51 models were analyzed. Table 1 reports the mean deviation (trueness) and standard deviation (precision) values for each impression technique. Both intraoral scanners—Trios 3Shape® and Carestream CS 3600®—showed lower mean deviation and variability compared with the conventional PVS impressions. Specifically, mean ± SD values were 13 ± 79 μm for Trios 3Shape®, 12 ± 82 μm for CS 3600®, and 25 ± 213 μm for PVS.

The ANOVA test (Table 3) revealed significant differences among the three techniques (p < 0.001). Pairwise comparisons using the Z-test (Table 4) confirmed that both digital systems were significantly more accurate than PVS (p < 0.001), while no statistically significant difference was found between the two scanners (p = 0.60).

Figure 1 illustrates the overall distribution of deviations. All techniques achieved accuracy within the clinically acceptable threshold of 30 μm. However, the digital systems displayed narrower dispersion and higher consistency, as evidenced by their lower SD values (Tables 5 and 6).

When analyzing signed deviations (Table 2), Trios 3Shape® tended to produce slightly more negative deviations, whereas PVS showed a prevalence of positive deviations, indicating dimensional overestimation. Carestream CS 3600® demonstrated a balanced distribution around zero."

"The present study provides additional quantitative evidence that, under standardized conditions, both Trios3Shape® and Carestream CS 3600™ achieve accuracy values well within the 30 µm clinical threshold, confirming their suitability for implant-supported restorations. This direct comparative data between these two scanners under identical conditions has not been reported before."

"Although the Best-Fit overlay used for surface superimposition is algorithm-driven, our workflow included a manual pre-alignment step and the evaluator was not blinded to the impression technique. These aspects may introduce a potential risk of bias and should be considered a limitation of the present study."

"Within the limitations of this in vitro study, all three impression techniques—TRIOS® 3Shape, CS 3600® Carestream Dental, and PVS analog impressions—demonstrated high intrinsic accuracy, with mean deviations below 30 μm from the reference model.

Both digital systems showed superior accuracy and consistency compared with the conventional PVS method. These findings suggest that intraoral scanning can provide more reliable and reproducible impressions, reducing potential errors related to the analog workflow.

The conclusions are strictly limited to in-vitro observations but clinical information can be drawn from these. From a clinical perspective, the higher trueness and precision of digital impressions may improve the passive fit of implant-supported restorations, enhance efficiency in laboratory communication, and increase patient comfort during impression procedures.

Further in vivo studies are needed to confirm these advantages under clinical conditions and to evaluate their long-term impact on prosthetic success and implant survival."

Reviewer 3 Report

Comments and Suggestions for Authors

Dear Authors,

            Using a laboratory scanner gold standard for partially edentulous models, the publication compares the accuracy of two intraoral scanners (Trios 3Shape, Carestream CS 3600) with conventional PVS impressions. The main conclusion is that, although all approaches are clinically acceptable (less than 30 µm), digital scanners provide better accuracy and far less variability than traditional methods. Rewrite the text as an in-vitro laboratory investigation in a clear and consistent manner. Eliminate clinical trial information (such as allocated and received interventions) and cut down on clinical patient details that aren't necessary. The origin and specifications of the master models should be the main topics of the introduction and methodology. Talk about the in-vitro model's limits, the inaccuracy that comes with the gold standard scanner, and the possible bias it introduces in a formal and noticeable way.

1. Although the abstract and introduction make reference to "patients," "surgeries," and "clinical conditions," the main approach (Section 2.6) is specifically an in-vitro investigation carried out using printed master models that were created using patient data. The study assesses the impression systems' accuracy in vitro rather than their clinical efficacy.
2. The "Excluded" and "Allocated" portions that were included prior to the introduction need to be eliminated because they are obviously an incorrectly placed piece of a clinical trial CONSORT flow diagram.
3. In the introduction, the lines 68–70 that read, "In this in vitro experimental protocol..." are a blatant contradiction to the previous and subsequent material that describes the recruitment of patients and the surgical procedures. According to lines 149–150, the study must always be presented in the report as an in-vitro laboratory inquiry utilising models taken from a clinical cohort.
4. Materials and techniques: The reported in-vitro accuracy experiment is essentially unrelated to the lengthy discussion of patient recruiting (2017–2022), inclusion/exclusion criteria, surgical techniques (flapless guided surgery), and ethical approval. It leads to misunderstanding. If patient scans were used to construct the master models, this should be briefly mentioned. A laboratory accuracy research should not use the framework of a large clinical trial. According to the text, "51 dental models from 42 patients." Did several models from the same patient receive separate treatment?
5. The description of the PVS group is confusing: "Allocated to PVS analog impression (=51) • Received allocated intervention (n=51) • Did not receive allocated intervention (n=51)". This makes no sense.
6. Given that the mean values (13µm, 12µm, and 25µm) are much smaller than the corresponding standard deviations (79µm, 82µm, and 213µm), does this indicate a skewed distribution or the presence of outliers in the data? Since the conventional mean ± SD can be misleading here, would reporting the median and interquartile range (IQR) provide a more accurate representation? Notably, despite a "good" mean, the large standard deviation—especially the 213µm for PVS—highlights that many data points greatly deviate from the reference and warrants clearer emphasis in the results.
7. The description of scanner technologies, while informative, is lengthy and reads like manufacturer marketing material. It could be condensed.
8. The conclusion that "IN VIVO studies are needed" is correct but highlights the core flaw of the manuscript. The conclusions should be strictly limited to the in-vitro findings.

Your Peer-reviewer

Author Response

We sincerely thank the reviewer for the valuable and precise comments, which significantly improved the clarity and methodological consistency of our manuscript. Below we detail all the modifications implemented in accordance with each suggestion.

Study design and general consistency
The entire manuscript has been rewritten and consistently presented as an in-vitro laboratory investigation. All references to clinical procedures, patient recruitment, or interventions have been removed or rephrased.

Title
The title has been modified to clearly indicate the in-vitro nature of the study: “Accuracy of Intraoral Scanners versus Polyvinyl Siloxane Impression in Implant Rehabilitations: An In Vitro Comparison.”

Abstract
The Abstract has been revised to explicitly clarify that the study was in-vitro and to provide information on the model origin and ethical aspects. The following sentences were added:

“The study was based on 3D-printed master models derived from partially edentulous clinical cases previously treated at our department (2017–2022). All cases required at least two implants.”

“Ethical approval and trial registration were not applicable to the present in-vitro investigation, as no patients were directly involved in the experimental phase. The digital data used to generate the laboratory master models originated from a separate clinical study conducted at ASST Santi Paolo e Carlo, Milan (Ethics Committee approval no. 1361, July 12, 2017; ClinicalTrials.gov registration, Unique Protocol ID 1361).”

Materials and Methods 
The introductory paragraph was rewritten as follows: “This in-vitro study was based on 51 master models obtained from partially edentulous clinical cases previously treated at the Implantology and Prosthodontics Department of the University of Milan. Digital and conventional impressions were compared under standardized laboratory conditions. The master models were generated from digital patient data (CBCT and intraoral scans), but no patients were directly involved in the experimental phase; therefore, ethical approval and informed consent were not required for this in-vitro analysis.”

“Ethical approval and trial registration were not applicable to the present in-vitro investigation, as no patients were directly involved in the experimental phase. The digital data used to generate the laboratory master models originated from a separate clinical study conducted at ASST Santi Paolo e Carlo, Milan (Ethics Committee approval no. 1361, July 12, 2017; ClinicalTrials.gov registration, Unique Protocol ID 1361).”

Additionally, all paragraphs referring to patient recruitment (2017–2022), inclusion/exclusion criteria, surgical procedures, and clinical follow-up have been removed to avoid confusion with clinical trials.

Section 2.6 – In-vitro procedure
The opening sentence was modified to reflect the correct laboratory context: “For each printed master model, implant scan bodies were positioned before scanning with the Trios 3Shape® scanner.”

CONSORT diagram
The previous CONSORT-style allocation diagram has been entirely removed from the manuscript, as it was inconsistent with the in-vitro design. A simplified methodological workflow has been retained in the text instead.

Conclusions
The Conclusions section was fully revised to limit statements to the in-vitro context. The new version reads: “Within the limitations of this in-vitro study, all three impression techniques—TRIOS® 3Shape, CS 3600® Carestream Dental, and PVS analog impressions—demonstrated high intrinsic accuracy, with mean deviations below 30 μm from the reference model. Both digital systems showed superior accuracy and consistency compared with the conventional PVS method. These findings suggest that intraoral scanning can provide more reliable and reproducible impressions, reducing potential errors related to the analog workflow.

Summary of changes:

• Manuscript fully rewritten as an in-vitro investigation

• All clinical details removed

• CONSORT diagram deleted

• Abstract, Methods, and Conclusions updated accordingly

• Ethical statement clarified

We sincerely appreciate the reviewer’s insightful comments, which helped us refine the manuscript and improve its internal coherence and scientific rigor.

Reviewer 4 Report

Comments and Suggestions for Authors

prosthesis-3927221 comments

It is a well written in-vitro study on an interesting topic. There are some points however that have to be corrected and a revision is required.

• The title is “Evaluation of the accuracy of intraoral scanner versus analog impression in implant supported rehabilitations: an in vitro  study”. In the exclusion criteria however the authors state that fully edentulous arches were excluded.  The authors should explain if impressions in patients with implants in fully edentulous arches were taken. If these patients were not included, it should be included in the title that the study was focused in partial impressions.
• In the beginning of the results the mean deviation was the measured dependent values. The authors should explain how these values were calculated.
• The authors state “… the average displacement values of the three methods with either a positive or negative sign”. How was the positive and negative sign was determined???
• In general the results should be reported in a more understandable way. The findings and statements should be based on the tables and diagrams and should be related accordingly to allow correlation. As they are written the conclusions are mixed with the findings.
• In the Discussion the authors state “…Discrepancies with other studies can be attributed”. Which are these studies? Furthermore the differences with the other studies should be reported.
• At the end of the discussion the limitations of the study should be reported.
• In the Conclusions the authors state “ IN VIVO studies are needed to assess the accuracy of IOS systems and conventional …”. This phrase should be in the discussion
• Last but not least: the authors should explain and comment in extent the comparison in accuracy of intraoral scanning and conventional impression in implants in fully edentulous patients.

                                                                                                                           .

Author Response

comment 1 -

• The title is “Evaluation of the accuracy of intraoral scanner versus analog impression in implant supported rehabilitations: an in vitro  study”. In the exclusion criteria however the authors state that fully edentulous arches were excluded.  The authors should explain if impressions in patients with implants in fully edentulous arches were taken. If these patients were not included, it should be included in the title that the study was focused in partial impressions.

We thank the reviewer for this valuable comment. We confirm that fully edentulous arches were excluded from the experimental dataset. The master models were derived exclusively from partially edentulous clinical cases, each including at least two implants per segment. To improve clarity, we have revised the title to specify that the study focused on partially edentulous implant rehabilitations: "Accuracy of intraoral scanners versus Polyvinyl Siloxane Impression in partially edentulous implant rehabilitations: An In Vitro Comparison."

comment 2 - 

• In the beginning of the results the mean deviation was the measured dependent values. The authors should explain how these values were calculated.

To respond to this clarification "In the beginning of the results the mean deviation was the measured dependent values. The authors should explain how these values were calculated.",  this sentence has been inserted in point 2.7: "For each impression technique, the surface deviation between the test model and the reference model was computed using the Best-Fit alignment algorithm implemented in GOM Inspect (Zeiss®, Oberkochen, Germany). The software calculated the linear distance between corresponding surface points of the two superimposed STL files. The mean deviation for each model was defined as the arithmetic mean of the absolute distance values (in μm) of all surface points, while the standard deviation described the dispersion of these values around the mean. Thus, each master model provided one mean deviation value, representing its overall trueness relative to the reference scan."

comment 3 - 

• The authors state “… the average displacement values of the three methods with either a positive or negative sign”. How was the positive and negative sign was determined???

To respond to the doubt about this incomplete authors phrase, “… the average displacement values of the three methods with either a positive or negative sign”. How was the positive and negative sign was determined???" The authors added in results: "The positive and negative signs were automatically assigned by the GOM Inspect software based on the direction of deviation between the test and reference surfaces. Positive values indicated that the test surface was located outward (above) relative to the reference model, whereas negative values indicated that the test surface was positioned inward (below) the reference surface. These signed deviations allowed qualitative visualization of the directionality of errors in the color-mapping analysis."

comment 4 -

• In general the results should be reported in a more understandable way. The findings and statements should be based on the tables and diagrams and should be related accordingly to allow correlation. As they are written the conclusions are mixed with the findings.

We thank the reviewer for this helpful observation. We have revised the Results section to improve clarity and readability. The findings are now presented in a structured manner, directly referencing the corresponding tables and figures. Interpretative statements have been moved to the Discussion section, while the Results now focus exclusively on the descriptive and statistical outcomes. These changes make the data presentation clearer and the correlation between text, tables, and diagrams more straightforward.

comment 5 - 

• In the Discussion the authors state “…Discrepancies with other studies can be attributed”. Which are these studies? Furthermore the differences with the other studies should be reported.

We thank the reviewer for this comment. We have now specified which studies showed different results and explained the methodological reasons for such discrepancies. In particular, we referred to Roig et al. (2020) and Nedelcu et al. (2018), who reported higher accuracy for conventional techniques in full-arch conditions, and to Imburgia et al. (2017)  who found results similar to ours in short-span implant models. A new paragraph has been added in the Discussion to clarify these differences.

comment 6 - 

• At the end of the discussion the limitations of the study should be reported.

We appreciate the reviewer’s suggestion. A new paragraph has been added at the end of the Discussion. It highlights the main constraints of the present in-vitro design, including the absence of in-vivo conditions, lack of operator blinding, restriction to partially edentulous cases, and the limited number of scanners tested. This section also outlines directions for future research:

"The present study has several limitations that should be acknowledged. First, the investigation was conducted in vitro, under controlled laboratory conditions that do not fully replicate the clinical environment ; factors such as saliva, soft-tissue mobility, and patient movement may affect impression accuracy in vivo. Second, the operator was not blinded to the impression technique, which may introduce a potential measurement bias during scanning and data processing. Third, the sample included only partially edentulous models with short implant spans; therefore, the results cannot be directly extrapolated to full-arch rehabilitations. Moreover, only two intraoral scanners were evaluated, and both the scanning strategy and alignment method (Best-Fit algorithm) could influence the deviation outcomes. Finally, the study assessed trueness and precision only through surface superimposition analysis, without evaluating the clinical fit of the prosthetic restorations.

Future in-vivo studies involving different IOS technologies, full-arch scenarios, and standardized scanning protocols are needed to validate these findings under clinical conditions."

comment 7 -

• In the Conclusions the authors state “ IN VIVO studies are needed to assess the accuracy of IOS systems and conventional …”. This phrase should be in the discussion

We thank the reviewer for this observation. The Conclusions section has been revised to focus solely on the main findings and clinical implications of the study.Although the Best-Fit overlay used for surface superimposition is algorithm-driven, our workflow included a manual pre-alignment step and the evaluator was not blinded to the impression technique. These aspects may introduce a potential risk of bias and should be considered a limitation of the present study: "The present investigation has several limitations that should be acknowledged. First, it was conducted in vitro, under controlled laboratory conditions that do not fully replicate the clinical environment; factors such as saliva, soft-tissue mobility, and patient movement may affect impression accuracy in vivo. Second, the operator was not blinded to the impression technique, potentially introducing measurement bias during scanning and data processing. Third, the sample included only partially edentulous models with short implant spans; therefore, the results cannot be directly extrapolated to full-arch rehabilitations. Moreover, only two intraoral scanners were evaluated, and both the scanning strategy and alignment method (Best-Fit algorithm) could influence deviation outcomes. Finally, the study assessed trueness and precision only through surface superimposition, without evaluating the clinical fit of the resulting prosthetic restorations. Within the limitations of this in-vitro investigation, further in-vivo studies are required to validate these findings under clinical conditions. Future research should compare different intraoral scanning systems, full-arch situations, and standardized scanning strategies to better assess the reproducibility and clinical applicability of digital versus conventional impressions in implant-supported rehabilitations. It is also important to consider that IOS devices are continually updated with new hardware and software versions, while research on polyvinyl siloxanes (PVS) has largely reached its development limit. Therefore, future comparisons are likely to increasingly favor digital systems as technological improvements continue."

comment 8 -

• Last but not least: the authors should explain and comment in extent the comparison in accuracy of intraoral scanning and conventional impression in implants in fully edentulous patients.

We thank the reviewer for this comment. Fully edentulous arches were intentionally excluded from this study, which focused exclusively on partially edentulous implant rehabilitations. A short paragraph has been added in the Discussion to clarify this methodological choice and to briefly comment on the literature regarding fully edentulous cases, emphasizing that the present findings cannot be extrapolated to full-arch implant scenarios.

"This study specifically focused on partially edentulous implant rehabilitations, as these represent the most frequent clinical scenarios and allow for standardized short-span scanning procedures. Fully edentulous arches were intentionally excluded because they involve different technical challenges—such as longer scanning spans and fewer anatomic landmarks—which can increase cumulative stitching errors and reduce accuracy.

The available literature indicates that, despite continuous technological improvements, intraoral scanning of fully edentulous arches still shows variable results compared with conventional impressions, especially in full-arch implant cases. For this reason, the present results should be interpreted strictly within the context of partially edentulous models, without direct extrapolation to full-arch rehabilitations."

Round 2

Reviewer 3 Report

Comments and Suggestions for Authors

The authors have provided a response to the reviewers' comments. The document satisfies requirements for publication.

Your Peer-reviewer

Reviewer 4 Report

Comments and Suggestions for Authors

The authors have made meticulous work and they have made all suggested corrections in their revised manuscript. They have also cleared several issues that were questionable in the previous version. All the above mentioned have been reported in details in a point-by point resonce letter.

The manuscript has been significantly improved and can be accepted.