Comparison of Three‐Dimensional Accuracy Between Digital and Conventional Definitive Impressions of Edentulous Arches: A Clinical Study

ABSTRACT

Background and Aims

This clinical study evaluated the three‐dimensional discrepancies between digital and conventional definitive impressions of edentulous arches for complete denture fabrication.

Methods

Ten edentulous patients (n = 16 arches: 8 maxillary and 8 mandibular) were enrolled. For each case, three digital data sets were acquired: (1) a direct intraoral scan (IOS), (2) a scan of a conventional definitive impression obtained using a 3D‐printed custom tray (CI3D), and (3) a scan of a conventional definitive impression made with an acrylic special tray (CI). All STL files were first reversed in Geomagic Design X and then aligned using best‐fit algorithms in Geomagic Control X, with IOS considered the reference data set. Statistical analyses were performed using SPSS and Python for supporting data visualization. Dimensional discrepancies were evaluated using absolute 3D distance and root mean square (RMS) deviation. Paired‐sample t‐tests were used to compare CI3D and CI deviations from IOS. Subgroup analyses were conducted for maxillary and mandibular arches, and correlation analysis was performed between deviation metrics.

Results

No statistically significant differences were found between CI3D and CI in their deviations from the IOS reference. The overall mean difference between groups was −0.11 ± 1.62 mm, and the RMS difference was −0.10 ± 0.68 mm (p = 0.75 and 93, respectively). Subgroup analysis revealed no significant differences in maxillary or mandibular arches (all p > 0.5). A moderate positive but not significant correlation (r = 0.31, p = 0.21) was observed between mean deviation in CI and RMS deviation in CI3D.

Conclusion

Within the limitations of this clinical study, which included edentulous ridges with minimal to moderate bone loss (Cawood and Howell Class III), 3D‐printed and acrylic tray‐based conventional definitive impressions demonstrated comparable dimensional accuracy to intraoral scanning, showing no statistically significant or clinically relevant differences between techniques.

1. Introduction

Edentulism, the complete loss of natural teeth, presents significant challenges impacting masticatory function, nutrition, speech, aesthetics, and overall quality of life [1]. Unrestored edentulism encompasses profound social, psychological, and emotional consequences, highlighting the importance of effective prosthetic solutions [2, 3]. Various treatment options are available for patients with total edentulism, ranging from conventional removable prostheses to contemporary implant‐supported fixed or removable prostheses [4].

A highly accurate prosthesis fabrication, regardless of its type, hinges on the use of a precise impression technique to create exact working models and replicate the oral cavity accurately [5]. Conventional definitive impressions have been the traditional method for fabricating complete removable dentures for over two centuries [6]. While they offer certain advantages, such as cost‐effectiveness and versatility, they possess several limitations, including patient discomfort, which can lead to anxiety or reluctance during the procedure, inaccuracy in capturing fine surface details, time‐consuming, material limitations in terms of dimensional stability, tear resistance, or flow properties, workflow management challenges; and, to sum up, technique sensitivity [7, 8]. Digital dentistry represents a transformative shift in dental practice, leveraging cutting‐edge technology to enhance precision, efficiency, and patient outcomes [9]. At the forefront of this revolution is intraoral scanning that utilizes advanced optics and software algorithms to capture detailed, three‐dimensional images of the oral cavity. The rapid adoption of digital dentistry, propelled by computer‐aided design/computer‐aided manufacturing (CAD/CAM) systems and 3D printing, seamlessly integrates intraoral scans for highly accurate and customized prostheses, improving patient care and efficiency [10].

Intraoral scanning offers several advantages over conventional definitive impression‐taking techniques, including real‐time visualization, allowing clinicians to instantly assess the quality of the scan and make necessary adjustments on the spot [11]. In addition to providing accurate digital casts and reducing the risk of distortion linked to impression materials, intraoral scanning can optimize workflows by swiftly rescanning missed areas [12]. This potential enhancement in efficiency and accuracy may decrease errors and the necessity for retakes [13]. Furthermore, intraoral scanning offers a more comfortable and patient‐friendly experience by eliminating the discomfort associated with conventional definitive impressions, particularly for patients with nausea, breathing difficulties, or sensitive gag reflexes, leading to increased patient satisfaction and compliance [14]. Additionally, intraoral scanning enables selective recording of relevant areas, allowing clinicians to focus on specific regions of interest while excluding unnecessary information [15]. This targeted approach not only improves workflow efficiency but also reduces data processing time [16, 17], enhancing overall productivity in the dental office [18, 19]. Additionally, it enables straightforward transmission of information to laboratory technicians and the storage of data, minimizing the necessity for storing physical impressions and poured models [20].

Despite these advantages, intraoral scanning presents certain limitations. Capturing accurate impressions in fully edentulous patients can be challenging due to the absence of natural teeth and the presence of mobile soft tissues, which complicate scanner stabilization. Additionally, the smooth and reflective nature of denture‐bearing areas, often covered with saliva and nonkeratinized mucosa and lacking distinct anatomical landmarks, can affect the image‐stitching process and compromise scan accuracy [21, 22, 23]. Furthermore, when attempting to record tissues in a mucostatic condition, intraoral scanners (IOS) may face difficulties in accurately capturing the functional depth of the vestibule, particularly in posterior maxillary regions where access is limited by the size and design of the IOS tip [13, 24].

A comprehensive review of the current literature reveals a growing body of research investigating the accuracy and efficacy of intraoral scanning in fabricating complete removable dentures. While some studies report favorable outcomes and clinical acceptability [25, 26, 27, 28], others highlight challenges and discrepancies in accuracy [22, 29, 30]. These methodological differences and variations in scanner performance underscore the need for further research to validate the clinical utility of intraoral scanning in this context. Therefore, the aim of the present study was to compare the dimensional accuracy of digital intraoral scans and conventional definitive impressions for the fabrication of complete removable dentures in edentulous patients. The null hypothesis was that there would be no statistically significant difference among the study groups.

2. Methods and Materials

This study was approved by the Ethics Committee of the Tehran University of Medical Sciences under the ethics code IR.TUMS.DENTISTRY.REC.1401.024. All patients were fully informed about the study and gave their informed consent before being included. The patients referred to the Prosthodontic Department of the Dental School at the International Campus of Tehran University of Medical Sciences between March 2022 and September 2023, who were candidates for complete dentures, were considered eligible to be included in the study.

The initial step involved a comprehensive screening process to determine whether potential participants satisfied the predefined inclusion criteria: complete edentulism in either the maxilla or mandible, accompanied by moderate bone resorption (Cawood and Howell Class III [31]), absence of local infectious lesions, age of 21 years or older, and the absence of systemic diseases [29]. Exclusion criteria included the presence of teeth or root remnants (exposed or submerged), lack of understanding of the study's objectives and procedures, and refusal to provide informed consent [32]. Consequently, following the preliminary screening and based on the estimated sample size, ten patients were enrolled; six were fully edentulous in both maxillary and mandibular arches, while two were edentulous in the maxillary arch and another two in the mandibular arch. This resulted in eight maxillary and eight mandibular arches being examined.

All ten patients underwent three impression‐making methods, all performed by a single experienced prosthodontist (S.H.), resulting in three distinct study groups (n = 16 arches), as follows.

Group 1—Direct intraoral scanning (IOS): An optical scan was made from patients' edentulous upper or lower jaws using an intraoral scanner (CS 3800, Carestream Dental, Atlanta, GA, USA) by following a predetermined scan strategy [29]. Lower jaw scans began at the right retromolar pad, tracing the jaw's anatomy in a zig‐zag pattern to the opposing side (Figure 1A). Upper jaw digitization began at the right tuberosity, proceeding in a zig‐zag manner to the anterior region (Figure 1B) [29]. All scans were exported in standard tessellation language (STL) format (Figure 2).

Figure 1.

(A) Scan strategy in the maxilla; the black dotted line represents the primary scan path, the green line the additional anterior path; (B) scan strategy in the mandible; the black dotted line represents the primary path, the green line the additional scan paths of the labial, buccal, and lingual vestibules.

Figure 2.

Group 1 intraoral digital scan. Representative intraoral digital scans of the edentulous maxillary and mandibular arches acquired using the IOS device.

Group 2—Conventional definitive impression using 3D‐printed special tray: Using the previously generated STL files from Group 1, 16 special trays were first designed in the Exocad software program (Exocad DentalCAD; Exocad GmbH) (Figure 3A,B) and then fabricated using a biocompatible Class I resin material (NextDent Tray, NextDent B.V., Soesterberg, The Netherlands) and a DLP 3D printer machine (Digident, Mobtakeran Mechatronic Ark, Tabriz, Iran) (Figure 3C,D). The trays were then cleaned for 3 min in ethanol (> 90%) to remove any excess resin, using an ultrasonic bath, and finally post‐cured at 60°C for 10 min. Custom tray functionalization or border molding was performed using Impression Compound Green Sticks (Kemdent, Swindon, United Kingdom), and definitive impressions were made using a zinc oxide‐eugenol‐free impression paste (Cavex Outline; Cavex Holland B.V., Haarlem, The Netherlands). The resulting impressions were then digitally scanned using the same intraoral scanner used in the first group (Figure 3E,F).

Figure 3.

Group 2 conventional definitive impression with 3D‐printed customized tray. (A) Digital design of the maxillary tray; (B) digital design of the mandibular tray; (C) 3D‐printed maxillary tray; (D) 3D‐printed mandibular tray; (E) digitalized maxillary impression; (F) digitalized mandibular impression.

Group 3—Conventional definitive impression: In the conventional method, initial impressions were made using rimlock stainless steel stock trays and irreversible hydrocolloid impression material (Blueprint Xcreme, Dentsply Sirona, Konstanz, Germany) and then poured using a Type III dental stone (Elite Model, Zhermack, Badia Polesine, Italy). Special trays were fabricated with a visible light‐cure custom tray material (Neoloy, DENTSPLY International Inc., York, USA) on the corresponding stone models. Functionalization of custom trays was executed with Impression Compound Green Sticks, and definitive impressions were made using the same zinc oxide‐eugenol‐free impression paste (Figure 4A,B). The definitive impressions were then digitally scanned using the same intraoral scanner from the initial group (Figure 4C,D). In Groups 2 and 3, a direct scan of the impression was chosen due to the anticipated reduction in shadowing effect resulting from the lack of undercuts and voids in edentulous arches [33].

Figure 4.

Group 3 conventional definitive impression with acrylic special tray. (A) Conventional maxillary impression; (B) digitalized maxillary impression; (C) conventional mandibular impression; (D) digitalized mandibular impression.

For each patient, the IOS data set and the scans of the corresponding conventional definitive impressions from Group 2 (CI3D) and Group 3 (CI) were exported using the intraoral scanner's software (CS ScanFlow, Carestream Dental, Atlanta, GA, USA) in STL format, which is compatible with and importable into reverse engineering and inspection software for further analyses [34]. The CI3D and CI STL files were first reversed in Geomagic Design X (3D Systems, Rock Hill, SC, USA) to ensure correct surface orientation [35]. Subsequently, for each arch, the CI3D and CI data sets were individually superimposed onto the IOS reference using best‐fit alignment in Geomagic Control X (3D Systems, Rock Hill, SC, USA) (Figure 5). All files were trimmed to eliminate peripheral areas not present in all files, as well as their nonmatching areas caused by practical aspects related to obtaining the IOS (mobile tissue stretching) and the conventional definitive impression (mobile tissue compression and folding at the margin of impression), which could have impaired alignment and, consequently, measurement accuracy [35]. Once aligned, the 3D distance deviation between IOS and CSs was quantified using the absolute value of the dimensional error (change) and the root mean square (RMS), calculated as the square root of the mean of squared point‐to‐point deviations between the data sets, as automatically computed by Geomagic Control X. The RMS value represents the cumulative deviation across the entire arch (Figure 5). The cumulative RMS‐based deviation was selected as it provides an overall quantitative representation of surface accuracy, minimizing the influence of localized discrepancies such as border misalignments, which may appear as blue areas in color maps. This method has been consistently applied in prior investigations assessing 3D trueness of intraoral and conventional scans [15, 36, 37]. In all analyses, the IOS data set was set as the reference, while CI3D and CI data sets were treated as the measured data sets. Negative deviations indicated that the measured data set surface was positioned inward relative to the IOS reference, whereas positive deviations indicated an outward position. Color deviation maps were generated in Geomagic Control X using a standardized scale ranging from –1.0 mm to +1.0 mm. Green regions represented deviations within the clinically acceptable threshold (± 0.1 mm), blue indicated inward deviations, and red indicated outward deviations relative to the IOS reference data set (Figure 5).

Figure 5.

Representative superimposition of intraoral and conventional definitive impressions. (A) Maxillary arch superimposition showing the spatial deviations between the intraoral scan and conventional impression. (B) Mandibular arch superimposition illustrating the deviation patterns between the intraoral scan and conventional impression. Green areas indicate deviations within the clinically acceptable threshold (± 0.1 mm), while blue and red areas represent localized inward and outward deviations, respectively.

The required sample size was calculated based on Russo et al. [38] using PASS11 software (NCSS, Kaysville, UT, USA) for paired samples. Assuming a significance level (α) of 0.05, power (1 − β) of 80%, and a minimum detectable mean difference of 0.1 mm in dimensional deviation, the estimated sample size was 16 arches. Statistical analysis was performed using SPSS version 26.0 (IBM Corp., Armonk, NY, USA) and Python v3.11 (SciPy and Seaborn libraries) for supporting data visualization. All data sets were tested for normality using the Shapiro–Wilk test. To assess whether the type of conventional definitive impression technique influenced dimensional accuracy relative to the IOS reference, paired‐sample t‐tests were used to compare deviations between CI3D and CI. Comparisons were made for both mean absolute deviation and RMS deviation. Effect sizes (Cohen's d) were calculated to evaluate the magnitude of differences, and 95% confidence intervals (CIs) were reported to indicate measurement precision. For exploratory analyses, subgroup comparisons were performed between maxillary and mandibular arches, and correlations between mean and RMS deviations were assessed using the Pearson correlation coefficient (r). A two‐tailed significance threshold of p < 0.05 was adopted for all tests. Descriptive statistics (mean, standard deviation, and 95% CI) were reported for all comparisons. Boxplots were generated to visually compare the distributions of dimensional deviations across techniques. The overall workflow of the study, from data acquisition to deviation analysis and outcome evaluation, is illustrated in Figure 6.

Figure 6.

Workflow of the study illustrating the sequential procedures from data acquisition through data set alignment and 3D deviation analysis to outcome evaluation.

3. Results

Ten patients (six females and four males) with a mean age of 65.4 ± 7.2 years (range 56–78 years), all of Iranian ethnicity, were enrolled in this clinical study. In total, 16 edentulous arches were examined, including 8 maxillary (8/16) and 8 mandibular arches (8/16). For each arch, three sets of digital data were acquired: an intraoral scan (IOS), a scan of a definitive impression made using a custom 3D‐printed tray (CI3D), and a scan of a definitive impression made with an acrylic special tray (CI). For each arch, the CI3D and CI data sets were independently superimposed onto the IOS reference data set using best‐fit alignment in the metric software. This process generated two sets of deviations, IOS versus CI3D and IOS versus CI, from which the absolute mean deviation values and RMS deviations were calculated to quantify dimensional accuracy.

To assess whether the type of conventional definitive impression technique influenced dimensional fidelity relative to the IOS reference, paired t‐tests were performed on deviation values obtained from IOS–CI3D and IOS–CI. The Shapiro–Wilk test confirmed normality of the data distribution. Paired t‐tests revealed no statistically significant differences between CI3D and CI for either mean deviations (−0.11 ± 1.62 mm, 95% CI: −0.97 to 0.78 mm; p = 0.75, Cohen's d = 0.07) or RMS deviations (−0.10 ± 0.68 mm, 95% CI: −0.46 to 0.56 mm; p = 0.93, Cohen's d = 0.15), indicating negligible effect sizes and comparable accuracy relative to the IOS reference (Figures 7 and 8). Full results are summarized in Table 1. Negative deviation values represent regions where the surface of the measured data set (CI3D or CI) was located inward relative to the IOS reference, while positive deviations indicate outward positioning. This distinction facilitates interpreting the spatial distribution of deviations across the arch surfaces. The deviation color maps (Figure 5) demonstrated that most surface areas appeared green, indicating dimensional differences within the clinically acceptable threshold (± 0.1 mm), while limited blue and red areas represented localized inward and outward deviations, respectively [39].

Figure 7.

Comparison of average dimensional deviations (mean) for IOS–CI and IOS–CI3D superimpositions in maxillary and mandibular arches.

Figure 8.

Comparison of RMS dimensional deviations for IOS–CI and IOS–CI3D superimpositions in maxillary and mandibular arches.

Table 1.

Comparison of dimensional accuracy between conventional impressions relative to intraoral scan.

Metric	Mean difference (mm)	Standard deviation (mm)	t	p	95% CI lower	95% CI upper	Shapiro–Wilk p
Mean difference (IOS/CI − IOS/CI3D)	−0.11	1.62	−0.27	0.75	−0.97	0.78	0.23
RMS difference (IOS/CI − IOS/CI3D)	−0.1	0.68	−0.58	0.93	−0.46	0.56	0.93

Abbreviations: CI, scan of final impression with conventional acrylic special tray; CI3D, scan of final impression with 3D‐printed tray; IOS, intraoral scan (reference); RMS, root mean square.

Subgroup analysis revealed no statistically significant differences between maxillary and mandibular arches in terms of dimensional accuracy. For maxillary arches, the mean deviation for IOS–CI was −0.11 ± 2.63 mm (95% CI: −2.31 to 2.09 mm) and for IOS–CI3D was −0.55 ± 1.44 mm (95% CI: −1.76 to 0.66 mm; paired t‐test, p = 0.55). RMS deviations were 2.44 ± 1.02 mm (95% CI: 1.59–3.29 mm) for IOS–CI and 2.43 ± 1.06 mm (95% CI: 1.56–3.30 mm) for IOS–CI3D (p = 0.96). For mandibular arches, the mean deviation for IOS–CI was 0.10 ± 1.56 mm (95% CI: −1.20 to 1.40 mm) and for IOS–CI3D was 0.32 ± 1.72 mm (95% CI: −1.12 to 1.76 mm; p = 0.60). RMS deviations were 2.51 ± 0.63 mm (95% CI: 1.99–3.03 mm) for IOS–CI and 2.32 ± 0.62 mm (95% CI: 1.80–2.84 mm) for IOS–CI3D (p = 0.52). Additionally, a moderate positive correlation was observed between the mean deviation for IOS–CI and RMS deviation for IOS–CI3D (r = 0.31, p = 0.21), indicating a trend where subjects with greater mean discrepancies in the acrylic special tray group also tended to show greater overall deviation in the 3D‐printed tray group, although this was not statistically significant. No strong outliers were observed. Boxplot visualizations (Figure 9) confirmed the high similarity between IOS–CI3D and IOS–CI deviations, supporting the main finding that both conventional techniques produced comparable accuracy relative to intraoral scanning.

Figure 9.

Boxplot comparison of dimensional deviations between CI (conventional tray) and CI3D (3D‐printed tray), both referenced to IOS. (Left Panel) Mean deviation (mm); (Right Panel) RMS deviation (mm). No statistically significant differences were found (p > 0.05 in both comparisons).

4. Discussion

This clinical study investigated whether dimensional discrepancies between intraoral scanning (IOS) and two conventional definitive impression techniques—one using a 3D‐printed custom tray (CI3D) and the other using an acrylic special tray (CI)—differed significantly in fully edentulous arches. The results confirmed that both conventional techniques produced comparable deviations when referenced to IOS, thereby supporting the null hypothesis. No significant differences in surface deviation were found between CI3D and CI, and this equivalence held across both maxillary and mandibular arches.

The similarity in outcomes between CI3D and CI can be attributed to multiple clinical and technical factors. First, edentulous arches present unique challenges for capturing soft tissue details [13, 40, 41], but also offer large, continuous tissue surfaces, which reduce the risk of stitching errors in digital scanning and increase the reliability of conventional definitive impression capture [42]. When all scans are referenced to IOS and aligned using robust surface‐matching algorithms, local deviations due to tray type or impression material are largely normalized, resulting in comparable global deviations. Moreover, the use of border‐molded custom trays in both CI3D and CI—albeit through different fabrication methods—may have equalized their performance [41, 43, 44]. Border molding is essential in capturing functional depth and peripheral extensions of soft tissues [45], and in this study, it was standardized across techniques using impression compound sticks. Therefore, any theoretical advantages of the 3D‐printed tray in precision or fit may have been neutralized by this common step [45]. Additionally, the scanning process applied to both definitive impressions (CI3D and CI) was conducted with the same intraoral scanner, under the same operator and environmental conditions, further minimizing technique‐dependent variability [25, 46]. Despite the generally increased difficulty of capturing mandibular edentulous arches [41, 44, 47], due to greater soft tissue mobility and fewer anatomical landmarks, no significant differences in dimensional deviation were observed between maxillary and mandibular scans in this study. This outcome may be attributed to the selection of patients with minimally to moderately resorbed ridges, which provided more stable tissue and reference points, as well as the use of standardized protocols, exclusion of peripheral areas prone to distortion, and alignment focused on stable tissue regions. However, it should be noted that functional impressions required for complete denture fabrication cannot yet be fully achieved through digital techniques. Current fully digital workflows either bypass this step or incorporate the digitalization of a conventional impression [48].

The use of color deviation maps provided a qualitative complement to the numerical RMS analysis by visually illustrating how surface deviations were distributed across the edentulous arches. The predominance of clinically acceptable deviation zones supports the reliability of both impression techniques, confirming that the quantitative findings are not limited to isolated measurement points but represent the overall surface accuracy. Localized under‐ or over‐contoured regions are expected due to the lack of distinct anatomical landmarks, soft tissue mobility, and surface reflectivity in edentulous areas, rather than methodological limitations. Previous studies have also highlighted the value of color deviation maps in validating dimensional accuracy assessments by offering a more intuitive understanding of the clinical relevance of deviations [49].

Although conventional definitive impressions are widely regarded as the clinical gold standard for edentulous cases [26], IOS was selected as the reference data set in this study to investigate its potential as a clinically reliable alternative. Recent evidence has shown that modern intraoral scanners achieve high trueness and precision in capturing edentulous arches [25, 50], supporting their suitability for use as a reference under controlled conditions. This methodological choice allowed for standardized alignment in Geomagic software while minimizing variability associated with impression materials and pouring processes. From a digital standpoint, intraoral scanning for edentulous arches has historically faced skepticism due to the lack of prominent anatomical landmarks, mobile mucosa, and difficulties in capturing vestibular extensions [13]. However, recent improvements in scanner optics, real‐time rendering algorithms, and antireflective calibration have enhanced their ability to capture soft tissue accurately [17, 51]. In this study, the IOS device demonstrated high consistency when used as a reference model, which is supported by recent literature showing favorable accuracy of modern intraoral scanners even in full‐arch and edentulous applications [52, 53, 54].

The present findings, showing no significant difference in dimensional accuracy between intraoral scanning and conventional impressions, are consistent with previous investigations on edentulous arches. Li et al. [15] found comparable trueness and precision among various impression‐making methods, confirming that intraoral scanners can achieve accuracy equivalent to conventional techniques. Al‐Dulaijan et al. [36] similarly reported that the TRIOS scanner exhibited precision comparable or superior to vinyl polysiloxane impressions, depending on the device used. Kontis et al. [11] observed that while indirect digitization of impressions yielded the highest trueness, intraoral scanning still provided clinically acceptable results, with deviations mainly in peripheral regions—consistent with the localized deviations noted in this study. Likewise, Kahya Karaca and Akca [28] found no significant differences between digital and conventional impressions for maxillary complete edentulism, except minor discrepancies at the palate and borders. The findings of the present study also align with previous research by Russo et al. [26], who reported that dimensional deviations between intraoral scans and conventional definitive impressions in edentulous arches were minimal and clinically acceptable. Similarly, Cao et al. [55] concluded that, when conducted under controlled clinical conditions, both digital and conventional definitive impression techniques yielded comparable accuracy. Collectively, these results support that intraoral scanning offers clinically comparable accuracy to conventional impressions in completely edentulous arches, provided that appropriate scanning strategies are used and peripheral tissues are properly managed.

In contrast, other studies—such as those by Hack et al. [29], Patzelt et al. [22], and Chebib et al. [43]—have identified potential limitations in intraoral scanning of edentulous arches, particularly in areas with mobile mucosa, inadequate palatal seal capture, or significant alveolar ridge resorption. These anatomical and physiological factors may compromise the accuracy of IOS data. The present study addressed these concerns by trimming peripheral scan areas and focusing alignment on stable tissue zones, thereby minimizing the influence of scanner access limitations or soft tissue distortion [47]. Moreover, variations in study design—including whether impressions or poured casts were scanned using desktop or intraoral scanners [41]—and the type of scanner used must be acknowledged as potential sources of heterogeneity. Schlenz et al. [30] observed that, although modern intraoral scanners achieved mean full‐arch deviations under 100 µm—generally considered clinically acceptable [50]—notable limitations still exist, particularly in long‐span or full‐arch applications. It should be noted, however, that their study was conducted in dentate or partially edentulous jaws with short edentulous spans, which are not directly comparable to fully edentulous cases. Finally, the year of publication must also be considered when interpreting differences among studies. The continuous evolution of optical technologies has significantly improved the accuracy of newer IOS systems, as supported by recent literature involving full‐arch implant‐ and mucosa‐supported prostheses [25, 28, 50, 56, 57, 58].

A possible explanation for the slightly better consistency seen in the IOS–CI comparison may be the broader mucostatic effect of conventional acrylic special trays [59]. Their passive extension and less intimate adaptation might introduce a more uniform compression of soft tissue, resulting in smoother contours that align more easily with digital scans. Meanwhile, custom 3D‐printed trays, despite being tailored, might produce slight over‐adaptation in localized zones, especially if the digital tray design was not adequately offset or if material shrinkage occurred during printing and curing [60]. These subtle discrepancies, although not statistically significant, may influence tissue distortion and explain the minor variations observed in deviation trends [60].

Clinically, these results reaffirm the validity of IOS as a primary method for impression‐making in edentulous patients, particularly when combined with standardized scanning protocols. The elimination of impression materials, tray fitting, and physical storage adds efficiency to the clinical workflow [36, 51]. Additionally, for prosthodontic treatment planning and digital denture fabrication, the ability to obtain accurate digital models directly from the patient can reduce chairside time and improve patient comfort [45, 61]. Importantly, this study suggests that when a digital approach is not feasible, either conventional technique, whether based on 3D‐printed or acrylic special trays, can be reliably used, without compromising dimensional fidelity.

Based on the statistical findings, no significant differences in dimensional accuracy were detected between CI and CI3D relative to the IOS reference. Therefore, the null hypothesis was accepted. Nonetheless, this study has several limitations. The relatively small sample size, although statistically justified, may have reduced sensitivity for detecting subtle differences, particularly in specific anatomical regions. The analysis was restricted to impressions and digital scans, without assessment of the complete prosthetic workflow; therefore, potential discrepancies in denture adaptation and retention resulting from the fabrication process were not evaluated. The patient cohort included only individuals with minimal to moderate bone loss, limiting generalizability to cases with severe ridge resorption. Regional deviation analysis was not performed, preventing identification of site‐specific discrepancies (e.g., ridge crest, vestibule, palate). Additionally, the intraoral scan was selected as the reference standard, despite known challenges and potential limitations in capturing fully edentulous jaws; thus, results may differ with other reference models or in the presence of significant soft tissue mobility. Finally, the use of a single scanner type and operator, while standardizing methodology, may restrict applicability to different devices, clinical conditions, or operator experience levels.

5. Conclusion

Within the limitations of this clinical study, which included edentulous ridges with minimal to moderate bone loss (Cawood and Howell Class III):

• 1.The dimensional accuracy of conventional definitive impressions made with 3D‐printed custom trays and conventional acrylic special trays was comparable when evaluated against intraoral scanning in edentulous arches.
• 2.No statistically significant differences were observed between intraoral scanning and either conventional technique, in both maxillary and mandibular subgroups.
• 3.Intraoral scanning can be potentially considered a clinically reliable alternative for capturing edentulous arch morphology in the context of complete denture fabrication.

Author Contributions

Conceptualization: Hassan Azangoo Khiavi and Sareh Habibzadeh. Methodology: Sareh Habibzadeh, Hassan Azangoo Khiavi, and Wasan Abdulsamad Rahman. Software: Seyed Ali Mosaddad. Validation: Sareh Habibzadeh and Hassan Azangoo Khiavi. Formal analysis: Wasan Abdulsamad Rahman. Investigation: Sareh Habibzadeh and Wasan Abdulsamad Rahman. Resources: Hassan Azangoo Khiavi and Sareh Habibzadeh. Data curation: Sareh Habibzadeh and Seyed Ali Mosaddad. Writing – original draft preparation: Wasan Abdulsamad Rahman and Sareh Habibzadeh. Writing – review and editing: Seyed Ali Mosaddad and Sareh Habibzadeh. Visualization: Seyed Ali Mosaddad and Wasan Abdulsamad Rahman. Supervision: Hassan Azangoo Khiavi. Project administration: Sareh Habibzadeh. All authors have read and approved the published version of the manuscript. All authors have read and approved the final version of the manuscript. Sareh Habibazdeh had full access to all the data in this study and takes complete responsibility for the integrity of the data and the accuracy of the data analysis.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not‐for‐profit sectors. No financial relationships or conflicts of interest are declared. As no external funding was involved, the funding source had no role in the study design, data collection, analysis, interpretation, writing of the manuscript, or the decision to submit it for publication.

Disclosure

The lead author (Hassan Azangoo Khiavi) affirms that this manuscript is an honest, accurate, and transparent account of the study being reported; that no important aspects of the study have been omitted; and that any discrepancies from the study as planned have been explained.

Ethics Statement

The Research Ethics Committee of Tehran University of Medical Sciences authorized the present study (IR.TUMS.DENTISTRY.REC.1401.024).

Consent

All patients who participated in the research provided written informed consent after being informed about the study's details.

Conflicts of Interest

The authors declare no conflicts of interest.

Khiavi H. A., Rahman W. A., Mosaddad S. A., and Habibzadeh S., “Comparison of Three‐Dimensional Accuracy Between Digital and Conventional Definitive Impressions of Edentulous Arches: A Clinical Study,” Health Science Reports 9 (2026): 1–11, 10.1002/hsr2.71775.

Data Availability Statement

The data supporting this study's findings are available from the corresponding author upon reasonable request.