Evaluation of relative deviations in virtual articulators for simulating occlusal contacts at increased vertical dimensions

Abstract

PURPOSE

Accurate planning for occlusal vertical dimension (OVD) increase is crucial for prosthodontic success. This study evaluates the effects of different OVD increase amounts (2 mm or 5 mm), methods (articulator-based or intraoral) and virtual facebow use on occlusal contacts.

MATERIALS AND METHODS

Twenty-one individuals were initially recruited; three dropped out, leaving 18 participants. Digital impressions, bite records in maximum intercuspal position (MIP) and facial scans were obtained. Each participant received eight occlusal splints, varying by OVD increase method, amount and facebow use. Centric relation contacts and MIP were recorded using articulator paper and a scannable bite material. These scans were superimposed with Group D (2 mm intraoral increase without facebow) serving as the reference baseline. Deviations in the X (lateral), Y (anteroposterior) and Z (superoinferior) directions were analyzed using non-parametric tests with Bonferroni correction for pairwise comparisons.

RESULTS

No significant deviations were observed in the X-direction. However, significant deviations occurred in the Y and Z-directions in certain groups, particularly with a 5 mm intraoral increase. Effect size analysis indicated that most Y-direction differences were minor, whereas Z-direction deviations at higher OVD increases were more clinically relevant.

CONCLUSION

Regardless of the method, increasing the OVD by 2 or 5 mm and using a virtual facebow did not significantly affect the number of occlusal contacts. However, spatial deviations of contacts were detected, with articulator-based 5 mm increase producing fewer deviations than intraoral methods. Within the limitations of this study, the virtual facebow did not substantially reduce deviations.

INTRODUCTION

One of the most important prerequisites for aFatih Istanbul succesfull prosthetic rehabilitation is evaluating the occlusal vertical dimension (OVD) and acquiring accurate maxillomandibular relationships.¹,² The OVD can decrease over time due to factors such as aging, tooth loss, severe attrition or erosion of the teeth or a deep overbite. In such cases, prosthetic restoration of the OVD is performed to achieve ideal occlusal relationships and dentofacial harmony.³,⁴

The OVD can be restored by intraorally recording the desired increase with a de-programmer (anterior jig) or tinfoil, and transferring the records to an articulator. Alternatively, a bite registration can be taken in the maximal intercuspal position (MIP) and adjusted on an articulator using the incisal pin. However, both techniques have different limitations. The success of the intraoral record depends on factors such as the width of the scanned region (interocclusal distance), the type of device used and the clinician’s experience.⁵,⁶,⁷ On the other hand, the success of OVD restoration depends heavily on accuracy replicating the patient’s mandibular movements on the articulator. Traditionally, facebows have been indispensable in this process, as they transfer the spatial relationship of the maxilla to the hinge axis, allowing for individualized opening and closing curves on the articulator. However, advances in digital technology, particularly the development of virtual articulators have introduced new possibilities. Virtual articulators can also simulate mandibular movements and allow adjustments to the three-dimensional (3D) mounting position. This raises questions about the necessity of analog facebows.⁸,⁹ Additionally, face scanning systems act as simplified virtual facebows. They allow the transfer of maxillomandibular relationships to a digital environment while capturing the patient’s facial features, providing an alternative.¹⁰,¹¹

Despite numerous studies, the most accurate and precise digital method for complex prosthetic treatments, such as OVD restorations, is still unclear. Furthermore, clinical studies comparing virtual environments to real clinical conditions or evaluating the ability of virtual articulators to replicate occlusal contacts accurately are lacking. Currently, there is no consensus on whether OVD increases should be performed intraorally or on semi/fully adjustable articulators, or on the necessity of using a facebow during these procedures. Based on these considerations, this study’s null hypothesis is that, for prosthetic restorations requiring an OVD increase, the use of a virtual facebow, the amount of the OVD increase and the method of implementation (intraorally or on an articulator) do not improve the clinical workflow or affect occlusal contacts.

MATERIALS AND METHODS

This study was conducted in accordance with international laws on clinical experimentation and the research protocol was approved by the Istanbul Medipol University Ethics Committee with protocol number 10840098-604.01.01-E.9705. The principles of the Declaration of Helsinki were followed at all stages. Written informed consent was obtained to confirm voluntary participation.

The sample size was calculated using G*Power version 3.1.9.2. Calculations were performed for a one-tailed test with an expected Cohen’s effect size of 0.8, 80% power and a type I error rate of 5%.⁷ This analysis indicated that 16 participants were sufficient; however, to account for potential dropouts, the target sample size was increased to 21 participants. Ultimately, three participants could not complete the study for reasons unrelated to the study protocol (two withdrew due to scheduling difficulties and one was excluded because a gag reflex was triggered during splint placement, which prevented reliable measurements). Therefore, the study was completed with 18 participants. The participants were aged between 18 – 35 years old and were undergraduate students of Istanbul University Faculty of Dentistry. The participants consisted of 14 females and 4 males with Angle Class I occlusion, no missing teeth, and no temporomandibular disorders. All individuals demonstrated cooperation in performing mandibular movements.

To ensure consistent head positioning across all bite registrations, each participant was seated upright with the Frankfurt horizontal plane aligned parallel to the floor. A standardized dental chair with an adjustable headrest was used to provide posterior support and minimize movement. Participants were instructed to fixate on a designated visual point to control for head movement during record acquisition. The Trios 3 wireless intraoral scanner (3Shape A/S, Copenhagen, Denmark) was used to create virtual models of the dental arches. Scanning procedures followed the manufacturer’s recommended protocols. To obtain a buccal interocclusal record, a MIP scan was performed bilaterally from the canines to the molars in a downward sweeping motion and the resulting file was saved as “MIP-0”. To increase the OVD intraorally, tin foil (commonly used in Gerber’s Resilience Test) was folded to the desired thickness using a mechanical caliper (Mitutoyo, Kanagawa, Japan). The foil was folded to a thickness of either 2 mm or 5 mm and symmetrically positioned under the buccal cusps of the first premolars on both sides to achieve the desired increase. Each foil was measured at three different points with a digital caliper to verify uniform thickness prior to intraoral placement (2.0 ± 0.05 mm or 5.0 ± 0.07 mm). During bilateral manipulation, the foils were positioned symmetrically and occlusal contacts were visually confirmed to ensure parallel seating. In a subset of cases, the foil thickness was remeasured after removal, confirming minimal compressibility under occlusal load (deviation of less than 0.1 mm). These steps were taken to ensure reproducibility and consistent achievement of the intended vertical increase. This method enabled increasing the vertical dimension regardless of the overjet or overbite relationship in the anterior region. Occlusion was gently guided by bilateral manipulation and participant was instructed to relax their jaw muscles to ensure a repeatable centric relation (CR) position. A single experienced prosthodontist examiner (E.B) performed all clinical and scanning procedures. This approach was intended to minimize variability related to the operator across the protocol. After verifying the accuracy and quality of the occlusal records visually, two occlusal relationships were captured as STL files: One with a 2 mm increased vertical dimension (MIP-2) and one with a 5 mm increased vertical dimension (MIP-5).

Facial scans were performed using the AFT System One (AFT Dental System, SL, Sevilla, Spain), which includes a tablet-controlled 3D sensor camera (Bellus3D Face Camera Pro) and an intraoral transfer fork. The transfer fork is an intraoral piece with reference numbers and points that can be seen extraorally. The fork was stabilized with low-viscosity polyvinyl siloxane material (Aquasil Soft Putty and Aquasil Ultra LV, Dentsply Sirona, Charlotte, NC, USA) to ensure proper CR. Beyron’s arbitrary hinge axis points were located and marked bilaterally on the subject’s skin to ensure proper positioning of the models in the virtual articulator.¹² Each participant underwent a facial scan in controlled lighting. Scans were repeated three times to ensure quality. The scans were saved in .obj format. Subsequently, the bite fork was scanned using a desktop laboratory scanner (Amann Girrbach GmbH) and saved as a .stl file (Fig. 1).

Fig. 1. Three different bite records were performed. MIP-0 record was used for OVD increase on virtual articulators, while MIP-2 and MIP-5 were used in subgroups for intraoral OVD increases and virtual articulator adjustments (A). Digital bite record was captured using folded tin foil with thicknesses of 2 mm and 5 mm (B). Bite fork was set with polyvinyl siloxane material on the upper jaw and a facial scan was performed. To integrate the collected data, the bite fork was scanned using an intraoral scanner (C). MIP: Maximum intercuspal position.

Eight different occlusal splint (OS) designs were fabricated for each individual. The groups were categorized based on the method of vertical dimension adjustment method (articulator or intraoral), the OVD increase amount and the use of a digital facebow (Fig. 2). Groups A, C, E and G involved facebow transfer. The .obj and the .stl files were imported into the OS design module in Exocad 2.4 (Exocad GmbH, Darmstadt, Germany) and the face scan was aligned with the intraoral transfer piece. The maxillary and mandibular models were then positioned on an Artex CR (Amann Girbach AG, Koblach, Germany) semi-adjustable virtual articulator. Additionally, in groups A, B, E and F (where the vertical dimension was increased on the articulator); the bite registration recorded as “MIP-0 mm” was used as the baseline. The vertical dimension was increased by opening the incisal pin to obtain the required distance (2 or 5 mm) between the buccal cusp tip of the right mandibular first premolar and the central fossa of the right maxillary first premolar tooth. In groups C, D, G, and H (where the vertical dimension was increased intraorally), the recorded occlusal registrations were directly transferred to the virtual articulator. To allow a direct evaluation of the virtual facebow’s effect on static occlusion, the articulator’s dynamic occlusion features were not used; average values were used instead.

Fig. 2. Flowchart illustrating the characteristics of the eight occlusal splint groups. Groups were categorized according to the method of OVD increase (articulator-based vs intraoral), the amount of increase (2 mm vs 5 mm), and the use of a virtual facebow (VF) (+VF: facebow used; -VF: facebow not used).

The OSs were designed using CAD software with the following parameters: an offset of 0.1 mm and a minimum wall thickness of 1 mm. The designs were exported as .stl files and prepared for 3D printing using Preform version 3.6.1 (Formlabs, Sommerville, MA, USA). The OSs were printed and post-processed using Dental LT Clear Resin (Formlabs, Sommerville, MA, USA) on a Form 2 3D printer (Formlabs, Somerville, MA, USA), according to the manufacturer’s instructions. Each splint underwent a standardized quality control protocol prior to printing to verify design integrity. After post-processing, all OSs were visually and digitally inspected for signs of deformation or surface irregularities. Then, the fabricated OSs were placed intraorally to verify their stability and retention. Subjects were instructed to relax their jaw muscles, after which the mandible was guided into the CR by bilateral manipulation without applying any force. In this position, the occlusal contact points on the flat OS were recorded using 40-micron blue Artifol (Bausch, Cologne, Germany). Subsequently, without manual guidance, subjects were asked to gently close in MIP and these contacts were recorded with 40-micron red Artifol. The difference between the two markings clearly distinguished the two positions. The marked contact points were photographed and the number of occlusal contact points in each position was recorded. Scannable polyvinyl siloxane bite registration material (Occlufast Rock, Zhermack, Italy) was applied to the OSs and an occlusal record was taken in CR after a five-minute deprogramming period. To ensure accuracy, the mandibular position was verified by comparing multiple registrations. Once the material had set, it was carefully removed while ensuring that it remained attached to the OS. A powder (CAD-CAM Scan Spray White, MATKIMYA, Istanbul, Turkey) was applied to improve the scanning quality. The OSs were then scanned using the TRIOS 3 (3Shape A/S, Copenhagen, Denmark). Each OS was saved and exported as an STL file. They were transferred to the Geomagic Control X software (3D Systems, Rock Hill, SC, USA). All OSs were oriented in the same X, Y and Z directions. The relative deviations of the occlusal contact points were analyzed based on these digital records. The coordinate system was defined by a global best-fit registration of the control (Group D) and test OSs using Geomagic Control software. The origin was set at the centroid of the superimposed occlusal surfaces and the axes were oriented following the software’s best-fit alignment. Accordingly, the X, Y, and Z directions represented mediolateral, anteroposterior, and superoinferior displacements, respectively, relative to this standardized alignment.

Group D was designated as the control group and was superimposed on the OSs of the other groups. The ‘global registration’ function identified 10,000 points of alignment between the control and test groups. After alignment, the 3D comparison function generated color-coded maps with maximum and critical values of 0.1 mm and 0.01 mm, respectively. These color maps displayed the deviation patterns between the control and study models. For each subject, the most prominent occlusal contact points were identified bilaterally on teeth 4 – 7, corresponding to the highest cusp vertices or clearly marked occlusal contact areas. These points were consistently selected as measurement landmarks across all groups. Deviations of these landmarks were then calculated in the X (lateral), Y (anteroposterior), and Z (superoinferior) directions, relative to the standardized global best-fit alignment (Fig. 3K). To minimize intra-examiner variability, a single examiner performed all, following a standardized protocol. Prior to the main analysis, the examiner repeated the point selection procedure twice on a subset of cases and confirmed reproducibility. Thereafter, single measurements were consistently applied. For each subject, the deviation values obtained from the eight posterior teeth were averaged in each direction prior to statistical analysis. The resulting comparison data were compiled and statistically analyzed (Fig. 3).

Fig. 3. A-E. MIP-0 record was placed on a virtual articulator and the OVD was increased by 2 mm and 5 mm between the first premolars (A). Collected digital data was aligned with the intraoral scan and the bite fork (B-C), bite fork was further aligned with the facial scan data (D). Models were then combined with the facial scan and set according to anatomical landmarks of each subject for the virtual facebow groups (E). F-J. OSs were 3D printed (F), evaluated intraorally for both contact points (G) and bite record material was applied (H). Digital data was then obtained (I) and analyzed to measure distances using 3D comparison software (J). Schematic illustration of the coordinate system showing X (lateral), Y (anteroposterior), and Z (superoinferior) directions relative to the dental arch (K).

The data were analyzed using SPSS 25.0 (IBM Corp. Armonk, NY, USA). Descriptive statistics for continuous variables were calculated and reported as mean ± standard deviation (SD), along with the minimum, and maximum values. Nonparametric tests were used for group comparisons. The Friedman test was used to assess differences among the groups. When the Friedman test indicated a significant overall difference, post-hoc pairwise comparisons between groups were performed using the Wilcoxon signed-rank test. A Bonferroni correction was applied by dividing the significance threshold (P < .05) by the number of pairwise tests (n = 28), yielding an adjusted threshold of P < .0018. All pairwise results were evaluated against this corrected significance level. Additionally, Spearman’s rho correlation coefficients were calculated to assess potential associations between continuous variables. Effect sizes were calculated using Cohen’s d to complement statistical significance testing, and interpreted according to conventional thresholds (small: 0.2, medium: 0.5, large: 0.8). A significance level of P < .05 was used for all analyses.

RESULTS

Changes in occlusal contact points in the X, Y, Z directions were compared among the OS groups. Table 1 summarizes the descriptive statistics of these deviations, representing relative differences calculated using Group D (the control group) serving as the reference baseline.

Table 1. Relative deviations of occlusal contact points (in mm) in the X, Y and Z directions relative to the control group.

Directions	Groups	Mean	S.D	S.E	Minimum	Maximum
X	A	-0.0080	0.42763	0.03564	-1.71	1.95
	B	0.0160	0.46969	0.03914	-1.91	2.77
	C	0.325	0.33699	0.02808	-0.71	3.48
	E	0.290	0.51184	0.04265	-2.07	2.91
	F	0.0036	0.39821	0.03318	-0.90	3.51
	G	0.266	0.55631	0.04636	-1.35	3.85
	H	0.374	0.54728	0.04561	-1.47	3.97
Y	A	-0.089	0.26914	0.02243	-0.58	0.99
	B	0.0822	0.32698	0.02725	-1.06	1.31
	C	0.0141	0.27911	0.02326	-0.64	2.99
	E	0.0242	0.39463	0.03289	-0.63	2.72
	F	0.0261	0.31381	0.02615	-0.51	3.01
	G	-0.1185	0.53037	0.04420	-1.65	3.18
	H	-0.1005	0.53090	0.04424	-1.66	3.23
Z	A	0.3342	0.58395	0.04866	-1.03	2.96
	B	0.4361	0.64151	0.05346	-0.54	3.03
	C	-0.0316	0.33436	0.02786	-1.55	1.35
	E	0.1622	0.70237	0.5853	-1.35	3.39
	F	0.0314	0.46740	0.03895	-1.38	1.47
	G	-0.4386	0.70778	0.05898	-2.12	1.84
	H	-0.4667	0.74437	0.06203	-2.74	1.69

Values are presented as mean ± standard deviation (SD), standard error (SE), minimum, and maximum. Deviations represent relative differences calculated with Group D (2 mm intraoral increase without facebow) as the reference baseline.

No significant differences were found among the groups regarding occlusal contact point deviation in the X-direction (lateral direction) (P > .05) (Fig. 4). As shown in Table 1, Group F had the smallest mean deviation from the control group, while Group H had the largest. Effect size estimation confirmed that these differences were negligible (Cohen’s d values ranged from -0.10 to 0.05 versus the control group), indicating minimal clinical impact.

Fig. 4. Box-plot of X-direction (lateral) deviation (in mm) in occlusal contact points relative to the control group; circles represent mild outliers (data points beyond 1.5× interquantile range from the 25th or 75th percentile).; asterisks represent extreme outliers (data points beyond 3× interquantile range from the 25th or 75th percentile).

In contrast to the X-direction, significant differences were found among the groups for the Y-direction (anteroposterior direction) deviations (Fig. 5). Post-hoc analysis using Wilcoxon signed-rank tests (with Bonferroni adjustment) revealed differences between certain pairs of groups. Group C, which had the smallest Y-direction change, differed significantly from Group G (which had the largest anterior deviation). A negative mean value indicates that the occlusal contact in Group G shifted slightly posteriorly relative to the control. However, effect sizes were generally small (d < 0.3) except for groups G and H, which demonstrated moderate effects (d = 0.49 and d = 0.45, respectively). These results suggest that, despite statistical significance, most Y-direction deviations were minor in magnitude.

Fig. 5. Box-plot of Y-direction (anteroposterior) deviation (in mm) in occlusal contact points relative to the control group; circles represent mild outliers (data points beyond 1.5× inter-quantile range from the 25th or 75th percentile).; asterisks represent extreme outliers (data points beyond 3× inter-quantile range from the 25th or 75th percentile).

Deviations in the Z-direction (superoinferior direction) of the occlusal contact points also differed significantly among the groups (Fig. 6). Group C showed the least deviation, whereas Group H showed the greatest deviation. Group C showed a moderate effect size (d = 0.77), Group F showed a medium effect size (d = 0.57), and Groups G and H showed large effects (d > 1.1) relative to the control group.

Fig. 6. Box-plot of Z-direction (superoinferior) deviation (in mm) in occlusal contact points relative to the control group; circles represent mild outliers (data points beyond 1.5× inter-quantile range from the 25th or 75th percentile).; asterisks represent extreme outliers (data points beyond 3× inter-quantile range from the 25th or 75th percentile).

The Friedman test showed no statistically significant differences in the number of occlusal contacts observed in CR and MIP among the groups (P > .05) (Fig. 7). Additionally, no significant correlation was found between the magnitude of occlusal contact point deviations in any direction and the number of occlusal contacts. Therefore, these outcomes should be interpreted as relative deviations among groups rather than absolute accuracy values.

Fig. 7. Comparison of the number of occlusal contacts in CR and MIP across groups.

DISCUSSION

The accurate adjustment of OVD plays a critical role in prosthetic and restorative dentistry, yet the optimal method for achieving this adjustment remains a subject of debate. This study investigated alternative clinical workflows for restoring OVD and evaluated their relative advantages. The findings indicated that, although using a facebow did not provide significant benefits, there were clinically notable differences in occlusal contact deviations between intraoral and articulator-based approaches with varying degrees of OVD increase. Therefore, the null hypothesis was partially rejected.

Although the participants had full dentition with minimal occlusal wear, they provided a stable neuromuscular and occlusal baseline that allowed the effects of an increased OVD to be isolated. The OVD was reversibly increased using flat OSs to avoid permanent intervention. Since the aim was to examine how occlusal contact patterns respond to controlled increases of 2 and 5 mm, using symptom-free participants allowed for clearer evaluation of the occlusal contact points and the digital workflow accuracy.⁴,⁹ The 2 and 5 mm target OVD increases were selected based on clinical relevance and literature evidence regarding patient adaptability. A 2 mm increase is commonly applied, as it generally falls within the average freeway space of dentate individuals and is often sufficient to evaluate the neuromuscular response or gain minimal restorative clearance. A 5 mm increase represents a more pronounced change, yet remains within the range considered biologically acceptable. Increases between 2 and 5 mm have been reported to be well tolerated, with only minor adverse effects observed in the short term.⁷,¹³ Moreover, Abduo¹⁴ demonstrated that a 5 mm increase does not cause significant or irreversible changes in condylar position, reinforcing the biomechanical safety of such an intervention. These findings support the reliability of implementing an OVD increase of up to 5 mm.

The Trios system, used in this study, has been shown to produce more accurate interocclusal records than other digital scanners.¹⁵ It has been reported that multiple digital interocclusal records reduce the tilting effect for full-arch restorations.⁵,¹⁶ Therefore, bilateral interocclusal records were taken to minimize the tilting effect occurring on the contralateral side. However, it should be acknowledged that these digitally recorded occlusal contacts primarily represent static relationships measured under controlled conditions and may not fully correspond to dynamic clinical functions, such as mastication or parafunctional activities. While such static contacts are well-suited for evaluating the spatial accuracy of occlusal relationships following vertical dimension changes, dynamic assessments are necessary to determine functional reliability and clinical stability.¹,¹⁵ The Artex-CR virtual articulator was used to design OSs. Virtual articulators have been shown to be as accurate and precise as mechanical articulators with regard to static occlusal contacts. Increasing the OVD by opening the incisal pin was shown not to result in any significant differences between articulators.¹⁷,¹⁸

Occlufast Rock was applied to the OSs to analyze the 3D displacement of the occlusal contacts. This material was chosen due to its high hardness, dimensional stability and scannability, thus ensuring accurate and reproducible bite registrations. Although Occlufast CAD could have been used as an alternative, Occlufast Rock was shown to provide better reproducibility in locating occlusal contacts.⁹ To ensure dimensional stability across repeated measurements, all OSs were fabricated using Dental LT Clear resin, a biocompatible photopolymer known for its long-term mechanical durability and dimensional consistency.¹⁹ The original .stl files of each splint were digitally reviewed and compared with scan data obtained after applying the bite registration material to verify structural integrity. No significant deviations were observed, indicating that the OSs maintained their dimensional accuracy across repeated measurements. These procedures ensured that occlusal contact assessments were not affected by potential material distortion. Additionally, all key procedural steps—including vertical dimension increase, scan data acquisition, and occlusal contact registration—were validated using mechanical measurements, standard manufacturer protocols, and digital post-processing verification to ensure methodological integrity across the workflow.

Although compressible materials, such as tin foil may appear less stable than rigid jigs, using them for bite registration has advantages when it comes to preserving the physiological nature of mandibular movements. When placed bilaterally, the foil allows the mandible to move freely under neuromuscular control, without being restricted or deflected by anterior stops. This is particularly important when evaluating changes in occlusal contact following increases in vertical dimension, as it minimizes the risk of forced condylar displacement and enhances the reproducibility of CR records. Previous studies have shown that tin foil-based bite registrations are as accurate as conventional methods, without significantly affecting condylar position or occlusal contact location. To minimize variability related foil’s compressibility, its thickness was verified with a caliper before each use and bilateral placement ensured symmetry. Nevertheless, small deviations due to material properties cannot be entirely excluded. Therefore, the bilateral foil technique provides a reversible, physiologically compatible, and methodologically sound alternative for controlled OVD increase.²⁰,²¹,²²

Group D was selected as the control group due to its clinical relevance and methodological clarity. It involved a 2 mm intraoral OVD increase, commonly within the freeway space of dentate individuals and is therefore considered a physiologically neutral and reversible condition. This configuration minimized potential sources of error from articulator-based mounting or facial scan alignment, enabling a focused evaluation of vertical changes alone under stable neuromuscular conditions. Previous studies have shown that CR records using 2 mm tin foil in the premolar region are highly reproducible and act as a neuromuscular deprogramming aid, enhancing registration accuracy.²¹,²² Similarly, Revilla-León et al.² demonstrated that maxillomandibular records obtained with a 2 – 3 mm interocclusal gap produced significantly better trueness and precision than records at MIP. Therefore, Group D provided the most physiologically and technically reliable baseline for evaluating the influence of added workflow components. Importantly, several studies have found no significant advantages of facebow usage in terms of occlusal contacts, patient comfort, or prosthetic stability in dentate individuals; in some cases, simplified approaches without facebow even yielded better outcomes.⁸,²³ Thus, omitting the facebow in the reference condition allowed us to establish a neutral and reproducible baseline while avoiding additional variables. Taken together, these factors justify selecting Group D as the most physiologically and technically reliable reference for evaluating the relative impact of facebow transfer and articulator-based versus intraoral OVD increases.

When the groups with intraoral OVD increases were compared with the control group separately, statistically similar behavior in the Y-direction was observed, regardless of the amount of increase. This finding indicated that maxillomandibular relationships were reproducibly maintained during intraoral recordings. Groups A, B, E, and F also showed comparable results, indicating that when the OVD was increased on the articulator, similar results were obtained in terms of occlusal contact point deviation, regardless of the use of a virtual facebow or the amount of increase. When the bite was registered in MIP with 2 mm and 5 mm increases, the occlusal contacts showed similarites, as they were determined by the same articulator trajectory. This demonstrated the consistency of the virtual articulator software. Groups E and F exhibited the highest deviations in the Z-direction; however, these deviations followed a more uniform and predictable pattern due to the consistent trajectory imposed by the virtual articulator. In contrast, Groups G and H, which used an intraoral OVD increase, showed more dispersed and multidirectional deviations across both Z and Y directions, indicating greater variability and positional instability. These results suggest that virtual OVD adjustments may offer greater clinical reliability despite the absolute magnitude of vertical deviation. The increased interocclusal distance during intraoral registration may reduce the accuracy of digital scanners, potentially introducing a tilting effect, or cause unintentional mandibular retrusion and rotation during the procedure.²,²⁴

In the Y-direction displacement, Groups C and F showed the least deviation from the control group, likely due to the use of identical articulator trajectories and individual mandibular opening-closing curves superimposed onto those of the control group. Conversely, Groups G and H showed the greatest occlusal contact deviations in the Y and Z directions. Although, Group G recorded the highest absolute Z-direction deviation (0.466 mm), the use of a virtual facebow in Group G did not significantly reduce this value (0.438 mm), suggesting limited efficacy of virtual facebow transfer in mitigating vertical discrepancies during intraoral OVD increases. While multiple group comparisons were statistically significant, effect size estimation revealed that most deviations—particularly in the X- and Y-directions—were associated with negligible or small effects (Cohen’s d < 0.3), indicating limited clinical impact. Only Groups G and H demonstrated moderate effects in the Y-direction, and moderate-to-large effects in the Z-direction (d = 0.57 – 1.20), suggesting that vertical deviations at higher OVD increments may be more clinically relevant. The observed micron-level differences remained within the clinically acceptable threshold of < 0.5 mm, as defined by the AAO,²⁵ and other reports have suggested that deviations below 1 mm are not critical.¹⁵ However, the consistent detection of moderate-to-large effect sizes in the Z-direction underscores that even subtle vertical alterations can systematically influence occlusal relationships. Thus, small OVD increases appear biomechanically safe and clinically stable, whereas larger increases (≈5 mm) demand careful monitoring and individual assessment to ensure long-term occlusal stability and patient comfort.

Collectively considering directional differences, the smallest deviations were consistently observed in the X-direction, reflecting the inherent stability of transverse mandibular relationships and the bilateral symmetry of the recording procedure. The Y-direction showed moderate variability, particularly in intraoral registrations, likely related to subtle mandibular retrusion, rotational effects, or tilting artifacts at increased vertical openings. The Z-direction demonstrated the greatest deviations, as vertical opening amplified scanner-related inaccuracies and the biomechanical sensitivity of mandibular repositioning. These findings suggest that while transverse and anteroposterior relationships remain largely intact during controlled OVD increases, vertical stability represents the most technique-sensitive dimension and requires careful clinical consideration.²⁶,²⁷

The smallest occlusal deviations were observed in Group C (0.014 mm in the Y-direction and 0.031 mm in the Z-direction), which supports the conclusion that a virtual facebow offers no clinical advantage when OVD changes do not exceed 2 mm. The use of a virtual facebow was not statistically significant across all groups, indicating that it can be considered unnecessary for healthy individuals with Angle Class I occlusion. However, the fundamental rationale for using a facebow is to maintain the precise distances from the cusp tips to the mandibular condyles.⁸ Nevertheless, the use of flat-surfaced OSs in this study may have limited the assessment of the facebow’s full impact, since such OSs do not replicate the typical cusp-fossa relationships seen in natural occlusion.

Morneburg and Pröschel²⁶ reported that an increase of 2 mm in OVD on an articulator resulted in occlusal deviations of up to 0.4 mm in premolars and molars. Meanwhile, an increase of 4 mm resulted in deviations of 0.3 – 0.8 mm in 44% of cases. These findings highlight the importance of identifying the true hinge axis as OVD increases. Although digital face scans provide more accurate data than analog facebows, determining the hinge axis remains critical.²⁸ The virtual facebow, which relied on the Beyron point as the average hinge axis, introduced limitations in reproducing individual hinge axes, which may have contributed to the observed deviations. However, previous studies suggest that deviations of less than 5 mm are not clinically significant.¹²,²⁹,³⁰ For future studies, an alternative to using a kinematic facebow is to place button-like markers on the upper and lower teeth during face scanning. The positional coordinates of these markers could then be recorded at different mandibular opening positions, allowing for more accurate tracking of mandibular trajectories and improving the resulting data.

In a study by Inoue et al.,⁴ no statistically significant differences were observed in vertical measurements between the gingival margins of premolars when the OVD was increased by 0, 3, or 5 mm. This aligns with our findings, suggesting that increasing the OVD by up to 5 mm on a virtual articulator is clinically acceptable for anatomically healthy patients without any variation, eliminating the need for additional procedures such as the use of a facebow.

There were no statistically significant differences in the number of occlusal contacts among the groups, indicating that as the amount of OVD increased, the use of facebow and the method of OVD increase did not significantly affect the number of occlusal contacts. Positional changes in occlusal contacts were independent of the number of occlusal contact points.

Diagnostic esthetic functional splints (DEFS), also referred to as Munich splints, aim to provide insights into occlusal design and aesthetics using CAD-CAM technologies without the need for tooth preparation.³¹ From this perspective, the present study serves as a pilot investigation into the effects of occlusal recording methods and the necessity of using a virtual facebow to define the horizontal and vertical relationships in DEFS fabrication. Due to various software limitations, flat occlusal surfaces were used instead of surfaces corresponding to the teeth’s morphology. Since the purpose of the facebow is to accurately transfer the distances between the tooth cusps and the mandibular condyles, using OSs with flat surfaces may have limited the facebow-related results.

While the digital workflow used in this study is efficient and repeatable, it has several limitations. Although facial scanning captures a static representation of the patient’s head and occlusal plane, it remains sensitive to postural variation, facial expression, and soft tissue tension. Despite standardized scanning conditions, minor discrepancies are inevitable.² Similarly, virtual articulators cannot replicate the full dynamics of mandibular movement, including muscle coordination or joint loading. They rely on static interocclusal records and a simulated hingeaxis, based on average anatomical values. Without true kinematic hinge-axis registration, deviations of 4 – 6° in condylar parameters have been reported, leading to clinically significant anteroposterior discrepancies at higher OVD increases.¹²,²⁶ Although this protocol does not claim micron-level precision, the consistency of the workflow across all groups, coupled with each participant serving as their own control, ensured that any procedural variability was evenly distributed. Thus, the observed differences in occlusal contact patterns likely reflect the true biomechanical consequences of OVD changes rather than inconsistencies introduced by measurement variability from the recording process. Furthermore, the study’s design, which involved eight different occlusal splints per participant, enabled within-subject comparisons of vertical dimension increases and the role of virtual facebow integration under consistent neuromuscular and anatomical conditions. This approach minimized inter-individual variability and allowed for the meaningful interpretation of how specific workflow components influence occlusal outcomes. While some reported deviations fall within clinically acceptable limits, their consistent statistical significance highlights the sensitivity of digital occlusal registrations to subtle protocol variations. Rather than advocating for clinical changes based solely on these findings, the results provide a methodological reference for understanding how incremental modifications to digital protocols may affect occlusal contact distribution.

Using an intraoral scanner (TRIOS 3) for reference model acquisition represents another limitation. Although laboratory scanners generally provide superior trueness and precision in full-arch scans, the intraoral scanner was chosen to reflect a clinically relevant digital workflow.³² Moreover, since all analyses were performed within-subject, each participant served as their own control, minimizing potential systematic bias. Nevertheless, this limitation should be considered when interpreting the absolute deviation values. It should be emphasized that the present study did not assess absolute accuracy, as no objective gold standard was available. Instead, all analyses were performed relative to Group D, which was chosen as a clinically relevant baseline rather than a validated reference of accuracy. Accordingly, the outcomes should be interpreted as relative deviations among groups, providing insight into intergroup variability rather than absolute measures of accuracy.

Another limitation is the potential for subjectivity when selecting a single prominent occlusal contact point per tooth. Although two preliminary trials confirmed consistency and all measurements were conducted by one calibrated examiner, the absence of formal reproducibility testing remains a constraint. Future studies should incorporate intraclass correlation coefficient (ICC) analysis or region-of-interest averaging to further improve reliability. In addition, for each subject, deviation values from the eight posterior teeth were averaged across directions prior to statistical analysis. This strategy ensured stable within-subject comparisons and enhanced the robustness of the results; however, such averaging may have obscured tooth-specific patterns of deviation. Future studies designed to perform tooth-level analyses could provide more detailed insights into site-specific variability.

This study included healthy participants of similar age groups with Angle Class I occlusion and minimal occlusal variation. Participants had no temporomandibular disorders, which ensured stable and reproducible CR records.¹,³³ Excluding patients with Angle Class II/III occlusion reduced variation in mandibular opening-closing trajectories and contributed to the close alignment between virtual face scan data and standard semi-adjustable articulator positioning. However, further studies incorporating dynamic occlusal simulations and with diverse samples including cases with temporomandibular joint disorders or developmental anomalies are needed to comprehensively evaluate occlusal deviations using virtual facebows and to further validate these results. Thus, while the homogeneity of the sample improved internal validity and protocol consistency, it inherently limits the external validity of the findings when applying them to more heterogeneous clinical populations.

CONCLUSION

Based on the results obtained from healthy participants with Angle Class I occlusion, the following conclusions were drawn: neither the use of a virtual facebow nor the method of increasing the OVD significantly affected the number of occlusal contacts for both 2 mm and 5 mm increases in OVD. Increasing the OVD by up to 5 mm using a virtual facebow based on facial scanning did not result in a clinically meaningful improvement in reproduction of occlusal contacts. For 2 mm OVD increases, intraoral increase resulted in the least occlusal deviation. In contrast, for 5 mm increases, articulator-based methods demonstrated greater accuracy, particularly in the anteroposterior and superoinferior directions. Overall, the magnitude of deviations was relatively small; however, their clinical relevance, particularly in long-term prosthetic outcomes, remains to be established. While the virtual facebow did not show a significant advantage under these experimental conditions, it may be valuable in more complex clinical cases and warrants further investigation.