Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2026 Jun 5;37(8):961–969. doi: 10.1111/clr.70140

Accuracy and Efficiency of Dynamic Computer‐Assisted Implant Surgery Using Three Different Registration Methods—A Laboratory Study

Toshiki Nojiri 1,✉, Kevser Pala 2,3, Ryo Yamamoto 1, Akihiro Fukutoku 1, Kazuhiro Kon 1, Shigemi Nagai 2, German O Gallucci 2
PMCID: PMC13446255  PMID: 42249609

ABSTRACT

Introduction

Accurate implant placement with Computer‐Assisted Implant Surgery (CAIS) is critical to ensure long‐term success. Dynamic‐CAIS systems enhance surgical precision through real‐time feedback, though comparative data on their accuracy and efficiency remain limited. This study evaluated how different registration methods in d‐CAIS systems influence implant placement accuracy and procedural time.

Materials and Methods

In this laboratory study, three registration methods were assessed: ND group, XC group, and XM group. Five experienced operators placed 25 implants per group (75 total) in partially edentulous maxillary models. Postoperative CBCT scans were used to measure deviations at the implant entry point, apex, vertical depth, and angle. One‐way ANOVA and Tukey's HSD test were used for statistical analysis.

Results

Mean 3D deviations at the implant entry point were 1.06 mm (ND), 0.71 mm (XC), and 0.90 mm (XM); at the apex 1.23 mm, 0.83 mm, and 1.07 mm, respectively. Vertical depth deviations at the apex were similar across groups: 0.61 mm (ND), 0.59 mm (XC), and 0.64 mm (XM). Angular deviation was highest in the ND group (2.96°) compared to XC (1.17°) and XM (1.02°), with a significant difference between ND and XM (p = 0.028). The average procedural time was shortest in the XM group (7.20 ± 2.48 min), though differences were not statistically significant.

Conclusions

Although registration methods minimally affect overall accuracy, system‐specific factors may influence angular deviation. Future studies should prioritize clinical trials with larger sample sizes and explore the influence of operator experience and learning curves.

Keywords: accuracy, computer‐assisted implant surgery, dynamic navigation

1. Introduction

Implant treatment is widely recognized as a reliable option for replacing missing teeth due to its high success rate and long‐term functionality (Albrektsson et al. 1986; Pjetursson et al. 2007, 2004). However, accurate implant placement is essential because surgery is performed near critical anatomical structures such as nerves, blood vessels, and the maxillary sinus (Fugazzotto et al. 2015; Juodzbalys et al. 2010, 2013; Kalpidis and Setayesh 2004; Lin et al. 2014). Moreover, implant positioning must correspond to the planned prosthetic restoration. To meet these requirements, computer‐assisted implant surgery (CAIS) has been introduced to improve the precision and safety of implant placement (Buser et al. 2004; Fortin et al. 2006; Jorba‐Garcia et al. 2025; Misch et al. 2001; Rocci et al. 2003; Schneider et al. 2009; Tahmaseb et al. 2014).

CAIS integrates radiographic data from cone beam computed tomography (CBCT), digital impressions, and prosthetic planning to create a virtual patient model, enabling accurate preoperative planning and identification of anatomical landmarks (Pedrinaci et al. 2024). This workflow facilitates a prosthetically driven implant positioning and allows transfer of the planned implant location to the patient with greater precision than freehand surgery. Two main approaches exist: static CAIS (s‐CAIS) and dynamic CAIS (d‐CAIS). In s‐CAIS, CAD/CAM‐generated surgical guides control osteotomy direction, depth, and position. According to the 2018 ITI Consensus, mean deviations were approximately 1.2 mm at the implant platform, 1.5 mm at the apex, and 3.5° angularly, remaining within clinical safety limits (Block and Emery 2016; Haemmerle et al. 2009; Kang et al. 2024; Wang, Shaheen, et al. 2022; Younis et al. 2024).

Despite its accuracy, s‐CAIS has limitations. Surgical guides can obstruct irrigation, increasing the risk of thermal bone injury during drilling, and friction between the drill and guide sleeve can reduce tactile feedback, limiting assessment of bone quality. Guide coverage can also restrict visibility, making verification of final implant depth difficult. In posterior regions, limited mouth opening may hinder guide placement and affect accuracy. Additionally, fabrication of patient‐specific guides increases cost and treatment time, and damage or misfit may require refabrication. Most notably, s‐CAIS relies on preoperative planning and offers limited intraoperative flexibility when anatomical variations or complications occur (Block and Emery 2016; Tahmaseb et al. 2014; Widmann and Bale 2006).

Dynamic CAIS (d‐CAIS), in contrast, does not require surgical guides but uses an optical tracking system with stereoscopic cameras that detect fiducial markers attached to the surgical handpiece and patient. Through triangulation algorithms, the system continuously displays the bur's position, depth, and angulation relative to the surgical plan (Block and Emery 2016). This approach maintains an unobstructed surgical field, allowing adequate irrigation and improved tactile feedback during drilling. Importantly, d‐CAIS enables real‐time adjustment of implant position, depth, and angulation, allowing clinicians to respond to anatomical variations or unexpected findings (Emery et al. 2016; Mandelaris et al. 2018). It is also advantageous in posterior regions or cases with limited mouth opening because no bulky guides are required, potentially reducing treatment time and costs while avoiding guide‐related complications. Consequently, interest in d‐CAIS has increased, and systematic reviews report clinically acceptable accuracy (Kang et al. 2024; Pellegrino et al. 2021; Wang et al. 2021; Wei et al. 2021).

A critical determinant of accuracy in d‐CAIS is patient registration, which aligns the preoperative virtual implant plan with the actual intraoral anatomy (Mandelaris et al. 2018; Pei et al. 2024; Wang et al. 2021; Wu et al. 2024). Each d‐CAIS platform employs distinct registration protocols to link the coordinate system of the optical tracking system with that of the 3D surgical plan (Block and Emery 2016; Emery et al. 2016; Mandelaris et al. 2018; Panchal et al. 2019). For example, the Navident system (ClaroNav, Toronto, Canada) uses pair‐point and surface registration, where a tracer tool records anatomical landmarks such as cusps or incisal edges. The X‐Guide system (X‐Nav Technologies, USA) provides two methods: X‐Mark, which traces tooth landmarks, and X‐Clip, a radiographic marker–based protocol in which a clip with radiopaque markers is attached to the arch before CBCT acquisition and automatically aligned during surgery through optical detection within the DICOM dataset.

Although these registration approaches differ technically, they all aim to ensure accurate correspondence between digital planning and clinical execution. However, comparative studies evaluating implant placement accuracy across different d‐CAIS registration protocols under standardized conditions remain limited. Therefore, the present study aimed to evaluate implant placement accuracy and time efficiency associated with different d‐CAIS systems and their respective patient registration methods under standardized in vitro conditions.

2. Materials and Methods

2.1. Study Design

This laboratory study was conducted at the Division of Removable Prosthodontics and Oral Rehabilitation, Department of Prosthodontics, School of Dentistry, Iwate Medical University. Tests were performed by five dentists, each with over 5 years of experience in implant treatment. Two d‐CAIS systems with a total of three registration methods were evaluated: (1) Navident (ND group), (2) X‐Guide with X‐Clip registration (XC group), (3) X‐Guide with X‐Mark registration (XM group). In each group, each operator placed five implants consecutively according to the assigned navigation system. Consequently, 25 implants were placed per group.

The primary outcome of this study was the implant placement accuracy. Additionally, the time required for implant placement was recorded, and the average time taken by each operator to place five implants was calculated as a representative value. The order of navigation system use was randomly assigned and applied to each operator in a pre‐determined sequence prior to the experiment. A total of 25 implants were placed in 25 models in each group.

2.2. Data Acquisition

Seventy‐five maxillary partially edentulous models were used for this study. The utilized models consisted of polyurethane resin‐based rigid foam. This material was produced by mixing polyurethane with foaming additives to approximate the hardness of bone and casting the mixture into silicone molds. The models were sufficiently radiopaque on CBCT to allow reconstruction of their morphology, and the implants could be clearly visualized for deviation measurements. CBCT imaging (KaVo OP 3D Vision, KaVo Dental GmbH, Germany) and extraction of DICOM data were performed for all models. In the XC group, before CBCT imaging, the thermoplastic resin of a dedicated device (X‐Clip) was adjusted to the model, and the X‐Clip was positioned to cover the right maxillary canine to the right maxillary central incisor on the model and used during CBCT imaging. Subsequently, an intraoral scanner (Trios4; 3Shape, Copenhagen, Denmark) was used to digitize the surface morphology of all models, and the data were exported in standard tessellation language (STL) format.

2.3. Data Registration and Implant Placement

The DICOM and STL data from each model were imported into the navigation system‐specific software (ND group: Navident 2.0, ClaroNav; XM and XC groups: Nobel DTX Studio, Nobel Biocare Services AG). A single operator, proficient in using the software, digitally planned the implant position and created an implant placement plan for an implant in the maxillary right first molar position (Figure 1). The model was mounted on a phantom head to simulate clinical conditions.

FIGURE 1.

FIGURE 1

Open in a new tab

Implant placement planning for each system: (a) Nobel DTX Studio; (b) Navident.

The stability of each instrument used for d‐CAIS was thoroughly verified to ensure that no unintended movement or instability occurred. The optical tracking system was positioned to maintain continuous and unobstructed tracking of all instruments. Subsequently, the patient‐specific physical model and the digitally planned implant position were registered according to the manufacturer's recommended registration method, which utilized either digital marker–based patient registration with fiducial markers (e.g., radiopaque metal spheres in X‐clip) or pair‐point registration based on clearly identifiable anatomical landmarks visible on both the analog model and the virtual planning data (Figure 2).

FIGURE 2.

FIGURE 2

Open in a new tab

Registration for each system: (a) The tracer traces the hard tissue surface. (b) The optical tracking system captures the movement of the tracer, and the software recognizes the surface morphology of the hard tissue. (c) Monitor during registration. (d) The X‐Mark probe tool traces the hard tissue surface. (e) The optical tracking system captures the position of the X‐Mark probe tool, and the software recognizes the surface morphology of the hard tissue. (f) Monitor during registration. (g) X‐Clip attached to the dentition. (h, i) The optical tracking system recognizes the fiducial marker (radiopaque metal ball) attached to the X‐Clip, and the software automatically performs registration.

For the pair‐point registration in the ND group, the right maxillary second molar, right canine, right central incisor, left second premolar, and left second molar were used as anatomical landmarks. In the XM group, anatomical landmark‐based registration was performed using the right maxillary second molar central fossa, first premolar distal marginal ridge, left lateral incisor distal angle, canine cusp, second premolar central fossa, and second molar central fossa. In the XC group, digital marker–based patient registration was used, in which the same fiducial marker device (X‐Clip), that was attached during CBCT imaging, was positioned identically on the model, covering the right maxillary first premolar to the right central incisor. Instrument calibration for both the drill axis and drill tip was performed prior to each procedure. Implant osteotomy and implant placement were then carried out under d‐CAIS. Before each procedure, the instrument tip was used to verify calibration accuracy by touching specific surfaces of the model's dentition. A total of 75 implants (Nobel Parallel CC RP Ø4.3 × 13; Nobel Biocare Services AG, Zurich, Switzerland) were inserted. The time required for implant placement, excluding the setup process such as registration, was measured. For each implant, the time recording started at the beginning of drilling with the first drill and ended when the implant was fully seated in its final position.

2.4. Measurement of Results

After implant placement, postoperative CBCT scans were obtained for each model. The resulting DICOM datasets were imported into the corresponding analysis software. The preoperative planning data and the postoperative DICOM data of the placed implants were superimposed to evaluate implant placement accuracy. For all placed implants, the following deviation parameters between the planned implant position and the actual implant position were measured: three‐dimensional deviation at the implant entry point, three‐dimensional deviation at the implant apex, vertical depth deviation at the implant apex, and angular deviation (Figure 3). All measurements were performed by a single operator who was blinded with regard to the group allocation.

FIGURE 3.

FIGURE 3

Open in a new tab

Measurement parameters for deviation between preoperative implant placement simulation and actual implant positioning: ① Three‐dimensional deviation at the implant entry point; ② Three‐dimensional deviation at the implant apex; ③ Vertical depth deviation; ④ Angular deviation.

2.5. Statistical Analysis

The normality of each group's data was planned to be assessed using the Shapiro–Wilk test. If normality was confirmed (p > 0.05), parametric testing using one‐way analysis of variance (ANOVA) would be conducted to compare implant placement accuracy and procedure time among the three groups (ND, XC, and XM). In the event of statistically significant differences (p < 0.05), Tukey's Honestly Significant Difference (HSD) test would be applied for post hoc multiple comparisons. If the assumption of normality was violated, nonparametric alternatives were applied. Specifically, the Kruskal–Wallis test was used to assess overall differences among the three groups, followed by pairwise comparisons using the Mann–Whitney U test with Bonferroni correction to control for multiple testing. Each operator placed five implants per group using the assigned navigation system, and the mean deviation and placement time per operator were used as representative values for statistical analysis. All results are expressed as mean ± standard deviation (SD). For visualization purposes, box‐and‐whisker plots (showing the median, Q1, and Q3) are included as Figures 4, 5, 6 to illustrate the data distribution.

FIGURE 4.

FIGURE 4

Open in a new tab

The deviations of measurement parameters. Box plots represent the deviations for each group (XC, XM, ND) across the measurement parameters. The lower and upper bounds of each box correspond to the first (Q1) and third (Q3) quartiles, respectively. The horizontal lines within each box indicate the median, and the “X” markers represent the mean values. Whiskers show the minimum and maximum values within each group. (a) Three‐dimensional deviation at the implant entry point. (b) Three‐dimensional deviation at the implant apex. (c) Vertical depth deviation.

FIGURE 5.

FIGURE 5

Open in a new tab

Angular deviation across groups. Box plots represent the deviations for each group (XC, XM, ND) across the measurement parameters. The lower and upper bounds of each box correspond to the first (Q1) and third (Q3) quartiles, respectively. The horizontal lines within each box indicate the median, and the “X” markers represent the mean values. Whiskers show the minimum and maximum values within each group.

FIGURE 6.

FIGURE 6

Open in a new tab

Comparison of implant placement time across groups. Box plots represent the deviations for each group (XC, XM, ND) across the measurement parameters. The lower and upper bounds of each box correspond to the first (Q1) and third (Q3) quartiles, respectively. The horizontal lines within each box indicate the median, and the “X” markers represent the mean values. Whiskers show the minimum and maximum values within each group.

All statistical analyses were performed using Python (statsmodels version 0.14.1; SciPy version 1.13.0). One‐way ANOVA and post hoc Tukey HSD tests were conducted using statsmodels, while the Shapiro–Wilk test for normality was performed using scipy.stats. The statistical analyst was blinded to group assignments to minimize bias during data evaluation.

2.6. Sample Size and Power Analysis

A post hoc power analysis was performed using G*Power software (version 3.1.9.7) to evaluate the adequacy of the sample size for detecting statistically significant differences among the three groups. The analysis was based on a one‐way ANOVA (fixed effects, omnibus) with the following parameters:

  • Effect size (Cohen's f): 0.40

  • α error probability: 0.05

  • Total sample size: 75 (25 implants per group)

  • Number of groups: 3

The achieved statistical power was calculated to be 0.869, which exceeds the commonly accepted threshold of 0.80. This result indicates that the sample size used in the study was sufficient to detect moderate to large effect sizes. Among the four outcome parameters assessed—coronal, apical, apico‐coronal (depth), and angular deviations—this analysis was based on angular deviation, as it was the only parameter that demonstrated a statistically significant difference among groups. Therefore, it was considered the most appropriate reference for power estimation.

3. Results

The deviation between the preoperative planned implant position and the actual implant position was compared among the three navigation systems. Prior to parametric analysis, the Shapiro–Wilk test confirmed that all outcome variables met the assumption of normality (p > 0.05 for all groups). No statistically significant differences were observed in the three‐dimensional entry point deviation or time efficiency. To provide additional insight into data distribution, box‐and‐whisker plots (median, first quartile (Q1), and third quartile (Q3)) are included in Figures 4, 5, 6. In addition to the mean and standard deviation, the median and interquartile ranges (Q1/Q3) are reported below to enhance transparency of data distribution. The results are presented in Table 1.

TABLE 1.

Implant placement accuracy for the three d‐CAIS registration methods (ND, XC, XM).

Outcome measure ND (mean ± SD; median [Q1–Q3]; range) XC (mean ± SD; median [Q1–Q3]; range) XM (mean ± SD; median [Q1–Q3]; range) p
3D deviation at entry (mm) 1.06 ± 0.23; 1.06 [0.92–1.28]; 0.76–1.28 0.71 ± 0.22; 0.82 [0.58–0.86]; 0.36–0.94 0.90 ± 0.24; 0.90 [0.82–1.10]; 0.54–1.14 0.105
3D deviation at apex (mm) 1.23 ± 0.31; 1.24 [0.94–1.40]; 0.92–1.66 0.83 ± 0.18; 0.90 [0.72–0.96]; 0.56–1.00 1.07 ± 0.29; 1.07 [0.82–1.28]; 0.70–1.30 0.094
Vertical depth deviation (mm) 0.61 ± 0.17; 0.64 [0.50–0.74]; 0.38–0.80 0.59 ± 0.20; 0.52 [0.46–0.70]; 0.38–0.88 0.64 ± 0.21; 0.62 [0.62–0.82]; 0.30–0.82 0.928
Angular deviation (°) 2.96 ± 1.81; 3.00 [0.84–3.44]; 0.84–5.64 1.17 ± 0.48; 1.18 [0.84–1.56]; 0.56–1.72 1.02 ± 0.19; 1.02 [0.90–1.20]; 0.80–1.24 0.028 a
Open in a new tab

Note: Values are presented as mean ± SD, median (Q1–Q3), and range. p‐values from one‐way ANOVA; angular deviation post hoc tested with Tukey HSD.

Abbreviations: ND, Navident; XC, X‐Clip; XM, X‐Mark.

a

Post hoc Tukey HSD: ND>XC (p = 0.047); ND>XM (p = 0.030); XC = XM (p = 0.885).

3.1. Three‐Dimensional Deviation at the Implant Entry Point

The mean deviation at the entry point was 1.06 ± 0.23 mm (range: 0.76–1.28 mm) in the ND group, 0.71 ± 0.22 mm (range: 0.36–0.94 mm) in the XC group, and 0.90 ± 0.24 mm (range: 0.54–1.14 mm) in the XM group. The median (Q1/Q3) values were 1.06 mm (0.92/1.28) for ND, 0.82 mm (0.58/0.86) for XC, and 0.90 mm (0.82/1.10) for XM (Figure 4). ANOVA revealed no statistically significant difference in entry point deviation among the three groups (p = 0.105).

3.2. Three‐Dimensional Deviation at the Implant Apex

The mean deviation was 1.23 ± 0.31 mm (range: 0.92–1.66 mm) in the ND group, 0.83 ± 0.18 mm (range: 0.56–1.00 mm) in the XC group, and 1.07 ± 0.29 mm (range: 0.70–1.30 mm) in the XM group. The median (Q1/Q3) values were 1.24 mm (0.94/1.40) for ND, 0.90 mm (0.72/0.96) for XC, and 1.07 mm (0.82/1.28) for XM (Figure 4). ANOVA revealed no statistically significant difference in apex deviation among the three groups (p = 0.094).

3.3. Vertical Depth Deviation at the Implant Apex

The average vertical depth deviation was 0.61 ± 0.17 mm (range: 0.38 to 0.80 mm) in the ND group, 0.59 ± 0.20 mm (range: 0.38 to 0.88 mm) in the XC group, and 0.64 ± 0.21 mm (range: 0.30 to 0.82 mm) in the XM group. The median (Q1/Q3) values were 0.64 mm (0.50/0.74) for ND, 0.52 mm (0.46/0.70) for XC, and 0.62 mm (0.62/0.82) for XM (Figure 4). ANOVA revealed no statistically significant differences in vertical deviation among the groups (p = 0.928).

3.4. Angular Deviation

The average angular deviation was 2.96° ± 1.81° (range: 0.84°–5.64°) in the ND group, 1.17° ± 0.48° (range: 0.56°–1.72°) in the XC group, and 1.02° ± 0.19° (range: 0.80°–1.24°) in the XM group. The median (Q1/Q3) values were 3.00° (0.84/3.44) for ND, 1.18° (0.84/1.56) for XC, and 1.02° (0.90/1.20) for XM (Figure 5). ANOVA revealed a statistically significant difference among groups (p = 0.028). Post hoc analysis using Tukey's HSD test showed that the ND group had significantly higher angular deviation compared to both the XC (p = 0.047) and XM (p = 0.030) groups, while no significant difference was found between the XC and XM groups (p = 0.885).

3.5. Time

The average time required for implant placement was 13.27 ± 5.73 min (range: 10.10–23.47) in the ND group, 9.39 ± 2.36 min (range: 5.70–12.30) in the XC group, and 7.20 ± 2.48 min (range: 4.20–10.45) in the XM group. The median (Q1/Q3) values were 10.51 min (10.25/11.30) for ND, 9.41 min (9.30/9.45) for XC, and 8.31 min (7.28/10.27) for XM (Figure 6). Although the Kruskal–Wallis test revealed a statistically significant difference among the three groups (p = 0.038), pairwise comparisons with Bonferroni correction did not identify significant differences between any two groups.

4. Discussion

CAIS is an essential tool for achieving safe and highly accurate implant placement. The accuracy of implant placement achieved in this study showed three‐dimensional deviations at the entry point, implant apex, and vertical depth ranging from 0.71 to 1.06 mm, 0.83 to 1.23 mm, and 0.59 to 0.64 mm, respectively, all of which were significantly below the clinically acceptable safety threshold of 2 mm. Furthermore, angular deviations were within a range of 1.02°–2.96°, suggesting that clinically acceptable precision has been achieved.

In an in ex vivo study by Emery et al. (2016), deviations at the entry point, implant apex, and angle were reported as 0.46, 0.48 mm, and 1.09°, respectively, which were slightly smaller than those observed in this study. In their study, Emery et al. employed a single surgeon experienced in dynamic navigation, placing implants in models under clinical simulation conditions using a dynamic navigation system. Although efforts were made to minimize bias by involving other individuals in data analysis, the results may have been influenced by the experience and skill of the sole participating surgeon. The discrepancy in results between their study and ours may also be attributed to the optimized conditions of the ex vivo environment, which can potentially enhance accuracy. Conversely, Mediavilla Guzman et al. (2019) reported deviation values of 0.85, 1.18 mm, and 4.00°, aligning closely with the results obtained in this study. Furthermore, a prospective in vivo study by Younis et al. (2024) demonstrated d‐CAIS related deviations at the entry point, implant apex, and angle of 0.99, 1.14 mm, and 3.66°, respectively, reflecting a high level of accuracy within clinically acceptable limits and supporting the findings of this study. Additionally, a systematic review by Jorba‐García et al. (2021) reported mean deviations of 0.75, 1.09 mm, and 2.84° at the entry point, implant apex, and angle, respectively, further corroborating the outcomes of this study. These findings suggest that the results of this study are clinically valid, and d‐CAIS systems offer a safe and reliable method for implant placement, comparable to or even exceeding the precision of conventional s‐CAIS and freehand techniques.

The accuracy of d‐CAIS is potentially influenced by various factors, including the quality of DICOM and STL data acquired, the data processing and registration procedures, the calibration of instruments, and additional patient‐specific factors such as movement and the intraoral environment. Among these factors, registration plays a critical role in aligning the implant placement plan in the virtual space with the three‐dimensional coordinates of the actual surgical field, thus ensuring accuracy in d‐CAIS. Given its significance, recent studies have focused on examining the factors within registration that may affect the precision of d‐CAIS systems. For instance, Ma et al. (2022) investigated the impact of two different patient registration methods—using radiographic markers versus intraoral anatomical landmarks—on accuracy. Their findings indicated no significant difference in accuracy between the two methods, and the range of deviation was comparable to the results in this study. Similarly, Wu et al. (2024) examined the influence of registration regions on accuracy, concluding that the choice of registration region did not significantly impact the precision of implant placement, with results similar to those found in this study. In this study, two types of d‐CAIS systems were used, with the patient registration methods further divided for analysis. No significant differences in accuracy were observed between groups, except for angular deviations, and the range of accuracy was consistent with previous studies. These findings suggest that registration may have a limited impact on the accuracy of implant placement. However, the registration process itself is complex and requires meticulous attention to detail to ensure optimal performance.

In this study, the ND group exhibited significantly larger angular deviations compared to the other groups. A possible reason for this discrepancy could be the difference in software versions between the systems. The software utilized by X‐Guide is relatively newer compared to the Navident software used in this study, and the enhanced tracking performance and quality of visual feedback in X‐Guide may have contributed to improved accuracy. Another contributing factor might be the stability of the optical marker/tracker–patient. In X‐Guide, the optical marker/tracker–patient is securely attached to the dental arch with a robust arm, whereas Navident's optical marker/tracker–patient is made with a wire‐based structure that, while easier to adjust, may lack stability. This lack of stability in Navident's optical marker/tracker–patient could lead to marker movement during surgery, potentially affecting both visual feedback from the guide and overall accuracy. Additionally, the time required for implant placement tended to be longer with Navident, likely for similar reasons. These technical factors may have impacted the performance of the ND group, suggesting that future studies should investigate the effects of the latest software versions to address these differences.

Furthermore, it is essential to interpret the results of this study in light of its limitations. This study was conducted in an in vitro environment, which differs from an actual clinical setting. In vitro model experiments provide accurate data by eliminating external factors encountered in clinical settings, allowing a focused evaluation of device and operator performance. Additionally, experiments conducted under controlled conditions ensure high reliability and reproducibility. On the other hand, under such conditions lack variables commonly encountered in clinical scenarios, such as the effects of soft tissues, patient movement, bleeding, and the moist intraoral environment, all of which can impact implant placement accuracy (Jorba‐García et al. 2021; Wei et al. 2021). These factors may influence implant accuracy in clinical environments; therefore, future studies should consider these variables through clinical trials. Additionally, this study included five operators with extensive experience in implant treatments and a total of 75 implants were incorporated into the experiment based on an appropriate study design to ensure data validity and reliability. As a result, it is considered that highly reliable data comparable to other studies of similar scale were obtained. However, the use of d‐CAIS systems require a certain learning curve (Spille et al. 2022; Wang, Liu, et al. 2022). Although the operators in this study were proficient in implant treatments, it cannot be ruled out that less experienced operators or those using d‐CAIS for the first time may exhibit different levels of accuracy. Therefore, further research involving a wider range of operators is essential to verify reproducibility in clinical settings while accounting for the influence of the learning curve. Such studies will contribute to promoting the clinical adoption of d‐CAIS systems.

In this study, we evaluated the accuracy of implant placement using a d‐CAIS system and confirmed that deviations at the entry point, implant apex, and angle remained within clinically acceptable ranges. Comparisons with other studies suggest that our findings demonstrate clinical effectiveness. Additionally, we considered differences in patient registration methods, concluding that their impact on accuracy is likely minimal; however, it is also suggested that software version and instrument stability may influence specific accuracy parameters. Future research should focus on comprehensively assessing the accuracy and effectiveness of navigation systems through clinical trials involving a larger number of practitioners and patients. For dynamic guides to achieve widespread clinical adoption, it is necessary to integrate both ex vivo and in vivo data and continue evaluation in real surgical environments. Such cumulative research is expected to lead to further improvements in d‐CAIS systems and greater precision in implant treatments.

One limitation of the present study was that the power analysis was conducted post hoc rather than a priori. While angular deviation yielded a statistically significant difference among groups and was confirmed to have sufficient power (0.869), the other parameters—coronal, apical, and depth deviations—did not reach statistical significance. Nevertheless, post hoc analyses conducted for all parameters suggested generally sufficient power across outcomes, supporting the overall adequacy of the sample size. Future studies should include parameter‐specific a priori sample size calculations based on pilot data or estimated effect sizes to further enhance methodological rigor and allow for more robust interpretation of nonsignificant findings.

5. Conclusion

High accuracy was achieved within clinically acceptable limits with all three tested d‐CAIS systems. All observed deviations fell within 2 mm, supporting the potential of d‐CAIS systems to provide a safe and reliable implantation technique.

Author Contributions

Toshiki Nojiri: conceptualization, methodology, data curation, investigation, writing – original draft, writing – review and editing. Kazuhiro Kon: investigation, visualization, writing – review and editing. Akihiro Fukutoku: investigation, writing – review and editing, formal analysis. Kevser Pala: writing – original draft, writing – review and editing. Ryo Yamamoto: investigation, visualization, writing – review and editing. Shigemi Nagai: supervision, writing – review and editing. German O. Gallucci: supervision, writing – review and editing.

Ethics Statement

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Data S1: CRIS guidelines (checklist for reporting in vitro studies).

CLR-37-961-s001.docx (17.1KB, docx)

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

References

  1. Albrektsson, T. , Zarb G., Worthington P., and Eriksson A. R.. 1986. “The Long‐Term Efficacy of Currently Used Dental Implants: A Review and Proposed Criteria of Success.” International Journal of Oral & Maxillofacial Implants 1, no. 1: 11–25. [PubMed] [Google Scholar]
  2. Block, M. S. D. M. D. , and Emery R. W. D. D. S.. 2016. “Static or Dynamic Navigation for Implant Placement—Choosing the Method of Guidance.” Journal of Oral and Maxillofacial Surgery 74, no. 2: 269–277. 10.1016/j.joms.2015.09.022. [DOI] [PubMed] [Google Scholar]
  3. Buser, D. , Martin W., and Belser U. C.. 2004. “Optimizing Esthetics for Implant Restorations in the Anterior Maxilla: Anatomic and Surgical Considerations.” International Journal of Oral & Maxillofacial Implants 19: 43–61. [PubMed] [Google Scholar]
  4. Emery, R. W. , Merritt S. A., Lank K., and Gibbs J. D.. 2016. “Accuracy of Dynamic Navigation for Dental Implant Placement–Model‐Based Evaluation.” Journal of Oral Implantology 42, no. 5: 399–405. 10.1563/aaid-joi-d-16-00025. [DOI] [PubMed] [Google Scholar]
  5. Fortin, T. , Bosson J. L., Isidori M., and Blanchet E.. 2006. “Effect of Flapless Surgery on Pain Experienced in Implant Placement Using an Image‐Guided System.” International Journal of Oral & Maxillofacial Implants 21, no. 2: 298–304. [PubMed] [Google Scholar]
  6. Fugazzotto, P. D. D. S. , Melnick P. R. D. M. D., and Al‐Sabbagh M. D. D. S. M. S.. 2015. “Complications When Augmenting the Posterior Maxilla.” Dental Clinics of North America 59, no. 1: 97–130. 10.1016/j.cden.2014.09.005. [DOI] [PubMed] [Google Scholar]
  7. Haemmerle, C. H. F. , Stone P., Jung R. E., Kapos T., and Brodala N.. 2009. “Consensus Statements and Recommended Clinical Procedures Regarding Computer‐Assisted Implant Dentistry.” International Journal of Oral & Maxillofacial Implants 24: 126–129. [PubMed] [Google Scholar]
  8. Jorba‐Garcia, A. , Pozzi A., Chen Z., et al. 2025. “Glossary of Computer‐Assisted Implant Surgery and Related Terms. First Edition.” Clinical and Experimental Dental Research 11, no. 4: e70148. 10.1002/cre2.70148. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Jorba‐García, A. , González‐Barnadas A., Camps‐Font O., Figueiredo R., and Valmaseda‐Castellón E.. 2021. “Accuracy Assessment of Dynamic Computer–Aided Implant Placement: A Systematic Review and Meta‐Analysis.” Clinical Oral Investigations 25, no. 5: 2479–2494. 10.1007/s00784-021-03833-8. [DOI] [PubMed] [Google Scholar]
  10. Juodzbalys, G. , Wang H.‐L., and Sabalys G.. 2010. “Anatomy of Mandibular Vital Structures. Part II: Mandibular Incisive Canal, Mental Foramen and Associated Neurovascular Bundles in Relation With Dental Implantology.” Journal of Oral & Maxillofacial Research 1, no. 1: e3. 10.5037/jomr.2010.1103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Juodzbalys, G. , Wang H.‐L., Sabalys G., Sidlauskas A., and Galindo‐Moreno P.. 2013. “Inferior Alveolar Nerve Injury Associated With Implant Surgery.” Clinical Oral Implants Research 24, no. 2: 183–190. 10.1111/j.1600-0501.2011.02314.x. [DOI] [PubMed] [Google Scholar]
  12. Kalpidis, C. D. R. , and Setayesh R. M.. 2004. “Hemorrhaging Associated With Endosseous Implant Placement in the Anterior Mandible: A Review of the Literature.” Journal of Periodontology 75, no. 5: 631–645. 10.1902/jop.2004.75.5.631. [DOI] [PubMed] [Google Scholar]
  13. Kang, S. , Hou Y., Cao J., Li S., Xue P., and Jiang Y.. 2024. “Comparison of Implantation Accuracy Among Different Navigated Approaches: A Systematic Review and Network Meta‐Analysis.” International Journal of Oral and Maxillofacial Implants 39, no. 3: 455–467. 10.11607/jomi.10381. [DOI] [PubMed] [Google Scholar]
  14. Lin, M.‐H. , Mau L.‐P., Cochran D. L., Shieh Y.‐S., Huang P.‐H., and Huang R.‐Y.. 2014. “Risk Assessment of Inferior Alveolar Nerve Injury for Immediate Implant Placement in the Posterior Mandible: A Virtual Implant Placement Study.” Journal of Dentistry 42, no. 3: 263–270. 10.1016/j.jdent.2013.12.014. [DOI] [PubMed] [Google Scholar]
  15. Ma, F. , Sun F., Wei T., and Ma Y.. 2022. “Comparison of the Accuracy of Two Different Dynamic Navigation System Registration Methods for Dental Implant Placement: A Retrospective Study.” Clinical Implant Dentistry and Related Research 24, no. 3: 352–360. 10.1111/cid.13090. [DOI] [PubMed] [Google Scholar]
  16. Mandelaris, G. A. , Stefanelli L. V., and DeGroot B. S.. 2018. “Dynamic Navigation for Surgical Implant Placement: Overview of Technology, Key Concepts, and a Case ReportDynamic Navigation for Surgical Implant Placement: Overview of Technology, Key Concepts, and a Case Report.” Compendium of Continuing Education in Dentistry 39, no. 9: 614–621. [PubMed] [Google Scholar]
  17. Mediavilla Guzman, A. , Riad Deglow E., Zubizarreta‐Macho A., Agustin‐Panadero R., and Hernandez Montero S.. 2019. “Accuracy of Computer‐Aided Dynamic Navigation Compared to Computer‐Aided Static Navigation for Dental Implant Placement: An In Vitro Study.” Journal of Clinical Medicine 8, no. 12: 2123. 10.3390/jcm8122123. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Misch, C. E. , Bidez M. W., and Sharawy M.. 2001. “A Bioengineered Implant for a Predetermined Bone Cellular Response to Loading Forces. A Literature Review and Case Report.” Journal of Periodontology 72, no. 9: 1276–1286. 10.1902/jop.2000.72.9.1276. [DOI] [PubMed] [Google Scholar]
  19. Panchal, N. , Mahmood L., Retana A., and Emery R.. 2019. “Dynamic Navigation for Dental Implant Surgery.” Oral and Maxillofacial Surgery Clinics of North America 31, no. 4: 539–547. 10.1016/j.coms.2019.08.001. [DOI] [PubMed] [Google Scholar]
  20. Pedrinaci, I. , Gallucci G. O., Lanis A., Friedland B., Pala K., and Hamilton A.. 2024. “Computer‐Assisted Implant Planning: A Review of Data Registration Techniques.” International Journal of Periodontics & Restorative Dentistry 45: 652–665. 10.11607/prd.7127. [DOI] [PubMed] [Google Scholar]
  21. Pei, X. , Weng J., Sun F., Ma Y., Iao S., and Liu X.. 2024. “Accuracy and Efficiency of a Calibration Approach in Dynamic Navigation for Implant Placement: An In Vitro Study.” Journal of Dental Sciences 19, no. 1: 51–57. 10.1016/j.jds.2023.06.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Pellegrino, G. , Ferri A., Del M., Prati C., Gandolfi M. G., and Marchetti C.. 2021. “Dynamic Navigation in Implant Dentistry: A Systematic Review and Meta‐Analysis.” International Journal of Oral and Maxillofacial Implants 36, no. 5: E121–E140. 10.11607/jomi.8770. [DOI] [PubMed] [Google Scholar]
  23. Pjetursson, B. E. , Brägger U., Lang N. P., and Zwahlen M.. 2007. “Comparison of Survival and Complication Rates of Tooth‐Supported Fixed Dental Prostheses (FDPs) and Implant‐Supported FDPs and Single Crowns (SCs).” Clinical Oral Implants Research 18, no. S3: 97–113. 10.1111/j.1600-0501.2007.01439.x. [DOI] [PubMed] [Google Scholar]
  24. Pjetursson, B. E. , Tan K., Lang N. P., Bragger U., Egger M., and Zwahlen M.. 2004. “A Systematic Review of the Survival and Complication Rates of Fixed Partial Dentures (FPDs) After an Observation Period of at Least 5 Years—I. Implant‐Supported FPDs.” Clinical Oral Implants Research 15, no. 6: 625–642. 10.1111/j.1600-0501.2004.01117.x. [DOI] [PubMed] [Google Scholar]
  25. Rocci, A. , Martignoni M., and Gottlow J.. 2003. “Immediate Loading in the Maxilla Using Flapless Surgery, Implants Placed in Predetermined Positions, and Prefabricated Provisional Restorations: A Retrospective 3‐Year Clinical Study.” Clinical Implant Dentistry and Related Research 5, no. S1: 29–36. 10.1111/j.1708-8208.2003.tb00013.x. [DOI] [PubMed] [Google Scholar]
  26. Schneider, D. , Marquardt P., Zwahlen M., and Jung R. E.. 2009. “A Systematic Review on the Accuracy and the Clinical Outcome of Computer‐Guided Template‐Based Implant Dentistry.” Clinical Oral Implants Research 20, no. S4: 73–86. 10.1111/j.1600-0501.2009.01788.x. [DOI] [PubMed] [Google Scholar]
  27. Spille, J. , Helmstetter E., Kuebel P., et al. 2022. “Learning Curve and Comparison of Dynamic Implant Placement Accuracy Using a Navigation System in Young Professionals.” Dentistry Journal 10, no. 10: 187. 10.3390/dj10100187. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Tahmaseb, A. , Wismeijer D., Coucke W., and Derksen W.. 2014. “Computer Technology Applications in Surgical Implant Dentistry: A Systematic Review.” International Journal of Oral & Maxillofacial Implants 29, no. Supplement: 25–42. 10.11607/jomi.2014suppl.g1.2. [DOI] [PubMed] [Google Scholar]
  29. Wang, F. , Wang Q., and Zhang J.. 2021. “Role of Dynamic Navigation Systems in Enhancing the Accuracy of Implant Placement: A Systematic Review and Meta‐Analysis of Clinical Studies.” Journal of Oral and Maxillofacial Surgery 79, no. 10: 2061–2070. 10.1016/j.joms.2021.06.005. [DOI] [PubMed] [Google Scholar]
  30. Wang, X. , Shaheen E., Shujaat S., et al. 2022. “Influence of Experience on Dental Implant Placement: An In Vitro Comparison of Freehand, Static Guided and Dynamic Navigation Approaches.” International Journal of Implant Dentistry 8, no. 1: 42. 10.1186/s40729-022-00441-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Wang, X.‐Y. , Liu L., Guan M.‐S., Liu Q., Zhao T., and Li H.‐B.. 2022. “The Accuracy and Learning Curve of Active and Passive Dynamic Navigation‐Guided Dental Implant Surgery: An In Vitro Study.” Journal of Dentistry 124: 104240. 10.1016/j.jdent.2022.104240. [DOI] [PubMed] [Google Scholar]
  32. Wei, S. M. , Zhu Y., Wei J. X., Zhang C. N., Shi J. Y., and Lai H. C.. 2021. “Accuracy of Dynamic Navigation in Implant Surgery: A Systematic Review and Meta‐Analysis.” Clinical Oral Implants Research 32, no. 4: 383–393. 10.1111/clr.13719. [DOI] [PubMed] [Google Scholar]
  33. Widmann, G. , and Bale R. J.. 2006. “Accuracy in Computer‐Aided Implant Surgery—A Review.” International Journal of Oral & Maxillofacial Implants 21, no. 2: 305–313. [PubMed] [Google Scholar]
  34. Wu, B. Z. , Ma F. F., Yan X. Y., and Sun F.. 2024. “Accuracy of Different Registration Areas Using Active and Passive Dynamic Navigation Systems in Dental Implant Surgery: An In Vitro Study.” Clinical Oral Implants Research 35, no. 8: 888–897. 10.1111/clr.14192. [DOI] [PubMed] [Google Scholar]
  35. Younis, H. , Lv C., Xu B., et al. 2024. “Accuracy of Dynamic Navigation Compared to Static Surgical Guides and the Freehand Approach in Implant Placement: A Prospective Clinical Study.” Head & Face Medicine 20, no. 1: 30. 10.1186/s13005-024-00433-1. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Data S1: CRIS guidelines (checklist for reporting in vitro studies).

CLR-37-961-s001.docx (17.1KB, docx)

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Articles from Clinical Oral Implants Research are provided here courtesy of Wiley

RESOURCES