Boxuan Xu, Wei-Shao Lin, Jianguo Tan, Li Chen
AR/MR-guided implant placement appears to achieve accuracy broadly comparable to existing digital navigation techniques, although the evidence base remains predominantly preclinical and highly heterogeneous. Future development should prioritize the optimization of tracking systems, alongside more robust clinical validation to establish its reliability.
OBJECTIVES: To systematically evaluate the accuracy of dental implant placement guided by augmented reality (AR) or mixed reality (MR) and to identify key clinical and technical factors influencing deviations.
DATA: This systematic review adhered to PRISMA guidelines and was registered in PROSPERO (CRD420251178583). Studies were assessed for methodological quality using the ROBINS-I tool. Data synthesis was performed via random-effects meta-analysis, with subgroup analyses.
SOURCES: Electronic (PubMed, Scopus, Web of Science, IEEE Xplore, and Google Scholar) and manual searches.
STUDY SELECTION: This review included English articles published before August 15, 2026, that assessed the performance of AR- or MR-guided systems in the placement of dental implants.
RESULTS: Twenty-one studies (15 in vitro, 2 clinical, 3 case reports, 1 cadaver) met the inclusion criteria. The single-arm meta-analysis of 14 studies showed pooled accuracy estimates for AR/MR-guided placement as follows: coronal deviation of 0.924 mm, apical deviation of 1.231 mm, and angular deviation of 2.932° (95% confidence intervals [CIs]: 0.598-1.249 mm, 0.871-1.592 mm, and 2.031-3.833°, respectively). Comparative analysis of 9 studies revealed that AR/MR guidance was significantly more accurate than freehand placement for coronal and angular deviations (p < 0.001), showed statistically significant but clinically small differences versus static guides (apical MD -0.084 mm; angular MD -0.319°; p = 0.001), and showed no statistically significant difference when compared with dynamic navigation (p > 0.05). Subgroup analyses suggested potential differences by tracking system, soft tissue, implant site, and display device, although these were limited by small subgroup sizes and should be regarded as hypothesis-generating.
CONCLUSIONS: AR/MR-guided implant placement appears to achieve accuracy broadly comparable to existing digital navigation techniques, although the evidence base remains predominantly preclinical and highly heterogeneous. Future development should prioritize the optimization of tracking systems, alongside more robust clinical validation to establish its reliability.
CLINICAL SIGNIFICANCE: AR/MR-based navigation may offer ergonomic advantages and spatial guidance comparable to existing systems, but its accuracy has been demonstrated predominantly in benchtop settings; practitioners should expect greater deviations under real clinical conditions.