The aim of this study was to evaluate the crown accuracy and time efficiency provided by a complete digital workflow for multiple implant-supported single crowns, using an auxiliary occlusal device, compared to a workflow using conventional impression and stone cast-based CAD/CAM crowns. The complete digital workflow on multiple implants with the use of an auxiliary occlusal device demonstrated superior crown accuracy, especially on occlusal surface. Laboratory time was significantly shortened in complete digital workflow, while clinical time showed no difference.
To fabricated successful prostheses, it is crucial to accurately reproduce the interocclusal relationship and transfer it to the articulator.[18, 19] Accurately transferring the interocclusal relationship to the CAD-CAM software program is essential for providing clinically acceptable restorations with accurate occlusal morphology , reducing the need for chairside occlusal adjustment. However, when multiple teeth are missing, the accuracy of VIR may not meet clinical standard, rendering a complete digital workflow on multiple implants invalid.[1, 20, 21] Previous research[17] explored the use of healing abutments to improve interocclusal relationship accuracy in IOS on stone models, however no clinical study of a complete digital workflow on multiple implant-supported crowns was reported. Therefore, this study represents a significant contribution to the field. The results demonstrate that complete digital workflows reduce the need for occlusal adjustment (in both depth and size) of crowns, as well as laboratory time, compared to conventional workflows. These findings suggest that model-free digital workflows may be suitable for multiple-implant supported restorations, expanding their indications.
In model-free digital workflows, there are generally two approaches to designing and fabricating restorations. The first approach, which is well-established, involves implant-supported screw-retained restorations comprising prefabricated Ti-base abutments and monolithic zirconia crowns. Several studies[5, 22, 23]have reported this full digital approach to be feasible, accurate, and efficient for single implants. However, when it comes to restoring crowns on multiple adjacent implants, particularly when implants are not perfectly parallel, clinicians face challenges in determining the correct insertion path for delivering adjacent screw-retained crowns, especially in the absence of a physical model in the full digital workflow. The second approach involves implant-supported cement-retained restorations, which utilize customized abutments and monolithic zirconia crowns. The use of customized abutments allows for adjustments to the insertion path, resulting in easier and faster delivery of cement-retained crowns. Previous studies[4] have reported and demonstrated the accuracy and efficiency of this approach for single implant-supported crowns. Therefore, customized abutments and cement-retained crowns were chosen to restore on multiple implants in digital and control groups.
The inclusion of a complete digital workflow without auxiliary occlusal devices as an ideal control group was initially considered. However, a preliminary study revealed that crowns produced through this approach sometimes required massive adjustments, while others exhibited infraocclusion and necessitated refabrication in the dental laboratory. Consequently, it was deemed inappropriate to select the complete digital workflow without auxiliary occlusal devices as the control group. Previous studies[4, 23, 24] investigating crown accuracy in digital workflows have utilized the conventional workflow as the control group. This conventional workflow involves conventional impressions, implant-level scanning, abutment CAD/CAM, abutment-level scanning, crown CAD/CAM, and crown delivery. Separate laboratory scans were performed on implant level and abutment level, as it is the most commonly employed method for fabricating cement-retained implant-supported restorations. The mean crown accuracy with the conventional workflow, using the occlusion record recommended in prior studies[25], was found to be 479.59±203.63 µm in this study. This accuracy level is comparable to the crown accuracy observed in conventional workflows for single restorations, which recorded measurements from 330.7µm to 485±194µm in previous studies[4, 5, 26].
In this study, proximal adjustment was a secondary outcome measure. It was observed that the mesial surface required more adjustment compared to the distal surface in both the test and control groups. In cases with free-end saddles, no adjustment was deemed necessary on the distal surface of the distal crown, which could be attributed to less need for distal adjustment. Additionally, when the contact between two implant-supported crowns was excessively tight, clinicians might choose to adjust either the distal surface of the mesial crown or the mesial surface of the distal crown based on their personal preference, which could have an impact on the results. However, since the primary focus of this study was the accuracy of occlusion, meticulous clinical instructions regarding proximal adjustment were not established. Overall, proximal adjustments showed no significant difference in two groups, indicating similar accuracy between digital and conventional impressions in terms of the positions of implants and neighboring teeth.
The study compared the clinical chair-side times for digital and conventional impressions and found no significant difference between the test and control groups. However, the time required for digital impressions was approximately 5 minutes longer than for conventional impressions. This result contradicts previous clinical studies [4, 5, 23, 27] which reported significantly shorter times for intraoral scanning (IOS). One possible explanation for the longer digital impression time could be the use of auxiliary devices, such as installing and uninstalling them, and adjusting their occlusion, which may have contributed to the prolonged process. Despite the longer impression time, the crown delivery process was more time-efficient in the digital workflow than in the conventional workflow, as less occlusion adjustment was needed in the digital group. Overall, the digital workflow did not demonstrate time-efficient superiority over the conventional workflow.
The benefits of the digital workflow for laboratory procedures were obvious; IOS obviates the need for plaster casting and model scanning, while the split-file technique allows designing and milling abutments and crowns in one step, circumventing the need for a second digital model. Several manual fabrication steps were skipped, so the laboratory time was dramatically reduced from 105.92 minutes in the control group to 46.08 minutes in the test group. However, it should be reminded that digital workflow requires highly experienced dental technicians because the model-free workflow eliminates the possibility of abutment and crown adjustment on the stone model. The dental technicians should consider abutment marginal depth, path of insertion, proximal contact, and occlusal morphology of restorations in one virtual design. And when the abutments and crowns are milled, only the abutment-crown marginal fit could be adjusted. The inappropriate design on abutment and crown morphology could not be realized until the clinician try them intraorally. The importance of technicians’ experience in CAD/CAM process was mentioned by previous reports[28, 29], and should be highlighted in this study for designing restorations on multiple implants is more complex and challenging.
The advantages of the digital workflow for laboratory procedures were evident in this study. Intraoral scanning (IOS) eliminates the need for plaster casting and model scanning, while the split-file technique enables designing and milling abutments and crowns in one step, thus avoiding the need for a second digital model. These steps reduce several manual fabrication processes, resulting in a significant reduction in laboratory time from 105.92 minutes in the control group to 46.08 minutes in the test group.
However, it is worth noting that the digital workflow requires highly experienced dental technicians because the model-free workflow eliminates the possibility of adjusting the abutment and crown on the stone model. Technicians must consider the abutment marginal depth, path of insertion, proximal contact, and occlusal morphology of the restorations in one virtual design. When the abutments and crowns are milled, only the abutment-crown marginal fit can be adjusted. Any inappropriate design on the abutment and crown morphology may not be recognized until the clinician tries them intraorally. The importance of technicians' experience in the CAD/CAM process has been highlighted in previous reports [28, 29] and this study underscores the complexity and challenges involved in designing restorations on multiple implants.
Several limitations of this study should be acknowledged. Firstly, the sample size was relatively small, which may affect the generalizability of the findings. Secondly, the study didn’t analyze free-end saddle and non-free-end saddle separately, which might affect the result differently, regardless of whether a digital or conventional workflow. Further study should focus on the effect of auxiliary devices in digital workflow for free-end saddle and non-free-end saddle cases individually. Thirdly, while we have shown the validity of the digital workflow for two adjacent implant-supported crowns, further studies are necessary to investigate its applicability for more than two posterior implants.