The cumulative effect of error in the digital workflow for complete‐arch implant‐supported frameworks: An in vitro study

• Published in: Clinical oral implants research 2022-09, Vol.33 (9), p.886-899
• Main Authors: Pan, Yu, Tsoi, James Kit Hon, Lam, Walter Y. H., Zhao, Ke, Pow, Edmond H. N.
• Format: Article
• Language: English
• Subjects: CAD‐CAM; Coordinate measuring machines; Data acquisition; Dental implants; Dental prosthetics; dimensional measurement accuracy; edentulous; Error analysis; Fabrication; Gypsum; imaging; Inspection; intraoral scanning; jaw; Maxilla; Quadrants; Rehabilitation; Resins; Scanners; three‐dimensional; Transplants & implants; Variance analysis; Workflow
• ISSN: 0905-7161; 1600-0501
• DOI: 10.1111/clr.13968

Purpose To investigate the error accumulation and distribution through various stages of the digital workflow for complete‐arch implant‐supported framework fabrication. Materials and methods A resin model of edentulous maxilla with 6 dental implants was scanned using an intraoral scanner for 10 times (Complete‐digital group). Ten conventional gypsum casts were made and digitized by a laboratory scanner (Analogue‐digital group). Five implant frameworks were designed and milled using CAD‐CAM technique for each workflow. Inter‐implant distances and angles of the resin model (reference) and frameworks were measured by a coordinate measuring machine, while the scans and virtual frameworks were examined by an inspection software. Effect of type of workflow and the individual stage on the accuracy of the frameworks were analysed by Two‐way ANOVA. Results The expanded uncertainty of both workflows was ~150 μm and ~ 0.8°. The accuracy of the CAD stage was the highest. In the complete‐digital workflow, the greatest distortion was found in the data acquisition stage, while in the analogue‐digital workflow, it was found in the CAM stage. Compared with the analogue‐digital group, the complete‐digital group showed a significant higher precision in the first quadrant, but lower trueness in the second quadrant in data acquisition, and a significantly lower precision in the second quadrant at the CAD stage. Conclusions Linear distortions of the complete‐digital and analogue‐digital workflows were clinically acceptable, while angular distortions were not. Distortions were generally derived from data acquisition and CAM stage. The CAD precision depended on the distortions derived from data acquisition. The complete‐digital workflow was not as accurate as the analogue‐digital in complete‐arch implant rehabilitation.