Effect of arch width on the accuracy of complete-arch digital implant impressions acquired with intraoral scanning, intraoral photogrammetry, and extraoral photogrammetry

This in vitro study aimed to evaluate the effect of maxillary arch width on the trueness and precision of complete-arch digital implant impressions obtained using IOS protocols with or without prefabricated auxiliary devices and to compare the results with extraoral and intraoral photogrammetry systems. Three edentulous maxillary master models representing wide, medium, and narrow arch widths were fabricated according to predefined intercanine and intermolar dimensions. Four multi-unit implant analogs were placed in each model to simulate an All-on-4 configuration. Six digital acquisition protocols were evaluated: CEREC Primescan without auxiliary devices (IOS-P), CEREC Primescan with prefabricated auxiliary devices (IOS-PPAD), Straumann SIRIOS without auxiliary devices (IOS-S), Straumann SIRIOS with prefabricated auxiliary devices (IOS-SPAD), extraoral photogrammetry (EPG), and intraoral photogrammetry (IPG). Each model was scanned with each protocol ( n = 17). Reference datasets were obtained using an industrial optical scanner. RMS values and angular deviations (ADs) were calculated to assess trueness, while pairwise intra-group RMS values calculated from repeated scans were used to assess precision. Arch width significantly affected RMS trueness and AD values ( P < 0.001), but not RMS precision ( P = 0.098). Impression protocol and the interaction between arch width and protocol significantly affected all outcome variables ( P < 0.001). The widest arch model showed the highest RMS and AD values. IPG showed the lowest overall RMS trueness value, while EPG and IPG formed the lowest statistical group for pairwise RMS precision values. Prefabricated auxiliary devices reduced RMS and AD values for both IOS systems compared with their corresponding non-assisted protocols. All RMS trueness values remained below the predefined 100-µm threshold; however, non-assisted IOS protocols exceeded the 0.4° AD threshold in the wide arch model. Arch width affected the trueness of complete-arch digital implant impressions, while precision was mainly influenced by the acquisition protocol. Prefabricated auxiliary devices improved IOS performance but did not completely eliminate the effect of wider arch configurations. Arch width and interimplant span should be considered when selecting a digital acquisition workflow. Photogrammetry-based protocols and PAD-assisted IOS workflows may provide more reliable options for wider complete-arch implant situations.