Statement of problemThe alignment of 3-dimensional (3D) files involves selecting a reference area before performing a local best fit alignment during the digital scan superimposition and is essential for comparing digital scans. Scan alignment relies on both reference area location and the alignment algorithm. However, a consensus on the impact of different reference areas on intraoral scanning accuracy is lacking. PurposeThe purpose of this in vitro study was to assess the impact of 3 superimposition reference areas on the accuracy of 3 intraoral scanners for a partially dentate maxilla. Material and methodsA Kennedy class II resin cast was scanned using 3 intraoral scanners (Primescan, TRIOS 3, and Emerald) outputting 30 digital scans (10 per scanner). Test scans from intraoral scanners were subsequently compared with a reference digital standard tessellation language file generated by a laboratory scanner with validated accuracy. The files were superimposed using best fit alignment for each intraoral scanner using 3 different superimposition reference areas (whole region of interest, palate, and all teeth). Accuracy was assessed by using a 3D analysis program (Geomagic Control X; 3D systems) for each scanner at 4 preselected areas. Test and reference scan differences were depicted on color maps and quantified via root mean square deviations. Differences were analyzed using regression analysis with the post hoc student t test and Bonferroni correction (α=.05). ResultsThe TRIOS 3 and Emerald produced positive deviations in the palatal color maps, whereas Primescan produced more uniform color maps, regardless of the superimposition strategy used. Primescan exhibited the best accuracy (trueness and precision) in both palatal and bounded edentulous areas, regardless of the superimposition reference area. The TRIOS 3 recorded the highest distal extension trueness (ranging from 42.9±7.7 µm to 65 ±19.5 µm), and Primescan achieved the highest precision (ranging from 28.5 ±9.8 µm to 48.9 ±16.9 µm), regardless of the superimposition area. Emerald demonstrated the highest teeth trueness (ranging from 31.6 ±6.8 µm to 69.6 ±11.5 µm), while Primescan produced the highest precision (ranging from 17.9 ±6.1 µm to 30.7 ±9.2 µm), regardless of the reference area used. ConclusionsThe chosen reference area for best fit alignment significantly influenced digital scan accuracy (P<.001). Primescan displayed the highest palatal and bounded edentulous area accuracy, with TRIOS 3 recording the highest distal extension trueness. Emerald recorded the highest teeth trueness and Primescan recorded the highest distal extension and tooth precision. All conclusions were independent of the superimposition strategy used.