Precision and accuracy of suggested maxillary and mandibular landmarks with cone-beam computed tomography for regional superimpositions: An in vitro study

•We determined accuracy and precision of maxillary and mandibular landmarks in 3 dimensions.•These landmarks are to be used in regional superimpositions in 3 dimensions.•Several landmarks had high accuracy and precision; others had large mean errors.

IntroductionOur objective was to identify and evaluate the accuracy and precision (intrarater and interrater reliabilities) of various anatomic landmarks for use in 3-dimensional maxillary and mandibular regional superimpositions.MethodsWe used cone-beam computed tomography reconstructions of 10 human dried skulls to locate 10 landmarks in the maxilla and the mandible. Precision and accuracy were assessed with intrarater and interrater readings. Three examiners located these landmarks in the cone-beam computed tomography images 3 times with readings scheduled at 1-week intervals. Three-dimensional coordinates were determined (x, y, and z coordinates), and the intraclass correlation coefficient was computed to determine intrarater and interrater reliabilities, as well as the mean error difference and confidence intervals for each measurement.ResultsBilateral mental foramina, bilateral infraorbital foramina, anterior nasal spine, incisive canal, and nasion showed the highest precision and accuracy in both intrarater and interrater reliabilities. Subspinale and bilateral lingulae had the lowest precision and accuracy in both intrarater and interrater reliabilities.ConclusionsWhen choosing the most accurate and precise landmarks for 3-dimensional cephalometric analysis or plane-derived maxillary and mandibular superimpositions, bilateral mental and infraorbital foramina, landmarks in the anterior region of the maxilla, and nasion appeared to be the best options of the analyzed landmarks. Caution is needed when using subspinale and bilateral lingulae because of their higher mean errors in location.