[Measurement errors in 2D cephalometrics]

The objective of this study is to analyze the influence that two parameters, the source-subject distance and the type of receptor, have on the precision of localization of the 19 most frequently employed cephalometric landmark points and then to analyze the repercussions of localization errors on cephalometric measurements. The correct use of profile cephalograms (norma lateralis) for clinical purposes depends essentially on the precision of the identification and localization of the different landmark points that are required for the elaboration of the various cephalometric analyses. For this study we asked 53 orthodontists, selected at random, to identify 19 cephalometric points on standard profile or digitized films taken at distances of 1 m 50 or of 4 m. The results we obtained show no statistically significant differences in relation to the source-subject distance or to the type of receptor. Each cephalometric landmark presented a characteristic dispersion cloud; the most reliable points were primarily located on the median sagittal plane. The cephalometric measurements that demonstrated the greatest variability were associated with positions of the incisor apices, which makes their value doubtful in the establishment of treatment plans and in ongoing evaluations of treatment progress. For similar reasons, the worth of the Frankfort plane can be questioned when it is compared to other, much more reliable cranial reference planes especially S-Na.