This was a pilot study designed to estimate the feasibility of measuring plane-to-plane MPA on a 3D image. Several studies have reported no significant difference and high concordance between 3D and 2D cephalometry. In a comparison of the 2D and 3D CBCT methods, Nalçaci et al. reported no statistically significant difference for the measurements of 12 angles (SNA, SNB, ANB, SND, NA-Pog, AB-NPog, Ns-Ba, IMPA, FMIA, SN Ans-Pns, L1-APog, and L1- NB) but did note a significant difference in the measurements of two angles (U1-NA , U1-SN).19
Yitschaky et al. reported high compatibility between 2D and 3D CT cephalometry in linear and angular measurements, excluding angular measurements that included the sella turcica anatomic landmark.11
Oh et al. compared the angle measurements of 3D reconstructed computed tomography and 2D conventional LCR images. This study discovered that the MPA measured from 2D images was larger than the line-to-line MPA measured on 3D reconstructed computed tomography images for all patients; however, high concordance was still noted.10
Zamora also reported that no statistically significant differences were observed between the angular and linear measurements obtained through LCR and those obtained through CBCT.20 Jung et al. compared 2D and 3D CBCT midsagittal projection cephalometric measurements. Their study noted no significant difference between 3D projected midsagittal plane measurements and 2D cephalometric measurements in patients with plane asymmetry (menton deviation < 2 mm). Although measurements differed after reorientation, these differences were not clinically significant.15
However, the aforementioned studies did not consider the 3D plane-to-plane MPA for measurement and did not conduct comparisons with the facial asymmetry group.
Gateno et al. reported that facial asymmetry affected both 2D and 3D cephalometric measurements. In their study, line-to-line gonial angle (condylion-gonion-menton) measurement was distorted when it was measured using 2D cephalometry, but this distortion was not observed in the 3D measurements. Plane-to-plane occlusal planeFrankfort horizontal angle measurements were distorted in planes with asymmetry in both 2D and 3D cephalometry, although the magnitude of the distortion was larger in 2D cephalometry. In the study conducted by Gateno et al., the 3D occlusal plane-Frankfort horizontal plane-toplane angle was more distorted in roll rotation than in yaw rotation (0.34° distortion in 10° yaw rotation).14
In this study, no significant difference was observed between the measurement values of 2D FMA, 3D FMA and 3D MP, and the difference in mean values between the groups was <0.8°. In extreme case that demonstrated in Figure 6
, the difference between 3D FMA and 3D MP was 0.32 mm. According to study of Kamoen et al., the clinical significance error of FMA is 0.8o. 21
The amount of error has no clinical significance at all. Therefore, the differences in our results fall within the clinically acceptable range of measurement error in the range of menton deviation up to 12 mm and Go discrepancy up to 8 mm. Thus, the 3D MP measurement method could be used to analyze patients with symmetric and asymmetric planes.
Our study has several limitations. First, the sample in this pilot study was relatively small and included only 10 patients in each group. Second, patients with facial deformities, such as cleft lip or palate, and a history of facial surgery or trauma were excluded to reduce the identification errors of 2D-LCR tracing. Finally, in the asymmetry group, only menton deviation and bilateral gonion discrepancy were used for classification. More cases could be included in future study to confirm the factors that may affect the methods of 3D MP measurement in asymmetry patients.
An example of case with extreme asymmetry (menton deviation: 10.8 mm, bilateral Go discrepancy: 7.1 mm). View Hi-Res Image