PDF
Citation
Print
Marietta Krüsi, Demetrios J. Halazonetis, Theodore Eliades, Vasiliki Koretsi
Covariance patterns between ramus morphology and the rest of the face: A geometric morphometric study.
- Korean J Orthod 2023;53:185-193
I read the paper titled “Covariance patterns between ramus morphology and the rest of the face: A geometric morphometric study” by Krüsi et al.1 with much interest, and I have some comments and concerns that might help. Primarily, no hypothesis is tested in this study and the descriptive findings regarding the ramus are limited to statistical shape-space. For example, the authors did not mention the coronoid process and the angle of the mandible in their narrative, both of which are dependent on muscle development. On the other hand, they also omitted the epigenetic potential of condylar stem cells2 and cranial base orthocephalization,3 both of which play significant roles in determining clinical mandibular morphology and spatial orientation. In effect, the ramus width might simply be a by-product of other proactive developmental processes; perhaps the developmental ontogeny of the mandible described in this article is based on, some might say, outdated concepts. For example, the functional matrix hypothesis was left unfinished,4 but later Singh5 was able to integrate genetics6 and epigenetics7 to describe mandibular behavior inter alia, using the Spatial Matrix Hypothesis, which integrates temporo-spatial patterning8 to describe developmental compensation and postural (clinical) decompensation.
Another issue with the earlier studies is both the materials and the methods that were available at that time. For example, Enlow and Hans9 used cephalographic data, and while the authors recognize the deficiencies of cephalometric analysis, their validation is based on different reasons. In other words, even though conventional two-dimensional (2D) cephalometric analysis is limited by fixed references planes, homologous landmarks etc., the main issue with this data is that it compresses a three-dimensional (3D) object unto a 2D image, which does not accurately reflect the subject of study in real space. Consequently, the ramus widths derived in this study may not reflect clinical reality. The reader is left wondering why 3D data, which is now more readily available, was not used in this study. In addition, while the methods used in this present study appear useful, they lack graphical outputs that earlier geometric morphometric studies successfully deployed on mandibular allometry several years ago.10 This omission represents both a deficiency in literature review and is a less useful presentation for clinical orthodontists. Finally, it is well known that mandible exhibits heterogeneity, based on factors such as cranial base morphology and ethnicity11 inter alia. Thus, by not discretizing into the various mandibular classes, the results of the present study might represent a sophisticated academic exercise, which appears to have reduced utility in clinical orthodontics. In any case, I encourage the authors to pursue their endeavor so that ramus widths, etc. can be used for modern diagnostic and treatment planning procedures as well as predictive orthodontic modeling.
We thank Dr. Singh for his insightful comments and for his interest in our work. The topic of mandibular growth, and its control by genetics, epigenetics and function, has a long history and is still under extensive investigation by the orthodontic community. It is therefore understandable that we could not cover it comprehensively, but presented only selected aspects relevant to our specific research question. We thank Dr. Singh for providing additional information from his own scientific work to the readers of the journal.
Regarding the use of 3D data, we fully agree that this would have been ideal. However, such data were not available due to radiation concerns, obliging us to use conventional cephalometric records that have driven most orthodontic research. We presented our results using contemporary visualization methods of geometric morphometrics, such as scatter plots in shape-space and extremes of variability patterns. The latter looks similar to a cephalometric tracing and is familiar to orthodontists. Of course, numerous alternatives exist, some not being in vogue, others of no clear advantage. Finally, subdividing the sample into discreet and arbitrary classes was in contrast to our methodology, as we were interested in evaluating co-variation patterns over the whole and continuous spectrum of shape variability.
XML Download