I read with interest the July 2011 article about the location of the temporomandibular joint (TMJ) in normal occlusion (Vitral RWF, da Silva Campos MJ, Rodrigues AF, Fraga MR. Temporomandibular joint and normal occlusion: is there anything singular about it? A computed tomographic evaluation. Am J Orthod Dentofacial Orthop 2011;140:18-24). Although the authors should be applauded for several aspects of their study, it has some limitations to which I want to draw attention.

Traditionally, the anterior, posterior, and superior joint spaces in the midsagittal plane are important to evaluate the centralization of the condyle in the fossa. With the development of 3-dimensional (3D) evaluation techniques, we can study the spatial locations of the condyle and fossa in 3 dimensions. That is, after the images of the condyle and fossa have been established, the minimum distances from all points on the condyle to their corresponding points on the fossa are calculated automatically. In the 3D image, distances are marked by different colors, which symbolize a certain range of distances. This can provide a visual and quantifiable position of the condyle in the fossa.

The articular disc plays an important role in the pathology of some TMJ diseases. When we analyze the relationship between the occlusion and the TMJ, the articular disc must be involved. For bone evaluation, a computed tomography (CT) scan is advised, whereas, for articular disc evaluation, a magnetic resonance imaging (MRI) scan is preferred.

Recently, scholars began to develop a method of image reconstruction of the TMJ (including the condylar process, the fossa, and the articular disc) on the basis of an MRI and CT image fusion technique. That is, during CT and MRI scans, several reference markers are attached to the patient. In the image-processing procedure, the images (condylar process, fossa, and reference markers) from the CT scans and the images (articular disc and reference markers) from the MRI scans are superimposed and fused on the basis of the reference markers.

The results of the analysis of the dimensional and positional symmetries of the condylar processes are also influenced by the locations of the midline of the face and the midline of the dentition. However, these midlines are not always coincident, even in persons with normal occlusion. Therefore, if we try to determine the symmetries of 2 condylar processes from CT scans, it is important to make certain that the midlines of the face and dentition are coincident.

It is difficult to analyze certain distances or angles on 3D models. For example, when we try to determine the midsagittal plane on the 3D model (which may be the most important factor), we must determine at least 3 points to define this plane. But point selection on the 3D model itself will lead to a prominent discrepancy. Creating a high-quality 3D cephalometric analysis is not yet possible. There is still much work to do.