We thank Dr Li for the letter and agree with the closing sentence, that “there is still much work to do.” The influence of occlusion on temporomandibular joint (TMJ) morphology is controversial.

Our article, published in July 2011, is part of a research series started over 10 years ago in which we investigated the structural characteristics of the TMJ in groups of patients with normal occlusion, Class II subdivision, Class I, and Class II Division 1 and Class III malocclusions. Much has changed since our early studies, and previously unavailable techniques, such as cone-beam computed tomography, have found a place in orthodontic research. Since we aimed to perform the same analysis in all groups for further comparisons, the same methodology had to be used for all groups of malocclusions. The next step of our research series, which is being concluded, is to compare all the data obtained from all groups evaluated.

The anterior, superior, and posterior articular spaces were determined according to Dr Li’s statement: the minimum distances from all points on the condyle to their corresponding points on the fossa. Since the analysis was made with bidimensional images, the spaces were defined as the shortest distances between the condyle and the mandibular fossa.

We agree that the articular disc plays an important role in the pathology of some TMJ diseases. However, patients with TMJ disorders were not included in any of our samples. Computed tomography scanning is the method of choice for assessment of the bony structures of the TMJ. Since we did not aim to evaluate the articular disc, magnetic resonance imaging was not considered in our study.

As for the analysis of positional symmetries of the condylar process, in both the anteroposterior and mediolateral directions, we did not use any occlusal or facial references. The condyles were assessed in axial sections obtained with the midsagittal plane as the reference. In an asymmetric sample, such as Class II subdivision subjects, for example, in which there is rotation of the dental arches, it is not practical to relate the midline of the dentition with the midline of the face. We believe that our conclusions might stimulate further studies with 3-dimensional models.

• 1.

  Evaluation of the concentric positions of the condyles in their respective mandibular fossae showed noncentralized positions for both normal occlusions and malocclusions.

• 2.

  No singular characteristic in the TMJs of the normal occlusion group was verified.

TMJ reconstruction imaging with a computed tomography-magnetic resonance imaging fusion technique, as proposed by Dr Li, will undoubtedly advance our understanding by gathering information on soft and bony tissues in the same high-definition imaging examination.