The aim of the study is to perform a three-dimensional (3D) evaluation of the dentoskeletal changes in pubertal Class II malocclusion subjects treated with the Herbst appliance (HAG), in comparison with a Class II comparison group (CG). 3D virtual models generated from three time-point CBCTs (T0, baseline; T1, immediately after Herbst insertion; and T2, 8 months after) of 25 HAG patients will be evaluated. Virtual models obtained from 25 Class II malocclusion patients, in the same stage of biologic maturation and skeletal abnormality, but with no orthopedic treatment will be constructed for the CG subjects. These CG patients are under orthodontic treatment, but only with teeth alignment. Voxel based registration on the anterior cranial fossa will be used to assess maxillary and mandibular displacement/articular fossa remodeling; regional registration on the mandibular corpus will be performed to evaluate mandibular growth and mandibular dental changes; and regional registration on the anterior region of the maxilla will be performed to evaluate maxillary growth and maxillary dental changes.

1. BACKGROUND Class II malocclusion is highly prevalent worldwide, and it is a major reason for orthodontic treatment to be initiated. This type of malocclusion can be restricted to the dentoalveolar region, but the majority of Class II patients exhibit a strong skeletal component to the malocclusion, including most frequently mandibular skeletal retrusion and an increased lower anterior facial height. With increased severity of the mandibular deficiency, the orthodontist is challenged to treat the underlying skeletal problem appropriately.

   For growing patients, the use of fixed and removable mandibular advancement appliances has been advocated for many decades. A variety of fixed Herbst appliance designs have achieved worldwide acceptance in that they eliminate most major patient compliance factors. Today the Herbst appliance (HA) is by far the most frequently used mandibular jumping device in the United States, with more Herbst appliance being fabricated than all other functions appliances combined.

   The Herbst appliance originally was introduced by Emil Herbst in the early years of the twentieth century, but it did not achieved worldwide clinical acceptance until its re-introduction by Hans Pancherz 70 years later. Since then, a significant number of clinical and scientific studies has been conducted about HA, particularly by Pancherz and colleagues.

   Two-dimensional cephalometric studies have reported increases in the length of the mandible and in forward displacement of the mandible following HA in comparison to matched untreated controls. On average, 2 mm of mandibular length gain (measured from Gonion to Pogonion) and 1.5 degree of improvement in the SNB angle can be observed following Herbst appliance therapy.

   Investigations in experimental animals have provided the histological evidence of the changes in the condyles, rami, and articular fossae when the mandible is advanced in a laboratory setting. Previous reports have showed an increased proliferation of the condylar cartilage in monkeys that had their mandible advanced with the Herbst appliance. These adaptations occurred primarily in the posterior and posterosuperior regions of the condyle. They also reported significant bone deposition along the posterior border of the mandibular ramus during the early part of the experimental period as well as significant deposition of new bone on the anterior surface of the postglenoid spine at the articular fossae. It should be noted, however, that the post-glenoid spine is not well-defined in humans.

   Clinical investigations using Magnetic Ressonance Imaging (MRI) and lateral radiographs have showed that changes in the mandibular condyles and articular fossae may occur with Herbst therapy. It has been demonstrated that following Herbst therapy some articular fossa remodeling can be seen at the anterior surface of the postglenoid spine, which causes a relocation of the articular fossae in a downward and forward direction. But such fossa adaptation in Herbst patients is less extensive than in experimental animals, with much individual variation in response evident.

   Although most Herbst studies are convergent as to the extent of skeletal and dentoalveolar adaptations, some questions still remain:

   • How much mandibular growth can be achieved with therapeutic mandibular advancement?
   • How much mandibular rotation and skeletal bite opening occurs?
   • How much of the original mandibular advancement is maintained long-term?
   • Is there is an adaptive behavior of the articular fossa after mandibular advancement, with bone remodeling occurring in the fossa.

   In the last decade, a new methodology using 3D virtual modeling has been developed allowing a change in the paradigm of assessing the skeletal changes associated with growth and treatment; this technology that has opened new horizons in scientific investigations of dentofacial orthopedics. Recently, the American Journal of Orthodontists and Dentofacial Orthopedics published the first 3D report on Herbst treatment, a pilot study in which 7 Class II subjects treated with the Herbst appliance were compared with 7 Class II subjects treated with fixed appliances and intermaxillary elastics. It was reported that Class II patients undergoing Herbst treatment demonstrated anterior displacement of the condyles and articular fossae along with maxillary restraint when compared with the Class II patients used for comparison.

   The aim of the present prospective controlled clinical trial is to investigate the skeletal changes in the mandible and articular fossae in Class II patients undergoing Herbst appliance therapy, using 3D virtual models in comparison to a control group comprising Class II subjects.

2. PATIENTS AND METHODS b.1) Study design

This is a non-randomised controlled clinical trial, with intervention. Two groups are been planned: 1) Class II malocclusion subjects treated with Herbst appliance (HAG); and 2) Class II malocclusion subjects treated only with teeth alignment. The two groups are matched by stage of skeletal maturation, type of malocclusion, chronological age, and gender. There is no blinding for the outcome evaluators. Due to ethical concerns, the comparison group will be composed of Class II subjects that need previous alignment and leveling of the teeth before orthopedic phase of their treatment.

b.2) Participants - settings and location where the data are collected The treatment are been performed in the Orthodontic Clinic of Pontifical Catholic University of Minas Gerais (Belo Horizonte, Brazil). This is a private university, but the patients are mostly low-income subjects. Data has been collected from August 2013 and finished on September 2015.

b3) Sample size Based on the standard deviation of 1.85 mm reported by Pancherz et al.28, an alpha significance level of 0.05 and a power of 0.80 to detect changes of 1.5 mm, a preliminary sample size of 25 patients per group was calculated.

Therefore, 25 individuals have received a Herbst appliance at the beginning of their treatment with a one-step full activation - Herbst Appliance Group (HAG); while 25 skeletal Class II malocclusion individuals, with indication for HA therapy, but with other clinical conditions that required prior treatment before HA insertion, have been allocated to the CG. The two groups have been matched by chronological age (between 12 years and 16 years-old), by stage of skeletal maturation and by the stage of dental development (permanent dentition).

b4) Appliance design

The Herbst appliance design included bilateral telescoping arms (3M Abzil, São José do Rio Preto, Brazil) articulated with pivots that were positioned in both the maxillary and mandibular arches. The pivots were welded to a rigid cantilever wire (.040-in stainless steel) extending from the lower first permanent molar bands (TP Orthodontics, La Porte, IN) to the canine regions of the mandible. In the maxillary arch, the pivots were welded to bands on the permanent first molars. A Hyrax expander (Morelli Ortodontia, Sorocaba, Brazil) and a .040-in stainless steel (SS) lower lingual arch were added to the HA structure to improve appliance stability and transverse dimensional control. Two .028-in SS wires were used as occlusal rests in the permanent second molars to avoid their extrusion after activation. The occlusal rests were removed when they interfered with the occlusion to avoid bite opening.

b5) Method of registration

Cone beam computed tomography (CBCT), intra-oral pictures, and extra-oral pictures have been collected in three time-points for HAG (T0, T1, and T2), and in two time-point for CG (T0, and T2), as a regular protocol of orthodontic record for these clinical treatments.

b6) Method of measurement

Analysis of serial CBCT images to evaluate changes between T0, T1, and T2 include 3D analysis methodology using ITK-SNAP (open-source software, www.itksnap.org); SLICER (open-source software, www.slicer.org; and VAM - Vectra Analysis Model software version 3.7.6 (Canfield Scientific Inc., Fairfield, NJ). The tridimensional image analysis procedures included: (1) approximation of scans; (2) construction of 3D label models; (3) voxel based image registration; and (4) quantitative and qualitative assessments using 3D mesh surface models.

Quantitative assessment of the positional changes between the 3D surface models of the cranial base, maxilla, and mandible will be performed using point-to-point landmarks measurements (VAM software), and virtual analytics (SLICER software).

Interactive visual analytics include graphic displays with qualitative assessments using semitransparent overlays of the 3D surface models at the different time-points and quantitative assessments of the surface displacements using color-coded surface distance maps. All visual analytics assessments will be performed using two modules (Model-to-Model Distance and Shape Population Viewer) in the SLICER software.

Voxel based registration on the anterior cranial fossae will be used to assess maxillary and mandibular displacement/articular fossae remodeling; regional registration on the mandibular symphysis will be performed to evaluate mandibular growth and mandibular dental changes; and regional registration on the anterior region of the maxilla will be performed to evaluate maxillary growth and maxillary dental changes.

b7) Method of analysis

For the primary and secondary outcome measures it will be used Statistical Package for the Social Science (SPSS) software (version 16.0 Chicago, IL). Descriptive statistics will be presented for both groups. Statistical differences between the two groups will be calculated with Mann-Whitney test (comparison between the difference of the means of the HAG and CG). The comparison between the changes within each group will be performed using the Wilcoxon's test.

Inclusion Criteria:

• Class II malocclusion with mandibular retrognathism
• Convex Profile
• Pubertal period
• Permanent dentition
• Parental permission with a signed consent form

Exclusion Criteria:

• Syndromic patients
• Previous orthopedic treatment