The efficacy of maxillary protraction protocols with the micro-implant-assisted rapid palatal expander (MARPE) and the novel N2 mini-implant—a finite element study

Background Maxillary protraction with the novel N2 mini-implant- and micro-implant-assisted rapid palatal expander (MARPE) can potentially provide significant skeletal effects without surgery, even in older patients where conventional facemask therapy has limited skeletal effects. However, the skeletal effects of altering the location and direction of force from mini-implant-assisted maxillary protraction have not been extensively analyzed. In this study, the application of the novel N2 mini-implant as an orthopedic anchorage device is explored in its ability to treat patients with class III malocclusions. Methods A 3D cranial mesh model with associated sutures was developed from CT images and Mimics modeling software. Utilizing ANSYS simulation software, protraction forces were applied at different locations and directions to simulate conventional facemask therapy and seven maxillary protraction protocols utilizing the novel N2 mini-implant. Stress distribution and displacement were analyzed. Video animations and superimpositions were created. Results By changing the vector of force and location of N2 mini-implant, the maxilla was displaced differentially. Varying degrees of forward, downward, and rotational movements were observed in each case. For brachyfacial patients, anterior micro-implant-supported protraction at −45° or intermaxillary class III elastics at −45° are recommended. For dolicofacial patients, either anterior micro-implants at −15° or an intermaxillary spring at +30° is recommended. For mesofacial patients with favorable vertical maxillary position, palatal micro-implants at −30° are recommended; anterior micro-implants at −30° are preferred for shallow bites. For patients with a severe mid-facial deficiency, intermaxillary class III elastics at −30° are most effective in promoting anterior growth of the maxilla. Conclusions By varying the location of N2 mini-implants and vector of class III mechanics, clinicians can differentially alter the magnitude of forward, downward, and rotational movement of the maxilla. As a result, treatment protocol can be customized for each unique class III patient.


Background
Facemask has shown some effectiveness in modifying growth and eliminating surgery in select patients [1]; however, there are many limitations. Unwanted dental side effects include proclination of maxillary incisors and extrusion and mesial tipping of the maxillary molars [2][3][4][5]. In addition, the rotation of the occlusal plane with conventional facemask has been observed leading to the development of an adjustable facemask that allows altering the line of force to achieve the desired skeletal movement [6].
To overcome these limitations, some researchers propose the use of miniplates for skeletal anchorage [7,8], which has resulted in greater skeletal effects, even in older patients. Numerous studies have shown the effectiveness of the use of miniplates placed in different locations within the maxilla in the clinical treatment of class III patients [9,10]. While Lee et al. [10] showed that the site of miniplate placement should be specifically considered between the infrazygomatic crest and the lateral nasal walls, Kim et al. [9] showed that the placement of miniplates in the palatal area resulted in wider stress distribution and more forward displacement compared to miniplates placed at the infrazygomatic crest area and conventional tooth-borne appliances.
With the use of micro-implants, it is now possible to achieve skeletal anchorage without the surgical procedures involved with placement and removal of miniplates. While micro-implant-supported facemask has shown evidence of clinical success [11,12], few studies have been published on this technique and few have analyzed how micro-implants can be utilized for maximal clinical effectiveness [11,13,14].
With the recent development of the novel N2 miniimplant (3-mm diameter, 2-mm length, tapered shape), it is believed that the short length of the implant reduces the risk of damaging anatomic structures during placement and therefore does not need to be placed interradicularly [15][16][17]. The aim of this study is to examine the ability of the novel micro-implant-assisted rapid palatal expander (MARPE) and N2 mini-implant to serve as an orthopedic anchorage device in creating favorable maxillary protraction protocols in lieu of the more invasive miniplates.
A finite element approach was used to simulate conventional facemask therapy, MARPE, and seven novel N2 mini-implant-supported maxillary protraction protocols to evaluate the corresponding stress patterns and skeletal changes. Our objective was to explore the ability of the novel N2 mini-implant to be used as an orthopedic anchorage and to understand how different placement locations and force directions can be used to correct different types of class III malocclusions.

Methods
The finite element method (FEM) model was generated from CT volumetric data (slice thickness of 0.300 mm) of a 42-year-old male patient of the Department of Biomedical Sciences at Ohio University, where informed consent was obtained prior to data collection. DICOM raw data was extracted from the CT and imported into Mimics 13.1 software (Materialise, Leuven, Belgium) to reconstruct a 3D model (Wu Laboratory, UCLA Bioengineering).
Utilizing Mimics, sutures 1.5-2 mm in width were manually generated ( Fig. 1): mid-palatal, pterygomaxillary (2), zygomaticomaxillary (2), zygomaticotemporal (2), median nasal, and lateral nasal (2). The FEM model generated from the ANSYS software yielded 91,933 elements and 344,451 nodes. Nodes along the foramen magnum and on the center of the forehead were constrained in all degrees of freedom, with zero displacement and zero rotation.
The 3D mesh was then imported into ANSYS 12.0 (ANSYS Inc, Canonsburg, PA) FEM software program. The FEM model was defined to be linear elastic and isotropic. The bone and tooth structures in the model were assigned properties of compact bone: Poisson's ratio of ν = 0.3 and a Young's modulus of E = 1.37 × 10 3 (kg/mm 2 ) [11,18]. Sutural elements were assigned values of connective tissue: ν = 0.49 and E = 6.8 × 10 −2 (kg/mm 2 ) [11,13]. Table 1 shows different locations and directions of force application, which simulate eight clinical protocols for maxillary protraction. The location of force delineates where the elastics of the facemask appliance pull from in simulation A and the location of micro-implant placement in simulations B-H. The direction of force (denoted by the [angle]) delineates protraction in relation to the occlusal plane. Values of 1000 g per side were applied for all simulations.
Simulation A mimics conventional facemask therapy, with force applied to the buccal of the first maxillary molars, angled 30°below the occlusal plane (Fig. 2). Simulation B models a micro-implant-supported hyrax with facemask shown in Fig. 3. The forces are applied 3 mm lateral to the mid-palatal suture, at a 30°angle below the occlusal plane (Fig. 4). For simulations C, D, and E, facemask therapy directly from anterior micro-   implants placed between the canine and first premolar roots is being modeled. The direction of force application is respectively 15°, 30°, and 45°below the occlusal plane (Fig. 5). Simulation F also has forces applied in the anterior between the canine and first premolar roots, but is directed 30°above the occlusal plane, simulating use of an intermaxillary spring pushing the maxilla forward and upward (Fig. 6). Finally, simulations G and H model the use of intermaxillary elastics from posterior maxillary micro-implants to anterior mandibular microimplants. Point of force application is in between the roots of the second premolar and first molar, directed 30°and 45°below the occlusal plane ( Fig. 7).
With ANSYS software: 1. Tensile and compressive stresses were measured separately in all simulations. 2. Video animations were developed. 3. Superimpositions were created to depict the skeletal displacement as a result of altering the location and direction of force application.

Results
First principle stress distributions depicting tensile stress are shown (Figs. 8 and 9). The color scale at the bottom of the figure shows the distribution of stress from the lowest (blue) to the highest (red). For all simulations, there is a high concentration of tensile stress directly posterior to the location of force application. In addition, tensile stresses concentrate at the pterygoid plates in all simulations, with slightly lower levels in simulation E. Tensile stresses are also evident in the medial orbit of all simulations, with greater levels in simulations E, F, and H. Other simulations have several areas of stress concentration unique to the simulation. In simulation A, tensile stresses are also present in the maxillary buttress. In simulations E and H, tensile stresses congregate near the lateral orbit, as well as the frontal process distal to the zygoma. The most unique pattern of tensile stress involves simulation F, with additional tensile stresses involving the forehead, orbit, zygomaticomaxillary suture, frontal process of the zygoma, and extending from the pterygoid plates backward to the foramen magnum.
Third principle stresses depicting compressive stress are also shown (Fig. 10). Dark blue areas show areas of high compressive stress.
For all simulations, there is a high concentration of compressive stress directly anterior to the location of force application. All simulations display compressive stresses in the forehead, with significantly less in simulation F. There is also a general concentration of stresses lateral to the infraorbital foramen, although this is minimal in simulations E and H. The distribution of stresses is distinctly unique in simulation F, with areas of compressive stress involving the lateral border of the orbit, medial superior border of the orbit, maxillary buttress, zygoma, and in the frontal bone distal to the zygoma.
Animation videos and superimpositions of each simulation were created. (simulations A-H). For the superimpositions, the "before" image is shown in blue, while the "after" image is displayed in a range of colors that directly correspond to the amount of Y-displacement (pure protraction) following force application (Figs. 11,12,13,14,15,16,17,18). As the color approaches red in the rainbow spectrum of colors, there is more Ydisplacement of the skull model. Table 2 summarizes the significant findings from each simulation video and superimposition.

Discussion
The hypothesis being tested in this study is that different placement locations and forces applied on the novel N2 mini-implants can translate into different types of maxillary protraction. The development of different clinical protraction protocols was achieved through the various simulations.
For all simulations, 1000 g of protraction forces were applied, as studies have shown that 500-1500 g is an appropriate force load for maxillary protraction [19]. As expected, tensile stresses were visible posterior to the site and direction of force application, while compressive stresses were observed anterior. This phenomenon was observed in all simulations. This is logical, as pulling anteriorly will tend to compress the anterior and place tension on the posterior.
The video animations and superimpositions clearly portray the skeletal effects in each simulation. Because the attachment of the maxilla to the skull is complex, the center of rotation changes depending on the location and vector of force. Varying the location and direction  of force produces translation, clockwise rotation, or counter-clockwise rotation of the maxillary complex. For example, a force that is located more anteriorly in the maxilla with a strong downward vector will tend to tip the anterior segment downward. The contour plots showing first and third principle stresses also correspond with the general rotation or translation of the maxilla.
However, simulations that result in the same rotation or translation movement do not necessarily have the same anterior-posterior movement, vertical movement, and rotation. A simulation may result in upward movement of the maxilla, while another results in downward movement. In addition, the amount of forward displacement also depends on the location and direction of force. By varying the location and vector of force, the effect on sutural expansion changes. The complex anatomy of the skull and circummaxillary sutures directly influences the skeletal response from different maxillary protraction protocols. As a result, the video animations and superimpositions provide valuable additional information that can be applied to treatment of patients in the clinical setting.
Several simulations display counter-clockwise rotation of the maxilla. Conventional facemask therapy has the point of force application furthest from the maxillary fulcrum of rotation. As a result, there is significant counter-clockwise rotation. The simulation is an oversimplification, as the force is applied directly to the maxillary molars, and incorporation of periodontal ligament (PDL) into the FEM model would have revealed extrusion, buccal tipping, and mesial movement of the maxillary molars and proclination of the maxillary incisors. Also, force dissipation to the PDL results in greater dentoalveolar compensations and fewer skeletal effects. PDL could not be incorporated into the FEM model, as the element size was larger than the width of PDL. Even with PDL, the simulation will not show alveolar bone remodeling as the teeth are displaced, and the dental movement will be underestimated when using the FEM. As a result, incorporation of PDL still produces an imperfect model.
Anterior  , which simulates facemask protraction from anterior micro-implants, displaces the maxillary complex significantly downward and backwards, which is counterintuitive, given that the goal of maxillary protraction is to move the maxilla forward. In this case, the class III will be corrected by the clockwise rotation of the  elastics between micro-implants is most effective in promoting the anterior growth of maxilla.
The clinical application of these results is evident, given that no two patients are identical. Even though our overall goal in treating class III growing individuals may be to protract the maxilla, each patient presents with different clinical findings. Patients may be brachyfacial or dolicofacial, may present with a deep bite or open bite, may have varying severities of skeletal A-P discrepancy, and may differ in the vertical positions of the maxilla resulting in varying degrees of gingival display. As a result, the location and direction of pull from micro-implants can be altered for maximal clinical effectiveness, as each simulation resulted in a different skeletal effect.
While the study simulates the skeletal response to force application, no simulation is as reliable as real life, so there are some limitations involved with the study. First, the model utilized in this study was generated from a CT scan of a 42-year-old male patient. Even though the literature states that there are no differences in using an adult model for FEM studies [20], sutural morphology and bone anatomy may change slightly throughout growth.
The material properties of bone in this study were assumed to be linear elastic, isotropic, and time independent. This is a simplification, as real bone is more  anisotropic and time dependent. Bone growth cannot be simulated; we can only show how the internal strain can promote or restrict growth. In addition, no distinction was made between cancellous and cortical bone. The material properties of sutures, while based on several publications, are still an approximation and may not be 100% coincident with connective tissue. Suture morphology was estimated, and due to limitations in computer memory and model element size, PDL could not be accurately incorporated into the model. Logically, the PDL should absorb a proportion of the forces resulting in displacement of the teeth. In addition, FEM cannot simulate the resorption and deposition of bone that should occur in the dentoalveolar bone.
However, it may be difficult to create a perfect model with ideal properties. Anatomy and material properties change from one person to another, thus making the problem more complex.
Without PDL, soft tissues, and remodeling within the model, the teeth are essentially ankylosed within the dentoalveolar bone. While the simulation does not accurately predict the dental effects [21][22][23], it does, however, demonstrate the skeletal effects, which is of primary concern when it comes to growth modification procedures like maxillary protraction. This study is valid in comparing how location and vector of force alters the skeletal effects. As a result, the location and direction of

Conclusions
Depending on the location and direction of force, the maxillary complex rotates clockwise, counter-clockwise, and/or translates anteriorly and vertically. By varying the location and vector of class III mechanics, orthodontists can differentially alter the magnitude of forward, downward, and rotational movement of the maxilla. As a result, we can customize the location of micro-implants and direction of force based on the patient's needs.