Maxillary protraction anchored on miniplates versus miniscrews: three-dimensional dentoskeletal comparison

Abstract

Objective

This retrospective study aimed to compare the three-dimensional (3D) outcomes of the novel miniscrew-anchored maxillary protraction (MAMP) therapy and the bone-anchored maxillary protraction (BAMP) therapy.

Methods

The sample comprised growing patients with skeletal Class III malocclusion treated with two skeletal anchored maxillary protraction protocols. The MAMP group comprised 22 patients (9 female, 13 male; 10.9 ± 0.9 years of age at baseline) treated with Class III elastics anchored on a hybrid hyrax expander in the maxilla and two mandibular miniscrews distally to the permanent canines. The BAMP group comprised 24 patients (14 female, 10 male; 11.6 ± 1.1 years of age at baseline) treated with Class III elastic anchored in two titanium miniplates in the infra-zygomatic crest and two miniplates in the mesial of the mandibular permanent canines. Three-dimensional displacements were measured in the pre- and post-treatment cone-beam computed tomography scans superimposed on the cranial base using the Slicer Automated Dental Tools module of 3D Slicer software (www.slicer.org). Mean differences (MD) between groups and 95% confidence interval (CI) were obtained for all variables. Intergroup comparison was performed using the Analysis of Covariance (P < .05).

Results

Both groups showed improvements after treatment. The MAMP group showed a smaller anterior (MD: −1.09 mm; 95% CI, −2.07 to −0.56) and 3D (MD: −1.27 mm; 95% CI, −2.16 to −0.74) displacements of the maxilla after treatment when compared with BAMP. Both groups showed negligible and similar anteroposterior changes in the mandible (MD: 0.33 mm; 95% CI, −2.15 to 1.34). A greater increase in the nasal cavity width (MD of 2.36; 95% CI, 1.97–3.05) was observed in the MAMP group when compared with BAMP.

Limitations

The absence of an untreated control group to assess the possible growth impact in these findings is a limitation of this study.

Conclusion

Both BAMP and MAMP therapies showed adequate 3D outcomes after treatment. However, BAMP therapy produced a greater maxillary advancement with treatment, while MAMP therapy showed greater transversal increases in the nasal cavity.

Introduction

Facemask therapy is the most common orthopedic management of Class III malocclusion in growing patients [1]. Ideal treatment timing for facemask therapy is in the deciduous and early mixed dentition due to a greater orthopedic effect than in the permanent dentition [2–4]. In late mixed or early permanent dentition, facemask therapy caused limited maxillary advancement while considerable dentoalveolar effects, including the mesial movement of the maxillary posterior teeth, are observed [2, 5].

The use of skeletal anchorage as miniplates and miniscrews has revolutionized the possibilities of orthopedic approaches for Class III malocclusion orthopedic treatment. An innovative maxillary protraction orthopedic approach using titanium miniplates as anchorage was described in 2009 [6]. The bone-anchored maxillary protraction (BAMP) therapy comprises Class III elastics attached to titanium miniplates placed in the infra-zygomatic maxillary crest and between the mandibular canines and lateral incisors, bilaterally [6]. BAMP therapy produced an anterior displacement of the maxilla in the late mixed and early permanent dentition [6, 7]. A mean maxillary advancement of 3.5 mm and minimal undesirable dentoalveolar effects were found after BAMP therapy [7]. Although BAMP therapy was an effective treatment option, the economic costs considering the miniplates and surgical procedures might be a limitation for low/medium socioeconomic level families. In addition, the miniplate placement requires a more invasive surgical procedure performed by a maxillofacial surgeon.

BAMP-derived therapies have been described in the past years. Wilmes et al. described the use of Class III elastics anchored on a hybrid expander in the maxilla and a mentoplate in the anterior region of the mandible [8]. To simplify the therapy and reduce the costs, Miranda et al. proposed the miniscrew-anchored maxillary protraction (MAMP) therapy [9]. MAMP is a BAMP-derived therapy replacing all miniplates with miniscrews as anchorage and including rapid maxillary expansion [9]. In MAMP therapy, Class III elastics are attached to the maxillary first molars anchored in a hybrid hyrax expander to mandibular miniscrews [9]. MAMP therapy increased 1.37° and 1.89 mm in the SNA angle and maxillary length, respectively. The maxillomandibular relationship improved after MAMP, with an increase of 1.82 mm in the Wits appraisal and a restriction of the mandibular displacement (−0.29° change in the SNB after treatment) [10]. The miniscrews have small costs and a simplified placement and removal surgical intervention. The miniscrews placement can be accomplished by the orthodontist. Conversely, miniplates can present higher stability rates during treatment and can be maintained for the retention period [7, 11, 12].

Both MAMP and BAMP therapies are adequate options to treat growing Class III patients [7, 10]. However, there is no previous study in the literature comparing the midface protraction and dentoskeletal changes produced by MAMP and BAMP therapies.

Objective

The aim of this study was to compare the three-dimensional (3D) outcomes of the novel MAMP therapy with the BAMP therapy. The hypothesis was that MAMP and BAMP therapy present similar 3D facial changes after treatment.

Material and methods

Study design

This is a retrospective study composed of two comparison groups. This study was approved by the institutional research ethical committee of Bauru Dental School, University of São Paulo (protocol number: 67610717.7.0000.5417) and by the Institutional Review Board of the University of Michigan, School of Dentistry (HUM00217429). All participants and parents signed a written informed consent before treatment.

Study size

Sample size calculation considered a power of 80%, an alpha error of 5%, a minimum difference to be detected of 1.5 mm in the maxillary anteroposterior displacement, and a standard deviation of 1.66 mm observed in a previous study [13]. A minimum data sample of 21 patients was required for each group.

Setting

The MAMP group comprised secondary data from patients prospectively treated at the Orthodontic Clinic of Bauru Dental School, University of São Paulo, between 2017 and 2022. The BAMP group comprised secondary data from patients consecutively treated at a private practice in Brussels, Belgium, between 2007 and 2015.

Participants

The sample comprised secondary data from patients prospectively treated with MAMP and BAMP therapy. The selection criteria were patients of both sexes in the late mixed or early permanent dentition with skeletal Class III malocclusion and maxillary deficiency (Wits appraisal of less than −1 mm), treated with MAMP or BAMP therapy. The exclusion criteria were craniofacial syndromes and a history of previous orthodontic treatment. Our sample comprised all patients meeting the inclusion criteria who were prospectively treated with either MAMP at our institution or BAMP therapy in Belgium during the study period. No eligible patients were excluded.

The MAMP group comprised 22 patients (9 female, 13 male; mean age of 10.9 years at baseline, SD = 0.9). The mean initial Wits appraisal in this group was −5.2 mm (SD = 2.1). MAMP therapy consisted of Class III elastics anchored on a hybrid hyrax expander (PecLab Ltda., Belo Horizonte, MG, Brazil) in the maxilla and on two miniscrews (PecLab Ltda., Belo Horizonte, MG, Brazil) in the mandible placed distally to the permanent canines, bilaterally (Fig. 1A and B). Parents were oriented to activate the expander screw 1/4 turn twice a day for 14 days. Class III elastics were recommended full-time using 150 g on each side in the first month and 250 g in the following months.

Figure 1.

MAMP therapy (A,B) and BAMP therapy (C,D).

The BAMP group comprised 24 patients (14 female, 10 male; mean age of 11.6 years at baseline, SD = 1.1). The mean initial Wits appraisal in this group was −4.8 mm (SD = 2.8). BAMP group was treated with Class III elastics anchored on two miniplates (Bollard, Tita-Link, Brussels, Belgium) in the posterior region maxilla positioned in the infra-zygomatic crest and two miniplates (Bollard, Tita-Link, Brussels, Belgium) placed bilaterally in the mandible mesially to the permanent canines (Fig. 1C and D). Full-time Class III elastics were recommended 3 weeks after miniplate surgery initially using 150 g on each side in the first month and increasing to 250 g in the following months.

In cases where miniscrews or miniplates became unstable before achieving a positive overjet, they were replaced, and treatment continued. All patients, regardless of such complications, were included in the analysis.

Variables

Cone-beam computed tomography (CBCT) scans were obtained immediately before (T1) and after treatment (T2). Three-dimensional displacements were assessed using previously validated methods of cranial base superimposition of the pre- and post-treatment CBCT scans [14, 15]. Two open-source software packages ITK-SNAP, version 3.8 (www.itksnap.org) [16], and 3D Slicer, version 5.1.0 (www.slicer.org) [17] were used. The Slicer Automated Dental Tools module was used to perform artificial intelligence-based automatic segmentation of craniofacial structures, automatic landmark location, and automatic quantification of the 3D displacements [18]. Initially, all scans were oriented using the Frankfurt horizontal plane perpendicular to the midsagittal plane [19]. Secondly, segmentations of the craniofacial structure were obtained using the automatic multi-anatomical skull structure segmentation algorithm [20]. A voxel-based registration was performed to superimpose the T2 scan over the oriented T1 scan, using the cranial base as a reference. Landmarks were placed automatically in the maxilla, mandible, and dental structures (Fig. 2). After automatic placement, the examiner refined all landmark locations [21]. The quantitative assessments were performed using the 3D Slicer ‘AQ3DC’ tool. Figures 3 and 4 show the superimpositions of pre- and post-treatment CBCT images on the cranial base for the MAMP group. Figures 5 and 6 show the pre- and post-treatment cranial base superimpositions for the BAMP group. Three patients in the MAMP and two patients in the BAMP group were excluded from the mandibular assessments because they were not in maximum intercuspation during the CBCT scan. Three-dimensional displacements and the amount of anteroposterior and supero-inferior displacements of each landmark from T1 to T2 surfaces were calculated. Maxillary and mandibular linear and angular measurements were also assessed. The gonial angle was calculated using the pitch rotation of the angle formed by the midpoint between the right and left condyle, the midpoint between the right and left gonion, and the menton landmark. Positive values corresponded to anterior and inferior displacements and labial/buccal inclination. Negative values corresponded to posterior and superior displacements and lingual inclination. Landmarks were placed in the incisal border and the root apex of the maxillary and mandibular central incisors to calculate the labiolingual inclination (pitch rotation). For the analysis of the labiolingual inclination, the average between the right and left sides was considered.

Figure 2.

Landmarks placement. A point (A): The most posterior point of the concavity of the anterior region of the maxilla. Right and Left Orbitales (ROr and LOr): Landmark placed at the most inferior point on the lower portion of the right and left orbits contour. Right and left nasal cavity (RNC and LNC): landmark placed at the most inferior and external point of the concavity of the right and left nasal cavity. Right and left zygomatic bones (RZyg and LZyg): landmarks placed at the most inferior portion of the right and left zygomatic bones. Anterior nasal spine (ANS): landmark placed at the anterior nasal spine. Posterior nasal spine (PNS): landmark placed at the posterior nasal spine. Pogonion (Pog): landmark placed at the most anterior point of the symphysis. Menton (Me): landmark placed at the most inferior point of the chin. Right and left condyle (RCo and LCo): landmark placed at the most superior and central point of the right and left condyle. Right and left gonion (RGo and LGo): landmark placed in the projection of a virtual bisector of a line adjacent to the mandibular base and posterior border of mandible in the right and left side. Right and left maxillary central incisor occlusal (UR1O and UL1O): landmark placed in the middle of the incisal edge of the maxillary right and left permanent central incisor. Right and left maxillary first premolar (UR4O and UL4O): landmark placed at the most central point in the crown occlusal surface of the right and left maxillary first premolar. Right and left maxillary first permanent molar (UR6O and UL6O): landmark placed at the most central point in the crown occlusal surface of the right and left maxillary permanent first molar.

Figure 3.

Cranial base superimposition of the pre- (white) and post- (semitransparent, red) treatment 3D maxillary surface models in a sagittal view (right side) for patients treated with MAMP therapy.

Figure 4.

Cranial base superimposition of the pre- (white) and post- (semitransparent, red) treatment 3D mandibular surface models in a sagittal view (right side) for patients treated with MAMP therapy.

Figure 5.

Cranial base superimposition of the pre- (white) and post- (semitransparent, blue) treatment 3D maxillary surface models in a sagittal view (right side) for patients treated with BAMP therapy.

Figure 6.

Cranial base superimposition of the pre- (white) and post- (semitransparent, blue) treatment 3D mandibular surface models in a sagittal view (right side) for patients treated with VAMP therapy.

Statistical analysis

Repeated measurements on 30% of randomly selected patients were performed after a 30-day interval. The intraclass correlation coefficient (ICC) was used to assess the intra-examiner reproducibility. Mean differences (MD; MAMP-BAMP) were obtained for all measurements. Normal distribution was verified using the Shapiro-Wilk test. Intergroup comparisons of age, sex ratio, and maturational growth stage at baseline were assessed using t-test and chi-square test, respectively. Intergroup comparisons were performed using analysis of covariance, considering the initial age as a covariate. The baseline values were also used as a covariate when present. The open-source statistical software JAMOVI software (version 2.3) (https://www.jamovi.org.) and SPSS Statistical Software Package (Version 21.0; SPSS, Chicago, IL) were used. Results were considered significant at P < .05.

Results

The ICC ranged from 0.833 to 0.999 demonstrating good to excellent intra-examiner reliability [22].

Participants

The study groups were similar regarding sex and maturational growth stage but statistically different regarding initial age and treatment time (Table 1). The BAMP group demonstrated greater initial age and treatment time when compared with the MAMP group. Initial maxillary and mandibular lengths were similar between groups (Table 1).

Table 1.

Intergroup comparisons for sex ratio, age, maxillary and mandibular length, and maturational growth stage at baseline, treatment time, and overjet correction.

Variable		MAMP		BAMP		P
		(n = 22)		(n = 24)
		Mean	SD	Mean	SD
Initial age (y)		10.9	0.9	11.6	1.1	.026†*
Treatment time (m)		11.0	3.7	13.8	2.99	.015†*
		n	%	n	%
Sex	Female	9	40.9	14	58.3	.238‡
	Male	13	59.1	10	41.7
Com-A (mm)		74.83	4.89	73.13	4.21	.245†
Com-Gn (mm)		73.48	5.93	74.06	4.81	.732†
Maturational growth stage	Pre-peak (%)	16 (72.7)		15 (62.5)		.460‡
	Peak (%)	6 (27.3)		9 (37.5)
	Post-peak (%)	0 (0)		0 (0)
Overjet Correction	Yes (%)	18 (81.8)		21 (87.5)		.592‡
	No (%)	4 (18.1)		3 (12.5)

^† t test;

^‡Chi-square test; *Statistically significant at P < .05.

m, midpoint. y, years; m, months; mm, millimeters.

Main results

Similar frequencies of overjet correction were found between groups (Table 1). In the MAMP group, the overjet was fully corrected in 18 out of 22 patients (81.8%). The BAMP group demonstrated the overjet correction in 21 out of 24 patients (87.5%).

Both groups showed improvements after treatment (Figs 3–6). Table 2 shows the intergroup comparison of the anteroposterior, supero-inferior, and 3D displacements. BAMP demonstrated greater anterior displacements of the maxilla when compared with MAMP (Table 2). A significantly smaller anterior (MD: −1.09 mm; 95% CI, −2.07 to −0.56), 3D (MD: −1.27 mm; 95% CI, −2.16 to −0.74) displacements of the maxilla after treatment were observed for MAMP when compared with BAMP. In addition, the Orbitale (MD: −0.74; 95% CI, −1.31 to −0.37), Zygomatic (MD: −1.08; 95% CI, −1.87 to −0.56) and Upper Incisor (MD: −0.78; 95% CI, −2.26 to −0.07) landmarks also showed significant smaller anterior displacement in the MAMP group compared to BAMP.

Table 2.

Intergroup comparisons of 3D displacements (analysis of covariance).

Variable	MAMP		BAMP		Mean difference	95% CI		P
	Mean	SD	Mean	SD		L	U
Anteroposterior displacement (mm)
A	1.92	1.24	3.01	1.24	−1.09	−2.07	−0.56	.001*
Orm	0.63	0.76	1.37	0.74	−0.74	−1.31	−0.37	.001*
Zygm	0.80	0.92	1.88	1.16	−1.08	−1.87	−0.56	<.001*
Mx.1m	3.16	1.68	3.94	1.92	−0.78	−2.26	−0.07	.037*
Com	−0.38	0.62	−0.51	0.80	0.13	−0.51	0.52	.977
Gom	−0.42	1.01	−1.46	1.41	1.04	−0.22	1.49	.141
Pog	0.95	2.72	0.62	2.35	0.33	−2.15	1.34	.641
Supero-inferior displacement (mm)
A	0.15	0.85	0.34	1.37	−0.19	−0.83	0.62	.776
Orm	−0.29	0.39	0.08	0.84	−0.37	−0.77	0.07	.100
Zygm	0.07	0.63	1.10	1.05	−1.03	−1.40	−0.33	.002 *
Mx.1m	0.45	1.36	0.62	1.74	−0.17	−0.79	1.10	.746
Com	−0.08	0.37	0.27	0.78	−0.35	−0.68	0.21	.303
Gom	1.45	1.28	1.61	2.03	−0.16	−1.08	1.39	.804
Pog	1.41	2.13	0.71	2.93	0.70	−0.82	2.92	.264
3D displacement (mm)
A	2.19	1.18	3.46	1.11	−1.27	−2.16	−0.74	< .001*
Orm	0.92	0.67	1.75	0.70	−0.83	−1.30	−0.43	< .001*
Zygm	1.23	0.88	2.52	1.07	−1.29	−1.95	−0.70	< .001*
Mx.1m	3.48	1.73	4.54	1.68	−1.06	−2.33	−0.19	.022 *
Com	0.87	0.49	1.14	0.78	−0.27	−0.74	0.22	.290
Gom	2.14	1.16	2.95	1.55	−0.81	−1.33	0.58	.430
Pog	3.71	2.40	3.39	2.07	0.32	−1.51	1.71	.901

^*Statistically significant at P < .05.

Mx, Maxillary; Md, mandibular; m, midpoint. (−) values denote medial/posterior/superior displacements; (+) values denote lateral/anterior/inferior displacements.

Both groups showed similar anteroposterior changes in the mandible. In both groups, the condylion and gonion landmarks displaced toward the posterior while the pogonion displaced anteriorly.

The maxillary vertical changes were minimal in both groups, except for the Zygomatic bone which showed a greater inferior displacement in the BAMP group when compared with MAMP (MD: −1.03; 95% CI, −1.40 to −0.33).

Table 3 shows the comparison of transversal and linear changes between groups. A greater increase in the nasal cavity (MD of 2.36; 95% CI, 1.97–3.05), intermolar (MD: 4.06; 95% CI, 4.00–6.24), and interpremolar (MD: 3.12; 95% CI, 2.91–5.23) widths were observed for the MAMP group compared with BAMP. The mandibular length showed a similar increase between groups.

Table 3.

Intergroup comparisons of 3D changes for lateral displacements and linear changes (analysis of covariance).

Variable	MAMP		BAMP		Mean difference	95% CI		P
	Mean	SD	Mean	SD		L	U
Lateral displacements (mm)
OrR-OrL	1.72	1.48	1.50	1.30	0.22	−0.72	1.06	.701
ZygR-ZygL	2.36	1.36	1.63	2.00	0.73	−0.30	2.42	.125
NCR-NCL	2.19	0.86	−0.17	0.77	2.36	1.97	3.05	< .001*
Mx.4R- Mx.4L	2.96	1.43	−0.16	1.96	3.12	2.91	5.23	< .001*
Mx.6R- Mx.6L	3.83	1.66	−0.23	1.84	4.06	4.00	6.24	< .001*
Linear changes (mm)
Com-A	2.50	1.51	3.47	1.26	−0.97	−1.99	0.09	.072
Com-Gn	1.39	2.51	0.90	2.10	0.49	−1.67	1.51	.922

^*Statistically significant at P < .05.

Mx, Maxillary; Md, mandibular; _m, midpoint;  _R, right;  _L, left.

Covariates: initial age and baseline characteristic.

Table 4 shows the intergroup comparison for angular changes. The MAMP group showed a smaller counterclockwise rotation of the palatal plane when compared with BAMP (MD: −1.12; 95% CI, −1.89 to −0.58). The mandibular plane showed a greater counterclockwise rotation after BAMP therapy. A greater decrease in the gonial angle was observed for the BAMP group when compared with MAMP (MD: 1.59; 95% CI, 0.89–2.24). MAMP therapy caused a slight lingual inclination of the mandibular incisors, while BAMP caused a slight labial inclination. No differences in the labiolingual inclination of maxillary incisors were observed between groups.

Table 4.

Intergroup comparisons of 3D angular changes (analysis of covariance).

Variable		MAMP		BAMP		Mean Difference	95 per cent CI		P
		Mean	SD	Mean	SD		L	U
ANS-PNS		0.36	0.72	1.48	1.25	−1.12	−1.89	−0.58	< .001*
Gom-Me		0.10	1.45	1.04	1.50	−0.94	−2.12	−0.72	.037*
Com-Gom-Me		−0.74	1.06	−2.33	0.80	1.59	0.89	2.24	< .001*‡
Labiolingual Inclination	Mx.1	2.09	4.01	1.16	4.48	0.93	−2.46	2.76	.907
	Md.1	−0.58	4.58	1.92	2.35	−2.50	−5.17	−0.04	.046*

^*Statistically significant at P < .05.

Mx, Maxillary; Md, mandibular; _m, midpoint;  _R, right;  _L, left; ANS: anterior nasal spine; PNS: posterior nasal spine; Go: gonion; Me; menton; (−) values denote lingual inclination and clockwise rotation; (+) values denote labial inclination and counterclockwise rotation.

Covariates: initial age;  ^‡initial age and baseline characteristic.

Discussion

The orthopedic protraction of the maxilla for Class III malocclusion treatment can promote changes in the maxillofacial structure. For this reason, three-dimensional analysis offers the possibility of assessing these changes in the three space planes. The voxel-based superimposition of the cranial base for growing patients was previously validated and used in previous studies [13, 14, 23–28]. This method allows a 3D visualization of changes promoted by treatment and facial growth in relation to the cranial base. Previous studies demonstrated high reproducibility levels of the methods using voxel-based cranial base superimposition [23–28]. Validated artificial intelligence-based tools were also incorporated in this study to perform the segmentation of craniofacial structures and landmark locations [18, 20, 21]. The AMASSS and ALICBCT tools were used to segment the cranial base, maxilla, and mandible and automatically locate landmarks in the craniofacial complex, respectively [20, 21]. These tools present the benefit of simplifying and enhancing the efficiency of 3D assessments with high performance and accuracy [18, 20, 21]. In this study, good to excellent intra-examiner reliability was found for all measurements.

This study compared an innovative protocol of maxillary protraction using miniscrews with a gold-standard method using skeletal anchorage. In BAMP therapy, maxillary protraction is anchored on miniplates placed in the infrazygomatic maxillary crest and in the supra-menton region between the lateral incisor and permanent canines. The new method is a BAMP-derived therapy where miniplates are replaced by miniscrews and rapid maxillary expansion (RME) is incorporated into the protocol. In MAMP therapy, the anchorage is provided by palatal miniscrews incorporated in a hybrid expander in the maxilla and in the mandible by miniscrews placed between the permanent canines and first premolars. Full-time Class III elastics are used in both therapies to promote maxillary traction forces. Previous studies demonstrated that both Class III orthopedic approaches using miniplates and miniscrews as anchorage effectively treat Class III malocclusion during adolescence [7, 10]. The present study outcomes confirm that both MAMP and BAMP therapies promote adequate orthopedic effects after treatment.

MAMP therapy demonstrated smaller anterior displacements of the maxilla when compared with BAMP at the level of A point, orbitale region, zygomatic bone, and maxillary incisors. A study by Nguyen et al. demonstrated a mean 3D displacement of 3.7 mm for the maxilla, 3.7 mm for the zygomatic bone, and 4.7 mm for the maxillary incisors after BAMP therapy [13]. A possible explanation for the difference in the midface changes found between the groups in our study is that MAMP used a hybrid anchorage in the maxilla. In contrast, BAMP used a pure skeletal anchorage. Therefore, in MAMP therapy, part of the efforts of Class III elastics are lost with dental effects of mesial migration of the maxillary first molars and labial inclination of maxillary incisors [10]. The force vector in both therapies is located below the maxilla’s center of resistance (CR). However, the BAMP force vector is closer to CR than the MAMP, which can favor a greater displacement of the maxilla. In addition, previous studies have demonstrated that miniscrews can move according to orthodontic load which could have affected the maxilla’s anterior displacement in MAMP therapy [29–31]. The differences between MAMP and BAMP in the anterior displacement of midface varied only between 0.7 and 1.08 mm. When evaluating the anterior displacement of the A-point between groups, there is a mean difference of 1.09 mm, with a 95% confidence interval ranging from −2.07 to −0.56. Therefore, the upper (−0.56) and lower limit (−2.07) of the 95% CI indicates that the clinical decision between MAMP and BAMP therapy should also consider the invasiveness and financial cost of both therapies, given that the upper limit falls within values that may be considered not clinically relevant. BAMP and MAMP therapies were similar regarding the frequency of overjet correction after treatment. In addition, the total anterior displacement of the maxilla in MAMP is probably the association of the anterior displacement of the maxilla usually observed after RME and the anterior displacement produced by the Class III elastics.

No differences in the anterior displacement of the mandible were observed after MAMP and BAMP therapies. Previous studies demonstrated that the anterior mandibular displacements were adequately controlled during BAMP therapy. Comparing the BAMP therapy outcomes with an untreated Class III sample, greater increases in total mandibular length were observed in the untreated group [7]. Cephalometric analysis showed a decrease of 0.29° for the SNB angle after MAMP therapy [10]. In addition, no mandibular ramus and gonial angle changes were observed after MAMP treatment [10]. In a previous 3D assessment, a restriction of the chin position as well as a distal displacement of the posterior ramus was observed after BAMP [32].

Maxillary and mandibular vertical changes were similar in both groups, except for the zygomatic bone, which demonstrated a greater inferior displacement in the BAMP group when compared with the MAMP group. The possible explanation is that a greater anterior displacement of the midface in the BAMP group could lead to a greater vertical displacement considering the face displaces forward and downward during natural growth. Another assumption is that the maxillary miniplate is placed near the zygomatic bone in BAMP therapy. In addition, previous studies have shown that BAMP therapy results in a counterclockwise rotation of the maxilla, which explains the greater zygomatic landmark inferior displacement of 1.1 mm compared to MAMP [6, 13].

Differences between the groups were observed for the transversal displacements of the maxilla. A greater increase in the nasal cavity width was observed for the MAMP group when compared with BAMP. In addition, MAMP therapy produced a greater transversal increase in the maxillary first premolar and molar widths of ~3 mm. These differences were expected considering an RME was performed as part of the MAMP protocol.

Statistical differences were observed for angular changes. The BAMP group demonstrated a greater counterclockwise rotation of the palatal plane (ANS-PNS) when compared with MAMP. In addition, a greater closure of the gonial angle (Co_m-Go_m-Me) and a counterclockwise rotation of the mandibular plane was observed with BAMP when compared with MAMP. Differently from our findings, a previous cephalometric assessment demonstrated a 1° clockwise rotation of the palatal plane, a 4.1° closure of the gonial angle, and a counterclockwise rotation of the mandibular plane after treatment with BAMP [7]. Using MAMP therapy, a previous study demonstrated a minimal increase in the gonial angle (0.14°) after treatment, a slight counterclockwise rotation of the palatal plane (0.4°), and a 1° increase in the FMA angle [10]. No intergroup differences were observed in the labiolingual inclination of the maxillary incisors after treatment with MAMP and BAMP. However, BAMP therapy demonstrated a slight labial inclination of the mandibular incisors compared to a mild lingual inclination observed in the MAMP group. A labial inclination of the mandibular incisors after BAMP therapy was also demonstrated in a previous study [7].

This study highlights the benefits of BAMP and MAMP therapy from a 3D assessment. BAMP demonstrated greater maxillary displacements after treatment when compared with MAMP. The advantage of BAMP therapy is that the miniplates can be maintained in the long term as active retention, throughout the remaining growth spurt. Cases where a critical stability of maxillary protraction is expected as in a hyperdivergent facial pattern [33, 34] might benefit from BAMP therapy due to the possibility of long-term retention. This assumption should be evaluated in future studies. In cases with a severe maxillary deficiency, BAMP may be recommended due to greater maxillary displacement after treatment. In addition, BAMP therapy should also be used when long-term retention is necessary. In cleft lip and palate patients where a greater instability of the maxillary displacement and unfavorable growth are expected, BAMP long-term retention is highly recommended [35]. The advantage of MAMP therapy is that it involves a less invasive surgical intervention without needing a maxillofacial surgeon and the lower financial costs. Another advantage of MAMP is the transversal increase of the maxilla, which is indicated for Class III patients with maxillary constriction. MAMP should be recommended when there is a need for a lower-cost therapy with a light or moderate maxillary deficiency and favorable growth pattern. Countries with restricted access to Bollard miniplates may benefit from MAMP as an alternative treatment option. In addition, there is a need for long-term evaluations to assess the stability of the orthopedic effects of MAMP and BAMP therapies. Future studies should also analyze the 3D mandibular changes using as reference stable structures in the chin, to allow a throughout comparison of the growth changes and bone modeling in the ramus after both therapies.

Limitations

One limitation of this study is the absence of a control group to assess the possible growth impact of these findings. However, obtaining CBCT scans of untreated skeletal Class III malocclusion patients raises ethical issues. Another limitation is related to ethnical differences between the BAMP and MAMP groups in this study. The results should be interpretable with caution considering that MAMP group was treated in the Bauru Dental School, University of São Paulo, while BAMP group was treated at a private practice in Brussels, Belgium. The differences in the initial age and treatment time highlight this limitation. The ethnical differences can explain the differences in the initial age of the patients included in this sample, even though the dental age was similar (inclusion criteria). To overcome this, an analysis of covariance was used, considering the initial age as a covariate.

Generalisability

The generalizability of this study’s results should be limited to growing patients with Class III malocclusion with similar age ranges and similar initial dentoskeletal features.

Conclusions

1. BAMP therapy showed a greater maxillary advancement than MAMP therapy.

2. MAMP therapy showed greater transversal increases than BAMP therapy.

Author contributions

Felicia Miranda (Conceptualization [equal], Data curation [equal], Formal analysis [equal], Funding acquisition [equal], Investigation [equal], Methodology [equal], Resources [equal], Writing—original draft [lead]), Daniela Garib (Conceptualization [equal], Data curation [equal], Formal analysis [equal], Funding acquisition [equal], Methodology [equal], Project administration [equal], Resources [equal], Supervision [equal], Writing—original draft [equal]), Ivan Silva (Data curation [equal], Resources [equal], Writing—review & editing [equal]), José Carlos Bastos (Data curation [equal], Investigation [equal], Resources [equal], Visualization [equal], Writing—review & editing [equal]), Aron Aliaga-Del Castillo (Data curation [equal], Formal analysis [equal], Resources [equal], Writing—review & editing [equal]), Marilia Yatabe (Data curation [equal], Investigation [equal], Methodology [equal], Writing—review & editing [equal]), Hugo De Clerk (Formal analysis [equal], Investigation [equal], Resources [equal], Writing—review & editing [equal]), and Lucia Cevidanes (Conceptualization [equal], Formal analysis [equal], Funding acquisition [equal], Methodology [equal], Project administration [equal], Supervision [equal], Writing—review & editing [equal])

Conflict of interest

All authors declare that they have no conflicts of interest.

Funding

This study was financed, in part, by the São Paulo Research Foundation (FAPESP), Brasil. Process Number #2020/01698-5 and #2021/10792-8, and by the NIDCR R01 DE024450.

Ethical approval

This study was approved by the institutional research ethical committee of Bauru Dental School, University of São Paulo, (protocol number: 67610717.7.0000.5417) and by the Institutional Review Board of the University of Michigan, School of Dentistry (HUM00217429).

Data availability

The data underlying this article will be shared on reasonable request to the corresponding author.