Introduction
This study aimed to evaluate and compare the palatal dimensional and morphologic characteristics (symmetry or asymmetry) before and after the treatment of functional posterior crossbite (FPXB) in mixed and permanent dentition.
Methods
Forty-four patients with a diagnosis of transverse maxillary deficiency and FPXB underwent maxillary expansion: 22 in mixed dentition (mixed dentition study group [MD-SG], mean age = 8.6 ± 1.3 years) and 22 in permanent dentition (permanent dentition study group [PD-SG], mean age = 13.3 ± 1.1 years). Two age-matched control groups (mixed dentition control group and permanent dentition control group) were included. Digital models at T0 (baseline) and T1 (12-18 months postexpansion) were analyzed for palatal dimensions, volume, and symmetry. Deviation analysis and percentage matching were performed between the original and the mirrored models. Statistical analyses assessed intragroup, intergroup, and intertiming differences.
Results
At T0, linear and volumetric measurements were greater at the noncrossbite side (nCBs) than the crossbite side (CBs) in patients with FPXB compared with controls ( P <0.001). The CBs/nCBs volumetric difference was greater in PD-SG than MD-SG ( P = 0.001), whereas surface matching was higher in MD-SG ( P = 0.004). At T1, CBs/nCBs dimensional asymmetry decreased in both FPXB groups ( P <0.001), and the surface matching improved ( P <0.001). However, MD-SG remarkably showed greater posttreatment changes than PD-SG (volumetric data: P = 0.012; surface data: P = 0.009).
Conclusions
Patients with FPXB in permanent dentition could exhibit greater maxillary asymmetry than in mixed dentition. After reestablishment of normal occlusion, the asymmetry reduced in both MD-SG and PD-SG, though the latter group retained more residual asymmetry, potentially affecting final maxillary morphology.
Highlights
-
•
Patients with functional posterior crossbite present palate asymmetry compared with controls in mixed and permanent dentition.
-
•
After rapid maxillary expansion and reestablishment of normal occlusion, palate asymmetry improves.
-
•
Late treatment of functional posterior crossbite could retain residual asymmetry, affecting maxillary arch morphology.
Unilateral posterior crossbite is an asymmetrical malocclusion that accounts for 80%-97% of posterior crossbites in the mixed dentition. , It may result from a unilateral maxillary constriction or by relative mandibular expansion, a condition defined as true unilateral posterior crossbite. More frequently, it is caused by occlusal interferences in centric relation, which lead to a functional mandibular shift toward the crossbite side (CBs) upon closure into centric occlusion. , This condition is recognized as functional posterior crossbite (FPXB). In most instances, the interferences arise from a mild skeletal or dentoalveolar maxillary constriction, whereas, less frequently, they are a consequence of excessive transverse mandibular growth.
In patients with transverse maxillary deficiency, the primary treatment option involves the skeletal expansion of the maxilla. , Besides its role in re-establishing adequate maxillary skeletal dimensions, maxillary expansion also facilitates the correction of mandibular posture and function in patients with FPXB. , Regarding the timing of intervention, early treatment of FPXB has been advocated to prevent asymmetrical mandibular growth. , However, recent studies indicate that patients with FPXB do not exhibit an increased risk of mandibular morphologic asymmetry, or at least not to a clinically relevant extent, when compared with controls without crossbite or with bilateral crossbite. , On the contrary, FPXB can be associated with a morphologic asymmetry of the maxilla, because patients with FPXB in mixed dentition might present a symmetrical contraction of the basal bone and the skeletal asymmetry of the alveolar processes, with the CBs being narrower than the noncrossbite side (nCBs). ,
Two recent studies have also shown that maxillary asymmetry improves 1 year after maxillary expansion. , These studies were conducted on patients in mixed dentition, and their findings indicate that palatal expansion may offer the additional benefit of intercepting and mitigating maxillary asymmetry before the transition to the permanent dentition. In this context, understanding whether the baseline asymmetry is stable or can be corrected after maxillary expansion has significant clinical relevance when considering the future application of orthodontic biomechanics in permanent dentition or other dental rehabilitation procedures. However, there are no studies in the literature that have addressed the symmetry or asymmetry changes of the palate after maxillary expansion, comparing patients treated in mixed and permanent dentition.
In this regard, the aim of the present study was to comparatively investigate the dimensional and morphologic changes of the palate in patients affected by FPXB after maxillary expansion performed in both permanent and mixed dentition. Therefore, a specific 3-dimensional (3D) imaging technology was used to analyze the maxillary asymmetry from digital scans. The present study could provide additional diagnostic insights into the clinical orthodontic management of maxillary asymmetry in growing patients. The null hypothesis was the absence of significant differences between the data recorded before and after maxillary expansion in mixed and permanent dentition.
Material and methods
The present retrospective study received approval from the institutional ethical committee of the University of Catania (protocol No. 154/2022/PO- A.M.M.O.C.) and has been carried out following the Helsinki Declaration on medical protocols and ethics.
The study group (SG) of the present retrospective study consisted of 44 patients with a diagnosis of transverse maxillary deficiency and treated with maxillary expansion as part of their comprehensive orthodontic therapy at the Department of Orthodontics of the University of Catania between January 2013 and December 2024. The diagnosis of transverse maxillary deficiency was based on clinical judgment considering palatal morphology and dimension, even in association with the presence of buccal corridors. The study sample was retrieved from the orthodontic digital archive according to the following criteria: the inclusion criteria included a unilateral posterior crossbite of at least 2 posterior teeth, mandibular shift towards the crossbite site ≥2 mm in centric occlusion and not in centric relation (FPXB), Class I or Class II edge-to-edge molar relationship, and pretreatment (T0) and posttreatment (T1) records (≥12 and ≤18 months after expander removal); the exclusion criteria included an anterior crossbite, missing teeth, temporomandibular disorder, orthodontic treatment before T0 and T1 records acquisition, Class III (ANB <0°), cranial deformities or syndromic conditions, and asymmetrical design of the expander. Patients with Class III malocclusions were excluded because of the reported higher occurrence of mandibular asymmetry with this malocclusion. During the retrospective selection process, patients were distinguished into 2 groups according to the dentition stage: mixed dentition SG (MD-SG) and permanent dentition SG (PD-SG). As the number of patients retrieved in the MD-SG almost doubled the PD-SG, a web application ( www.randomizer.org ) was used to randomly select the same number of patients in the MD-SG. The MD-SG included 10 males and 12 females with a mean age of 8.6 ± 1.3 years, and the PD-SG included with a mean age of 13.3 ± 1.1 years. Two control samples were age-matched to the study samples, respectively—the mixed dentition control group (MD-CG: 9 males and 13 females; mean age 9 ± 1.2 years) and the permanent dentition control group (PD-CG: 10 males and 12 females; mean age 13.5 ± 0.9 years). The inclusion and exclusion criteria were the same as the SGs, plus the absence of posterior crossbite (either bilateral or functional crossbite). The same web application ( www.randomizer.org ) was used to randomly select the same number of patients in the 2 control groups. Supplementary Figure 1 shows the flow chart with a detailed description of the retrospective sample selection. In the present study, the STROBE (Strengthening the Reporting of Observational Studies [case-control studies] in Epidemiology) guideline was used as a reporting template ( Supplementary Table I ).
Because of differences in the dentition stages, the 2 groups underwent maxillary expansion using a Hyrax expander with distinct configurations. MD-SG had bands on primary second molars (alternatively on the permanent first molars) with extended arms to maxillary deciduous canines (alongside the enamel-gingival junction of posterior teeth); PD-SG had bands on permanent first molars with extended arms or bands on permanent canines ( Supplementary Fig 2 ).
The expansion protocol involved 3 activations per week (alternate days) in the MD-SG and 1 activation per day in the PD-SG until overexpansion was achieved, that is, when the mesiopalatal cusps of the maxillary first molars were in contact with the buccal cusps of the mandibular first molars. Skeletal expansion was confirmed by the appearance of a central incisor diastema in both groups. The appliance remained in place for 6-8 months as retention, and during this period, patients did not receive other orthodontic appliances. Dental impressions with bite registrations were taken before treatment (T0) and after 12-18 months (T1) without the appliance in place as part of new diagnostic records. According to the inclusion criteria, patients included in both MD-SG and PD-SG did not receive an orthodontic appliance (both in the maxillary and mandibular arches) during the interval time between the end of the retention period and the new records acquisition (T1). Plaster models at T0 and T1 were digitized using a D2000 3D desktop scanner (3Shape, Copenhagen, Denmark).
All maxillary digital models were imported into Ortho Analyzer (3Shape A/S, Copenhagen, Denmark) to perform 2-dimensional measurements (hemilinear distances) and the segmentation of the palate.
Two-dimensional measurements—To analyze the transverse dimensions of the maxillary arch, a median palatal plane (MPP) was generated on digital casts through 2 landmarks identified along the median palatal raphe (14) ( Fig 1 ): point 1 = the point on the median palatal raphe adjacent to the second ruga; point 2 = the point on the median palatal raphe 1 cm distal to point 1. Afterwards, the emi-palate distance between the midpoint at the dentogingival junction of the canine (D1) and first molar (D2) was calculated on both CBs and nCBs.
The MPP was defined using 2 landmarks along the median palatal raphe. The first landmark (point 1) was positioned at the raphe level adjacent to the second ruga, whereas the second landmark (point 2) was placed 1 cm distal to the first. D1 and D2 indicate the linear distances from the midpoint of the dentogingival junction of the canine and first molar to the MPP, respectively.
For 3D measurements, the same software was used to perform the segmentation of the palate at T0 and T1. The anatomy of the palate was isolated by generating a gingival plane passing through all the most apical points of the dentogingival junction of all teeth, from the right first molar to the left first molar ( Fig 2 , A and B ). To verify the morphologic changes (symmetry or asymmetry) and perform surface analysis of the palate, a specific 3D imaging technology involving superimposition of T0 and T1 palate digital models was carried out, according to a consolidated digital methodology. , The procedure involved 4 steps: (1) mirroring (3-Matic Medical software, version 13; Materialise NV, Leuven, Belgium): the segmented palate was duplicated and mirrored using the MPP as reference, that is the line passing through a point placed at the level of the second rugae and a second point 1 cm distal, along the palatal raphe ( Fig 3 , A and B ); (2) surface registration (3-Matic Medical software): the original and the mirrored palate models were superimposed via preliminary registration using MPP as reference plane, and a final registration was executed using the best-fit alignment feature in the software ( Fig 3 , C ); (3) volumetric assessment (3-Matic Medical software): the total palate volume was calculated at T0 and T1 along with hemilateral volumes (CBs and nCBs), the latter obtained using the same MPP used as reference for the mirroring procedure ( Fig 3 , D ); (4) deviation analysis and matching percentage calculation (Medit-Link version 3.3.6; Medit Corp Ltd, South Korea): the mean and maximum values of the linear distances (Euclidean distance) between the surfaces of the 2 palatal models, measured across 100% of the surface points. The analysis was complemented by the visualization of the 3D color-coded maps, set at a 0.5 mm range of tolerance ( green ), to better evaluate and locate the discrepancy between the model surfaces ( Fig 4 ). These values represented the degree of correspondence between the original and the mirrored models and, therefore, provided quantitative data on the morphologic characteristics of the palate detected at T0 and T1.
Segmentation of the palate and generation of the maxillary reference model. The anatomy of the palate was isolated by generating a gingival plane passing through all the most apical points of the dentogingival junction of all teeth, from the right first molar to the left first molar (A), and afterward, the palatal vault model was created (B) .
A, The mirroring process of the palate and the superimposition between the original (reference) and the mirrored model. Definition of the MPP; B, Generation of duplicated mirrored model of the palate ( pink , original; green , mirrored); C, Superimposition of the original and mirrored models using the MPP plane and its perpendicular plane as reference, and adjustment using best-fit alignment algorithm; D, Hemivolumetric assessment using the same MPP plane as reference.
Color-coded map generated from deviation analysis between the original and mirrored models of the palate in MD-CG and PD-CG (T0), MD-SG, and PD-SG (T0 and T1). The RGB colored scale bar (millimeters) is shown on the right ( red , maximum positive deviations; blue , maximum negative deviations; green , tolerance range, set to 0.5 mm). The number of polygons for surface representation was set to the maximum of 100,000; the corresponding polygons of the selected reference areas were automatically superimposed.
The entire workflow was carried out by a single expert operator (M.G.P.). After 4 weeks, the same measurements were repeated by the same operator to obtain data for intraoperator reliability. A second expert operator (A.L.G.) performed the digital workflow to obtain data for interoperator reliability assessment.
A pilot study was performed on 20 patients (MD-CG = 10 and PD-CG = 10) satisfying the inclusion and exclusion criteria to evaluate sample size power. The analysis showed that 17 patients for each group were required to detect a mean difference of 12,962 mm 3 between the intertiming mean difference (T1-T0) of the hemivolumetric differences (CBs– nCBs = asymmetry), with a power of 80% and a significance level of 0.05. However, we were able to include 22 patients per group, which increased the robustness of the data findings.
Statistical analysis
Descriptive statistics were designed to assess the demographic and clinical characteristics of the study sample that could represent confounders influencing data outcomes. In this regard, mean age and categorical characteristics (sex and skeletal maturity) among the MD-SG, MD-CG, PD-SG, and PD-CG were compared using the one-way analysis of variance and Chi-square test ( Table I ).
Table I
Demography and clinical characteristics of the sample of the study
| Sample characteristics | Mixed dentition | Permanent dentition | Significance | ||
|---|---|---|---|---|---|
| MD-SG (n = 22) | MD-CG (n = 22) | PD-SG (n = 22) | PD-CG (n = 22) | ||
| Mean/n | Mean/n | Mean/n | Mean/n | ||
| Mean age | 8.6 (± 1.3) | 9 (± 1.2) | 13.3 (± 1.1) | 13.5 (± 0.9) | P <0.05 |
| Gender | |||||
| Male | 10 | 9 | 8 | 10 | NS |
| Female | 12 | 13 | 14 | 12 | |
| Skeletal maturity | |||||
| CVMS 1 | 18 | 19 | 4 | 4 | P <0.05 |
| CVMS 2 | 4 | 3 | 12 | 11 | |
| CVMS 3 | 0 | 0 | 6 | 7 | |
| Skeletal sagittal classification | |||||
| Class 1 | 15 | 16 | 13 | 14 | NS |
| Class 2 | 7 | 6 | 9 | 8 | |
| Overjet | 3.6 (± 1.9) | 3.4 (± 1.5) | 2.8 (± 2) | 3.3 (± 1.6) | NS |
| Overbite | 2.6 (± 2.2) | 2.8 (± 1.9) | 4.1 (± 1.8) | 3.2 (± 1.5) | P <0.05 |
| Intermolar width | 29.77 (± 2.3) | 32.05 (± 1.3) | 31.35 (± 2.4) | 34.11 (± 1.6) | P <0.05 |
| Retention time | 6.9 (± 0.6) | – | 7.2 (± 0.9) | – | NS |
Note. Values are mean (± standard deviation); skeletal classification defined according to the ANBˆ value.
NS , not significant; CVMS, cervical vertebral maturation stage.
Preliminary data analysis was performed using the Shapiro-Wilk test and the Levene test to assess data distribution and equality of variance. As the data showed a normal data distribution, parametric tests were used. Paired-samples t test was used to assess the following: (1) hemilateral linear measurements and volumes between CB and nCB sides (intratiming assessment) at T0 and T1 in both groups, (2) intertiming comparison of the mean differences of CBs/nCBs side between T0 and T1 in both groups, and (3) intertiming comparison of the surface’s matching of percentage between T0 and T1 in both groups. Independent-samples t test was used to compare the following: (1) the mean difference of CBs/nCBs sides at the baseline between each group and the respective control group, (2) the mean differences between D1 and D2 measurements at T0 and T1, (3) the mean difference of CBs/nCBs between the 2 groups (intergroup mean difference evaluation) at T0 and T1, and (4) the mean difference of the T0-T1 surface’s matching of percentage between the 2 groups. Cohen d measure was performed to calculate the effect size of intertiming differences recorded in MD-SG and PD-SG. Statistical significance was set at P <0.05. Intraexaminer reliability was assessed using the intraclass correlation coefficient. Datasets were analyzed using SPSS (version 24; IBM, Armonk, NY).
Results
No differences were found between intraoperator readings, with excellent correlation indexes ranging 0.898-0.921 for linear measurements, 0.918-0.942 for volumetric measurements, and 0.933-0.950 for surface analysis. Similarly, no differences were found between interoperator readings, with excellent correlation indexes ranging 0.875-0.904 for linear measurements, 0.902-0.927 for volumetric measurements, and 0.920-0.935 for surface analysis.
In both MD-SG and PD-SG, linear and volumetric measurements at the nCBs were significantly greater than those at the CBs (MD-SG: D1 = nCBs 1.34 ± 0.92 mm > CBs; D2 = nCBs 1.80 ± 1.45 mm > CBs; volume = nCBs 332.52 ± 156.98 mm 3 > CBs and PD-SG: D1 = nCBs 1.07 ± 0.63 mm > CBs; D2 = nCBs 2.18 ± 0.96 mm > CBs; volume = nCBs 492.23 ± 140.27 mm 3 > CBs) ( Tables II and III ). These differences between CBs and nCBs were significantly greater than those observed in the corresponding control groups ( Supplementary Table II ). In MD-SG and PD-SG, the mean differences between both sides observed in the D2 region were statistically greater than those found in the D1 region ( Table II and Supplementary Table II ). The superimposition of the original and mirrored palatal models revealed a lower percentage of agreement in both groups compared with controls (MD-SG = 69.46% ± 11.30%; PD-SG = 59.76% ± 9.63%) ( Supplementary Table II ). These data supported the presence of a dimensional and morphologic asymmetry in the maxilla for both mixed and permanent dentition stages (CBs < nCBs) when compared with controls.
Table II
Comparative assessment of emi-lateral palatal volumetric measurements recorded in the MD-SG and PD-SG
| Group | Timing | Side | D1 (mm) | Difference | P value | Mean difference of the differences | P value | P value | P value |
|---|---|---|---|---|---|---|---|---|---|
| MD-SG | T0 | CBs | 10.87 (± 1.13) | 1.34 (± 0.92) | P <0.001 | 0.98 (± 0.62) | P <0.001 | P = 0.041 | Baseline T0 (MDG vs PDG): P = 0.283 |
| nCBs | 12.20 (± 1.08) | ||||||||
| T1 | CBs | 13.87 (± 1.30) | 0.36 (± 0.66) | P = 0.053 | |||||
| nCBs | 14.23 (± 1.14) | ||||||||
| PD-SG | T0 | CBs | 11.61 (± 1.61) | 1.07 (± 0.63) | P <0.001 | 0.58 (± 0.39) | P <0.001 | P = 0.041 | Baseline T0 (MDG vs PDG): P = 0.283 |
| nCBs | 12.68 (± 1.70) | ||||||||
| T1 | CBs | 14.96 (± 1.10) | 0.49 (± 0.53) | P = 0.011 | |||||
| nCBs | 15.45 (± 1.27) | ||||||||
| Group | Timing | Side | D2 (mm) | Difference | P value | Mean difference of the difference | P value | P value | P value |
| MD-SG | T0 | CBs | 13.98 (± 1.43) | 1.80 (± 1.45) | P <0.001 | 0.81 (± 0.78) | P <0.001 | P = 0.772 | Baseline T0 (MDG vs PDG): P = 0.330 |
| nCBs | 15.78 (± 1.34) | ||||||||
| T1 | CBs | 16.72 (± 0.94) | 0.99 (± 1.13) | P <0.001 | |||||
| nCBs | 17.71 (± 1.26) | ||||||||
| PD-SG | T0 | CBs | 14.59 (± 1.24) | 2.18 (± 0.96) | P <0.001 | 0.83 (± 0.73) | P <0.001 | P = 0.772 | Baseline T0 (MDG vs PDG): P = 0.330 |
| nCBs | 16.76 (± 1.33) | ||||||||
| T1 | CBs | 17.57 (± 1.17) | 1.35 (± 1.29) | P <0.001 | |||||
| nCBs | 18.92 (± 1.33) |
Stay updated, free dental videos. Join our Telegram channel
Leave a Reply
VIDEdental - Online dental courses
