Authors: Huixin Xue, Yezi Qi, Xinxin Ni, Yan Feng, Jun Lin
Categories: Scientific Research Report, Thin palatal bone, Three-dimensional evaluation, Personalized treatment
Source: International Dental Journal
Authors: Huixin Xue, Yezi Qi, Xinxin Ni, Yan Feng, Jun Lin
Maxillary transverse deficiency (MTD) affects approximately 10% of adults, but conventional expansion treatments are often ineffective and risky in mature patients due to suture fusion. Although microimplant-assisted rapid palatal expansion (MARPE) improves outcomes, it relies on sufficient bone thickness, posing a challenge for individuals with thin palatal bone. This study evaluates a digitally personalized MARPE approach, guided by cone-beam computed tomography-based bone mapping, for adults with MTD and thin bone (<2.5 mm).
A retrospective analysis was conducted on 18 patients. Custom MARPE appliances were designed using computer-aided design/computer-aided manufacturing technology, with microimplants placed strategically using surgical guides. Skeletal, dental, and zygomatic changes were assessed via pre- and postexpansion cone-beam computed tomography and cephalograms.
All patients were successfully treated without implant failure. Significant skeletal expansion was anterior nasal spine width increased by 5.19 mm, posterior nasal spine by 4.49 mm, and maxillary basal width by 4.05 mm (all P < .01). Dental side effects were minimal. Significant three-dimensional remodelling of the zygomaticomaxillary complex was observed, along with sagittal improvements in SNA and ANB angles (0.57° and 1.54°, respectively) and increased anterior facial height (2.00 mm, P < .01). All measurements showed high reliability (intraclass correlation coefficient >0.90).
Personalized MARPE demonstrated promising results for MTD in patients with thin palatal bone, providing significant skeletal expansion with minimal adverse dental effects.
This study provides evidence that a digitally personalized MARPE protocol, guided by CBCT, is a feasible and well-tolerated treatment approach for adults with maxillary transverse deficiency and critically thin palatal bone. It provides significant skeletal expansion with minimal dental side effects, offering a viable solution for a patient population that has limited and high-risk treatment options.
Maxillary transverse deficiency (MTD) is a common craniomaxillofacial deformity in orthodontics, with a prevalence of approximately 10% among adults.^1^^,^^2^ It presents as narrow dental arches, unilateral or bilateral posterior crossbite, dental crowding, and high palatal vault, impairing occlusal function. It may also adversely affect facial aesthetics, leading to issues such as narrow nasal root, deepened nasolabial folds, and underdevelopment of the zygomatic and paranasal regions.^3^
Rapid maxillary expansion (RME) is a classical treatment that applies force to the midpalatal suture to widen the maxilla. However, with age, suture fusion increases resistance to expansion, reducing RME efficiency in adults and potentially causing adverse effects such as bone loss and gingival recession.^2^^,^^4^ Surgically assisted rapid palatal expansion (SARPE) was developed to address this limitation but involves significant surgical trauma and patient discomfort.
To address these challenges, microimplant-assisted rapid palatal expansion (MARPE) was introduced, utilizing microimplants to provide enhanced skeletal anchorage.^5^ Modifications by various researchers have led to design variations, including the maxillary skeletal expander, which involves placing bands on the maxillary first molars and implanting microimplants in the posterior region of the hard palate corresponding to the first molars and on both sides of the midpalatal suture.6, 7, 8 MARPE delivers direct force to the maxilla, effectively expanding the midpalatal suture, zygomaticomaxillary suture, and pterygomaxillary junction, and has proven successful in treating adult MTD patients.^9^^,^^10^
A critical factor influencing MARPE success is primary microimplant stability, which heavily depends on palatal bone thickness and density.^11^ Systematic reviews by Winsauer et al^12^ indicate that microimplants with a diameter of 2 mm and length of 10 to 14 mm require at least 5 mm of bone support to ensure retention. Gahleitner et al^13^ suggested a minimum palatal bone thickness of 4 mm. Ichinohe et al^14^ reported higher success rates of midpalatal suture opening when microimplants engage more than 4.5 mm of palatal bone, with cortical thickness exceeding 1.5 mm being critical for stability. In cases of insufficient bone volume, there is an increased risk of implant failure, nasal mucosa perforation, tooth-driven expansion, and other complications.^15^^,^^16^
Common MARPE microimplant placement sites are located 3 to 6 mm lateral to the midpalatal suture in the first molar region, while patients with thin bone may require anteriorly adjusted positioning.^8^^,^^17^ Several cone-beam computed tomography (CBCT) studies have assessed palatal bone thickness to investigate optimal bony anchorage sites for MARPE.^13^^,^^18^^,^^19^ However, it is important to note that the findings of this study may not be universally applicable to patients with thin palatal bone. Therefore, personalized design enables the determination of appropriate microimplant placement locations in these patients, aiming to achieve treatment efficacy and safety. This includes a comprehensive assessment of bone thickness distribution in the palate and paramedian regions, and exploring alternative bony anchorage sites based on individual anatomy.
Advances in digital technology – including 3D imaging, computer-aided design, and computer-aided manufacturing – have greatly enhanced MARPE treatment planning.^11^^,^^20^ In patients with thin palatal bone, digital tools enable precise assessment of bone morphology, optimized microimplant positioning, and fabrication of customized expanders, thereby improving stability and reducing the risk of failure.
This study employs a digital workflow to design and fabricate patient-specific MARPE appliances for individuals with MTD and thin palatal bone. Using CBCT and lateral cephalograms, we will evaluate morphological changes in the palate, maxilla, and anchor teeth before and after personalized MARPE treatment, with particular focus on three-dimensional changes in the zygomaticomaxillary complex. The aim of this research is to assess the clinical effectiveness of digitally planned, bone thickness-guided MARPE therapy in patients with MTD and thin palatal bone.
This was a single-arm, single-centre, retrospective cohort study. The data were collected from the medical records of patients who had been treated according to a standardized personalized MARPE protocol. All treatment procedures, including appliance fabrication and the expansion regimen, were performed as part of routine clinical care prior to the conception of this study.
This retrospective study included 18 patients (7 males and 11 females; mean 22.56 ± 4.90 years) who underwent treatment at the Department of Orthodontics, the First Affiliated Hospital of Zhejiang University School of Medicine, between June 2024 and December 2024. The study protocol was approved by the hospital’s Ethics Committee (NO: 2025-0328).
Inclusion criteria were as (1) diagnosis of MTD, defined as a maxillary–mandibular basal bone width discrepancy of less than 5 mm or mandibular basal bone wider than the maxillary basal bone; (2) critically thin palatal bone, defined as a thickness of <2.5 mm^15^ at the paramedian site (3 mm lateral to the midpalatal line,^21^ passing through the maxillary first molar) in the mid-to-posterior palate, as measured on CBCT according to the method of Patni et al,^22^ rendering the site unsuitable for conventional microimplant placement; (3) cervical vertebral maturation stage at stage 5 or 6, indicating postpeak growth; (4) absence of periodontal disease; and (5) no history of prior orthodontic treatment.
Exclusion criteria (1) history of craniofacial trauma or presence of craniofacial syndromes; (2) psychiatric disorders, systemic diseases, or any other conditions contraindicating orthodontic treatment; (3) history of orthognathic surgery or orthodontic/orthopaedic treatment; and (4) concurrent use of facemask protraction therapy.
Sample size calculation was performed using PASS 15.0 software (NCSS), with a two-sided alpha (α) of 0.05, and statistical power of 80% for a paired test. Based on the primary outcome of midpalatal suture expansion, and referencing previous studies by Sharma et al,^23^ Ventura et al,^24^ and Lázaro-Abdulkarim et al,^25^ a mean difference of 2.00 mm and a standard deviation of 1.50 mm for the change in midpalatal suture width were selected. This calculation indicated that a minimum of 6 participants was required.
Data CBCT scans and alginate impressions of the maxillary arch were acquired for each patient. The plaster models were poured and then scanned using a BLZ scanner (BLZ Tech) to generate digital models.
Data fusion and virtual The CBCT data and digital dental models were merged using 3Shape Appliance Designer software (3Shape). The palatal bone thickness was evaluated using the software’s measuring tools. Areas with insufficient bone thickness (<2.5 mm) in the median and paramedian sites (3 mm lateral to the midpalatal line, passing through the maxillary first molar) of the mid-to-posterior palate were identified.^15^^,^^21^^,^^22^ Subsequently, sites with adequate bone volume were located in the palatal region (usually close to the alveolar bone) to determine the insertion points and trajectories for the posterior microimplants, carefully avoiding critical structures such as tooth roots. The anterior microimplant sites were planned within the anterior median and paramedian regions.
Appliance Based on the predetermined microimplant positions, the MARPE screw expander was digitally designed. Mucosal-adapting abutments that matched both the microimplants and the individual’s palatal contours were created. Molar bands and connecting arms extending from the abutments to the bands were also designed digitally.
Virtual simulation and The integrated CBCT data was used to simulate appliance placement and microimplant insertion. This critical step verified positioning accuracy, ensured bicortical engagement of the microimplants, and prevented damage to adjacent structures such as the nasal cavity, maxillary sinus, and incisive canal.
Manufacturing: The final appliance was manufactured using computer-aided design/computer-aided manufacturing technology and 3D printing, ensuring physical conformity to the digital plan.2. MARPE treatment process (clinical application):
Appliance delivery and guided The custom-made expander was cemented onto the maxillary first molars. Under local anaesthesia (4% articaine), the osteotomy sites were prepared using a pilot drill under saline irrigation. The four microimplants (BMK; 2 mm diameter; 12- or 14-mm length) were then inserted through the guide channels of the expander (Figure 1), precisely transferring the virtual surgical plan to the clinical procedure.Fig. 1Personalized MARPE on the plaster model and in the mouth.Fig 1
Standardized expansion After a 1-week consolidation period for initial microimplant stability and soft tissue healing,26, 27, 28 a standardized activation protocol was four activations per day (1/4 of a turn, 90° each) during the first week, followed by two activations per day thereafter. This regimen was consistently applied, with patients maintaining activation records and progress monitored by the same clinician.
Lateral cephalograms and CBCT scans were acquired on the day of appliance placement and after expansion completion. The active expansion phase was concluded upon reaching a predetermined clinical endpoint, which was defined by the establishment of a positive transverse relationship with mild overcorrection. This was achieved when the palatal cusps of the maxillary molars occluded on the buccal cusps of the mandibular molars, ensuring the correction of the posterior crossbite while accounting for potential relapse.^29^ Postexpansion images were acquired on the same day this endpoint was reached, prior to the commencement of the retention period. All images were obtained using consistent
All CBCT images were acquired using the same device (NewTom VGi) with the following FSV: 110 kV/1.61 mA, SSV: 110 kV/1.47 mA, exposure 3.6 s, axial slice 0.300 mm, axial slice 0.300 mm, and field of full. During imaging, patients were seated in a natural head position, gazing forward, holding handles with both hands, breathing calmly, maintaining centric occlusion, and with the tongue tip placed against the palate at the end of swallowing.
Lateral cephalograms were acquired using the same 2D imaging system (Planmeca ProMax 2D, Planmeca). Patients stood in natural head position during imaging, with eyes looking forward, chin rest stabilizing head position, and centric occlusion maintained.
To ensure objective assessment, all measurements were conducted by a single investigator who was blinded to the patient identity and treatment time point. This was achieved by anonymizing and randomizing all CBCT and cephalometric images prior to analysis. All measurements were performed twice at 1-week intervals, and the mean values were used for subsequent statistical analysis.
CBCT data in DICOM format were imported into Dolphin Imaging 11.9 software (Dolphin Imaging) for 3D reconstruction and head orientation calibration. Three reference planes were the midsagittal plane (MSP), axial palatal plane (APP), and molar coronal plane (MCP). The MSP passed through anterior nasal spine (ANS), PNS, and N points; the APP was perpendicular to the MSP and passed through ANS and PNS; the MCP was perpendicular to both MSP and APP and passed through PNS^30^^,^^31^ (Figure 2). Pre- and post-treatment reconstructed skulls were aligned based on the anatomical structures of the anterior cranial base. Pre- and post-treatment models were superimposed based on anterior cranial base structures, and the coordinate system was transferred with high precision (error <0.05 mm) using Point N and bilateral Porion points as references (Figure 3).Fig. 2Head position calibration. (A) Establish the midsagittal plane passing through the ANS and PNS points in the top view; (B) establish the midsagittal plane passing through the ANS and N points in the front view; (C) establish the transverse plane passing through the ANS and PNS points in the side view; (D) oblique lateral view after completion of head position calibration.Fig 2Fig. 3Superimposition of pre- and post-treatment models.Fig 3
Four planes were defined for (1)APP: A plane passing through ANS-PNS and perpendicular to the MSP.(2)Molar coronal plane (MCP): A coronal plane passing through the furcation of the right maxillary first molar and parallel to the MCP.(3)Zygomatic axial section (ZAS): A plane passing through the vertical midpoint of the zygomaticotemporal sutures and the vertical midpoint of the temporal articular tubercles (Figure 4).Fig. 4Zygomatic axial section and zygomatic coronal plane on CBCT.Fig 4Fig. 5Measurement of maxillary expansion efficiency. (A) Measurement of expansion efficiency on the axial palatal plane; (B) measurement of expansion efficiency on the molar coronal plane.Fig 5Fig. 6Measurement of dental effects.Fig 6Fig. 7Measurement of the zygomaticomaxillary complex. (A) Measurement of distance on the zygomatic axial section; (B) measurement of angle on the zygomatic axial section; (C) measurement of distance on the zygomatic coronal plane; (D) measurement of angle on the zygomatic coronal plane.Fig 7(4)Zygomatic coronal plane (ZCP): A coronal plane passing through the lowest point of the zygomaticomaxillary arch and the midpoint of the frontozygomatic suture (Figure 4).Multiplanar measurements were performed to evaluate maxillary expansion, dental effects, and zygomaticomaxillary complex changes (Table 1). Sagittal and vertical changes in the zygomaticomaxillary complex were assessed on lateral cephalograms (Table 2).Table 1Parameters evaluated on CBCT.Table 1Measurement categoryAbbreviationFull nameDefinitionExpansion efficiencyANSWAnterior nasal spine widthDistance between left and right ANS points on APPPNSWPosterior nasal spine widthDistance between left and right PNS points on APPSOASuture opening angleMedial angle between left and right ANS-PNS lines on APP; (+) posterior intersection, (–) anterior intersection^9^ (Figure 5A)MBWMaxillary basal widthDistance between the two points where the line connecting the buccal root apexes of the maxillary first molars intersects the buccal alveolar cortical bone on MCPNBWNasal bone widthDistance between widest points of bilateral nasal cavities on MCPU6DMolar distanceDistance between central fossae of bilateral maxillary first molars on MCP^36^ (Figure 5B)Dental effectsU6HMolar vertical heightVertical distance from palatal cusp of maxillary first molars to APP on MCPU6AMolar inclination angleMedial angle between the line connecting the central fossa to the furcation of maxillary first molars and APP on MCPABHAlveolar bone heightVertical distance from buccal alveolar crest of maxillary first molars to APP on MCPABTAlveolar bone thicknessDistance between buccal and palatal cortical bone at the level of maxillary first molar furcation on MCP (Figure 6)Zygomaticomaxillary complex changesAMDAnterior maxillary distanceDistance between the most anterior points of bilateral maxillae on ZASPZDPosterior zygomatic distanceDistance between the most lateral points of the bilateral zygomaticotemporal sutures on ZASPTDPosterior temporal distanceDistance between the apical points of bilateral temporal articular tubercles on ZASZTAZygomaticotemporal angleAngle formed by lines connecting the most lateral point of the zygomaticotemporal suture to the most anterior maxillary point and the apical points of temporal articular tubercle on ZASZPAAngle of the zygomatic process of the temporal boneAngle formed by a line connecting the most posterior point of bilateral temporal articular tubercles, and a line connecting the most posterior point on the temporal articular tubercle to the most external point on the zygomaticotemporal suture on ZAS^41^ (Figure 7A,B)UZDUpper zygomatic distanceDistance between most lateral frontozygomatic suture points on ZCPLZDLower zygomatic distanceDistance between most lateral zygomaticomaxillary suture points on ZCPFETAFrontoethmoidal angleOuter superior angle formed by the line connecting the lowest point of crista galli and the outermost points of bilateral frontozygomatic suture points on ZCPFZAFrontozygomatic angleAngle formed by the line connecting the outermost point of frontozygomatic suture with the lowest point of crista galli and the outermost point of zygomaticomaxillary suture on ZCPZMAZygomaticomaxillary angleAngle formed by the outermost point of zygomaticomaxillary suture, the outermost point of the frontozygomatic suture, and the point where the cortical bone of the maxillary sinus floor merges with the nasal floor on ZCPMIAMaxillary inclination angleMedial angle between the line connecting the most lateral point of the maxilla to the fusion point of the nasal floor and maxillary sinus floor, and the projection of the MSP on the MCP^37^ (Figure 7C,D)Table 2Cephalometric measurements on lateral cephalograms.Table 2MeasurementDefinitionSignificanceSagittal measurementsSNA (°)Angle between subspinale (Point A) and the anterior cranial base (SN plane)Maxillary prominenceSNB (°)Angle between supramental (Point B) and the SN planeMandibular prominenceANB (°)Angle formed by Point A, nasion (Point N), and Point BRelative jaw positionA-Np (mm)Distance from Point A to the perpendicular line from Point NMaxillary prominenceB-Np (mm)Distance from Point B to the perpendicular line from Point NMandibular prominenceWits (mm)Distance between the perpendicular projections of Points A and B onto the occlusal planeJaw relationshipA-Ptm (mm)Distance between the perpendicular projections of the pterygomaxillary fissure (Ptm) and Point A onto the Frankfort horizontal (FH) planeMaxillary lengthANS-Ptm (mm)Distance between the perpendicular projections of Ptm and anterior nasal spine (ANS) onto the FH planeMaxillary lengthVertical measurementsPP-SN (°)Angle between the palatal plane (PP) and the SN planeMaxillary plane inclinationMP-SN (°)Angle between the mandibular plane (MP) and the SN planeMP inclinationOP-SN (°)Angle between the occlusal plane and the SN planeOcclusal plane inclinationN-ANS (mm)Distance from Point N to ANSUpper anterior face heightN-Me (mm)Distance from Point N to Menton (Me)Total anterior face heightANS-Me (mm)Distance from ANS to Menton (Me)Lower anterior face heightANS-Me/N-Me (%)Ratio of lower anterior face height to total anterior face heightLower anterior facial height ratioN-ANS/N-Me (%)Ratio of upper anterior face height to total anterior face heightUpper anterior facial height ratio
Statistical analysis was conducted using SPSS 27.0 (IBM Corporation). Intraclass correlation coefficient (ICC) analysis was conducted on duplicate measurements to assess reliability, with an ICC value greater than 0.75 indicating good reliability. The mean values of duplicate measurements were used for subsequent statistical analysis. Normality was evaluated using the Shapiro–Wilk test. Paired t tests (normal distribution) or Wilcoxon signed-rank tests (non-normal distribution) were used for pre- and post-treatment comparisons. A P value of less than .05 was considered statistically significant.
A total of 18 patients with thin palatal bone were included in this study. Descriptive data for the population are given in Table 3. Successful midpalatal suture expansion was achieved in all patients, with no instances of microimplant loosening or failure. ICC analysis for duplicate measurements demonstrated high reliability, with all measurement items exceeding 0.90.Table 3Baseline characteristics of study population.Table 3VariableValueAge (y)22.56 ± 4.90Gender Male7 Female11Amount of transverse deficiency (mm)2.94 ± 3.76Palatal bone thickness (mm)1.78 ± 0.45
As shown in Table 4, following personalized MARPE treatment, patients with thin palatal bone showed significant expansion in ANS width, posterior nasal spine width, maxillary basal width, nasal bone width, and intermolar distance (P < .01). The suture opening angle also increased slightly (P < .01).Table 4Measurement results of maxillary expansion efficacy.Table 4Measurement itemBefore treatmentAfter treatmentChangeP valueANSW (mm)0.005.19 ± 2.595.19 ± 2.59<.001⁎⁎⁎PNSW (mm)0.004.49 ± 1.814.49 ± 1.81<.001⁎⁎⁎SOA (°)0.000.60 ± 0.690.60 ± 0.69.002⁎⁎MBW (mm)62.77 ± 5.3666.82 ± 5.074.05 ± 1.87<.001⁎⁎⁎NBW (mm)34.52 ± 3.2838.26 ± 3.373.74 ± 2.29<.001⁎⁎⁎U6D (mm)49.29 ± 3.6855.25 ± 3.335.97 ± 2.47<.001⁎⁎⁎⁎⁎P <.01.⁎⁎⁎P <.001.
As presented in Table 5, the vertical height changes of the left and right molars were not significant (P > .05). The right molar inclination angle increased (P < .01), while the change in left molar inclination angle was not statistically significant (P > .05). Changes in buccal alveolar bone height on the left and right sides were negligible (P > .05). Right alveolar bone thickness decreased by 0.13 ± 0.21 mm (P < .05), while the change in left alveolar bone thickness was not statistically significant (0.08 ± 0.30 mm, P > .05).Table 5Measurement results of dental effects.Table 5Measurement itemBefore treatmentAfter treatmentChangeP valueL-U6H (mm)23.09 ± 2.1623.13 ± 2.120.04 ± 0.65.809R-U6H (mm)23.10 ± 2.2423.15 ± 2.070.05 ± 0.80.784L-U6A (°)96.50 ± 6.4397.13 ± 5.740.64 ± 1.99.193R-U6A (°)98.18 ± 4.7299.79 ± 5.051.60 ± 2.88.030L-ABH (mm)13.35 ± 2.4113.38 ± 2.420.03 ± 0.26.679R-ABH (mm)13.16 ± 2.3513.15 ± 2.29–0.01 ± 0.32.912L-ABT (mm)13.80 ± 1.6913.71 ± 1.76–0.08 ± 0.30.270R-ABT (mm)13.75 ± 1.6013.62 ± 1.66–0.13 ± 0.21.018L-, left side; R-, right side.⁎P < .05.
As shown in Table 6, after personalized MARPE expansion, anterior maxillary distance, posterior zygomatic distance, posterior temporal distance, upper zygomatic distance, lower zygomatic distance, zygomatic process angles, frontozygomatic angles and maxillary inclination angles increased significantly (P < .05). The right zygomaticotemporal angle decreased by 0.70° ± 1.17° (P < .05), while the left zygomaticotemporal angle decrease was not statistically significant (1.45° ± 4.47°, P > .05). The frontoethmoidal angle decreased (P > .05), while the left and right zygomaticomaxillary angles showed little increase (P > .05), both showing no statistical significance.Table 6Axial and coronal changes in the zygomaticomaxillary complex.Table 6Measurement itemBefore treatmentAfter treatmentChangeP valueAxialAMD (mm)16.35 ± 1.9418.84 ± 2.002.49 ± 1.07<.001⁎⁎⁎PZD (mm)119.14 ± 7.14121.50 ± 7.042.36 ± 1.32<.001⁎⁎⁎PTD (mm)122.85 ± 5.47123.12 ± 5.610.27 ± 0.34.003⁎⁎L-ZTA (°)131.13 ± 7.01129.68 ± 4.26–1.45 ± 4.47.186R-ZTA (°)131.10 ± 2.82130.40 ± 3.22–0.70 ± 1.17.020*L-ZPA (°)87.25 ± 3.9789.10 ± 4.131.85 ± 1.81<.001⁎⁎⁎R-ZPA (°)86.24 ± 3.4387.79 ± 3.631.54 ± 1.28.001⁎⁎CoronalUZD (mm)101.77 ± 5.33102.13 ± 5.500.36 ± 0.37.001⁎⁎LZD (mm)92.10 ± 5.6595.72 ± 5.403.62 ± 1.35<.001⁎⁎⁎FETA (°)164.70 ± 4.48164.35 ± 4.84–0.35 ± 1.46.323L-FZA (°)76.07 ± 2.6978.05 ± 3.041.98 ± 1.24<.001⁎⁎⁎R-FZA (°)76.51 ± 3.4078.29 ± 4.091.77 ± 1.49<.001⁎⁎⁎L-ZMA (°)103.41 ± 5.63103.44 ± 6.000.03 ± 1.91.947R-ZMA (°)102.98 ± 5.34103.00 ± 5.560.02 ± 1.19.935L-MIA (°)97.82 ± 3.7899.72 ± 4.891.90 ± 2.08.001⁎⁎R-MIA (°)97.11 ± 5.1299.04 ± 5.471.93 ± 1.51<.001⁎⁎⁎L-, left side; R-, right side.⁎P < .05.⁎⁎P < .01.⁎⁎⁎P < .001.
Cephalometric analysis results are presented in Table 7. In the sagittal dimension, the SNA angle, ANB angle, A-Np, A-Ptm, and ANS-Ptm increased (P < .05), while the SNB angle and B-Np decreased (P < .05). Wits appraisal increased by 1.06 ± 2.90 mm (P > .05), but did not reach statistical significance.Table 7Sagittal and vertical cephalometric changes in the zygomaticomaxillary complex.Table 7Measurement itemBefore treatmentAfter treatmentChangeP valueSagittalSNA (°)82.49 ± 3.3483.07 ± 3.330.57 ± 0.92.017SNB (°)81.36 ± 5.0380.43 ± 4.69–0.93 ± 1.23.005⁎⁎ANB (°)1.13 ± 3.522.67 ± 2.861.54 ± 1.07<.001⁎⁎⁎Wits (mm)–4.63 ± 6.22–3.57 ± 4.751.06 ± 2.90.139A-Np (mm)–1.84 ± 2.34–0.24 ± 2.291.59 ± 1.07<.001⁎⁎⁎B-Np (mm)–3.68 ± 6.55–4.44 ± 5.93–0.77 ± 1.19.014A-Ptm (mm)43.22 ± 3.9344.78 ± 4.111.56 ± 1.39<.001⁎⁎⁎ANS-Ptm (mm)45.12 ± 4.0646.65 ± 4.081.53 ± 1.60.001⁎⁎VerticalPP-SN (°)6.52 ± 3.077.21 ± 3.210.69 ± 1.50.068MP-SN (°)36.90 ± 7.0038.27 ± 7.021.37 ± 1.54.001⁎⁎OP-SN (°)16.09 ± 4.9916.96 ± 4.830.87 ± 2.42.146N-ANS (mm)50.47 ± 5.3051.04 ± 4.670.57 ± 1.25.070N-Me (mm)117.78 ± 9.77119.79 ± 10.012.00 ± 2.45.003⁎⁎ANS-Me (mm)68.16 ± 5.4369.55 ± 5.821.39 ± 2.35.023*ANS-Me/N-Me (%)57.91 ± 1.8658.08 ± 1.670.17 ± 1.17.556N-ANS/N-Me (%)42.81 ± 1.7842.62 ± 1.700.20 ± 1.02.425⁎P < .05.⁎⁎P < .01.⁎⁎⁎P < .001.
In the vertical dimension, mandibular plane-SN angle, N-Me, and ANS-Me increased significantly (P < .05), while PP-SN and occlusal plane-SN angles increased by 0.69° ± 1.50° (P > .05) and 0.87° ± 2.42° (P > .05), respectively, showing no statistical significance. Changes in the upper anterior facial height ratio and lower anterior facial height ratio also showed no statistical significance (P > .05).
For a concise overview, the key findings regarding skeletal, dental, and zygomatic changes are summarized in Table 8.Table 8Summary of key skeletal, dental, and zygomatic changes following personalized MARPE treatment.Table 8Measurement categoryParameter (abbreviation)Change (mean ± SD)Clinical and biological significanceSkeletal expansionAnterior nasal spine width (ANSW)5.19 ± 2.59 mmPrimary indicator of successful anterior midpalatal suture opening.Posterior nasal spine width (PNSW)4.49 ± 1.81 mmPrimary indicator of successful posterior midpalatal suture opening.Suture opening angle (SOA)0.60° ± 0.69°Reflects the characteristic V-shaped pattern of maxillary expansion.Dental effectsMolar inclination angle (U6A)0.64° ± 1.99° (L)1.60° ± 2.88° (R)Indicates minimal buccal tipping, highlighting the skeletal-dental nature of expansion.Zygomaticomaxillary complex changesAnterior maxillary distance (AMD)2.49 ± 1.07 mmLateral expansion of the maxilla, the zygomatic bone, and the whole zygomatic arch.Posterior zygomatic distance (PZD)2.36 ± 1.32 mmAngle of the zygomatic process of the temporal bone (ZPA)1.85° ± 1.81° (L)1.54 ± 1.28° (R)Suggests a rotation of the zygomatic complex relative to the temporal bone.Frontozygomatic angle (FZA)1.98° ± 1.24° (L)1.77° ± 1.49° (R)Outward rotation of the zygomatic complex in the coronal plane.ANB angle1.54° ± 1.07°Suggests a mild forward displacement of the maxilla relative to the mandible.Anterior facial height (N-Me)2.00 ± 2.45 mmIndicates a slight increase in vertical dimension.
To visually assess the distribution and variability of the primary skeletal expansion outcomes, Figure S1 presents boxplots of the individual changes in ANSW and PNSW. The plots demonstrate a consistent treatment effect across the cohort. While potential outliers were observed in both the ANSW and PNSW data, it is noteworthy that they originated from the same patient.
A sensitivity analysis was therefore performed to evaluate the robustness of our primary findings. As detailed in Table S1, the results were compared between the full dataset (n = 18) and the dataset with the outlier excluded (n = 17). The analysis confirmed that the exclusion of this outlier did not substantially alter the mean expansion values or their statistical significance (P < .001 for both outcomes). This confirms that the significant skeletal expansion effect observed is robust and not driven by any single individual’s data.
During the active expansion phase, all patients reported an expected sensation of tension across the palate. This subjective sensation corresponded radiographically to the progressive opening of the midpalatal suture. A number of patients experienced transient palatal pain, which was effectively managed with analgesics (eg, ibuprofen or Saridon) as needed. Furthermore, transient discomfort in the regions of the nasal base and zygoma was reported by a minority of patients, a finding that is consistent with the biomechanical involvement of the zygomaticomaxillary and other circummaxillary sutures. Notably, a subset of patients spontaneously reported an improvement in nasal breathing following expansion, describing a subjective sensation of increased nasal airflow. All reported symptoms were self-limiting and resolved without further clinical intervention following the cessation of active expansion.
With the continuous advancement of digital technology, an increasing number of studies have explored the application of imaging techniques to optimize the selection and positioning of microimplants. CBCT has emerged as a valuable tool for precise microimplant placement. Several scholars have suggested that CBCT imaging enables detailed assessment of palatal bone anatomy, thereby facilitating optimal microimplant positioning.^11^^,^^15^^,^^20^ Beyond imaging evaluation, digital technologies – particularly when integrated with 3D printing – allow the fabrication of patient-specific MARPE appliances, offering improved expansion efficiency and stability. In this study, posterior microimplants were designed and anchored in the part of the midpalatal region that is close to the alveolar bone for patients with thin palatal bones exhibiting insufficient thickness in the posterior midline areas. Utilizing digital workflows for precise planning and device customization, we manufactured personalized MARPE appliances and evaluated their clinical efficiency using CBCT and lateral cephalograms.
In this study, a total of 18 patients with thin palatal bone were included, all of whom achieved successful midpalatal suture expansion without microimplant loosening or failure. These preliminary results indicate that palatally anchored patient-specific MARPE is a promising and clinically applicable approach, which demonstrated an acceptable safety profile within the constraints of this study’s cohort.
The anterior and posterior midpalatal suture expansion widths were 5.19 ± 2.59 mm and 4.49 ± 1.81 mm, respectively, with the posterior nasal spine expanding to approximately 85.93% of the anterior expansion. This near-parallel expansion pattern is consistent with the findings of previous studies, which generally report greater anterior than posterior expansion, with posterior expansion typically ranging between 70% and 90% of the anterior expansion.^32^^,^^33^ The suture opening angle was 0.60° ± 0.69°, aligning with findings by Colak et al^30^ in their study on maxillary skeletal expander, which reported a mean expansion angle of 0.57° (range: –0.8° to 1.3°) in 50 patients. Consequently, even in subjects with thin palatal bone, personalized MARPE achieved effective expansion while maintaining the characteristic anterior-wider-than-posterior expansion pattern.
A substantial augmentation in basal bone width was also observed. Maxillary basal width increased by 4.05 ± 1.87 mm, and intermolar width increased by 5.97 ± 2.47 mm, yielding a bone-to-dental expansion ratio of 67.84%. This efficiency exceeds that of tooth-borne expanders and is consistent with other MARPE studies. Weissheimer et al^34^ reported that RME resulted in skeletal contributions of 54.7% anteriorly and 39.2% posteriorly relative to dental expansion. Similarly, Lin et al^6^ discovered that MARPE yielded a skeletal expansion ratio ranging from 57.5% to 77.0%. However, this value is considerably larger than the 35.6% mean skeletal component of expansion reported in a previous meta-analysis.^35^ The presence of variations among studies may be attributable to differences in appliance design, microimplant position, patient age, and evaluation criteria.
Furthermore, the maxilla exhibited an inverted-V expansion pattern in the coronal plane, consistent with previous reports and resonates with the findings presented by Lázaro-Abdulkarim et al in their systematic review.^25^^,^^36^ The nasal bone exhibited an increase in width of 3.74 ± 2.29 mm, which is slightly less than the increase observed in the basal and intermolar regions. This pattern may be related to rotational displacement of the zygomaticomaxillary complex, which will be discussed in later sections.^37^ Studies indicate that conventional tooth-borne expanders often result in a fan-shaped coronal expansion, whereas MARPE facilitates more parallel expansion, likely due to enhanced skeletal anchorage and patient-specific design.^6^^,^^38^
Park et al^38^ reported that MARPE not only effectively opens the midpalatal suture but may also induce buccal tipping of molars, bending of palatal and alveolar bone, and separation of adjacent sutures. For instance, the application of MARPE treatment in patients aged 16 to 26 years resulted in an average increase of 5.8° in buccal inclination of the maxillary first molars. In the present study, the right molar inclination exhibited a significant increase of 1.86° ± 3.11° (P < .05), while the left side demonstrated a change that did not reach statistical significance, with a value of 0.64° ± 1.99° (P > .05). As a hybrid tooth- and bone-borne appliance, MARPE transmits forces to both skeletal and dental structures. In adults, with greater skeletal resistance, a portion of the expansion force may still act on the anchored teeth, leading to dental tipping and alveolar bending. Moreover, the mechanical effects of MARPE are not limited to transverse expansion but may also induce rotational changes in the maxilla, thereby further influencing tooth position and alveolar morphology.
Changes in molar inclination may be accompanied by vertical displacement, such as extrusion of the palatal cusps. Qi et al^36^ reported a 0.86 mm increase in vertical height of the palatal cusps of the maxillary first molars. In contrast, the present study did not identify significant vertical changes in molar position. This discrepancy may be attributable to variations in patient characteristics, expansion protocol, and force system design. It is noteworthy that the MARPE apparatus utilized by Qi et al did not incorporate digital planning, a factor which may have influenced the precision and homogeneity of expansion. In conclusion, the impact of MARPE on molar vertical position appears to be negligible.
Changes in molar inclination may also affect periodontal structures, particularly the alveolar bone. Sperl observed that tooth-borne RME in adolescents frequently led to periodontal changes at the buccal bone plate of the first molars, with 47.7% exhibiting bone dehiscence and 15.6% showing bone fenestration.^39^ In the present study, changes in buccal alveolar bone height were not statistically significant. A slight reduction in alveolar bone thickness was observed on the right side (0.13 ± 0.21 mm, P < .05), while the change on the left (0.08 ± 0.30 mm, P > .05) was not significant.
The findings indicate that personalized MARPE in patients with thin palatal bone elicits predominantly mild dental and periodontal responses in the short term. These observations support its potential as a clinically applicable technique. Nevertheless, patients should be informed about potential changes in molar inclination and alveolar bone thickness, which may impose an additional burden on periodontal tissues. It is recommended that close monitoring is undertaken during treatment in order to minimize the potential for complications.
Conventional tooth-borne expanders typically induce minimal transverse displacement of the zygomatic complex. Baccetti et al^4^ reported zygomatic width increases of only 0.4 and 0.3 mm in early-treated and adolescent patients, respectively, following Haas expander therapy. Ong et al^40^ observed a 1.4 mm increase in zygomatic width subsequent to tooth-borne RME in adolescents. In contrast, MARPE produces more substantial zygomatic displacement due to its greater force transmission to midfacial skeletal structures.^37^ In the present study, the lower zygomatic distance exhibited an increase of 3.62 ± 1.35 mm (P < .01), while the posterior zygomatic distance demonstrated a rise of 2.36 ± 1.32 mm (P < .01). These findings suggest a substantial transverse expansion of the zygomatic complex.
Studies suggest that, during the process of MARPE, the maxillary base and zygomatic complex undergo not only transverse displacement but also rotational movement around a centre of resistance.^33^^,^^37^^,^^41^ The upper and lower zygomatic distances exhibited increases of 0.36 ± 0.37 mm (P < .01) and 3.62 ± 1.35 mm (P < .01), respectively, thereby corroborating the findings of preceding studies. Almaqrami reported a 2.95 mm increase in lower zygomatic distance,^33^ while Cantarella et al^37^ noted increases of 0.5 and 4.6 mm in the upper and lower zygomatic distances, respectively. The greater displacement in the lower zygomatic region suggests outward rotation of the zygomatic bone in the coronal plane. This rotation was further supported by significant increases in the frontozygomatic angles on both the left (1.98° ± 1.24°, P < .01) and right (1.77° ± 1.49°, P < .01) sides. In contrast, the frontoethmoidal angle decreased by 0.35° ± 1.46° (P > .05), indicating stable relationships among the ethmoid, frontal, and zygomatic bones and suggesting a centre of rotation near the frontozygomatic suture.^37^ The zygomaticomaxillary angles showed minimal change (left: 0.03° ± 1.91°; 0.02° ± 1.19°; P > .05), indicating maintained structural relationship between the zygoma and maxilla. Increases in the maxillary inclination angles (left: 1.90° ± 2.08°; 1.93° ± 1.51°; P < .01) further support rotational displacement of the maxillary base, consistent with the observed inverted-V expansion pattern.
In the axial plane, the anterior maxillary distance increased by 2.49 ± 1.07 mm (P < .01), the posterior zygomatic distance by 2.36 ± 1.32 mm (P < .01), and the posterior temporal distance by 0.27 ± 0.34 mm (P < .01). The zygomatic process angles exhibited an increase of 1.85° ± 1.81° (P < .01) and 1.54° ± 1.28° (P < .01) on the left and right sides, respectively. The zygomaticotemporal angles demonstrated a slight decrease on the right (0.70° ± 1.17°, P < .05) and left (1.45° ± 4.47°, P > .05), with the latter showing no statistical significance. These findings are generally consistent with those reported by Cantarella et al,^41^ which suggested that the zygomaticomaxillary complex pivots around a centre of rotation situated in the vicinity of the zygomatic process of the temporal bone. Therefore, the observed expansion pattern suggests that personalized MARPE may involve a rotation centre comparable to what has been described for conventional MARPE in the literature. This treatment approach addresses MTD and may also exert an influence on midfacial width.
A number of studies have previously reported on the occurrence of sagittal and vertical displacement of the maxilla following rapid expansion. da Silva Filho et al^42^ observed a downward and backward rotation of the PP subsequent to RME. A meta-analysis by Lin et al^43^ indicated that SARPE resulted in forward, downward, and clockwise rotation of the maxilla. Yılmaz et al^7^ found that microimplant-assisted expansion exhibited an augmentation in the SNA angle, accompanied by negligible vertical alterations. In the present study, the SNA angle increased by 0.57° ± 0.92° (P < .05), the SNB angle decreased by 0.93° ± 1.23° (P < .01), and the ANB angle increased by 1.54° ± 1.07° (P < .01). These findings are in alignment with those reported by Song et al,^44^ indicating a slight forward displacement of the maxilla and backward rotation of the mandible. These findings were further supported by increases in A-Np (1.59 ± 1.07 mm, P < .01) and decreases in B-Np (0.77 ± 1.19 mm, P < .05).
Sagittal and vertical changes in the zygomaticomaxillary complex may be related to its rotational movement during expansion. Song et al^44^ used a 3D coordinate system to measure the displacement of skeletal landmarks and found that the entire complex moved anteriorly and inferiorly, with greater inferior displacement posteriorly and near the midline – consistent with a rotational pattern. The observed increases in A-Ptm (1.56 ± 1.39 mm, P < .01) and ANS-Ptm (1.53 ± 1.60 mm, P < .01) may reflect lengthening of the maxilla in the sagittal plane due to rotational expansion.
Vertically, the mandibular plane-SN angle exhibited an increase of 1.37° ± 1.54° (P < .01), while N-Me and ANS-Me demonstrated increases of 2.00 ± 2.45 mm (P < .01) and 1.39 ± 2.35 mm (P < .05), respectively. Backward rotation of the mandible contributed to the decrease in SNB and increase in gonial angle and anterior facial height.^33^^,^^45^
Anterior and downward displacement of the maxilla may improve facial profiles in Class III patients, but clockwise rotation of the mandible could adversely affect high-angle cases. Notwithstanding these general trends, individual variations in anatomy, occlusion, and growth potential complicate predictions of sagittal and vertical changes. In the present study, a decline in SNA was observed in four patients, while an increase in SNB was noted in two. Wertz reported that following RME, some patients exhibited downward and backward maxillary displacement, while others showed downward and forward movement.^46^ Farfel et al^47^ found no significant sagittal, vertical, or soft tissue changes after SARPE. Almaqrami et al^33^ reported forward maxillary displacement in 47 of 78 patients and backward displacement in 31 patients. McNamara et al^48^ observed mandibular advancement following RME in early mixed dentition Class II cases, which contradicts the findings of the present study. Consequently, the identification of patients who would benefit from maxillary expansion in terms of sagittal and vertical correction remains a complex and unpredictable process.
Beyond the skeletal corrections, the observed transverse expansion of the maxilla and zygomatic complex holds significant implications for both airway function and soft tissue aesthetics. This finding is supported by the umbrella review of Ventura et al,^24^ which demonstrated that MARPE effectively modifies the upper airway by increasing nasal airflow and reducing nasal resistance. The widening of the nasal cavity and alveolar processes not only contributes to reduced airflow resistance and improved nasal breathing,^29^^,^^49^ but may also lead to a subtle increase in alar base width, potentially exerting an influence on nasal soft tissues, as demonstrated by a meta-analysis of Liu et al^50^ and a study of Mehta et al.^51^ These combined skeletal and soft tissue changes create a favourable environment for enhanced airway patency while simultaneously influencing perioral aesthetics. Future prospective studies incorporating standardized facial photographs or 3D facial scans are warranted to quantitatively assess these beneficial soft tissue changes and their correlation with airway improvements.
The present study noted some degree of asymmetric expansion between the right and left sides. This observation aligns with previous reports of asymmetric expansion patterns in MARPE therapy. A study by Elkenawy et al^31^ found that approximately 51.6% of patients (16 out of 31) exhibited statistically significant asymmetric expansion. The authors proposed that this phenomenon might be attributed to several factors, including regional variations in bone density along the midpalatal suture and circummaxillary structures, microimplant stability, inherent patient craniofacial asymmetry, and pre-existing occlusal patterns. However, it is noteworthy that these initial asymmetries appear to be effectively managed during subsequent orthodontic treatment. Research by Barton et al^52^ demonstrated that after comprehensive fixed appliance therapy, no significant long-term skeletal or dental asymmetries remained in MARPE-treated patients. This suggests that the observed initial asymmetries represent transient biomechanical responses to expansion forces rather than permanent structural changes, and can be adequately addressed through standard orthodontic finishing procedures. In addition, the observed variability in certain parameters, as reflected by standard deviations exceeding the mean values in some measurements, warrants careful consideration. This statistical pattern may be attributed to inherent biological diversity in craniofacial anatomy (eg, morphology of the zygomatic arch) and individualized patient responses to maxillary expansion.^41^ Such physiological variability can be magnified in studies with limited sample sizes, reducing the statistical power to detect more consistent effects. Future investigations incorporating larger, multicentre cohorts and standardized three-dimensional assessment protocols will be essential to better characterize the true range of anatomical variation and enhance the generalizability of these findings.
A key consideration in interpreting this study is the absence of a concurrent control group. As a single-arm investigation, our findings primarily describe the outcomes achievable with personalized MARPE in a cohort of patients with thin palatal bone. However, the added value of our personalized approach can be contextualized by comparing our results to the existing literature on conventional MARPE or alternative treatments in similar populations. For instance, studies on conventional MARPE in patients with unfavourable bone conditions have reported higher rates of failure.^16^ In contrast, our cohort demonstrated consistent skeletal expansion with the personalized guides, suggesting that the preoperative CBCT-based planning and guided surgery may enhance the predictability of the orthopaedic outcome and mitigate the risks associated with thin bone. Nonetheless, we emphasize that such cross-study comparisons are indirect and must be interpreted with caution. Future prospective, controlled studies are needed to directly validate the comparative efficacy of this personalized protocol against conventional methods.
This study has several limitations that should be considered when interpreting the results. First, as a single-arm study without a control group, the findings presented here establish a baseline for the performance of personalized MARPE in this challenging population but cannot establish its superiority or inferiority to alternative treatment modalities. Second, the single-centre design and the fact that all treatments were performed by a single operator may limit the generalizability of the findings and introduce operator-specific bias. Third, the relatively modest sample size, though sufficient for the primary outcome, reduces the statistical power for analysing all secondary parameters and subgroup effects.^53^ Fourth, although a blinding procedure (anonymization and randomization of images) was implemented for measurements, the discernible morphological change resulting from the expansion itself may have compromised complete blinding to the treatment timepoint. However, the reliance on objective quantitative measurements and the high demonstrated intraexaminer reliability are believed to mitigate the potential for significant measurement bias. Fifth, the short-term follow-up restricts our analysis to the immediate expansion phase and does not allow for assessment of long-term stability, relapse, or potential late-onset complications. Finally, despite standardized protocols, potential measurement biases inherent to CBCT landmark identification, though mitigated by a calibrated examiner, cannot be entirely eliminated. Future multicentre, randomized controlled trials with larger sample sizes and long-term follow-up are necessary to confirm these preliminary findings.
This study investigated the application and efficiency of personalized MARPE in patients with thin palatine bones. The findings confirm the potential of personalized MARPE to produce widening of the midpalatal suture, increased maxillary basal bone width, and enhanced maxillary transverse development. In addition, the observed impact of the device on molar inclination and alveolar bone was limited in this cohort. Therefore, personalized MARPE provides a promising treatment for MTD patients with thin palatal bone.
Huixin Xue: Writing – original draft, formal analysis. Yezi Qi: Resources, methodology. Xinxin Ni: Software. Yan Feng: Writing – review and editing. Jun Lin: Conceptualization.
This study was approved by the Ethics Committee of the First Affiliated Hospital, Zhejiang University School of Medicine (NO: 2025 − 0328). The study was explained, and informed consent was obtained from all of the patients. As for adolescents, informed consent was obtained from their legal guardians.
All data generated or analysed during this study are included in this published article.
This work was supported by the Zhejiang Provincial Natural Science Foundation (Grant No. LMS25H140004).
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this article.