The aim of this 2-arm parallel trial was to compare the dentoskeletal effects of the expander with differential opening (EDO) and the Hyrax expander in the mixed dentition.
Patients aged 7-11 years with maxillary dental arch constriction and Class I or Class II sagittal relationships were randomly allocated into 2 study groups. The experimental group comprised 22 patients (10 males, 12 females) with a mean age of 8.46 years treated with the EDO. The comparison group was composed of 24 patients (6 males, 18 females), mean age of 8.92 years treated with the conventional Hyrax expander. One complete turn per day for 6 days was performed for the posterior screw of the EDO and for the Hyrax expander. The anterior screw of the EDO was activated 1 complete turn per day for 10 days. The primary outcomes were the anterior opening of the midpalatal suture, changes on the interincisal diastema width, maxillary dental arch widths, arch perimeter, arch length, palatal depth, inclination of maxillary posterior teeth and on dental arch shape, and the amount of differential expansion in the anterior region compared with the posterior region of the maxillary dental arch. Computer-generated randomization was used. Allocation was concealed with sequentially, numbered, sealed, and opaque envelopes. Blinding was applicable for outcome assessment only. Occlusal radiographs of the maxilla were obtained at the end of the active expansion phase (T2). Intraoral photographs were obtained immediately pre-expansion (T1) and at T2. Digital dental models were obtained at T1 and 6 months after the active expansion period (T3). Intergroup comparisons of T1-T2 changes were performed using multiple linear regression analysis ( P < 0.05). The independent variables were both treatment and the starting forms. Bonferroni correction for multiple tests was applied.
The experimental group showed a significantly greater opening of the anterior region of the midpalatal suture, a greater increase of the interincisal diastema width, and greater increases of the intercanine distance and inter–first deciduous molar distance than the Hyrax expander. The experimental group showed a significant differential expansion between the anterior and posterior regions, whereas the Hyrax group produced a similar expansion in the canine and molar regions. Serious harm was not observed.
The EDO was capable of promoting greater orthopedic and dental changes in the anterior region of the maxilla than the conventional Hyrax expander. Similarity between the 2 expanders was observed for changes in the posterior region width, arch perimeter, arch length, palatal depth, and posterior teeth inclination.
The dentoskeletal effects of the expander with differential opening (EDO) was compared with the Hyrax expander in the mixed dentition.
The EDO promoted greater orthopedic changes in the anterior region of the maxilla.
The EDO promoted greater increases of intercanine and interdeciduous molar distances.
The EDO showed differential expansion between the anterior and posterior regions.
The Hyrax expander promoted similar expansion in the canine and molar regions.
Rapid maxillary expansion (RME) is the most common orthopedic procedure used to treat maxillary constriction and posterior crossbites. The dentoskeletal effects of RME are well documented in the orthodontic literature. Conventional RME expanders open the midpalatal suture and increase maxillary widths and the arch perimeter with a parallel-opening screw, centrally positioned in the palate. Alternatively, the fan-type expander concentrates changes in the anterior region of the dental arch with negligible changes in the molar region.
Approximately one third of patients with maxillary constriction have a greater transversal deficiency in the intercanine width than the intermolar width. In these cases, conventional RME expanders would overexpand the molar region to correct the intercanine width because the screws have a parallel opening. This undesirable effect could cause a significant decrease of the buccal alveolar bone plate thickness with an increased risk of bone dehiscences and gingival recessions. Additionally, previous studies on the long-term stability of conventional RME showed greater relapse of the intercanine distance than the interpremolar and intermolar distances.
Recently, a novel orthopedic maxillary expander was proposed aiming to promote greater expansion on the anterior than on the posterior region. The expander with differential opening (EDO) has 2 parallel-opening screws, 1 anteriorly and the other posteriorly positioned in the palate. Different amounts of activation in the anterior and posterior expansion screws determine a trapezoid-shaped opening of the appliance diverging toward anterior. A recent study analyzed the dentoskeletal effects of the EDO in patients with complete bilateral cleft lip and palate (BCLP). Patients with BCLP do not have midpalatal suture, and therefore the effects of EDO might be different in noncleft individuals. No previous studies evaluated the dentoskeletal outcomes of EDO in noncleft patients.
Specific objectives and hypotheses
The aim of this study was to compare the dentoskeletal outcomes of the EDO and the conventional Hyrax expander during the mixed dentition in noncleft individuals. The null hypothesis was that there is no difference for the dentoskeletal effects between the EDO and the Hyrax expander.
Material and methods
Trial design and any changes after trial commencement
This single-center study was a randomized clinical trial with 2-parallel arms. This randomized clinical trial followed the Consolidated Standards of Reporting Trials’ statement and guidelines, and no changes in methods after trial commencement were required.
Participants, eligibility criteria, and settings
This study was ethically analyzed before trial commencement by the Research Institutional Board of the Bauru Dental School, University of São Paulo, and was approved under protocol number 1.292.365. Additionally, the protocol of this study was registered in the ClinicalTrials.gov with the identifier NCT02810353 .
Recruitment of patients occurred in the Clinic of Orthodontics, Bauru Dental School, University of São Paulo from May to November 2015. Eligibility criteria included patients of both sexes in the mixed dentition, with ages ranging from 7 to 11 years diagnosed with maxillary constriction and Class I or Class II sagittal relationships. Exclusion criteria were the presence of cleft lip and palate, craniofacial syndromes, carious lesions, and history of previous orthodontic treatment.
Patients who met the eligibility criteria during recruitment were invited to participate in the study. Written consent terms were obtained from the patients and legal guardians. At this time, 1 researcher opened an envelope that contained a card with the name of 1 of the 2 types of expanders. Therefore, patients were randomly allocated into 1 of the 2 study groups. The treatment of the patients from both groups was performed by the same operator (ACMA).
The experimental group comprised patients treated with the EDO ( Fig 1 , A – C ). The EDO was composed of 2 11-mm prefabricated screws, 1 posteriorly positioned on the palate at the level of the first permanent molars and the other anteriorly positioned at the level of the first deciduous molars (Great Lakes Orthodontics, Tonawanda, NY). Orthodontic bands were preferentially adapted on the maxillary first permanent molars, and clasps were bonded on the maxillary deciduous canines. When the maxillary first permanent molars were partially erupted, or the distal aspect of the crown was covered by gingiva, the maxillary second deciduous molars were banded, and a wire extension was soldered on the palatal aspect of the first permanent molars. Both expander screws were concurrently activated for 6 days, with an activation protocol of half a turn in the morning and half a turn in the evening. Afterward, only the anterior screw was activated for an extra 4-day time with the same protocol. After the active expansion period, the expander was kept in the mouth as a retainer for 6 months. At the end of the retention phase, the expanders were removed, and a removable retention plate was installed.
The comparison group comprised patients who underwent RME using the conventional Hyrax expander ( Fig 1 , D – F ). The Hyrax expander was composed of a 11-mm screw centrally positioned on the palate. Similar to the experimental group, either maxillary first permanent molars or maxillary second deciduous molars were banded, and circumferential clasps were bonded on the maxillary deciduous canines. The screw was activated half a turn in the morning and half a turn in the evening for 6 days. After the active expansion period, the expander was kept in the oral cavity as a retainer for 6 months. At the end of the retention phase, the expander was removed, and a removable retention plate was installed.
Maxillary occlusal radiographs were obtained at the end of the expansion active phase (T2). The radiographic images were taken according to the biosecurity and radioprotection requirements, using the Insight occlusal radiographic film (Kodak Company, Rochester, NY) and a dental x-ray machine of 10 mA and 70 kV. The radiographic films were manually processed in a darkroom, using the temperature or time technique. Standardized frontal intraoral photographs were taken orthostatically for each patient immediately pre-expansion (T1) and at the end of the active expansion phase (T2); for example, 6 or 10 days after the appliance’s installation. The photographs were taken at a distance of 30 cm from the patients using a Canon T1i digital camera (Canon EOS Digital Rebel Inc, Tokyo, Japan), 100-mm macro lens, circular flash, f 11, shutter speed 1/125, and ISO 200. Standardized conventional dental models were obtained for each patient immediately pre-expansion (T1) and 6 months after expansion (T3). The maxillary dental arch models were digitized using a 3Shape R700 3D scanner (3Shape A/S, Copenhagen, Denmark), and the obtained 3-dimensional images were saved in an .stl file format.
Outcomes (primary and secondary) and any changes after trial commencement
The primary outcomes of this study were the dimension of anterior midpalatal suture opening ( Pr-Pr ‘); the changes in the interincisal diastema, maxillary dental arch widths ( c-c , d-d , e-e, and 6-6 ), arch perimeter and length, palatal depth and inclination of posterior teeth ( Ic , Ie, and I6 ), dental arch shape, and the amount of differential expansion in the anterior region compared with the posterior region of the maxillary dental arch. No outcome changes occurred after trial commencement.
The dimension of the anterior midpalatal suture opening was digitally measured on the maxillary occlusal radiographs using Dolphin Imaging software, version 11.0 (Dolphin Imaging and Management Solutions, Chatsworth, Calif), as shown in Figure 2 .
The width of the interincisal diastema was measured using a modification of a method proposed in a previous study. Initially, the mesiodistal width of the clinical crown of the maxillary right central incisor of each patient was manually measured on the pre-expansion conventional dental models using a caliper. Using a digital millimetric ruler, the photograph was resized in Microsoft PowerPoint 2013 (Microsoft Corporation, Redmond, Wash) according to the actual size of the measured tooth. The interincisal diastema was measured using Dolphin Imaging software, as shown in Figure 3 .
The measurements of maxillary dental arch widths, arch perimeter and length, palatal depth, and inclination of posterior teeth were performed on the pre- and postexpansion digital dental models using the OrthoAnalyzer 3D software (3Shape A/S), as shown in Figures 4-6 .
The dental arch shape was evaluated using the Geomagic Wrap 2015 software (Raindrop Geomagic Inc, Morrisville, NC) and the Microsoft Excel 2013 (Microsoft Corporation). In the Geomagic Wrap 2015 software, Andrews’ facial axis points were set on the maxillary teeth. These points were decomposed in 3 cardinal directions generating values on the x-, y-, and z-axis. Considering the y-axis values referred to the depth dimension (cervico-oclusal plane), only the x-axis values (transversal plane) and the z-axis values (sagittal plane) were tabulated. Microsoft Excel 2013 was used to graphically determine the mean maxillary dental arch shape for both study groups at T1 and T2, using the interpolar function.
Sample size calculation
A minimum difference of 2 mm in the intercanine distance, a standard deviation of 1.65, an alpha error of 5%, and a test power of 80% were considered for sample size calculation. Twenty participants were required in each group.
Interim analyses and stopping guidelines
Randomization (random number generation, allocation concealment, implementation)
A simple electronically generated randomization was performed before trial commencement using the Random Allocation Software program. Randomization ensured patients’ allocation in both groups with a 1:1 ratio. Allocation concealment involved numbered, sealed, and opaque envelopes prepared before trial commencement. One envelope was sequentially opened for each participant during recruitment. Each envelope contained a card with the name of 1 expander. The initials of the name of the participant, the type of expander, and the date of allocation were identified in the external surface of the envelope. One operator was responsible for the randomization process, allocation concealment, and implementation.
Double-blinding was not possible because the operator and patients were aware of the type of expander that was being installed. However, blinding was accomplished during outcome assessment once the maxillary occlusal radiographs, the photographs, and the digital dental models were unidentified during analysis.
One operator (ACMA) performed all the measurements and repeated them in 30% of the sample at least 1 month later. The intraexaminer error was assessed using the intraclass correlation coefficient.
Statistical analysis (primary and secondary outcomes, subgroup analyses)
Kolmogorov-Smirnov tests were used to verify normal distribution of variables. Considering that the variables showed normal distribution, parametric tests were used. Intergroup comparisons for initial age were performed with t tests. A chi-square test was used for intergroup comparison regarding sex ratio. Intergroup comparisons of T1-T2 changes were performed using multiple linear regression analysis. The independent variables were both treatment and the starting forms to adjust the comparisons to possible differences at T1. Results were regarded significant at P < 0.05. Bonferroni correction for multiple tests (tests performed on a set of 12 measurements) was applied. Differential expansion assessment was performed, comparing the difference between c-c and 6-6 width change in both groups using t tests. Associated 95% confidence intervals were calculated. All statistical tests were conducted with SPSS Statistics for Windows, version 25.0 (IBM, Armonk, NY).
One hundred sixty-one participants were recruited from May to November 2015; 105 patients (65.21%) were excluded because they did not meet the eligibility criteria. Fifty-six patients were randomized in a 1:1 ratio to the study groups (experimental group, 28; comparison group, 28). The trial ended when the sample size allowed a dropout rate of approximately 30%. Figure 7 shows the participants’ flow chart with reasons of losses and exclusions before and after randomization.