Evaluation of the density of the midpalatal suture after maxillary expansion: a comparative observational study

Abstract

Background

This study aimed to evaluate the effects of surgically assisted rapid maxillary expansion (SARME) and rapid maxillary expansion (RME) groups on midpalatal suture (MPS).

Methods

CBCT records who underwent RME and SARME between 2013 and 2024 were included in the study. CBCT axial sections taken preoperatively (T0) and after a 3-month retention period (T1) were evaluated using the MPS. Fractal Analysis (FA) method using the ImageJ program and compared between the groups.

Results

9 patients underwent SARME (%37.5) and 15 patients underwent RME (%62.5). FA values ​​of the SARME and RME groups at T0 were found to be 1.02 ± 1.17 and 1.46 ± 0.09, respectively. FA values ​​of the SARME and RME groups at T1 were found to be 0.98 ± 1.08 and 1.32 ± 0.08, respectively. The difference between T1 and T0 in the SARME and RME groups was 0.02 ± 0.09 and 0.34 ± 0.08, respectively. When FA differences were compared between the groups, no statistically significant difference was found. (p > 0.05)

Conclusion

The potential effect of increasing retention time on the clinical recovery process has been clarified. In patients who underwent RME and SARME, after 3 months of retention, MPS density decreased compared to the initial density. The findings suggest that increasing the retention time in both RME and SARME groups for increased ossification. FA provides a useful method for evaluating skeletal effects of RME and SARME.

Introduction

Transverse maxillary narrowness is a condition characterized by decreased horizontal growth of the maxillary dental arch, and is a common problem in orthodontic practice [1]. Common clinical findings in adults and children include unilateral or bilateral posterior crossbite, high dental arch, tooth inclination, anterior tooth crowding, and respiratory distress [2, 3]. Before deciding which type of maxillary expansion technique to apply to treat transverse discrepancies, it is crucial to assess the maturation of the midpalatal suture (MPS) [4]. Before maturation, the MPS is abundant in connective tissue and has a wide space between the maxillary bones. Since this connective tissue is not completely calcified, the suture area is radiolucent. During maturation, connective tissue begins to form at the suture edges and becomes a mixture of calcified bone and noncalcified tissue. As a result, the density ratio of the MPS increases during maturation. During adolescence, the suture area calcifies to resemble cortical bone, and the bone spicules become increasingly intertwined [5]. MPS separation has been proven to be an effective method for the treatment of maxillary stenosis and moderate maxillary crowding. This technique allows for the expansion of the maxilla, providing dentoalveolar and skeletal corrections and optimizing treatment outcomes [6, 7].

Rapid maxillary expansion (RME) is a conventional approach used to address transverse maxillary deficiency. By applying a high-magnitude transverse force rapidly to the upper teeth, it successfully induces maxillary expansion, leading to the separation of the MPS [8]. When the appliance is activated, it first creates pressure on the periodontal ligament, then tipping of the alveolar processes and anchor teeth follows. Finally, a distraction force is created in the MPS and diastema is seen between the central incisors [9]. The appliance must be kept in the mouth for 3–4 months to ensure that the width obtained becomes stable and the gap is filled with bone [7]. Surgically assisted rapid maxillary expansion (SARME) is another commonly used treatment option to surgically correct transverse discrepancies between the upper and lower jaws. This treatment is primarily used in young adults or adults with completely fused MPS or in cases where there is a discrepancy between the upper and lower jaws of > 5 mm in width and conservative orthodontic techniques cannot be used [10].

Cone beam computed tomography (CBCT) is an important radiographic technique that helps analyze, diagnose and measure the relationship between the maxillofacial skeletal anatomy and the structures surrounding this structure [11]. The main advantages of CBCT include its ability to capture images in sagittal, axial, and coronal planes, perform reconstructions without distortion, create 3D images, and use these images for modeling with specialized software [12]. CBCT has a lower radiation dose and cost compared to computed tomography (CT) [13]. The Hounsfield units (HU) scale is an effective method to quantitatively measure maxillary bone density via CBCT images [14, 15]. This scale was used to assess bone maturation at the MPS following RME without surgery [16]. Several studies have assessed bone density in HU at the MPS after SARME [17]. However, due to significant differences in data collection protocols and highly heterogeneous methodologies, new methods are needed to better understand the process of mineralizing the newly formed bone tissue in the MPS [4]. Fractal analysis (FA) is a mathematical method used to describe, analyze and measure structural patterns and complex shapes, especially bones [18]. Although its use in dentistry is increasing, this method can measure changes in the trabecular structure after jaw bone regeneration and assess the morphological pattern of the jaw bones and their potential variation over time [19, 20].

Previous studies on maxillary expansion methods (RME and SARME) and MPS have important limitations that justify the present study. One of these limitations is methodological heterogeneity; differences in data collection, imaging protocols, and analytical techniques make the results variable and often incomparable. This prevents the establishment of standard clinical guidelines. Furthermore, short-term follow-up in previous studies introduces the problem of not being able to evaluate the long-term remodeling and ossification processes of the midpalatal suture. The aim of this study was to evaluate changes in MPS and alveolar bone density after RME and SARME treatment using CBCT images and to investigate the hypothesis that retention time affects outcomes.

Materials and methods

Ethics committee approval was obtained before the commencement of the study from Eskişehir Osmangazi University Non-Interventional Clinical Research Ethics Committee (decision no: 54, dated:23.07.2024; ethics committee approval protocol number: 2024 − 253). A total of 24 patients with transverse maxillary narrowing who applied to Eskişehir Osmangazi University, Faculty of Dentistry, Department of Orthodontics and Oral and Maxillofacial Surgery between 2013 and 2024 were included in this study. The scientific ethical compliance of this study was ensured by informing the participants and/or their parents, both verbally and in writing, about the purpose and methodology of the study, and obtaining their signed consent.

Inclusion criteria

Patients diagnosed with transverse maxillary narrowing.

Patients who underwent either RME or SARME between 2013 and 2024.

Availability of preoperative (T0) and 3-month postoperative (T1) CBCT images for analysis.

Patients without systemic diseases or syndromes that could affect bone regeneration or healing.

Patients not taking regular medications that could interfere with skeletal or dental responses.

Exclusion criteria

Patients with systemic diseases or syndromes that might impact the study outcomes.

Patients who were taking medications regularly that could affect bone density or suture healing.

Patients who did not complete the 3-month retention period or had incomplete CBCT records.

Sample size

In this study, the required sample size calculation was made using G*Power 3.1.9.4 software. A two-tailed t-test was selected to compare the means of two independent groups. In the calculation, the effect size (Cohen’s d) was determined as 1.4, α error probability as 0.05 and power (1-β) as 0.80. The sample sizes for RME and SARME were calculated as 11 and 9, respectively. In this direction, a total of 24 samples were included.

Interventions and procedures

Rationale for RME and SARME selection

The choice between RME and SARME was determined based on the patient’s age, clinical evaluation, and radiographic assessment of the MPS maturity. RME was performed on patients with open or partially fused MPS, typically observed in younger individuals where non-surgical expansion was expected to achieve sufficient maxillary widening. SARME was indicated for patients with fully fused MPS or advanced skeletal maturity, commonly seen in older adolescents or adults, where surgical intervention was necessary to enable effective expansion. This decision-making process ensured that the most appropriate technique was applied to maximize treatment efficacy and minimize complications.

RME protocol

After the placement of mini screws and molar bands, impressions were taken with alginate. The appliance was cemented using dual-cure glass ionomer cement for molar bands and light-cured composite filling material for recesses. Hyrax expansion screws were activated twice daily (0.25 mm per turn), continuing until the desired transverse width was achieved. The appliance was retained in situ for 3 months to stabilize the expansion.

SARME protocol

The operations were performed under general anesthesia. A horizontal incision was made with a number 15 scalpel approximately 5 mm above the mucogingival junction, covering the right and left first molars of the maxilla. The mucoperiosteal flap was elevated upwards, and the apertura piriformis, anterior nasal spine, nasal cavity, infraorbital foramen, zygomaticomaxillary junction, and the tuberosity and pterygomaxillary junction were made completely visible with the help of subperiosteal dissection in the posterior region. After adequate dissection and removal of the soft tissue, an osteotomy was performed at least 5 mm apical to the tooth roots and approximately 10°-15° to the occlusal plane. This procedure was performed in a region starting from the edge of the piriformis aperture, covering the anterior lateral nasal wall, the lateral wall of the maxillary sinus, and the zygomaticomaxillary resistance zone, extending to the pterygomaxillary junction. After the completion of the lateral wall osteotomies, a vertical osteotomy was performed using piezoelectric surgery, starting from the anterior nasal spine level and extending between the central tooth roots, along the MPS. Care was taken not to damage the tooth roots during the procedure. The osteotomy line was carefully deepened with the help of an osteotome and a hammer, and depth control was provided by palpating the palatal surface of the premaxilla with the index finger of one hand. In continuation of this procedure, an osteotomy was performed using a chisel and the MPS was divided (Fig. 1). Afterwards, the incision lines were closed with 4.0 vicryl after bleeding control.

Fig. 1

This figure illustrates the post-operative lateral osteotomy areas following SARME. The highlighted regions include key anatomical landmarks such as the anterior nasal spine and zygomaticomaxillary junction, demonstrating the extent of surgical intervention

After the latent period, the hyrax screws of the expansion devices were activated 2 turns per day (0.25 mm per turn). After the desired expansion was achieved, the expansion was terminated and the device was kept in the mouth for retention.

The patients included in our study were divided into two groups as SARME group and RME group. Low-dose CBCT images were obtained from the patients preoperatively (T0) and at 3 months after retention (T1). The CBCT images were obtained in standingposition by using CBCT machine. FA evaluations were performed on CBCT routinely requested from patients before and after treatment.

Outcome measures

Primary outcome

Changes in MPS density and complexity were evaluated using FA. FA was conducted on CBCT images using the ImageJ software, following the protocol established by White and Rudolph [21].

CBCT acquisition

Images were obtained using a Planmeca Promax 3D mid (Helsinki, Finland) CBCT machine. Axial sections with a thickness of 0.3 mm were analyzed.

Fractal analysis

Fractal analysis was performed using the image J program. Image J is an open source, Java-based image processing and analysis software developed by the National Institutes of Health (NIH). ImageJ has a comprehensive toolkit for processing 2D and 3D images, which is why it is widely used in medical and scientific research. The program can process various image formats, apply filters, perform measurements, and perform advanced mathematical calculations. It can be downloaded for free from https://imagej.nih.gov.

To measure the structural complexity of the mid-palatal suture, a mathematical method called box-counting was used. This method analyzes how many boxes “fill” the complex pattern in an image by overlaying it with a grid of squares (or boxes) of different sizes. In this way, it becomes possible to numerically measure the complexity of the shape and the density of detail.

On axial sections, a 20 × 100 pixel rectangular ROI (Region of Interest) was selected parallel to the long axis of the midpalatal suture, encompassing the bone tissue where the midpalatal suture was visualized (Fig. 2). Image preparation for fractal analysis followed the protocol established by White and Rudolph [21]. While preparing the image for fractal analysis, there is a stage where the differences between the original image and the image we will process are determined. For this reason, the ROI was duplicated (Fig. 3a). A Gaussian blur filter was applied to reduce noise in the image (Fig. 3b). To highlight the structural features of the mid-palatal suture, the blurred image was subtracted from the original ROI (Fig. 3c). 128 Gy tones were added to the image for a more controlled conversion to black and white (Fig. 3d). The processed image was converted to a binary (black and white) image (Fig. 3e). To improve this processed image, erosion was applied to remove small artifacts and then dilation was applied to highlight prominent features (Fig. 3f and g). The invert option was applied, the pixel values ​​in the image are reversed (Fig. 3h). Skeletonization step was applied to obtain the representation of the basic contours of the midpalatal suture as lines (Fig. 3i). The skeletonized ROI was analyzed using the box-counting method, a tool in the ImageJ ‘Analysis’ menü (Fig. 4). The fractal value of the structural complexity of the midpalatal suture was obtained.

Fig. 2

The selected ROI on the MPS in CBCT axial slices is shown. This region represents the primary measurement area used for fractal analysis. The image contains a 20 × 100 pixel rectangle used as the starting point of the analysis process

Fig. 3

The image processing steps applied in ImageJ software are shown in detail. (a) Duplicated ROI, (b) Blurred image, (c) Subtracted image, (d) Added image 128 Gy values (e) Thresholded image (f) Eroded image, (g) Dilated image, (h) Inverted image, (i) Skeletonized image

Fig. 4

Fractal size calculation from processed ROI

All image analyses were conducted by the same observer. To assess intraobserver reliability, images from ten randomly chosen patients were re-evaluated after a two-week interval. Pearson correlation coefficients were calculated to evaluate the consistency of the measurements. No statistically significant differences were found between the initial and repeated measurements (p > 0.05), confirming consistent results. Intraclass Correlation Coefficient (ICC) was used for interobserver reliability. There is consistency between the measurements (p = 0.001). Retrospective design limitations, small sample size can be identified as potential bias in the study.

Statistical analysis

For the statistical analysis, the IBM SPSS Statistics Version 20 package program (IBM Corp.; Armonk, NY) was used. The Shapiro-Wilk test was used to assess the normality of data distribution for all variables. Independent t-tests were applied to compare FA values ​​between the SARME and RME groups for normally distributed variables. The Mann-Whitney U test was used for non-normally distributed variables. The test results were considered statistically signifcant if p was < 0.05. One-way analysis of variance with covariates (UNIANOVA) was used, considering age and gender as covariates. This statistical method adjusted pre-post differences by taking into account these confounding variables.

Results

The age range of the individuals was between 10 and 28 (mean 16.46 ± 5.09). 11 (45.8%) of the individuals were male and 13 (54.2%) were female. Of the female patients, 10 (76.9%) underwent RME, 3 (23.1%) underwent SARME; of the male patients, 5 (45.5%) underwent RME, and 6 (54.5%) underwent SARME. Age, T0 and T1 FA values, and T1-T0 FA values of female and male patients are shown in Table 1. SARME was applied to 9 patients (37.5%), and RME was applied to 15 patients (62.5%). T0 FA values were 1.02 to 1.46 (mean 1.17 ± 0.09), T1 FA values were 0.98 to 1.32 (mean 1.08 ± 0.08), and T1-T0 FA values were 0.02 to 0.34 (mean 0.09 ± 0.08). The difference values of the SARME and RME groups with T0 and T1 FA are shown in Table 2. When the groups were evaluated in terms of FA between T0-T1, it was seen that there was no statistically significant difference (p > 0.05).

Table 1 Gender-Based descriptive statistics

Table 2 Descriptive statistics and evaluation of differences between treatment groups

Although there was no statistically significant difference in FA values ​​between the SARME and RME groups (p > 0.05), a decrease in FA values ​​was observed in both groups at the end of the 3-month retention period compared to the initial values. When age-related ossification tendencies were evaluated using FA values, it was determined that younger patients in the RME group showed stronger bone remodeling, while adult patients in the SARME group had lower ossification rates. Pre-post differences adjusted for age and gender were not statistically significant between the groups (p = 0.288). This suggests that potential effects of age and gender were minimized by including them in the analysis.

Discussion

In the literature reviews, it has been reported that both RME and SARME are effective methods in the fixing of transversal maxillary deficiency. In addition, the limitations of the studies include short-term follow-up and small number of patients [22]. This study aimed to evaluate the ossification in MPS during the 3-month retention period by creating SARME and RME groups. MPS density can be a useful indicator for clinical applications. The first priority area of ​​use is to determine whether surgically assisted or traditional expansion methods are appropriate in adolescents and young adults. The second area of ​​use is to estimate skeletal effects before treatment [5]. The MPS density was examined the MPS density to evaluate the skeletal effects after the two methods and compared them between the two groups. To better understand the effectiveness of these methods, it is important to consider the imaging techniques used for evaluating MPS changes.

A CBCT-based method can be used to analyze the MPS. This method allows in vivo imaging of the MPS without superimposing anatomical structures. Therefore, CBCT provides both qualitative and quantitative measurements, allowing the evaluation of changes in the MPS in response to applied forces [12]. Measurements were evaluated using CBCT. In the literature, post-treatment MPS has been evaluated with various methods such as HU. Petrick et al. [23] reported that the amount of HU increased compared to the initial values ​​by taking a CT scan of the patient 7 months after SARME surgery. However, they reported that more clinical follow-up studies are needed to clarify the retention time required for transverse stability after SARME and that the sutures are generally thin, short and difficult to visualize, and that the suture gaps do not contain bone structures, which may affect HU measurements. While CBCT provides detailed imaging, the FA method offers a complementary approach for analyzing structural changes over time. The assessment of MPS ossification by FA is becoming a diagnostic parameter that helps to adjust the treatment to minimize the undesirable effects of RME treatment. If a pretreatment CBCT scan image is available, it has been reported that determining the MPS density ratio should be preferred as it is a more meaningful way to understand the response to expansion treatment [5]. FA evaluation was performed by transferring the images obtained from CBCT images to the ImageJ program. Ossification in the pre-treatment and 3rd month retention phase was compared between the groups. Although there was no statistically significant difference between the groups, lower FA values ​​were found in both groups compared to the initial values.

The importance of the retention period becomes evident when evaluating its role in treatment stability and minimizing recurrence. The retention period after MPS expansion plays a fundamental role in treatment; during this period, MPS ossification occurs and the possibility of recurrence is minimized [24]. The biological behavior of MPS in patients undergoing SARME has not been fully elucidated, particularly regarding the correct retention time to allow complete remodeling of the area. This issue requires further investigation to improve treatment stability [25, 26]. The retention time after RME is a critical choice in clinical practice; according to the literature, this duration ranges between 2 and 12 months [27]. The frequently applied method is to apply for 3 months. This period is a reasonable period for RME, but might not occur sufficient for SARME [28] The mean bone density at the MPS 6 or 7 months after SARME was lower than the preoperative baseline values, suggesting that a longer retention period should be considered after expander activations. However, no specific ideal time for this retention period has been specified [23, 29]. Although the present study assessed the effects of a 3-month retention period, the findings suggest that this duration may be insufficient, particularly for SARME patients. Given the slower ossification process observed in adults with fully fused sutures, future studies should evaluate whether extending the retention period to 12 months provides greater stability and minimizes relapse risk. A longer follow-up period would also allow for a more comprehensive understanding of the long-term bone remodeling process, contributing to refined clinical recommendations on retention duration.

Ok et al. [5] also applied RME to 20 patients aged between 8 and 13 years (mean 10.64) and evaluated the thinning in MPS using FA method. They reported a decrease in FA size and thinning in MPS thickness in CBCT images taken before the operation and during the retention period at 4 months after the activation period. Although a statistical comparison was made regarding bone loss in MPS, they reported that the clinical significance of the reported bone loss was controversial. Garib et al. [30] applied RME to 22 patients aged between 6 and 9 years in their study in Brazil. They obtained CT images from the patients before expansion and on the 30th day after activation to examine the changes. However, they did not take into account the data belonging to the retention period. In addition, the time interval evaluated in previous studies was quite low compared to other studies. Ozturk et al. [31] evaluated the new bone formation in the MPS after the retention period (average of 112 days) in patients treated with RME using the FA method. They compared the FA value of the MPS before and after the procedure. According to the comparison results, they reported that there was a statistically significant difference between them. When the FA values ​​were compared between the genders, they also found no statistically significant difference. According to the research results, although the FA value of the new bone formed after RME did not reach the initial level, they reported that relapse could be prevented by starting fixed orthodontic treatment after RME. The authors reported that the evaluation of CBCT images with the FA method would be a reliable method to evaluate MPS ossification in patients treated with RME and to determine the time to start fixed orthodontic treatment [31]. Based on the findings reported in the meta-analysis, RME has less bone density difference compared to SARME. Literature reports that bone healing during expansion is related to the patient’s age of onset. Given that SARME serves as a treatment option for adult patients who have completed growth, the age of the patients could potentially explain this variation [28]. These data may also have an impact on retention time; It is reported in different literature that the retention period varies between 2 and 12 months [32]. Findings in the literature indicate that MPS ossification occurs more rapidly in young patients. The findings suggest that standardizing retention periods irrespective of patient age or surgical intervention type may not be appropriate. In this study, when the 3rd month follow-up images were compared with the initial images, it was seen that the FA value in the RME group was lower than in the SARME group. However, when the fractal changes in the follow-up process in both groups were compared, no statistical difference was found. In addition, it was observed that the FA values in both the RME and SARME groups were lower than the pre-treatment fractal values and this was statistically significant. These differences indicate that not only age but also biological factors such as gender should be taken into consideration. For adult patients undergoing SARME, particularly those with fully fused sutures, extending the retention period beyond 3 months is recommended to enhance ossification and minimize relapse risk. The pre-post differences, when adjusted for age and gender, did not show statistical significance between the groups. This indicates that the inclusion of age and gender as covariates effectively controlled for their potential influence, ensuring that the observed outcomes were not significantly biased by these factors.

The study’s limitations include a short follow-up period and a small sample size, making it difficult to evaluate long-term treatment changes. Although the box counting method is a valuable method for assessing FA changes, it may also have its limitations. The retrospective design also constitutes the limations of the study.

Conclusion

The study findings confirm that the 3-month retention period may not be sufficient for complete ossification, particularly in adult patients undergoing SARME. The results highlight the importance of retention duration in ensuring treatment stability. Future studies should investigate optimal retention periods to improve ossification outcomes.