Evaluation of bony changes observed in mid-palatal suture after rapid maxillary expansion using fractal analysis

Introduction

Transverse deficiency of maxilla is one of the common reasons of malocclusion, and is characterized by crowding of teeth, deep palate, and posterior cross-bite [1]. This abnormality is treated with rapid maxillary expansion (RME) in children and adolescents [2]. In the past, occlusal radiographs have been used to evaluate the mid-palatal suture, while now, cone-beam computed tomography (CBCT) is employed with great success. CBCT is inexpensive and has many advantages, such as low radiation dose and accessibility. It can be safely used to examine bone tissues of the maxillofacial region by allowing 3-dimensional examination. Evaluation of mid-palatal suture fusion using CBCT permits accurate assessment of the mid-palatal suture while avoiding superimposition of other anatomical structures. In this way, surgical treatment can be avoided, and late adolescent and young adult patients can be treated with RME [3-5].
A retention period is required to prevent a recurrence after expansion. With prolongation of the retention period, mineralization of the mid-palatal suture increases and recurrence risk decreases. Measuring bone density is thought to be helpful in preventing a recurrence [6, 7].
Fractal analysis (FA) is a method used to investigate the complexity of fractal structures. Input data are images, and trabecular bone model is measured with box-counting method [8]. A smaller fractal dimension (FD) indicates that pores inside the bone are higher and more porous, whereas a higher FD indicates that the architecture of bone is more complex, dense, and pores inside the bone are fewer [9].
Currently, there is a growing interest in FA in many fields of science, including medicine. In dentistry, FA is used in many areas, including evaluation of bone density, dental implants, and bone grafts [10].
Although various authors recommend a retention period of 3 to 6 months after RME, there is still no consensus on this issue today [11].

Objectives

This study aimed to evaluate the new bone formed in the mid-palatal suture after retention period in patients treated with RME using FA.

Material and methods

This retrospective study was approved by the local ethics committee of our institution (protocol No: 135). This study was carried out in line with the Helsinki Declaration of 1975, as revised in 2000. Consent was obtained from patients’ legal guardians. In the study, mid-palatal suture images of patients who underwent RME treatment were analyzed using FD analysis.
Inclusion criteria: 1) CBCT taken before and after RME treatment; 2) maxillary transversal deficiency with posterior cross bite; 3) patients without any previous systemic disorders.
Exclusion criteria: 1) CBCT images with the mid- palatal suture area not suitable for evaluation (motion or metallic artefacts); 2) congenital deformity; 3) cleft lip/palate; 4) patients with incisive canal cyst.
CBCT images of 30 patients (18 females and 12 males) before treatment (T0) and after retention period (T1) were evaluated.
Tooth-borne (hyrax) type expander was used in this study (Figure 1). For over-expansion, the expansion period was continued until the palatal cusp of the maxillary first molar reached the buccal cusp of the mandibular first molar. Extent of the expansion was determined according to the needs of patients, with frequent clinical controls. Patients and their parents were requested to rotate the screw two-quarter turns per day, in the morning and evening. CBCT images were taken before (T0) and after retention (T1) period, during which no other orthodontic treatment was applied to patients.
CBCT images (Castellini, X Radius Trio Plus, Italy) were acquired from the same device, with 90 kVp, 8 mA, field of view (FOV) 16 × 18, slice thickness of 1 mm, and voxel size of 0.2 mm [3]. All radiological evaluations and FA measurements were performed by oral and maxillofacial radiologist with four years of experience (EMAO). For intra-examiner reliability and calibration of evaluations, the same observer reviewed the images two weeks after first evaluation.
Steps used to obtain the main image required for evaluation of the mid-palatal suture were as follows:

1. In axial plane, the antero-posterior reference line was set to be in the mid-palatal plane.
2. In coronal plane, the horizontal reference line was adjusted to pass through the center of palate in the maxillary first molar region, and intersected with the antero-posterior reference line.
3. The mid-palatal suture was drawn by marking the anterior and posterior nasal spine points of the palate in the sagittal plane. The axial view section, called the “main image” where the mid-palatal suture was traced, was obtained.
4. Horizontal line drawn along the mid-palatal suture was determined to be at the center of the distance between the floor of the nose and the roof of the mouth (Figure 2).

FD analysis was calculated by box-counting me­thod using ImageJ software (National Institutes of Health, Bethesda, MD, USA) [12]. After identification of the mid-palatal suture, the imaging area was designed as a 40 × 120 pixel rectangle at T0 and T1 on the axial CBCT image, extending from the posterior border of the incisive canal to the anterior border of the posterior nasal spine (Figure 3). Subsequent procedures included:

• the selected ROI was copied;
• the amplified image was blurred with a Gaussian filter (35 sigma) as an intensity correction step;
• blurred image was removed;
• 128 gray pixel values were added to each pixel position;
• an image was created, where substantial parts of the image represented specific areas with different brightness levels, such as the bone marrow and trabeculae;
• the image was binary with a brightness;
• the rendered image was eroded, thereby reducing noise;
• the image expanded, and the structures became more prominent;
• the trabecular bone was converted from white to black, and the bone marrow was converted from black to white using the software’s “invert” option;
• the resulting image was skeletonized;
• the process involved etching the structures until a single pixel line remained;
• the image was divided into 2, 3, 4, 6, 8, 12, 16, 32, and 64 pixel squares, with fractal box-counting algorithm of the software. Squares with trabeculae and the total number of frames were calculated for each square of different sizes (Figure 4).

Statistical analysis

IBM SPSS Statistics version 21.0 (IBM Corp., Armonk NY, USA) was employed to analyze the data. Descriptive statistics were performed for minimum-maximum values, and mean and standard deviations. Intra-examiner reliability was determined by calculating Cohen’s kappa (κ) value, and Shapiro-Wilk test was used to evaluate whether the distribution was normal. Paired t-test was applied to compare the groups. Statistical significance level was assumed as p < 0.05.

Results

A total of 30 patients, including 18 females and 12 males, participated in the current study. The average age of patients was 14.29 ± 1.62 years. The duration of expansion and retention periods varied between 85 and 146 days, with an average of 112.33 (± 19.03) days. Intra-examiner reliability was calculated as κ = 0.903. When the fractal value of the median palatine suture was compared before and after the procedure, a statistically significant difference was found among them (p < 0.001) (Table 1). Changes in fractal values were 0.13 ± 0.09 in males and 0.15 ± 0.07 in females, with no statistically significant difference (p = 0.403).

Discussion

RME is widely used in the treatment of patients with maxillary transversal deficiency. Retention period after RME plays an important role in stability. Although there is no consensus on the retention period, some authors suggest that this period should last for 3 or 6 months. However, a prolonged retention period reduces patient’s comfort [13-15]. The density of bone structures in the maxillofacial region can be evaluated with Hounsfield units (HU) obtained from computed tomography (CT). CBCT is used more frequently in evaluating the maxillofacial region due to its low cost and radiation dose compared with CT. Although various researchers have indicated methods of converting HU to gray values obtained from CBCT, the accuracy of HU values acquired from this transformation is still controversial [16, 17]. Studies with FA data from CBCT images have shown that FA is a reliable method for evaluating the density of bone tissue [18]. Naveda et al. [19] evaluated the repair of mid-palatal suture on CBCT after mini-screw-assisted rapid palatal expansion treatment in 24 patients, and observed a significant reduction in bone density after a 6-month retention period.
In the literature, after surgically-assisted RME, bone density was examined, and it was reported that the newly formed bone could not reach pre-surgical bone mineralization even after 6 or 7 months [6, 20, 21].
Schauseil et al. [22] evaluated the density of the mid-palatal suture on CBCT images after RME in 14 patients. They reported that the mid-palatal suture density decreased significantly after expansion. After six months of retention, the bone density did not return to its pre-RME value, but there was a significant increase.
Although the mineralization of the bone formed after distraction osteogenesis is satisfactory, it is less mineralized than normal bone, which may lead to a relapse [23]. Similar to previous studies, the ossification of the bone formed after the procedure did not reach its previous values in our study. This difference was also demonstrated numerically by the FA value. Even though ossification values did not reach previous rates, no relapse was observed in our patients, because their fixed orthodontic treatments were continued after RME.
Kwak et al. [24] evaluated the mid-palatal suture maturation with FA on CBCT, and found a strong correlation and statistically significant difference between FD and mid-palatal suture maturation. Similarly, in our study, a significant difference was found among T0 and T1 in the mid-palatal suture FD; FD value was lower in T1.
FA has the favor of being more objective and quantitative than other methods. Many studies have been done using FA, and its usefulness in dentistry has been proven. In this study, it was proposed that FA can be used to assess ossification during the retention period of the mid- palatal suture pre-operatively and post-operatively.
Weyrich et al. [25] investigated the effects of various expansion appliances on RME, and they results were similar after the retention period. Even though only one type of appliance was used in our study, it can be presumed that similar results can be obtained with other types of appliances.
This study has some limitations. FD differ according to methods used in studies on FA for bone diseases. Since FD may differ according to the calculation methods used, combining approaches used in clinical practice can simplify operator practice and eliminate differences between interpretations. Additionally, only one type of appliance (hyrax) was employed in the current study; therefore, no comparison could be made with different appliances.

Conclusions

Even though the FA value of the new bone formed after RME does not reach the initial rate, it can be specu­lated that a recurrence can be prevented by starting fixed orthodontic treatment after RME. Evaluation of CBCT images using the FA method is reliable to assess mid-palatal suture ossification in patients undergoing RME treatment, and to determine starting time of fixed orthodontic treatment.

Disclosures

1. Institutional review board statement: The study was approved by the Health Sciences NonInterventional Research Ethics Committee of the Ankara Medipol University No: 135.
2. Assistance with the article: None.
3. Financial support and sponsorship: None.
4. Conflicts of interest: The authors declare no potential conflicts of interest concerning the research, authorship, and/or publication of this article.

References