INTRODUCTION

Rapid maxillary expansion (RME) represents a well-proven method for treatment of maxillary transverse deficiency. Nevertheless, it can generate several problems in skeletally mature patients, such as fenestration of the buccal cortex, root resorption, lateral tipping of the teeth, and, finally, the inability to open the midpalatal suture.1,2 To date, the decision of whether to use surgically assisted RME (SARME) has been an age-based one, although no reliable correlation between chronological age and changes in the midpalatal suture in skeletally mature patients exists.3 Since SARME results in a higher morbidity than conventional RME,4,5 finding ways to determine if and to what extend the expansion of the maxillary complex will need surgical assistance or not is important to reducing the potential risks for the patient.6

A decrease in the midpalatal suture width, which can be age and functional related,7–9 is a potential prognostic factor for an increase of the resistance against conventional RME.10–12 For this reason, searching for a radiological device that solidly visualizes the midpalatal suture complex and enables one to measure the suture width is essential to the therapeutic approach.

As reported by Wehrbein and Yildizhan,13 the occlusal radiograph is not a reliable method with which to define the suture structure. Multislice spiral computed tomography (MSCT) has generally improved diagnostic imaging possibilities in medicine.14,15 The prototype of a flat-panel detector-based volumetric computed tomography (fpVCT) demonstrates an increase in spatial resolution and improves the evaluation of detailed skeletal and dental structures.16–18 It has already been reported16,19 that fpVCT is superior to the MSCT with regard to visualization of the sutural structures.

The aim of the present pilot study was to analyze the sutural width via fpVCT compared to the respective histological preparations in a porcine model.

MATERIALS AND METHODS

For the investigation palate specimens from five young (aged 12–18 weeks) and five old (aged 128–208 weeks) Sus scrofa domestica, German land race, were used. After sacrificing the animals, the maxillae were removed and the palatine mucosa was detached. Along the midpalatal suture line, six cylindric bone samples with a diameter of 1 cm were obtained and fixed in 4% buffered formaldehyde solution at ambient temperature.

All imaging was performed with the fpVCT prototype (General Electric Global Research, Niskayuna, NY), which comprised two flat-panel x-ray detectors based on amorphous silicon, with a dimension of 20.5 × 20.5 cm and with 1024 × 1024 pixels. The pixel size amounted to 200 × 200 µm, with a radiographic enlargement of 1.43 that created a resolution of 140 µm. All data were generated with the same protocol, which has been described in detail elsewhere.19 To explore the data sets, we used the voxtool 3.0.64u software implemented on an Advantage Workstation 4.2 (GE HealthCare, Chalfont, St. Giles, UK).

For further analysis the data sets were displayed as facial slices perpendicular to the suture and the palatal bone surface. This area was marked with a line on the palatal surface of a volume-rendering image of the respective specimen via the software. The line was oriented perpendicularly to the suture, as identifiable at the palatal bony surface. The image of the volume rendering with the reference line drawn in was stored as a reference image for the histological preparation. Six width measurements equally distributed along the suture were performed in the reference image plane.

The two observers were unaware of the experimental design and underwent standardized preparatory training. The images were evaluated in a randomized order regardless of the age of the specimens.

For the histological analysis the obtained bone specimens were embedded in methyl methacrylate (Technovit 9100, Heraus Kulzer, Hanau, Germany). In the reference plane parallel sections with a thickness of 5–7 µm were prepared using a LEICA RM 2165 microtome. For the histomorphometric analysis the sections were stained with alizarin/methylene blue.20 For the measurement of the sutural width in the histological images the software Axio Vision combined with a microscope, Axioskop 2 plus/Axio Cam MRc5 (Carl Zeiss, Oberkochen), was used. The measurements were carried out by the two observers in the same way described for the fpVCT images.

RESULTS

In general, the observations revealed differences between the younger and the older animals in the suture anatomy according to the course of the suture, with a higher interdigitation and smaller suture width in the old age group (Figure 1). Furthermore, the bony structures appear to be more compact in the older group. In the fpVCT images and in the histological sections a closure of the suture or obliterations in the suture line, respectively, could not be observed. The average values for the parameter “suture width” for the young and the old age groups via fpVCT and histomorphometric analysis are shown in Table 1.

View Figure

• Download as Powerpoint
• Download Figure

Table 1 Comparison of the Suture Width in the Younger and Older Age Groups Determined by Flat-Panel Volume Computed Tomography (fpVCT) and Histomorphometry

View Fullscreen

The data obtained by fpVCT as well as by histomorphometric analysis reveal a highly significant age dependency of the measured suture width (both P < .0001), with lower suture width values in older subjects (compared to the younger group).

The averaged suture widths measured in the fpVCT images show a distinctively higher mean compared to the histomorphometric data (young group  =  0.25 mm vs 0.18 mm; old group  =  0.15 mm vs 0.10 mm), with high statistical significance (P < .0001) (Figures 2 and 3).

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

The reader shows no significant influence in the fpVCT group (P  =  .311) or in the histomorphometry (P  =  .235) group.

DISCUSSION

SARME is a common treatment procedure that becomes necessary in numerous, especially adult, patients.6,21 However, it may cause severe complications.4,22 There is controversy with regard to which age is required for SARME, since many authors23–25 have shown successful nonsurgical RME in adults as well. Therefore, finding a congruent diagnostic procedure to indicate SARME is essential but not yet possible.26 However, most studies6,10–12 show that age-correlated morphological changes of the midface go along with an increased resistance against conservative RME, which necessitates SARME. Unfortunately, until today it has not been possible, especially in the borderline age group of 19–20 years,21,26 to identify patients who will need SARME before expansion, since anatomical changes of the midpalatal suture are subject to a considerably high interindividual variability and are not only correlated with the chronological age.3 Hand wrist radiographs typically used for the evaluation of skeletal maturity are not indicative for estimation of the condition in the midpalatal suture.27,28 The assumption that the total obliteration in the midpalatal suture in adults is the main factor for the increasing resistance in adult patients29,30 lacks evidence but is still popular in the actual literature.6,21

Knaup et al.8 found significant differences in the human midpalatal suture width between younger and older subjects. These findings correlate well with the radiological and histological observations made in our study. We found highly significant differences in mean values of the suture width between the younger and the older groups, with no parts of obliteration in the midpalatal suture. A previous three-dimensional micro-CT study27 demonstrated an increase in bone density in the region of the midpalatal suture in older subjects. According to these findings as well as to the observations made in the present study, the morphological changes in the condition of the sutural complex and the changes in the bony structures connected to the maxilla presumably cause the increase in resistance against the expansion force in older subjects.10,27,31 In addition to this, a functional influence on the sutural bone apposition and the general cranial suture morphology was demonstrated in many studies.7,32–34 In this context, Katsaros et al.7 revealed that a reduced masticatory function caused by a soft diet is accompanied by less bone apposition in the facial sutures of rats. Since many authors found that craniofacial morphology affects the bite force with, for example, smaller values for the long-faced type,35,36 a different strategy in maxillary transverse deficiency therapy for different craniofacial morphologies is reasonable. Taking age-dependent morphology as well as functional changes of the midpalatal suture into consideration, it appears to be meaningful to identify an easy-to-determine parameter with which to predict the feasible resistance against conventional RME. The parameter “suture width,” dependent on both function and age, is possibly of high prognostic value for the therapeutic strategy. According to the sutural anatomy, humans and pigs show comparable characteristics,37 so that the data acquired in pigs in the present study most likely show a high similarity to data in humans.

The present study shows that precise data from the sutural complex can be obtained with fpVCT. Numerous studies17,38–40 substantiate the good diagnostic capabilities of fpVCT for small bone structures. Obenauer et al.41 demonstrated that with the prototype of a fpVCT the imaging of different phantoms can be distinctively improved in comparison to the results afforded by MSCT.

Micro-CT allows a high local resolution of 5–50 µm and is an appropriate system for nondestructive visualization of fine bone architecture.42 It was used to evaluate animal and human craniofacial sutures, but micro-CT studies27,43,44 limited the sample size to approximately 1 cm and are therefore not applicable.

There are only very few comparative studies concerning radiological and histomorphometric analysis of suture preparations.13,44 Recinos et al.44 revealed a major morphological consistency with minor changes in the same structures of micro-CT images with histological sections of murine sutures. The authors saw a possible explanation for these differences in the fact that incompletely calcified regions may not be identified as fused with micro-CT because of their lower tissue density. The observed differences between fpVCT images and histomorphometric analyses in the present study could be explicable in the same manner. On average, we found distinctively smaller differences in suture width within the histological sections than in the respective fpVCT images. This could be explained by a reduced calcification of the bony sutural borderline adjacent to the sutural lumina, which can therefore not be correctly depicted by fpVCT. Crucial for the measurement of the suture widths is the identification of the bone-suture interface. As a result of the resolution of the fpVCT of about 0.150 mm, the bone border is blurred within this dimension as a result of the partial volume effect. To compensate for that, we used a threshold for bone segmentation of 60 HU, which is the average of the bone density and the density of the liquid inside the suture. The threshold was kept constant in all analyses; therefore, only the same absolute error is added to the results. Suture distances measured in the fpVCT data sets all appeared larger than the results obtained with histology, indicating that the threshold was chosen a little bit too high, which yielded overestimation of the partial volume effect with respect to the bone. Additionally, the differences result from the fact that the virtual sections of the fpVCT are thicker than the sections used for the histomorphometric analysis, causing a considerable decrease in resolution.32 Nevertheless, the present pilot study demonstrates that fpVCT is capable of determining the midpalatal suture width highly reliably within the examined different age groups, although the exact values do not always correspond to those obtained histomorphometrically.

The fpVCT tool used is a computed tomographic modality that until now has not been available for patient scanning because of the lack of variable collimation of the x-ray tube.41 In the available literature cone-beam CT technology is described by several author 45–47 as a means to examine RME effects;48,49 however, the morphological appearance of the suture structure itself or measurements of the width has not been described before.

CONCLUSIONS

• With fpVCT it is possible to determine the suture width in younger and older age groups of a porcine model. A closure of the suture or obliterations in the suture line could not be observed in either the young or the old age groups.

• Since the factor suture width is of possible prognostic value, changes in the therapeutic approach concerning the SARME are possible.