Age-dependent interactions of maxillary sutures during RME and their effects on palatal morphology

During rapid maxillary expansion (RME, sometimes called rapid palatal expansion [RPE]), the palatine process of the maxilla and the horizontal laminae of the palatine bone are separated by application of force in children and adolescents, while the pterygoid processes in the caudal region spreads laterally [39]. These combined effects lead to the therapeutically desired basal expansion in patients with pronounced maxillary constriction.

In adults, supportive surgery is necessary for successful RME treatment. This is thought to be caused by an age-related increase in the sutural bone density of the median palatal suture, reduced bone elasticity and fusion of the circummaxillary sutures during skeletal maturation [1, 11, 25, 30, 36]. Without pterygopalatomaxillary separation during surgically supported RME, only an anteriorly open triangular expansion can be achieved, whereas additional pterygopalatomaxillary separation can realise a parallel expansion [30]. Surgical separation of the maxilla from the sphenoid is also recommended as a preventive measure to avoid tension-related complications in the region of the skull base [18].

In children and adolescents, there is no unanimous opinion in the literature on when conventional RME treatment leads to which expansion mode. Many studies focused on the transverse plane and here especially on changes in the median palatal suture. However, in their systematic reviews, both Bazargani et al. [4] and Liu et al. [27] found primarily no consensus on whether RME treatment leads to triangular, i.e. anteriorly greater expansion, or to parallel expansion of the median palatal suture. However, patient age, measurement methods and recording techniques vary greatly in these studies.

The morphological changes of the tooth-bearing palate are of particular interest for orthodontic treatment. Kinzinger et al. [23] could demonstrate with a dental cast analysis for the first time that the therapeutic effects of rapid maxillary expansion on palatal morphology vary in patients with two different dentition stages. They concluded that RME should be performed in the early mixed dentition if parallel palatal expansion is desired. In later dentition stages, however, maxillary expansion occurs triangularly.

Increasing obliteration tendency or sutural bone density of the median palatal suture cannot be the sole cause of such different morphological treatment reactions of the tooth-bearing palate. Rather, an additional interaction with transversely running palatal sutures is to be suspected, which causes an age-dependent change in the position of rotation and resistance centres.

In the present study, the results of a three-dimensional (3D) radiological diagnosis are first compared with a limited cast analysis. Then, for the first time, the therapeutic effects of RME treatment on the morphology of the tooth-bearing palate are retrospectively quantified three-dimensionally with a larger number of patients subdivided according to chronological age using study casts. The combination of results is used to highlight and discuss the respective influence of the transverse palatine suture and the pterygopalatomaxillary junction [41] on the morphological changes.

The goals of the present study were to investigate the following questions:

An RME appliance with exclusively dental anchorage was used in all patients in the present study to ensure comparability. This appliance with a hyrax screw (palatal screw type S, Forestadent, Pforzheim, Germany with a lift height of 0.2 mm) was fixed with two prefabricated bands on the maxillary first molars and fitted with two occlusal rests on the first premolars or deciduous molars, respectively. The appliance was activated twice daily until the desired transverse arch expansion including a moderate overcorrection was achieved. The appliance then remained passively in situ for about 6 months.

In the transverse plane, the opening width of the median palatal suture was measured at projected connecting distances in the region of the anchorage teeth (Fig. 1a), the transverse expansion angle (Fig. 1b) was determined, and the anterior–posterior ratio was calculated.

In the frontal plane, the superior and inferior distances of the open intermaxillary suture were measured (Fig. 2a), their ratio (inferior/superior) calculated, and the frontal expansion angle determined (Fig. 2b).

Finally, a 3D reconstruction was made from the CBCT datasets for visual assessment of the surrounding sutures (intermaxillary, median palatine, transverse palatine and pterygopalatomaxillary sutures).

A total of 120 dental casts were measured, taken before insertion of the RME appliance (T1) and immediately after its removal (T2). The dental casts were digitised for the analysis.

The dental arch width was measured anteriorly at the premolars or deciduous molars and posteriorly at the first permanent molars according to Pont [34]. The width of the palate was measured between the most coronal points of the gingival margin at the first premolars or deciduous molars and the first permanent molars (gingival–alveolar plane). Starting from this plane, the width was determined in 2 mm steps ascending cranially up to 6 mm (skeletal–basal plane; Fig. 3). On these three exemplary vertical planes (dental, gingival–alveolar and skeletal–basal, i.e. gingival–alveolar +6 mm cranially), the ratio of the anterior–posterior width was determined to record the quality of the transverse expansion (values < 1 = inverse V‑shaped/delta-shaped; 0 = parallel, > 1 = V-shaped/triangular).

The palate height was determined by measuring the perpendicular junction of the raphe median line with the junction lines of the most coronal points of the gingival margin at the first premolars or deciduous molars and the first permanent molars. Anteriorly and posteriorly, the palatal height was measured median and 5 mm right (first quadrant) and left (second quadrant) paramedian of the palatal centre (Fig. 3). The ratio of anterior–posterior height was calculated median and paramedian (values < 1 = relatively greater increase in posterior height; 0 = equal change in anterior and posterior height, > 1 = relatively greater increase in anterior height).

The width ratio and the height ratio indicate changes in the palatal depth in the sagittal plane.

The measurements of 9 patients with CBCT scans were correlated with the results of the cast analysis on the three defined levels (dental, gingival–alveolar and skeletal–basal) and their anterior–posterior ratio to allow a direct comparison between the results of the width measurements on the dental cast and the RME effects upon the median palatal suture.

Normal distribution was evaluated visually and with the Shapiro–Wilk test after data collection. Treatment-associated changes in variables were analysed for intragroup comparisons using the linked t‑test. Differences between the groups were assessed using analysis of variance (ANOVA). Post hoc testing was performed using the Tukey test. Homogeneity of variance was confirmed using the Levene test. Mean and standard deviation (SD) as well as the confidence interval (CI) were reported for each variable. Statistical significance was assumed at p-values < 0.05. The significance level was defined as follows: p ≥ 0.05 not significant, p < 0.05 significant, p < 0.01 highly significant, p < 0.001 highly significant. Of the dental casts, 25% were randomly selected and measured again by the same investigator after 3 months to determine the combined method error (MF) according to Dahlberg [9]. The error of the method for linear (height, width) and angular measurements was calculated with the formula MF = √(∑d²/2n), with d as the difference between two measurement results and n as the number of repeated measurements. The MF in the present study was < 1 for all measurements (height 0.61 mm, width 0.55 mm).

The CBCT scans of the patients from PG1 and PG2 show an almost parallel opening of the median palatal suture in the transverse plane wherever measured. However, the size of the opening is smaller in the patient from PG2 after an identical number of hyrax screw activations. It is striking that the paired pterygopalatomaxillary sutures were still partially open in the 7.3-year-old patient from PG1, but completely closed in the 10.2-year-old patient from PG2. The scan of the 13.3-year-old patient from PG3, on the other hand, shows a V-shaped opening of the median palatal suture which decreases from anterior to posterior, and closed pterygopalatomaxillary sutures. The age-dependant different expansion modes are also confirmed in the frontal views: parallel expansion of the intermaxillary sutures starting from the palate to the floor of the nose in the two younger patients from PG1 and PG2, but a triangular expansion with a cranially tapering opening in the older patient from PG3 (Figs. 5, 6 and 7).

The comparison of the three patients reveals fundamental differences in the behaviour of the transverse palatal suture after RME. In the two younger patients, a continuous opening of the transverse palatal suture is visible in the transverse plane on both sides. This is confirmed by the 3D reconstruction. In addition to the clearly opened intermaxillary and transverse palatal sutures, a bilaterally opened transverse palatal suture is visible in the transverse plane from both coronal and caudal sides. In the two younger patients from PG1 and PG2, the palatal sutures are consequently open in a cruciform shape and divide the hard palate into four areas. In the oldest patient from PG3, on the other hand, the transverse palatal suture cannot be depicted continuously in the scans of the horizontal plane. It only appears open over a short distance at the intersection with the median palatal suture. The 3D image shows a comparatively smaller, but triangular opening of the median palatal suture. Since the transverse palatal suture can only be visualised medially and appears obliterated laterally, the bony palate appears to be divided into two parts by the V‑shaped opened median palatal suture.

The width ratio a/p, i.e. the quotient of the width differences between T1 and T2, was determined on the three defined levels (dental, gingival–alveolar, skeletal–basal). In PG1, there is a parallel to slightly dorsal V‑shaped transverse expansion on all levels. The ratio of the palatal width from anterior to posterior thus remained largely unchanged in PG1. The situation is different in PG2 and especially in PG3. Here, the increase in palatal width is significantly greater anteriorly than posteriorly at all levels (PG2: p < 0.001, p = 0.006, p = 0.009; PG3: p = 0.003, p < 0.001, p = 0.002) and thus occurs in a ventral V‑shape. This is particularly pronounced in PG3 at the gingival–alveolar and skeletal–basal levels. Accordingly, the differences between PG1 and PG3 are significant at all levels (p < 0.001, p = 0.022, p = 0.021; Tables 4 and 5).

In PG 1, the height at each measurement point increases significantly both anteriorly and posteriorly. In the two older patients (PG 2 and PG 3), however, with one exception (PG 3: 5 mm left lateral), the change in posterior height is insignificant. In the sagittal, anterioposterior comparison, the height increase in the older patients (PG 2 and PG 3) occurs more pronounced anteriorly than posteriorly. In PG 1, on the other hand, the median height increase is almost the same. It is striking that almost everywhere the height increases are greater on the right and left sides than directly at the raphe median line. There are no significant differences in the palatal height change between the groups (Tables 6 and 7).

The height ratio a/p, i.e. the quotient of the height differences between T1 and T2, was determined on three sagittal planes. In all patients and in all groups, the height changes were more pronounced anteriorly than posteriorly. An increase of the ratio occurred from PG 1 to PG 2 to PG 3, both median and 5 mm right and left paramedian. The results of the height measurements allow the conclusion that reduced height changes with increasing age in the posterior region are the reason for this. Since the intergroup comparison revealed significant changes in only two places, the results are to be interpreted as tendencies (Tables 8 and 9).

The forces and moments generated by hyrax screw activation act not only on the median palatal suture but also on the surrounding sutures [4, 14, 21, 22, 35, 37], especially the palatine bones and the pterygoid process of the sphenoid bone. The tensions are initially concentrated on the anterior palate, then run dorsally along the median palatal suture and via the palatine bone to the sphenoid bone, the zygomatic process and the medial orbital walls [7]. The forces generated after therapeutic opening of the median palatal suture do not drop significantly, allowing the conclusion that the main resistance to palatal expansion lies not in the median palatal suture itself but in the surrounding maxillary connections [19, 46]. However, different, age-related structural changes must be considered.

In the case of the median palatal suture, it has been possible to qualify and quantify postnatal development based on histological studies of human specimens [24, 28, 32, 42]. Melsen [28] divided the sutural ossification process from birth to the age of 18 years into three phases. With increasing age, she describes a reduction in sutural width with increasing meandering interlocking of the two palatal processes. Further histological examinations of human palates prove an age-dependent obliteration process from posterior to anterior with generally very low obliteration values, but with strong inter- and intraindividual variations [24, 25, 32, 42].

The transverse palatine suture separates the palatine process of the maxillary bone from the horizontal plate of the palatine bone and crosses the median palatine suture in its dorsal region. In 1977, Persson and Thilander [32] found in a histological study of human specimens that the transverse palatine suture begins to obliterate later than the posterior, but earlier than the anterior part of the median palatal suture. However, the extent of obliteration seems to be less in a direct age comparison.

In 1982, Melsen and Melsen [29] also described the postnatal development of the palatomaxillary region from the newborn to the age of 27 in a human cadaver study. They found a change in morphology of the transverse palatal suture occurring before and during pubertal growth. A slightly wavy course develops into a distinct squamous (overlapping) suture.

Tschechne [40] was able to specify these age-related morphological changes of transverse palatal suture based on a study with 155 human skulls in 2005: He found that the position of the transverse palatal suture changes with age in both sagittal and vertical directions. While the maxilla moves ventrally, the transverse palatal suture develops a bone deficit. Its sutural growth then fills the bone deficit caused by maxillary displacement. This sutural growth initially takes place on the ventral, maxillary side. Until the age of 7, this growth is five times as large as on the palatal side, until the age of 12 it is still 2.5 times as large. In the next few years, sutural growth changes both qualitatively and quantitatively: by the age of 19, the maxillary side of the suture grows only one-tenth as much as the palatal side. In addition, sutural growth between 12 and 19 years represents only about 20% of the total growth until the 19th year of age. It can be assumed that the markedly reduced growth rate from the age of 12 increases the tendency to obliterate and thus the rigidity of the transverse palatal suture significantly. This is also evident in the oldest patient with CBCT scan in this study.

Timms [39] described the anatomical proximity of the paired palatine processes of the maxilla and palatine bone with the pterygoid process of the sphenoid bone. Consequently, he investigated a possible age-dependent correlation between expansion of the dentoalveolar arch and the pterygoid hamulus by intraoral measurements. Timms found that the pterygoid processes of the sphenoid bone spread laterally outwards in the lower region. Despite only a weak correlation, there was a tendency for increasing age to be a factor in the gradual reduction of this basic movement.

However, Ghoneima et al. [14] found no significant changes in pterygopalatomaxillary sutures after RME treatment in 20 patients aged between 8 and 15 years in a CT study. This is also evident in the CBCT scans evaluated in this study. In 8 patients between 8 and 14 years, the pterygopalatomaxillary sutures are bilaterally closed. Only the youngest patient, 7.3 years old, shows partially open sutures (Table 1, Fig. 5a).

Melsen and Melsen [29] described the role of the pterygopalatine sutures in preventing posterior expansion and concluded that this suture limits the extent of expansion and dictates the expansion pattern under RME treatment.

From the results of the studies by Timms [39], Ghoneima et al. [14] and Melsen and Melsen [29], the different findings of the median palatal sutures in relation to the obliteration findings of the pterygopalatomaxillary sutures can thus be explained, especially from the first compared to the second, but also to the third patient example of this study.

Thus, the morphology of the median palatal suture itself does not appear to have a limiting factor on the opening mode. The results of the present clinical study combined with the radiological case studies indicate that the age-related changes in the transverse pterygopalatine and palatomaxillary sutures are decisive for the quality (parallel or V‑shaped) of the median palatine suture expansion and thus also for the morphological changes of the maxillary palatal vault.

The effects of skeletal expansion with RME appliances are largely determined by the point of force application and its relation to the maxillary rotation centres. According to various studies, these are located either in the dorsal part of the median palatine suture or close to the frontomaxillary sutures [2, 6, 15, 38, 44, 45]. However, the marked effects of ageing on palatal morphology suggest that both the position of the centre of resistance and the centre of rotation of the maxilla are not fixed, but change. Melsen and Melsen [29] have already suggested that the centre of resistance of the maxillary complex changes during postnatal development, showing a shift towards the fused sutural region from the juvenile to the pubertal period.

In younger patients up to age 10 (PG1), an almost parallel median palatal suture opening occurs, accompanied by and corresponding to morphological changes in the palatal vault, particularly in the ventral part, and around the palatine processes of the maxilla. There seems to be a mutual interaction between the median and the transverse palatal sutures: The younger the patients are, the more the median palatal suture opens, this influences the transverse palatine suture, which also opens.

Lione et al. [26] interpreted CT data and demonstrated that in younger patients RME treatment significantly increases the distance between the lateral pterygoid processes. However, if the palatomaxillary and the pterygopalatine sutures are also morphologically altered due to age as described in the studies by Melsen and Melsen [29] and Tschechne [40], the bones involved behave as one unit under the therapeutically applied force systems [39]. As a result, the opening of the median palatal suture is met with increased resistance in the posterior region with increasing age [13, 43]. The maxillary centre of resistance is shifted dorsocranially [5] and the centre of rotation is shifted ventrally, resulting in a V-shaped opening of the median palatal suture and a corresponding influence on the morphology of the palatal vault (PG3). As the sphenoidal bone is not paired, the pterygoid process is bent laterally [44] and tensions are created around the sphenoidal bone [17, 20].

The interaction of the various centres of rotation is the cause of palatal height and shape changes after RME treatment. Especially the centres of rotation in the frontal plane near the frontomaxillary sutures cause the bone of the hard palate to rotate and pivot laterally, consequently leading to a relative paramedial increase in palatal height. In younger patients (PG 1), this effect occurs not only in the palatal processes of the maxilla but also in the horizontal plate of the palatine bone. With increasing age and cascading obliterations of the transverse pterygopalatine and palatomaxillary sutures, only the osseous maxillary parts are affected to a significant extent.

The present study was the first to evaluate the effects of rapid maxillary expansion on the morphology of the maxillary palate in relation to patient age. The changes were analysed in the three spatial planes on dental casts.

The palatal width increased significantly in all cases at all measurement points. While the early treatment group shows a greater width increase posteriorly than anteriorly, this effect is reversed the two later treatment groups. The widening is then significantly greater anteriorly than posteriorly on all levels. This is also confirmed by the results of the width ratio calculation.

The palatal height increases significantly during treatment anteriorly in all three groups, but in the posterior region median and right and left paramedian only in the youngest patient group. With increasing age, the height changes in the posterior region become smaller, resulting in different alterations of palatal depth. Overall, the treatment-induced changes are clearly more pronounced in width than in height.

A parallel and more even opening of the suture from the palate to the nasal floor is age-dependent and happens only in patients with an early treatment begin, whereas a later treatment start leads to a V-shaped opening with a decrease in widening. Treatment success thus significantly depends upon the age at treatment begin.

Visual and metric analyses of CBCT datasets allow the interpretation that age-dependent obliteration tendencies of the sutures have a decisive influence on the biomechanics of the forced skeletal expansion of the maxilla. Of the surrounding structures, in addition to the pterygopalatomaxillary sutures, the transverse palatine suture has a key role for the different morphological effects. Its age-progressive obliteration causes a craniodorsal shift of the maxillary centre of resistance and a ventral shift of the centre of rotation, which is responsible for an altered suture expansion mode while using the same force application.

The partially possible comparison of selected measurement on dental casts and CBCT scans indicates that conclusions about the type of suture opening can be drawn from cast measurements alone. For this indication, radiation- and cost-intensive CBCT datasets appear to be dispensable.