Abstract
To evaluate the stress pattern in the craniofacial skeleton in a patient with unilateral cleft deformity of the secondary palate and alveolus in response to various techniques of surgically assisted rapid maxillary expansion (SARME).
Three patient-specific composite skull models were developed for finite element model analysis. The details of the modeling procedure have been described in Part I of this series. The finite element analysis was performed on each model with a specified SARME technique in combination with RME using Abaqus (6.7).
The ideal form of surgery in SARME for patients with unilateral cleft deformity of the secondary palate and alveolus would be complete unilateral LeFort I with pterygoid dysjunction in combination with midpalatal split, followed by isolated midpalatal split and zygomatic buttress osteotomies.
A more invasive SARME technique can significantly reduce the resultant stresses. However, this benefit should be weighed against the risk of increasing complications associated with more extensive surgeries. When a more conservative surgical technique is selected, it would be preferable to perform a midpalatal split rather than zygomatic buttress osteotomies, as indicated by the stress-strain distribution and displacement pattern associated with different SARME techniques.
INTRODUCTION
Surgically assisted rapid maxillary expansion (SARME) is frequently utilized to expand the maxilla in adolescents and adult patients.1 This technique is selected because of the increasing thickness of the bony structures, with corresponding reduction in the elasticity that occurs with skeletal maturity and, to a lesser extent, the increasing ossification of the median palatal suture. Thus, various combinations of osteotomies have been proposed to ‘weaken’ the midfacial skeletal structures to allow for maxillary expansion through orthopedic transpalatal devices.
However, there is no agreement in the literature regarding the extent to which a subtotal LeFort I surgical osteotomy can be expected to consistently achieve stable correction. Shetty et al.2 concluded that complete midpalatal and pterygo-maxillary osteotomies are essential for predictable skeletal expansion in adults. Exclusive use of bilateral zygomatic buttress osteotomies, which was previously advocated, appears to be inadequate. Holberg et al.3 demonstrated lowest stress values associated with a complete osteotomy at the LeFort I level. While a complete LeFort I osteotomy with a midpalatal osteotomy would allow for unrestricted transverse expansion, the question is whether any combination of a subtotal 2-segment Lefort I will achieve the desired result. The question has clinical relevance in terms of peri-operative management and financial impact on health care. Kusakabe et al.4 showed that separating the pterygo-maxillary junctions assisted in the bodily displacement of the lateral maxillary segment, which would facilitate correction of maxillary arch constrictions in the adult unilateral cleft lip patient.
As a natural consequence, the stress-strain distribution varies with the specific surgical approach utilized with SARME. The different osteotomy cuts used for SARME modify the magnitude and distribution of stress with expansion forces and can be expected to result in different biological and mechanical responses to expansion. However, the literature5–7 contains few clinical references concerning the application of these surgical techniques to unilateral cleft palate patients, and these techniques have yielded disparate results. In addition, there have been no published data on finite element analysis evaluating SARME in unilateral cleft patients.
MATERIALS AND METHODS
The objective of this finite element research is to evaluate the stress pattern in the craniofacial skeleton in individual unilateral cleft palate in response to various techniques of SARME. The techniques evaluated in the present study were the following: midpalatal suture split (MPS); zygomatic buttress osteotomy with pterygo-maxillary dysjunction (ZB); and LeFort I osteotomy with pterygo-maxillary dysjunction in combination with midpalatal split (LeFort I/MPS).
All of the described surgical techniques were performed on the cleft side only. The clinical relevance of the present research is that as variations in the surgical technique lead to different stress-strain distributions on the maxillofacial skeletal structure, the technique or combination of techniques that exhibits the lowest magnitude and the most uniform distribution of stresses with desired expansion pattern is the procedure of choice for SARME.
Methodology
With institutional review board approval, the archived computed tomographic scan (slice increment, 0.625; pixel size, 0.488; matrix 512 × 512 pixels) of a patient with unilateral cleft was used in this study. The patient was 16 years old, male, and had previously undergone repair of the lip and palate during infancy, with subsequent alveolar bone graft.
Three composite skull models were used to represent the craniofacial structures. The details of the modeling procedure have been described in Part I of this series (Figure 1). Each model consisted of approximately 131,000 tetrahedron elements. The volume mesh from Abaqus (6.7) (SIMULIA, Providence, RI) was imported to MIMICS (Materialise, Leuven, Belgium) to assign the material properties (Table 1). The material properties were assigned to the cortical bone, cancellous bone, and teeth based on the Hounsefield units. For the boundary conditions, a zero displacement and rotation was imposed on nodes around the foramen magnum.
The finite element analysis was performed on each model with a specified SARME technique in combination with RME using Abaqus (6.7) (SIMULIA). To simulate the expansion, the displacement between the node of premolars and molars was simulated in 0.25-mm increments, with total expansion measuring 2 mm on each side. The biomechanical response of the model was analyzed in terms of Von Mises (VM) stress and displacements.
RESULTS
The results are summarized in Tables 2–7 and Figures 2–7. With ZB only, the maximum lateral expansion was observed in the second molar; this expansion decreased progressively in the anterior aspect. An asymmetric expansion pattern was observed between the cleft and the non-cleft segment. The expansion had far-reaching effects on the base of the skull and the cranial vault. The maximum stresses were in the order of 1800 N/mm2 (1 N/mm2 = 1 MPa). The region of maximum stress was the primary palate. The center of rotation of the cleft segment was in the premolar region.
With MPS only, a uniform pattern of lateral and overall displacement was observed, with anterior and posterior displacements being equal in magnitude. The maximum stress was reduced to 480 N/mm2 and was observed in the molar-premolar region on the cleft side. Other areas of high stress were the zygomatic buttress, inferior orbital rim of the non-cleft side, and lateral nasal wall on both sides. With LeFort I/MPS, a more uniform displacement pattern was observed compared with other two osteotomy combinations. The displacement pattern was more symmetrical when cleft and non-cleft sides were compared. The maximum stresses were drastically reduced to 230 N/mm2 which were observed in the molar-premolar region and the zygomatic buttress on the non-cleft side.
With SARME osteotomies there were high VM stresses associated with the base of the skull, which were confined to the region of the medial pterygoid plate, scaphoid fossa, and palato-vaginal canal. These stresses were highest with ZB and were significantly reduced with LeFort/MPS and MPS only.
When comparing the dental expansion among the three protocols, the expansion at the dental level was greater with lateral osteotomies only. When comparing the expansion at the alveolar bases (maxillary tuberosity), the expansion at the alveolar bases was greatest with LeFort I/MPS, followed by MPS only. The least lateral expansion at the alveolar bases was observed with ZB.
DISCUSSION
A composite skull model was used for finite element model analysis. Previous studies8,9 and the authors' imaging experience have shown that the computed tomographic images are not reliable for generating three-dimensional (3D) models of teeth; the 3D laser scanner (LPX 1200 Roland Laser Scanner, Roland DGA Corporation, Irvine, Calif) was used to generate the 3D model of the teeth from the maxillary stone model. The reliability of this scanner in generating 3D models has been confirmed in a previous study.10
The study of stress-strain distribution and displacement pattern in the present study revealed considerable differences in the effectiveness of these surgical measures. More extensive surgery for SARME resulted, to a varying extent, in reductions in these stresses. While surgical weakening of the zygomatico-alveolar crest apparently produces only a minor reduction of the induced stresses, a complete LeFort I osteotomy from the piriform aperture to the pterygo-palatal junction along with a midpalatal split is far more effective. Though LeFort I/MPS osteotomy demonstrated the greatest protective effect for the cranial base and the midface, it is more extensive than a conventional lateral osteotomy.
Displacement Pattern
A more symmetric expansion of the two maxillary halves was observed with LeFort I/MPS, followed by MPS; isolated ZB led to asymmetric expansion of the two maxillary halves. When comparing the dental expansion among the three protocols, the expansion at the dental level was greater with lateral osteotomies only. When comparing the expansion at the alveolar bases (maxillary tuberosity), the expansion was greatest with the LeFort I/MPS, followed by MPS only. The least lateral expansion at the alveolar bases was observed with isolated lateral osteotomies. These findings point to the fact that more lateral bending and rotation of maxillary halves was observed with isolated lateral osteotomies, whereas more bodily displacement of maxillary halves was evident with MPS and LeFort I/MPS osteotomy. It can be suggested that lateral bending and rotation would be more prone to relapse following SARME.
An increase in lateral nasal width, and consequently an improvement in the nasal airway, can be expected with SARME. With the ZB only there was practically no increase in the lateral nasal width. Although only 2 mm of expansion was simulated, it can be suggested that even with more expansion, as created in the clinical situation, there would be minimal improvement in the nasal airway following SARME when only the ZB is performed. A similar response has also been observed in older individuals following SARME.11 Mossaz et al.12 reported that when SARME is carried out unilaterally the maxillary width increase is primarily confined to the operated side, while the opposite side serves as anchorage. However, in our study both sides showed increases in nasal cavity width.
It was interesting to note the anterior displacement pattern of the naso-maxillary complex with SARME. With MPS only (ie, no lateral osteotomy), the teeth and the alveolar bases were displaced in a posterior direction. With ZB only, in general there was a tendency for posterior displacement of the maxillary complex. With LeFort I/MPS, the maxillary complex was displaced in the anterior direction. It is evident from the present findings that a more conservative surgical approach would worsen the maxillary retrusion already present in cleft cases. Hence, with a more conservative surgical approach for SARME, an active maxillary protraction therapy would be beneficial.
The lateral bending of the medial and lateral pterygoid plates was observed with all three techniques. The lateral bending was greater in the inferior aspect compared to the superior aspect, indicating that the articulation with the base of the skull offers resistance to lateral expansion in this region. The lateral bending of the pterygoid plates was maximal for the MPS only. This could be related to the combination of lateral rotation and bodily displacement of the naso-maxillary complex. The zygomatic complex was displaced in the supero-lateral direction following LeFort I/MPS and isolated MPS. With ZB only, the zygomatic complex was displaced infero-laterally.
Stress Pattern
The present study confirms that more conservative osteotomies for SARME lead to higher stresses in the craniofacial region. With ZB, the anterior primary palate offered maximum resistance to lateral expansion, followed by the infra-orbital region on the non-cleft side and the zygomatic buttress bilaterally. Even after ZB, the stress at the zygomatic buttress on the cleft side was greater when compared to the stress on the non-cleft side. This could be related to the high stresses at the incomplete primary palate, which could be transmitted to the zygomatic buttress region of the cleft side. With isolated MPS, the maximum resistance to expansion was offered by zygomatic buttress of the non-cleft and the cleft sides. The magnitude of VM stress at the zygomatic buttress was less with ZB compared with isolated MPS. However, the maximum stresses in other areas were significantly reduced with isolated MPS compared to ZB. This agrees with the findings of Shetty et al.,2 who showed the midpalatal suture as the principal obstruction to maxillary expansion, as demonstrated by their experiments on a human photoelastic analogue skull.
The least magnitude of stress was observed with LeFort I/MPS. This agrees with the findings of Holberg et al.3 It must be kept in mind that the results of the present study cannot be directly compared to those of the study by Holberg et al.,3 as the latter study was performed on the non-cleft case and, also, the midpalatal suture was assumed to be patent. With LeFort I/MPS, the areas offering resistance to expansion were the infra-orbital region and the zygomatic buttress of the non-cleft side, followed by the zygomatic buttress of the cleft side.
With all techniques of SARME, the medial pterygoid plate and its surrounding structures (scaphoid fossa) offered significantly more resistance to expansion, compared to the lateral pterygoid plate. Hence, the dysjunction of the medial pterygo-maxillary junction seems critical when performing the SARME procedure in adults. Lateral pterygo-maxillary dysjunction could be avoided.
Clinical Implications
It can be concluded that the ideal form of surgery in SARME for patients with unilateral cleft deformity of the secondary palate and alveolus would be complete unilateral LeFort I with pterygoid dysjunction in combination with MPS. This agrees with the findings of Holberg et al.,3 who demonstrated that following SARME, the lowest stress values in the non-cleft case were associated with a complete osteotomy at the LeFort I level. Hence, more extensive surgery for SARME can significantly reduce the resultant stresses and possibly minimize the associated complications. However, this benefit should be weighed against the fact that with increasing invasiveness of surgery, the risk of complications also increases.13,14 Very conservative surgical relief, however, entails the risk of inadequate protection of the cranial base and midface, with the increased risk of consequent complications.15
Posttreatment results following LeFort I multiple-piece osteotomies to widen the maxilla may lead to a significant degree of relapse, along with the risk of compromising the palatal mucosa and its vascularity when excessive widening is performed.16,17 Hence, sometimes a more conservative approach may be desired to avoid the complications of LeFort I osteotomy. For a conservative surgical technique, a MPS osteotomy should be selected over a ZB, which also involves a risk of intraoperative bleeding18 and normally necessitates general anesthesia. However, some of these complications of ZB can be avoided if the pterygo-maxillary junction is not invaded.19
Studies20 reporting transverse maxillary distraction in unilateral cleft cases have indicated that the vector of the distraction should ideally be slightly oblique and not perpendicular to the midline palatal suture, similar to the maxillary constriction in non-cleft patients, which allows for a greater expansion of the more collapsed anterior, rather than the posterior, part of the maxilla. This vector can be achieved by placing abutment plates of the transpalatal multicomponent bone-anchored distractor20 asymmetrically, which increases the risk of loosening the abutment plate screws during the activation period. By following the appropriate surgical osteotomy, a more uniform pattern of displacement can be achieved and the risks associated with asymmetric placement of distraction devices can be avoided.
CONCLUSIONS
The ideal form of surgery in SARME patients with unilateral cleft deformity of the secondary palate and alveolus would be complete unilateral LeFort I with pterygoid dysjunction in combination with MPS, followed by isolated MPS and isolated lateral osteotomies.
A more extensive surgery for SARME can significantly reduce the resultant stresses and possibly minimize the complications associated with SARME.
However, when a more conservative surgical technique is desired, it would be preferable to perform a MPS rather than ZB, as indicated by the stress-strain distribution and the displacement pattern associated with different SARME techniques.