The study was approved by the Ethics Committee of Chinese PLA General Hospital. All patients provided informed consent.
This work was supported by the Key Research and Development Plan for Social Development Project of Hainan Provincial Science and Technology Department (SQ2016SHFZ0112) and Scientific Research Support Funds of PLA General Hospital (2017FC-TSYS-2015).
Mandibular defects often follow re-sectioning of benign and malignant maxillofacial tumors involving the mandible.1 Although the development and popularization of microsurgery and free flap technology provides an option for effective mandibular reconstruction during removal of the lesion, some patients still fail to receive timely repair because of limiting medical conditions, high risk of recurrence, or unwillingness to increase surgical trauma. These patients often present with compensatory rotation of the ramus and condyle of the bilateral mandibles, as well as disorganized occlusion, limited mouth opening, and lower facial deformity (the lowest one-third of the face). Mandibular reconstruction in this group of patients is a challenge for clinicians, and, in the absence of standardized repair methods, reconstruction mainly depends on a clinician's experience, thus resulting in uneven outcomes in different patients. In this study, accurate restoration of the physiologic position of the residual bone segment of the mandible is an urgent problem. Therefore, we attempted to standardize the recovery of the position of the mandibular condyle and ramus via computer-aided design; importantly, we designed a 3-dimensional (3D)-printed template to accurately perform the operation and restore the continuity of the mandible.
The study population consisted of 5 patients who were admitted to the Department of Oral and Maxillofacial Surgery of PLA General Hospital during 2016 to 2017. At the time of admission, all patients had undergone re-sectioning operations >6 months prior in other hospitals, and were relapse free at their final follow-up. The baseline data of the patients are shown in Table 1.
Computer-Aided Determination of the Physiologic Position of the Condyle and Residual Stump of the Mandible
Computed tomography (CT) data (in DICOM format) of the maxillofacial region were collected from each patient preoperatively. Data partitioning was performed using ProPlan CMF 2.0 (Materialise, Leuven, Belgium) to separate the maxilla and stumps of the bilateral mandibles; this was followed by a 3-dimensional reconstruction. Then, anatomical markers in the CT images (Table 2) were utilized to reposition the bilateral mandibular ramus stumps. Detailed procedures were as follows: first, the vertical position of the condyle was determined. With reference to the standard superior joint space (2.8 mm) on the temporomandibular joint (TMJ),2 the selected reference landmark comprised the distance from the most lateral point of the condylar arc to the most concave point of the acetabulum. The condyle was vertically moved until the superior joint space returned to normal (Fig. 1A). Second, the position of the condyle on the sagittal plane was determined. The selected landmark was the midpoint of the anterior and posterior slopes of the TMJ by reference to the standard anterior (2.06 mm) and posterior (2.30 mm) joint space, as previously reported.3 We, therefore, moved the condyle parallel to the sagittal plane until both the anterior and posterior joint spaces returned to normal (Fig. 1A). Third, the axial position of the condyle on the horizontal plane was determined. The intercondylar axis angle, formed by the intersection of the right condylar axis [defined as the line passing through the condylus medialis (CM) and condylus lateralis (CL) of the right condyle] and left condylar axis (defined as the line passing through CM and CL of the left condyle) at the anterior border of the foramen magnum (ICAa) is 145° to 160°,3 as previously reported; in this study, the corrected angle was set to 155° (Fig. 1B). Fourth, we adjusted the anterior-inferior angle, formed by a line tangent to the posterior border of the mandibular ramus (ie, the line between articulare [Ar] and gonion [Go], and the Sella-Nasion plane [∠SNAG]) to be 94° (Fig. 1C).4 Fifth, the lateral and medial positions of the gonion on the coronal plane were adjusted until the ratio of zygomatic width to mandible width (Zy-Zy/Go-Go) was 1.33/1 (Fig. 1D).5,6 The aforementioned procedures were repeated to ensure that the quantitative values were within a reasonable range (Table 3).
Virtual Surgery and 3D-Printed Surgical Template
An osteotomy line of the mandible was positioned and prepared for virtual osteotomy with the aid of data segmentation. Then, division of the fibula (both left and right fibula) was implemented through data partitioning. A 3D model was constructed on the left or right fibula, as per patient needs. The resulting fibular model and lesion of the mandible were overlapped to determine the appropriate position of the fibula, length of the body, position of the mandibular angle, height of the elevation, and angle of the mandible. Accordingly, the virtual osteotomy line of the fibula was generated. The 3D model of the reconstructed fibula was further used to prepare a positioning template for mandibular reconstruction. CT data of the mandible and fibula were input into the forward engineering software, 3-matic (Materialise), and 2 surgical templates for osteotomies of the fibula and mandible were manufactured following the planned osteotomy lines. Another surgical template, allowing the accurate positioning of the free fibula flap in the mandibular reconstruction as planned, was designed based on the shape, angle, and position of the reconstructed fibula. A high-accuracy 3D printer was used to print osteotomy and positioning templates for the fibula and mandible.
Main measure outcomes included: Shape accuracy: Maxillofacial CT data, within 2 weeks postoperatively, were collected and imported together with the preoperative data into the reverse engineering software, geomagic control, for preoperative and postoperative image matching, as well as analysis of reconstruction accuracy. Location accuracy: based on the reconstructed 3D model, postoperative results concerning the reposition of the mandible were compared to normal physiologic values. A 1-sample t-test was then used for data analysis in SPSS software.
The error of actual surgical sites was within 4 mm in all 5 patients; the mean error was within 3 mm in 94.3% of surgical sites.
Postoperative measurements of bilateral condyles and mandibular rami were compared to the predicted values. These comparisons showed that the mean width of the superior joint space was 2.94 ± 0.60 mm, which did not differ from the normal width of 2.8 mm; the mean width of the anterior joint space was 1.45 ± 0.84 mm, which was significantly lower than the normal width of 2.06 mm; the mean width of the posterior joint space was 2.90 ± 0.75 mm, which was significantly larger than the normal width of 2.30 mm; the mean value of ICAa was 160.5° ± 7.1°, which was significantly higher than the set value of 155°; the value of ∠SNAG was 93.1° ± 3.9°, which was not significantly different from the normal value of 94°; and the Zy-Zy/Go-Go ratio was 1.326 ± 0.36, which was not significantly different from the normal value of 1.33.
A male patient, 27 years of age, was diagnosed with mandibular osteoblastoma and underwent surgical removal with concurrent free iliac bone graft in 2016. The patient suffered from postoperative bone graft infection and received secondary surgery for removal of the bone graft and internal fixator, 2 months after initial surgery. He was admitted to our hospital in 2017. Preoperative computer-aided design for mandibular reconstruction and postoperative outcomes is shown in Figures 2–8.
The mandible is defined as the lower one-third of the facial skeleton, which is highly important in maintenance of the facial pattern and chewing function. Mandibular defect is a commonly observed craniofacial defect, mainly caused by tumor and trauma; as this often leads to severe facial deformities and dysfunctions that involve chewing, speaking, swallowing, and breathing, the defect seriously threatens a patient's quality of life. Restoration of the shape and function of the mandible is challenging for maxillofacial surgeons.
Mandibular reconstruction is considered pivotal in the repair of orofacial defects, as it can restore a patient's facial shape, and facilitate recovery of occlusal and voice functions. Mounting evidence indicates that mandibular reconstruction with vascularized fibula has become a preferred treatment for mandibular defects.7 Moreover, there has recently been an increase in the accuracy of mandibular repair resulting from the rapid development of computer-aided design technology.8
An ideal reconstruction of the mandible, with continuity of anatomical structure and normal orofacial functions, is expected to restore the continuity of the mandible; the normal physiologic shape of the mandible; the physiologic position of the condyle in the articular fossa; and the occlusal relationship (as much as possible).
Concurrent reconstruction is undoubtedly the optimal choice for mandibular repair, and the ongoing development of computer technology continues to elevate the accuracy of positioning for the mandibular condyle and ramus.9 For patients with normally positioned bilateral mandibular condyles and rami, the position of the fibula can be determined by image fitting; for those with the normally positioned unilateral condyle and ramus, image simulation is used to position the condyle and ramus; and for bilateral mandibular defects or deformities, matched data of the mandible can be acquired in the database. However, obvious dislocations of the condyle and fractured ends, as well as occlusal dislocation, hinder recovery of the physiologic shape of the mandible in patients undergoing 2nd-stage reconstruction, who are unable to implement concurrent reconstruction for a variety of reasons. Therefore, we attempted to restore the physiologic positions of the mandibular condyle and ramus in patients with obsolete mandibular defects through a computer-aided approach, and then planned for actual surgeries. Successful recovery of mandibular shape was achieved in our study.
Notably, bilateral condyles under strong muscular traction cannot be repositioned as planned, in the absence of postoperative intermandibular traction. In our study, all patients were subjected to postoperative intermandibular traction and exhibited expected outcomes. Undeniably, concurrent mandibular reconstruction, rather than 2nd-stage reconstruction, is still the preferred choice and has achieved desired outcomes with a great reduction in the difficulties of both surgical design and procedures. In addition, initial CT data (obtained prior to the initial surgery) are available for repositioning the ramus stumps in the 2nd-stage reconstruction through image fitting or registration of the upper and lower jaws.
Our method for 2nd-stage mandibular reconstruction is only recommended in cases with an inability for concurrent reconstruction, or in the absence of initial CT data. In the present study, we repositioned the mandibular condyle and ramus by computer simulation, then implemented actual osteotomy and reconstructive surgeries. In this process, there are some precautions: preoperative design for mandibular reconstruction, especially the selection of anatomical landmarks and adjustment of the occlusal relationship, needs to be performed autonomously by experienced clinicians; this is essential during the selection of anatomical landmark points and adjustment of the occlusal relationship. The surgical procedures should be in strict accordance with the designed template to ensure accurate implementation of the surgical schedule. First, the physiologic position of the condyle should be determined within the articular fossa; subsequent adjustment should be performed using the center of the condyle as the origin of coordinates, especially following adjustment of the angle created by the posterior border of the mandibular ramus and axial rotation of the mandible. The adjustment process needs to be repeated at least twice to correct manually selected points and associated errors. Considering that the ICAa is not sufficiently accurate because of condylar reconstruction, the axial position of the mandible should generally be further adjusted based on the relationship between the midline and the tooth position. As a result of local bone remodeling, soft-tissue remodeling, and scar formation, we found that the coracoid process often affected the degree of opening after reconstruction, so we performed coronoidectomy in all patients. The surgery restored the continuity of the mandible and the mandible morphology to an approximately normal state. However, due to the asymmetry of muscle strength on both sides, the reconstructed mandible cannot remain in the preferred position. Thus, to balance against the abnormal muscle strength, type I maxillomandibular elasticity traction was necessary for all patients, for at least 1 month. Because of long-term malpositioning of the jaw bone, all patients had significant compensatory tooth movement. The purpose of postoperative orthodontics was to modify the compensation. Therefore, we used fixed orthopedic technology for decompensation. A type I occlusal relationship was the goal of correction. After orthodontic treatment, the dental column defects were repaired with removable denture or implant repair.
In summary, in cases of obsolete mandibular defects, the most important step toward good facial shape and orofacial function is to restore the physiologic positions of the mandibular condyle and ramus. The combined use of computer-aided design and physiologic parameters can reposition the mandibular condyle and ramus; on this basis, 3D-printed surgical templates can be designed and fabricated to standardize the accurate reconstruction of the mandible in cases of obsolete mandibular defects.
1. Guerrier G, Alaqeeli A, Al Jawadi A, et al. Reconstruction of residual mandibular defects by iliac crest bone graft in war-wounded Iraqi civilians, 2006-2011. Br J Oral Maxillofac Surg
2. Dalili Z, Khaki N, Kia SJ, et al. Assessing joint space and condylar position in the people with normal function of temporomandibular joint with cone-beam computed tomography. Dent Res J (Isfahan)
3. Wang MQ. Oral Anatomy and Physiology. Beijing: People's Medical Publishing House; 2012.
4. Li M, Gui L, Zhang ZY, et al. Intraoral approach for the correction of prominent mandibular angle. Zhonghua Yixue Meixue Meirong Zazhi
5. Zhou Z, Liu DL, Liu ZG, et al. Measure of mandible of Han nationality people and pertinence analysis. Zhongguo Meirong Yixue Zazhi
6. Wang X, Zhang ZK. Cephalometric radiography of Chinese beauty crowd. Zhonghua Kouqiang Yixue Zazhi
7. Ferreira JJ, Zagalo CM, Oliveira ML, et al. Mandible reconstruction: History, state of the art and persistent problems. Prosthet Orthot Int
8. Sieira Gil R, Roig AM, Obispo CA, et al. Surgical planning and microvascular reconstruction of the mandible with a fibular flap using computer-aided design
, rapid prototype modelling, and precontoured titanium reconstruction plates: a prospective study. Br J Oral Maxillofac Surg
9. Toto JM, Chang EI, Agag R, et al. Improved operative efficiency of free fibula flap mandible reconstruction with patient-specific, computer-guided preoperative planning. Head Neck
Keywords:© 2018 by Mutaz B. Habal, MD.
Computer-aided design; 3-dimensional-printed template; mandibular reconstruction