Value of preoperative mandibular plating in reconstruction of the mandible

Transcription

1 ORIGINAL ARTICLE Value of preoperative mandibular plating in reconstruction of the mandible Eitan Prisman, MD, FRCSC, * Stephan K. Haerle, MD, MSc, FRCSC, Jonathan C. Irish, MD, Michael Daly, MSc, Brett Miles, MD, Harley Chan, PhD Clinical Instructor, Division of Otolaryngology, Vancouver General Hospital, University of British Columbia. Gordon & Leslie Diamond Health Care Centre, 4th Floor-2775 Laurel Street, Vancouver, British Columbia Canada V5Z 1M9. Accepted 14 May 2013 Published online 29 May 2013 in Wiley Online Library (wileyonlinelibrary.com). DOI /hed ABSTRACT: Background. The purpose of this study was to evaluate the efficacy of preoperative versus intraoperative mandibular contouring using rapid prototyping technology. Methods. Ten patients requiring mandibular reconstruction had a preoperative mandibular plate contoured to a fabricated 3D mandibular model based on preoperative imaging. A traditional intraoperative plate was also contoured. Two surgeons blinded to the study compared the plates with respect to conformance, surface-area contact, and best overall match. A cost benefit analysis was then performed. Results. The average time to contour was seconds and seconds for the preoperative and intraoperative plates, respectively (p 5.83). Interobserver analysis revealed no statistically significant differences in conformance (p 5.38) or surface area contact (p 5.14). In 7 of 9 cases, the preoperative plate was selected for the final reconstruction. In 1 case, an intraoperative plate was not contoured because of the lateral extent of the tumor. Conclusion. In cases of mandibular distortion secondary to disease, pathologic fracture or defects involving multiple mandibular subsites this method is particularly advantageous. VC 2013 Wiley Periodicals, Inc. Head Neck 36: , 2014 KEY WORDS: preoperative mandibular plating, mandible reconstruction, contouring, head and neck cancer, rapid prototyping, stereolithography INTRODUCTION Mandibular reconstruction after partial or total mandibulectomy for squamous cell carcinoma (SCC), ameloblastoma, or osteoradionecrosis as sequelae after irradiation is critical for subsequent oral rehabilitation. 1 Historically, free bone grafts were used for mandibular reconstruction; however, this technique is limited by higher rates of bone resorption and infection, especially in the oncological setting requiring adjuvant therapy. 2 Microvascular reconstruction with osseous free tissue transfer has since emerged as the preferred technique for mandibular reconstruction in these situations. 3 Utilizing free tissue transfer has provided patients with improved oral competence, dental restoration, and mastication, as well as improved esthetic appearance. 4 6 Traditionally, a titanium plate is fashioned to the native mandible before segmental resection and later used as a template for reconstruction. This technique may be a challenging and time-consuming process with reports of intraoperative mandibular plate adaptation taking up to 60 minutes. 7 This time span may *Corresponding author: E. Prisman, Head and Neck and Microvascular Surgery, Department of Otolaryngology Head and Neck Surgery, Mount Sinai School of Medicine, One Gustave L. Levy Place, Box 1189, New York, NY eitan.prisman@utoronto.ca Contract grant sponsor: TECHNA-Guided Therapeutics (GTx) Program at the University Health Network, including The Kevin and Sandra Sullivan Chair in Surgical Oncology, The Hatch Engineering Fellowship Fund, The RACH Fund, and the Princess Margaret Hospital Foundation. represent a significant impact on operating room utilization and on patient care. Preoperatively contoured mandibular reconstruction plates may offer a significant advantage with regard to medical and economical aspects. Three dimensional (3D) anatomic models generated with CT data linked to the rapid prototyping of stereolithography have been used to preoperatively plan the reconstruction in orthopedic or craniofacial surgery Stereolithography is a process that polymerizes a liquid resin using an ultraviolet laser based on a virtual model. 11,12 This fabrication process is anatomically accurate and has been used to allow adaptation of mandibular reconstruction plates. 12 A European multicenter clinical trial analyzing the clinical application of anatomic models suggested several clinical applications including surgical planning and simulation, assistance in consenting patients, producing surgical implants, and providing a template for resection. 13 More recently, preoperative adaptation of mandibular reconstruction plates has allowed for simultaneous in situ osteotomies and rigid fixation of osseous segments during harvest. This has resulted in significantly reduced operative times. 7,14 The purpose of this prospective study was to investigate the efficacy of preoperative mandibular contouring compared to intraoperative contouring with respect to optimizing mandibular contour and to determine which cases would be most appropriate to apply this technology. PATIENTS AND METHODS Ten prospectively enrolled patients undergoing mandibular resection with osseous free flap reconstruction at the 828 HEAD & NECK DOI /HED JUNE 2014

2 PREOPERATIVE PLATING IN MANDIBLE RECONSTRUCTION FIGURE 1. Segmented mandible generated from preoperative CT shown in coronal (A), sagittal (B), and axial views (C), and as a virtual 3D surface (D). (E) The printed 3D mandible is shown next to a titanium reconstruction plate. (F) The preoperative contoured plate conformed to the physical model. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.] University Health Network, Toronto, Ontario, Canada, between March 2009 and January 2011, were included in this study. The Institutional Review Board approved the study protocol. All patients were referred to our institution for further management of their malignant cancer (n 5 7), benign cancer (n 5 1), or osteoradionecrosis of the mandible after primary irradiation (n 5 2). The pathological diagnosis and corresponding mandibular defects were defined preoperatively according to a previously published study by Urken et al. 15 Preoperative diagnostic head and neck CT scans of the mandible were transferred and imported to a Windows-based computer workstation with 3D modeling software (MIMICS v12.0, Materialise, Ann Arbor, MI) and a 3D computer image of the mandible was constructed based on image segmentation methods. According to the institutional protocol, the acquired images have a pixel size of mm 2 and a slice thickness of 2 mm. Manual editing was performed on the tri-planer views (axial, sagittal, and coronal) for the fine details of segmentation. Using an exported mesh model, the mandibular segmentation was then printed using rapid prototyping Fused Deposition Modeling (Figure 1). In this study, a Vantage (Stratasys Inc., Eden Prairie, MN) rapid prototyping machine was used to fabricate the 3D mandible templates. The material selected for fabricating the mandible template was polycarbonate, which can be sterilized in an autoclave and permits intraoperative use. The 3D printing time for a mandible structure was 30 to 60 minutes depending on the complexity of the defect. The cost of the in-house fabrication was $110 U.S. dollars. Mandibular non-locking reconstruction plates (2.0 mm; Stryker Inc., Kalamazoo, MI) were contoured to the stereolithographic model by the primary surgeon at least 24 hours before surgery. In 5 patients, a surgical oncology resident in training also contoured an additional mandibular plate. The plating time for both the experienced head and neck surgeon and the resident were recorded. The plates were randomly marked, sterilized, and brought to the operating room on the day of the procedure. During surgery, the time for contouring an additional intraoperative mandibular plate with identical specifications was recorded. The senior surgeon involved in the case contoured the preoperative and intraoperative plates. Two experienced head and neck surgeons, who were blinded to the plating method, visually and manually assessed both mandibular plates as they were applied to the native mandible at the time of the operation, and then completed a questionnaire. This consisted of 3 specific questions (Figure 2): (1) How well does the plate match the contour of the mandible?; (2) Evaluate the percentage of surface area contact between the plate and mandible; and (3) Which plate is the overall best match? For the first question, a 5-level Likert scale was used to evaluate the performance of the plate. The second question rated the plates for surface area contact between the reconstruction plate and the native mandible using a visual analog scale. Statistical analysis Recorded data included the time for preoperative and intraoperative contouring, and the evaluation of the performance of the different plates (preoperatively vs intraoperatively contoured). Differences were calculated using a Wilcoxon signed-rank test. Any p values <.05 were considered significant. All statistical analysis was performed using the SAS 9.3 (SAS Institute Inc., Cary, NC). HEAD & NECK DOI /HED JUNE

3 PRISMAN ET AL. FIGURE 2. Questionnaire for blinded surgical observers rating preoperative and intraoperative mandible plates. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.] RESULTS Ten patients diagnosed with SCC (n 5 5), osteoradionecrosis (n 5 2), ameloblastoma (n 5 1), adenoid cystic carcinoma (n 5 1), or mucoepidermoid cystic carcinoma (n 5 1) were included in this study. The characteristics of the patients and tumors are shown in Table 1. All patients underwent segmental mandibulectomy and subsequent reconstruction of the mandible. The choice of free tissue transfer for mandible reconstruction is outlined in Table 1. The average time to contour the preoperative mandibular plate by the experienced head and neck surgeon was seconds compared to seconds for the intraoperative plate contoured by the same surgeon (Figure 3A; p 5.83). The average time for the surgical resident to preoperatively contour was seconds. The analysis of the plate matching Likert scale (question 1) revealed a median score of 4 for both the intraoperative (interquartile range [IQR] 1) and preoperative (IQR 0) plates. There were no statistically significant differences between intraoperative and preoperative plating from the 2 observers (p 5.38). Regarding the second question, the median score and IQR of the percentage of surface contact for intraoperative and preoperative plates were both 8 (70% to 80% surface contact) and 1.25, respectively (p 5.14). For question 3, both observer 1 and observer 2 scored 5 of 9 preoperative plates as the best plate for the surgery. In only 2 cases, the observers did not show the same preference. In this case, the primary surgeon performing the reconstruction made the final selection. In both cases, this translated to the preoperative plate being selected (Figure 3B). Intraoperative plating was not available in 1 case because of the severity TABLE 1. Summary of patient and tumor characteristics. Diagnosis Defect Reconstruction ORN R, B, SH Fibula SCC R, B Fibula SCC R, B, SH Fibula SCC R, B, SH Fibula ORN R, B, SH, SH Fibula MCA R, B, SH Scapular SCC B, SH, SH Scapular AM R, B, SH Fibula ACC R, B, SH Fibula SCC R, B, SH Fibula Abbreviations: ORN, osteoradionecrosis; R, ramus; B, boday; SH, symphyseal; SCC, squamous cell carcinoma; MCA, mucoepidermoid carcinoma; AM, ameloblastoma; ACC, adenoid cystic carcinoma. 830 HEAD & NECK DOI /HED JUNE 2014

4 PREOPERATIVE PLATING IN MANDIBLE RECONSTRUCTION FIGURE 3. (A) Contouring time per case for experience surgeon (preoperative and intraoperative) and resident (5 preoperative cases). (B) Choice of best plate (preoperative or intraoperative) by 2 blinded surgical observers. of the tumor extending laterally, which prevented contouring; therefore, the preoperative plate was the only viable option for the surgery (Figure 4). In summary, in 7 cases, the preoperative plate was selected for the final reconstruction, in 2 cases the intraoperative plate was chosen, and in 1 case the preoperative plate was the only option. Cost analysis Excluding the capital costs of the 3D modeling software (MIMICS v12.0, Materialise) and the Vantage (Stratasys Inc) rapid prototyping machine, which may be accessible either in-house or at local engineering firms, the variable cost of producing a stereolithographic model was calculated as follows. The variable cost considers salaried time required to produce a virtual mandible segmentation using the 3D modeling software, in addition to rented time from a third party with a Vantage rapid prototyping machine, as well as the cost of the polycarbonate material. This translates to an average cost of $ per case to produce the 3D model. Similarly, excluding the capital costs of the operating room, the variable cost per unit time of operating at our institution was estimated at $10.05 per minute. This was derived based on the variable cost per unit time for anesthesia supplies, and support staff including nurses, operating room attendants, and clerks. The costs for surgical and anesthesia staff were not included. Assuming an average time for intraoperative contouring of 833 seconds, this corresponds to a cost of $ Subtracting the cost of fabricating a 3D model, this translates to an average additional cost of $8 per case for using a 3D template. Repeating the calculation with the average time for contouring of a resident in training, this computes to a net savings of $16.65 per case in favor of a 3D template. Alternatively, when in-house fabrication is not feasible, having a third party produce a stereolithographic model costs $ Previous publications have estimated the cost per minute of operating room time to be, on average, $66/minute. 16 Assuming an average time for intraoperative contouring of 833 seconds, this corresponds to a cost of $916. Subtracting the cost of the 3D model fabrication, this translates to an average additional cost of $184 per case for using a 3D template. Repeating the calculation with the average time for contouring by a resident in training, this computes to an average additional cost of $22 per case for using a 3D template. DISCUSSION Accurately restoring mandibular contour is essential to achieving function including occlusion, mastication, deglutition, and an adequate airway. Although advances in locking plate technology provide additional stability to the reconstructed mandible, with less emphasis on precise contouring, inaccurate contouring of the reconstruction plate has been associated with malocclusion and poor esthetics. 3 Stereolithography has been integrated into surgical planning to aid in reconstructing complex anatomic defects with evidence suggesting a reduced operative time and improved surgical planning and performance. 11 This study is the first to objectively compare preoperative mandibular reconstruction plates contoured to a 3D rapid prototyping model with the standard intraoperative mandibular plate adaptation in 10 prospective mandibular osseous free tissue transfer reconstructions. There was no FIGURE 4. (A) Case 8 of ameloblastoma with bony neoplasm extending laterally from the outer cortex. (B) The virtual image of the mandible presenting the challenge of intraoperative contouring. (C) The rapid prototype model with the bony deformity. (D) Digitally altered virtual mandible to remove bony neoplasm and allow for preoperative contouring. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.] HEAD & NECK DOI /HED JUNE

5 PRISMAN ET AL. FIGURE 5. (A) Lateral extent of squamous cell carcinoma preventing intraoperative contouring without violating tumor margins. (B) Virtual mandible with soft tissue component subtracted allowing for preoperative contouring. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.] statistically significant difference in the objective measurements used to compare the preoperative and intraoperative plates using the Likert scale and the visual analog scale to evaluate conformance and surface-area contact, respectively. However, the preoperative mandibular plate was selected in 7 of 9 cases, by the study blinded reconstructive surgeon, and was judged to be more accurate than the intraoperative plates. Furthermore, there are certain circumstances in which a preoperative plate may provide a more accurate reconstruction with significant savings in time as detailed below. Lateral extent of the neoplasm Pathologic processes that significantly deform the osseous mandible may prevent accurate intraoperative contouring of the mandibular reconstruction plates (Figure 4). Similarly, when a malignant process either extends medially to involve the mandible (ie, epithelial carcinoma), or extends laterally from the mandible (Figure 5), exposing the outer cortex of the mandible will require the surgeon to violate the tumor in order to contour an intraoperative plate. This is further complicated by the mandibular deformation caused by the disease process. Intraoperatively, this may be overcome by applying a temporary external fixation device to the proximal and distal portions of the mandible, or to place the patient in intermaxillary fixation in order to maintain the spatial orientation of the native mandible. 17 However, this is a laborious and time-consuming process. Alternatively, the surgeon may choose to resect the mandibular segment and freehand the adaptation of the reconstructive plate, the accuracy of which depends on the experience of the surgeon. The versatility of rapid prototype modeling allows for virtual subtraction of the malignant soft tissue or modification of the deformed outer cortex of the mandible, which thereby provides an epoxy model representing the osseous structure of the mandible for preoperative plate contouring (Figure 4C). This can be performed efficiently using available software. Alternatively, the model can be printed and then physically modified to allow for preoperative plate adaptation. Pathologic fracture Pathological fractures secondary to osteoradionecrosis (case 5), or malignant disease, further complicate the ability to accurately restore the premorbid mandibular contour (Figure 6). Often, an earlier pretreatment CT scan can be used to produce the virtual and stereolithographic model, which can then be used as a template to preoperatively contour a mandibular reconstruction plate. FIGURE 6. (A) Case 5 of osteoradionecrosis complicated by a pathological fracture. (B) Preoperative CT scan with an intact mandible used to create the virtual mandible shown in (C). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.] 832 HEAD & NECK DOI /HED JUNE 2014

6 PREOPERATIVE PLATING IN MANDIBLE RECONSTRUCTION Cost analysis In these 10 prospective mandibular reconstruction cases, the average additional cost of producing a 3D model translated to $184 if the primary surgeon performed the preoperative contouring, and $22 if performed by the surgical resident. However, the time to contour a reconstruction plate increases in relation to the length and complexity of the defect. Case 5 had the longest mandibular segmental resection and included 4 mandibular subsites (Table 1). Compared to the average, this case required an additional 5 minutes to contour an intraoperative mandibular reconstruction plate. In this particular case, the use of a preoperative plate would translate to a net savings of $154. Furthermore, cases that are complicated by a pathological fracture present considerable more difficulty to contour an intraoperative plate. For example, in case 9, the intraoperative contouring time was 6 minutes longer than the average, and the use of a 3D model would translate to a net savings of $220. This technique is most efficacious in cases of laterally eroding tumors that prevent the surgeon from applying a traditional intraoperative plate (case 8). In these cases, an external fixator is indicated to restore mandibular contour. However, even in expert hands, this technique requires a considerable amount of additional time, 17 and the use of a 3D model would be particularly cost effective. Conversely, the application of routine preoperative mandibular contouring, especially in cases with short or uncomplicated segmental mandibulectomy defects, is not cost effective. Limitations The cost analysis is limited by including 10 prospective cases of mandibular reconstruction, and not specifically selecting out more complicated mandibular reconstructions, such as cases 5, 8, and 9. Furthermore, the maximal benefit of this technique was not captured in this study. In case 8, the additional time required to assemble an external fixator was never measured, precisely because of the additional time it would require, and therefore the added intraoperative contouring time was not measured. In addition, the variability in operative room costs and relatively small number of cases restricts the analysis. Accuracy of the reconstructed plates was objectively measured clinically by independent and experienced blinded observers, although it was not compared with imaging studies such as Panorex or CT image overlay. The limitations of preoperative contouring include the difficulty to synchronize the position of the reconstructed plate on the preoperative stereolithographic model with the exact position on the native mandible. Several measurements from constant landmarks can be made to optimize this synchronization, including the distance of the plate to the inferior edge of the mandible, mandibular angle, or dentition. CONCLUSION Although routine application of this technology is certainly not cost effective, particularly in cases with short or uncomplicated segmental mandibulectomy defects, there are several circumstances when preoperative plating may be particularly advantageous. These situations include when the pathology either prevents exposure of the native mandible or distorts the bony architecture of the mandible (case 8), or when the mandible is complicated by a pathologic fracture (case 5). In these situations, there may be considerable decrease in operative time, which may translate to financial savings. These advantages are likely to be enhanced in the hands of less experienced surgeons. Furthermore, our data indicate that plates adapted preoperatively are more optimal in terms of restoring the contour of the mandible based on the assessment of 2 independent blinded observers. Further clinical study is required to determine whether this translates to improved mandibular form and function postoperatively. REFERENCES 1. Bak M, Jacobson AS, Buchbinder D, Urken ML. Contemporary reconstruction of the mandible. Oral Oncol 2010;46: Genden E, Haughey BH. Mandibular reconstruction by vascularized free tissue transfer. Am J Otolaryngol 1996;17: Mehta RP, Deschler DG. Mandibular reconstruction in 2004: an analysis of different techniques. Curr Opin Otolaryngol Head Neck Surg 2004;112: Hidalgo DA. Fibula free flap: a new method of mandible reconstruction. Plast Reconstr Surg 1989;84: Urken ML, Buchbinder D, Costantino PD, et al. Oromandibular reconstruction using microvascular composite flaps: report of 210 cases. Arch Otolaryngol Head Neck Surg 1998;124: Cordeiro PG, Disa JJ, Hidalgo DA, Hu QY. Reconstruction of the mandible with osseous free flaps: a 10-year experience with 150 consecutive patients. Plast Reconstr Surg 1999;104: Ro EY, Ridge JA, Topham NS. Using stereolithographic models to plan mandibular reconstruction for advanced oral cavity cancer. Laryngoscope 2007;117: Mankovich NJ, Cheeseman AM, Stoker NG. The display of threedimensional anatomy with stereolithographic models. J Digit Imaging 1990;3: Brown GA, Milner B, Firoozbakhsh K. Application of computer-generated stereolithography and interpositioning template in acetabular fractures: a report of eight cases. J Orthop Trauma 2002;16: Essig H, Rana M, Kokemueller H, et al. Pre-operative planning for mandibular reconstruction a full digital planning workflow resulting in a patient specific reconstruction. Head Neck Oncol 2011;3: Jacobs PF. Rapid prototyping & manufacturing: fundamentals of stereolithography. New York, NY: McGraw Hill; Bouyssie JF, Bouyssie S, Sharrock P, Duran D. Stereolithographic models derived from x-ray computed tomography. Reproduction accuracy. Surg Radiol Anat 1997;19: Erben C, Vitt KD, Wulf J. First statistical analysis of data collected in the Phidias validation study of stereolithography models. Phidias Newsletter 2000;5: Toro C, Robiony M, Costa F, Zerman N, Politi M. Feasibility of preoperative planning using anatomical facsimile models for mandibular reconstruction. Head Face Med 2007;3: Urken ML, Weinberg H, Vickery C, Buchbinder D, Lawson W, Biller HF. Oromandibular reconstruction using microvascular composite free flaps. Report of 71 cases and a new classification scheme for bony, soft-tissue, and neurologic defects. Arch Otolaryngol Head Neck Surg 1991;117: Shippert RD. A study of time-dependent operating room fees and how to save $ by using time-saving products. Am J Cosmetic Surg 2005; 22: Ung F, Rocco JW, Deschler DG. Temporary intraoperative external fixation in mandibular reconstruction. Laryngoscope 2002;112: HEAD & NECK DOI /HED JUNE