[Skip to Navigation]
Sign In
Figure 1.  Maxillectomy for Malignant Tumor With Model
Maxillectomy for Malignant Tumor With Model

The inset shows the flap.

Figure 2.  Fibula Cutting Guides and Model of a Planned Reconstruction
Fibula Cutting Guides and Model of a Planned Reconstruction

The inset shows the cutting guide.

Figure 3.  Base View of Patient From Figure 1 Showing Projection of Maxillary Reconstruction
Base View of Patient From Figure 1 Showing Projection of Maxillary Reconstruction
Table 1.  Patient Characteristics Between VSP and Pre-VSP Groups
Patient Characteristics Between VSP and Pre-VSP Groups
Table 2.  Surgical Characteristics Between VSP and Pre-VSP Groups
Surgical Characteristics Between VSP and Pre-VSP Groups
1.
Myers  EN, Snyderman  CH.  Operative Otolaryngology E-Book: Head and Neck Surgery. Elsevier; 2017.
2.
Vural  E, Hanna  E.  Extended lateral rhinotomy incision for total maxillectomy.   Otolaryngol Head Neck Surg. 2000;123(4):512-513. doi:10.1067/mhn.2000.108277 PubMedGoogle ScholarCrossref
3.
Liu  Z, Yu  H, Wang  D, Wang  J, Sun  X, Liu  J.  Combined transoral and endoscopic approach for total maxillectomy: a pioneering report.   J Neurol Surg B Skull Base. 2013;74(3):160-165. doi:10.1055/s-0033-1338260 PubMedGoogle ScholarCrossref
4.
Rivera-Serrano  CM, Terre-Falcon  R, Duvvuri  U.  Combined approach for extensive maxillectomy: technique and cadaveric dissection.   Am J Otolaryngol. 2011;32(5):417-421. doi:10.1016/j.amjoto.2010.07.023 PubMedGoogle ScholarCrossref
5.
Futran  ND.  Primary reconstruction of the maxilla following maxillectomy with or without sacrifice of the orbit.   J Oral Maxillofac Surg. 2005;63(12):1765-1769. doi:10.1016/j.joms.2005.08.014 PubMedGoogle ScholarCrossref
6.
Genden  EM, Okay  D, Stepp  MT,  et al.  Comparison of functional and quality-of-life outcomes in patients with and without palatomaxillary reconstruction: a preliminary report.   Arch Otolaryngol Head Neck Surg. 2003;129(7):775-780. doi:10.1001/archotol.129.7.775 PubMedGoogle ScholarCrossref
7.
Zhang  WB, Wang  Y, Liu  XJ,  et al.  Reconstruction of maxillary defects with free fibula flap assisted by computer techniques.   J Craniomaxillofac Surg. 2015;43(5):630-636. doi:10.1016/j.jcms.2015.03.007 PubMedGoogle ScholarCrossref
8.
Kosutic  D, Uglesic  V, Knezevic  P, Milenovic  A, Virag  M.  Latissimus dorsi–scapula free flap for reconstruction of defects following radical maxillectomy with orbital exenteration.   J Plast Reconstr Aesthet Surg. 2008;61(6):620-627. doi:10.1016/j.bjps.2007.11.004 PubMedGoogle ScholarCrossref
9.
Kääriäinen  M, Kuuskeri  M, Gremoutis  G, Kuokkanen  H, Miettinen  A, Laranne  J.  Utilization of three-dimensional computer-aided preoperative virtual planning and manufacturing in maxillary and mandibular reconstruction with a microvascular fibula flap.   J Reconstr Microsurg. 2016;32(2):137-141. doi:10.1055/s-0035-1563396PubMedGoogle Scholar
10.
Pang  JH, Brooke  S, Kubik  MW,  et al.  Staged reconstruction (delayed-immediate) of the maxillectomy defect using CAD/CAM technology.   J Reconstr Microsurg. 2018;34(3):193-199. doi:10.1055/s-0037-1607394 PubMedGoogle ScholarCrossref
11.
Modest  MC, Moore  EJ, Abel  KMV,  et al.  Scapular flap for maxillectomy defect reconstruction and preliminary results using three-dimensional modeling.   Laryngoscope. 2017;127(1):E8-E14. doi:10.1002/lary.26351 PubMedGoogle ScholarCrossref
12.
Cordeiro  PG, Santamaria  E.  A classification system and algorithm for reconstruction of maxillectomy and midfacial defects.   Plast Reconstr Surg. 2000;105(7):2331-2346. doi:10.1097/00006534-200006000-00004 PubMedGoogle ScholarCrossref
13.
Okay  DJD, Genden  E, Buchbinder  D, Urken  M.  Prosthodontic guidelines for surgical reconstruction of the maxilla: a classification system of defects.   J Prosthet Dent. 2001;86(4):352-363. doi:10.1067/mpr.2001.119524 PubMedGoogle ScholarCrossref
14.
Brown  JS, Shaw  RJ.  Reconstruction of the maxilla and midface: introducing a new classification.   Lancet Oncol. 2010;11(10):1001-1008. doi:10.1016/S1470-2045(10)70113-3 PubMedGoogle ScholarCrossref
15.
Lund  V, Howard  DJ, Wei  WI.  Endoscopic resection of malignant tumors of the nose and sinuses.   Am J Rhinol. 2007;21(1):89-94. doi:10.2500/ajr.2007.21.2957 PubMedGoogle ScholarCrossref
16.
Parida  PK, Gupta  AK.  Medial maxillectomy: a comparative study as a surgical procedure.   Otolaryngol Head Neck Surg. 2008;138(2):192-199. doi:10.1016/j.otohns.2007.10.018 PubMedGoogle ScholarCrossref
17.
Vergez  S, Martin-Dupont  N, Lepage  B, De Bonnecaze  G, Decotte  A, Serrano  E.  Endoscopic vs transfacial resection of sinonasal adenocarcinomas.   Otolaryngol Head Neck Surg. 2012;146(5):848-853. doi:10.1177/0194599811434903 PubMedGoogle ScholarCrossref
18.
Foster  RD, Anthony  JP, Singer  MI, Kaplan  MJ, Pogrel  MA, Mathes  SJ.  Microsurgical reconstruction of the midface.   Arch Surg. 1996;131(9):960-965. doi:10.1001/archsurg.1996.01430210058011PubMedGoogle ScholarCrossref
19.
Triana  RJ  Jr, Uglesic  V, Virag  M,  et al.  Microvascular free flap reconstructive options in patients with partial and total maxillectomy defects.   Arch Facial Plast Surg. 2000;2(2):91-101. doi:10.1001/archfaci.2.2.91 PubMedGoogle ScholarCrossref
20.
Rodriguez  ED, Bluebond-Langner  R, Park  JE, Manson  PN.  Preservation of contour in periorbital and midfacial craniofacial microsurgery: reconstruction of the soft-tissue elements and skeletal buttresses.   Plast Reconstr Surg. 2008;121(5):1738-1747. doi:10.1097/PRS.0b013e31816b13e1 PubMedGoogle ScholarCrossref
21.
Modabber  A, Gerressen  M, Ayoub  N,  et al.  Computer-assisted zygoma reconstruction with vascularized iliac crest bone graft.   Int J Med Robot. 2013;9(4):497-502. doi:10.1002/rcs.1557 PubMedGoogle ScholarCrossref
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words
    Original Investigation
    April 1, 2021

    Association of Virtual Surgical Planning With External Incisions in Complex Maxillectomy Reconstruction

    Author Affiliations
    • 1Department of Otolaryngology–Head and Neck Surgery, Mayo Clinic, Rochester, Minnesota
    • 2Department of Otolaryngology–Head and Neck Surgery, Mayo Clinic, Jacksonville, Florida
    • 3Department of Otorhinolaryngology, Mayo Clinic, Rochester, Minnesota
    • 4Department of Radiology, Mayo Clinic, Rochester, Minnesota
    • 5Department of Anatomic Modeling, Mayo Clinic, Rochester, Minnesota
    JAMA Otolaryngol Head Neck Surg. 2021;147(6):526-531. doi:10.1001/jamaoto.2021.0251
    Key Points

    Question  Does the use of virtual surgical planning (VSP) and 3-dimensional (3-D) modeling decrease the need for external incisions in maxillectomy reconstruction?

    Findings  This retrospective cohort study of 38 patients who underwent maxillectomies requiring microvascular reconstruction compares one group of patients from an era prior to VSP with one group from an era using VSP and 3-D modeling to design, contour, and inset the reconstructive flap. The patients in the VSP group had a very low rate of lateral rhinotomy despite having equally extensive defects as the patients in the pre-VSP group.

    Meaning  This study suggests that VSP may help with planning, execution, placement, and fixation for complex free flap maxillectomy defect reconstruction through a transoral minimally invasive approach.

    Abstract

    Importance  Maxillectomy can commonly be performed through a transoral approach, but maxillectomy defect reconstruction can be difficult to precisely design, contour, and inset through this approach.

    Objective  To evaluate whether the use of virtual surgical planning (VSP) and 3-dimensional (3-D) modeling is associated with a decrease in the requirement of lateral rhinotomy (LR) for patients undergoing total and partial maxillectomy reconstruction.

    Design, Setting, and Participants  This retrospective cohort study was conducted among patients undergoing subtotal or total maxillectomy with microvascular free flap reconstruction with or without VSP and 3-D modeling at a single tertiary care academic medical center between January 1, 2008, and October 3, 2019.

    Interventions  Maxillectomy and free flap reconstruction with or without VSP.

    Main Outcomes and Measures  Necessity of LR or other external incision for contouring, placement, and fixation of reconstruction as well as surgical complications.

    Results  Fifteen patients (12 men [80%]; mean age, 64 years) underwent maxillectomy with free flap reconstruction without VSP. Eight patients (53%) in this group underwent total maxillectomy, and 4 patients in this group (27%) underwent partial maxillectomy. Twenty-three patients (18 men [78%]; mean age, 58 years) underwent maxillectomy with free flap reconstruction and VSP and 3-D modeling. Twelve of these patients (52%) underwent total maxillectomy, and 11 (48%) underwent partial maxillectomy. Lateral rhinotomy was necessary for 1 patient (4%) in the VSP group vs 12 patients (80%; 95% CI, 54%-98%) in the pre-VSP group. There were no LR complications in the VSP group vs 6 in the pre-VSP group. Among both groups, 14 patients underwent fibula free flap, 22 patients underwent subscapular system free flap, and 2 patients underwent cutaneous or osteocutaneous radial forearm free flap. There were no flap failures in the LR group and 1 flap failure in the group without LR.

    Conclusions and Relevance  This cohort study suggests that the use of VSP and 3-D modeling for maxillectomy reconstruction is associated the a decrease in the need for external incisions without compromising reconstructive flap utility.

    Introduction

    The maxilla is a complex facial bone that provides supporting structure to the orbital contents, anchors the ipsilateral upper dentition, and separates the oral cavity from the nasal tract, while providing cosmetic and functional structure to the midface. The maxilla may develop benign and malignant tumors of the oral cavity and nasal cavity lining, paranasal sinuses, minor salivary glands, dentition, bone, and orbital contents. These processes frequently require total or partial maxillectomy. Maxillectomy has traditionally been performed through external incisions via lateral rhinotomy (LR) or Weber-Fergusson modifications that allow access to the zygomaticomaxillary suture, orbital rim, and floor and sometimes beyond.1,2

    Many surgeons have evolved away from external incisions secondary to facial disfigurement, scar contracture, wound breakdown, and fistula formation.3 The transoral and endoscopic techniques that have been developed allow for adequate visualization of the maxillary boundaries, to allow for even total maxillectomy to be performed without external incisions.4 When combined with transconjunctival incisions and upper lid incisions, even total maxillectomy with or without orbital exenteration can be performed without LR incisions.3,4

    Reconstruction of maxillectomy defects with microvascular free tissue transfer demonstrated a major improvement in alleviating the functional and cosmetic morbidity of maxillectomy.5,6 Placing adequate vascularized soft tissue into the defect helps restore facial contour and isolate the oral and nasal cavities, and vascularized bone restores cheek and orbital support and allows for the restoration of dentition. However, microvascular flaps for maxillary reconstruction are bulky and dimensionally complex. Multiple different tissue types must be precisely placed and affixed to the surrounding tissue, and positioning them properly, even with the adequate exposure provided through ample external facial incisions, is often challenging.7,8 Use of virtual surgical planning (VSP) has improved the design, execution, and results of maxillary reconstruction.7,9,10 The precision for the complex osteotomies that are often required, the positioning and insetting of the tissue in precise alignment, and the planning for osseointegrated implants have all shown improvement with the use of preoperative computer-aided design.11 Still, many surgeons, even when using VSP, still perform maxillectomy and free flap reconstruction through an LR incision, and many series describing VSP for this process still demonstrate patients with external incisions.7,8 The primary reason for the continued use of LR incisions for maxillectomy and free flap reconstruction is that even a precisely planned reconstruction is still challenging to inset and fixate through transoral incisions.

    We have found that the combined use of VSP and 3-dimensional (3-D) printing allows for precise plate contouring and fixation of the maxillary reconstruction ex vivo on the 3-D model. With the use of VSP and the production of cutting guides for the maxilla and the reconstructive flap, the fit of the reconstructive flap into the defect is more precise. Plates can be precisely contoured to the model, or custom plates can be designed with the ideal position of screws. The flap can then be inset into the model on the back table to ensure a precise fit and then inset transorally into the defect without the added exposure provided by transfacial incisions. We sought to evaluate the feasibility of this process and the improvement in morbidity by comparing a cohort of patients undergoing maxillectomy and reconstruction before the implementation of VSP (the pre-VSP group) with a cohort of patients undergoing maxillectomy and reconstruction after the implementation of VSP (the VSP group).

    Methods
    Study Population

    We performed a retrospective medical record review to compare techniques and outcomes for all patients who underwent partial or total maxillectomy with microvascular reconstruction between January 1, 2008, and October 3, 2019. The class of maxillectomy and defect, the type of flap used, the use of LR, and the complications with the rhinotomy were compared between a group of 23 patients who underwent preoperative VSP and an earlier group of 15 patients who did undergo VSP. This study was approved by the Mayo Clinic institutional review board. Patients provided written consent.

    For 12 of the 15 patients in the pre-VSP group, surgery was performed through intraoral incisions and LR or Weber-Fergusson incisions. Maxillectomy was performed with frozen section clearance of the margins in all cases regardless of approach, and reconstruction was performed simultaneously with a microvascular free flap dictated by the extent of the defect, era, and surgeon preference.

    In the VSP group, high-resolution computed tomography (CT) was performed for both the maxilla and the donor site for the free flap. The surgical planning consisted of a meeting with the radiologist to review the image and plan the resection of the tumor and the ensuing defect. Decisions were made regarding use of CT technology, dual-energy CT and computed tomography angiography or multienergy scanning, unique metal reduction artifact algorithms, the possible need for magnetic resonance imaging and magnetic resonance angiography, and patient positioning mimicking that in the operating suite. The surgeon and radiologist meet and review the specific case along with reconstruction goals. During the modeling stage, the radiologist transfers the images to the anatomical modeling laboratory on site and imports them into Mimics software (Materialise). The imaging data are segmented, allowing for different anatomical structures such as bone, vessels, tumor, brain, and soft tissues of the neck to be separated. A virtual 3-D anatomical model is created containing several parts. This model is exported into a stereolithography file that then undergoes any needed postprocessing in 3-Matic software (Materialise). The models are then exported as new stereolithography files into Objet Studio (Stratasys Ltd) to be placed on a build tray. Life-size models are printed using varied liquid photopolymers that allow multicolor, multiple flexibility, and clear versions using a PolyJet Connex 350 3-D printer (Stratasys Ltd). Cutting guides were similarly produced to allow for the precise placement and orientation of the osteotomies on the maxilla as well as on the bone of the reconstructive flap. These cutting guides are an integral part of the planning and execution to allow for the avoidance of LR. With VSP, the osteotomies of the defect can align precisely with the osteotomies of the reconstructive segments and allow for the precise placement of the reconstruction without having to modify the cuts in situ.

    During the surgical stage, the sterilized model and the cutting guides were brought into the operating room. The maxilla was exposed by transoral incisions and a preseptal transconjunctival incision if necessary. Canthotomy was avoided to prevent subsequent scarring and malpositioning of the lower lid. The cutting guides were affixed to the maxilla with temporary screws, and the maxillary cuts were made with a sagittal saw and osteotomes. Cuts were made on the fibula or scapula with similar use of the cutting guides. A 3-D model of the defect was also produced by making identical cuts on the maxillary model, if a model incorporating the defect had not been printed (Figure 1).

    Cutting guides were printed to allow for precise cutting and contouring of the reconstruction to fit the defect (Figure 2). A critical next step in the process was contouring the standard reconstructive plates and affixing the reconstructive flap to the model on the back table (Figure 1 and Figure 3). This step allowed for confirmation of the precise placement and fit of the flap to the defect. Once this step was performed, the flap could be taken free from the model and placed precisely into the defect on the patient and fixated without the need for wide exposure of the surrounding tissue of the maxilla. Because the location of the plates and screw holes had been planned on the model, fixation was performed with miniplate screws placed directly with straight drills and a screwdriver through the transoral incision and transconjunctival incisions. A transfacial percutaneous screw placement was rarely necessary.

    Statistical Analysis

    Statistical analysis was performed to determine the 95% CIs of the difference in the number of class I or II and class III or IV maxillectomy defects between the groups, as well as to determine the same 95% CIs in the difference in the number of rhinotomies performed in each group.

    Results

    Fifteen patients (12 men [80%]; mean age, 64 years) underwent maxillectomy with free flap reconstruction without VSP between 2008 and 2014, when VSP was seldom being used by our group for maxillectomy planning, and 23 patients (18 men [78%]; mean age, 58 years) underwent maxillectomy with VSP and 3-D modeling of the resection and defect (Table 1). The extent of the maxillectomies and the types of reconstruction are listed in Table 1. Twelve of the patients (80%) without VSP underwent an LR for completion of the maxillectomy and placement of the free flap. In the pre-VSP group, 8 patients (53%) underwent total maxillectomy, and 4 patients (27%) underwent partial maxillectomy. There were 9 class III or IV defects and 6 class I or II defects among the patients in the pre-VSP group. There was no significant difference in the classes of maxillectomies performed between the VSP group (48%) and the pre-VSP group (53%) (difference, 5% [95% CI, −38% to 27%]).

    In the VSP group, 11 (48%) had a class III or IV maxillectomy defect, and 12 (52%) had a class I or II defect. Twelve of these patients (52%) underwent total maxillectomy, and 11 (48%) underwent partial maxillectomy. One patient in the VSP group (4%) received an LR secondary to an extensive and bulky tumor that involved a large portion of the zygoma that was difficult to access without a transfacial incision. Twelve patients (80%) in the pre-VSP group received an LR.

    Among both groups, 15 patients underwent a fibular free flap for maxillectomy reconstruction, 21 patients underwent a subscapular system free flap, and 2 patients underwent an osteocutaneous radial forearm free flap (Table 2). Two patients underwent vein grafting to allow for tension-free vascular anastomosis. There were no flap failures in the LR group. One patient in the group without a LR developed arterial insufficiency of the flap on postoperative day 3, and the flap could not be revascularized and salvaged.

    Six (50%) of the 12 patients in the pre-VSP group who underwent LR developed complications from their LR incision. These complications included delayed LR dehiscence with fistula formation in 4 patients who underwent radiotherapy as part of their treatment and LR contracture with lip and nasal distortion in 2 patients. No patients in the VSP group experienced LR complications. There were no flap failures in the LR group, and 1 flap failure in the group without LR.

    Discussion

    The 2 maxillae have been described as the most important bones of the facial skeleton because of their multifaceted roles in facial support, cosmesis, mastication, and respiration.12 Maxillectomy, either total or subtotal, is often required for benign and malignant tumor eradication. Traditionally, maxillectomy is performed through an LR incision; this incision is sometimes combined with sublabial incisions, infraorbital or transconjunctival incisions, lip incisions, and even temporal or preauricular incisions to gain exposure to the maxilla and its junctions with the zygoma and sphenoid bones. The types of maxillectomy are nearly as varied as the tumors that occur in this bone, and the extent of maxillectomy is dictated by the extent of the neoplastic process. Maxillectomy and maxillectomy defects have been classified by many authors, and these classifications are typically based on both the medial-to-lateral and inferior-to-superior extent of resection.12-14 The widely used classification system of Brown and Shaw14 divides defects vertically into class I (inferior maxillectomy with no oral-antral fistula), class II (inferior maxillectomy with oral or antral fistula), class III (total maxillectomy with orbital rim and floor resection), and class IV (extended total maxillectomy with orbital resection), and horizontally into subclasses a (central palate), b (hemipalate) c (onto contralateral palate), and d (total hard palate). Class I or II defects can typically be resected through transoral incisions and often reconstructed with prosthetics or simple local tissue flaps.13

    As the extent of the maxillectomy increases, the need for more lateral and superior exposure is required. Although external incisions such as LR and orbital extensions satisfy these requirements, there is a cost to pay for this exposure. Even well-placed and well-executed rhinotomy and lip split incisions leave a lasting scar that can easily be visualized after the operation. Most patients requiring extensive maxillectomy also undergo postoperative radiotherapy and chemotherapy. Even with free tissue reconstruction, the entire extent of the transfacial rhinotomy incision is not supported on its undersurface with bone. This lack of support, coupled with soft tissue atrophy and chronic inflammatory changes in the underlying and radiated intranasal cavity, can be associated with late contracture, wound breakdown, and nasal facial fistula formation.3 Through several decades of evolution of craniofacial techniques and prefabrication of plating systems, modern maxillectomy has largely eliminated external incisions.15,16 Following endoscopic medial maxillectomy, transoral and endoscopic approaches have been applied to more extensive procedures than just medial or inferior maxilla resection.17 Visualization of the zygomaticomaxillary junction and the zygomaticosphenoid communication can typically be adequately achieved without LR and external incisions, and if simply removing the maxilla was the only endeavor, most patients could be spared LR, even in total maxillectomy.

    Resecting the maxilla through a transoral incision is common for most experienced head and neck surgeons and is often not the most challenging portion of maxillectomy. Because of the vital functional and aesthetic roles of this complex 3-D bone, the crux of maxillectomy is in reconstructing the more extensive defects. Isolating the mouth from the nasal cavity, reestablishing the dentoalveolar complex, supporting the orbit, and restoring cheek contour can ideally be accomplished only with free tissue transfer of bone and soft tissue. However, the free tissue transfers that can accomplish these goals, withstand postoperative radiotherapy, and provide adequate soft tissue fullness are often bulky during their initial harvest and inset.5,18 The major series detailing reconstruction of advanced maxillectomies demonstrate the use of LRs and extended LRs with wide exposure purely to adequately inset and internally fixate the flap.18-20 Reports of extended maxillectomy with only transoral and transnasal incisions admit that these approaches increase the difficulty of intraoperative reconstruction because of the more restricted space.3,4 Often, the patients in these reports of maxillectomy with only transoral incisions undergo reconstruction by obturators without bone and soft tissue reconstruction.3 However, obturators, although acceptable for smaller inferior maxillectomy defects, are inadequate to achieve maximal rehabilitation of complex and extensive maxillary defects.6 Our finding has been that even total maxillectomy can be well performed with clear margins through transoral incisions and orbital incisions. Still, the inset of the flap without VSP and modeling is challenging without external incisions. Even in cases of subtotal maxillectomy, prior to VSP, external incisions were often necessary to achieve adequate placement and fixation of the reconstruction.

    Virtual surgical planning has been shown to improve the efficiency and precision of maxillary reconstruction by free tissue transfer.9,11 Surgical planning and the use of intraoperative models and cutting guides have been widely adopted for these reasons. Even with the acknowledged benefits associated with using only mucosal incisions for maxillectomy, most authors advocating the use of VSP still place the reconstruction through an LR.7,21 This study demonstrates the utility of combining minimally invasive approaches to the maxilla, VSP, medical modeling, and the use of medical modeling in the operating room to eliminate the LR and the potential complications of fistula formation and unsightly facial scarring. The technique requires some surgical experience because placement of the bulky reconstruction and fixation through limited incisions can be challenging. It also requires some additional time, resources, and expense that may not be widely available at every facility. Finally, problems may arise when the extent of the tumor is unrecognized preoperatively and deviations from the operative plan are necessary. Adjustments to the resection can be made in the operating room, and those adjustments can be translated to the model and the osteotomies of the reconstructive flap. Cutting guides often must be modified to be low profile and to contour precisely to the bone to be used through a transoral incision. When the cutting guides are abandoned, however, the reconstruction may not fit precisely. Some groups have adopted delayed reconstruction with VSP to remedy the uncertainty of adequate margins in the operating room.10

    Maxillectomy reconstruction with vascularized tissue entails placing the reconstruction in the midface and anastomosing the vessels somewhere in the neck. The distance between the flap vessels and the recipient vessels can exceed the length of the vascular pedicle. A vein graft is often used to add extra vascular length, with one reconstructive series using a vein graft in 9 of 12 defects involving the palate.19 Vein grafts were used in only 2 of the 38 patients in this series. High reliance on bone-bearing flaps with well-planned pedicles (scapular tip flap based on angular artery and well-planned distal fibula free flap) and maximizing donor vessel length during the neck dissection may partially account for this finding, but VSP also helps plan the orientation of the flap and the location of the vascular anastomosis. This preoperative evaluation and planning may decrease the uncertainty of vessel reach during the operation. Avoidance of LR requires the creation of a subcutaneous cheek tunnel for passage of the flap vessels into the neck. This tunnel must be of ample size to avoid compression of the pedicle in the immediate postoperative period when edema of the cheek is greatest. Although the rate of flap failure in this series is low (3%), the single flap failure was associated with vascular compromise secondary to complications in the cheek tunnel that caused vascular compression. We create a subcutaneous cheek tunnel that will accommodate several fingers, that will carefully pass the vessels through this tunnel under direct vision to ensure correct orientation, and that will now place a Penrose drain through the tunnel to avoid this problem.

    Limitations

    This study has some limitations, including its retrospective and epochal design, and the inherent limitations of selection bias that result from that type of reporting. Maxillectomy requiring free flap reconstruction is not an exceedingly common event, and the lessons learned are largely from experience and adjustments in technique, as evidenced by all the major series reported being retrospective studies. It may be that our decreased use of LR was in part associated with increased experience as much with the use of VSP and 3-D modeling. In addition, it may be that the decreased complications in the VSP group were due to improved technique as well as the elimination of LR. Finally, the comparison groups were small, making matching of the tumors and flaps in the 2 groups challenging. Still, we think that the comparison between the pre-VSP era and the era of 3-D model printing demonstrates some valuable lessons that can improve both cosmesis and function.

    Conclusions

    Preoperative VSP, use of 3-D printed models and cutting guides, and intraoperative fixation of the reconstruction to the model may eliminate the use of LR for even extensive maxillectomy reconstruction. Avoidance of LR may decrease the rate of maxillectomy-related complications.

    Back to top
    Article Information

    Accepted for Publication: February 4, 2021.

    Published Online: April 1, 2021. doi:10.1001/jamaoto.2021.0251

    Corresponding Author: Eric J. Moore, MD, Department of Otolaryngology–Head and Neck Surgery, Mayo Clinic, 200 First St SW, Rochester, MN 55905 (moore.eric@mayo.edu).

    Author Contributions: Dr Moore had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

    Concept and design: E. J. Moore, Van Abel, Martin, Morris, Alexander.

    Acquisition, analysis, or interpretation of data: E. J. Moore, Price, Van Abel, Janus, E. T. Moore, Martin, Morris.

    Drafting of the manuscript: E. J. Moore, E. T. Moore, Morris.

    Critical revision of the manuscript for important intellectual content: E. J. Moore, Price, Van Abel, Janus, Martin, Morris, Alexander.

    Statistical analysis: E. T. Moore.

    Administrative, technical, or material support: Janus, Martin, Morris, Alexander.

    Supervision: E. J. Moore, Price, Van Abel, Morris.

    Conflict of Interest Disclosures: None reported.

    Additional Contributions: We thank the patient in Figure 3 for granting permission to publish this information.

    References
    1.
    Myers  EN, Snyderman  CH.  Operative Otolaryngology E-Book: Head and Neck Surgery. Elsevier; 2017.
    2.
    Vural  E, Hanna  E.  Extended lateral rhinotomy incision for total maxillectomy.   Otolaryngol Head Neck Surg. 2000;123(4):512-513. doi:10.1067/mhn.2000.108277 PubMedGoogle ScholarCrossref
    3.
    Liu  Z, Yu  H, Wang  D, Wang  J, Sun  X, Liu  J.  Combined transoral and endoscopic approach for total maxillectomy: a pioneering report.   J Neurol Surg B Skull Base. 2013;74(3):160-165. doi:10.1055/s-0033-1338260 PubMedGoogle ScholarCrossref
    4.
    Rivera-Serrano  CM, Terre-Falcon  R, Duvvuri  U.  Combined approach for extensive maxillectomy: technique and cadaveric dissection.   Am J Otolaryngol. 2011;32(5):417-421. doi:10.1016/j.amjoto.2010.07.023 PubMedGoogle ScholarCrossref
    5.
    Futran  ND.  Primary reconstruction of the maxilla following maxillectomy with or without sacrifice of the orbit.   J Oral Maxillofac Surg. 2005;63(12):1765-1769. doi:10.1016/j.joms.2005.08.014 PubMedGoogle ScholarCrossref
    6.
    Genden  EM, Okay  D, Stepp  MT,  et al.  Comparison of functional and quality-of-life outcomes in patients with and without palatomaxillary reconstruction: a preliminary report.   Arch Otolaryngol Head Neck Surg. 2003;129(7):775-780. doi:10.1001/archotol.129.7.775 PubMedGoogle ScholarCrossref
    7.
    Zhang  WB, Wang  Y, Liu  XJ,  et al.  Reconstruction of maxillary defects with free fibula flap assisted by computer techniques.   J Craniomaxillofac Surg. 2015;43(5):630-636. doi:10.1016/j.jcms.2015.03.007 PubMedGoogle ScholarCrossref
    8.
    Kosutic  D, Uglesic  V, Knezevic  P, Milenovic  A, Virag  M.  Latissimus dorsi–scapula free flap for reconstruction of defects following radical maxillectomy with orbital exenteration.   J Plast Reconstr Aesthet Surg. 2008;61(6):620-627. doi:10.1016/j.bjps.2007.11.004 PubMedGoogle ScholarCrossref
    9.
    Kääriäinen  M, Kuuskeri  M, Gremoutis  G, Kuokkanen  H, Miettinen  A, Laranne  J.  Utilization of three-dimensional computer-aided preoperative virtual planning and manufacturing in maxillary and mandibular reconstruction with a microvascular fibula flap.   J Reconstr Microsurg. 2016;32(2):137-141. doi:10.1055/s-0035-1563396PubMedGoogle Scholar
    10.
    Pang  JH, Brooke  S, Kubik  MW,  et al.  Staged reconstruction (delayed-immediate) of the maxillectomy defect using CAD/CAM technology.   J Reconstr Microsurg. 2018;34(3):193-199. doi:10.1055/s-0037-1607394 PubMedGoogle ScholarCrossref
    11.
    Modest  MC, Moore  EJ, Abel  KMV,  et al.  Scapular flap for maxillectomy defect reconstruction and preliminary results using three-dimensional modeling.   Laryngoscope. 2017;127(1):E8-E14. doi:10.1002/lary.26351 PubMedGoogle ScholarCrossref
    12.
    Cordeiro  PG, Santamaria  E.  A classification system and algorithm for reconstruction of maxillectomy and midfacial defects.   Plast Reconstr Surg. 2000;105(7):2331-2346. doi:10.1097/00006534-200006000-00004 PubMedGoogle ScholarCrossref
    13.
    Okay  DJD, Genden  E, Buchbinder  D, Urken  M.  Prosthodontic guidelines for surgical reconstruction of the maxilla: a classification system of defects.   J Prosthet Dent. 2001;86(4):352-363. doi:10.1067/mpr.2001.119524 PubMedGoogle ScholarCrossref
    14.
    Brown  JS, Shaw  RJ.  Reconstruction of the maxilla and midface: introducing a new classification.   Lancet Oncol. 2010;11(10):1001-1008. doi:10.1016/S1470-2045(10)70113-3 PubMedGoogle ScholarCrossref
    15.
    Lund  V, Howard  DJ, Wei  WI.  Endoscopic resection of malignant tumors of the nose and sinuses.   Am J Rhinol. 2007;21(1):89-94. doi:10.2500/ajr.2007.21.2957 PubMedGoogle ScholarCrossref
    16.
    Parida  PK, Gupta  AK.  Medial maxillectomy: a comparative study as a surgical procedure.   Otolaryngol Head Neck Surg. 2008;138(2):192-199. doi:10.1016/j.otohns.2007.10.018 PubMedGoogle ScholarCrossref
    17.
    Vergez  S, Martin-Dupont  N, Lepage  B, De Bonnecaze  G, Decotte  A, Serrano  E.  Endoscopic vs transfacial resection of sinonasal adenocarcinomas.   Otolaryngol Head Neck Surg. 2012;146(5):848-853. doi:10.1177/0194599811434903 PubMedGoogle ScholarCrossref
    18.
    Foster  RD, Anthony  JP, Singer  MI, Kaplan  MJ, Pogrel  MA, Mathes  SJ.  Microsurgical reconstruction of the midface.   Arch Surg. 1996;131(9):960-965. doi:10.1001/archsurg.1996.01430210058011PubMedGoogle ScholarCrossref
    19.
    Triana  RJ  Jr, Uglesic  V, Virag  M,  et al.  Microvascular free flap reconstructive options in patients with partial and total maxillectomy defects.   Arch Facial Plast Surg. 2000;2(2):91-101. doi:10.1001/archfaci.2.2.91 PubMedGoogle ScholarCrossref
    20.
    Rodriguez  ED, Bluebond-Langner  R, Park  JE, Manson  PN.  Preservation of contour in periorbital and midfacial craniofacial microsurgery: reconstruction of the soft-tissue elements and skeletal buttresses.   Plast Reconstr Surg. 2008;121(5):1738-1747. doi:10.1097/PRS.0b013e31816b13e1 PubMedGoogle ScholarCrossref
    21.
    Modabber  A, Gerressen  M, Ayoub  N,  et al.  Computer-assisted zygoma reconstruction with vascularized iliac crest bone graft.   Int J Med Robot. 2013;9(4):497-502. doi:10.1002/rcs.1557 PubMedGoogle ScholarCrossref
    ×