30 Reconstruction of the Maxilla
Reconstruction of the maxilla continues to pose a challenge to the head and neck oncologic surgeon due to long-term effects of adjuvant therapies, longer patient survivorship, and an increase in patient expectations regarding form and function. Multiple classification schemes have been proposed, with the aim of classifying anatomic and functional defects and guiding reconstruction strategies. Examples of defects and reconstructive approaches are described in a defect-oriented manner, and postoperative care is reviewed.
The maxilla plays an integral role both in aesthesis and function of the oral cavity. It provides structural support to oral and orbital cavities, and dominates projection of the midface. Functionally, defects of the maxilla can impact speech, swallowing, and mastication. Goals of reconstruction are multifactorial, and should include: oronasal separation, maintenance of occlusion, restoration of the orbital floor, re-establishment of the load-bearing vertical buttress, patency of the nasolacrimal duct and nasal airway, support and suspension of adynamic facial soft tissue, and adequate cosmesis. 1
Rehabilitation of both structural and functional aspects of maxillary defects can be accomplished by any of several prosthetic options or by surgical reconstruction. In the past, prosthetic rehabilitation was the predominant choice in most cases. However, in the last three decades, the advantages of surgical reconstruction have been realized through the evolution of new techniques and surgical design technology. In spite of this, appropriate selection of patients for either prosthetic or reconstructive methods remains an important point of discussion.
In light of the notorious intricacy associated with maxillary defects and their reconstruction, multiple complex classification systems have been developed to facilitate logical reconstruction strategies. However, no single classification scheme has been adopted above all others, prompting ambiguous interpretation in reconstructive results. In addition, the controversy surrounding optimal orbitomaxillary and nasomaxillary reconstruction techniques remains, particularly with respect to prevention or management of postoperative sequelae. This heterogeneity in reconstructive planning has the unfortunate effect of lack of clinical outcomes with sufficient data to drive the creation of well-accepted guidelines in this area. That being said, there is a plethora of literature supporting common principles in maxillary reconstruction, in order to achieve acceptable outcomes. This chapter will identify and discuss appropriate reconstruction strategies for the patient with a maxillary defect, perioperative considerations, and prevention and management of surgical sequelae.
30.2 Diagnosis and Evaluation
A thorough approach to assessment in maxillary reconstruction involves a complete history, physical examination, and review of biopsy and imaging results. All components warrant consideration, as they may provide insight on the most appropriate reconstructive strategy. For example, a more aggressive pathology may warrant a more aggressive resection margin, or delayed reconstruction (e.g., sinonasal undifferentiated carcinoma or sarcoma, respectively). A physical examination indicating orbital rim or floor involvement may direct the use of a different free tissue option (e.g., proptosis due to orbital floor invasion). Advanced tumor stage evident on imaging studies may portend the need for adjuvant radiation therapy, which may change the extent of soft tissue incorporated in the reconstruction.
Appropriate imaging will have significant impact on treatment planning, oncologic resection, and therefore reconstructive planning. In general, computed tomography (CT) imaging is considered a gold standard for evaluation of bony destruction. Specific structures to examine include the orbital floor, nasolacrimal duct, ethmoid sinus, posterior maxillary wall, and extent of hard palate erosion. Magnetic resonance imaging (MRI) can be considered for evaluation of the integrity of specific structures, such as the subcutaneous fat, infraorbital nerve, periorbita, extraconal fat, extraocular muscles, and optic nerve. Positron emission tomography-computed tomography (PET/CT) technology should be utilized in cases with high risk of regional or distant metastasis.
30.3 Anatomic Considerations/Relevant Anatomy
Structural support of the midface involves both vertical and horizontal buttresses. Vertical buttresses include the nasomaxillary, zygomaticomaxillary, and pterygomaxillary buttresses. 2 These supports are vertically suspended from a thick frontal bar, and are reinforced by horizontal zygomatic, alveolar, and palatine buttresses. Regardless of the defect, regeneration of these buttresses is important to endure occlusal load, and restore facial form. In general, restoration of two of the three vertical buttresses will provide more optimal results with regards to form and function, but certainly a wide variability of reconstructive options exist with acceptable outcomes. If the goal of the reconstructive plan is eventual occlusal restoration with dental implants, then the vertical buttress construct becomes more critical in terms of the quantity and location of bone, and the increased occlusal load associated with these types of reconstructions.
While the vertical buttresses provide support for mastication, the horizontal buttresses provide support for the globe, determine the height of the occlusal plane, and contribute to aesthetic qualities of the midface. Certainly from the functional standpoint, the correct position of the globe is critical, and therefore restoration of osseous structure after ablation of the orbital rim/floor is optimal. Additional alloplastic reconstruction (i.e., titanium mesh) can supplement this osseous reconstruction. It should be noted that vascularized osseous reconstruction will help withstand the effects of radiation therapy, and provide more predictable results, when compared to non-vascularized tissue, or alloplastic-only constructs. From a functional standpoint, provided the oral cavity defect is closed, reconstruction of the osseous structure of the maxilla is critical only in cases where dental implants are planned. Careful attention to the position of osseous tissue is vital to treatment plan success in these situations.
30.4 Classification of Maxillary Defects
Classification systems can allow a defect to be succinctly deconstructed into its component parts, concentrate discussion around problem areas, focus on reconstructive options, and compare short- and long-term results. The intricacy of the anatomy and consequent defects is mirrored in the variety and number of published classification systems, none of which is universally accepted. 3 – 6
Cordeiro et al 6 described a four-part classification scheme distinguishing defects based on an increasing surface area-to-volume (SA:V) ratio. Type I defects involve one or two walls of the maxilla, leaving the palate and orbital floor intact. Type II defects, or subtotal maxillectomy defects, involve resection of the maxilla, but leave the orbital floor intact. The authors subdivides Type II defects into IIA defects (involving less than 50% of the transverse palate) and Type IIB defects (involving greater than 50% of the transverse palate). Type III defects, or total maxillectomy defects, involve the entire wall of the maxilla and are subdivided into Type IIIA and Type IIIB, depending on whether the orbital contents are resected. Type IV orbitomaxillary defects involve resection of the entire orbital cavity contents, leaving dura and brain exposed. The authors produced a detailed algorithm for the type of free flap reconstruction according to defect type, 7 , 8 yet a significant critique of the classification scheme is the failure to consider functional dental restoration.
Okay et al 4 described a classification that illustrated maxillary defects in both horizontal and vertical planes, based on biomechanical stability for obturator retention. This was the first classification system to consider the status of the zygomatic arch and the orbital floor, and provide insight into dental restoration. In this classification, Class IA defects involve any portion of the hard palate save for the tooth-bearing alveolus. Class IB defects involve any portion of the maxilla with preservation of both canine teeth. Defects are classified as Class II if they engage only one canine or if they involve less than 50% of the hard palate. Class III defects involve resection of both canines or greater than 50% of the hard palate. Subclasses “F” and “Z” pertain to involvement of the orbital floor and zygomatic arch. The hallmark of the Okay classification is the focus on the setting necessary to accommodate effective prosthetic reconstruction. For example, Class IA and Class IB defects can be addressed with soft tissue alone, whereas Class III defects require osseous substrate to retain an obturator.
From the reconstructive standpoint, Brown and colleagues 5 perhaps introduced the most widely accepted classification which described defects according to its vertical and horizontal components. In this system the vertical component (Classes I-IV) is coupled with an appraisal of the orbit, and a horizontal component (Classes A-C). Class I defects involve resection of the alveolar ridge or palate with no oroantral defect; Class II defects involve an alveolar and antral defect but do not include the orbital floor or rim; Class III defects involve an alveolar and antral defect including the orbital floor with or without periorbita or skull base; Class IV defects involve an alveolar and antral defect including the orbital floor and orbital cavity contents with or without skull base involvement. Horizontal defects can remain on one side of the midline (Class A), cross the midline or involve the septum (Class B), or involve the entire palate (Class C). Brown and Shaw 9 expanded the existing classification system in 2010 with Class V orbitomaxillary and Class VI nasomaxillary defects. In addition, the dentoalveolar or functional side of the defect was reorganized to the following: Class A defects involving an isolated palate defect; Class B with a defect less than or equal to half of the unilateral maxilla; Class C with a defect less than or equal to half bilateral or transverse anterior maxilla; and Class D defect comprising greater than half of the maxilla. The authors reviewed 147 cases of reconstructed midface and maxillary oncologic defects reconstructed with both forearm (osteocutaneous and fasciocutaneous) and thoracodorsal angular artery (TDAA) latissimus perforator free flaps, reaching the conclusion that an obturator can be considered for Classes I, II, A, and B defects not involving the orbit, and that free flap reconstruction is recommended for more extensive defects such as Classes II, D, III-VI.
While no single classification offers a superior option for reconstructive planning, familiarity with the above three classification systems can avoid common reconstructive errors, and provide improved outcomes by preoperatively identifying critical considerations when formulating a treatment plan.
30.5 Virtual Surgical Planning (VSP)
Three-dimensional (3D) printers and medical modeling have become increasingly available, allowing for accurate measurement of both vertical and horizontal buttresses for adequate projection and improved symmetry. Previous studies have boasted both shorter ischemia and total operative time, 10 , 11 as well as increased accuracy and reproducibility of planned osteotomies in maxillary reconstruction. 12 Despite an increase in incurred surgical costs 11 and preoperative planning, this technology has been widely utilized in most academic centers for complex cases, and does provide an improved platform for treatment planning. It is also an excellent tool for learning and anticipating issues related to the reconstructive plan preoperatively.
30.6 Reconstructive Options
Goals of reconstruction of the maxilla include oronasal separation, restoration of structural integrity and midfacial contour, and restitution of a scaffold for suspension of soft tissues. In?short, speech, swallowing, and mastication must be re-established?durably.
In the past, maxillectomy defects, regardless of the extent, were reconstructed with prosthetic obturation. The advantages of this option included partial restoration of facial contour as well as immediate dental restoration in some cases. Other advantages include accurate customization to existing anatomy and shorter operative times. However, disadvantages to this approach frequently included inadequate prosthesis-tissue seal, need for manual dexterity to manipulate the prosthesis, and chronic irritation and pain. 13 In addition, the often subsequent need for obturator revision due to maxillary cavity contraction negatively impacted patient satisfaction. 14 A study by Moreno et al 15 compared 73 obturated and 40 reconstructed patients and demonstrated that reconstruction provided better speech and swallowing outcomes, particularly in patients with defects involving a large palatal component. Even in small to medium defects, the superiority of free flaps in improving quality of life was shown by Genden et al. 16 Brown and Shaw, 9 using the 2010 classification modification, recommended obturation for Classes I, II, A, and B defects only. However, even with small defects, patient dissatisfaction with obturation has spurred the development of multiple local or pedicle flaps as substitutes.
30.6.1 Partial Hard Palate Defects
With smaller maxillary defects, cosmetic and functional deficits are less common, and the focus becomes expeditious return to function, while imparting minimal morbidity to the patient. This focus can be realized with local or regional flaps. However, careful patient selection is of the utmost importance in ensuring a good outcome, as these techniques are limited by pedicle length and soft tissue bulk.
Brown Class Ia defects, or partial defects of the hard palate, may be addressed with local, pedicled, or free flaps, depending on the size and location. Local and pedicled options may include a palatal island flap, buccal fat pad flap, temporalis muscle flap, or facial artery musculomucosal (FAMM) flaps.
The palatal island flap, based on the greater palatine vessels, may be used to resurface defects up to 15 cm2. 17 This is commonly used for small Brown Class Ia defects, and can restore the oronasal barrier in through-and-through defects due to the thick mucosa of the hard palate. It should be noted that patients should expect moderate donor site pain during re-epithelialization which occurs within 8-12 weeks.
The buccal fat pad is situated between the masseter and buccinator muscles, and consists of a main body and four extensions: buccal, pterygoid, superficial temporal, and deep temporal. The parotid duct and zygomatic and buccal branches of the facial nerve cross the lateral surface of the pad. Its rich vascularity from the maxillary, superficial temporal, and facial arteries allows it to be used as an axial-pattern pedicled flap. Stuzin et al, who noted a mean weight and volume of 9.3 g and 9.6 mL, respectively, characterized the buccal fat pad flap in terms of its size. 18 As such, only defects less than 12 cm2 are amenable to reconstruction with this method. 19 The utility of this flap is primarily in small posterior maxillectomy defects with no intraoral communication requiring coverage for a denture. Amin et al 20 described their experience in 24 patients with a variety of maxillary defects, wherein the buccal fat pad flap was covered with a split-thickness skin graft and an acrylic plate. Complete healing and partial dehiscence were observed within 4 weeks postoperatively in 18 and 6 patients, respectively. Of the 10 patients that underwent radiation therapy, none experienced wound breakdown.
The FAMM flap 21 can be used for lateral hard palate defects in edentulous patients. The FAMM flap is an axial flap based on the terminal branches of the facial artery as it courses deep to the buccal mucosa, lateral to the buccinator muscle, and medial to the muscles of facial expression. As such, the flap itself is composed of buccal mucosa and buccinator muscle. The flap should be designed with a wide base, and the pedicle can be identified using a Doppler probe. Careful attention to the location of the pedicle in dentate patients can prevent vascular compression and flap compromise.
Temporalis muscle flaps can be used when the maxillary defect involves more than just the palate, but only a small amount of soft tissue bulk is needed for bony coverage. For patients in whom a free flap may not be a reasonable option, a temporalis flap, with or without vascularized bone, may be used. Specifically, vascularized calvarial temporalis flaps can be used to re-contour the orbital floor or superior or anterior wall of the maxilla due to the natural curvature of the bone. 22 However, it should be noted that this requires tunneling the flap for a significant distance and/or partially resecting the zygomatic arch, necessitating identification of the frontal branch of the facial nerve. In addition, temporal hollowing, with the resultant aesthetic deformity, and postoperative trismus are donor site considerations.
For large Brown Class Ia defects spanning the entire non-tooth-bearing region of the palate, a fasciocutaneous radial forearm free flap (RFFF) can be considered. 14 This option boasts of a long, vascular pedicle, reliable anatomy, and a thin, pliable skin paddle to create a steadfast oronasal partition that will not hinder denture retention. Several other options, including myofascial free tissue transfer (i.e., serratus or vastus lateralis), may be considered in these situations.