HEAD AND NECK
CHAPTER 37 MANDIBLE RECONSTRUCTION
JOSEPH J. DISA AND EVAN MATROS
The mandible contributes to airway stability, is important in speech, deglutition, and mastication, and largely determines the shape of the lower face. Consequently, functional and aesthetic goals are equally important considerations in mandible reconstruction. Specific functional goals include preservation of temporomandibular joint function with maximal opening ability and maintenance of occlusion. In more severe cases in which many teeth are missing, restoration of normal interarch distance and alignment is critical for the facilitation of subsequent dental rehabilitation. Key aesthetic goals include symmetry, preservation of lower facial height and anterior chin projection, and correction of submandibular soft-tissue neck defects.
The vast majority of segmental mandible defects are caused by cancer. Squamous cell (epidermoid) carcinoma is the etiology in the majority of cases with the mandible commonly invaded by adjacent tongue or floor of mouth tumors. Osteogenic sarcoma is the second most common cause of segmental mandibular defects resulting from cancer resection and the most common primary bone tumor. Mucoepidermoid carcinoma, adenoid cystic carcinoma, leiomyosarcoma, and fibrous histiocytoma are examples of other tumors. A small number of segmental mandibular defects result from extensive benign cystic or fibrotic bone disease. Gunshot wounds are the most common traumatic cause, but their number is small compared with tumors. Segmental loss because of infection is rare, but can occur after complications of mandible fractures.
Mandible defects requiring reconstruction are sometimes caused by bone loss alone (e.g., osteoradionecrosis). However, the majority of defects usually include adjacent intraoral soft tissue as well as submandibular soft tissue. Some bone defects include external skin loss instead of mucosa, and the most complex include bone, mucosa, and skin.
Two classification schemes have been proposed for mandible defects. The most practical describes bone loss in terms of central segments (designated C and defined as lying between the two canine teeth), lateral segments (L), and hemimandible segments (H).1 Hemimandible and lateral segments are similar except that the former includes the condyle, whereas lateral segments do not. A defect commonly is a combination of more than one segment, for example, LC, HC, or LCL. Although this description may appear tedious, it is actually useful as a common language to standardize the variable reconstructive problems posed by these entities (Figure 37.1).
Mandible reconstruction can be accomplished by a variety of means, including nonvascularized bone grafts, metal plates, pedicled flaps, and free flaps. Nonvascularized grafts, such as an iliac crest segment, can be used for a short bone gap (<3 cm) in a setting of benign disease. This is a rare application. Although conceptually and technically simple, this method relies on creeping substitution for long-term mandible stability.
Pedicled flaps include the trapezius and pectoralis osteomyocutaneous flaps. The primary attraction of these donor sites is that they lie adjacent to the head, thus permitting their movement into this area without disconnecting their blood supply. Although this is an attractive concept, there are several important drawbacks. First, use of these flaps enlarges the size of the primary wound considerably compared with harvesting tissue from a distant donor site. This increases the potential for morbidity at the site of the reconstruction. More importantly, a significant portion of flap volume is used just to reach the recipient site. The distal portion of the flap, which is used for the actual reconstruction, often has a marginal blood supply and is at risk for ischemic necrosis. Perhaps the greatest limitation of these flaps is that they do not provide enough tissue in the proper configuration to be useful. The bone available with the pectoralis major muscle (rib) and the trapezius (spine of the scapula) is limited compared with free-flap alternatives. In addition, the osseous components of these flaps have poor blood supply, derived only from the periosteum, resulting in high rates of non-union and limiting the surgeon’s ability to perform shaping osteotomies. Although the pectoralis has been used to reconstruct the anterior mandible and the trapezius to reconstruct the lateral mandible, these flaps are generally not recommended as primary methods of mandible reconstruction.
Prosthetic mandible replacement has evolved as an alternative method of reconstruction that still has legitimate, but limited, application. Mesh trays made of Dacron or metal were introduced in the 1970s as scaffolds that were filled with bone graft chips and used for segmental bone defects. Long-term follow-up has shown this method to be ineffective. Problems with extrusion and bone graft dissolution commonly occurred. Metal reconstruction plates developed as a result of orthopedic hardware advances in other areas. These plates are available today in a variety of lengths and styles.
Metal reconstruction plates offer advantages of decreased operating time and avoidance of a bone graft donor site. They have important disadvantages: risk of exposure or infection; risk of plate fracture; preclusion of dental reconstruction; and a thin shape that does not provide adequate bulk for reconstruction. These disadvantages are particularly problematic in the setting of radiation therapy. Another important drawback is the functional limitation seen with the use of metal plates for hemimandible defects that include the condyle. The prosthetic condyle is a poor substitute for the native structure. The long-term effects of a metal condyle in the native glenoid fossa are unknown, and occlusion is often poorly maintained with a metal plate that includes a condyle. As a result of these disadvantages, the first choice for reconstruction of segmental mandibular defects is with vascularized bone flaps. However, prosthetic reconstruction may be useful in scenarios when bone reconstruction is not possible such as extensive oncologic resection, absence of suitable bone flaps, or presence of significant medical comorbidities.
When reconstruction of segmental mandibular defects is performed with reconstruction plates, adequate soft-tissue coverage is critical to prevent plate extrusion. The pectoralis major myocutaneous flap is commonly used for this purpose; however, plate exposure still occurs, particularly with anterior reconstructions in which the tension on the flap is greatest. One in three plate reconstructions fails when a pedicled flap is used for coverage.
The most reliable soft-tissue coverage for a reconstruction plate is provided by a free flap, which provides abundant tissue and can be inset without tension. The vertical rectus flap, forearm flap, or the anterior lateral thigh (ALT) flaps are commonly used for this purpose and flap selection is guided by the volume of soft tissues required for reconstruction. The sole advantage of this approach (reconstruction plate plus forearm flap) is that it is somewhat quicker to perform than an osteocutaneous free flap. The disadvantage is that it combines the worst features of its two component parts: the risks of infection and exposure from a foreign body and the risk of free-flap failure. The combination of a soft-tissue free flap and a reconstruction plate is probably best reserved for lateral defects in those patients who are poor candidates for an osteocutaneous free flap. Elderly patients with multiple medical comorbidities may benefit from a shorter operative procedure and are more likely to accept the permanent dentition defect than are young patients.2
FIGURE 37.1. Examples of mandibular defects. A. A typical lateral defect of the L type. Note how the intermaxillary fixation maintains occlusion for flap insetting. B. An LCL anterior defect showing the mobile lateral mandibular segments, submental soft-tissue defect, and the lack of reference points to guide accurate flap insetting.
Osteocutaneous free-flap reconstruction is often the most effective method of mandible repair. These flaps have both soft tissue and bone components, which are available in an optimal configuration for solving specific composite tissue problems. This technique is ultimately dependent on the integrity of the microvascular anastomoses for success. Fortunately, the favored donor sites all have excellent flap pedicle qualities (vessel diameter and length), and the head and neck area generally has excellent recipient vessels available, despite previous surgical treatment and radiation. Free-flap survival rates are approximately 95%.3
FREE-FLAP DONOR-SITE SELECTION
Since early in the development of free-flap mandible reconstruction, there have been multiple donor sites from which to choose. Rib, metatarsal, and ilium were among the first flaps developed.4 The ilium had been the most popular of the three owing to its abundant bone, which even resembles a hemimandible when harvested in a particular way and was the workhorse for the first decade of free-flap mandible reconstruction. Further evolution has led to the development of the radius, scapula, and fibula donor sites.5 These additional options have increased the flexibility and precision of the technique as the specific assets and limitations of each donor site have become clear.
A review of 155 free-flap mandible reconstructions at Memorial Sloan-Kettering Cancer Center has shown that the fibula is currently the donor site of choice for most patients (Table 37.1).6 The radius, the scapula, and the ilium (to a diminishing extent) are better choices in a few specific settings. Each has unique advantages and disadvantages. A comparison of the donor sites is helpful in selecting the proper flap for a particular problem (Table 37.2). Some have better bone qualities, some have better skin, and some have significant disadvantages that make them seldom the flap of choice despite their good qualities (Figure 37.2).
The ilium has abundant bone but has a predetermined shape that makes flap shaping inherently less precise than other donor-site options. It may be useful in some hemimandible reconstructions because its shape most closely resembles this portion of the mandible. The ilium is said to have a segmental blood supply from the deep circumflex iliac artery, although this is debatable on a practical level. This type of vascular anatomy is preferred because it allows segmental osteotomies with survival of each portion of the flap. Long ilium flaps, however, tend to have less robust, even marginal, circulation at the distal end of a multiply osteotomized flap.
The skin island available with the ilium does not have a reliable circulation in many patients. In addition, the soft-tissue component of the flap is often bulky and lacks mobility with respect to the bone. This makes insetting difficult and limits usefulness of the soft-tissue component of the flap. Some authors propose including a portion of the internal oblique muscle with the flap as an alternative source of soft tissue. The muscle is covered with a skin graft when used inside the oral cavity.
Closure of the ilium donor site is arduous and there is a possibility of hernia formation or late attenuation of the lateral abdominal wall. This donor site is painful and limits early mobilization of the patient. Splitting the ilium and leaving its outer rim intact is proposed as a means of facilitating the closure process, but this makes flap harvest more tedious.
FIGURE 37.2. Free-flap donor sites for mandible reconstruction. A. Scapula. B. Ilium. C. Radius. D. Fibula. Note the relative amounts of skin, the relationship of the pedicle to the bone, and the bone configurations available.
Today, the indications for use of the ilium are limited. Perhaps the best indication is a short lateral or hemimandible segment not requiring mucosal lining replacement. The problems with the ilium described above often make other donor sites preferable, even for this type of defect (Figure 37.3).
The radius has the best quality skin island compared with other donor-site alternatives. It is thin, pliable, and abundant. The vascular pedicle is also ideal, with long, large-diameter vessels capable of reaching the opposite side of the neck for difficult recipient vessel problems. The bone, in contrast, is the worst compared with other choices. The radius must be carefully split during harvest to prevent postoperative fracture at the donor site, and some authors have advocated primary bone grafting and plating of the radius donor site to decrease the incidence of this complication. Length is generally limited to a segment located between the insertion of the pronator teres and the brachioradialis muscles (approximately 10 cm), although some authors describe taking longer pieces. The bone thickness is marginal for later placement of osseointegrated implants for dental rehabilitation.
There is insufficient soft tissue available with this flap to provide the necessary bulk to fill submandibular neck defects. The donor-site appearance is often poor postoperatively owing to a need for skin graft closure and the additional proximal forearm scar necessary for obtaining adequate pedicle length (Figure 37.4).
The best indication for a radius free flap is a bone defect that is limited to the ramus and the proximal body with a large associated intraoral soft-tissue defect. The split radius is adequate to restore mandibular continuity. Dental rehabilitation is usually superfluous posteriorly, and so the thin nature of the bone is not a factor. The cheek soft tissues are thick and maintain facial contour despite this flap’s inherent lack of bulk. The skin island is ideal for resurfacing a large posterior mucosal defect. Reconstruction of most anterior defects is a relative contraindication to the use of the radius flap because adequate soft tissue and bone volume are essential in this area for the best functional and aesthetic reconstruction.
The scapula offers the greatest amount of soft tissue compared with other donor sites. It is possible to include a skin island as long as 30 cm and to include the entire latissimus dorsi muscle if needed. The skin island is somewhat thick compared with the forearm donor site. A useful feature of this flap is that the bone and the soft-tissue components (skin and latissimus dorsi muscle) are independent of each other except for a common vascular pedicle. Up to 14 cm of bone is available from the lateral scapula. The bone does not have a segmental blood supply; therefore, multiple osteotomies can be hazardous to the viability of portions of the flap. The blood supply of the proximal scapular segment is derived from branches of the circumflex scapular artery, while the distal most portion is supplied by branches arising from the thoracodorsal vessels. The primary disadvantage of this flap is its donor-site location, requiring delay in flap harvest until after the resection. The patient typically has to be repositioned several times throughout the operation. Shoulder function is sometimes compromised following scapula flap harvest. Patients can exhibit weakness and decreased range of shoulder motion (Figure 37.5). In addition, the pedicle length is somewhat short (6 to 8 cm), precluding access to the contralateral neck vessels.
FIGURE 37.3. A. Design of osteocutaneous iliac crest flap. Note the design of the lateral mandible reconstruction and position of the skin island. B. Panorex of iliac crest in place. C. Postoperative appearance.
The best indication for a scapula free flap in mandible reconstruction is a bone gap associated with a large soft- tissue defect. This applies most to patients who require simultaneous intraoral and external soft-tissue replacement. The priority in these cases of advanced local disease is to achieve uncomplicated primary wound healing. The precision of the bony reconstruction is often a secondary concern. The result is compromised whenever a skin island is placed externally on the face owing to color mismatch and partial facial nerve paralysis associated with the defect. Although rarely indicated, a combined scapula and latissimus dorsi flap is useful for large defects, including those resulting from a radical neck dissection. The latissimus dorsi restores neck contour and protects the exposed vessels. This can actually produce an elegant result, but constitutes a massive effort when performed in conjunction with a mandible reconstruction. The scapula flap is also a reasonable choice for straight lateral segments when the fibula flap is not available.
FIGURE 37.4. Radial forearm osteocutaneous flap (A and B). A. Design of an osteocutaneous forearm flap. B. Note the limited diameter of bone relative to available skin.
FIGURE 37.5. Resection and reconstruction using a scapula osteocutaneous flap (A and B). A. Planned resection of cheek and mandible. B. Full-thickness defect, including skin, mandible, mucosa, and associated soft tissues. C. Osteocutaneous scapula flap. D. Postoperative appearance.
The fibula donor site has many advantages.5 The bone is available with enough length to reconstruct any mandible defect. The straight quality of the bone with adequate height and thickness constitutes the ideal bone stock for precisely shaping a mandible flap. Unlike the ilium, there are no nuances of shape that limit the flap contouring process. Also unlike other donor sites, the periosteal blood supply is functionally of a segmental type. Osteotomies can be planned wherever necessary and can be placed as close as 1 cm apart without concern for bone viability. The vascular pedicle has sufficient length and is of large diameter. The flexor hallucis longus muscle located along the posterior border of the bone is ideal for filling adjacent soft-tissue defects in the submandibular portion of the upper neck. The skin island available with the fibula is reliable in approximately 91% of patients. It is thicker than the forearm skin, but thinner than the scapula skin. A large skin paddle can be harvested for complex defects, but the donor site will require skin graft closure. Of all potential donor sites, the fibula is the most convenient because it is located farthest from the head and neck area.
The main disadvantage is the unreliability of the skin blood supply in 9% of cases,5 although the incidence of this complication has been debated. There are no reliable preoperative tests to identify the patients who are at risk for an inadequate skin blood supply. Despite this problem, it is uncommon to be faced with a need for skin and to have none available. The forearm or anterolateral thigh flaps can be used as a second free flap, should the need for extra skin unexpectedly arise, combining the best features of both flaps. This practice is actually preferable to using a single flap, such as the scapula, in which neither the bone nor the skin is in the ideal configuration.
The fibula is indicated for all anterior defects and most lateral defects. It is the flap of choice for the majority of mandible defects except for a few special situations in which the radius or scapula may constitute a better choice.6 It is particularly well suited to anterior defects because the skin island can be used to reconstruct the floor of the mouth. The flexor hallucis longus muscle is perfectly situated to fill in the dead space within the anterior arch of the mandible and restore upper neck contour.
Depending upon the defect location, soft-tissue flaps serve one of two purposes in mandible reconstruction. They can be used either as an adjunct to an osteocutaneous flap with insufficient skin or in lieu of osseous reconstruction altogether.
For anterior mandible defects, osseous flaps are mandatory to provide sufficient rigidity to prevent the Andy Gump deformity, with airway collapse, drooling, and facial distortion. A simultaneous soft-tissue flap may be indicated when the fibula skin island is inadequate to close an anterior intraoral or external soft-tissue defect. The most commonly reported soft-tissue flap used in combination with the fibula is the radial forearm flap, but the anterolateral thigh, rectus abdominis, and pectoralis major flaps can be used as well.
On the contrary, for lateral defects, soft-tissue flaps alone are useful in specific reconstructive scenarios. These flaps reliably fill the dead space, ensure a watertight closure, and prevent tethering of intraoral structures. The remaining native mandible provides functional support and maintains facial proportions. For example, defects involving both the lateral mandible and extensive intraoral soft tissues located in difficult locations such as the retromolar trigone, palate, and oropharynx are more easily closed with a soft-tissue flap. Furthermore, large composite defects can be closed with a single-folded double-island soft-tissue flap such as an ALT or vertical rectus abdominis myocutaneous, rather than with the fibula flap skin island that has limited size and rotational degrees of freedom around the intermuscular septum. Soft-tissue free-flap reconstructions are also useful in cases where the temporomandibular joint has been resected, creating a large defect in the posterior skull base. This approach may also be adopted in patients who are suboptimal candidates for two-flap surgery, such as those with significant comorbidities or advanced disease and in those lacking suitable bone donor sites. Regardless of the reconstructive scenario, primary healing of the incisions is a priority in the head and neck cancer patients, many of whom need to receive adjuvant therapy. Although like tissue is not being replaced with like, retrospective series demonstrate equivalent aesthetic and functional outcomes of soft-tissue flaps compared with osseous flaps, with the exception of an inability to place osseointegrated dental implants.7,8
A reconstructive algorithm for the mandible using osseous and/or soft-tissue flaps is presented in Figure 37.6.
The most serious postoperative problems in patients undergoing mandible reconstruction are cardiopulmonary. Pneumonia, arrhythmias, and myocardial infarction are life threatening problems for which this patient population is at risk due to prior smoking. As a result, the patient’s medical history should be reviewed and medical clearance obtained for preoperative optimization and assessment of surgical risk.
Preoperative consultation with the dental service is valuable in the management of mandible reconstruction patients. Intermaxillary fixation, intraoperative tooth extraction, custom fabrication of various splints, and other ancillary procedures are best performed with forethought and the help of interested colleagues. This also sets the stage for the patient’s postoperative dental rehabilitation with either conventional dentures or osseointegrated implant technology.
Two specific preoperative studies contribute to improved aesthetic results.9 A 1:1 computed tomography (CT) or magnetic resonance imaging scan of the mandible taken in the transverse plane at a level just below the tooth roots is the basis for fabrication of a template showing the full-size shape of the mandible. A lateral cephalogram will allow fabrication of a second template showing the shape of the mandible in the sagittal plane. These images can be transferred to acrylic plastic or used directly as radiographic film cut-outs to assist in the flap-shaping process. Together with the surgical specimen as a reference, this permits the bone to be completely shaped at the donor site, while the vascular pedicle remains intact (Figure 37.7) and contributes to improved accuracy in reconstruction.
FIGURE 37.6. Reconstructive algorithm for mandibular defects. Depending upon bony defect location and quantity of soft-tissue resection, an osseous flap, soft-tissue flap, or a combination of both can be used to reconstruct the mandible.
FIGURE 37.7. Preoperative planning (A and B). Templates are fashioned from tracings of a lateral cephalogram (left) and a 1:1 scale axial plane CT scan of the mandible (right). These templates serve as valuable references during flap shaping.
Technology from other fields is finding new applications in mandible reconstruction with goals of improving accuracy, precision, and efficiency.10 Through the use of computer-aided design and manufacturing (CAD–CAM), reconstructive and ablative surgeons can perform virtual operations in advance of the actual surgery with the assistance of engineers. The first step is osteotomy design, including those made by the ablative team, based on high-resolution three-dimensional CT images of the native mandible (Figure 37.8). Next, virtual osteotomies that optimize bone apposition are planned using CT images of the patient’s fibula to reconstruct the contour of the excised mandible. Using computer-aided manufacturing, cutting jigs for the native mandible and fibula, and a contoured reconstruction plate are generated for the operating field. Intraoperatively, the cutting jig is shifted along the fibula to optimize skin island position and pedicle length. Closing wedge osteotomies are made through cutting slots without additional measurement. The remainder of the operation proceeds in a standard fashion. Advantages of such technology are evident in cases where tumor distortion of the mandible precludes accurate specimen measurement or for reconstruction of anterior defects where anatomic orientation of ramus fragments cannot be reliably maintained.
Routine use of preoperative imaging is not necessary for the fibula donor site in the majority of patients. The main indications for preoperative imaging are signs and symptoms of peripheral vascular disease or an abnormal pedal pulse examination. Improvements in noninvasive imaging techniques, such as CT angiography or magnetic resonance angiography, have obviated the need for angiograms. Furthermore, because fibula reconstruction has been successfully performed even with overt peroneal artery atherosclerotic disease, its presence does not necessarily rule out the use of this donor site.11 Reconstructive surgeons should also be aware of the rare vascular anomaly of a dominant peroneal artery (peroneal arteria magna), in which harvest of the fibula flap could lead to leg ischemia. The precise incidence of this congenital variation is not well established.
With rare exception, all patients should have a tracheostomy for safety. It is often possible to begin donor-site dissection at the same time as the ablative portion of the procedure. If there is significant doubt as to the extent of the disease, it is better to wait until the situation is clarified before beginning.
Flap shaping can be performed while ablation is in progress with the aid of the templates described previously. The surgical specimen is also a valuable visual aid. Measurements of total graft length can be obtained, as well as measurements to identify where osteotomies are best made to duplicate mandible shape.9 Subtle nuances in shape can be appreciated by direct examination of the specimen. Typical locations or fibula osteotomies include the parasymphyseal, midbody, and mandibular angle regions.
Bony fixation can be accomplished with the use of miniplate fixation.12 This method is efficient, safe, and strong. Preformed reconstruction plates have been preferred by others, but this method does not allow subtle nuances of mandible shape to show when a bulky plate is applied to the outer surface of the flap. When hardware requires removal to facilitate osseointegrated dental implants, the mini-plate technique limits the exposure necessary by allowing for removal of only the hardware in the region of the implants. Other methods, such as interosseous wires, do not provide enough resistance to torsional stress in a multiply osteotomized bone flap. Intermaxillary fixation is used only as an adjunctive form of fixation. Its primary role is to maintain occlusion during the insetting of lateral flaps (Figure 37.1A). External fixators, previously popular for stabilizing the lateral segments when reconstruction is deferred, are rarely indicated in mandible reconstruction.
Lateral defects differ from anterior defects in terms of the approach to shaping the flap. In the case of the fibula, ilium, and scapula, the angle of the mandible is generally planned where the vascular pedicle enters the bone (Figure 37.9). This provides maximum pedicle length to reach the recipient vessels in the neck. This is where the first osteotomy is made in the bone, with the second osteotomy made to form the curve in the midbody. The ramus height is determined by measurements taken from the specimen. The condyle can often be harvested from the surgical specimen and then mounted directly onto the flap. Frozen-section examination of bone scrapings are performed to rule out tumor in the condyle. It must not be used if doubt exists. This method is better than the alternative of transecting the ramus high and leaving the condyle in situ. It is difficult to fix the fibula flap to the condyle in this situation.13
FIGURE 37.8. Application of computer-aided design and manufacturing in mandible reconstruction. Virtual osteotomies of the native mandible are planned by the ablative surgeon on high-resolution three-dimensional CT images (left). Virtual osteotomies of the patient’s fibula are planned to reconstruct missing portions of the mandible (center). Cutting jigs for the native mandible and fibula are generated using computer-aided manufacturing. Closing wedge fibula osteotomies are made intraoperatively without further measurement. Proximal and distal cutting slots on both the mandible and fibula jigs match exactly (right). (Image provided by Medical Modeling, Golden, CO.)
FIGURE 37.9. Osteotomies and fixation of typical lateral and central flaps (A and B). A. A typical lateral flap (with transplanted condyle). B. Typical anterior flap. Note the location of the pedicle in each.
Anterior flap shaping begins by planning the location of the central segment so as to maximize flap pedicle length (Figure 37.9). The central segment usually measures 2 cm. An osteotomy is made on each end in two planes. The body segments curve away from the central segment in both a posterior and superior direction. The body segments are usually of unequal length. It is important to use the transverse template to accurately reproduce the splay of the body segments away from the central segment and each other. As in the case of lateral flap, it is best to leave the ends of the flap long and make the final osteotomies, which determine overall flap fit, at the time of flap insetting.
The recipient site is prepared prior to dividing the flap pedicle. The ends of the mandible segments are dissected in a subperiosteal plane for approximately 2 cm to allow room for mini-plate fixation. Recipient vessels are identified, and the intraoral wound closed as much as possible. Intermaxillary fixation is placed in the case of a lateral flap, and the remaining portion of the mandible is placed in fixation to maintain occlusion during flap insetting.
Lateral flaps containing a condyle are inset by seating the condyle first and then finalizing the overall flap length. An additional osteotomy is often required near the midline to recreate the curve of the mandible in this area. The transverse template is useful during the insetting process as an aid to achieving symmetry. To avoid the cheek appearing either “bowed-out” or “caved-in,” depending on the type of error (Figure 37.10), the angle of the mandible flap must be set at the correct distance from the midsagittal plane during insetting.
Anterior flaps are more difficult to inset correctly. Often the only visual guides are the maxillary arch and the midline (Figure 37.1B). The lateral segments usually cannot be stabilized because of a lack of dentition. It is easy to make errors that can result in prognathism, retrognathia, increased or decreased lower facial height, asymmetry caused by a twist in the flap, or a shift in the midline to one side as a result of unequal lengths of the mandible body.9 It is also important to establish the correct interarch distance in anticipation of later dental reconstruction (Figure 37.11).
After the bone is inset, the microvascular anastomoses are performed. The facial artery is often selected as a recipient vessel, although the external carotid (end-to-side) and the superior thyroid artery are also good choices. The lingual artery should be used with caution if the contralateral lingual artery has been previously ligated. The external jugular vein can be used as the recipient vein if it is suitable for use. An end-to-end anastomosis is easily performed. The internal jugular vein can also be used and has the advantage of decreased tendency for kinking. Often both of the peroneal vena comitantes are anastamosed to the internal and external jugular veins.
FIGURE 37.10. Postoperative result after reconstruction of a lateral defect (A and B). A. Postoperative views of a patient after lateral reconstruction. Note the symmetry. B. The Panorex shows the flap with miniplate fixation and osseointegrated implants.
Final wound closure follows completion of the microvascular anastomoses. Watertight closure of the intraoral wound is critical. A leak will often lead to contamination of the miniplates, which can develop into an orocutaneous fistula that results in considerable morbidity. Despite all efforts, however, patients with head and neck cancers are at risk for salivary leakage, particularly if they have a history of radiation therapy. Fortunately, the viability of the flap is usually not threatened, particularly if the drainage is controlled by opening the wound and performing good wound care with intra- and extra-oral packing. Suction drains are always placed as the neck flaps are closed, but they are carefully positioned away from the microvascular anastomoses. A feeding tube is routinely placed at the conclusion of the procedure.
FIGURE 37.11. Postoperative results after reconstruction of a central defect (A and B). A. Postoperative views of an anterior reconstruction. B. The Panorex for this patient shows the flap and miniplate fixation.
Early patient mobilization is encouraged. Ambulation is possible within the first postoperative week even when the fibula is used. Weight bearing is usually limited for 3 weeks, when the fibula flap donor site is skin grafted. Tube feedings are begun within 48 hours of the conclusion of the procedure. Irrigation of the oral cavity for hygiene is generally begun after 3 days. The tracheostomy is left in place for 10 to 14 days, or until wound healing is assured and it is clear that additional surgery is not needed.
Free-flap monitoring is not precise in most cases of free-flap mandible reconstruction because the bone is buried and the skin island, if present, usually lies within the oral cavity. A conventional or implantable Doppler ultrasonography device often can be used to monitor the vascular pedicle of the flap. A clear arterial signal usually can be obtained over the mandible away from the neck vessels. Frequently, a venous hum also can be heard. Skin island bleeding in response to pinprick is a useful confirmatory test when there is doubt about flap viability.
Physical therapy to address donor-site problems is rarely necessary. Follow-up studies of the mandible are limited to periodic Panorex radiographs. Radionuclide scans are not necessary.
There are three categories of potential complications in mandible reconstruction: general medical problems, head and neck wound problems, and donor-site problems. Pulmonary and cardiac problems are the most common source of general medical complications. Free-flap failure is the single most important complication pertaining to the head and neck area. Fortunately, most failing flaps can be salvaged, and the actual incidence of total flap loss is generally less than 5%. Reconstruction plate exposure and intraoral wound dehiscence (which may lead to orocutaneous fistula formation) constitute other serious problems. Donor-site complications are uncommon and rarely require additional surgery. They include abdominal wall attenuation (ilium donor site), seroma (scapula), exposed tendons or fracture (radius), and delayed skin graft healing (fibula). Cellulitis can occur at the donor site or in the head and neck area. Despite the complexity of mandible reconstruction cases, the incidence and severity of postoperative complications are low.
OTHER POSTOPERATIVE ISSUES
Many patients undergoing mandible reconstruction require postoperative radiation therapy. The presence of microvascular anastomoses does not affect the timing of postoperative radiotherapy.Radiation can begin as soon as complete wound healing is assured and the patient has recovered sufficiently. This usually requires at least 4 weeks.
Mandible reconstruction is functionally incomplete without dental restoration. A small percentage of patients can be fitted with conventional dentures if there is dentition present on either side of the defect. Patients with reconstruction plates cannot have dentition restored over the plate.
Osseointegrated implant reconstruction is the method of choice for dental restoration when conventional dentures are not feasible. These implants are placed into the bone flap and serve as a permanent foundation upon which a dental prosthesis is mounted.14 These implants are usually placed at a later date when the patient has proven to be free of disease, usually no sooner than 6 months, and more appropriately 1 year, after mandible reconstruction. Three to four implants are usually sufficient for a long dental defect.
Implant placement is usually an office procedure. Preparation of the site of implantation usually requires a preliminary procedure to remove bone flap miniplates, which may interfere with implant placement. The skin island overlying the bone is thinned at the same time. The implants must be correctly aligned during placement. They are uncovered after osseointegration has occurred, usually about 4 months later. The implants are lengthened by the placement of abutment collars on which the dental prosthesis is mounted (Figure 37.12).
Osseointegrated implants are usually unsuitable for placement in irradiated bone flaps. There is concern that the implants will not integrate into irradiated bone or, if they do, that they may not remain stable once loaded. It has been proposed that implants be placed immediately at the time of mandible reconstruction. Although they are more likely to integrate prior to irradiation, their long-term fate will not necessarily be better once the bone is irradiated. Moreover, there are issues of threatened flap viability and proper implant alignment when they are placed immediately. This practice adds more time to an already lengthy procedure.
Although the size and number of series on pediatric mandible reconstruction is limited, some general concepts can be gleaned. First, fibula transfer can be performed safely with low complication rates in children as young as 1 year old. Second, since the fibular epiphysis is not routinely included in the reconstruction, normal growth of the transferred bone does not occur.15 Anticipatory modifications can be made at the time of primary reconstruction, including substitution of lag screws for titanium plates, use of absorbable plates, and preservation of fibula length for later use by overlapping it on the native mandible. Finally, as the child grows secondary procedures may be necessary, such as distraction osteogenesis or sagittal split osteotomy of the neomandible, or secondary addition of osseous and soft-tissue flaps.
FIGURE 37.12. Dental restoration using osseointegrated implants. A. Postoperative Panorex shows three osseointegrated implants in a previously placed fibula flap. B. The abutment collars are shown with the metallic superstructure in place. C. The completed dental prosthesis is shown after it has been applied to the superstructure.
Mandible reconstruction is most commonly performed for tumors, of which squamous cell carcinoma and osteogenic sarcomas are the most common types. Lateral and anterior defects constitute two distinct types of reconstructive problems. Reconstruction is most commonly performed either with reconstruction plates and regional flaps or with microvascular free flaps.
Free flaps usually yield the best functional and aesthetic results. These lengthy procedures consist of multiple subcomponents, including bone harvesting, shaping, and fixation; insetting; microvascular anastomoses; and soft-tissue closure. Preoperative planning, such as cardiopulmonary screening and fabrication of shaping templates, increases safety and improves the aesthetic and functional results of free-flap reconstruction. Among the most important complications in mandible reconstruction are reconstruction plate exposure, free-flap failure, and serious cardiopulmonary problems.
1. Boyd JB, Gullane PJ, Rotstein LE, Brown DH, Irish JC. Classification of mandibular defects. Plast Reconstr Surg. 1993;92(7):1266-1275.
2. Singh B, Cordeiro PG, Santamaria E, et al. Factors associated with complications in microvascular reconstruction of head and neck defects. Plast Reconstr Surg. 1999;103(2):403-411.
3. Hidalgo DA, Disa JJ, Cordeiro PG, Hu QY. A review of 716 consecutive free flaps for oncologic surgical defects: refinement in donor-site selection and technique. Plast Reconstr Surg. 1998;102(3):722-732; discussion 733-734.
4. Taylor GI. Reconstruction of the mandible with free composite iliac bone grafts. Ann Plast Surg. 1982;9(5):361-376.
5. Hidalgo DA. Fibula free flap: a new method of mandible reconstruction. Plast Reconstr Surg. 1989;84(1):71-79.
6. 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(5):1314-1320.
7. Hanasono MM, Zevallos JP, Skoracki RJ, Yu P. A prospective analysis of bony versus soft-tissue reconstruction for posterior mandibular defects. Plast Reconstr Surg. 2010;125(5):1413-1421.
8. Mosahebi A, Chaudhry A, McCarthy CM, et al. Reconstruction of extensive composite posterolateral mandibular defects using nonosseous free tissue transfer. Plast Reconstr Surg. 2009;124(5):1571-1577.
9. Hidalgo DA. Aesthetic improvements in free-flap mandible reconstruction. Plast Reconstr Surg. 1991;88(4):574-585; discussion 586-587.
10. Sharaf B, Levine JP, Hirsch DL, et al. Importance of computer-aided design and manufacturing technology in the multidisciplinary approach to head and neck reconstruction. J Craniofac Surg. 2010;21(4):1277-1280.
11. Disa JJ, Cordeiro PG. The current role of preoperative arteriography in free fibula flaps. Plast Reconstr Surg. 1998;102(4):1083-1088.
12. Hidalgo DA. Titanium miniplate fixation in free flap mandible reconstruction. Ann Plast Surg. 1989;23(6):498-507.
13. Hidalgo DA. Condyle transplantation in free flap mandible reconstruction. Plast Reconstr Surg. 1994;93(4):770-781; discussion 782-783.
14. Frodel JL Jr, Funk GF, Capper DT, et al. Osseointegrated implants: a comparative study of bone thickness in four vascularized bone flaps. Plast Reconstr Surg. 1993;92(3):449-455; discussion 456-458.
15. Guo L, Ferraro NF, Padwa BL, Kaban LB, Upton J. Vascularized fibular graft for pediatric mandibular reconstruction. Plast Reconstr Surg. 2008;121(6):2095-2105.