TITLE: MANDIBULAR RECONSTRUCTION
SOURCE: UTMB Department of Otolaryngology
RESIDENT PHYSICIAN: Kelly Sweeney Clements, MD
FACULTY PHYSICIAN: Christopher H. Rassekh, MD
SERIES EDITOR: Francis B. Quinn, Jr., MD
DATE: April 23, 1997

|Return to Grand Rounds Index|


"This material was prepared by resident physicians in partial fulfillment of educational requirements established for the Postgraduate Training Program of the UTMB Department of Otolaryngology/Head and Neck Surgery and was not intended for clinical use in its present form. It was prepared for the purpose of stimulating group discussion in a conference setting. No warranties, either express or implied, are made with respect to its accuracy, completeness, or timeliness. The material does not necessarily reflect the current or past opinions of members of the UTMB faculty and should not be used for purposes of diagnosis or treatment without consulting appropriate literature sources and informed professional opinion."

Introduction

Reconstruction of mandibular defects after trauma or tumor resection is one of the most challenging problems facing reconstructive surgeons. The mandible plays a major role in airway protection and support of the tongue, lower dentition, and the muscles of the floor of the mouth permitting mastication, articulation, degluttition, and respiration. It also defines the contour of the lower third of the face. Interruption of mandibular continuity, therefore, produces both a cosmetic and functional deformity. The resulting dysfunction after loss of part of the mandible varies from minimal to major. Loss of mandibular continuity results in deviation of the mandible toward the resected side due to the unopposed pull of the remaining muscles of mastication and soft tissue contracture and scar formation. There is limited range of motion when attempting lateral and protrusive movements of the jaw with a return to midline on opening or closing secondary to the remaining contralateral muscles of mastication. In addition, malocclusion and problems with proprioception occur.

When undertaking mandibular reconstruction, the restoration of bony continuity alone should not be considered the measure of success. The functions of chewing, swallowing, speech articulation, and oral competence must also be addressed. The ultimate goal of mandibular reconstruction is to return the patient to their previous state of function. In order to achieve this goal, the reconstructive surgeon must attempt to restore bony continuity and facial contour, maintain tongue mobility, and attempt to restore sensation to the denervated areas. Oral rehabilitation postoperatively is important to improve the patients ability to manipulate the food bolus, swallow, and articulate speech. Dental rehabilitation must also be addressed.

Historical Background

Free bone grafting was the first method of reconstructing mandibular defects and was initially reported by Bardenheuer in 1881. Particulate bone and block bone grafts were used as early as 1954 by Converse. In the 1960's, mandibular reconstruction depended largely on the use of cancellous bone packed into stainless steel or vitallium trays as well as free cortical bone grafts. Early failures from these procedures were related to infection secondary to contamination by saliva and to problems with the recipient bed available for the free bone grafts. Late failures from these procedures resulted from inadequate revascularization and resorption of the graft. Failure rates as high as 50% and complication rates of 80% were reported with these procedures.

In 1971, Conley introduced the osteomyocutaneous flap. Although early results were discouraging, success rates of 50 - 80% were obtained with additional experience using pectoralis flaps with attached rib grafts. Alloplastic material was initially intended for temporary support. However, with the advent of pedicle flap procedures and free tissue transfer, combined with the advent of steel and titanium reconstruction plates in the 70's and 80's, the reconstructive options using these materials increased. In the 1980's, the popularity and increased utilization of vascularized free tissue transfer further revolutionized mandibular reconstruction. Success rates with free flaps of greater than 90% are reported. In 1989, Urken introduced the sensate free flap to head and neck reconstruction. This development is noteworthy as sensation in oromandibular reconstruction appears to have a significant beneficial effect on recovery of function.

Primary Verses Delayed Reconstruction

In previous decades, delayed reconstruction of mandibular defects was favored over primary reconstruction secondary to the belief that primary reconstruction could potentially mask tumor recurrence. In addition, success rates following primary reconstruction were not very high. This was noted by Lawson et al who, in 1982, reported a success rate of 90% for delayed reconstruction versus 46% for primary reconstruction. In addition, oral contamination of primary reconstruction resulted in unacceptably high complication rates from infection. Further support for this belief was evidenced by Komisar who noted in 1990 that patients undergoing primary reconstruction experienced longer and more frequent hospitalizations due to multiple staged procedures or complications secondary to reconstruction. He also reported that patients undergoing immediate reconstruction with a free nonvascularized bone graft had similar results as unreconstructed patients with regard to swallowing and mastication, but did have improved cosmesis. He concluded that there was no functional benefit obtained with immediate restoration of mandibular continuity.

Improved treatment modalities and the advent of microvascular free tissue transfer techniques have resulted in the resurgence of primary oromandibular reconstruction. More effective imaging and more effective adjuvant therapies have improved mapping and control of primary tumors. In addition, the availability of large vascularized segments of bone which can be transferred have allowed wider resections with oncologically sound margin size reducing the tendency to conserve tissue at the expense of adequate margins. Microvascular reconstruction of bone, soft tissue, and skin can now be used for primary reconstruction with a high rate of success. The ability to restore sensation to flaps has further enabled patients to achieve marked improvements in oral competence, speech, and swallowing. Also, dental implants can be placed successfully in vascularized bone flaps to achieve dental restoration which further enhances postoperative functional ability.

In 1991, Shockley, Weissler, and Pillsbury published a retrospective review of 19 patients who underwent primary mandibular reconstruction using reconstruction plates and noted a 79% success rate. Success in this study was defined by retention of the plate without need for further surgical intervention. Functional results were not assessed secondary to the retrospective design of the study. They did note, however, that the functional status of the patient appeared to correlate more with performance status and the extent of soft tissue loss than with the size of the mandibular defect. He concluded that immediate reconstruction of mandibular defects using reconstruction plates does not replace the use of free flaps but should be remembered as an alternative that offers fast and reliable reconstruction with no donor site morbidity and excellent facial contour.

Recent studies on patients undergoing immediate free flap reconstruction of oromandibular defects show that these patients achieve a superior level of function without a significant increase in complications or hospital stay. In addition, by reconstructing the mandibular defect primarily, the problems of drift of the remaining mandibular segments and contracture of the surrounding soft tissues are avoided. Primary reconstruction also decreases the risk of facial nerve injury when attempting to reconstruct angle, ramus, or condyle defects. Identification of the nerve can be quite difficult in a previously operated and/or irradiated bed. In light of the above findings, primary reconstruction of oromandibular defects with vascularized bone transfer is now considered the procedure of choice for oromandibular reconstruction. However, in certain circumstances such as lack of microvascular surgical skills or equipment or due to inherent patient factors, it may still be necessary to employ alternative methods of reconstruction and/or delayed reconstruction for oromandibular defects.

Defect Evaluation

The first step in undertaking mandibular reconstruction involves careful evaluation of the patients anatomy in order to define the full extent of the existing or proposed defect. Both bony and soft tissue components must be examined so that the surgeon can conceptualize the tissue components which require reconstruction. The detailed definition of the defect provides the surgeon with a framework for selecting the best method for reconstruction.

Defects which are lateral and limited to the mandibular body often cause only minimal cosmetic and functional deformity. Patients may compensate for lateral defects and reconstruction may not be necessary. With increasing extent of bony loss, however, severe functional and cosmetic deformities result which necessitate reconstruction in order to restore quality of life. Furthermore, when evaluating defects that involve the mandibular ramus, it is important to note if the patient has a proximal segment of bone, a functioning temporomandibular joint, or a condylar neck to which the graft may be secured. Radiographic analysis of the bony mandibular anatomy can be very helpful when formulating a plan for oromandibular reconstruction. Computed tomography with bone windows, 3D CT, panorex films, and magnetic resonance imaging add additional information in terms of amount of actual bone and soft tissue loss and the relationships of the remaining mandibular segments to the proposed and/or existing defect.

When evaluating patients with existing mandibular defects, the quality and quantity of the remaining soft tissue is important. When considering the use of nonvascularized bone grafts, the ideal soft tissue bed would have enough bulk, vascularity, and cellularity in order to incorporate the bone graft. However, tissue loss, scar contracture, and prior irradiation often make secondary reconstruction difficult and decrease the chances of success. In this setting, the use of hyperbaric oxygenation should be considered.

Many different methods of classifying mandibular defects have been described in the literature. The HCL classification system involves using three uppercase (H,C,L) and three lowercase (o,m,s) letters. Defects designated as H are lateral defects which include the condyle but do not cross the midline. L defects are basically H defects with the condyle excluded. C defects consist of the central component of the mandible including the four incisors and two canine teeth. These letters can be used in combination to describe the extent of the defect. For example, an angle to angle defect would be described as LCL. The small letters are used to describe the extent of the soft tissue requirement. The letter m designates that mucosa is needed, s indicates that skin is needed, sm denotes that skin and mucosa are needed, and o implies that only bone is needed.

Alloplastic Materials

Alloplastic implants can be used as a temporary spacer to span the missing segment of bone until a definitive reconstructive procedure can be performed. In patients who are unsuitable for other reconstructive procedures due to poor health or other reasons, these implants may be used as permanent fixation. These materials are also used to fixate vascularized free bone grafts to remaining mandibular segments.

Many different types of alloplastic materials have been used for mandibular reconstruction with varying rates of success. Organic calcium salts have been used with limited success. Synthetic materials such as methylmethacrylate, proplast and teflon which are nonbiodegradeable and biocompatible have also been used, but success has been limited by breakdown of the overlying tissue with subsequent extrusion of the implant. In 1992, Goode reported the use of tobramycin-impregnated methacrylate in four patients with good results and noted that the methacrylate could be molded to fill any size defect and that the slow release of tobramycin decreased infection rates. Wires and pins have been used to maintain the preoperative relationship of resected margins. However, lack of stability with these wires and pins leads to infection and rapid extrusion.

A variety of reconstruction trays made of Dacron, stainless steel, vitallium and titanium have been used as cribs for autogenous bone grafts. The meshwork design of these devices allows ingrowth of host blood vessels promoting osteoneogenesis. The trays are generally strong enough to support the mandible without the need for intermaxillary fixation, but they should be reserved for patients undergoing secondary reconstruction. The trays can be removed after six months or when good radiographic evidence of bone healing is noted.

Mandibular reconstruction plates and screws are the most widely used alloplastic devices for mandibular reconstruction. The most common metals used in the fabrication of these plates are stainless steel, vitallium, and titanium. Vitallium is an alloy of cobalt, chromium, and molybdenum. This type of plate initially seemed to be ideal, however, the low malleability can make application difficult. AO stainless steel and AO titanium reconstruction plates were developed in an attempt to find a mandibular reconstructive option that was fast, single-staged, and reliable while maintaining oral function and form. These plates have been used with varying rates of success. The development of the titanium hollow osseointegrated reconstruction plate (THORP) was an attempt to address the failures of the older plating systems. This plate has a hollow screw made of titanium with perforations along the screw body which allow bone ingrowth and result in increased plate stability at the bone-screw interface. An expansion bolt within the screw head allows the plate to be anchored to the interosseous screw instead of being compressed to the underlying mandible. This prevents pressure necrosis of the underlying bone decreasing the potential for plate failure at the screw-bone interface.

Reconstruction plates are usually shaped before the mandibular resection and applied afterwards. By bending these plates and placing drill holes in the proximal and distal mandible segments before the mandibulotomy, the surgeon can more confidently establish the proper relationships of the remaining mandibular segments after removal of the involved bone.

Placement of mandibular reconstruction plates does not contraindicate the use of post-operative radiation therapy. In 1991, Gullane reported an analysis of 64 cases evaluating the interface radiation dose using both stainless steel and titanium plates with a parallel beam radiation technique. He noted that the radiation dose at the plate-bone interface increased only 15% at the 6-mV level with the excess tissue dose scatter extending only 1.1 mm to the surrounding soft tissue.

Pedicled and free flaps may be combined with plate reconstruction for soft tissue supplementation and to minimize the possibility of postoperative complications. The pectoralis myocutaneous flap is the most commonly used pedicled flap for this purpose. The plate is usually placed first, and the muscular pedicle is then suspended from the plate. Additionally, the skin paddle can be used for reconstruction of intraoral or lip and chin reconstruction. Free flaps can also be used to supplement plate reconstruction in a similar manner. When draping soft tissue flaps over reconstruction plates, it important to avoid compression of the vascular pedicle by the plate.

In 1994, Boyd et al published a study of two patient groups in an attempt to define the role of reconstruction plates for bone replacement in mandibular reconstruction. The first group consisted of 15 patients reconstructed with radial forearm osteocutaneous flaps and 16 reconstructed with radial forearm fasciocutaneous flaps and reconstruction plates. The second group consisted of 40 patients reconstructed with radial forearm fasciocutaneous flaps and reconstruction plates (21 stainless steel plates, 19 THORP plates). Success was defined as reconstructions that survived and functioned until the time of review or death. He noted that vascularized bone was more successful than plates in terms of reconstruction success and minimizes days of life lost. However, overall success of reconstruction plates was 78.9%. THORP plates were noted to be more durable than steel plates. Anterior reconstructions were noted to predispose to plate exposure, while lateral reconstructions were noted to respond well to reconstruction using plates and radial forearm flaps. He concluded that radial forearm osteocutaneous flaps have significantly fewer complications and increased success rates than reconstruction plates used with radial forearm fasciocutaneous flaps. However, mandibular reconstruction using plates covered with vascularized soft tissue remain an effective alternative.

Early plate failure in the first six weeks after surgery is most often due to technical variations in plate application such as overprojection or unstable application of the plate which can lead to soft tissue breakdown and infection. Local wound care and IV antibiotics are usually sufficient treatment for these complications. Rarely, debridement or secondary closure may be required. Extensive loss of tissue requires an alternative reconstruction plan which most often involves microvascular soft tissue transfer. The plate does not need to be removed if stable. Unstable plates should be replaced because healing in this setting is very unlikely.

Late plate complications are more common. Exposure of the plate at this point suggests instability of the reconstruction and the surgeon must plan for alternative reconstruction. Exposure after 12 - 18 months can occur from resorption of bone around the hardware with resultant plate instability. In this setting, microvascular free tissue transfer should be considered. Fracture of the plate can also occur as a late complication and may require plate replacement.

Free Bone Grafting

Nonvascularized autogenous bone grafts can be used for reconstruction of small to medium size mandibular defects. These can be harvested from the patients calvarium, rib, ilium, tibia, fibula, scapula, humerus, radius, and metatarsus and provide viable and immunocompatible osteoblastic cells.

Bone graft healing occurs in two phases. Initially, new osteoid is deposited by osteoblastic cells which survive the transplantation process. The amount of bone formed is directly proportional to the number of viable osteoblasts transferred. In order to ensure adequate bone formation, it is important to provide the maximum amount of cells per given volume. The bone formed during this phase tends to be poorly organized. This continues for about four weeks and ultimately determines the size of the resulting new bone. The second phase contributes very little to the new bone mass. It begins about two weeks after implantation and continues indefinitely. It involves the revascularization, remodeling, and reorganization of the previously formed bone by osteoblasts and osteoclasts. This process is mediated by bone morphogenic protein which is found most abundantly in cortical bone.

Cancellous bone grafts, consisting of medullary bone and bone marrow, contain the highest percentage of viable osteoblasts. These grafts become revascularized rapidly due to their particulate structure and large surface area. This results in a higher percentage of surviving cells after transplantation. In contrast, cortical grafts consisting of lamellar bone struts contain large numbers of osteoclasts. These cells rarely survive the transplantation process due to the time delay required for revascularization. Corticocancellous grafts contain both cortical bone and underlying cancellous bone providing osteoblastic cells as well as strength necessary for bridging discontinuous defects. Cancellous bone grafts produce sufficient phase one healing and can be used in cases with small defects where phase two healing is encouraged by the surrounding periosteum and bone. These are not adequate for larger defects. Cortical bone grafts rarely survive transplantation due to the lack of revascularization and are not used. The combination of particulate cortical bone and cancellous marrow provides the best potential for osteogenesis. The particulate nature of the graft allows rapid revascularization. However, structural support in the form of an alloplastic tray is required because of the lack of rigidity of this type of graft.

Allogenic mandible, rib, or iliac crest has been used occasionally for mandibular reconstruction. The allograft is usually hollowed and functions as a biodegradable tray for particulate corticocancellous bone grafts or as supplementation for autogenous bone grafting when insufficient bone is available. In 1990, Lowlicht published a report on 20 patients who underwent this type of reconstruction using an allogenic crib/particulate bone cancellous marrow graft protocol and noted an overall success rate of 81%. He found this method of mandibular reconstruction to be reliable particularly in irradiated tissue beds. The benefits of this method included low immunogenicity of the graft, high concentration of transplanted osteocytes, and complete bioresorbability of the tray with transmission of increasing stress to the autogenous graft which can facilitate phase II osteogenesis.

In 1993, Constantino et al published a report on the use of distraction osteogenesis as a potential method of reconstructing mandibular defects. His study involved making 2.5 cm unilateral segmental mandibular body defects in three dogs followed by creation of a transport disc of bone cut from one end of the defect with the vascular supply preserved by maintaining the periosteum across the transport disk osteotomy site. A distraction appliance was then applied to the mandible via an external incision. The movement of the transport disk across the defect was noted to leave a regenerative callus at a rate of 1 mm/day. The dogs were noted to have normal function one year later. Tests showed that the mandible maintained 77% of the pre-operative strength. He concluded that distraction osteogenesis is capable of forming stable bone in mandibular defects at a rate of 1.0 mm/day, and that the bone formed is comparable in thickness to native mandible.

Vascularized Pedicled Bone Transfers

In the 1970's, the superiority of vascularized bone in mandibular reconstruction was demonstrated. Animal studies illustrated that vascularized bone resists infection and extrusion, and heals to surrounding bone in a manner similar to fracture healing. This occurred even in irradiated fields. These findings initially lead to attempts of wrapping free bone grafts in regional cutaneous flaps with delayed transfer after revascularization of the bone. Mixed results have been reported with tissue transferred in this manner.

In 1980, Ariyan and Cuono reported the use of a pectoralis major pedicled myocutaneous flap transferred with a segment of the underlying fifth rib. Latissimus dorsi with attached rib has also been used. In 1980, Panje introduced the trapezius osteomyocutaneous flap reporting an 87% success rate in 27 patients. This flap can provide up to 12 x 2.5 cm of scapular bone for reconstruction of mandibular defects. The medial scapular spine is used in combination with either a superiorly based trapezius flap based on the paraspinous perforators and the occipital artery, or an island trapezius flap based on the transverse cervical vessels. Sacrificing trapezius function solely for mandibular reconstruction is not usually recommended. However, this flap may be useful in certain patients who have already undergone a neck dissection with sacrifice of cranial nerve XI. A temporalis muscle pedicled flap with attached split calvarial bone graft has also been described.

Pedicled bone transfers are used infrequently because of several technical problems. These flaps tend to be difficult to harvest and have a limited arc of rotation. There is also limited mobility of the bone graft relative to the soft tissue portion of the flap. The blood supply of the bone portion is often tenuous after transfer limiting the flaps reliability. Additionally, the bone available is not very thick which limits dental rehabilitation. It is important, however, for head and neck surgeons to be aware of the reconstructive possibilities of these pedicled flaps because they may prove useful in certain situations.

Free Vascularized Tissue Transfer

Microvascular free tissue transfer has provided the reconstructive surgeon with the opportunity to more carefully address the aesthetic and functional reconstruction of oromandibular defects based on the wide variety of bone and soft tissue available. There are several characteristics that the ideal free flap should have. The tissue should be of sufficient length, width, and height for reconstruction of the proposed defect and should be well vascularized with a pedicle of adequate length. In addition, the bone portion of the flap should be similar in contour to the native mandible and should be easily shaped without vascular compromise. The skin paddle should be thin, pliable, and sensate, and there should be minimal donor site morbidity. Accessibility of the donor site is also important to allow for a two team approach during the procedure to decrease operative time.

Rib:

Rib tissue was the first microvascular bone transfer described in the literature. This flap is based on the intercostal vessels with the fifth and sixth ribs harvested. These flaps have limited available bone, and the vascular supply of the skin paddle is unreliable. In addition, unacceptable complications have been reported with harvest. For these reasons, these flaps are rarely used.

Iliac Crest:

The iliac crest provides a curved piece of bone of up to 16 cm in length. It is possible to fashion an ascending ramus using this flap without the need for osteotomies because of the wide flat shape. The iliac crest is also ideal for receiving osseointegrated implants. The main blood supply to the bone comes from the deep circumflex iliac artery. The skin paddle is supplied by an array of perforators from along the medial aspect of the crest. One variation of this flap involves using the internal oblique muscle for intraoral reconstruction. This muscle can be left to epithelialize without a skin graft, and the skin paddle can be used to reconstruct external skin defects or as an external monitor. In young female patients, the iliac crest may be the most acceptable donor site because the cosmetic deformity is minimal. The site tends to be painful for the first 48 hours and in some patients may remain tender for several months. In addition, some patients experience gait disturbances after surgery which may persists for several months. Other rare complications include neuropraxia and hernia formation which can be avoidable by using meticulous technique.

In 1991, Urken et al published a report of 10 patients who underwent reconstruction of through and through mandibular and soft tissue defects using the internal oblique iliac crest free flap. There were no flap failures in this series. Bone scans on postoperative day four or five showed good uptake in all segments of the bone graft. Complications included one death secondary to an esophageal perforation and one patient with persistent aspiration who required a total laryngectomy. The authors noted that the major limiting factor with the iliac crest flap has been the quality of the soft tissue component. The addition of the internal oblique muscle to this flap, therefore, provided more reliable vascularity and a greater degree of mobility of the soft tissue component relative to the bone. They concluded that the internal oblique iliac crest free flap is particularly useful for reconstruction of mandibular defects which include through and through skin/soft tissue defects.

Fibula:

The fibula provides the longest segment of bone with 20 to 30 cm available for harvest. In addition, the segmental blood supply of the bone permits multiple osteotomies. This flap can be used to span an angle to angle defect. The bone is also of adequate width and height to allow for placement of osseointegrated dental implants. The arterial supply to this flap consists of the peroneal artery. The skin paddle is supplied by perforators running in the intercompartmental septum and has less bulk than the iliac crest paddle but more than a radial forearm flap. Many surgeons recommend obtaining an arteriogram preoperatively to rule out an absent peroneal artery and to establish the position of the vascular perforators. Donor site morbidity with this flap is minimal unless the distal osteotomy is within 6 cm of the ankle. In addition, the location of the flap will allow simultaneous harvest by a second team at the time of tumor resection. However, there is a risk of peroneal nerve injury with resultant foot drop or weakness in plantar flexion of the great toe which can be avoided with meticulous dissection.

Initial reports on the fibula flap questioned the reliability of the skin island. Much debate centered around this topic with some authors recommending including a cuff of soleus or flexor hallucis longus muscle with the skin paddle in order to assure adequate vascularity. This technique, however, resulted in a somewhat bulky soft tissue paddle which made reconstruction more difficult. In 1995, Jones et al attempted to end the ongoing debate by reporting their results of a study designed to determine the reliability of the skin island associated with the fibula flap. They designed flaps over the distal third of the fibula in 60 cadavers and completely isolated the septum and ligated all muscle perforators prior to dye injection. They noted that all of the flaps had 100% reliable perfusion of the skin island after injection of the proximal peroneal artery. Next, they performed 34 fibular flaps and designed the skin island over the distal third of the leg based only on the septal perforators with no muscle incorporation. They concluded that the reliability of the skin paddle could be assured by designing it more distally over the lower third of the leg, performing Doppler studies of the perforators before surgery, and designing osteotomies to protect the septocutanous perforators.

In 1993, Hidalgo and Rekow published a report on 60 patients who underwent fibula free flap reconstruction of mandibular defects for an average bone gap of 9.4 cm. In this series, the skin island was harvested with the entire length of septum to include all of the septocutaneous vessels. A 90% rate of reliability of the skin island was noted. The authors concluded that the fibula was a good alternative for mandibular reconstruction due to the large amount of bone and soft tissue available and the limited donor site morbidity.

In 1994, Cheung et al reported results of 12 fibula flaps used to reconstruct anterior mandibular defects. The authors noted that the anterior arch of the mandible is a critical area in mandibular function and facial appearance, providing support for the tongue and maintaining support of the lateral portions for effective mastication. The success rate in this series was 100%. In addition, the functional and cosmetic results were rated as excellent or good in 75% of the patients. They recommended use of the fibula osteocutaneous free flap for reconstruction of angle to angle mandibular defects.

Radial Forearm:

The radial forearm provides the best quality of skin for oral reconstruction. It is thin, pliable, and abundant and has the potential to restore intraoral sensation. However, the bulk of the flap may be inadequate for large soft tissue defects. This flap is based on the radial artery and 10 - 12 centimeters of up to 40% of the radial circumference may be harvested. Bone harvested with this flap should be taken as a boat shaped graft to eliminate right angle weak points which results in a small amount of wasted bone. Postoperative immobilization is important to allow for healing of the remaining radial bone. The donor site is covered with a skin graft which may not be well tolerated in a young patient secondary to scar formation and the resultant cosmetic deformity. The osteotomy reduces the strength of the radius to three point bending up to 74% with pathologic fractures reported in 23% of patients. When fractures occur, the patient may experience limited range of motion and decreased pinch and grip strength. Because of the limited bone supply available with the radial forearm flap, other alternatives should probably be used for defects requiring a substantial amount of bone and soft tissue.

Scapula:

The scapular free flap is based on the circumflex scapular artery and vein which give off periosteal perforators to supply the lateral border of the scapula. The quality of the skin paddle is inferior to the radial forearm flap but it is extremely versatile. Scapular and parascapular skin paddles may be harvested based on circumflex scapular branches. Inclusion of the thoracodorsal system of the subscapular artery will allow latissimus dorsi and serratus anterior with associated ribs to be included. This wide array of soft tissue components and 12 centimeters of lateral scapula bone available make this flap attractive for reconstruction of large composite head and neck defects. The major disadvantages of this flap include the need for the patient to be positioned on his side which results in difficulty with the two team approach. In addition, the amount and quality of bone available is less than some of the other sites which can limit dental rehabilitation. Donor site morbidity is usually mild with cosmesis dependent on the amount of skin harvested and the need to skin graft the donor site.

Clavicle:

In 1996, Seikaly and others introduced the clavipectoral osteomyocutaneous free flap as an additional option for reconstruction of mandibular defects. Cadaver dissections with ink injections revealed that the clavicle was supplied mainly by the clavicular branch of the thoracoacromial artery with a minor contribution from the deltoid branch of the thoracoacromial artery. In addition, an average of 16 centimeters of bone is available for harvest. Two different soft tissue donor sites were described based on either the clavicular head of the pectoralis major muscle and overlying skin supplied by the deltoid and clavicular branches of the thoracoacromial artery or the sternocostal head of the pectoralis major muscle and skin supplied by the pectoral branch of the thoracoacromial artery. Advantages associated with this flap include the availability of the donor site during tumor resection, surgical anatomy that is familiar to the head and neck surgeon, and minimal functional and cosmetic donor site morbidity. One disadvantage of this flap is the relatively short pedicle which may require the use of interposition grafts for both the arterial and venous anastomosis. The authors concluded that the clavipectoral osteomyocutaneous free flap is an excellent source of well vascularized, potentially innervated soft tissue and bone for reconstruction of mandibular defects.

The use of simultaneous free flaps has been reported by several authors. Large defects may be reconstructed in this manner, taking advantage of the best aspects of respective flaps. Large mandibular defects can be reconstructed using a fibula bone flap while the oral cavity can be reconstructed using a radial forearm skin paddle. To avoid excess ischemic time when using multiple free flaps, the first flap should be revascularized prior to severing the second flap from the donor vessels.

In 1995, Futran et al published a review comparing three different types of reconstruction plates used for fixation of vascularized bone grafts in mandibular reconstruction in 95 patients, 48 with AO stainless steel plates, 25 with AO titanium plates, and 22 with titanium hollow screw reconstruction plates (THORP). Because of the hollow design, the THORP has increased surface area available for contact with the osteocytes which allows bony ingrowth and promotes osseointegration. Plate stability is also noted to increase during the implantation period when using this type of plate. Three plate fractures, eight plate exposures, and two cases of nonunion were reported in the AO stainless steel plate group. In the AO titanium group, one plate exposure and one instance of loose screws occurred. The only complication in the THORP group was a plate exposure in one patient. The authors concluded that the AO THORP is the best method to rigidly fixate vascularized bone grafts because of the advanced design and potential for osseointegration, and because fewer screws are needed for adequate fixation.

Sensate Flaps

In 1989, Urken introduced the concept of sensate free flaps by transferring the associated cutaneous nerve with the skin paddle. The basic principle of sensate flaps is that a portion of skin is transferred with the associated cutaneous sensory nerve. An anastamosis is then performed between the cutaneous sensory nerve and a sensory nerve in the reconstructed region in order to provide the central connection for the sensory input. It is important that the recipient nerve be transected in the normal course of tumor resection and not simply sacrificed for reconstructive purposes. The radial forearm flap can be transferred with the antebrachial cutaneous sensory nerve, and the fibula flap can be transferred with the lateral sural cutaneous nerve. Recent studies have shown good recovery of sensation in these free flaps. Return of sensation does not appear to be affected by age, location of defect, or radiation therapy. The restoration of sensation in microvascular free flaps has been shown to have beneficial effects in recovery of oral cavity and pharyngeal function.

Dental Rehabilitation

Full restoration of masticatory function requires intraoral reconstruction of a denture bearing surface. Scarring and the inability to obtain a thick pliable surface over the neomandible have frustrated efforts to obtain dental rehabilitation. The use of endosseous dental implants has markedly improved the ability to provide a stable anchorage for dentures. In 1965, the first patient was successfully fitted with a fixed prosthesis supported by permanent implants. Endosseous implants are now the treatment of choice for restoration of dentition and mastication in edentulous patients.

There are different types of dental prostheses available for use. Conventional tissue borne prostheses (full and partial dentures) are the least stable, but may be satisfactory if anchored to remaining dentition. The implant-assisted prosthesis is a removable denture supported by two or more endosseous implants. Loading forces are shared equally between the mucosal surface and the implants. This appliance is more stable than traditional dentures. The implant-borne prosthesis consists of a retrievable, fixed denture connected to an abutment by endosseous screws. It requires at least four evenly distributed implants and is the most stable form of dental rehabilitation. It can be removed periodically if necessary by the dentist or prosthodontist for maintenance or cleaning.

The minimum bone height required for implants is 10 millimeters. The bone graft must be wide enough to provide adequate bone around the implant. The implants function as tooth root analogues and are usually placed in a two stage procedure. In the first stage, the recipient bone is drilled and the implant is placed. These should remain stable and unloaded for up to 6 months to permit full osseointegration. After successful osseointegration has occurred as evidenced by clinical and radiographic evaluation, the second stage is performed by elevating the mucosa and placing the transmucosal abutment. The mucosa is then closed around the abutment, and an occlusal dressing is placed. Two weeks later, the prosthesis is constructed and can be anchored directly to the implants or to a removable overdenture. Long term retention rates approach 85% at ten years.

In 1989, Urken et al published a report on several patients who underwent mandibular reconstruction with microvascular free bone grafts and have been successfully rehabilitated with osseointegrated implants. The authors noted that implants can be placed at the time of reconstruction with a second stage of prosthesis insertion four to six months later. Reasons for placing the implants at the time of the reconstruction include reliable vascularity, wide access, ability to assess relationships for accurate placement, elimination of additional procedures, and earlier restoration of dental rehabilitation. The authors concluded that primary placement of osseointegrated implants in well vascularized neomandibles reconstructed with microvascular free flaps is a safe, effective, and rapid method of achieving early dental rehabilitation. Similarly, in 1994, Kraut et al concluded that the incorporation of osseointegrated implants into the armamentarium of the head and neck surgeon has facilitated functional reconstruction of patients with mandibular and maxillary defects.

Quality of Life Issues

Attention to quality of life issues and functional assessment of patients has become increasingly important. Success with mandibular reconstruction can no longer be measured solely on proper wound healing and nonextrusion of the implant. Form, function, and cosmetic results must also be considered. In patients who have had reconstruction of mandibular defects, functional assessment should include evaluation of the patients overall well being, swallowing ability, oral competence, speech ability, dental and chewing rehabilitation, and cosmetic results. Studies show that functional assessment of patients reconstructed with microvascular free tissue transfer demonstrate clear superior function over non- reconstructed patients without significantly increased hospital stays.

The increased popularity of vascularized tissue transfer for reconstruction of mandibular defects has prompted some questions concerning cost-effectiveness, morbidity, and suitability of these procedures for elderly patients or those with incurable disease. In an attempt to further clarify these issues Talesnik et al published a report in 1995 on 39 patients after reconstruction of mandibular defects. There were 21 patients who received pectoralis major myocutaneous flaps, 14 received osteocutanous free tissue transfer, and 4 underwent reconstruction using plates and free tissue transfer for soft tissue coverage. They noted that length of surgery and duration of intensive care monitoring were longer for the free flap groups, but the length of hospitalization was similar. Systemic complications were more frequent in the free flap groups, but flap problems were more common in the pedicled flap group. Facial appearance was rated as higher in the free flap groups. There was no significant difference between the groups when comparing speech, social functioning, or oral function. The cost of hospitalization was higher in the free flap groups. However, this difference decreased when cost of subsequent hospitalizations for complications was considered. The authors concluded that free tissue reconstruction provides more predictable aesthetic results and faster return to normal social functioning than pedicled tissue transfer, and that the combination of reconstruction plate/free tissue transfer is a cost-effective, efficient method of reconstruction giving good results and low morbidity.

Conclusion

Oromandibular reconstruction is a complex procedure with many available options. The reconstructive surgeon must address several concerns before undertaking any reconstructive attempts. First, the decision regarding primary versus secondary reconstruction must be made after considering all variables and patient factors. The ablative team must take care to carefully preserve adequate vessels if microvascular tissue transfer is planned. The soft tissue repair must provide sufficient bulk and ensure adequate tongue mobility. In addition, the reconstruction must provide sufficient durability and strength to allow resumption of daily activities. A water-tight closure must be achieved to avoid problems with infection or fistula formation. The training and availability of the reconstructive team must be considered, as well as the time required for reconstruction, the patients long term prognosis, and the type of defect and resultant donor site morbidity. Lastly, the overall health status and the desires of the patient must be taken into consideration. Keeping these concerns in mind, it is important for the surgeon to be familiar with a wide range of reconstructive alternatives so that the best procedure for each patient can be chosen.

BIBLIOGRAPHY

Ariyan S, Cuono CB. Myocutaneous Flaps for Head and Neck Reconstruction. Head Neck Surg 2(4):321 - 345, 1980.

Baker G. Oseeointegrated Implants for Dental Rehabilitation Following Ablative and Reconstructive Surgery. Op Tech Otol Head Neck Surg 4(2):115 - 122, 1993.

Boyd JB, Mulholland RS, Davidson, J, et al. The Free Flap and Plate in Oromandibular Reconstruction: Long-Term Reivew and Indications. Plas Reconstr Surg 95(6):1018 - 1028, 1995.

Boyd JB. Osteocutanous Free Flap Options in Oral Cavity Reconstruction. Op Tech Otol Head Neck Surg 4:104 - 114, 1993.

Buchbinder D, Urken ML. Mandibular Reconstruction. In Bailey BJ (ed). Head and Neck Surgery - Otolaryngology, Philadelphia, Lippincott, 1993.

Cheung SW, Anthony JP, Singer MI. Restoration of Anterior Mandible With the Free Fibula Osseocutaneous Flap. Laryngoscope 104:105 - 113, 1994.

Converse JM, Campbell RM. Bone Grafting in Surgery of the Face. Surg Clin North Am 34:375 - 401, 1954.

Conley J. Use of Composite Flaps Containing Bone for Major Repairs in the Head and Neck. Plast Reconstr Surg 49:522 - 526, 1972.

Costantino PD, Friedman CD, Shindo ML, et al. Experimental Mandibular Regrowth by Distraction Osteogenesis. Crch Otolaryngol Head Neck Surg 119:511 - 516, 1993.

Futran ND, Urken ML, Buchbinder D, Moscoso JF, Biller HF. Rigid Fixation of Vascularized Bone Grafts in Mandibular Reconstruction. Arch Otolaryngol Head Neck Surg 121:70 - 76, 1995.

Gilbert RW, Dorion D. Management of the Failed Mandibular Reconstruction Plate. Op Tech Otol Head Neck Surg 4(2):159 - 164, 1993.

Gilbert RW, Dorion D. Near-Total Mandibular Reconstruction: The Free Vascularized Fibular Transfer. Op Tech Otol Head Neck Surg 4(2):145 - 148.

Goode RL, Reynolds BN. Tobramycin-Impregnated Methylmethacrylate for Mandible Reconstruction. Arch Otolaryngol Head Neck Surg 118:201 -204, 1992.

Gullane PJ. Primary Mandibular Reconstruction: Analysis of 64 Cases and Evaluation of Interface Radiation Dosimetry on Bridging Plates. Laryngoscope 101:1 - 24, 1991.

Hidalgo DA, Rekow A. A Review of 60 Consecutive Fibula Free Flap Mandible Reconstructions. Plas Reconstr Surg 96(3):585 - 596, 1995.

Irish JC, Gullane PJ. Plating Techniques for Mandibular Reconstruction. Op Tech Otol Head Neck Surg 4(2):96 - 103, 1993.

Jones NF, Monstrey S, Gambier BA. Reliability of the Fibular Osteocutaneous Flap for Mandibular Reconstruction: Anatomical and Surgical Confirmation. Plast Reconstr Surg 97(4):707 - 716, 1996.

Komisar A. The Functional Result of Mandibular Reconstruction. Laryngoscope 100:364 - 374, 1990.

Kraut RA, et al. Endosteal Implants Following Tumor Surgery and Avulsive Trauma. Laryngoscope 104: 504 - 512, 1994.

Lawson W, Blaek SM, Loscalzo LJ, Biller HF, Krespi YP. Experience with Immediate and Delayed Mandibular Reconstruction. Laryngoscope 92:5 - 10, 1982,

Lowlicht RA, Delacure MD, Sasaki CT. Allogeneic (Hojograft) Reconstruction of the Mandible. Laryngoscope 100:837 - 843, 1990

Moscoso JF, Urken ML. The Internal Oblique-Iliac Crest Osseomyocutaneous Free Flap for the Reconstruction of Compostie Oromandibular Defects. Op Tech Otol Head Neck Surg 4(2):132 - 135, 1993.

Panje WR, Morris MR. Oral Cavity and Oropharyngeal Reconstruction. In Cummings (ed). Otolaryngology - Head and Neck Surgery. St. Louis, Mosby, 1993.

Schockley WW, Weissler MC, Pillsbury HC. Immediate Mandibular Replacement Using Reconstruction Plates. Arch Otolaryngol Head Neck Surg 117:745 - 749, 1991.

Seikaly H, Calhoun K, Rassekh C, Slaughter D. The Clavipectoral Osteomyocutanous Free Flap. Otol Head Neck Surg in press (accepted for publication 1996).

Talesnik A, Markowits B, Calcaterra T, Ahn C, Shaw W. Cost and Outcome of Osteocutaneous Free- Tissue Transfer versus Pedicled Soft-Tissue Reconstruction for Composite Mandibular Defects. Plas Reconstr Surg 97(6):1167 - 1177, 1996.

Urken ML, Buchbinder D. Oromandibular Reconstruction. In Cummings (ed). Otolaryngology - Head and Neck Sugery. St Louis, Mosby, 1993.

Urken ML, Buchbinder D, Weinberg H, et al. Functional Evaluation Following Microvascular Oromandibular Reconstruction of the Oral Cancer Patient: A Comparative Study of Reconstructed and Nonreconstructed Patients. Laryngoscope 101:935 - 950, 1991.

Urken ML, Buchbinder D, Weinberg H, Vickery C, Sheiner Al, Biller HF. Primary placement of osseointegrated implants in Microvascular Mandibular Reconstruction. Otolaryngol Head Neck Surg 101(1):56 - 73, 1989.

Urken ML, Moscoso JF. Sensate Cutanous Flaps in Oral Cavity and Pharyngeal Reconstruction. Op Tech Otol Head Neck Surg 4(2):141 - 144, 1993.

Urken ML, Weinber H, Vickery C, Buchbinder D, Lawson W, Biller HF. Oromandibular Reconstruction Using Microvascular Composite Free Flaps. Arch Otolaryngol Head Neck Surg 117:733 - 744, 1991.

Urken ML, Weinberg H, Vickery C, Buchbinder D, Lawson W, Biller HF. The Internal Oblique-Iliac Crest Free Flap in Composite Defects of the Oral Cavity Involving Bone, Skin, and Mucosa. Laryngoscope 101:257 - 270, 1991.

Wei FC, Seah CS, Tsai YC, Liu SJ, Tsai MS. Fibula Osteoseptocutaneous Flap for Reconstruction of Composite Mandibular Defects. Plas Reconstr Surg 93(2):294 - 304, 1994.