Master Techniques in Otolaryngology - Head and Neck Surgery: Reconstructive Surgery, 1ed.

16. Management of the Isolated Condylar Defect

Neal D. Futran

INTRODUCTION

The need for surgical reconstruction of an isolated defect of the condyle and ramus of the oromandibular complex is usually the result of resection of a neoplasm involving the temporomandibular joint (TMJ). This entity is uncommon and can arise as a primary mass of the TMJ, as an extension of a neoplasm from surrounding tissue, or from metastases from a distant site. A detailed history, clinical examination, imaging studies, and tissue diagnosis are paramount in surgical planning.

Surgical resection of tumors of the TMJ is challenging as a result of various factors including preservation of the facial nerve, need for skull base resection, and/or necessity of lateral temporal bone resection. Reconstruction of these defects is also challenging, as functionally the TMJ plays a key role in jaw movement, mastication, and facial contour.

Resection of tumors involving the TMJ results in various defects, with most resulting in resection of the condyle of the mandible, and possibly the meniscus and glenoid fossa. The decision-making process and various reconstructive options for defects of the condyle are discussed; however, this chapter focuses on the reconstruction of the isolated condylar defect using the fibula free flap.

Primary lesions of the TMJ in a variety of ways often mimic much more common disorders of the TMJ. Patients may give a history of unilateral facial swelling or asymmetry, obstruction of the external auditory meatus, middle ear effusion, malocclusion, chronic dysfunction of the TMJ, pain, or aural fullness. Patients less commonly may describe hypesthesia in the distribution of the mandibular branch of Cranial Nerve V. This variability often results in a delay in diagnosis with patients often receiving an initial diagnosis of TMJ, or in fewer patients a diagnosis of parotitis. Table 16.1 lists a differential diagnosis of masses involving the TMJ.

TABLE 16.1 A Differential Diagnosis of Masses Involving the Temporomandibular Joint. Benign and Malignant Masses of the TMJ

Patients presenting with a malignant tumor originating from surrounding structures with secondary involvement of the TMJ may present with features similar to primary masses in the TMJ. More commonly, however, they will present with features of malignancy such as significant pain, otalgia, dysphagia, trismus, facial weakness, or infiltration of the skin.

Each patient’s past medical and surgical history should be reviewed in detail. The patient’s general health and his or her ability to tolerate a general anesthesia should be evaluated. Special attention should be given to any conditions that might compromise wound healing, such as diabetes or malnutrition. A history of peripheral vascular disease should also be investigated, as this may be pertinent in any patient in whom free tissue transfer is being considered.

Any mass involving the head and neck requires a thorough evaluation. Suspicion of a mass involving the posterior mandible and TMJ may not be obvious. Attention to facial symmetry or unilateral facial swelling may reveal an underlying mass. The parotid gland should be palpated for discrete masses or lesions (Fig. 16.1). The TMJ should be assessed, noting dysfunction of the joint (clicking, locking, crepitus) or trismus. The skin should be examined for any suspicious cutaneous lesions. Examination of the ears should be performed with evaluation for unilateral middle ear effusion or decreased hearing. Any finding of unilateral effusion warrants assessment with flexible nasopharyngoscopy. The oral cavity and oropharynx should be examined looking for any suspicious lesions. Any residual dentition should also be evaluated. In addition, the oropharynx should be examined for deviation of the tonsil or lateral pharyngeal wall, as this may represent a mass arising from the parapharyngeal space. Evaluation of the cranial nerves should be performed as their involvement may indicate malignancy, perineural spread, or involvement of the skull base. Special attention should be given to the facial nerve and any weakness should be documented as this may alter surgical planning. Fixation of the skin to the mass or considerable pain suggest a malignant tumor.

FIGURE 16.1 This figure demonstrates a tumor involving the parotid gland and TMJ (arrow).

Although multiple options exist for reconstruction of the TMJ, an understanding of when to use certain reconstructive techniques is critical to surgical success and patient outcome (Fig. 16.2). Perhaps most important in the decision-making process is assessment of the glenoid fossa. This can usually be evaluated preoperatively with the use of imaging studies but should also be assessed intraoperatively. If the glenoid fossa remains intact after completion of surgical ablation and confirmation of surgical margins, an osseous microvascular free flap can be used for reconstruction. Prior to inset of the free flap, the capsule of the glenoid fossa should be assessed, and if intact, the free flap may be inset. However, if no meniscus is present, soft tissue should be placed in the glenoid fossa. Multiple types of soft tissue have been used, all with some degree of success. Some examples include acellular dermis, free fascia, periosteum, or temporalis muscle. Absence of soft tissue between the end of an osseous free flap and the glenoid fossa results in bone on bone reconstruction, possibly resulting in ankylosis and subsequent trismus.

FIGURE 16.2 An algorithm for reconstruction of the TMJ.

If surgical ablation to clear surgical margins requires resection of the lateral skull base, glenoid fossa, and condyle, reconstruction with an osseous microvascular flap is contraindicated. Instead, a soft tissue flap should be used to protect the brain from a prosthetic or bony condylar reconstruction. As described below, a flap of adequate bulk should be used to restore appropriate facial contour, and maxillomandibular fixation (MMF) will be required with gradual use of guiding elastics to allow occlusal guidance.

There are no contraindications to reconstruction of the condyle using the fibula donor site unless the donor site lacks the three-vessel distal runoff required to harvest the flap.

Any lesion of the head and neck found on clinical examination and in agreement with the patient’s history should be biopsied appropriately.

Since primary tumors of the TMJ are rare, the head and neck surgeon more commonly encounters tumors involving the TMJ as a result of locoregional spread. These tumors may originate from the parotid gland, skin, or oral cavity. Imaging starts with a CT scan of the neck with contrast (Fig. 16.3) as this provides good soft tissue detail and excellent detail regarding bony involvement. MRI may also be necessary to provide improved soft tissue detail, to demonstrate potential neural involvement, and to provide additional information regarding skull base involvement. If involvement of the lateral temporal bone is suggested on initial CT scan of the neck, a dedicated temporal bone CT with and without contrast should be ordered. In patients with residual dentition, a Panorex may be beneficial to assess the quality of the remaining teeth. Establishing a histopathologic diagnosis is critical in providing the correct treatment plan. Tumors of the oral cavity, parotid gland, or skin can often be directly evaluated and biopsied in the clinic. If not directly accessible, histopathologic evaluation can be obtained through fine needle aspiration biopsy (FNAB), with or without the aid of imaging modalities. If the diagnosis is still in question, open biopsy may be necessary.

FIGURE 16.3 A. CT scan demonstrating a mass in the right parotid involving the TMJ. B. CT scan demonstrating a primary tumor of the TMJ.

Primary tumors of the TMJ are less accessible to biopsy than are tumors of the parotid, oral cavity, or skin. These tumors are typically benign hyperplastic lesions of the TMJ or malignancy of some tissue derivative of the TMJ. Given the anatomic location, imaging is the least invasive method of evaluating these neoplasms. A review of the literature by Shintaku et al. suggests that a panoramic radiograph is a good screening study for hyperplastic lesions of the TMJ being 97% sensitive and 45% specific. CT imaging is 70% sensitive and 100% specific in defining bony abnormalities. MRI is useful in evaluating TMJ lesions as it also assesses the surrounding soft tissue and is able to differentiate benign and malignant bony TMJ tumors with 44% sensitivity and 95% specificity. Imaging modalities combining functional and morphologic modalities, such as single positron emission CT and positron emission tomography, demonstrate improved sensitivity in assessing hyperplastic conditions of the TMJ. Single positron emission CT has demonstrated 100% sensitivity and specificity when assessing a hyperplastic condyle. Although these imaging modalities aid in treatment planning, histopathologic evaluation may also be necessary. Given the anatomic location of the TMJ and lack of access, FNAB with the aid of CT may be employed, or open biopsy may be required.

The peroneal artery and its paired venae comitantes supply the fibula flap and its overlying skin paddle. Routinely, these vessels are ligated during flap harvest. In some patients, perfusion to the donor extremity is dependent on the integrity of these vessels. Peripheral arterial occlusive disease (PAOD) of the lower extremity may prevent the use of the fibula flap due to inadequate collateral circulation to the donor extremity and also render the pedicle vessels unusable. Vascular anomalies have also been reported whereby sacrifice of the peroneal vessels would result in ischemia in the donor leg, as the peroneal vessels are always present. Assessment of the vasculature of the lower extremity should be performed preoperatively to prevent possible complications in the donor site.

Previously, lower extremity angiography was the “gold standard” for peripheral vascular assessment of the lower extremities. Although accurate, angiography is expensive, is invasive, and carries the risk of complications including embolus, hemorrhage, vessel injury, thrombosis, and contrast reaction. Magnetic resonance angiography and CT angiography have also been used to assess lower extremity vasculature. Although accurate and noninvasive, these studies are expensive and unable to determine the presence and quantity of cutaneous perforators. Color flow Doppler (CFD), an inexpensive, noninvasive ultrasound examination, uses both brightness node (B-mode) ultrasound and Doppler signal measurements to evaluate both quantity of blood flow and blood vessel patency in vessels as small as 1 mm in diameter. When compared to angiography, CFD is as sensitive and specific for vascular anatomy and PAOD. Examination of lower extremity vasculature using CFD in patients considered for fibula free tissue transfer is accurate in evaluating the vasculature of the lower extremity, diagnosing PAOD, and confirming patency of the flap pedicle vessel and donor site viability. Moreover, accurate identification and marking of cutaneous perforators from the peroneal artery can be achieved, allowing the surgeon to anticipate the viability of the skin paddle.

Resection of the mandible usually is a result of advanced stage squamous cell carcinoma or cancer of the parotid gland. More typical malignancies of the oral cavity rarely involve the TMJ, which may be explained by exit of the neurovascular bundle below the condyle or due to limited lymphatics in this region. Surgical resection of the typical cancer of the oral cavity involves osteotomies made inferior to the muscle attachments to the condyle with resection to obtain clear margins involving the body of the mandible, angle, and lower ramus (Fig. 16.4). The lateral pterygoid muscle and capsular attachments are preserved, and usually, a reconstruction plate can be fixed to the remnant of the condyle.

FIGURE 16.4 A surgical specimen demonstrating complete resection of the tumor of the oral cavity and the adjacent mandible.

The necessity of resecting the TMJ to achieve appropriate tumor margins for an advanced cancer of the oral cavity or parotid gland or primary tumor of the TMJ results in a challenging defect for the reconstructive surgeon. The TMJ is a diarthrodial joint playing a key functional role in jaw movement, mastication, and mandibular form. Unique to the TMJ is its ability in jaw opening to perform rotational movement followed by gliding motion, thereby allowing for maximal divergence. The condyle of the mandible, with its lateral pterygoid attachments, is separated from the glenoid fossa by the joint capsule and meniscus. Disruption of the condyle or meniscus can result in incapacitating instability of the joint, trismus, and chronic pain. Reconstruction of the TMJ provides a complex challenge to the surgeon to restore TMJ function, facial symmetry, occlusion, and mastication. A variety of reconstructive techniques to address defects of the TMJ and/or the condyle have been described in the literature.

Prosthetic Implants

Prosthetic implants have been used to replace the missing condyle. Although the use of a prosthetic implant is technically feasible, and may be considered for patients who are free flap candidates, it has fallen out of favor. Both titanium and stainless steel implants have been associated with complications including fracture of the plate or erosion of the skull base, heterotopic bone formation, intracranial perforation, and facial paralysis. Other alloplastic materials, such as Proplast-Teflon TMJ implants, have also been associated with implant migration, implant fracture, local erosion, middle cranial fossa perforation, and foreign body reaction. The use of alloplastic implants is generally limited to defects of the condyle and ramus and not the large mandibular defects common to head and neck cancer.

Soft Tissue Reconstruction

Although vascularized bone is now accepted as the standard tissue for reconstruction of the mandible, there are certain circumstances, which may preclude the use of these methods.

Patients with significant comorbidities and poor overall prognosis may benefit from less complex reconstruction allowing for more rapid recovery and less morbidity. Also, as described above, osseous reconstruction is contraindicated in patients requiring resection of the glenoid fossa. In contrast to defects of the anterior mandible, where rigid reconstruction using bone is of paramount importance to achieve optimal functional and aesthetic outcome, lateral defects are associated with significantly less morbidity; in fact most of the morbidity associated with these defects results from inadequate soft tissue lining and bulk, not lack of bone. Use of the rectus abdominis free flap (RAFF) to reconstruct defects of the posterior mandible and condyle has been described (Fig. 16.5). The anterolateral thigh flap has similar characteristics of bulk and minimal donor site morbidity, and other vascularized soft tissue flaps (both free and pedicled) may also be considered.

FIGURE 16.5 A. Surgical defect with condyle, ramus, and glenoid fossa absent. Temporal lobe of brain exposed. B. Harvested RAFF. C. Flap with majority of flap skin deepitheliazed. D. Postoperative result at 1 year. E. Postoperative occlusion at 1 year. F. Postoperative interincisal opening at 1 year.

Although this type of reconstruction has demonstrated good outcomes regarding speech and swallowing, disadvantages include worsened cosmesis due to blunting of the mandibular angle and varying degrees of malocclusion in dentate patients. Blunting of the mandibular angle can be addressed by secondary procedures addressing soft tissue fullness in the neck. Malocclusion, although difficult to eliminate, can be minimized with use of guiding elastics.

Fibula Free Flap

The fibula free flap has become the most popular tissue for reconstruction of the mandible providing up to 25 cm of bone if needed. This flap, based on the peroneal artery, and its associated venae comitantes, can be used as an osseous or osteocutaneous flap. The skin paddle is reliable and is based on septo-or musculocutaneous perforators. The fibula is dense cortical bone that can tolerate multiple osteotomies allowing precise contouring of the neomandible. Its successful use in reconstruction of the TMJ and condyle is well documented. Preoperative assessment, harvest, and inset of this flap for the isolated defect of the condyle are described in detail below.

Morbidity of the donor site of the fibula free flap is minimal. Closure of the donor site can usually be performed primarily; however, a split-thickness skin graft may be necessary as tight closure can result in compartment syndrome. Postoperatively, patients remain in a splint for the first 4 to 5 days and then are encouraged to ambulate with assistance. In general, most patients are able to resume full activity within a few weeks. The iliac crest is another choice when considering a vascularized bone flap and is most commonly used when significant atherosclerotic and peripheral vascular disease preclude use of a fibula free flap, as the deep circumflex iliac vessels supplying the iliac crest are likely to be patent. The description of this flap is beyond the scope of this chapter.

Usually, the leg ipsilateral to the defect is selected for flap harvest. This allows for the vascular pedicle to exit at the angle or body of the mandible for anastomosis to the vessels in the neck. It also presents the lateral aspect of the fibula for placement of fixation screws. Prior to harvest, approximation of the intermuscular septum is performed by assessment of the topographical anatomy and identifying the fibular head superiorly and the lateral epicondyle of the ankle inferiorly. A line is then drawn connecting these two anatomical landmarks. An elliptical fibular free flap skin paddle is designed over the distal two-thirds of the intermuscular septum, as the dominant septocutaneous perforators are located in this region. Flap harvest is then performed with a tourniquet pressure of 350 mm Hg. Incision of the anterior limb of the skin paddle is performed, carrying the incision through the skin and subcutaneous tissue. The peroneus longus is identified, and the fascia over this muscle is incised. Subfascial dissection is then performed from anterior to posterior allowing identification of the intermuscular septum and any septocutaneous perforators, which may be present. Care is taken not to violate the intermuscular septum as this may compromise the viability of the skin paddle. After leaving a cuff of muscle, the peroneus longus, peroneus brevis, and extensor hallucis longus are elevated off of the fibula allowing identification of the contents of the anterior compartment: the deep peroneal nerve, anterior tibial artery, and anterior tibial vein. With further medial dissection, the interosseus septum is identified. The interosseus septum is then incised proximally and distally along its length. The fibula is then divided with a bone cutting saw leaving at least 7 cm of proximal and distal fibula bone in place. Preservation of 7 cm of proximal and distal fibula bone ensures stability of the knee and ankle joint, respectively. Division of the fibula allows lateral retraction of the bone facilitating the remainder of the flap harvest. Division of the tibialis posterior muscle is performed along its entirety exposing the peroneal artery and venae comitantes. The distal peroneal artery and venae comitantes are ligated distally. Care should be taken to remain in the middle of the chevron oriented fibers of the tibialis posterior as dissection too close to the medial surface of the fibula may result in damage to the peroneal artery and venae comitantes. Incision of the posterior limb of the skin paddle is now performed with the incision carried through the skin and subcutaneous tissue allowing identification of the soleus muscle. The fascia of the soleus muscle is incised and a cuff of soleus muscle taken to protect any musculocutaneous perforators, which may perfuse the skin paddle. The fibula flap remains attached in the leg by the flexor hallucis longus. Moving from distal to proximal, the flexor hallucis longus is divided leaving a cuff of muscle attached to the flap. At this point, the flap remains attached by the proximal peroneal artery and venae comitantes. Prior to division of the pedicle, the tourniquet is released, flap perfusion assessed, and hemostasis achieved. The proximal pedicle is then divided and the graft used for reconstruction.

Exposure is performed though either a lip-splitting or cervical incision. After completion of the resection, exploration of the neck is performed and recipient vessels are identified in the neck (Fig. 16.6A). Recipient vessels may also be identified during neck dissection. Once recipient vessels have been identified in the neck and cleared of surrounding tissue, the proximal peroneal artery and paired venae comitantes of the fibula free flap are divided (Fig. 16.6B). The flap is transferred to the defect in the head and neck. As previously discussed, the use of a microvascular osseous flap should only be performed if the glenoid fossa is intact.

FIGURE 16.6 A. Surgical defect with the glenoid fossa intact. B. Harvested fibula free flap. C. Placement of acellular dermis in the glenoid fossa to replace the TMJ meniscus. D. Placement of the reconstruction plate. E. Contouring of the fibula flap to neocondyle. F. Wiring of neocondyle into the root of the zygomatic arch. G. Flap deepithelialized for cheek bulk. Small portion is left intact inferiorly to act as flap monitor. H. Postoperative result at 1 year. I. Occlusion with maximum intercuspation. J. Panoramic radiograph at 1 year.

Prior to insetting the free flap, the glenoid fossa capsule must be assessed. If absent, placement of soft tissue, such as a free fascial graft, temporalis flap, or acellular dermis, is required to prevent scar or callus formation within the reconstructed joint and subsequent ankylosis (Fig. 16.6C).

Maintenance of occlusion is critical to the ultimate success of reconstruction. Prior to inset of the free flap, the dentate patient is placed in MMF. For edentulous patients, the mandible is positioned midline to the maxilla. If feasible, reconstruction plates may be fashioned from the native mandible prior to resection, thereby obviating the need for intraoperative MMF.

A pear-shaped burr is used to smooth the outer cortex of the residual native mandible facilitating maximal contact of mandible to the reconstruction plate. A locking reconstruction plate is bent to the contour of the mandible, restoring occlusion after inset of the fibula flap. If desired, a prefabricated plate may also be used. The reconstruction plate is secured to the native mandible using appropriately sized bicortical screws (Fig. 16.6D).

The length of the total bone defect is measured and marked on the distal portion of the fibula. Measurement of the distance from the superior aspect of the condyle to the inferior aspect of the angle of the mandible ensures correct mandibular height. Prior to removing excess proximal fibula, the periosteum is cleaned from the proximal portion of the fibula, thereby freeing the pedicle. The excess fibula is then removed. Osteotomies are performed carefully to provide correct width, projection, and mandibular height. The distal portion of the fibula fitting into the glenoid fossa is contoured with a cutting burr to a condylar shape (Fig. 16.6E). The distal fibula bone is maintained superior to the reconstruction plate to facilitate its fitting into the joint. The fibula is secured to the reconstruction plate with nonlocking monocortical screws. If possible, each segment of fibula should be secured with two screws. Once in place, the neocondyle is secured in the joint space. Holes are drilled through the anterior lip of the glenoid fossa and through the neocondyle, and a 24-gauge wire is used to secure the two in place (Fig. 16.6F). A heavy nonabsorbable suture may also be used if preferred. The soft tissue paddle is then inset and deepithelialized as necessary (Fig. 16.6G). The microvascular anastomoses are then performed in the usual fashion. Figure 16.6H–J demonstrates the postoperative result at 1 year.

Condyle Transplant

When oncologically safe, reconstruction of the TMJ using an osseous microvascular flap with a nonvascularized condylar transplant has also been described. At the time of reconstruction, the condyle, which has been removed with the resected specimen, is attached to the free flap with miniplates. This technique allows precise positioning of the osseous microvascular flap into its native position in the glenoid fossa. Facial symmetry is improved due to accurate graft length, proper positioning of the angle of the mandible, and subsequent restoration of preoperative posterior facial height. Dual joint function of the mandible may be preserved, allowing potential maintenance of preoperative occlusion. Partial resorption of the condylar transplant occurs in some patients; however, this has not been associated with a decrease in function, interincisal opening, position of the mandible, or abnormal symptoms such pain, trismus, or ankylosis. The degree of postoperative interincisal opening achieved is inversely proportional to use of radiation therapy and the amount of soft tissue resected, particularly in the infratemporal fossa.

Postoperatively, patients are maintained on a soft diet for 3 months to allow for the appropriate healing. I use elastic bands to guide occlusion when necessary. In the event of malocclusion, I typically manage this aggressively with orthognathic guidance using either arch bars with elastic bands or a retainer. The donor site is typically managed with conservative wound care.

A retrospective case series of 518 consecutive mandibular reconstructions performed at our institution identified 44 patients requiring TMJ resection and reconstruction. Flap selection was performed according to the algorithm described above. Perioperative complications were assessed. Of the 44 patients, there were no flap failures. Malocclusion occurred in 3/7 dentate patients reconstructed with soft tissue and in 2/24 dentate patients reconstructed with bone. Patients were treated with guiding elastics for a mean of 23 days. Other perioperative complications included two cases of temporary facial nerve paralysis and three instances of wound infection.

Other reported complications during free flap reconstruction of the TMJ include cerebrovascular accident, myocardial infarction, hypoperfusion of the free flap due to venous thrombosis of the microvascular anastomosis, and hematoma in the neck.

Functional outcomes after TMJ reconstruction with microvascular reconstruction are poorly documented. The same retrospective case series performed at our institution also assessed functional results after free flap reconstruction of the TMJ in regard to diet, speech, and maximal interincisal distance. Overall, 93% of patients resumed an oral diet. Thirty-eight percent of patients (16 dentate and 1 edentulous) achieved a regular diet, 54% (14 dentate and 10 edentulous) achieved a soft diet, and 3 patients remained PEG dependent. All 44 patients were intelligible over the phone. Maximal interincisal distance ranged from 2 to 4.5 cm with a mean of 3.6 cm. Patients who received radiation therapy experienced an average loss of 1.1 cm of interincisal opening. No statistical differences were found in any variable in patients undergoing fibula reconstruction with or without condyle autotransplantation.

• TMJ replacement with osseous free tissue transfer is reliable, is reproducible, and restores function. Optimal results rely on the new condyle being properly seated and maintained in the glenoid fossa.

• Soft tissue flaps are successful in situations not suitable for bony reconstruction of the condyle.

• Malocclusion may result if guiding elastics in dentate patients are not used to achieve a stable occlusion.

• Postoperative radiotherapy, though necessary in many cases, is a critical factor that limits joint function in this group of patients.

• Standard head and neck surgical set

• Reciprocating saw

I gratefully acknowledge the contributions of Thomas J. Gernon, MD.

SUGGESTED READING

Lindqvist C. Erosion and heterotopic bone formation after alloplastic temporomandibular joint reconstruction. J Oral Maxillofac Surg 1992;50:942–949; discussion 950.

Lindqvist C. Rigid reconstruction plates for immediate reconstruction following mandibular resection for malignant tumors. J Oral Maxillofac Surg 1992;50:1158–1163.

Kroll SS. Reconstruction of posterior mandibular defects with soft tissue using the rectus abdominis free flap. Br J Plast Surg 1998;51:503–507.

Wax MK, Winslow CP, Hansen J, et al. A retrospective analysis of temporomandibular joint reconstruction with free fibula microvascular flap. Laryngoscope 2000;110:977–981.

Abramowicz S, Dolwick MF, Lewis SB, et al. Temporomandibular joint reconstruction after failed teflon-proplast implant: case report and literature review. Int J Oral Maxillofac Surg 2008;37:763–767.