Marginal mandibulectomy: Oncologic and nononcologic outcome☆

Patients and methods 

The charts of all patients treated by the authors, who underwent marginal mandibulectomy between July 1991 and July 2001, were reviewed. Data collection included demographics; physical examination; tumor characteristics; clinical and radiologic relation to the mandible; surgical technique, with special attention to neck dissection and ligation of facial artery; radiation therapy; osteoradionecrosis; fractures; and recurrence.

Results 

Twenty-six patients underwent marginal mandibulectomy. One patient died of cardiac complications 2 weeks after surgery and was excluded from the study. The remaining 25 patients, whose ages ranged from 45 to 86 years old, were included in the study. Twenty-two patients had the primary tumor in the oral cavity, 1 in the oropharynx, and 2 had metastatic neck tumor abutting the inferior/lateral border of the mandible. The number of patients staged T1 to T4 were 6, 13, 2, 2, respectively. Two patients with metastatic neck disease had unknown primary tumors. All patients had tumor limited to the soft tissue, or attached to periosteum, without evidence of gross bony involvement.

Patient follow-up ranged from 1 to 9 years, with the median follow-up of 3.6 years. The patients underwent a vertical, horizontal, or oblique marginal mandibulectomy. Twenty-three patients underwent neck dissections, with 7 patients having bilateral lymphadenectomy via a visor flap. Bilateral facial artery ligation was performed in 6 patients and unilateral ligation on the remainder of the neck dissections. Eleven patients completed postoperative radiation therapy, 2 patients had had previous radiation, and 1 patient started but could not complete the course of radiotherapy.

Two patients with close margins had local recurrence at the primary site for a local control rate of 92%. Of the eleven patients who received postoperative radiotherapy, 4 patients (36.4%) developed ORN of the mandibular segment within 1 to 2 years after surgery. Two of these patients subsequently developed pathologic fractures. All 4 patients with ORN had 1 or both facial arteries ligated.

Two other patients who did not receive postoperative radiotherapy developed fractures in the mandibulectomy site—soon after the initial surgery. One patient had bilateral neck dissection, with bilateral facial artery ligation, and periosteal stripping. The other patient had had previous radiation for a prior, unrelated, squamous cell carcinoma of the pharynx.

Discussion 

This study corroborates the growing body of evidence in the literature asserting that marginal mandibulectomy is an oncologically sound procedure with very good local control rates.

It has been previously shown that local recurrence/control rates after segmental or marginal mandibulectomy are comparable.3, 4, 9 When recurrences do occur, they predominantly occur in the soft tissue, rather than the mandibular bone.10

Marginal mandibulectomy can provide not only good oncologic outcome but also greatly improves the patient's postoperative function and cosmesis by preserving a viable segment of mandible. To maintain the long-term viability of the remnant mandibular rim and prevent pathologic fractures, it is important to be aware of the blood supply to the mandible.6

Throughout youth and middle age, the predominant arterial supply to the mandible is via the inferior alveolar artery. However, Bradley et al11 showed that with maturity (>50 years of age) the subperiosteal vessels become the dominant blood supply to the mandible.11 These vessels are composed of the branches of the buccal, lingual, and facial arteries.12 Thus, it is important not only to maintain the integrity of the remaining periosteum itself but also to preserve its blood supply. The facial artery is commonly ligated during neck dissection, although not always necessarily. Komisar and Barrow6 stressed the importance of maintaining the periosteal blood supply, including the facial artery. In their series, no patients developed pathologic fractures or ORN.6

In contrast, our series included 4 pathologic fractures—2 associated with ORN and 2 secondary to poor technique. In all 4 cases, 1 or both of the facial arteries were ligated compromising the blood supply to the remaining mandible. In 1 patient, the entire periosteum was stripped, leaving behind no major blood supply.

Of the 11 patients who underwent postoperative radiotherapy, all 11 had at least a unilateral facial artery ligation at the time of neck dissection. Within 2 years, 4 of 11 patients developed ORN; 2 with consequent fractures. Ionizing radiation can lead to injury and subsequent fibrosis to vessels. Thus, preservation of the blood supply is important to maintain the viability of the remaining margin of bone. However, because clinically apparent level 1 metastases are often the main findings justifying neck dissection, preservation of the facial artery is not always oncologically possible.

We now strongly recommend preservation of the facial artery, if oncologically possible, in patients undergoing neck dissection with marginal mandibulectomy. This goal is certainly achievable during elective neck dissection for the N0 neck.

Conclusion 

Our results confirm that marginal mandibulectomy provides an oncologically sound approach to tumors abutting the mandible, with very good local control rates. Attention should be given to the preservation of the periosteum of the remaining segment, as well as the facial arteries, to prevent the immediate postoperative morbidity of mandibular fracture. When postoperative radiation is given, the surgeon should take into account the long-term complication of osteoradionecrosis.