Introduction

Oral tumors are common in dogs and account for 6% of all canine tumors.¹ Surgical resection is the most economical, rapid, and curative treatment available for most localized tumors. Bone reconstruction after bony resection is rarely performed in dogs because they can tolerate aggressive surgeries such as mandibulectomy, maxillectomy, and orbitectomy. However, the mechanical and cosmetic abnormalities may be unacceptable to some owners.^2,3 In human medicine, frozen bone treated with liquid nitrogen is used as autograft material for patients with primary and metastatic bone tumors and mandibular tumors.^4,5 The purpose of this paper is to present a liquid nitrogen-treated autograft technique that was used to preserve mandibular function and cosmetic appearance after the wide local resection of oral neoplasms in a dog.

Case Report

A 10 yr old intact female German shepherd dog was referred to Iwate University Veterinary Teaching Hospital with a large mandibular mass. Two mo prior to this presentation, the owner had noticed a small mass (approximately 1.5 cm) on the right mandible of the dog. The mass grew rapidly by approximately 10 cm in 2 mo, and the patient produced blood-tinged saliva. Results of a complete blood count and serum biochemical parameters were within the normal reference range.

The tumor presented as a large, irregular, raised, cauliflower-like ulcerated mass. The tumor size was measured macroscopically. The length, width, and height of the mass were 10.0 cm, 4.0 cm, and 3.6 cm, respectively (Figure 1A). There was no evidence of pulmonary metastasis on three-view thoracic radiographs and helical computed tomography (CT) scans. The tumor extended from canine tooth to the first molar of the right-side mandible, but it did not extend over the midline. The CT scan revealed a periosteal reaction in the mandible from the third incisor tooth to the fourth premolar (Figure 1B). Palpation and CT imaging detected no lymphadenomegaly. Fine-needle aspiration was performed on the left and right mandibular lymph nodes. There was no evidence of metastatic disease.

View Figure

• Download as Powerpoint
• Download Figure

An incisional biopsy was performed. Histopathology revealed a peripheral odontogenic fibroma and malignant melanoma. The tumor was classified as a stage III oral tumor, based on the TNM staging system (T3b, N0, M0). A hemimandibulectomy was proposed; however, the owner refused having the dog undergo this procedure and insisted on preserving its mandible. Therefore, reconstruction of the mandible with a liquid nitrogen-treated frozen bone autograft was considered. This operation was approved by the Animal Experimental Ethics Committee of Iwate University (accession number A 201230). The owner provided informed consent and a letter of acceptance before the dog underwent treatment.

The dog was premedicated with intravenous midazolam and fentanyl citrate^a and then anesthetized with propofol^b. Anesthesia was maintained with sevoflurane^c. Fentanyl and remifentanil^d were used in the perioperative period for pain control.

The tumor was resected en bloc between the mandibular symphysis and the first and second right molars by unilateral rostral mandibulectomy using an oscillating saw. To obtain 2-cm surgical margins, the surgeon determined the incision lines intraoperatively with the naked eye and with CT images. Before freezing the resected mandibular segment, the tumor and soft tissue were removed from it by using a periosteal elevator and rongeur. This procedure was performed outside of the dog and at another operating table. Surgical gloves and instruments were changed to prevent tumor seeding during the freezing treatment in liquid nitrogen. The resected mandible was frozen in liquid nitrogen (-196°C) for 20 min, thawed at room temperature (25°C) for 15 min, and thawed in water (30°C) for 10 min.^5,6 The bone was repositioned with a seven-hole locking compression plate^e with 2.7 mm locking screws and wires. The mandibular symphysis was fixed by intraosseous wire. The repaired mandible was covered with a labial mucosal–submucosal flap and a sublingual mucosa. Blunt and sharp dissections were used to separate the labial mucosa and submucosa from the remainder of the lip. The flap and sublingual mucosa were sutured with simple interrupted sutures of a synthetic absorbable material^f. An esophagostomy tube was inserted for nutritional management. Two days after surgery, the dog could ingest food orally. One wk later, the incisors were lost. However, the other teeth remained.

The surgically obtained tissue samples were examined histopathologically. The tumor was diagnosed as a peripheral odontogenic fibroma and malignant melanoma. These were two separate tumors that were adjacent histopathologically, rather than a collision tumor. Adjuvant therapies were recommended; however, the owner declined them.

To assess bone healing and lung metastasis, CT scan was performed under general anesthesia. Thirty-five days after reconstruction surgery, callus formation was evident on CT (Figure 1C). The periosteal reaction, which was present before surgery, disappeared. A portion of the cortical junction was also present on CT imaging. Seventy-five days after implantation, CT demonstrated the presence of a cortical junction (Figure 1D).

One hundred fifty-nine days after reconstruction surgery, the malignant melanoma recurred at the rostral of the mandible and invaded the left rostral mandibular bone and the rostral part of the graft. In addition, the tumor progressed across the midline to the left canine tooth area. There was no evidence of pulmonary metastasis on thoracic radiographs and CT imaging. The left and right mandibular lymph nodes were not enlarged. There was no lymph node fixation. One hundred sixty-nine days after the reconstruction surgery, a subtotal hemimandibulectomy was consequently performed. The extirpated specimens were examined histopathologically. New bone tissue had formed in the gap between the frozen bone and the host bone (Figure 2A). Osteocytes, osteoblasts, and blood vessels were present in the regenerated bone. Histopathology showed good synostosis. There was a mixture of new bone tissue with vital osteocytes and graft bone (i.e., dead bone), which was characterized by empty osteocyte lacunae (Figure 2B).

View Figure

• Download as Powerpoint
• Download Figure

Long-term follow-up information from the owner was obtained by telephone because the owner had moved, and it was hard for the owner to come to the hospital. The owner said that the dog had increased salivation and dropped food and water while eating and drinking after the subtotal hemimandibulectomy. After the mandibular reconstruction, the dog could eat and drink normally, and hypersalivation did not occur. The dog died 312 days after the reconstruction surgery. According to the owner, the dog was healthy and did not have any recent complications. The owner did not elect to have an autopsy performed.

Discussion

By using liquid nitrogen-treated bone, reconstruction of the mandible was performed, and good synostosis was observed in the dog reported in this paper. The cosmetic appearance was preserved. She recovered the ability to eat and drink 2 days after the repair. The reconstruction did not cause apparent pain or discomfort to the dog. There were no expected complications (e.g., infection, fracture, nonunion, or failure of the graft resulting from using dead bone) except tumor recurrence. The incisors were lost as a minor complication because periodontal tissue devitalization probably occurred by freezing. If this dog had survived longer, other teeth may also have been lost. Synostosis was confirmed by CT imaging and macroscopic examination. Histological examination of the specimens revealed osteogenesis at the host–graft junction. A mixture of dead bone and new bone was observed. The graft bone acts as a scaffold or template for a new bone formation.⁷ The process of bone graft incorporation involves “creeping substitution,” whereby gradual vascular ingrowth, resorption, and replacement of necrotic bone progressively occur.^8,9 In our case, creeping substitution is possibly the main way in which bony union occurred. Therefore, the liquid-nitrogen-treated graft was incorporated with the host bone, and the mandibular function was preserved.

Reconstruction of mandibular defects has been reported in veterinary medicine. In previous reports, the mandibular defect was filled with autogenous cortical grafts (e.g., full-thickness rib segments or full cortical diaphyseal ulnar segments) or with an absorbable collagen sponge (e.g., recombinant human bone morphogenetic protein-2).^10,11 In comparison to these methods, the advantages of liquid nitrogen treatment include the following: perfect fit between graft and host bone, less invasive surgical procedures, and no need for cytokines like recombinant human bone morphogenetic protein-2.

Many methods have been developed to reuse resected bone for reconstruction, including irradiation, autoclaving, and pasteurization.^12–15 However, these methods require special equipment or strict thermal control. By contrast, liquid-nitrogen-treated bone methods use simple equipment, and liquid nitrogen is inexpensive. In addition, there is no significant difference in compression strength between the intact bone and the liquid-nitrogen-treated bone.⁶ The advantages of this method are its simplicity, its osteoconduction, its short treatment time, the perfect fit and sufficient biomechanical strength of the graft, and the fact that the use of a bone bank is unnecessary.¹⁶

Mandibular reconstruction is uncommon in dogs because client satisfaction is high (85%) with the cosmetic and functional outcome after mandibulectomy alone, and dogs usually tolerate it well.² However, reported postoperative complications include incisional dehiscence, increased salivation, mandibular drift and malocclusion, and difficulty prehending food.¹⁷ Mandibular reconstruction may improve the quality of life of dogs.

In the animal in this report, the malignant melanoma recurred 159 days after surgery, and the dog survived for 312 days after the first surgery. The reasons of recurrence were conceivable as a failure of margin assessment and probably represented insufficient surgical resection. There would be little possibility of failure of tumor devitalization. Local recurrences have also been reported in human medicine. Tsuchiya et al. concluded that recurrence happens in the soft tissue and probably represents satellite lesions from insufficient surgical resection.⁵ The tumor was devitalized by one freezing cycle of 20 min in this report. In human basic research, one freezing cycle and two freezing cycles were compared using human osteosarcoma tissue, which was placed in the cortex of a metatarsal bone of a Holstein cow (the shape of the cow's metatarsal bone is similar to that of the human tibia).⁶ The one-cycle liquid nitrogen method sufficiently achieves tumor cell extinction and produces the same results as those obtained with the two-cycle method. In addition, one freezing cycle of 20 min was safe in this clinical study.⁵ The canine mandible is thinner than the human tibia; therefore, the melanoma cells were probably killed sufficiently.

Surgical margin assessment is an issue to be addressed in the future. In human medicine, MRI is particularly advantageous in evaluating the oral cavity and depicting better the anatomy of this region, compared to CT.¹⁸ MRI is also the preferred imaging modality for evaluating bone marrow invasion, which are findings that are critical for treatment planning. Further study is necessary, but MRI may be useful to evaluate the surgical margins of oral tumors in veterinary medicine. In addition, the surgical margins have to be indicated with ink or some other method. The soft tissues that were incised (and later closed over the autograft) should have undergone a biopsy examination for any potential presence of residual tumor.

The cause of death could not be settled, but it may be that the dog died of a metastasis from the tumor. The most common cause of death with malignant melanoma is lung metastasis.¹⁷ Median survival times for dogs with oral melanoma treated with surgery is approximately 17–18 mo with stage I disease, 5–6 mo with stage II disease, and 3 mo with stage III disease.¹⁹ This dog lived a long time for having a large, rapidly growing melanoma.

Conclusion

Liquid-nitrogen-treated autograft was successfully applied to the mandibular reconstruction of a dog that had an oral melanoma. The synostosis was confirmed, but, unfortunately, the melanoma recurred. This mandibular reconstruction method has not been adequately established in veterinary medicine, and further evaluation is necessary.