Combined Dorsolateral and Intraoral Approach for the Resection of Tumors of the Maxilla in the Dog
This paper describes in detail a combined dorsal and intraoral approach for maxillectomy for tumors involving tissues more caudal to the third premolar. The only intraoperative complication was that of blood loss, with six out of 20 dogs requiring a single unit of blood. Histopathologically clean margins were obtained in 14 of the 20 cases, with a recurrence rate of 50% in these dogs and a median time to recurrence of 24 months. This represents an improvement in outcome over previously reported studies, and the authors postulate this is due to the better exposure and access to the area afforded by the combined approach over the standard intraoral approach.
Introduction
The most common tumors occurring on the maxilla and rostral skull of the dog are fibrosarcoma (FSA), osteosarcoma (OSA), squamous cell carcinoma (SCC), malignant melanoma (MM), and epulides.1–7 Many of these tumors (i.e., FSA, OSA, and SCC) are locally invasive, and the more caudally positioned tumors often only present when significant disease is present, probably because tumor growth in the caudal oral cavity is not readily noticed by the owner. These two factors combine to make surgical resection of these tumors a challenge. Many studies28–11 state that tumor-free margins are associated with a more favorable prognosis; however, only one study has correlated outcome with the histopathological appearance of the margins of resected tissue for maxillectomies in the dog.12 These authors found that, regardless of tumor type and other variables, dogs with more caudally positioned tumors had a worse prognosis, as did dogs where tumor-free surgical margins were not obtained.
The intraoral approach to maxillectomy was originally described in dogs by Withrow, et al.13 and Salisbury, et al.8 and appears to have been the technique used most commonly since.29–11 This technique works well for tumors located on the hard palate or near the dental arcade, but for more dorsally located tumors and for the more caudal tumors, the exposure afforded by this technique is poor.11
This paper describes in detail a modification of the existing procedure which allows greater exposure and better visualization of the extent of the tumor when making the dorsal osteotomy for a maxillectomy, and for the dorsal and caudomedial osteotomy when maxillectomy is combined with a partial orbitectomy. This paper also reports on the effectiveness of this procedure in facilitating the gaining of tumor-free margins, and it describes the intraoperative and postoperative complications associated with this procedure. The “combined approach” described in this paper has been alluded to in another paper14 in connection with performing total orbitectomies where the tumor extended to involve the oral cavity, but the technique has not been reported or comprehensively described as a specific modification for maxillectomy.
Materials and Methods
Medical records of dogs undergoing a combined-approach maxillectomy, with or without partial orbitectomy, were reviewed. Dogs were included in this retrospective study if they had surgery performed for histopathologically confirmed oral tumors where the caudal extent of the resection extended beyond the third premolar (PM3), the tumor did not cross the midline, and the tumor had not originated within the nasal cavity. Cases were selected from the period of April 1996 to December 2000. All surgeries were performed at the Animal Cancer Center, Colorado State University. All cases were evaluated preoperatively for tumor location, regional lymph-node involvement by fine-needle aspiration if lymph nodes were palpably enlarged, and pulmonary metastasis by thoracic radiography. Tumor location was based upon physical examination under general anesthesia and by computed tomography (CT). All dogs had a CT scan performed, and this was used to determine the extent of the tumor and thus the extent of resection required [Figure 1]. Margins of resected tissue were inked,a and a drawing of the resected tissue was made, with the inked margins indicated in order to allow the pathologist to properly orientate the specimen. All margins were examined histopathologically. A tumor-free margin, or “clean margin,” was defined as absence of microscopical tumor extending to the incised tissue margin. A “dirty margin” was defined as one where the neoplastic tissue extended to the edge of the resected tissue. All complications encountered during surgery and immediately afterward, while the animals were in the hospital, were recorded. The dogs were examined 2 weeks postoperatively, with particular attention being paid to the presence of dehiscence and ulceration of the upper lip by lower teeth. Follow-up data was recorded from telephone interviews or revisits to the Animal Cancer Center.
Surgical Method
The dog is placed in lateral recumbency with the affected side uppermost and a mouth gag in place on the lower side. The lateral side of the face is clipped and surgically prepared from the nasal planum, over the midline on the contralateral side, ventrally to the mucocutaneous junction, and as far caudal as dictated by the margins of the lesion. The eye is usually clipped and draped within the surgical field and is left open. The oral cavity is irrigated with 10% povidone-iodine solution, which is first diluted 1:10 with tap water. Perioperative antibiotic therapy consists of intravenous [IV] cephazolin sodium (22 mg/kg body weight, 20 to 30 minutes prior to surgery and every 90 minutes thereafter during surgery). As these are potentially painful procedures, postoperative pain relief is initiated preoperatively, with opioids being given via the IV (e.g., morphine at 0.5 mg/kg body weight) or intramuscular (IM) route (e.g., morphine at 1.0 mg/kg body weight) prior to surgery to “protect” the patient from the surgical stimulus.15 These are administered with a preoperative nerve block of the infraorbital nerve, utilizing bupivacaine.16
The dorsal incision is created first, through the skin just lateral to midline of the dorsal aspect of the nasal cavity. This is continued through the subcutaneous tissue, between the paired levator nasolabialis muscles, and down to bone. If removal of the zygomatic arch or tumor in this area is required, the incision is carried caudally by sweeping down below the eye, following the most prominent aspect of the zygomatic arch [Figure 2].
The ventral aspect of the globe is separated from the dorsal zygoma with a combination of sharp and blunt dissection, leaving the conjunctival sac intact while the masseter muscle is separated from the ventral aspect of the zygomatic arch with sharp and blunt dissection. The dorsal osteotomy for the maxillectomy can be prepared by reflecting the periosteum and associated soft tissue with a periosteal elevator. A second incision is made in the buccal mucosa immediately dorsal to the gingiva, but dictated by the margins of the tumor. This incision is made 1 cm from the tumor. The mucosa is sharply dissected down to the bone and undermined with scissors until connection with the dorsal incision is made. This necessitates severance of the infraorbital nerve and associated vessels. The vessels are individually ligated. This effectively creates a bipedicled flap of the skin, buccal mucosa, and associated soft tissues of the lateral aspect of the nasal cavity, which can be retracted both dorsally and ventrally [Figures 3A, 3B].
During this dissection, the facial vein is ligated at the most dorsocaudal aspect of the incision [Figure 4]. The facial vein from the level of the dorsal nasal tributary is preserved if possible, in order to facilitate venous drainage from the muzzle.
Retraction of this flap allows visualization of the whole of the lateral aspect of the maxilla, and the sites of the rostral, dorsal, and caudal osteotomies can be determined. The dorsal and cranial osteotomies in the maxilla are created at this time with an oscillating saw or osteotome. Next, an incision is made through the mucoperiosteum of the hard palate, immediately medial to the dental arcade, or as dictated by the margins of the tumor. The more medial the incision, the greater the likelihood of severing the greater palatine artery. If this occurs, it is then ligated. An osteotome is used to create an osteotomy through the palatine bone of the hard palate to delineate the cranial and caudal extent of the resection. An oscillating saw is then used to make the caudal osteotomy in the maxilla, joining the dorsal maxillary osteotomy and the caudal aspect of the osteotomy in the hard palate. If an inferior orbitectomy is combined with this approach, the final osteotomy is created through the medial aspect of the orbit with an osteotome. This is angled from the orbit to the medial aspect of the caudal molar tooth. The globe is protected with a small retractor to prevent iatrogenic damage during creation of the osteotomy sites. For the more caudal resections, the zygomatic arch also needs to be transected. This is performed early on in the procedure. Figure 5 shows an example of where the lateral cuts in the bone might be placed.
Hemorrhage is controlled after removal of the free segment of bone. This is achieved with a combination of cautery and metallic clips.b Hemorrhage can be brisk and profuse, especially as the infraorbital or maxillary artery is cut during the final osteotomy. In all cases, invasive blood pressure monitoring (via an arterial catheter placed in the dorsal metatarsal artery) should be performed, because this gives the most accurate measure of blood pressure and hemorrhage can be profuse with this type of surgery. The surgery site is copiously lavaged and suctioned, and the resected specimen is grossly inspected for adequacy of margins. If the margins appear inadequate, further tissue is resected. Closure is commenced dorsally by a simple continuous suture through the subcutaneous tissues with 3-0 polyglyconate,c followed by continuous closure of the skin with 3-0 nylon.d If a partial orbitectomy is performed, the masseter muscle is apposed to the fibrous tissue ventral to the globe, and the remaining tissues are closed routinely. Orally, the thick fibrovascular layer of the lip is sutured to the hard palate submucosa with interrupted mattress sutures of 2-0 polyglyconate, or it can be anchored via small holes placed through the bone of the palate. If tension exists during closure of the oral cavity, it is recommended that multiple small holes be placed in the palatine bone with a Kirschner wire to facilitate a strong, double-layered closure. The buccal mucosa and mucosa of the hard palate are opposed with a simple continuous suture pattern of 3-0 polyglyconate. No attempt is made to close the dead space between the oral and lateral nasal incision. Drains are not required, as the surgical site invariably opens into the nasal cavity and drainage occurs via the nasal orifice.
Further doses of opioids are administered as required during surgery and postoperatively; good analgesia is achieved using a continuous fentanyl infusion for 24 hours (2 to 6 μg/kg body weight per hour) and an injectable non-steroidal anti-inflammatory drug (NSAID) (such as carprofen [4.0 mg/kg body weight], meloxicam [0.2 mg/kg body weight], or ketoprofen [1.0 mg/kg body weight]) administered immediately postoperatively. The dogs are started on either oral morphine (0.5 to 1.0 mg/kg body weight per os [PO] up to tid for 3 to 5 days) or a fentanyl patch, concurrently with an oral NSAID for 7 to 10 days to provide analgesia. Antibiotics are continued overnight only. An Elizabethan collar is applied in the immediate postoperative period to minimize self trauma. Sneezing copious amounts of bloody mucus is expected in the immediate postoperative period, necessitating frequent cleaning; however, within 24 hours after surgery, hemorrhage has typically stopped. There may be a small amount of subcutaneous emphysema noted, but this generally does not get worse, and the skin over the surgical site can be seen to move with respiration in the early postoperative period. The patient is offered oral fluids and soft food the morning following surgery. Patients are discharged 24 to 48 hours following surgery.
Statistical Analyses
Descriptive statistics were used to describe the outcome in 20 dogs that underwent this surgery. Using a two-tailed Student’s t-test, surgical times and the drop in packed cell volume (PCV) were compared between the dogs that were given a blood transfusion and those that were not. The correlation coefficient between surgical time and drop in PCV was also calculated. The use of further statistical analysis was considered inappropriate due to the small number of dogs and differing tumor types.
Results
The technique described above was used on 20 dogs between April 1996 and December 2000. These dogs were referred to the Animal Cancer Center at Colorado State University for further treatment or diagnosis of masses located caudally on the maxilla. Tables 1 and 2 list the clinical details of each individual case. The most common reason for presentation to the referring veterinarian was a swelling noticed below and slightly rostral to one eye. The age range of the dogs was 3 to 15 years, with a mean age of 8.6 years. All of the dogs were ≥20 kg, except for one 10-kg dog, and the mean weight for all dogs was 31 kg. There were nine neutered males, seven neutered females, two intact males, and two intact females. Staging of the dogs prior to surgery showed no evidence of visible metastatic disease.
The tumor types involved were sarcoma/FSA (n=9, which included one myxosarcoma and one undifferentiated sarcoma), OSA (n=4), MM (n=1), SCC (n=2), multilobular osteochondrosarcoma (MLO; n=2), parosteal OSA (n=1), and acanthomatous epulis (n=1). Surgery was planned in all cases using a CT scan. This surgical technique was chosen most commonly for tumors located at the caudolateral maxilla, just below the ventral rim of the orbit, and in all cases the resection extended caudally beyond PM3.
The technique was performed in all cases as described. There were no technical problems associated with the surgery. The only intraoperative problems encountered were hypotension (mean blood pressure, <70 mm Hg for >5 minutes), thought to be due to blood loss (n=11), and blood loss that was regarded by the surgeon as significant without measurable hypotension (n=4). In six cases, all of which either were hypotensive or suffered significant blood loss, a transfusion (1 unit of cross-matched blood) was given during anesthesia. The average drop in PCV was 15.4%. There did not appear to be any correlation between duration of surgery and drop in PCV (coefficient of correlation, 0.03). The mean surgical time was 91 minutes for the group that was administered blood, and the mean drop in PCV was 12.8% (measured at the end of surgery, which included administered blood in some cases); the mean surgical time was 77 minutes for the group not given blood, and the mean drop in PCV was 16.6%. There was no significant difference between the two groups for the duration of surgery or the drop in PCV (P>0.05).
In each case, the surgeon was afforded excellent visualization of the planned resection margins. Histopathologically clean margins were obtained in 14 (70%) of the 20 cases as a result of the initial resection. Four of the nine FSA, one of the two SCC, and one of the two MLO cases all had “dirty resection” margins. These results were not predicted by the gross appearance (i.e., the margins appeared grossly clean). “Dirty” resection margins were not all consistently located in any one (e.g., caudal) area.
One dog died during radiation following incomplete resection of a FSA, approximately 2 weeks following surgery. Necropsy was not permitted. Of the remaining 19 dogs, three were documented to have lung metastases during follow-up; all had clean resections, one for MM and two for OSA. No local recurrence was documented in these cases; however, metastases were seen 3, 3.5, and 8 months post-surgery, and the dogs were euthanized shortly thereafter. Of the remaining 11 dogs with clean resections, seven were documented to suffer local recurrence (FSA, n=4; OSA, n=2; and epulis, n=1). The median time to local recurrence in this group was 24 months. Two of these dogs were lost to further follow-up, and the median survival time of the remaining five dogs was 31 months. The remaining four dogs with clean resections are alive and disease-free at a mean of 23.5 (median, 23.5) months postsurgery (55, 33, 14, and 11 months). Two dogs with clean resections and local recurrence underwent radiation therapy (one palliative, one curative intent). Sixty-four percent (9/14) of dogs having clean resections were alive 2 years following surgery.
In the five of 19 remaining dogs, all of which had “dirty resections,” one dog was euthanized shortly following surgery because the owners believed it never recovered properly from the anesthetic (the owners considered the dog’s behavior to be altered following anesthesia and surgery). One dog had no local recurrence at 2 months postsurgery, but then it was lost to follow-up. One dog (that had a SCC) underwent radiation therapy (51 Gy) and was euthanized 32 months after surgery for progressive hind-limb paralysis. The cause of the paralysis was not investigated. Two dogs suffered local recurrence at 2 and 12 months postsurgery, and they were euthanized 2 and 21 months postsurgery, respectively; both had FSAs. In the four of five dogs with “dirty resections” not lost to follow-up, the overall median survival time was 11.5 months (one MLO and three FSAs or undifferentiated sarcoma).
Six dogs received adjunctive therapy (i.e., chemotherapy for two OSA; radiation therapy for an OSA, SCC, and two FSA). All of these had clean resection margins except for the SCC and one FSA. One of these dogs was lost to follow-up; one died during radiation therapy; and the median survival of the other four was 10 months, with the dog having the SCC and dirty resection surviving the longest.
Long-term complications occurred in three dogs; one required surgery to close a 0.5-cm diameter oronasal fistula at 7 months postoperatively, and two dogs developed ulceration of the upper lip as a result of impingement of the lower teeth on the lip. Both of these dogs required the teeth involved to be filed down, and the teeth were capped in one case.
Discussion
Most authors state that the goal in resection of oral tumors is attaining tumor-free margins;128–11 however, only two studies have correlated histopathological status of the resection margins with local recurrence or survival.1217 One of these studies12 concerned maxillectomies, and tumor-free margins were found to have a significant positive influence on survival; dogs in which tumor cells were found histopathologically extending to the peripheral excised margin were 3.6 times more likely to die from the tumor compared to dogs where margins were considered tumor free. Also in that study, dogs with tumors located caudally (i.e., caudal to PM3) were 4.3 times more likely to die from the tumor compared to dogs with tumors located rostrally to PM3. The standard maxillectomy technique involves an intraoral approach to the tumor; however, it has been stated that this technique makes effective resection of caudolateral-based tumors difficult.1118
The combined technique for maxillectomy described in this paper greatly facilitates surgical resection of both the more caudally based maxillary tumors and the more laterally based maxillary tumors. If the combined approach is better for the more caudal and dorsal tumors, then it should result in clean margins more often, decreased local recurrence, and better survival times. This cannot be proven by this study, but review of the literature suggests that the combined approach facilitates resection of the more caudal and dorsal tumors. In the dogs presented here, tumor-free margins were obtained in 70% of cases overall. Schwarz, et al.12 obtained tumor-free margins in 71% of cases in a group of dogs of similar age and weight that were suffering from malignant oral tumors. However, in the study by Schwarz, et al.,12 between 43% and 71% of the malignant tumors (which varied depending on tumor type) were located rostral to the canines, and only 22% to 50% were located caudal to PM3; potentially the surgery was easier than for more caudally located tumors in these cases. In the group of dogs presented here, 100% of the tumors extended caudal to PM3. It is the authors’ opinion that the combined approach allows tumor-free margins to be obtained more frequently than if the standard approach is used for tumors of the caudal and caudolateral aspect of the maxilla. There is not enough detail in the published material to permit statistical comparison of the data from Schwartz, et al. with the present data, and a definitive statement about whether or not tumor-free margins are obtained more frequently with the combined approach must await further comparative study. Schwarz, et al.12 reported a local recurrence rate of 41% for malignant tumors in dogs having clean resections. The local recurrence rate in the dogs presented here was similar at 50%. In this group of dogs, the overall median survival time for all dogs that were not lost to follow-up or were not still alive at the time of writing, was 16 months. Schwarz, et al.12 reported a median survival time of 8 months for dogs undergoing surgery of the maxilla for malignant tumors. In this study, 64% of dogs were alive at 1 and 2 years; Schwarz, et al. reported figures of 68% and 50%. The impact of adjunctive therapy administered to six dogs in this report is difficult to assess due to the small numbers.
Blood loss was the only documented intraoperative complication, and this caused hypotension in 11 of the 20 cases. Six of the cases were deemed to require a blood transfusion due to the amount of blood loss. This has not been noted as a common complication in other reports of maxillec-tomy128–11 and is probably a result of the caudal nature of the resections carried out in this study. It is the authors’ opinion that a more caudally based resection of the maxilla results in greater hemorrhage than a slightly more cranially based one, as the vessels encountered are larger and hemorrhage occurs rapidly. The major contributor to this bleeding is the maxillary artery as it courses to become the infraorbital artery. It is very difficult to clip or ligate this artery prior to transection, so the most caudal osteotomies are performed last. Following prompt removal of the maxilla, the vessel is ligated or clipped. Other methods of hemostasis (e.g., direct pressure, topical application of adrenaline, application of gelatin-impregnated sponge, and use of bone wax) should be used as previously described.8 The authors do not consider temporary ligation of the carotid arteries to be necessary if blood is available; however, if blood is not available, this technique might be useful. In the cases presented here, there was no association between drop in PCV or the need for a blood transfusion and the total surgical time. However, all these surgeries were carried out as quickly as was safely possible, and it is the authors’ opinion that speed is very important in preventing even greater blood loss. Although there was no statistically significant difference in surgical time between the cases that received blood and those that did not in this series, the surgical time was longer in those that received blood. However, this is not reflected in the PCV value, because the data is skewed by the fact that PCVs were measured after surgery when many of the dogs receiving blood had received a significant proportion of the administered blood. The surgery described here is associated with significant blood loss, and this cannot be effectively stemmed until the mass is removed en bloc from the dog. Surgeons undertaking this procedure should consider the possibility that blood may be required intraoperatively.
Previous reports of maxillectomy involving ligation of the infraorbital artery have not documented any adverse effects on regional perfusion.28 The blood supply to the labial mucosa is via the lateral nasal artery and dorsal nasal arteries, both of which are branches of the infraorbital artery, and the angular artery of the mouth and the superior labial artery, both of which are branches of the facial artery. There is considerable anastomosis between arterial supply to the left and right muzzle and between branches of the infraorbital and facial arteries. Thus, when the infraorbital artery is ligated, the collateral circulation to the lip and muzzle is maintained from the facial artery and the contralateral infraorbital artery.19 Likewise, the mucosa of the hard palate is supplied by the paired major palatine arteries, and there are extensive anastomoses between the right and left major palatine arteries so that following ligation of one artery, mucosal circulation will be adequately maintained by the contralateral artery.
Other reported intraoperative complications associated with maxillectomy are disruption of the nasolacrimal duct and problems involving reconstruction. Immediate postoperative problems reported are difficulty breathing and subcutaneous emphysema. With caudal maxillectomies as described here, it is likely that the nasolacrimal duct will be damaged. Salisbury18 suggests placing a stent from the puncta of the nasolacrimal duct, through the nasal cavity, and out through the ipsilateral nostril if the nasolacrimal duct is damaged. This is left in place for 1 month to allow for epithelialization of a new duct. The authors of this report have not considered the excess tear spillage to be a significant enough problem to warrant such a procedure. In some dogs, tear spillage did occur. Owners of these dogs did not consider this a problem. Further work is needed to clarify the extent of the problem and how often tear spillage is associated with gross disruption of the duct, and the benefit of stenting. Due to the much better exposure over the standard intraoral approach to maxillectomy, reconstruction was not a problem in any of these cases. None of the cases here suffered respiratory difficulty postoperatively, but only half of the nasal cavity was entered. None of the dogs that underwent the procedure described here suffered postoperative emphysema, although the lateral muzzle skin does move slightly with air movement through the nose. This movement of skin is barely noticeable by 2 weeks postoperatively.
The main reported complication of maxillary surgery is that of dehiscence. In this series, it was documented in only one case. Schwarz, et al.12 found an overall dehiscence of 33%, with 80% of these occurring caudal to the canines, and similar findings have been reported by Harvey.20 The low incidence reported here might be due to better exposure and less tension on tissues at the time of closure as a result of raising the bipedicled flap. Occasionally reported is the problem of lower teeth impinging on the upper soft tissues following maxillectomy. This occurred in two dogs in this study, and one required burring down of the offending tooth to resolve the problem.
Conclusion
The combined approach to maxillectomy facilitates exposure and, therefore, resection and reconstruction for tumors located on or involving the lateral maxilla, caudal maxilla, or near the ventral rim of the orbit. There are no intraoperative complications apart from significant hemorrhage, and the postoperative complications appear to be minimal.
Yellow ink; Shandon Tissue Dyes, Pittsburgh, PA
Hemoclips; US Surgical Corporation, Norwalk, CT
Maxon; Sherwood, Davis and Geck, St. Louis, MO
Dermalon; Sherwood, Davis and Geck, St. Louis, MO
Citation: Journal of the American Animal Hospital Association 39, 3; 10.5326/0390294
Citation: Journal of the American Animal Hospital Association 39, 3; 10.5326/0390294
Citation: Journal of the American Animal Hospital Association 39, 3; 10.5326/0390294
Citation: Journal of the American Animal Hospital Association 39, 3; 10.5326/0390294
Citation: Journal of the American Animal Hospital Association 39, 3; 10.5326/0390294
Citation: Journal of the American Animal Hospital Association 39, 3; 10.5326/0390294
Representative computed tomography (CT) scan of a dog with a right-sided caudolateral maxillary tumor (i.e., fibrosarcoma) (CT at level of the carnasial tooth). Each slice of the CT is evaluated for presence of tumor, and the intra-oral (marked as 2) and dorsal (marked as 1) margins are defined. Rostral and caudal margins are also defined from examination of CT slices in those areas, and the resection is planned using a skull from a similar breed to aid in orientation. (Key: L=left; R=right; ETT=endotracheal tube; N=nasal cavity; R. Mand.=right mandible)
Photograph of lateral aspect of a cadaver dog’s head, showing the position of the lateral incision. The initial incision is in a rostrocaudal direction along the dorsolateral aspect of the nose, and it is continued caudally, sweeping below the eye, along the line of the zygomatic arch, if required. This is altered slightly, depending on the exact location of the tumor.
Photograph of cadaver specimen, showing the bipedicled flap that is created once the lateral incision in the skin and the intraoral incision in the mucosa of the buccal vestibule are joined.
Intraoperative view of a dog following creation of the bipedicled flap, showing the good access to the caudolateral maxilla.
Photograph of a dissection in a cadaver, showing retraction of the superficial levator nasolabialis (LNL) to show the course of the dorsal nasal (DNV), lateral nasal (LNV), and facial veins (FV). The facial vein is ligated where indicated (by X), but it is otherwise preserved during creation of the bipedicled flap to facilitate venous drainage from the area postoperatively.
Photograph of the skull of a dog, showing an example of location of the lateral osteotomies for tumor resection (the exact cuts will depend on where the tumor is located). The zygomatic arch (1) is generally cut first, if required; then the lateral (2) and dorsal (3) aspects of the maxilla and nasal bones are cut. Following these, the palatine bone is cut, and then the final cut (A) is made in the caudal maxilla or the medial orbit, depending on the extent of the tumor. If a partial orbitectomy is performed, the final cut is angled from position A to the caudomedial aspect of the desired resection in the oral cavity. This cut is most easily made using a large osteotome.
Contributor Notes
