Introduction

Mandibulectomy and maxillectomy are most commonly indicated for the treatment of oral tumors.^1–3 Additional indications for these procedures include severe trauma, osteomyelitis, lock jaw syndrome, open-mouth lower jaw locking, and dentigerous cysts.^4,5,6,7,8 Many of the reports^9,10,11 on mandibulectomy and maxillectomy have focused on reporting oncological outcomes following these surgical procedures (and not on complications from the procedures themselves). Surgical complications that have been associated with these procedures include hemorrhage,^11,12 aspiration pneumonia,^11,13,14 incisional dehiscence,^{1,2,3,11,12,13,15,16,17} mandibular drift and malocclusion,^{11,13,15,17,18} ranula formation,^11,13,18 anorexia,^11,17 nasal discharge or epistaxis,¹¹ difficulty prehending,^2,11,14 implant migration,^11,14 oronasal fistula formation,¹¹ and death.¹⁶ Additional complications, although not secondary to the procedure itself, have included tumor recurrence and metastasis.^{2,3,11,12,14,16,18}

Risk factors for complications associated with maxillectomy and mandibulectomy have been investigated to a limited extent. A recent study⁹ that evaluated complications in dogs undergoing partial maxillectomy revealed that tumor size, location (ie, caudal locations), and surgical approach were significantly associated with severe intraoperative hemorrhage requiring blood transfusion. This was the first study⁹ in veterinary medicine to focus on possible complications following orofacial surgery. However, the study⁹ had some limitations because it represented affected dogs from a specialized surgical oncology center, its focus was solely on maxillectomy, and a high percentage of affected dogs underwent a complete or caudal maxillectomy (51.8%). Some smaller studies^12,13,16,18 have speculated possible risk factors for potential surgical complications. For instance, in a study² that evaluated partial maxillectomy as a treatment for oral cancer in 61 dogs, the authors determined that caudal maxillectomy resulted in higher rates of partial wound dehiscence, compared with dogs that underwent rostral resections. One study⁹ cited that 40% of dogs that underwent a caudal maxillectomy received a blood transfusion. In contrast, another study¹⁹ cited that only 25% of dogs that underwent a caudal maxillectomy received a blood transfusion. Additionally, specific complications may be identified, but only rarely is the overall complication rate reported. This may be due to a low overall number of patients in these studies.^13,14,16,17

In a study,²⁰ complications following orthognathic surgery (corrective jaw surgery) are reported to occur in 19.2% of human patients. These complications include postoperative malocclusion, hemorrhage, inferior alveolar nerve injury, bad split (unfavorable fracture at osteotomy site), and infection. In that study,²⁰ complications were more common in men and associated with the increasing duration of surgery, increasing number of surgeries, surgical site involved, and type of osteotomy performed. Surgical site infection and aspiration pneumonia are other examples of common complications following oral surgery in humans.^21,22 One study²² documented hypoalbuminemia and increasing surgical time as important risk factors for postoperative complications following oral surgery. In patients with head and neck cancer, the rate of surgical site infection ranges from 10% to 45%, with previous chemotherapy, performance of a tracheotomy, free flap reconstruction, blood loss volume, and hypoalbuminemia found as risk factors.^{23,24,25,26,27} This type of investigation is recommended in veterinary medicine to inform both surgeons and pet owners of possible complications and risk factors for complications following these surgical procedures.

The purpose of the study reported here was to provide contemporary information about the complication rates and risk factors for complications following mandibulectomy and maxillectomy in dogs. Our hypotheses were that maxillectomies, specifically caudal procedures, would result in a greater risk of severe hemorrhage requiring a blood transfusion and carry a higher risk for incisional dehiscence or oral fistula formation. Additionally, we hypothesized a larger mass size would result in an increased risk of complications associated with both mandibulectomy and maxillectomy.

Materials and Methods

Results

Discussion

Findings in the present study indicated that there was a moderate risk of complication following mandibulectomy or maxillectomy, with 37.3% (171/459) of patients experiencing ≥ 1 complication. Of 271 total complications, 18 (7%) were considered catastrophic. In most studies that evaluated mandibulectomy and maxillectomy in veterinary medicine, specific complications are identified, but the overall complication rate is rarely reported. One study³ that examined mandibular resection as a treatment for oral cancer in dogs had an overall complication rate of 22%. Because performing mandibulectomy is often less technically challenging than maxillectomy, a lower complication rate may be expected. Interestingly, in the present study, both procedures had a similar overall complication rate in patients (mandibulectomy, 38% [106/279]; maxillectomy, 36.1% [65/180]). However, many of the complications associated with mandibulectomy were considered minor overall and therefore self-limiting (69.6% [110/158]). The maxillectomy procedure had fewer minor complications (49.5% [47/95]) and slightly more major and catastrophic complications (38.9% [37/95] and 11.6% [11/95], respectively). In a study¹¹ of curative-intent surgery for oral tumors in dogs, there was an overall complication rate of 16.2%. However, that study did not have a follow-up requirement, which could have led to an underestimation of the occurrence of surgical complications, compared with that of the present study. Another study⁴ of 8 dogs evaluated the occurrence of complications following partial maxillectomy for the treatment of oral tumors; there was a 100% complication rate within the first 48 hours after surgery and an 85% complication rate from 48 hours to 4 weeks after surgery. This was substantially higher than the complication rate for dogs (37.3% [171/459]) of the present study; this could be explained by the fact that a greater proportion of the maxillectomy procedures in the other study⁴ involved a caudal approach. Also, the other study⁴ recorded complications that were not reported in the present study, such as facial pawing and epiphora. These complications could not be included in the present study because of the retrospective nature and institutional variability in reporting. However, both studies did record many of the same complications, including epistaxis, facial swelling, orofacial fistula formation, incisional dehiscence, infection, and hemorrhage.

Contrary to our hypothesis, the size of a tumor was not associated with the occurrence of a complication on the basis of multivariable analysis findings. The surgical complexity required is expected to increase with larger tumor size, but this did not result in an increased risk of complications, including hemorrhage. An association between tumor size and excessive surgical bleeding followed by a blood transfusion has been reported in a previous study.⁹ It is important to keep in mind that the complexity of a lesion on the basis of size would also be affected by the size and breed of the dog. On multivariable analysis, only surgical time was independently associated with the occurrence of ≥ 1 complication. This finding is suspected to be secondary to an increase in surgical complexity, as resection or reconstruction could take longer. However, higher inherent risks such as hemorrhage or hemorrhage occurring persistently over a longer period could also extend surgical time. Prolonged surgical time has been associated with an increased risk of complications for oral surgery in human medicine.²³ To the authors' knowledge, no previous study evaluating mandibulectomy or maxillectomy procedures in veterinary medicine has demonstrated an increased risk of complications with surgical time.

Incisional dehiscence or oral fistula was the most common surgical complication in the present study, as demonstrated in previous studies.^3,11,13 Given the retrospective nature of the present study, these complications were combined because it was difficult to determine what was an actual fistula versus incisional dehiscence that resulted in a fistula. Three dogs developed orofacial fistulas following mandibulectomy. This finding was unexpected because oral fistulas are more commonly associated with maxillectomy procedures. For these dogs, it was difficult to discern whether the fistula formation was secondary to the surgical procedure itself, the underlying disease process, or the use of adjunctive treatment such as radiation treatment. Performing maxillectomy, as well as a procedure in the caudal aspect of the oral cavity, were both associated with an increased risk for occurrence of an incisional dehiscence or oral fistula in the univariable analysis. Only the maxillectomy procedure was confirmed on multivariable analysis as being associated with an increased risk for incisional dehiscence or oral fistula occurrence. It has been postulated that an increase in surgical site dehiscence of a maxillectomy procedure may be the result of an increased tumor size before detection by the owner or veterinarian, resulting in a more complex resection. However, this contradicts our finding that tumor size was not associated with an occurrence of a complication. Incisional dehiscence is suspected to be more common with caudal maxillectomy; however, this was not confirmed with multivariable analysis in the present study. This may be the result of a type II error, as only 43 of 459 (9.4%) dogs underwent caudal maxillectomy. In 1 study,² 80% of the dogs that developed wound dehiscence following partial maxillectomy had dehiscence that occurred caudally to the canine teeth. In the present study, a caudal procedure was defined as being performed caudal to the fourth premolar tooth.¹⁵ Our results may have been altered if the anatomic landmark for a caudal procedure had been different, making wound dehiscence a more frequent complication for caudal procedures. Most of the complications in the present study were minor (157/253 [62.0%]); therefore, surgical or medical intervention was not needed. However, dehiscence in the maxillary region is a serious condition because it can result in an oronasal fistula and further complications. Clients should be prepared for these possible complications and potential requirements for surgical revision.

Preoperative radiation therapy, chemotherapy, or both were also associated with an increased risk of surgical site dehiscence or oral fistula formation. To the authors' knowledge, the present study was the first to document this finding with veterinary orofacial surgery. However, this has been documented for humans who have undergone head and neck surgery and received preoperative chemotherapy.²⁷ In 1 study³³ of dogs with tumors, the risk of complications following cutaneous or mucosal flap surgery is higher for dogs that underwent preoperative radiation therapy. In the present study, it is also possible that neoadjuvant therapy was attempted in dogs with more advanced disease before surgical management, resulting in a selection bias. Radiation damage to healthy tissues (eg, stem cells, blood vessels, and others) can be progressive and permanent.²⁸ Both chemotherapy and radiation therapy can impede wound healing.²⁸ It is therefore important to consider the optimal timing for surgery following these treatments. It has been recommended that if radiation therapy is given preoperatively, surgery should be performed after acute radiation reactions have resolved (ie, generally 3 to 4 weeks, depending on the protocol).²⁸ The optimal timing of surgery following chemotherapy has not been assessed.

In the present study, univariate analysis revealed that a maxillectomy procedure and caudal resection were also associated with an increased risk of severe hemorrhage requiring blood transfusion. Similarly, MacLellan et al⁹ determined that dogs undergoing caudal maxillectomy or complete maxillectomy are significantly more likely to have had excessive surgical bleeding and 3 times as likely to have received a blood transfusion than dogs undergoing other maxillectomy procedures. This finding is suspected to be a result of the more complex vascular anatomy in this region and presence of larger vascular branches.⁹ Additional preparations could be made in these instances, such as determination of blood type, placement of multiple IV catheters for fluid or blood administration, and having both fluid volume support and blood products available. Temporary occlusion of the carotid artery on the side of resection could be considered, but this technique was not assessed in the present study.

Although tumor recurrence was not considered a surgical complication, risk factors were evaluated for tumor recurrence, given that a large proportion of patients were undergoing surgery for tumor removal (415/459 [90.4%]). Overall recurrence in the present study was low over a prolonged follow-up period (median follow-up time of 1,070 days), with only 54 (11.8%) dogs developing tumor recurrence and 1 (0.2%) dog with suspected but not confirmed tumor recurrence. In the multivariable logistic regression analysis, whether the tumor was malignant or benign and the largest tumor dimension were independently associated with tumor recurrence. As the tumor dimension increased, there was an increased risk of recurrence. This finding is similar to results of previous studies. Sarowitz et al¹¹ determined that a smaller tumor size (< 2 cm in diameter) is associated with a longer disease-free interval following curative-intent surgery. In the present study, malignant tumors were associated with increased odds of recurrence, compared with benign tumors, which follows what has been found in other studies.^11,12 Preoperative radiation therapy or chemotherapy did not decrease the likelihood of recurrence in the multivariable analysis. This finding may have been a result of the small number of dogs that received these treatments (n = 26).

Because the present study was a retrospective analysis performed at 8 referral institutions, there were several inherent limitations including variability of completeness of medical record information and a lack of standardization with surgical procedures, as these were performed by different surgeons; postoperative management including analgesia; and adjunctive treatment. Also, specifics regarding surgical treatment, such as the use of Rummel tourniquets or suture patterns used, were not investigated in the present study. A follow-up of 90 days was required; thus, some complications may have occurred after this time, resulting in an underestimation of surgical complications. Although most follow-up information was obtained from an in-hospital examination, a telephone conversation with the client was also sufficient, which could result in an underestimation of surgical complications. It is also possible that an overestimation of complications was made in the present study, as all complications that occurred following the procedure were included. Some of these complications may have been related to the anesthesia or case management versus the surgical procedure itself.

The present study provides new information regarding risk factors for the development of complications and provides veterinarians with contemporary information for patient outcomes when performing these specific procedures. These procedures are safe with a moderate risk for a complication but a low risk for a catastrophic complication. The hypothesis that both a maxillectomy procedure and a caudal resection increased the risk for incisional dehiscence or oral fistula was accepted. These procedures also increased the risk for severe hemorrhage requiring a transfusion. However, although the size of the lesion increased the risk for occurrence of a complication on the basis of univariable analysis, this effect was not confirmed on multivariable analysis. On the basis of our data, a larger mass as well as performing maxillectomy, particularly when involving a caudal resection, may carry an increased risk for complications. When attempting one of the procedures, the authors would recommend careful planning by use of imaging as well as care with reconstruction to minimize tension across the suture line. Blood products should be available for immediate use if necessary. Also, clients should be advised of these potential complications to allow for appropriate expectations.