Abstract
Introduction
Temporomandibular joint (TMJ) pseudoankylosis is an uncommon clinical finding that is often a result of trauma or neoplasia in the dog.1–4 Pseudoankylosis or false ankylosis occurs when extra-articular structures are affected, creating a fibrous or osseous callus between bone and soft tissue, 5 which compromises the mobility of the joint. A true ankylosis is where intraarticular structures are primarily affected. 6 Regardless of a diagnosis of true versus false ankylosis, the result of decreased mobility may include difficulty with mastication, prehension, panting and poor oral hygiene as well as discomfort.1,2
Successful correction of TMJ ankylosis and pseudoankylosis has been accomplished surgically3,4 with the goals of removing the site of ankylosis, increasing mobility, preventing synostosis 4 and improving patient quality of life. Computed tomography (CT) with three-dimensional (3D) reconstruction allows superior diagnosis and evaluation of spatial positioning to guide surgical planning when compared with skull radiographs.7–9 Previously described surgical treatments for TMJ ankylosis and pseudoankylosis include mandibular excisional ostectomy, 5 mandibular condylectomy, 10 gap arthroscopy,3,4 and segmental mandibulectomy. 11
This case report characterizes a combination of gap arthroplasty and interpositional arthroplasty in a 2-year-old dog that had a combined TMJ ankylosis and pseudoankylosis with minimal TMJ mobility. Although interpositional arthroplasty is one of the three common surgical procedures for correction of TMJ ankylosis in humans, 12 it is not commonly performed in veterinary patients. Interpositional arthroplasty was described in a feline case report of TMJ ankylosis and pseudoankylosis. 13 In addition, the use of a 3D printed model for surgical planning and intraoperative spatial positioning proved to be a valuable tool to assist in the surgical repair of this temporomandibular pathology.
Materials and Methods
Signalment and History
A 2-year-old, Labrador retriever mixed-breed, male intact dog was referred to the Dentistry and Oral Surgery Service at the University of Wisconsin, Veterinary Teaching Hospital for inability to open the mouth with a history of suspected orofacial trauma. The patient was able to prehend soft food, however the inability to open the mouth was noted when he was unable to pick up toys or pant.
Initial Evaluation
On physical examination, the patient had lateral facial asymmetry due to left-sided swelling over the zygomatic arch. He had retained normal maxillo-mandibular relationship rostrocaudally. A conscious oral evaluation revealed marked calculus and gingivitis in caudal arches and an inability to open the mouth. The patient otherwise appeared clinically healthy. He was sedated with dexmedetomidine a (2.5 mcg/kg IV) and butorphanol b (0.2 mg/kg IV) and a CT scan c was performed and the data was presented in digital imaging and communications (DICOM) file format (Figures 1 and 2). The CT scan indicated the presence of chronic, comminuted fractures of the left temporal bone, zygomatic arch, and left coronoid process of the mandible as well as secondary pseudoarthrosis of the zygomatic arch with the coronoid process, which were well characterized with a 3D volume rendering of the DICOM data (Figures 3 and 4). The anesthetized patient had an inter-incisal distance measured at 0.7 cm. The DICOM was converted to standard tesselation language (STL) file format and segmented to highlight the region of surgical interest with Mimics. d The segmented STL file was then 3D printed in polylactic acid e with a fused-deposition modeling 3D printer f (Figure 5).

CT scan at the level of the TMJ, demonstrating involvement of the left temporal bone and the condyle resulting in a true TMJ ankylosis.

CT scan rostral to the TMJ demonstrating involvement of the left zygomatic arch and the ramus and coronoid process of the left mandible resulting in pseudoankylosis.

3D volume rendering of the CT scan with a lateral view of the left TMJ ankylosis and pseudoankylosis.

3D volume rendering of the CT scan with a dorsal view comparing the normal right TMJ with the ankylosed left TMJ.

3D printed section of the patient skull with the pathology to be used for surgical planning and anatomic landmark identification during surgery.
Surgical Procedure
Six weeks following the CT scan, the patient presented for surgical correction by performance of gap arthroplasty and interpositional temporalis myofascial transposition. Pre-anesthetic examination and blood work, including chemistry, total protein, and packed cell volume, were unremarkable except for a mild hyperglobulinemia of 3.7 g/dL (Normal range: 2.2-3.5 g/dL) likely due to inflammation. The patient's blood type was dog erythrocyte antigen (DEA) 1 and he was deemed an acceptable anesthetic candidate and received an American Society of Anesthesiologist's (ASA) status of 2.
The patient was premedicated with methadone g (0.25 mg/kg IM) and dexmedetomidine a (2.4 mcg/kg IM) and induced with propofol h (4.9 mg/kg IV). A 20-gauge catheter was placed in the left cephalic vein and an isotonic, balanced electrolyte solution i was administered at a constant rate infusion of 5 mL/hour. A 22-gauge catheter was placed in the left dorsal pedal artery for invasive blood pressure monitoring. Intubation was performed with a rigid endoscope and a stylet to place an endotracheal tube due to the inability to fully open the mouth with an inter-incisal distance that measured 7 mm. As difficulty intubating was anticipated, the team was prepared to perform a tracheostomy in the even that this was necessary to secure a safe airway. A size 11 mm endotracheal tube j was successfully passed. The cuff was inflated and anesthesia was maintained with 1–2% isoflurane k and an oxygen flow rate of 2 L/minute on a circle system along with mechanical ventilation. Continuous anesthetic monitoring included electrocardiogram, arterial blood pressure, pulse oximetry and end-tidal CO2. The patient's temperature was maintained with a forced-air warming device. l
Following intubation, the patient was prepared for surgery with hair clipped and rough and sterile scrubs performed over the left lateral aspect of the face. Pre-operative analgesia was provided using an extraoral approach to the left inferior alveolar regional nerve and an ultrasound-guided left trigeminal nerve block with bupivacaine m (37.5 mg total). Intraoperative analgesia was provided with ketamine n (10 mg/kg/min) and fentanyl o (10 mcg/kg/HR) continuous rate infusions. Cefazolin p (22 mg/kg, IV) was administered every 60 minutes intraoperatively and once 6 hours postoperative. The patient was transported to the operating room, placed in right lateral recumbency and the surgical site was sterilized and draped for surgery.
Surgical exploration of the left TMJ, zygomatic arch and coronoid process was performed with a #15 surgical blade q making a curvilinear skin incision along the ventral border of the left zygomatic arch to the TMJ and along the angle of the jaw. Subcutaneous tissues were bluntly dissected. The zygomatic arch was located and exposed following sharp dissection of the platysma muscle. The periosteum of the zygomatic arch was elevated, followed by reflection of the temporalis and masseter muscle attachments. Throughout the surgical procedure, the 3D model was used to identify precise anatomical structures that were difficult to recognize due to previous trauma and excessive fibro-osseus callus formation.
A rostral osteotomy was performed in the zygomatic arch at the level of the orbital ligament using a piezotome r and the thickened zygomatic arch with pseudoarthrosis to the coronoid process of the mandible was resected gradually with rongeurs. s The masseter muscle was reflected from the coronoid process using a periosteal elevator t and a second osteotomy was made horizontally through the ramus of the mandible, at the level of the mandibular fossa, preserving the mandibular artery, vein and nerve. Cadaveric images have been included for reference, as the surgical anatomy of the patient was close to unrecognizable (Figures 6 and 7). The TMJ was disarticulated and a final osteotomy was made through the temporal bone medial to the retroarticular process. Large winged elevators u were used to elevate larger bone segments, while additional osteoplasty was performed with rongeurs to smooth any sharp boney edges. Upon disarticulation of the mandibular condyle, the masticatory branches of the trigeminal nerve and the maxillary artery were exposed. A superficial flap from the caudoventral aspect of the temporalis muscle was sharply dissected with fascia attached and reflected caudoventrally to facilitate attachment to the periosteum of the remaining temporal bone using 3-0 Biosyn v suture in a simple interrupted pattern. The role of the temporalis myofascial flap was to function as an autologous interpositional material in an effort to prevent continued osseous healing and further pseudoankylosis (Figures 8 and 9). Prior to closure, the surgical wound was flushed thoroughly with sterile saline.

Cadaveric example representing skin incision and fascial incision on ventral aspect of the zygomatic arch. The dashed lines represent the dorsal and ventral borders of the zygomatic arch prior to skin incision.

Cadaveric example of the osteotomy through the ramus of the mandible after removal of the zygomatic arch. Excess masseter muscle has been excised to enhance visualization for image, which is not done on the clinical patient.

Cadaveric example of the temporalis myofascial flap rotated caudoventrally, prior to suturing over the temporal osteotomy site.

Cadaveric example of temporalis myofascial flap rotated caudoventrally and sutured over the temporal bone osteotomy site.
Throughout the surgery, hemostasis was performed by digital pressure, bipolar cautery, w and ligatures as needed. The neurovascular structures in the region of the TMJ were identified during the surgery in an effort to avoid unnecessary surgical trauma. The palpebral branch of the facial nerve that courses along the dorsal aspect of the zygomatic arch and the auriculopalpebral nerve that lies caudal to the retroarticular process were both avoided. The zygomatic arch periosteum was attached to temporalis and masseter muscles surgically with 3-0 Biosyn v suture in a simple interrupted pattern. Nocita x (100 mg) was injected locally at the incision site. Subcutaneous tissue was closed in a simple continuous pattern with 3-0 Monocryl y suture. The dermis was closed with 3-0 Monosof z suture in a cruciate pattern.
Immediately following surgery, a CT was performed, confirming complete ostectomy of left coronoid process, the zygomatic process of the temporal bone, the mandibular condyle and the squamous portion of the temporal bone. The patient recovered uneventfully from anesthesia and meloxicam aa (0.2 mg/kg subcutaneously) was administered. He was hospitalized overnight on a fentanyl o CRI, maintenance volume of intravenous fluids i (2 mL/kg/hr) and soft food. The patient ate readily and opened and closed its mouth comfortably with no evidence of mandibular deviation. He was sent home on gabapentin bb (10 mg/kg orally every 8 hours) and meloxicam aa (0.1 mg/kg orally every 24 hours) for 5–6 days of analgesia with specific instructions to monitor the incision for signs of infection or dehiscence. It was also recommended that the patient transition to a dental diet and rubber chew toys to encourage use of the masticatory apparatus.
Results
Phone communication with the caregivers reported normal chewing behaviors and no indication of pain two days post-operatively. The patient was consciously evaluated 20 days post-operative for skin suture removal and evaluation of occlusion. During conscious examination, the inter-incisal distance was measured at 70 mm and the patient exhibited no indication of pain. A mild mandibular drift to the left could be observed when the mouth was opened, however this corrected as he closed his mouth. Left masseter and temporalis muscle atrophy was appreciated upon palpation. Physiotherapy, including the encouragement to play with and carry rubber toys was recommended along with a dental diet to encourage continued movement of the masticatory apparatus.
Three months post-operatively, the patient was anesthetized for castration, to address periodontal disease, and repeat the skull CT scan. The foster owner continued to report normal chewing behaviors with no evidence of discomfort. The patient had normal occlusion and anesthetized inter-incisal distance was measured at 76 mm. Marked temporalis and masseter muscle atrophy of the left side was noted. The CT results demonstrated no evidence of re-ankylosis and there were smooth, rounded ostectomy sites (Figure 10). The left caudal mandibular molar teeth (310 and 311) were surgically extracted due to advanced periodontal disease. The owner was educated to carry out a home oral hygiene routine as well as physiotherapy via rubber balls and to supply a dental diet. Annual, anesthetized oral evaluation and cleaning was recommended. The most recent contact with the owner was greater than 48 months after the surgery and the patient was reported to be doing well with normal range of motion, normal occlusion based on maxillary and mandibular canine interlock and the ability to pick up toys and pant.

3D volume rendering of CT scan 3 months after surgery with no evidence of re-ankylosis or regrowth of bone at osteotomy sites.
Discussion
TMJ gap arthroplasty with interpositional temporalis myofascial transposition is an appropriate surgical option for correction of TMJ pseudoankylosis in the canine. In the current case, the procedure provided immediate improvement in jaw mobility and quality of life. Potential intra- and post-operative complications with this surgical approach include hemorrhage from the maxillary and mandibular arteries, mandibular drift, neurovascular damage and paralysis, and synostosis. 4
To prepare for challenging surgeries and minimize risk, appropriate diagnostic imaging modalities should be employed. CT is a more sensitive technique for detection of TMJ abnormalities than is conventional radiography7,8,14,15 and thus should be utilized. In addition, CT images can be reconstructed into 3D images that are used for spatial positioning and preoperative planning. 3 Three-dimensional reconstructed images have been previously evaluated for surgical treatment planning in cases of pseudoankylosis and ankylosis of the TMJ with success.3,4 With the advancement and availability of 3D printers, these 3D images may be used to produce a real-life model of the skull. 4 Intraoperatively, the 3D printed skull was frequently evaluated and compared to the surgical site at various angles and it was critical in determining landmarks in otherwise indistinguishable tissue which was due to previous trauma. The use of the 3D model was imperative for the navigation and avoidance of potentially fatal complications and for resection of planned tissue.
With the wider availability and decreasing cost of 3D printing, this tool may be utilized in veterinary medicine for surgical planning as well as teaching. The use of 3D printed models is especially helpful to distinguish individual variations, evaluate pathology and enhance visualization of small spaces. In this case, a challenging procedure was successfully performed using a 3D printed skull with arguably better outcomes than other salvage procedures such as segmental mandibulectomy or ostectomy of the callus.
Additional large, retrospective case studies involving 3D printed models as well as determination of surgical treatment and outcomes of TMJ ankylosis and pseudoankylosis are recommended to evaluate the risks of re-ankylosis and other complications. In addition, long-term follow-up is recommended to evaluate for degeneration of contralateral TMJ, mandibular drift, and occlusal trauma.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
a.
Dexdomitor®, Zoetis, Parsippany-Troy Hills, New Jersey.
b.
Torbugesic®, Zoetis, Parsippany-Troy Hills, New Jersey.
c.
GE HiSpeed LX/i CT; GE Medical Systems, Milwaukee, Wisconsin.
d.
Mimics Innovative Suite, Materialize, Leuven, Belgium.
e.
Polylactic acid, Makerbot Industries, Brooklyn, New York.
f.
Ultimaker 3, Ultimaker, Waltham, Massachusetts.
g.
Methadone, Akorn Inc, Lake Forest, Illinois.
h.
Propofol, Zoetis, Parsippany-Troy Hills, New Jersey.
i.
PlasmaLyte, Baxter International, Deerfield, Illinois.
j.
Dynarex Corporation, Orangeburg, New York.
k.
Isoflurane, Dechra Pharmaceuticals, Northwich, United Kingdom.
l.
Bair hugger™, 3M™, St. Paul, Minnesota.
m.
Bupivacaine, Hospira, Inc., Lake Forest, Illinois.
n.
Ketamine, Hospira, Inc., Lake Forest, Illinois.
o.
Fentanyl, Abbott Laboratories, Chicago, Illinois.
p.
Cefazolin, Pfizer, New York City, New York.
q.
Keystone Industries, Gibbstown, New Jersey.
r.
Piezosurgery® touch unit, Mectron SpA, Carasco, Italy.
s.
HIS Small bone rongeurs, Henry Schein, Melville, New York.
t.
EX7 Periosteal elevator, Henry Schein, Melville, New York.
u.
Patterson Veterinary Supply, Devens, Massachusetts.
v.
Biosyn™, Medtronic, Minneapolis, Minnesota.
w.
Bipolar cautery, Bovie® Medical Corporation, Clearwater, Florida.
x.
Nocita®, Aratana Therapeutics, Leawood, Kansas.
y.
Monocryl®, Ethicon®, Bridgewater, New Jersey.
z.
Monosof™, Medtronic, Minneapolis, Minnesota.
aa.
Metacam®, Boehringer Ingelheim, Ingelheim, Germany.
bb.
Gabapentin, Amneal Pharmaceuticals, Bridgewater Township, New Jersey.
