Le Fort I osteotomy is the technique most applied worldwide in the treatment of dentoskeletal deformity involving the maxilla. Even though it is considered a very safe technique with good intra- and postoperative results, many complications have been described. This paper presents a case of carotid jugular fistula developed in a 22-year-old white male submitted to Le Fort I osteotomy for the treatment of anteroposterior maxillary deficiency, and discusses the possible aetiology and management of this serious complication.
The Le Fort I osteotomy, generally considered a safe procedure, is one of the techniques most commonly applied worldwide to correct deformities affecting the middle third of the face, and is also useful in cases that require access to skull base tumours. However, complications have been reported in the literature, the most common of which are haemorrhage, infection, and airway obstruction. Other complications have also been reported, such as perioperative dental trauma, exaggerated maxillary impaction with an unfavourable aesthetic result, oroantral and oronasal fistulae, nerve or duct damage, and, with lower incidence, velopharyngeal dysfunction and vascular complications.
A traumatic arteriovenous fistula is an abnormal communication between an artery and a vein caused by an incomplete tearing of both vessels. Most cases of traumatic arteriovenous fistulae involve the internal carotid artery and are more frequent after blunt or penetrating trauma, and more rarely as a result of surgical procedures such as orthognathic surgery .
Lesions to the internal carotid artery during Le Fort I osteotomy have been described in the literature. Although rare, it is a serious complication .
The aetiology of carotid jugular fistulae related to the treatment of dentoskeletal deformities is controversial, some authors attributing this complication to the sharp bony edge of the pterygoid complex detached during surgical fracture and others to the anatomical characteristics of the patient .
The following case report describes the occurrence of a right carotid–jugular arteriovenous fistula after a Le Fort I osteotomy performed for maxillary advancement and intrusion and discusses the possible causes and preventive measures to avoid this serious kind of complication.
A 22-year-old white male diagnosed with anteroposterior maxillary deficiency was admitted to the hospital for an orthognathic surgery procedure that consisted of a 5-mm maxillary advancement and a 2-mm intrusion. The usual Le Fort I osteotomies were performed, applying the reciprocating saw at the anterior and lateral aspect of the maxilla, the specific chisel at the nasal septum, and a curved chisel between the pterygoid process and the maxillary tuberosity. Nonetheless, downfracture of the maxilla was unsuccessful at the first attempt; all osteotomy cuts were then re-examined, and the same instruments and additional pressure (that is still compatible with the Le Fort I manoeuvre) were applied to complete the fracture. The bone interferences were removed, the maxilla easily positioned, and the procedure completed with the fixation of plates and screws from the 1.5-mm system (Tóride Indústria e Comércio Ltda., Mogi Mirim, São Paulo, Brazil).
The immediate postoperative period was uneventful and the patient was transferred from the post-anaesthetic recovery room to a standard occupancy room, completely awake and without any neurological symptoms. Eight hours after the surgical procedure, the patient presented one episode of nausea and vomiting, followed by lethargy. Unfortunately, the Oral and Maxillofacial Surgery (OMS) staff were not informed of these important events. Fifteen hours after the surgical procedure, the hospital nursing staff contacted the OMS staff, reporting that the patient presented left hemiplegia, hemiparaesthesia, and anisocoria. In addition to these signs, the patient was confused and complaining of severe headache and was promptly transferred to the intensive care unit (ICU) and the Neurosurgery staff were contacted. A cranial computed tomography (CT) scan was acquired, showing an ischemic area in the right middle cerebral artery region (MCA) without midline shift ( Fig. 1 A).
Angiography revealed a right internal carotid artery-to-jugular vein fistula at the level of the foramen lacerum, with high venous flow to the jugular bulb, superior and inferior petrous sinuses, cavernous sinus, and with reflux into the contralateral jugular vein ( Fig. 4 A,B).
After 3 days in the ICU, the patient was clinically stable, conscious, and maintaining left hemiplegia, hemiparaesthesia, and anisocoria. The patient complained of pulsating pain behind the right eye; however physical examination did not reveal chemosis, protosis, or a bruit at auscultation. Despite clinical stability, serial CT scanning revealed progressive enlargement of the hypodense area with midline shift ( Fig. 1 B). Owing to imminent herniation and risk of neurological decline, a decision was made to perform a decompressive craniectomy ( Fig. 1 C,D). A reverse question mark incision was made over the frontotemporoparietal region, extending down anterior to the tragus. A large hemicraniectomy was performed using a Midas Rex Legend drill (Medtronic Inc., Minneapolis, MN, USA) and additional bone was removed from the sphenoid wing, above the zygoma to decompress the basal cisterns. The dura was opened in a stellate fashion. To allow expansion of the oedematous brain tissue, the dura was left open and a large pericranium patch placed over it and anchored on five points. Finally, the galea layer and the skin were closed in a standardized fashion. One week after the craniectomy, the patient’s pupils were isochoric and he reported improvement of light touch sensibility on the left side of the body.
Twenty days after the craniectomy, the endovascular approach was planned as treatment for the fistula. Through an ipsilateral femoral approach, a vertebral 5F catheter was positioned in the left vertebral artery guided by the angiography. An Amplatz Super Stiff 0.035-mm Guidewire (Marlborough, Massachusetts, USA) was advanced beyond the fistula, followed by progression of the long inductor above the point of the fistula. The V12 6 × 38 mm stent was mounted on a balloon advanced over the 0.035-mm guide and positioned covering the carotid–jugular fistula. The stent was released under gradual balloon insufflation, aiming for the restitution of the usual calibre of the right internal carotid artery ( Fig. 4 C,D). After 30 days in hospital, the patient was discharged and sent for physical rehabilitation therapy.
At the 3-month follow-up, the patient showed significant improvement of hemiplegia and hemiparaesthesia, including being able to walk with the aid of crutches. A new CT scan revealed the stent position and a defect in the skull bones caused by decompressive craniectomy (Figs. 2 A–D and 4 C,D ). At the 36-month follow-up and after intensive physiotherapy, only moderate weakness and movement limitation in the left arm was present on the physical examination. The patient presented no difficulty in walking without aid and was able to drive without any car adaptations. It is important to note that the patient was extremely cooperative throughout the proposed treatment plan. The most recent CT scan shows the reconstruction of the bone defect caused by craniectomy with an acrylic prosthesis ( Fig. 3 E–H).
The occurrence of traumatic pseudoaneurysms and arteriovenous fistulae in the head and neck regions is rare, mostly associated with penetrating trauma caused by knife or gunshot wounds. Nonetheless, laceration because of a bone spur, or a skull base fracture involving the floor of the middle cranial fossa can also result in damage to the internal carotid artery . Furthermore, arteriovenous fistulae can appear in any kind of surgery involving the head and neck region, for example after a rhinoplasty or even a dental extraction. They may arise as a result of the simultaneous laceration of an artery and a vein .
Although relatively rare, major complications associated with the Le Fort I osteotomy, such as arteriovenous fistulae, brain abscesses, and even blindness have been described in the literature .
Recent papers have reported the occurrence of vascular damage related to the Le Fort I osteotomy, such as carotid-cavernous fistula and dissection of the internal carotid artery, as well as aneurysm, thrombosis, and laceration of the same artery . Other vascular lesions following orthognathic surgery have also been described, such as maxillary artery pseudoaneurysm following subcondylar osteotomy , or pseudoaneurysm of the facial artery following mandibular sagittal osteotomy .
The posterior portion of the sphenoid bone, which forms a bony protrusion below the foramen lacerum, has been hypothesized to be one of the possible causes of this type of lesion . Kang et al. studied the size, shape, and location of this bony protrusion, found in 71% of the adult skulls examined. The authors recommended caution during the pterygomaxillary separation, because if the base of the sphenoid bone is fractured, this protrusion can move upwards through the foramen lacerum and damage the internal carotid artery. Additionally, if the pterygomaxillary separation is incomplete and downfracture is attempted, the downward force can be transferred to the body of the sphenoid and the pterygoid plates, resulting in bad fractures and damage to the circumjacent structures. In summary, this bony protrusion can be displaced superiorly through the foramen lacerum during the surgical procedure damaging the anatomic structures related to this foramen. Anatomical variations of the base of the skull, such as incomplete ossification or bone defects next to the cavernous sinus, can contribute to the possibility of trauma to the internal carotid artery in its cavernous segment while performing maxillary orthognathic surgery . In our case, a right internal carotid artery-to-jugular vein fistula at the level of the foramen lacerum occurred following the Le Fort I osteotomy. The arterial-venous shunt caused a decrease of the blood flow in the middle cerebral artery. Probably because of the patient’s incomplete formation of the Circle of Willis ( Fig. 3 ), the blood flow deficit in the right cerebral hemisphere could not be compensated by the left carotid artery, thus causing the stroke.
To date, only seven reports of traumatic arteriovenous fistula related to orthognathic surgery were found via PubMed , Carneiro et al. being the most recent. This is a result of the rarity of this complication. The clinical transoperative signs observed in cases in which the carotid-jugular fistula was diagnosed after the surgical procedure performed for the treatment of dentoskeletal deformities are summarized in Table 1 . Difficulty in achieving the downfracture of the maxilla was evidenced in two cases , probably as a result of the incomplete pterygomaxillary osteotomy. The same difficulty in completing the downfracture was observed in the case described in this paper. In consequence, all the previously performed osteotomies were revised, applying the same instruments, before a new attempt was made to downfracture the maxilla. Unfortunately, even the easier performance of the downfracture made possible by the revision of the osteotomies did not prevent the occurrence of the carotid-jugular fistula.
|Article||Procedure||Affected side||Transoperative reports||Follow-up||After effects|
|Lanigan and Tubman||Le Fort I Osteotomy||Arteriovenous fistula present between the intercavernous portion of the left internal carotid artery and the cavernous sinus||Extreme difficulty in downfracturing the maxilla||8 months||Very slight diplopia on extremes of gaze to the left|
|Hes and de Man||Le Fort I Osteotomy||Fistula between the intracavernous portion of the right internal carotid artery and the cavernous sinus||No description of difficulties encountered during the surgery||2 years||None|
|Newhouse et al.||Le Fort I Osteotomy||Fistula involving the right internal carotid artery and the internal jugular vein at the base of the skull||Profuse arterial haemorrhage from the right posterior region after downfracture||3 months||Left leg hemiparesis|
|Suzuki et al.||Bilateral sagittal split osteotomy||Dissecting aneurysm of the left internal carotid artery||No difficulties were encountered during the surgery||1 year||Minimal right upper extremity paresis|
|Carneiro et al.||Surgically assisted rapid maxillary expansion||High-flow carotid cavernous fistula of the right internal carotid artery||No description of difficulties encountered during the surgery||6 months||None|
|Karaman et al.|| Functional endoscopic
|Carotid cavernous fistula of the left internal carotid artery||No description of difficulties encountered during the surgery||2 months||None|
|Levine and Super||Le Fort I osteotomy||Carotid cavernous fistula of the left internal carotid artery||Downfracture difficulty||12 months||Residual ophthalmoplegia on upward gaze of the left eye|
|Singhal et al.||Le Fort I osteotomy and bilateral sagittal split osteotomy||Right carotid artery dissection with thromboembolic middle cerebral artery occlusion||No description of difficulties encountered during the surgery||10 days||Dense left hemiparesis|
The traumatic carotid-jugular fistula is clinically presented with thrill on palpation and murmur on auscultation, with or without motor deficit, which may vary from hemiparesis to hemiplegia . As described in the case in this paper, the patient remained temporarily paralyzed on the left side of his body, because of a massive stroke caused by cerebral hypoperfusion, which manifested around 15 hours after the surgical procedure. Even after 36 months of intensive physiotherapy, moderate weakness and limitation of movement in the left arm were present on the physical examination.
To prevent damage to the descending palatine artery and diminish bone interferences during the Le Fort I osteotomy, Trimble et al. proposed the vertical posterior osteotomy through the maxillary tuberosity and not at the pterygomaxillary junction. This technical modification also has the potential to decrease the incidence of carotid-jugular fistulae for two possible reasons: (a) the downward force required to complete the fracture is diminished, and (b) the fracture runs mainly along the maxilla instead of along the pterygoid plate of the sphenoid bone. Furthermore, the use of instruments such as angulated serrated piezoelectric cutting tips to perform the complete osteotomy before attempting the downfracture, can also help in this matter.
Three final points: first, although the aetiology of the carotid-jugular fistula related to the Le Fort I osteotomy is still controversial, caution should be exercised during the downfracture manoeuvre, which is cited as a potential etiological factor. If any difficulty is encountered during downfracture, it is strongly recommended that all the osteotomies be revised before attempting to complete the manoeuvre.
Secondly, a delay in the referral of unusual symptoms may have the effect of exacerbating the patient’s clinical condition, indicating the absolute need for close monitoring in the early postoperative period.
Thirdly, Goffinet et al.’s findings led them to suggest the necessity of a multidisciplinary approach (involving maxillofacial, vascular, and neuroradiology teams) to the treatment of dento-midfacial deformities in relation to vascular malformations. Accordingly, they recommend this approach as the surest way to reduce the possibility of morbidity. Our findings amply confirm Goffinet’s suggestion, since the complication described here, although rare, was handled by vascular and neurosurgical teams, always accompanied by the maxillofacial team responsible for the orthognathic surgery.
The Ethics Committee issued the following statement “Accordingly the protocol (Resolution n° 466/12) of the School of Dentistry of Ribeirão Preto, University of São Paulo, the patient signed the informed consent that includes the authorization for use of image and medical information for scientific publication. The authors confirm that they have read the Helsinki Declaration and follow its guidelines”.