Abstract
Le Fort I osteotomy is a routine procedure for oral and maxillofacial surgeons. Following advances in instrumentation and anaesthesia, it is usually carried out safely as an elective procedure in hospitals with no adverse complications. Life-threatening complications are rare although the operation is performed in an area with an extensive vascular supply. The authors report a case of Le Fort I osteotomy that resulted in an unusual complication of cerebrospinal fluid leak.
Case report
A 19-year-old caucasian female was referred for assessment of her facial deformity. Her main complaint was a long face with excessive exposure of the maxillary teeth. She had a Class II Div I malocclusion on a Class II skeletal base. She was assessed in a joint orthognathic clinic and a routine orthognathic work-up was performed. Cephalometric tracing revealed an increased lower anterior facial height with a retrusive mandible. She underwent pre-surgical orthodontics including dental decompensation. The surgical plan consisted of a Le Fort I maxillary impaction and bilateral sagittal split osteotomy to advance the mandible to Class I relation.
The maxilla was impacted 4 mm anteriorly and 5 mm posteriorly. In the pterygoid region, unusually dense bone created some difficulty in impacting the maxilla, but the expected movement was achieved after bone removal with a bur and rongeurs. The mandible was advanced to a Class I occlusal relationship after performing a bilateral sagittal split osteotomy as planned. The operation was otherwise uneventful and so was her immediate recovery from anaesthesia.
Postoperatively, the patient was poorly motivated and did not tolerate oral intake well; she was kept in hospital for monitoring and intravenous fluids. On the third postoperative day the patient complained of a persistent clear fluid discharge from the nose. On examination there was an obvious discharge of clear fluid from the left nostril. A sample of the fluid was collected and sent for laboratory analysis to establish the diagnosis. The results demonstrated fluid high in glucose and was positive for beta 2 transferrin confirming the presence of a cerebrospinal fluid (CSF) leak. Neurological examination was unremarkable and plain film radiographs showed satisfactory position of the osteotomy segments with nothing else of note.
A CT cysternogram showed evidence of CSF leak from the floor of the left middle cranial fossa at the site of attachment of the pterygoid plates with CSF pooling in the left nostril ( Fig. 1 ). Following neurosurgical consultation, a decision was made to treat conservatively and allow the CSF leak to settle spontaneously. Postoperative antibiotic (Co-amoxiclav) was stopped at this point. The leak did not resolve after 10 days. Following further discussions with the neurosurgeons a lumbar drain was placed to decrease the CSF pressure and allow the leak to resolve. There was a reduction in the amount of drainage in the first 4 days but the leak failed to resolve and the lumbar drain was re-sited. On the sixth day following lumbar drain insertion, the CSF leak ceased and the patient was discharged. During her hospital stay she suffered no neurological symptoms or deficit. On her follow-up visits she had made a complete recovery.
Discussion
Le Fort I osteotomy, alone or in combination with a mandibular osteotomy, is carried out safely in a hospital setting on a daily basis. Surgical complications are relatively rare and the common early complications include nerve injury and malocclusion. Intra operatively it is not uncommon to observe brisk haemorrhage, especially following maxillary osteotomy due to mechanical disruption of blood vessels, including greater palatine artery and pterygoid venous plexus.
CSF is a colourless fluid that occupies the subarachnoid space and the ventricular system of the brain. Produced by the ependymal cells that line the ventricles of the brain and the central canal of the spinal cord, the principal role of the CSF is protection of the brain against mechanical and chemical injuries. CSF is produced at the rate of 500 ml/day, but a major portion of this is reabsorbed into the blood stream through arachnoid villi, which are finger-like projections of the arachnoid that project into dural venous sinuses, especially the superior sagittal sinus .
CSF leak may occur through the nose or ear from traumatic or iatrogenic causes. Untreated CSF leaks can result in meningitis in 25–50% of cases . Most traumatic CSF leaks cease spontaneously within 7 days but others may require active surgical intervention. CSF and nasal secretions are often indistinguishable and tests are required to confirm the diagnosis. The glucose content of fluid alone is not diagnostic as nasal secretions also contain glucose. A positive beta 2 transferrin assay is considered to be diagnostic of a CSF leak. Treatment includes bed rest, hydration and steroid. A lumbar puncture/drain is carried out with the intention of reducing the CSF pressure to arrest the leak. If the leak persists, open craniotomies to close the dural tear surgically may be required.
CSF leakage is not uncommon following trauma to the facial skeleton. The authors think that CSF leak as a complication of Le Fort I osteotomy is rare. A similar picture in Le Fort III osteotomy can be accounted for anatomically by the close proximity of the osteotomy cuts to the cranial base. In the present patient it is likely that the CSF leak occurred as a result of a fracture at the attachment of the pterygoid plate to the base of the skull, but the exact mechanism is not clear. Retrospectively, analysing the operative procedure, there are two possibilities: the CSF leak could have occurred at the time of pterygomaxillary dysjunction or at the time of removal of bone in the posterior maxilla and pterygoid region to allow maxillary impaction. Unusually thick bone in this region, in this patient, could have resulted in a bad split, which resulted in this unusual complication.
Unfavourable pterygoid plate fracture is well studied and documented . Precious et al. reported the incidence of pterygoid plate fracture as approximately 80%, irrespective of the technique used for pterygomaxillary dysjunction. They compared dysjunction with Obwegeser chisel and Tessier spreading forceps . Postoperative CT studies of the pterygomaxillary region following Le Fort I osteotomy have demonstrated that the incidence of unfavourable fractures of the pterygomaxillary region may be underestimated. It is possible that many of these unfavourable fractures are unnoticed as there is no associated CSF leak perhaps due to a local soft tissue seal. Renicke et al. reported an incidence of 58% for pterygoid plate fracture following Le Fort I osteotomy using postoperative CT scans . Lanigan et al. recommended the use of a micro-oscillating saw for pterygomaxillary dysjunction in Le Fort I osteotomies, to reduce the incidence of high level unfavourable pterygoid plate fractures .
Studies on cadaver models have demonstrated a higher than anticipated incidence of pterygoid plate fractures. Robinson and Hendy reported an incidence of 75% pterygoid plate fractures in their cadaveric models . They observed high and low level fractures of the pterygoid region. Hiranuma et al. in their strain distribution study showed that the medial pterygoid plate is vulnerable to fractures at the time of pterygomaxillary separation irrespective of the type of chisel used . Appropriate angulation of the chisel is the most significant and decisive factor in achieving a favourable separation at the pterygomaxillary region.
Ophthalmic complications, such as blindness, ocular palsy and diplopia, have been reported following Le Fort I osteotomy . Atypical fractures of the pterygoid region propagating to the orbital apex and the base of skull have been implicated as the cause of these complications. Girotto et al. reported similar incidents of CSF leak following Le Fort I osteotomy that resolved following placement of a lumbar drain .
The incidence of unfavourable fractures of the pterygoid plate following pterygomaxillary dysjunction for Le Fort I osteotomy is possibly underestimated. Most of these fractures probably go undetected clinically as there are no associated complications. Surgeons should be aware of the potential for an unfavourable pterygoid plate fracture and subsequent CSF leak in Le Fort I osteotomies, especially in those patients requiring posterior maxillary impaction. Early recognition of this complication and institution of appropriate treatment is key in the management of these cases to avoid the risk of meningitis.