SOURCE: UTMB Dept. of Otolaryngology Grand Rounds
DATE: March 12, 1997
RESIDENT PHYSICIAN: Debora P. Wilson, M.D.
FACULTY: Jeffrey T. Vrabec, M.D.
SERIES EDITOR: Francis B. Quinn, Jr., M.D.
The large majority (around 80%) of temporal bone fractures are classified as longitudinal. Longitudinal fractures are typically caused by a direct blow to the lateral skull, usually the parietal region. They commonly cross the middle ear space and external canal. This may lead to laceration of the external auditory canal, tympanic membrane rupture, and/or ossicular chain disruption. This explains the common finding of a conductive or mixed hearing loss in these patients. It is estimated that 20% of longitudinal fractures result in facial paralysis.
Transverse fractures are considerably less common than longitudinal fractures of the temporal bone (around 20%). They are frequently associated with more severe, even fatal trauma. They are seen in association with intense blows to either the occiput or a direct frontal blow. Severe sensorineural hearing loss and loss of vestibular function due to violation of the otic capsule is not uncommon. It is estimated that transverse fractures are associated with a 50% incidence of facial nerve paralysis.
The literature is somewhat confusing about the incidence of mixed temporal bone fractures. While some authors state that a "pure" longitudinal or transverse fracture is rarely seen, temporal bone fractures tend to follow these patterns and hence are classified as such. If a fracture is very comminuted and/or has fracture lines both parallel and perpendicular to the long axis of the petrous pyramid, they are called mixed.
An adequate history regarding details of the injury should be gathered from the patient, family members, and/or medical personnel attending to the patient. Of particular importance is the time of onset of any facial nerve weakness or paralysis. Physical examination should include a thorough head and neck exam including the cranial nerves. Look for evidence of a basilar skull fracture, which includes hemotympanum and Battle's sign. Also note any signs of bloody or serous discharge from the ear or nose. The external ear should be cleansed under aseptic conditions. Radiological evaluation should include a high resolution CTscan of the temporal bone. With serious penetrating injuries to the temporal bone such as gunshot wounds and stabbings, a carotid arteriogram should also be included.
Penetrating trauma to the temporal bone is generally more extensive and severe than blunt trauma. Nevertheless, they can both result in significant morbidity. The sequelae of temporal bone injury can include the following: hearing loss, facial nerve weakness or paralysis, dizziness, CSF otorrhea and rhinorrhea, and TMJ dysfunction. Less common but serious complications of temporal bone fractures include brain herniation and exsanguination.
Transverse fractures more frequently result in severe sensorineural hearing loss while longitudinal fractures are more likely to cause conductive or mixed hearing loss. Even without apparent temporal bone fracture, head injury alone can cause hearing loss, most likely due to labyrinth and/or cochlear concussions.
Evaluation of hearing loss is preferably done by audiometric testing that includes pure-tone air and bone conduction and speech discrimination scores. A presumptive evaluation can usually be done with tuning forks.
Sensorineural hearing loss has no effective medical treatment, unless there is evidence of perilymphatic fistula. A patient with a conductive hearing loss that persists after resolution of hemotympanum should be evaluated for ossicular chain discontinuity.7 High resolution CT scanning has been proven superior to MRI in evaluating temporal bone fractures and assessing the status of the ossicular chain.8 The most common site of ossicular dislocation is separation of the incudostapedial joint, followed by fracture of the stapes crura and dislocation of the incus with the malleus.
In a review by Tos on the prognosis of hearing loss in temporal bone fractures, he found that 80 % of cases of conductive hearing loss from longitudinal fractures resolved spontaneously, while cases of total sensorineural hearing loss due to transverse fractures showed no improvement 19
Facial nerve paresis or paralysis is frequently categorized as immediate-onset or delayed-onset. Weakness should be considered immediate if recognized within the first few hours after injury. The weakness should be classified according to the House- Brackman grading system.
Traditional topognostic testing to determine the site of injury is of minimal value. High resolution CT scanning is the diagnostic tool of choice for evaluating the facial nerve in temporal bone trauma. In a study published in 1990 by Haberkamp et al, Gadolinium- enhanced MRI was found to be helpful in accurately predicting the site of facial nerve injury. They found three patterns of nerve enhancement with Gd-enhanced scans: no enhancement, focal enhancement, and diffuse enhancement. It appeared that when focal enhancement of the nerve was produced by trauma, Gd-MRI scanning may be useful in revealing the lesion site. 9
Two studies thoroughly evaluate facial paralysis in the longitudinal fracture of the temporal bone. In 1974, Fisch published his review of 28 traumatic facial nerve injuries associated with longitudinal temporal bone fractures.10 He found that 93% of the facial nerve pathology was seen in the labyrinthine segment of the facial nerve. Lambert and Brackman in 1984 confirmed this finding.11 They reported in their series of 26 patients that 80 % facial nerve injuries found at surgical exploration were in the perigeniculate region. These two studies differ as to the type of injury to the facial nerve. Fisch found the highest rate of injury was due to intraneural hematoma (50%) followed by transection of the nerve (26%), and bony impingement upon the nerve (17%). In contrast, Lambert and Brackman found that most injuries in their study were due to bony impingement (45%) followed by contusion (36%), transection (9%), and intraneural hematoma (9%).10,11
While most agree that traumatic facial nerve paresis, or weakness can be closely observed, the management of traumatic facial nerve paralysis remains controversial.
In 1974, Fisch recommended basing the decision for surgery on the time of onset of paresis, the degree of paresis, the degree and evolution of degeneration as measured by electroneurography, and the degree and evolution of regeneration. He postulated that an immediate-onset paralysis with evidence of slow degeneration had a better prognosis than a delayed-onset paralysis with rapid degeneration. Fisch recommended early surgery for patients presenting with greater than 90% nerve degeneration by ENOG within six days of onset of palsy. He also recommended delayed surgery for patients with facial nerve paralysis six months after onset without any evidence of regeneration.10
In Lambert and Brackman's series published in 1984, they did not segregate patients according to time of onset of paralysis but rather on electrical testing by the maximal stimulation test or by electroneurography. They recommended surgery for all patients with facial nerve paralysis who meet electrical criteria based on electrodiagnostic testing irregardless of the time of onset of the paralysis.11
In 1986, Coker et al published a retrospective review of 29 cases of traumatic facial nerve paralysis. All patients who met electrical criteria for nerve degeneration based on nerve excitability testing or ENOG were surgically explored. In contrast to Fisch, Dr. Coker's group did not find that the rate of nerve degeneration was an indication of the severity of the injury or a predictor of facial nerve recovery. Dr. Coker's group developed a management algorithm based on their findings including recommendations for surgical approach. The surgical approach was dictated by the level of hearing . If the patient had serviceable hearing, facial nerve exploration was approached by a transmastoid and/or middle fossa approach. If the patient had near total sensorineural loss, the facial nerve was explored using a transmastoid/ translabyrinthine approach.12
In 1992, McKennan and Chole evaluated 32 patients with facial paralysis following temporal bone trauma. They segregated patients into groups based on the time of onset of their paralysis. Patients with delayed-onset paralysis were treated medically (usually with steroids) and electrical testing was not performed. The immediate-onset group was divided into a nonsurgical group and a surgical group. Surgical exploration was based on Hilger stimulation testing. All patients with immediate-onset paralysis met criteria for surgery based on electrical testing, however the three patients who were not surgically explored either did not consent or surgery was deferred due to other severe injuries. McKennan and Chole found an 84-94% complete recovery of facial nerve function in the delayed-onset group. They also observed that patients with immediate-onset paralysis due to penetrating temporal bone trauma had a high incidence of facial nerve transection that required repair. As to the patients with immediate-onset paralysis due to closed head injury, they concluded that further study was needed. A criticism by the editor of the journal in which the article was published was that because electrical testing was not done on the delayed- onset group, the status of nerve degeneration could not be evaluated. The excellent results found in the delayed-onset group could be because the injuries were not severe.13
The facial nerve may be injured by direct mechanical trauma or indirectly by heat generated by drilling. Heat damage can be lessened by constant suction-irrigation. If facial nerve paralysis is noted post-operatively, it may be due to the effects of lidocaine injection. If the paralysis persists after a few hours, surgical exploration is indicated. If the facial nerve is directly damaged during surgery, the extent of the injury should be fully assessed by exposing the injured segment by 5-10mm. The nerve should then be stimulated both proximally and distally. If the nerve can be stimulated with 0.05ma, the prognosis is good and no treatment is necessary. If the nerve can only be stimulated distal to the injured segment, then the nerve should be more widely exposed and synkinesis is likely.
Green et al reviewed a series of 22 patients with iatrogenic facial nerve injuries treated at the House Ear Clinic between 1963 and 1990. If the facial nerve was transected less than 50%, it was treated with decompression. If the transection was greater than 50%, the nerve was either reanastomosed or grafted. Green found that the most common location for iatrogenic injury was the lower tympanic segment. Results of the study indicated that the extent of injury to the nerve was underestimated either due to granulation tissue on the nerve or due to limitations on exposure of the nerve. Green found that patients who underwent primary anastomosis had a better outcome than patients that had cable grafting. His group did not find a benefit to early repair and they postulated that lack of tension on the anastomosis and nerve alignment were likely the most important factors.15 Bleeding
Iatrogenic injury to the sigmoid sinus or jugular bulb is usually easily controlled with placement of Gelfoam over the tear. If a larger tear is produced, pressure should be applied and the potential of an air embolism should be considered. The symptoms of an air embolism include hypotension, increased end expiratory CO2 and abnormal cardiac sounds. Immediate treatment includes placing the patient in a left lateral decubitus position with their head down. If cardiovascular collapse is imminent, direct aspiration of air from the vena cava is necessary (by the anesthesiologist). Bleeding from injury to the intratemporal carotid is difficult due to lack of exposure. Bleeding should initially be controlled with pressure. Distal and proximal control of the vessel is then necessary, with the potential need for ligation. Sensorineural Hearing Loss and Vestibular Injury
Sensorineural hearing loss and vertigo can occur from direct injury to the cochlea and/or labyrinths. They may also result from removal of cholesteatoma that has eroded the stapes footplate, round window, or semicircular canal. The vertigo is usually acute with gradual improvement over weeks due to compensation. CSF Leakage
Due to the proximity of the dura when operating on the temporal bone, a tear in the dura with subsequent CSF leak is a risk. A laceration in the dura noted intraoperatively should be repaired with sutures and covered with fascia or muscle to prevent a CSF leak. If a significant CSF leak that does not respond to conservative measures occurs post- operatively, surgical exploration may be necessary.
Insertional trauma from cochlear implantation has been studied. Kennedy confirmed damage to the spiral ligament from electrode insertion. He also emphasized the need to cease insertion of the electrodes at the point of first resistance, as widespread damage to the cochlea has been demonstrated.17 Welling et al studied and compared the insertional trauma caused by three different electrode designs. His group found that all three designs (Cochlear/Nucleus, Symbion/Inneraid, and Storz/UCSF) caused cochlear damage. The Nucleus device appeared to cause the least amount of damage and was the easiest to insert, while the Storz/UCSF coil was the most difficult to insert.18
2) Youngs, R., Deck, J., Kwok, P., Hawke, M.: Severe Sensorineural Hearing Loss Caused by Lightening. Arch Otolaryngol Head and Neck Surg, 114:1184-1187, 1988
3) Weinberger, D.G., Selesnick, S.H.: Roller Blade Falls - A new Case of Temporal Bone Fractures: Case Reports. J. of Trauma, 37:500-3, 1994
4) McGuirt, W.F., Stool, S.E.: Temporal Bone Fractures in Children: A Review with Emphasis on Long-term Sequelae. Clinical Pediatrics, 31:12-18, 1992
5) Kamerer, D.B.: Middle Ear and Temporal Bone Trauma. In: Head and Neck Surgery - Otolaryngology. B.J. Bailey (Ed)., J.B. Lippincott, Philadelphia, 1993.
6) Kinney, S.E.: Trauma. In: Otolaryngology - Head and Neck Surgery. C.W. Cummings (Ed)., Mosby, St. Louis, 1986.
7) Hasso, A.N., Ledington, J.A.: Traumatic Injuries of the Temporal Bone. Otolaryngologic Clinics of North America, 21: 295-317, 1988
8) Zimmerman, R.A., Bilaniuk, L.T., Hackney, D.B., Goldberg, H.I., Grossman, R.I.: Magnetic Resonance Imaging in Temporal Bone Fracture. Neuroradiology, 29:246-51, 1987
9) Haberkamp, T.J., Harvey, S.A., Daniels, D.L.: The Use of Gadolinium-Enhanced Magnetic Resonance Imaging to Determine Lesion Site in Traumatic Facial Paralysis. Laryngoscope, 100:1294-1300, 1990
10) Fisch,U.: Facial Paralysis in Fractures of the Petrous Bone. Laryngoscope, 84:2141-54, 1974.
11) Lambert, P.R., Brackmann, D.E.: Facial Paralysis in Longitudinal Temporal Bone Fractures: A Review of 26 Cases. Laryngoscope, 94: 1022-1026, 1984.
12) Coker, N.J., Kendall, K.A., Jenkins, H.A., Alford, B.R.: Traumatic Intratemporal Facial Nerve Injury: Management Rational for Preservation of Function. Otolaryngol Head Neck Surgery, 97: 262-269, 1987.
13) McKennan, K.X. and Chole, R.A.: Facial Paralysis in Temporal Bone Trauma. Am J Otol, 13: 167-172, 1992.
14) Avrahami, E.: CT of Intact but Nonfunctioning Temporomandibular Joints Following Temporal Bone Fracture. Neuroradiology, 36: 142-143, 1994.
15) Green, J.D., Shelton, C., Brackmann, D.E.: Surgical Management of Iatrogenic Facial Nerve Injuries. Otolarygol Head Neck Surg, 111: 606-610, 1994
16) Chole, R.A. and Brodie, H.A.: Surgery of the Mastoid and Petrosa. In: Head and Neck Surgery - Otolaryngology. B.J. Bailey (Ed.), J.B. Lippincott, Philadelphia, 1993
17) Kennedy, D.W.: Multichannel Intracochlear Electrodes: Mechanism of Insertion Trauma. Laryngoscope, 97:42-49, 1987.
18) Welling,D.B., Hinojosa, R., Gantz, B.J., Lee, J.T.: Insertional Trauma of Multichannel Cochlear Implants. Laryngoscope, 103: 995-1001, 1993.
19) Tos, M. Prognosis of Hearing Loss in Temporal Bone Fractures. J Laryngol Otol, 85:1147-1159, 1971
20) Caniff, J.P.: Otorrhea in Head Injuries. Br J Oral Surg, 8:203, 1971