Temporal bone fractures

Temporal Bone Fractures BENNY NAGERIS, MD, MAYNARD C. HANSEN, MD, WILLIAM G. LAVELLE, MD, FREDERICK A. VAN PELT, MD The diagnosis and monitoring of patients presenting to an emergency department with blunt temporal bone fracture and complications requiring acute management were reviewed for a four-month period. Of 104 trauma patients with closed head injury, 15 patients were diagnosed with temporal bone fracture, 12 of whom survived their injuries. Four patients developed cerebrospinal fluid (CSF) otorrhea and two patients developed facial nerve paralysis; all patients had resolution of complications with conservative management. Significant shortcomings in the initial evaluation and monitoring of patients with temporal bone fracture were identified. Specific and thorough facial nerve examinations were not initially conducted on temporal bone fracture patients and subsequent inpatient monitoring for facial nerve paralysis and CSF otorrhea was incomplete. The outcome of temporal bone fracture is discussed. This article reminds the emergency physician of the importance of initial diagnosis and documentation of temporal bone fractures. (Am J Emerg Meal 1995;13:211214. Copyright © 1995 by W.B. Saunders Company)

The temporal bone contains several neural and sensory structures that may be damaged in trauma. The possible complications involving these structures after temporal bone fracture are well documented in the literature, and their frequency is determined in part by the type of fracture; consequently, potential complications from temporal bone fracture must be carefully considered.l Although hearing loss is commonly associated with temporal bone fracture and may require surgical intervention, 2'3 the most important complications to be aware of in the acute trauma setting are facial nerve (CNVII) paralysis and cerebrospinal fluid (CSF) otorrhea or otorhinorrhea. The immediate onset of facial nerve paralysis could be an indication for emergent exploration and decompression; delayed onset suggests an edematous process and is treated conservatively with steroids and careful monitoring. 4'5 CSF otorrhea or otorhinorrhea, depending on the intactness of the tympanic membrane, occurs in 26% of temporal bone fractures and resolves spontaneously in more than 90% of cases. Indications for surgical repair, such as otorrhea prolonged beyond two weeks, recurrent leaks secondary to fistula formation, recurrent episodes of meningitis, and/or herniation of neural structures through the fracture, require patients to be closely monitored with careful documentation of the leak progression. 6 Although high-resolution computed tomography (CT) has From the Department of Otolaryngology-Head and Neck Surgery, University of Massachusetts Medical Center, Worcester, MA. Manuscript received January 31, 1994; revision accepted July 9, 1994. Address reprint requests to Dr Nageris, Department of Otolaryngology-Head and Neck Surgery, University of Massachusetts Medical Center, 55 Lake Ave North, Worcester, MA 01655. Key Words: Temporal bone fracture, CSF otorrhea, facial nerve paralysis. Copyright © 1995 by W.B. Saunders Company 0735-6757/95/1302-0021 $5.00/0

improved the radiological diagnosis of temporal bone fractures, 7 the diagnosis must often be based on clinical findings alone. Indeed, early evaluation of patients for complications of temporal bone fractures often requires clinical suspicion based on the relevant medical history and physical examination before radiological findings become available, s,9 Because of the relatively large number of blunt head injuries evaluated by the trauma team in our medical center, we were interested in determining the incidence of temporal bone fractures and of the complications requiring acute management, with specific focus on initial diagnosis, appropriate physical examination, and the subsequent monitoring and management of these patients. Initial history and physical evaluation begins by obtaining a complete history from the patient or witness regarding otorrhea, pressure or absence of facial movement, loss of consciousness, previous neuro-otologic deficit, and use of alcohol, drugs or other medications. Physical examination begins with inspection of periauricular soft tissue, external auditory canal, and tympanic membrane. If there is suspicion of CSF leak, the otologic examination should be conducted with sterile instruments. Note should be made of any external auditory canal skin laceration, deformity of the bony external auditory canal, tympanic membrane injury, hemotympanum, and the nature (bloody or clear) of any otorrhea. The presence of nystagmus should be recorded. Spontaneous nystagmus in the head-neutral position is usually pathologic. Horizontal nystagmus of greater intensity with eyes closed as opposed to eyes open indicates a peripheral end organ lesion whereas the opposite suggests a CNS etiology. Vertical nystagmus also implicates CNS pathology. If the patient's conditions allows, speech-reception threshold can be approximated for both ears with the Barany noisemaker, and the use of the 512 tuning fork can differentiate a conductive from sensorineural hearing loss with the Weber test. Complete or incomplete facial paralysis should be documented. Deep pain stimulation may be required if the patient is unresponsive. PATHOPHYSIOLOGY

Temporal bone fractures are classified as either longitudinal or transverse. The longitudinal is the most common (80%) and results from lateral trauma to the temporal and parietal area. The fracture line (Figure 1A) begins in the squamous portion of the temporal bone, extends along the posterosuperior external canal, through the middle ear, to the area of the foramen lacerum. Bleeding from the ear (from tympanic membrane and external auditory canal skin tear) and conductive hearing loss are the usual findings. The transverse fracture begins in the foramen magnum and extends in a direction perpendicular to the long axis of the temporal bone (Figure 1B). It may involve the internal 211

212

AMERICAN JOURNAL OF EMERGENCY MEDICINE • Volume 13, Number 2 • March 1995

FIGURE 1. Temporal bone fractures (arrows). (A) Longitudinal. (B) Transverse. (Reprinted with permission. ~3)

auditory canal, the capsule and facial canal, explaining the common presentations of hemotympanum, vertigo, sensorineural hearing loss, and facial nerve paralysis. Figure 2 shows longitudinal and transverse fractures on a CT scan image.

radiological findings consistent with the diagnosis of temporal bone fracture. This subset of charts was then further evaluated for documentation of appropriate physical findings used to diagnose CNVII paralysis and CSF otorrhea and to monitor these patients for progression of these complications.

METHODS

RESULTS

A retrospective chart review was conducted of all trauma cases evaluated by the trauma team during a four-month period. Charts of patients with head injuries were identified and subjected to a more detailed analysis for recorded historical, physical examination, and

Of 343 patients evaluated by the trauma team in a fourmonth interval, 104 (30.3%) had closed head injuries. Fifteen patients with temporal bone fracture were identified based on the clinical findings of hemotympanum, bleeding from the ear, CSF otorrhea, external auditory canal laceration, and/or Battle's sign; 80% (12/15) were confirmed by CT. Three of the 15 patients (20%) died in the ED or on the first day of hospitalization and were excluded from further study. C S F otorrhea developed in 33.3% (4/12) of surviving patients; in two patients, it was noted on initial presentation, and in the other two patients, it developed within the first nine days of admission. All cases apparently resolved spontaneously, although one patient developed meningitis. Two of the surviving patients (16.7%) developed facial nerve paralysis diagnosed as delayed in onset. An ears/nose/throat tENT) consultation was requested in only 50% of all temporal bone fracture cases. It is significant that the initial physical examination and the subsequent monitoring of all patients with temporal bone fractures was incomplete, particularly for ruling out complications requiring acute management. In 50% (5/10) of surviving patients with confirmed temporal bone fracture by CT, an ear exam was either not performed at all (1/15) or limited to the observation of bleeding from the ear (4/5); only 2 of these 5 patients received ENT consultations, and hemotympanum was demonstrated in both cases; the remaining 3 received no further ear examinations after admission. Although two patients were diagnosed with C S F otorrhea on initial physical examination and two other patients developed otorrhea during their admission, 2 of these 4 patients were not further examined for C S F leak resolution, and the

FIGURE 2. Axial CT of the temporal bone demonstrating a longitudinal fracture (straight arrow) and a transverse fracture (curved arrow).

NAGERIS ET AL • TEMPORAL BONE FRACTURES

2 with documented resolution failed to receive continued monitoring. Three of these 4 patients were discharged without a final examination for the absence of otorrhea. The most significant shortcoming in the management of all 12 surviving patients with temporal bone fractures was the complete absence of a specific CNVI1 examination on initial presentation; 58.3% (7/12) had no cranial nerve examination at all and 41.7% (5/12) were documented as " C N I I - X I I intact." The absence of an initial examination inappropriately eliminated the possibility of diagnosing immediate onset paralysis and may have resulted in the misdiagnosis of facial nerve paralysis in at least one of the two patients who developed paralysis, a patient who initially had presented with severe facial trauma and swelling. In both patients, the CNVII examination was not performed until significant paralysis had developed, and few of the other patients were monitored during their admission.

DISCUSSION Temporal bone fractures have been demonstrated to comprise 22% of all skull fractures 8 and were found in 14.4% of patients with closed head injuries in this study. Temporal bone fractures remain a clinical diagnosis in up to 20% of cases because of the limitations of radiological studies, and the importance of the initial history and physical examination cannot be underestimated. Hemotympanum with or without tympanic membrane laceration was found to be present in all surviving patients in whom the tympanic membrane had been visualized (75%); in the remaining patients, brisk bleeding from the ear limited the initial cursory attempt to visualize the tympanic membrane, and they did not undergo follow-up examination or receive an ENT consultation. Because the importance of the diagnosis of temporal bone fracture is to alert and prompt the trauma team to perform a more detailed physical examination to rule out associated complications requiring acute management, maximum effort should be made to visualize the tympanic membrane in all patients with head injuries, and those patients in whom visualization is impaired secondary to brisk bleeding should be assumed to have hemotympanum and temporal bone fracture until subsequent examinations demonstrate otherwise. CSF otorrhea is a relatively common complication of temporal bone fracture that fortunately resolves in more than 90% of cases. Nevertheless, all patients with temporal bone fractures require close monitoring for CSF otorrhea so that trends that indicate more aggressive surgical management are recognized. All of the 33% of patients who developed CSF otorrhea apparently experienced spontaneous resolution in spite of the absence of follow-up examination and documentation in half of the cases. Failure to document the progression of otorrhea in these cases could have jeopardized further management of this complication had this been necessary. Facial nerve paralysis can occur in 10% to 50% of temporal bone fractures, depending on the type of fracture. 9 All patients with suspected temporal bone fracture require a thorough, focused examination of CNVII. When an adequate examination is not possible secondary to severe facial trauma, documentation must justify its absence. The facial nerve can be examined in most trauma patients, conscious or

213

unconscious, by stimulation. The determination of immediate onset paralysis is of considerable importance because its presence is an indication for emergent nerve decompression as soon as the patient is stabilized. The latter has been reemphasized by Lambert and Brackman 1° on a large series of facial paralysis patients after temporal bone fracture. Barrsl notes that immediately after facial nerve injury, the proximal portion of the facial nerve begins vigorous axonal regeneration. Empty axonal tubules are seen as early as five days after trauma and are nearly complete at 21 days. These findings emphasize the need for early diagnosis of the paralyzed nerve as well as the need for early intervention in cases of immediate complete paralysis. Patients with delayed onset paralysis are managed conservatively and require close monitoring of the progression or resolution of paralysis .4 Facial nerve function was not initially evaluated in any of our patients with temporal bone fractures. CNVII paralysis described as delayed in onset developed in 16.7% (2/12) patients and their initial facial nerve examination was conducted only after significant paralysis had been noted. Although one patient required tarsorrhaphy, both patients had good return of facial function with conservative management. Although facial nerve paralysis may be readily apparent in conscious patients with temporal bone fractures in the absence of facial trauma, significant misdiagnosis can result in the unconscious patient or in the patient with multiple facial injuries and swelling unless specifically tested f o r ) 2 as may have occurred in one patient in our study. Regardless of the absence or the extent of associated injury, failure to perform an initial baseline CNVII exam inappropriately eliminates surgical intervention by mandating that all paralyses be classified as delayed. This can result in unfortunate patient outcomes as well as unfortunate medicolegal situations.

CONCLUSIONS The trauma review conducted on 343 patients over a fourmonth period demonstrated the absence of an adequate initial evaluation of facial nerve function in all 12 surviving patients with temporal bone fractures and inconsistent inpatient monitoring of facial nerve function and for CSF otorrhea in these patients.

REFERENCES 1. Goodwin WJ: Temporal bone fractures. Otolaryngol Clin North Am 1983;16:651-659 2. Olson JE, Shagets FW: Blunt Trauma of the Temporal Bone. St Louis, MO, Am Acad OtolaryngoI-Head & Neck Surg Found Inc, 1986, pp 1-73 3. Wennmo C, Spandow O: Fractures of the temporal bone-Chain incongruencies. Am J Otolaryngol 1993;19:38-42 4. Coker NJ, Kendall KA, Jenkins HA, Alford BR: Traumatic infratemporal facial nerve injury: Management rationale for preservation of function. Otolaryngol Head Neck Surg 1987;97: 262-269 5. Fisch U: Facial paralysis in fractures of the petrous bone. Laryngoscope 1974;84:2141-2154 6. Hicks GW, Wright Jr. JW, Wright III JW: Cerebrospinal fluid otorrhea. Laryngoscope 1980;90:1-25 (suppl 25) 7. Johnson DW: Temporal bone trauma: HRCT evaluation. Radiology 1984;151:411-415

214

AMERICAN JOURNAL OF EMERGENCY MEDICINE • Volume 13, Number 2 • March 1995

8. Cannon CR, Jahrsdoerfer RA: Temporal bone fractures. Arch Otolaryngol Head Neck Surg 1983;109:285-288 9. Waldron J, Hurley SEJ: Temporal bone fractures: a clinical diagnosis. Arch Emerg Med 1988;5:145-150 10. Lambert PR, Brackmann DE: Facial paralysis in longitudinal temporal bone fractures: A review of 26 cases. Laryngoscope 1984;94:1022-1026 11. Barrs DM: Traumatic facial nerve paralysis. In Gates (ed):

Current Therapy in Otolaryngology-Head and Neck Surgery (ed 5). St Louis, MO, Mosby, 1994, pp 118-123 12. Giasscock ME, Wiet RJ, Jackson CG, Dickins JRE: Rehabilitation of the face following traumatic injury to the facial nerve. Laryngoscope 1979 ;89:1389-1404 13. Nadol JB Jr, Schuknecht HF: Surgery of the Ear and Temporal Bone, Chapter 31 : Facial Nerve Injury and Decompression. New York, NY, Raven Press, 1993, p 332