The role of angiography in managing patients with temporal bone fractures: a retrospective study of 64 cases.
We conducted a retrospective study of the utility of angiography in the evaluation of patients with temporal bone fractures. Our study population was made up of 64 patients--58 males and 6 females, aged 14 to 75 years (mean: 35.3)--with a temporal bone fracture who had presented to a level I trauma center over a 1-year period. Records were reviewed and data were obtained on the mechanism of injury; the type of fracture; associated injuries, particularly neurocranial injuries detected on computed tomography (CT) of the head; and any angiographic findings that might have been obtained. Theprimary outcomes measures were the type of treatment administered (conservative vs. surgical) and mortality. Patients were assigned to 1 of 4 groups according to CT results and angiographic findings, if any: normal CT and no angiogram (group 1; n = 12), abnormal CT and no angiogram (group 2; n = 28), abnormal CT and an abnormal angiogram (group 3; n = 9), and abnormal CT and a normal angiogram (group 4; n = 15). Conservative treatment was administered to all 12 patients in group 1 and to 9 patients (60%) in group 4; surgical treatment was provided to two-thirds of the patients in both group 2 and group 3. Mortality was low in group 1 (n = 0), group 3 (n = 1; 11%), and group 4 (n = 1; 7%), but high in group 2 (n = 10; 36%). 1n fact, the key finding of this study was that mortality in the group with an abnormal CT and no angiogram (group 2) was significantly higher than mortality in the group with an abnormal CT and an abnormal angiogram (group 3) (p = 0.02), even though the injuries in the 2 groups were similarly severe and their management was similarly aggressive. We conclude that current guidelines for angiography may need to be expanded to include all patients who have CT evidence of neurocranial injury in order to detect those vascular injuries that need aggressive management and thus lower overall mortality.
Temporal bone fractures have been noted to occur in as many as 22% of patients who sustain head trauma. (1) The most common causes of these fractures are motor vehicle accidents, which account for as many as one-third of cases. (2) Included among the other common causes are assaults, falls, and motorcycle accidents. The initial diagnosis of these fractures can be made on physical examination and confirmed by computed tomography (CT).
When otolaryngologists are consulted to participate in the care of these patients, we are primarily concerned with the assessment of facial nerve function and hearing status and the detection of a cerebrospinal fluid (CSF) leak. However, the severity of the head injury in these patients often leads to delays in otolaryngologic evaluation and management.
Vascular injuries of the cranium in patients with temporal bone fractures are rare, but when they do occur, they often have serious implications and they may ultimately play a large role in the patient's prognosis. To the best of our knowledge, the incidence and nature of vascular injuries in patients with temporal bone fractures have not been reported in the otolaryngology or trauma literature. The study of choice for determining the presence and extent of vascular injuries in patients with head trauma, including those with temporal bone fractures, is angiography, (3) although CT is commonly used as a screening tool for vascular injury in many trauma centers. The presence of neurocranial injury, as well as the suggestion of a vascular injury on CT, may then prompt the clinician to obtain confirmatory angiography.
We conducted a study to attempt to determine if obtaining an angiogram is helpful in the management of patients with temporal bone fractures. Our hypothesis was that angiography is necessary in the workup of all patients with temporal bone fractures, regardless of the presence or absence of neurocranial abnormalities seen on head CT.
Patients and methods
After obtaining approval from the Institutional Review Board at the University of Tennessee (UT) Health Sciences Center, we conducted a retrospective chart review of all patients with temporal bone fractures who had been seen over a 1-year period at the UT Regional Medical Center in Memphis, which is a level I trauma facility. Charts were reviewed for patient demographics; the mechanism of injury; the type of fracture; associated injuries, particularly neurocranial injuries detected on CT of the head; and any angiographic findings that might have been obtained. Outcomes measures included the type of treatment (conservative vs. surgical) and mortality.
A total of 64 patients--58 males and 6 females, aged 14 to 75 years (mean: 35.3)--with temporal bone fractures were identified from the trauma registry during the study period. The three most common mechanisms of injury were motor vehicle accidents (47% of patients), assaults (23%), and gunshot wounds (14%) (figure 1). Fracture patterns included mixed/oblique in 45% of patients, comminuted in 20%, classic longitudinal in 20%, and classic transverse in 14%.
In an effort to define our patients in a manner that might help us confirm or disprove our hypothesis, we classified these 64 patients into 4 groups based on CT and angiographic findings:
[FIGURE 2 OMITTED]
* Group 1: Patients who had no evidence of neurocranial injury on CT and who had not undergone angiography (n = 12);
* Group 2: Patients who did have CT evidence of neurocranial injury but who had not undergone angiography (n = 28);
* Group 3: Patients with evidence of neurocranial injury on CT who had an abnormal angiogram (n = 9); and
* Group 4: Patients with evidence of neurocranial injury on CT whose angiogram was normal (n = 15).
As the preceding breakdown indicates, angiography had been performed on 24 patients, and 9 of them (38%) had abnormal findings. The most common abnormalities were pseudoaneurysm (n = 4), carotid dissection (n = 3), and vasospasm (n = 2);2 patients had 2 abnormalities each (figure 2). Demographic variables were equally distributed among the 4 groups of patients.
At our trauma center, the management of blunt and penetrating head injury is administered according to institutional protocols that have been developed by the surgical trauma department over time. In cases of blunt cerebrovascular injury, four-vessel cerebral angiography was performed whenever any of the following findings were observed: an unexplained neurologic deficit inconsistent with CT findings, the presence of Homer syndrome, a LeFort II or III fracture, a cervical spine injury, or a neck soft-tissue injury. In cases of penetrating head and neck injury, angiography was performed in all stable patients with platysmal violation, regardless of whether the injury occurred in zone I, II, or III. Four-vessel angiography was chosen over other modalities such as CT angiography and magnetic resonance angiography (MRA) per the protocols of our institution. At the time these patients were evaluated, the lack ofa multidetector CT scanner precluded the performance of CT angiography. The magnetic resonance imaging unit in place had a 0.2-Tesla field strength, which is not sufficient for good quality diagnostic MRA. These reasons likely played a role in why four-vessel angiography was performed as a first step in the protocol.
Also, our institutional protocols require low-osmolar iodinated contrast to be administered to patients with high creatinine levels prior to the performance of a cerebral angiogram. However, in this study, we did not encounter any patients who were in renal failure or who had an allergy to contrast. This is not altogether surprising, since our sample size was fairly small and most patients who present to trauma centers tend to be in good general health.
The data collected were analyzed using simple chi square ([chi square]) analysis for all categorical data and the unpaired Student t test for continuous variables. This analysis was performed with the help of the SAS statistical package (SAS Institute; Cary, N.C.).
Type of treatment. Conservative treatment was administered to all 12 patients with normal CT results (group 1) and to 9 of the 15 patients (60%) with an abnormal CT but a normal angiogram (group 4) (table). Surgery was performed on two-thirds of the patients with an abnormal CT and either no angiogram (group 2) or an abnormal angiogram (group 3).
Mortality. Mortality was lowest (0%) in those patients whose CT was negative (group 1) (table). Among patients with an abnormal CT, mortality was significantly higher in those patients who did not undergo angiography (group 2) than in those with an abnormal angiogram (36 vs. 11%; p = 0.02), even though their management was just as aggressive and their injuries were no more severe. Mortality among patients with an abnormal CT and a normal angiogram (group 4) was 7%.
The extent of the otolaryngologist's involvement in the management of temporal bone fractures is often dictated by the severity of the associated injuries. The presence of an intracranial vascular injury can have grave prognostic implications and can accordingly alter management. (4) For example, knowledge of vasospasm and its exact location on angiography can ensure a safer wound exploration in cases when it is warranted. Also, in a patient undergoing radical mastoidectomy for a CSF leak, prior knowledge of a thrombosed sigmoid sinus would be valuable for operative planning (figure 3). As far as we know, the presence and nature of accompanying vascular injuries have not been reported in the context of temporal bone fractures in either the otolaryngology or trauma literature. Therefore, we believe that our findings are especially relevant.
The demographics of our patient population are generally in concordance with those described in the trauma literature. (1) The fact that the demographics of our 4 study groups were also similar conferred a certain amount of validity to our intergroup comparisons. Also, the mechanisms of temporal bone trauma in our study parallel the common causes cited by Cannon and Jahrsdoerfer. (1)
The fact that 38% of the angiograms in our study were abnormal leads us to believe that angiography should be considered whenever there is CT evidence of neurocranial injury. Conservative management was more often performed in patients who had a negative CT and no angiogram (group 1) and in those with an abnormal CT and a normal angiogram (group 4). Given that all of the patients in group 1 had been treated conservatively and that the mortality in this group was 0%, we believe that angiography is not necessary when there is no evidence of neurocranial abnormality on CT. Mortality was highest (36%) in patients who had an abnormal CT and no angiogram (group 2) despite the fact that (1) these patients were managed just as aggressively as the patients with an abnormal CT and an abnormal angiogram and (2) their injuries were no more severe. We therefore believe that if the indications for angiography are broadened, more vascular injuries might be detected, and this in turn could lead to more aggressive management when indicated and ultimately lower overall mortality. Indeed, our findings suggest the need to expand the indications for angiography to all patients with CT abnormalities on initial evaluation.
When we first reviewed our data, one of our concerns was that the injuries in our group 2 patients might have been so severe that angiograms were not deemed to be warranted despite the CT evidence of intracranial injury. However, further review revealed that of the 10 fatalities in this group, only 2 patients likely had a dismal prognosis from the outset--1 patient with a brainstem laceration and 1 patient with diffuse axonal injury. All the other patients in this group simply did not meet the criteria for angiography.
[FIGURE 3 OMITTED]
Our study did have limitations that are inherent in all retrospective studies (i.e., the study design did not allow us to establish causality), but we believe that our findings are significant enough to warrant prospective studies in future.
In conclusion, angiography can play a valuable role in the management of patients with temporal bone fractures. Based on our findings, we believe that angiography is not necessary when there is no evidence of n eurocranial injury on CT. However, we also believe that the current indications for obtaining angiography should be expanded to include all cases in which there is evidence of neurocranial injury on CT in order to detect vascular injuries that might be managed aggressively in an effort to decrease overall mortality.
(1.) Cannon CR, Jahrsdoerfer RA. Temporal bone fractures. Review of 90 cases. Arch Otolaryngol 1983;109(5):285-8.
(2.) Brodie HA, Thompson TC. Management of complications from 820 temporal bone fractures. Am J Otol 1997;18(2):188-97.
(3.) Baker WE, Wasserman J. Unsuspected vascular trauma: Blunt arterial injuries. Emerg Med Clin North Am 2004;22(4):1081-98.
(4.) Teknos TN, Joseph MP, Megerian CA, et al. Carotid artery hemorrhage resulting from temporal bone fracture. Am J Otolaryngol 1997;18(5):338-40.
K. Asif Ahmed, MD; David Mison, MD; Wesley S. Whatley, MD; Rakesh K. Chandra, MD
From the Department of Otolaryngology, University of Tennessee Health Sciences Center, Memphis.
Corresponding author: K. Asif Ahmed, MD, 9695 Cutter Lane, Lakeland, TN 38002. Phone: (731) 422-0340; fax: (731) 422-0297; e-mail: firstname.lastname@example.org
Previous presentation: The information in this article was originally presented at the Southern Section meeting of the Triological Society; Jan. 12-14, 2006; Naples, Fla.
Table. Type of management approach and outcomes in the 4 groups Treatment Outcome * Conservative Surgical Survival Mortality Group n (%) n (%) n (%) n (%) 1 (n = 12) 12 (100) 0 12 (100) 0 Normal CT, no AG ([dagger]) 2 (n = 28) 9 (32) 19 (68) 18 (64) 10 (36) Abnormal CT, no AG 3 (n = 9) 3 (33) 6 (67) 8 (89) 1 (11) Abnormal CT, abnormal AG 4 (n = 15) 9 (60) 6 (40) 14 (93) 1 (7) Abnormal CT, normal AG * Mortality was significantly higher in patients with an abnormal CT who did not undergo angiography (group 2) than in those with an abnormal CT who had an abnormal angiography (group 3); [chi square] = 9.840944, p = 0.019968. ([dagger]) CT = computed tomography; AG = angiography. Figure 1. Chart shows the various mechanisms of injury in the 64 cases of temporal bone fracture (MVA = motor vehicle accident; GSW = gunshot wound). MA 47% Assault 23% GSW 14% Other 16% Note: Table made from bar graph.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||ORIGINAL ARTICLE|
|Author:||Ahmed, K. Asif; Mison, David; Whatley, Wesley S.; Chandra, Rakesh K.|
|Publication:||Ear, Nose and Throat Journal|
|Date:||May 1, 2009|
|Previous Article:||Latrogenic nasal saddling--a consequence of aphakia.|
|Next Article:||Patient history and CT findings in predicting surgical outcomes for patients with rhinogenic headache.|