TY - JOUR
T1 - Diagnosing mild traumatic brain injury
T2 - Where are we now?
AU - Dutton, Richard P.
AU - Prior, Kate
AU - Cohen, Robin
AU - Wade, Christine
AU - Sewell, John
AU - Fouche, Yvette
AU - Stein, Deborah
AU - Aarabi, Bizhan
AU - Scalea, Thomas M.
N1 - Copyright:
Copyright 2011 Elsevier B.V., All rights reserved.
PY - 2011/3
Y1 - 2011/3
N2 - Background: The brain acoustic monitor (BAM), an indicator of cerebral autoregulation, has previously shown high sensitivity but low specificity for computed tomographic (CT) abnormality in patients following the clinical diagnosis of traumatic brain injury. We assessed the utility of the BAM in diagnosing mild TBI (mTBI) in patients with and without normal findings of CT scan, a population for which there are a few objective markers of disease. Methods: We prospectively studied 369 patients with mechanism of injury consistent with TBI. The diagnosis was evaluated by five methods: (a) study enrollment (i.e., mechanism of injury), (b) signs of head trauma, (c) expert physician assessment, (d) presence of initial symptoms (loss of consciousness [LOC]; amnesia), and (e) BAM. All patients had a head CT scan. We compared the BAM screen results with the diagnosis of mTBI and BAM data from 50 normal volunteers and 49 trauma control patients not thought to have TBI. Results: None of the diagnostic methods correlated well with the others. Correlation between the methods ranged from 21% to 71%. BAM discriminated between patients with mTBI versus without TBI (p < 0.01) and patients with mTBI versus normal subjects (p < 0.001). There were 14 patients with new abnormal findings of CT scans. A history of LOC and physical signs of head injury were associated with a new abnormality on head CT (p < 0.05 and p < 0.01, respectively), whereas an abnormal BAM signal was suggestive (p = 0.08). The sensitivity of BAM abnormality for head CT abnormality was 100%, with a specificity of 30.14%. Conclusion: There is no gold standard for the diagnosis of mTBI. BAM screening is a useful diagnostic adjunct in patients with mTBI and may facilitate decision making. An abnormal BAM reading adds significance to LOC as a predictor of a new abnormality on head CT. In our study, opting not to CT scan patients with a normal BAM signal would have missed no new CT findings and no patients who required medical intervention for TBI, at a cost savings of $202,950.
AB - Background: The brain acoustic monitor (BAM), an indicator of cerebral autoregulation, has previously shown high sensitivity but low specificity for computed tomographic (CT) abnormality in patients following the clinical diagnosis of traumatic brain injury. We assessed the utility of the BAM in diagnosing mild TBI (mTBI) in patients with and without normal findings of CT scan, a population for which there are a few objective markers of disease. Methods: We prospectively studied 369 patients with mechanism of injury consistent with TBI. The diagnosis was evaluated by five methods: (a) study enrollment (i.e., mechanism of injury), (b) signs of head trauma, (c) expert physician assessment, (d) presence of initial symptoms (loss of consciousness [LOC]; amnesia), and (e) BAM. All patients had a head CT scan. We compared the BAM screen results with the diagnosis of mTBI and BAM data from 50 normal volunteers and 49 trauma control patients not thought to have TBI. Results: None of the diagnostic methods correlated well with the others. Correlation between the methods ranged from 21% to 71%. BAM discriminated between patients with mTBI versus without TBI (p < 0.01) and patients with mTBI versus normal subjects (p < 0.001). There were 14 patients with new abnormal findings of CT scans. A history of LOC and physical signs of head injury were associated with a new abnormality on head CT (p < 0.05 and p < 0.01, respectively), whereas an abnormal BAM signal was suggestive (p = 0.08). The sensitivity of BAM abnormality for head CT abnormality was 100%, with a specificity of 30.14%. Conclusion: There is no gold standard for the diagnosis of mTBI. BAM screening is a useful diagnostic adjunct in patients with mTBI and may facilitate decision making. An abnormal BAM reading adds significance to LOC as a predictor of a new abnormality on head CT. In our study, opting not to CT scan patients with a normal BAM signal would have missed no new CT findings and no patients who required medical intervention for TBI, at a cost savings of $202,950.
KW - Brain acoustic monitor
KW - Diagnosis
KW - Traumatic brain injury
UR - http://www.scopus.com/inward/record.url?scp=79952788049&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=79952788049&partnerID=8YFLogxK
U2 - 10.1097/TA.0b013e31820d1062
DO - 10.1097/TA.0b013e31820d1062
M3 - Article
C2 - 21610342
AN - SCOPUS:79952788049
SN - 0022-5282
VL - 70
SP - 554
EP - 559
JO - Journal of Trauma - Injury, Infection and Critical Care
JF - Journal of Trauma - Injury, Infection and Critical Care
IS - 3
ER -